diff options
Diffstat (limited to 'scraper/reports/first_pages.html')
| -rw-r--r-- | scraper/reports/first_pages.html | 48171 |
1 files changed, 48171 insertions, 0 deletions
diff --git a/scraper/reports/first_pages.html b/scraper/reports/first_pages.html new file mode 100644 index 00000000..1fa50094 --- /dev/null +++ b/scraper/reports/first_pages.html @@ -0,0 +1,48171 @@ +<!doctype html><html><head><title>First pages</title><link rel='stylesheet' href='reports.css'></head><body><h2>First pages</h2><table border='1' cellpadding='3' cellspacing='3'><tr><td>611961abc4dfc02b67edd8124abb08c449f5280a</td><td>Exploiting Image-trained CNN Architectures +<br/>for Unconstrained Video Classification +<br/><b>Northwestern University</b><br/>Evanston IL USA +<br/>Raytheon BBN Technologies +<br/>Cambridge, MA USA +<br/><b>University of Toronto</b></td><td>('2815926', 'Shengxin Zha', 'shengxin zha')<br/>('1689313', 'Florian Luisier', 'florian luisier')<br/>('2996926', 'Walter Andrews', 'walter andrews')<br/>('2897313', 'Nitish Srivastava', 'nitish srivastava')<br/>('1776908', 'Ruslan Salakhutdinov', 'ruslan salakhutdinov')</td><td>szha@u.northwestern.edu +<br/>{fluisier,wandrews}@bbn.com +<br/>{nitish,rsalakhu}@cs.toronto.edu +</td></tr><tr><td>610a4451423ad7f82916c736cd8adb86a5a64c59</td><td> Volume 4, Issue 11, November 2014 ISSN: 2277 128X +<br/>International Journal of Advanced Research in +<br/> Computer Science and Software Engineering +<br/> Research Paper +<br/> Available online at: www.ijarcsse.com +<br/>A Survey on Search Based Face Annotation Using Weakly +<br/>Labelled Facial Images +<br/>Department of Computer Engg, DYPIET Pimpri, +<br/><b>Savitri Bai Phule Pune University, Maharashtra India</b></td><td>('15731441', 'Shital A. Shinde', 'shital a. shinde')<br/>('3392505', 'Archana Chaugule', 'archana chaugule')</td><td></td></tr><tr><td>6156eaad00aad74c90cbcfd822fa0c9bd4eb14c2</td><td>Complex Bingham Distribution for Facial +<br/>Feature Detection +<br/>Eslam Mostafa1,2 and Aly Farag1 +<br/><b>CVIP Lab, University of Louisville, Louisville, KY, USA</b><br/><b>Alexandria University, Alexandria, Egypt</b></td><td></td><td>{eslam.mostafa,aly.farag}@louisville.edu +</td></tr><tr><td>61ffedd8a70a78332c2bbdc9feba6c3d1fd4f1b8</td><td>Greedy Feature Selection for Subspace Clustering +<br/>Greedy Feature Selection for Subspace Clustering +<br/>Department of Electrical & Computer Engineering +<br/><b>Rice University, Houston, TX, 77005, USA</b><br/>Department of Electrical & Computer Engineering +<br/><b>Carnegie Mellon University, Pittsburgh, PA, 15213, USA</b><br/>Department of Electrical & Computer Engineering +<br/><b>Rice University, Houston, TX, 77005, USA</b><br/>Editor: +</td><td>('1746363', 'Eva L. Dyer', 'eva l. dyer')<br/>('1745861', 'Aswin C. Sankaranarayanan', 'aswin c. sankaranarayanan')<br/>('1746260', 'Richard G. Baraniuk', 'richard g. baraniuk')</td><td>e.dyer@rice.edu +<br/>saswin@ece.cmu.edu +<br/>richb@rice.edu +</td></tr><tr><td>61084a25ebe736e8f6d7a6e53b2c20d9723c4608</td><td></td><td></td><td></td></tr><tr><td>61542874efb0b4c125389793d8131f9f99995671</td><td>Fair comparison of skin detection approaches on publicly available datasets +<br/>a. DISI, Università di Bologna, Via Sacchi 3, 47521 Cesena, Italy. +<br/><b>b DEI - University of Padova, Via Gradenigo, 6 - 35131- Padova, Italy</b></td><td>('1707759', 'Alessandra Lumini', 'alessandra lumini')<br/>('1804258', 'Loris Nanni', 'loris nanni')</td><td></td></tr><tr><td>61f93ed515b3bfac822deed348d9e21d5dffe373</td><td>Deep Image Set Hashing +<br/><b>Columbia University</b><br/><b>Columbia University</b></td><td>('1710567', 'Jie Feng', 'jie feng')<br/>('2602265', 'Svebor Karaman', 'svebor karaman')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>jiefeng@cs.columbia.edu +<br/>svebor.karaman@columbia.edu, sfchang@ee.columbia.edu +</td></tr><tr><td>6180bc0816b1776ca4b32ced8ea45c3c9ce56b47</td><td>Fast Randomized Algorithms for Convex Optimization and +<br/>Statistical Estimation +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2016-147 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-147.html +<br/>August 14, 2016 +</td><td>('3173667', 'Mert Pilanci', 'mert pilanci')</td><td></td></tr><tr><td>61f04606528ecf4a42b49e8ac2add2e9f92c0def</td><td>Deep Deformation Network for Object Landmark +<br/>Localization +<br/>NEC Laboratories America, Department of Media Analytics +</td><td>('39960064', 'Xiang Yu', 'xiang yu')<br/>('46468682', 'Feng Zhou', 'feng zhou')</td><td>{xiangyu,manu}@nec-labs.com, zhfe99@gmail.com +</td></tr><tr><td>612075999e82596f3b42a80e6996712cc52880a3</td><td>CNNs with Cross-Correlation Matching for Face Recognition in Video +<br/>Surveillance Using a Single Training Sample Per Person +<br/><b>University of Texas at Arlington, TX, USA</b><br/>2École de technologie supérieure, Université du Québec, Montreal, Canada +</td><td>('3046171', 'Mostafa Parchami', 'mostafa parchami')<br/>('2805645', 'Saman Bashbaghi', 'saman bashbaghi')<br/>('1697195', 'Eric Granger', 'eric granger')</td><td>mostafa.parchami@mavs.uta.edu, bashbaghi@livia.etsmtl.ca and eric.granger@etsmtl.ca +</td></tr><tr><td>61efeb64e8431cfbafa4b02eb76bf0c58e61a0fa</td><td>Merging Datasets Through Deep learning +<br/>IBM Research +<br/><b>Yeshiva University</b><br/>IBM Research +</td><td>('35970154', 'Kavitha Srinivas', 'kavitha srinivas')<br/>('51428397', 'Abraham Gale', 'abraham gale')<br/>('2828094', 'Julian Dolby', 'julian dolby')</td><td></td></tr><tr><td>61e9e180d3d1d8b09f1cc59bdd9f98c497707eff</td><td>Semi-supervised learning of +<br/>facial attributes in video +<br/>1INRIA, WILLOW, Laboratoire d’Informatique de l’Ecole Normale Sup´erieure, +<br/>ENS/INRIA/CNRS UMR 8548 +<br/><b>University of Oxford</b></td><td>('1877079', 'Neva Cherniavsky', 'neva cherniavsky')<br/>('1785596', 'Ivan Laptev', 'ivan laptev')<br/>('1782755', 'Josef Sivic', 'josef sivic')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>6193c833ad25ac27abbde1a31c1cabe56ce1515b</td><td>Trojaning Attack on Neural Networks +<br/><b>Purdue University, 2Nanjing University</b></td><td>('3347155', 'Yingqi Liu', 'yingqi liu')<br/>('2026855', 'Shiqing Ma', 'shiqing ma')<br/>('3216258', 'Yousra Aafer', 'yousra aafer')<br/>('2547748', 'Wen-Chuan Lee', 'wen-chuan lee')<br/>('3293342', 'Juan Zhai', 'juan zhai')<br/>('3155328', 'Weihang Wang', 'weihang wang')<br/>('1771551', 'Xiangyu Zhang', 'xiangyu zhang')</td><td>liu1751@purdue.edu, ma229@purdue.edu, yaafer@purdue.edu, lee1938@purdue.edu, zhaijuan@nju.edu.cn, +<br/>wang1315@cs.purdue.edu, xyzhang@cs.purdue.edu +</td></tr><tr><td>614a7c42aae8946c7ad4c36b53290860f6256441</td><td>1 +<br/>Joint Face Detection and Alignment using +<br/>Multi-task Cascaded Convolutional Networks +</td><td>('3393556', 'Kaipeng Zhang', 'kaipeng zhang')<br/>('3152448', 'Zhanpeng Zhang', 'zhanpeng zhang')<br/>('32787758', 'Zhifeng Li', 'zhifeng li')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td></td></tr><tr><td>614079f1a0d0938f9c30a1585f617fa278816d53</td><td>Automatic Detection of ADHD and ASD from Expressive Behaviour in +<br/>RGBD Data +<br/><b>School of Computer Science, The University of Nottingham</b><br/>2Nottingham City Asperger Service & ADHD Clinic +<br/><b>Institute of Mental Health, The University of Nottingham</b></td><td>('2736086', 'Shashank Jaiswal', 'shashank jaiswal')<br/>('1795528', 'Michel F. Valstar', 'michel f. valstar')<br/>('38690723', 'Alinda Gillott', 'alinda gillott')<br/>('2491166', 'David Daley', 'david daley')</td><td></td></tr><tr><td>0d746111135c2e7f91443869003d05cde3044beb</td><td>PARTIAL FACE DETECTION FOR CONTINUOUS AUTHENTICATION +<br/>(cid:63)Department of Electrical and Computer Engineering and the Center for Automation Research, +<br/><b>Rutgers, The State University of New Jersey, 723 CoRE, 94 Brett Rd, Piscataway, NJ</b><br/><b>UMIACS, University of Maryland, College Park, MD</b><br/>§Google Inc., 1600 Amphitheatre Parkway, Mountain View, CA 94043 +</td><td>('3152615', 'Upal Mahbub', 'upal mahbub')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('2406413', 'Brandon Barbello', 'brandon barbello')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>umahbub@umiacs.umd.edu, vishal.m.patel@rutgers.edu, +<br/>dchandra@google.com, bbarbello@google.com, rama@umiacs.umd.edu +</td></tr><tr><td>0da75b0d341c8f945fae1da6c77b6ec345f47f2a</td><td>121 +<br/>The Effect of Computer-Generated Descriptions on Photo- +<br/>Sharing Experiences of People With Visual Impairments +<br/><b>YUHANG ZHAO, Information Science, Cornell Tech, Cornell University</b><br/>SHAOMEI WU, Facebook Inc. +<br/>LINDSAY REYNOLDS, Facebook Inc. +<br/><b>SHIRI AZENKOT, Information Science, Cornell Tech, Cornell University</b><br/>Like sighted people, visually impaired people want to share photographs on social networking services, but +<br/>find it difficult to identify and select photos from their albums. We aimed to address this problem by +<br/>incorporating state-of-the-art computer-generated descriptions into Facebook’s photo-sharing feature. We +<br/>interviewed 12 visually impaired participants to understand their photo-sharing experiences and designed +<br/>a photo description feature for the Facebook mobile application. We evaluated this feature with six +<br/>participants in a seven-day diary study. We found that participants used the descriptions to recall and +<br/>organize their photos, but they hesitated to upload photos without a sighted person’s input. In addition to +<br/>basic information about photo content, participants wanted to know more details about salient objects and +<br/>people, and whether the photos reflected their personal aesthetic. We discuss these findings from the lens +<br/>of self-disclosure and self-presentation theories and propose new computer vision research directions that +<br/>will better support visual content sharing by visually impaired people. +<br/>CCS Concepts: • Information interfaces and presentations → Multimedia and information systems; • +<br/>Computer and society → Social issues +<br/>impairments; computer-generated descriptions; SNSs; photo sharing; self-disclosure; self- +<br/>KEYWORDS +<br/>Visual +<br/>presentation +<br/>ACM Reference format: +<br/>2017. The Effect of Computer-Generated Descriptions On Photo-Sharing Experiences of People With Visual +<br/>Impairments. Proc. ACM Hum.-Comput. Interact. 1, 1. 121 (January 2017), 24 pages. +<br/>DOI: 10.1145/3134756 +<br/>1 INTRODUCTION +<br/>Sharing memories and experiences via photos is a common way to engage with others on social +<br/>networking services (SNSs) [39,46,51]. For instance, Facebook users uploaded more than 350 +<br/>million photos a day [24] and Twitter, which initially supported only text in tweets, now has +<br/>more than 28.4% of tweets containing images [39]. Visually impaired people (both blind and low +<br/>vision) have a strong presence on SNS and are interested in sharing photos [50]. They take +<br/>photos for the same reasons that sighted people do: sharing daily moments with their sighted +<br/> +<br/>Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee +<br/>provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and +<br/>the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. +</td><td></td><td></td></tr><tr><td>0d88ab0250748410a1bc990b67ab2efb370ade5d</td><td>Author(s) : +<br/>ERROR HANDLING IN MULTIMODAL BIOMETRIC SYSTEMS USING +<br/>RELIABILITY MEASURES (ThuPmOR6) +<br/>(EPFL, Switzerland) +<br/>(EPFL, Switzerland) +<br/>(EPFL, Switzerland) +<br/>(EPFL, Switzerland) +<br/>Plamen Prodanov +</td><td>('1753932', 'Krzysztof Kryszczuk', 'krzysztof kryszczuk')<br/>('1994765', 'Jonas Richiardi', 'jonas richiardi')<br/>('2439888', 'Andrzej Drygajlo', 'andrzej drygajlo')</td><td></td></tr><tr><td>0db43ed25d63d801ce745fe04ca3e8b363bf3147</td><td>Kernel Principal Component Analysis and its Applications in +<br/>Face Recognition and Active Shape Models +<br/><b>Rensselaer Polytechnic Institute, 110 Eighth Street, Troy, NY 12180 USA</b></td><td>('4019552', 'Quan Wang', 'quan wang')</td><td>wangq10@rpi.edu +</td></tr><tr><td>0daf696253a1b42d2c9d23f1008b32c65a9e4c1e</td><td>Unsupervised Discovery of Facial Events +<br/>CMU-RI-TR-10-10 +<br/>May 2010 +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, Pennsylvania 15213 +<br/><b>c(cid:13) Carnegie Mellon University</b></td><td>('1757386', 'Feng Zhou', 'feng zhou')</td><td></td></tr><tr><td>0d538084f664b4b7c0e11899d08da31aead87c32</td><td>Deformable Part Descriptors for +<br/>Fine-grained Recognition and Attribute Prediction +<br/>Forrest Iandola1 +<br/><b>ICSI / UC Berkeley 2Brigham Young University</b></td><td>('40565777', 'Ning Zhang', 'ning zhang')<br/>('2071606', 'Ryan Farrell', 'ryan farrell')<br/>('1753210', 'Trevor Darrell', 'trevor darrell')</td><td>1{nzhang,forresti,trevor}@eecs.berkeley.edu +<br/>2farrell@cs.byu.edu +</td></tr><tr><td>0dccc881cb9b474186a01fd60eb3a3e061fa6546</td><td>Effective Face Frontalization in Unconstrained Images +<br/><b>The open University of Israel. 2Adience</b><br/>Figure 1: Frontalized faces. Top: Input photos; bottom: our frontalizations, +<br/>obtained without estimating 3D facial shapes. +<br/>“Frontalization” is the process of synthesizing frontal facing views of faces +<br/>appearing in single unconstrained photos. Recent reports have suggested +<br/>that this process may substantially boost the performance of face recogni- +<br/>tion systems. This, by transforming the challenging problem of recognizing +<br/>faces viewed from unconstrained viewpoints to the easier problem of rec- +<br/>ognizing faces in constrained, forward facing poses. Previous frontalization +<br/>methods did this by attempting to approximate 3D facial shapes for each +<br/>query image. We observe that 3D face shape estimation from unconstrained +<br/>photos may be a harder problem than frontalization and can potentially in- +<br/>troduce facial misalignments. Instead, we explore the simpler approach of +<br/>using a single, unmodified, 3D surface as an approximation to the shape of +<br/>all input faces. We show that this leads to a straightforward, efficient and +<br/>easy to implement method for frontalization. More importantly, it produces +<br/>aesthetic new frontal views and is surprisingly effective when used for face +<br/>recognition and gender estimation. +<br/>Observation 1: For frontalization, one rough estimate of the 3D facial shape +<br/>seems as good as another, demonstrated by the following example: +<br/>Figure 2: Frontalization process. (a) facial features detected on a query +<br/>face and on a reference face (b) which was produced by rendering a tex- +<br/>tured 3D, CG model (c); (d) 2D query coordinates and corresponding 3D +<br/>coordinates on the model provide an estimated projection matrix, used to +<br/>back-project query texture to the reference coordinate system; (e) estimated +<br/>self-occlusions shown overlaid on the frontalized result (warmer colors re- +<br/>flect more occlusions.) Facial appearances in these regions are borrowed +<br/>from corresponding symmetric face regions; (f) our final frontalized result. +<br/>The top row shows surfaces estimated for the same query (left) by Hass- +<br/>ner [2] (mid) and DeepFaces [6] (right). Frontalizations are shown at the +<br/>bottom using our single-3D approach (left), Hassner (mid) and DeepFaces +<br/>(right). Clearly, both surfaces are rough approximations to the facial shape. +<br/>Moreover, despite the different surfaces, all results seem qualitatively simi- +<br/>lar, calling to question the need for shape estimation for frontalization. +<br/>Result 1: A novel frontalization method using a single, unmodified 3D ref- +<br/>erence shape is described in the paper (illustrated in Fig. 2). +<br/>Observation 2: A single, unmodified 3D reference shape produces aggres- +<br/>sively aligned faces, as can be observed in Fig. 3. +<br/>Result 2: Frontalized, strongly aligned faces elevate LFW [5] verification +<br/>accuracy and gender estimation rates on the Adience benchmark [1]. +<br/>Conclusion: On the role of 2D appearance vs. 3D shape in face recognition, +<br/>our results suggest that 3D shape estimation may be unnecessary. +</td><td>('1756099', 'Tal Hassner', 'tal hassner')<br/>('35840854', 'Shai Harel', 'shai harel')<br/>('1753918', 'Eran Paz', 'eran paz')<br/>('1792038', 'Roee Enbar', 'roee enbar')</td><td></td></tr><tr><td>0d467adaf936b112f570970c5210bdb3c626a717</td><td></td><td></td><td></td></tr><tr><td>0d6b28691e1aa2a17ffaa98b9b38ac3140fb3306</td><td>Review of Perceptual Resemblance of Local +<br/>Plastic Surgery Facial Images using Near Sets +<br/>1,2 Department of Computer Technology, +<br/>YCCE Nagpur, India +</td><td>('9083090', 'Prachi V. Wagde', 'prachi v. wagde')<br/>('9218400', 'Roshni Khedgaonkar', 'roshni khedgaonkar')</td><td></td></tr><tr><td>0de91641f37b0a81a892e4c914b46d05d33fd36e</td><td>RAPS: Robust and Efficient Automatic Construction of Person-Specific +<br/>Deformable Models +<br/>∗Department of Computing, +<br/><b>Imperial College London</b><br/>180 Queens Gate, +<br/>†EEMCS, +<br/><b>University of Twente</b><br/>Drienerlolaan 5, +<br/>London SW7 2AZ, U.K. +<br/>7522 NB Enschede, The Netherlands +</td><td>('3320415', 'Christos Sagonas', 'christos sagonas')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{c.sagonas, i.panagakis, s.zafeiriou, m.pantic}@imperial.ac.uk +</td></tr><tr><td>0df0d1adea39a5bef318b74faa37de7f3e00b452</td><td>Appearance-Based Gaze Estimation in the Wild +<br/>1Perceptual User Interfaces Group, 2Scalable Learning and Perception Group +<br/><b>Max Planck Institute for Informatics, Saarbr ucken, Germany</b></td><td>('2520795', 'Xucong Zhang', 'xucong zhang')<br/>('1751242', 'Yusuke Sugano', 'yusuke sugano')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')</td><td>{xczhang,sugano,mfritz,bulling}@mpi-inf.mpg.de +</td></tr><tr><td>0d3bb75852098b25d90f31d2f48fd0cb4944702b</td><td>A DATA-DRIVEN APPROACH TO CLEANING LARGE FACE DATASETS +<br/><b>Advanced Digital Sciences Center (ADSC), University of Illinois at Urbana-Champaign, Singapore</b></td><td>('1702224', 'Stefan Winkler', 'stefan winkler')</td><td></td></tr><tr><td>0db8e6eb861ed9a70305c1839eaef34f2c85bbaf</td><td></td><td></td><td></td></tr><tr><td>0d0b880e2b531c45ee8227166a489bf35a528cb9</td><td>Structure Preserving Object Tracking +<br/><b>Computer Vision Lab, Delft University of Technology</b><br/>Mekelweg 4, 2628 CD Delft, The Netherlands +</td><td>('2883723', 'Lu Zhang', 'lu zhang')<br/>('1803520', 'Laurens van der Maaten', 'laurens van der maaten')</td><td>{lu.zhang, l.j.p.vandermaaten}@tudelft.nl +</td></tr><tr><td>0d3882b22da23497e5de8b7750b71f3a4b0aac6b</td><td>Research Article +<br/>Context Is Routinely Encoded +<br/>During Emotion Perception +<br/>21(4) 595 –599 +<br/>© The Author(s) 2010 +<br/>Reprints and permission: +<br/>sagepub.com/journalsPermissions.nav +<br/>DOI: 10.1177/0956797610363547 +<br/>http://pss.sagepub.com +<br/><b>Boston College; 2Psychiatric Neuroimaging Program, Massachusetts General Hospital, Harvard Medical School; and 3Athinoula A. Martinos</b><br/>Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School +</td><td>('1731779', 'Lisa Feldman Barrett', 'lisa feldman barrett')</td><td></td></tr><tr><td>0dbf4232fcbd52eb4599dc0760b18fcc1e9546e9</td><td></td><td></td><td></td></tr><tr><td>0d760e7d762fa449737ad51431f3ff938d6803fe</td><td>LCDet: Low-Complexity Fully-Convolutional Neural Networks for +<br/>Object Detection in Embedded Systems +<br/>UC San Diego ∗ +<br/>Gokce Dane +<br/>Qualcomm Inc. +<br/>UC San Diego +<br/>Qualcomm Inc. +<br/>UC San Diego +</td><td>('2906509', 'Subarna Tripathi', 'subarna tripathi')<br/>('1801046', 'Byeongkeun Kang', 'byeongkeun kang')<br/>('3484765', 'Vasudev Bhaskaran', 'vasudev bhaskaran')<br/>('30518518', 'Truong Nguyen', 'truong nguyen')</td><td>stripathi@ucsd.edu +<br/>gokced@qti.qualcomm.com +<br/>bkkang@ucsd.edu +<br/>vasudevb@qti.qualcomm.com +<br/>tqn001@eng.ucsd.edu +</td></tr><tr><td>0d3068b352c3733c9e1cc75e449bf7df1f7b10a4</td><td>Context based Facial Expression Analysis in the +<br/>Wild +<br/><b>School of Computer Science, CECS, Australian National University, Australia</b><br/>http://users.cecs.anu.edu.au/∼adhall +</td><td>('1735697', 'Abhinav Dhall', 'abhinav dhall')</td><td>abhinav.dhall@anu.edu.au +</td></tr><tr><td>0dd72887465046b0f8fc655793c6eaaac9c03a3d</td><td>Real-time Head Orientation from a Monocular +<br/>Camera using Deep Neural Network +<br/>KAIST, Republic of Korea +</td><td>('3250619', 'Byungtae Ahn', 'byungtae ahn')<br/>('2870153', 'Jaesik Park', 'jaesik park')</td><td>[btahn,jspark]@rcv.kaist.ac.kr, iskweon77@kaist.ac.kr +</td></tr><tr><td>0d087aaa6e2753099789cd9943495fbbd08437c0</td><td></td><td></td><td></td></tr><tr><td>0d8415a56660d3969449e77095be46ef0254a448</td><td></td><td></td><td></td></tr><tr><td>0dfa460a35f7cab4705726b6367557b9f7842c65</td><td>Modeling Spatial-Temporal Clues in a Hybrid Deep +<br/>Learning Framework for Video Classification +<br/>School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, +<br/><b>Fudan University, Shanghai, China</b></td><td>('3099139', 'Zuxuan Wu', 'zuxuan wu')<br/>('31825486', 'Xi Wang', 'xi wang')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('1743864', 'Hao Ye', 'hao ye')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')</td><td>{zxwu, xwang10, ygj, haoye10, xyxue}@fudan.edu.cn +</td></tr><tr><td>0d14261e69a4ad4140ce17c1d1cea76af6546056</td><td>Adding Facial Actions into 3D Model Search to Analyse +<br/>Behaviour in an Unconstrained Environment +<br/><b>Imaging Science and Biomedical Engineering, The University of Manchester, UK</b></td><td>('1753123', 'Angela Caunce', 'angela caunce')</td><td></td></tr><tr><td>0dbacb4fd069462841ebb26e1454b4d147cd8e98</td><td>Recent Advances in Discriminant Non-negative +<br/>Matrix Factorization +<br/><b>Aristotle University of Thessaloniki</b><br/>Thessaloniki, Greece, 54124 +</td><td>('1793625', 'Symeon Nikitidis', 'symeon nikitidis')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>Email: {nikitidis,tefas,pitas}@aiia.csd.auth.gr +</td></tr><tr><td>0db36bf08140d53807595b6313201a7339470cfe</td><td>Moving Vistas: Exploiting Motion for Describing Scenes +<br/>Department of Electrical and Computer Engineering +<br/><b>Center for Automation Research, UMIACS, University of Maryland, College Park, MD</b></td><td>('34711525', 'Nitesh Shroff', 'nitesh shroff')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{nshroff,pturaga,rama}@umiacs.umd.edu +</td></tr><tr><td>0d781b943bff6a3b62a79e2c8daf7f4d4d6431ad</td><td>EmotiW 2016: Video and Group-Level Emotion +<br/>Recognition Challenges +<br/>Roland Goecke +<br/>David R. Cheriton School of +<br/>Human-Centred Technology +<br/>David R. Cheriton School of +<br/>Computer Science +<br/><b>University of Waterloo</b><br/>Canada +<br/><b>University of Canberra</b><br/>Centre +<br/>Australia +<br/>Computer Science +<br/><b>University of Waterloo</b><br/>Canada +<br/>Tom Gedeon +<br/>David R. Cheriton School of +<br/>Information Human Centred +<br/>Computer Science +<br/><b>University of Waterloo</b><br/>Canada +<br/><b>Australian National University</b><br/>Computing +<br/>Australia +</td><td>('1735697', 'Abhinav Dhall', 'abhinav dhall')<br/>('2942991', 'Jyoti Joshi', 'jyoti joshi')<br/>('1773895', 'Jesse Hoey', 'jesse hoey')</td><td>abhinav.dhall@uwaterloo.ca +<br/>roland.goecke@ieee.org +<br/>jyoti.joshi@uwaterloo.ca +<br/>jhoey@cs.uwaterloo.ca +<br/>tom.gedeon@anu.edu.au +</td></tr><tr><td>0d735e7552af0d1dcd856a8740401916e54b7eee</td><td></td><td></td><td></td></tr><tr><td>0d06b3a4132d8a2effed115a89617e0a702c957a</td><td></td><td></td><td></td></tr><tr><td>0d2dd4fc016cb6a517d8fb43a7cc3ff62964832e</td><td></td><td></td><td></td></tr><tr><td>0d33b6c8b4d1a3cb6d669b4b8c11c2a54c203d1a</td><td>Detection and Tracking of Faces in Videos: A Review +<br/>© 2016 IJEDR | Volume 4, Issue 2 | ISSN: 2321-9939 +<br/>of Related Work +<br/>1Student, 2Assistant Professor +<br/>1, 2Dept. of Electronics & Comm., S S I E T, Punjab, India +<br/>________________________________________________________________________________________________________ +</td><td>('48816689', 'Seema Saini', 'seema saini')</td><td></td></tr><tr><td>0d1d9a603b08649264f6e3b6d5a66bf1e1ac39d2</td><td><b>University of Nebraska - Lincoln</b><br/>US Army Research +<br/>2015 +<br/>U.S. Department of Defense +<br/>Effects of emotional expressions on persuasion +<br/><b>University of Southern California</b><br/><b>University of Southern California</b><br/><b>University of Southern California</b><br/><b>University of Southern California</b><br/>Follow this and additional works at: http://digitalcommons.unl.edu/usarmyresearch +<br/>Wang, Yuqiong; Lucas, Gale; Khooshabeh, Peter; de Melo, Celso; and Gratch, Jonathan, "Effects of emotional expressions on +<br/>persuasion" (2015). US Army Research. Paper 340. +<br/>http://digitalcommons.unl.edu/usarmyresearch/340 +</td><td>('2522587', 'Yuqiong Wang', 'yuqiong wang')<br/>('2419453', 'Gale Lucas', 'gale lucas')<br/>('2635945', 'Peter Khooshabeh', 'peter khooshabeh')<br/>('1977901', 'Celso de Melo', 'celso de melo')<br/>('1730824', 'Jonathan Gratch', 'jonathan gratch')</td><td>DigitalCommons@University of Nebraska - Lincoln +<br/>University of Southern California, wangyuqiong@ymail.com +<br/>This Article is brought to you for free and open access by the U.S. Department of Defense at DigitalCommons@University of Nebraska - Lincoln. It has +<br/>been accepted for inclusion in US Army Research by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. +</td></tr><tr><td>0da4c3d898ca2fff9e549d18f513f4898e960aca</td><td>Wang, Y., Thomas, J., Weissgerber, S. C., Kazemini, S., Ul-Haq, I., & +<br/>Quadflieg, S. (2015). The Headscarf Effect Revisited: Further Evidence for a +<br/>336. 10.1068/p7940 +<br/>Peer reviewed version +<br/>Link to published version (if available): +<br/>10.1068/p7940 +<br/>Link to publication record in Explore Bristol Research +<br/>PDF-document +<br/><b>University of Bristol - Explore Bristol Research</b><br/>General rights +<br/>This document is made available in accordance with publisher policies. Please cite only the published +<br/>version using the reference above. Full terms of use are available: +<br/>http://www.bristol.ac.uk/pure/about/ebr-terms.html +<br/>Take down policy +<br/>Explore Bristol Research is a digital archive and the intention is that deposited content should not be +<br/>removed. However, if you believe that this version of the work breaches copyright law please contact +<br/>• Your contact details +<br/><b>Bibliographic details for the item, including a URL</b><br/>• An outline of the nature of the complaint +<br/>On receipt of your message the Open Access Team will immediately investigate your claim, make an +<br/>initial judgement of the validity of the claim and, where appropriate, withdraw the item in question +<br/>from public view. +<br/> </td><td></td><td>open-access@bristol.ac.uk and include the following information in your message: +</td></tr><tr><td>951368a1a8b3c5cd286726050b8bdf75a80f7c37</td><td>A Family of Online Boosting Algorithms +<br/><b>University of California, San Diego</b><br/><b>University of California, Merced</b><br/><b>University of California, San Diego</b></td><td>('2490700', 'Boris Babenko', 'boris babenko')<br/>('37144787', 'Ming-Hsuan Yang', 'ming-hsuan yang')<br/>('1769406', 'Serge Belongie', 'serge belongie')</td><td>bbabenko@cs.ucsd.edu +<br/>mhyang@ucmerced.edu +<br/>sjb@cs.ucsd.edu +</td></tr><tr><td>956e9b69b3366ed3e1670609b53ba4a7088b8b7e</td><td>Semi-supervised dimensionality reduction for image retrieval +<br/><b>aIBM China Research Lab, Beijing, China</b><br/><b>bTsinghua University, Beijing, China</b></td><td></td><td></td></tr><tr><td>956317de62bd3024d4ea5a62effe8d6623a64e53</td><td>Lighting Analysis and Texture Modification of 3D Human +<br/>Face Scans +<br/>Author +<br/>Zhang, Paul, Zhao, Sanqiang, Gao, Yongsheng +<br/>Published +<br/>2007 +<br/>Conference Title +<br/>Digital Image Computing Techniques and Applications +<br/>DOI +<br/>https://doi.org/10.1109/DICTA.2007.4426825 +<br/>Copyright Statement +<br/>© 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/ +<br/>republish this material for advertising or promotional purposes or for creating new collective +<br/>works for resale or redistribution to servers or lists, or to reuse any copyrighted component of +<br/>this work in other works must be obtained from the IEEE. +<br/>Downloaded from +<br/>http://hdl.handle.net/10072/17889 +<br/>Link to published version +<br/>http://www.ieee.org/ +<br/>Griffith Research Online +<br/>https://research-repository.griffith.edu.au +</td><td></td><td></td></tr><tr><td>959bcb16afdf303c34a8bfc11e9fcc9d40d76b1c</td><td>Temporal Coherency based Criteria for Predicting +<br/>Video Frames using Deep Multi-stage Generative +<br/>Adversarial Networks +<br/>Visualization and Perception Laboratory +<br/>Department of Computer Science and Engineering +<br/><b>Indian Institute of Technology Madras, Chennai, India</b></td><td>('29901316', 'Prateep Bhattacharjee', 'prateep bhattacharjee')<br/>('1680398', 'Sukhendu Das', 'sukhendu das')</td><td>1prateepb@cse.iitm.ac.in, 2sdas@iitm.ac.in +</td></tr><tr><td>951f21a5671a4cd14b1ef1728dfe305bda72366f</td><td>International Journal of Science and Research (IJSR) +<br/>ISSN (Online): 2319-7064 +<br/>Impact Factor (2012): 3.358 +<br/>Use of ℓ2/3-norm Sparse Representation for Facial +<br/>Expression Recognition +<br/><b>MATS University, MATS School of Engineering and Technology, Arang, Raipur, India</b><br/><b>MATS University, MATS School of Engineering and Technology, Arang, Raipur, India</b><br/>in +<br/>three +<br/>to discriminate +<br/>it +<br/>from +<br/>represents emotion, +</td><td></td><td></td></tr><tr><td>95f26d1c80217706c00b6b4b605a448032b93b75</td><td>New Robust Face Recognition Methods Based on Linear +<br/>Regression +<br/><b>Bio-Computing Research Center, Shenzhen Graduate School, Harbin Institute of Technology, Shenzhen, Guangdong Province, China, 2 Key Laboratory of Network</b><br/>Oriented Intelligent Computation, Shenzhen, Guangdong Province, China +</td><td>('2208128', 'Jian-Xun Mi', 'jian-xun mi')<br/>('2650895', 'Jin-Xing Liu', 'jin-xing liu')<br/>('40342210', 'Jiajun Wen', 'jiajun wen')</td><td></td></tr><tr><td>95f12d27c3b4914e0668a268360948bce92f7db3</td><td>Interactive Facial Feature Localization +<br/><b>University of Illinois at Urbana Champaign, Urbana, IL 61801, USA</b><br/>2 Adobe Systems Inc., San Jose, CA 95110, USA +<br/>3 Facebook Inc., Menlo Park, CA 94025, USA +</td><td>('36474335', 'Vuong Le', 'vuong le')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td></td></tr><tr><td>9547a7bce2b85ef159b2d7c1b73dea82827a449f</td><td>Facial Expression Recognition Using Gabor Motion Energy Filters +<br/>Dept. Computer Science Engineering +<br/>UC San Diego +<br/>Marian S. Bartlett +<br/><b>Institute for Neural Computation</b><br/>UC San Diego +</td><td>('4072965', 'Tingfan Wu', 'tingfan wu')<br/>('1741200', 'Javier R. Movellan', 'javier r. movellan')</td><td>tingfan@gmail.com +<br/>{marni,movellan}@mplab.ucsd.edu +</td></tr><tr><td>9513503867b29b10223f17c86e47034371b6eb4f</td><td>Comparison of optimisation algorithms for +<br/>deformable template matching +<br/><b>Link oping University, Computer Vision Laboratory</b><br/>ISY, SE-581 83 Link¨oping, SWEDEN +</td><td>('1797883', 'Vasileios Zografos', 'vasileios zografos')</td><td>zografos@isy.liu.se ⋆ +</td></tr><tr><td>955e2a39f51c0b6f967199942d77625009e580f9</td><td>NAMING FACES ON THE WEB +<br/>a thesis +<br/>submitted to the department of computer engineering +<br/><b>and the institute of engineering and science</b><br/><b>of bilkent university</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>master of science +<br/>By +<br/>July, 2010 +</td><td>('34946851', 'Hilal Zitouni', 'hilal zitouni')</td><td></td></tr><tr><td>956c634343e49319a5e3cba4f2bd2360bdcbc075</td><td>IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 36, NO. 4, AUGUST 2006 +<br/>873 +<br/>A Novel Incremental Principal Component Analysis +<br/>and Its Application for Face Recognition +</td><td>('1776124', 'Haitao Zhao', 'haitao zhao')<br/>('1768574', 'Pong Chi Yuen', 'pong chi yuen')</td><td></td></tr><tr><td>95ea564bd983129ddb5535a6741e72bb1162c779</td><td>Multi-Task Learning by Deep Collaboration and +<br/>Application in Facial Landmark Detection +<br/><b>Laval University, Qu bec, Canada</b></td><td>('2758280', 'Ludovic Trottier', 'ludovic trottier')<br/>('2310695', 'Philippe Giguère', 'philippe giguère')<br/>('1700926', 'Brahim Chaib-draa', 'brahim chaib-draa')</td><td>ludovic.trottier.1@ulaval.ca +<br/>{philippe.giguere,brahim.chaib-draa}@ift.ulaval.ca +</td></tr><tr><td>958c599a6f01678513849637bec5dc5dba592394</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Generalized Zero-Shot Learning for Action +<br/>Recognition with Web-Scale Video Data +<br/>Received: date / Accepted: date +</td><td>('2473509', 'Kun Liu', 'kun liu')<br/>('8984539', 'Wenbing Huang', 'wenbing huang')</td><td></td></tr><tr><td>950171acb24bb24a871ba0d02d580c09829de372</td><td>Speeding up 2D-Warping for Pose-Invariant Face Recognition +<br/><b>Human Language Technology and Pattern Recognition Group, RWTH Aachen University, Germany</b></td><td>('1804963', 'Harald Hanselmann', 'harald hanselmann')<br/>('1685956', 'Hermann Ney', 'hermann ney')</td><td>surname@cs.rwth-aachen.de +</td></tr><tr><td>59be98f54bb4ed7a2984dc6a3c84b52d1caf44eb</td><td>A Deep-Learning Approach to Facial Expression Recognition +<br/>with Candid Images +<br/><b>CUNY City College</b><br/>Alibaba. Inc +<br/><b>IBM China Research Lab</b><br/><b>CUNY Graduate Center and City College</b></td><td>('40617554', 'Wei Li', 'wei li')<br/>('1713016', 'Min Li', 'min li')<br/>('1703625', 'Zhong Su', 'zhong su')<br/>('4697712', 'Zhigang Zhu', 'zhigang zhu')</td><td>lwei000@citymail.cuny.edu +<br/>mushi.lm@alibaba.inc +<br/>suzhong@cn.ibm.com +<br/>zhu@cs.ccny.cuny.edu +</td></tr><tr><td>59fc69b3bc4759eef1347161e1248e886702f8f7</td><td>Final Report of Final Year Project +<br/>HKU-Face: A Large Scale Dataset for +<br/>Deep Face Recognition +<br/>3035141841 +<br/>COMP4801 Final Year Project +<br/>Project Code: 17007 +</td><td>('40456402', 'Haoyu Li', 'haoyu li')</td><td></td></tr><tr><td>591a737c158be7b131121d87d9d81b471c400dba</td><td>Affect Valence Inference From Facial Action Unit Spectrograms +<br/>MIT Media Lab +<br/>MA 02139, USA +<br/>MIT Media Lab +<br/>MA 02139, USA +<br/><b>Harvard University</b><br/>MA 02138, USA +<br/>Rosalind Picard +<br/>MIT Media Lab +<br/>MA 02139, USA +</td><td>('1801452', 'Daniel McDuff', 'daniel mcduff')<br/>('1754451', 'Rana El Kaliouby', 'rana el kaliouby')<br/>('2010950', 'Karim Kassam', 'karim kassam')</td><td>djmcduff@mit.edu +<br/>kaliouby@mit.edu +<br/>kskassam@fas.harvard.edu +<br/>picard@mit.edu +</td></tr><tr><td>59bfeac0635d3f1f4891106ae0262b81841b06e4</td><td>Face Verification Using the LARK Face +<br/>Representation +</td><td>('3326805', 'Hae Jong Seo', 'hae jong seo')<br/>('1718280', 'Peyman Milanfar', 'peyman milanfar')</td><td></td></tr><tr><td>59efb1ac77c59abc8613830787d767100387c680</td><td>DIF : Dataset of Intoxicated Faces for Drunk Person +<br/>Identification +<br/><b>Indian Institute of Technology Ropar</b><br/><b>Indian Institute of Technology Ropar</b></td><td>('46241736', 'Devendra Pratap Yadav', 'devendra pratap yadav')<br/>('1735697', 'Abhinav Dhall', 'abhinav dhall')</td><td>2014csb1010@iitrpr.ac.in +<br/>abhinav@iitrpr.ac.in +</td></tr><tr><td>590628a9584e500f3e7f349ba7e2046c8c273fcf</td><td></td><td></td><td></td></tr><tr><td>593234ba1d2e16a887207bf65d6b55bbc7ea2247</td><td>Combining Language Sources and Robust +<br/>Semantic Relatedness for Attribute-Based +<br/>Knowledge Transfer +<br/>1 Department of Computer Science, TU Darmstadt +<br/><b>Max Planck Institute for Informatics, Saarbr ucken, Germany</b></td><td>('34849128', 'Marcus Rohrbach', 'marcus rohrbach')<br/>('37718254', 'Michael Stark', 'michael stark')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td></td></tr><tr><td>59eefa01c067a33a0b9bad31c882e2710748ea24</td><td>IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY +<br/>Fast Landmark Localization +<br/>with 3D Component Reconstruction and CNN for +<br/>Cross-Pose Recognition +</td><td>('24020847', 'Hung-Cheng Shie', 'hung-cheng shie')<br/>('9640380', 'Cheng-Hua Hsieh', 'cheng-hua hsieh')</td><td></td></tr><tr><td>59e2037f5079794cb9128c7f0900a568ced14c2a</td><td>Clothing and People - A Social Signal Processing Perspective +<br/><b>Faculty of Mathematics and Computer Science, University of Barcelona, Barcelona, Spain</b><br/>2 Computer Vision Center, Barcelona, Spain +<br/><b>University of Verona, Verona, Italy</b></td><td>('2084534', 'Maedeh Aghaei', 'maedeh aghaei')<br/>('10724083', 'Federico Parezzan', 'federico parezzan')<br/>('2837527', 'Mariella Dimiccoli', 'mariella dimiccoli')<br/>('1724155', 'Petia Radeva', 'petia radeva')<br/>('1723008', 'Marco Cristani', 'marco cristani')</td><td></td></tr><tr><td>59dac8b460a89e03fa616749a08e6149708dcc3a</td><td>A Convergent Solution to Matrix Bidirectional Projection Based Feature +<br/>Extraction with Application to Face Recognition ∗ +<br/><b>School of Computer, National University of Defense Technology</b><br/>No 137, Yanwachi Street, Kaifu District, +<br/>Changsha, Hunan Province, 410073, P.R. China +</td><td>('3144121', 'Yubin Zhan', 'yubin zhan')<br/>('1969736', 'Jianping Yin', 'jianping yin')<br/>('33793976', 'Xinwang Liu', 'xinwang liu')</td><td>E-mail: {YubinZhan,JPYin,XWLiu}@nudt.edu.cn +</td></tr><tr><td>59e9934720baf3c5df3a0e1e988202856e1f83ce</td><td>UA-DETRAC: A New Benchmark and Protocol for +<br/>Multi-Object Detection and Tracking +<br/><b>University at Albany, SUNY</b><br/>2 School of Computer and Control Engineering, UCAS +<br/>3 Department of Electrical and Computer Engineering, UCSD +<br/>4 National Laboratory of Pattern Recognition, CASIA +<br/><b>University at Albany, SUNY</b><br/><b>Division of Computer Science and Engineering, Hanyang University</b><br/>7 Electrical Engineering and Computer Science, UCM +</td><td>('39774417', 'Longyin Wen', 'longyin wen')<br/>('1910738', 'Dawei Du', 'dawei du')<br/>('1773408', 'Zhaowei Cai', 'zhaowei cai')<br/>('39643145', 'Ming-Ching Chang', 'ming-ching chang')<br/>('3245785', 'Honggang Qi', 'honggang qi')<br/>('33047058', 'Jongwoo Lim', 'jongwoo lim')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>59d225486161b43b7bf6919b4a4b4113eb50f039</td><td>Complex Event Recognition from Images with Few Training Examples +<br/>Irfan Essa∗ +<br/><b>Georgia Institute of Technology</b><br/><b>University of Southern California</b></td><td>('2308598', 'Unaiza Ahsan', 'unaiza ahsan')<br/>('1726241', 'Chen Sun', 'chen sun')<br/>('1945508', 'James Hays', 'james hays')</td><td>uahsan3@gatech.edu +<br/>chensun@google.com +<br/>hays@gatech.edu +<br/>irfan@cc.gatech.edu +</td></tr><tr><td>5945464d47549e8dcaec37ad41471aa70001907f</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Every Moment Counts: Dense Detailed Labeling of Actions in Complex +<br/>Videos +<br/>Received: date / Accepted: date +</td><td>('34149749', 'Serena Yeung', 'serena yeung')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td></td></tr><tr><td>59c9d416f7b3d33141cc94567925a447d0662d80</td><td>Universität des Saarlandes +<br/>Max-Planck-Institut für Informatik +<br/>AG5 +<br/>Matrix factorization over max-times +<br/>algebra for data mining +<br/>Masterarbeit im Fach Informatik +<br/>Master’s Thesis in Computer Science +<br/>von / by +<br/>angefertigt unter der Leitung von / supervised by +<br/>begutachtet von / reviewers +<br/>November 2013 +<br/>UNIVERSITASSARAVIENSIS </td><td>('2297723', 'Sanjar Karaev', 'sanjar karaev')<br/>('1804891', 'Pauli Miettinen', 'pauli miettinen')<br/>('1804891', 'Pauli Miettinen', 'pauli miettinen')<br/>('1751591', 'Gerhard Weikum', 'gerhard weikum')</td><td></td></tr><tr><td>59bece468ed98397d54865715f40af30221aa08c</td><td>Deformable Part-based Robust Face Detection +<br/>under Occlusion by Using Face Decomposition +<br/>into Face Components +<br/>Darijan Marčetić, Slobodan Ribarić +<br/><b>University of Zagreb, Faculty of Electrical Engineering and Computing, Croatia</b></td><td></td><td>{darijan.marcetic, slobodan.ribaric}@fer.hr +</td></tr><tr><td>59a35b63cf845ebf0ba31c290423e24eb822d245</td><td>The FaceSketchID System: Matching Facial +<br/>Composites to Mugshots +<br/>tedious, and may not +</td><td>('34393045', 'Hu Han', 'hu han')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>59f325e63f21b95d2b4e2700c461f0136aecc171</td><td>3070 +<br/>978-1-4577-1302-6/11/$26.00 ©2011 IEEE +<br/>FOR FACE RECOGNITION +<br/>1. INTRODUCTION +</td><td></td><td></td></tr><tr><td>59420fd595ae745ad62c26ae55a754b97170b01f</td><td>Objects as Attributes for Scene Classification +<br/><b>Stanford University</b></td><td>('33642044', 'Li-Jia Li', 'li-jia li')<br/>('2888806', 'Hao Su', 'hao su')<br/>('7892285', 'Yongwhan Lim', 'yongwhan lim')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td></td></tr><tr><td>599adc0dcd4ebcc2a868feedd243b5c3c1bd1d0a</td><td>How Robust is 3D Human Pose Estimation to Occlusion? +<br/><b>Visual Computing Institute, RWTH Aachen University</b><br/>2Robert Bosch GmbH, Corporate Research +</td><td>('2699877', 'Timm Linder', 'timm linder')<br/>('1789756', 'Bastian Leibe', 'bastian leibe')</td><td>{sarandi,leibe}@vision.rwth-aachen.de +<br/>{timm.linder,kaioliver.arras}@de.bosch.com +</td></tr><tr><td>5922e26c9eaaee92d1d70eae36275bb226ecdb2e</td><td>Boosting Classification Based Similarity +<br/>Learning by using Standard Distances +<br/>Departament d’Informàtica, Universitat de València +<br/>Av. de la Universitat s/n. 46100-Burjassot (Spain) +</td><td>('2275648', 'Emilia López-Iñesta', 'emilia lópez-iñesta')<br/>('3138833', 'Miguel Arevalillo-Herráez', 'miguel arevalillo-herráez')<br/>('2627759', 'Francisco Grimaldo', 'francisco grimaldo')</td><td>eloi@alumni.uv.es,miguel.arevalillo@uv.es +<br/>francisco.grimaldo@uv.es +</td></tr><tr><td>59d8fa6fd91cdb72cd0fa74c04016d79ef5a752b</td><td>The Menpo Facial Landmark Localisation Challenge: +<br/>A step towards the solution +<br/>Department of Computing +<br/><b>Imperial College London</b></td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('2814229', 'George Trigeorgis', 'george trigeorgis')<br/>('1688922', 'Grigorios Chrysos', 'grigorios chrysos')<br/>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('1719912', 'Jie Shen', 'jie shen')</td><td>{s.zafeiriou, g.trigeorgis, g.chrysos, j.deng16, jie.shen07}@imperial.ac.uk +</td></tr><tr><td>59e75aad529b8001afc7e194e21668425119b864</td><td>Membrane Nonrigid Image Registration +<br/>Department of Computer Science +<br/><b>Drexel University</b><br/>Philadelphia, PA +</td><td>('1708819', 'Ko Nishino', 'ko nishino')</td><td></td></tr><tr><td>59d45281707b85a33d6f50c6ac6b148eedd71a25</td><td>Rank Minimization across Appearance and Shape for AAM Ensemble Fitting +<br/>2The Commonwealth Scientific and Industial Research Organization (CSIRO) +<br/><b>Queensland University of Technology</b></td><td>('2699730', 'Xin Cheng', 'xin cheng')<br/>('1729760', 'Sridha Sridharan', 'sridha sridharan')<br/>('1820249', 'Simon Lucey', 'simon lucey')</td><td>1{x2.cheng,s.sridharan}@qut.edu.au +<br/>2{jason.saragih,simon.lucey}@csiro.au +</td></tr><tr><td>59319c128c8ac3c88b4ab81088efe8ae9c458e07</td><td>Effective Computer Model For Recognizing +<br/>Nationality From Frontal Image +<br/>Bat-Erdene.B +<br/>Information and Communication Management School +<br/><b>The University of the Humanities</b><br/>Ulaanbaatar, Mongolia +</td><td></td><td>e-mail: basubaer@gmail.com +</td></tr><tr><td>59a6c9333c941faf2540979dcfcb5d503a49b91e</td><td>Sampling Clustering +<br/><b>School of Computer Science and Technology, Shandong University, China</b></td><td>('51016741', 'Ching Tarn', 'ching tarn')<br/>('2413471', 'Yinan Zhang', 'yinan zhang')<br/>('48260402', 'Ye Feng', 'ye feng')</td><td>∗i@ctarn.io +</td></tr><tr><td>59031a35b0727925f8c47c3b2194224323489d68</td><td>Sparse Variation Dictionary Learning for Face Recognition with A Single +<br/>Training Sample Per Person +<br/>ETH Zurich +<br/>Switzerland +</td><td>('5828998', 'Meng Yang', 'meng yang')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{yang,vangool}@vision.ee.ethz.ch +</td></tr><tr><td>926c67a611824bc5ba67db11db9c05626e79de96</td><td>1913 +<br/>Enhancing Bilinear Subspace Learning +<br/>by Element Rearrangement +</td><td>('38188040', 'Dong Xu', 'dong xu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('1686911', 'Stephen Lin', 'stephen lin')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td></td></tr><tr><td>923ede53b0842619831e94c7150e0fc4104e62f7</td><td>978-1-4799-9988-0/16/$31.00 ©2016 IEEE +<br/>1293 +<br/>ICASSP 2016 +</td><td></td><td></td></tr><tr><td>92b61b09d2eed4937058d0f9494d9efeddc39002</td><td>Under review in IJCV manuscript No. +<br/>(will be inserted by the editor) +<br/>BoxCars: Improving Vehicle Fine-Grained Recognition using +<br/>3D Bounding Boxes in Traffic Surveillance +<br/>Received: date / Accepted: date +</td><td>('34891870', 'Jakub Sochor', 'jakub sochor')</td><td></td></tr><tr><td>9264b390aa00521f9bd01095ba0ba4b42bf84d7e</td><td>Displacement Template with Divide-&-Conquer +<br/>Algorithm for Significantly Improving +<br/>Descriptor based Face Recognition Approaches +<br/><b>Wenzhou University, China</b><br/><b>University of Northern British Columbia, Canada</b><br/><b>Aberystwyth University, UK</b></td><td>('1692551', 'Liang Chen', 'liang chen')<br/>('33500699', 'Ling Yan', 'ling yan')<br/>('1990125', 'Yonghuai Liu', 'yonghuai liu')<br/>('39388942', 'Lixin Gao', 'lixin gao')<br/>('3779849', 'Xiaoqin Zhang', 'xiaoqin zhang')</td><td></td></tr><tr><td>92be73dffd3320fe7734258961fe5a5f2a43390e</td><td>TRANSFERRING FACE VERIFICATION NETS TO PAIN AND EXPRESSION REGRESSION +<br/>Dept. of {Computer Science1, Electrical & Computer Engineering2, Radiation Oncology3, Cognitive Science4} +<br/><b>Johns Hopkins University, 3400 N. Charles St, Baltimore, MD 21218, USA</b><br/>5Dept. of EE, UESTC, 2006 Xiyuan Ave, Chengdu, Sichuan 611731, China +<br/><b>Tsinghua University, Beijing 100084, China</b></td><td>('39369840', 'Feng Wang', 'feng wang')<br/>('40031188', 'Xiang Xiang', 'xiang xiang')<br/>('1692867', 'Chang Liu', 'chang liu')<br/>('1709073', 'Trac D. Tran', 'trac d. tran')<br/>('3207112', 'Austin Reiter', 'austin reiter')<br/>('1678633', 'Gregory D. Hager', 'gregory d. hager')<br/>('2095823', 'Harry Quon', 'harry quon')<br/>('1709439', 'Jian Cheng', 'jian cheng')<br/>('1746141', 'Alan L. Yuille', 'alan l. yuille')</td><td></td></tr><tr><td>920a92900fbff22fdaaef4b128ca3ca8e8d54c3e</td><td>LEARNING PATTERN TRANSFORMATION MANIFOLDS WITH PARAMETRIC ATOM +<br/>SELECTION +<br/>Ecole Polytechnique F´ed´erale de Lausanne (EPFL) +<br/>Signal Processing Laboratory (LTS4) +<br/>Switzerland-1015 Lausanne +</td><td>('12636684', 'Elif Vural', 'elif vural')<br/>('1703189', 'Pascal Frossard', 'pascal frossard')</td><td></td></tr><tr><td>9207671d9e2b668c065e06d9f58f597601039e5e</td><td>Face Detection Using a 3D Model on +<br/>Face Keypoints +</td><td>('2455529', 'Adrian Barbu', 'adrian barbu')<br/>('3019469', 'Gary Gramajo', 'gary gramajo')</td><td></td></tr><tr><td>924b14a9e36d0523a267293c6d149bca83e73f3b</td><td>Volume 5, Number 2, pp. 133 -164 +<br/>Development and Evaluation of a Method +<br/>Employed to Identify Internal State +<br/>Utilizing Eye Movement Data +<br/>(cid:2) Graduate School of Media and +<br/><b>Governance, Keio University</b><br/>(JAPAN) +<br/>(cid:3) Faculty of Environmental +<br/><b>Information, Keio University</b><br/>(JAPAN) +</td><td>('31726964', 'Noriyuki Aoyama', 'noriyuki aoyama')<br/>('1889276', 'Tadahiko Fukuda', 'tadahiko fukuda')</td><td></td></tr><tr><td>9282239846d79a29392aa71fc24880651826af72</td><td>Antonakos et al. EURASIP Journal on Image and Video Processing 2014, 2014:14 +<br/>http://jivp.eurasipjournals.com/content/2014/1/14 +<br/>RESEARCH +<br/>Open Access +<br/>Classification of extreme facial events in sign +<br/>language videos +</td><td>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('1738119', 'Vassilis Pitsikalis', 'vassilis pitsikalis')<br/>('1750686', 'Petros Maragos', 'petros maragos')</td><td></td></tr><tr><td>92115b620c7f653c847f43b6c4ff0470c8e55dab</td><td>Training Deformable Object Models for Human +<br/>Detection Based on Alignment and Clustering +<br/>Department of Computer Science, +<br/>Centre of Biological Signalling Studies (BIOSS), +<br/><b>University of Freiburg, Germany</b></td><td>('2127987', 'Benjamin Drayer', 'benjamin drayer')<br/>('1710872', 'Thomas Brox', 'thomas brox')</td><td>{drayer,brox}@cs.uni-freiburg.de +</td></tr><tr><td>928b8eb47288a05611c140d02441660277a7ed54</td><td>Exploiting Images for Video Recognition with Hierarchical Generative +<br/>Adversarial Networks +<br/>1 Beijing Laboratory of Intelligent Information Technology, School of Computer Science, +<br/><b>Big Data Research Center, University of Electronic Science and Technology of China</b><br/><b>Beijing Institute of Technology</b></td><td>('3450614', 'Feiwu Yu', 'feiwu yu')<br/>('2125709', 'Xinxiao Wu', 'xinxiao wu')<br/>('9177510', 'Yuchao Sun', 'yuchao sun')<br/>('2055900', 'Lixin Duan', 'lixin duan')</td><td>{yufeiwu,wuxinxiao,sunyuchao}@bit.edu.cn, lxduan@uestc.edu.cn +</td></tr><tr><td>926e97d5ce2a6e070f8ec07c5aa7f91d3df90ba0</td><td>Facial Expression Recognition Using Enhanced Deep 3D Convolutional Neural +<br/>Networks +<br/>Department of Electrical and Computer Engineering +<br/><b>University of Denver, Denver, CO</b></td><td>('3093835', 'Mohammad H. Mahoor', 'mohammad h. mahoor')</td><td>behzad.hasani@du.edu and mmahoor@du.edu +</td></tr><tr><td>92c2dd6b3ac9227fce0a960093ca30678bceb364</td><td>Provided by the author(s) and NUI Galway in accordance with publisher policies. Please cite the published +<br/>version when available. +<br/>Title +<br/>On color texture normalization for active appearance models +<br/>Author(s) +<br/>Ionita, Mircea C.; Corcoran, Peter M.; Buzuloiu, Vasile +<br/>Publication +<br/>Date +<br/>2009-05-12 +<br/>Publication +<br/>Information +<br/>Ionita, M. C., Corcoran, P., & Buzuloiu, V. (2009). On Color +<br/>Texture Normalization for Active Appearance Models. Image +<br/>Processing, IEEE Transactions on, 18(6), 1372-1378. +<br/>Publisher +<br/>IEEE +<br/>Link to +<br/>publisher's +<br/>version +<br/>http://dx.doi.org/10.1109/TIP.2009.2017163 +<br/>Item record +<br/>http://hdl.handle.net/10379/1350 +<br/>Some rights reserved. For more information, please see the item record link above. +<br/>Downloaded 2018-11-06T00:40:53Z +</td><td></td><td></td></tr><tr><td>92e464a5a67582d5209fa75e3b29de05d82c7c86</td><td>Reconstruction for Feature Disentanglement in Pose-invariant Face Recognition +<br/><b>Rutgers University, NJ, USA</b><br/>2NEC Labs America, CA, USA +</td><td>('4340744', 'Xi Peng', 'xi peng')<br/>('39960064', 'Xiang Yu', 'xiang yu')<br/>('1729571', 'Kihyuk Sohn', 'kihyuk sohn')</td><td>{xpeng.cs, dnm}@rutgers.edu, {xiangyu, ksohn, manu}@nec-labs.com +</td></tr><tr><td>927ba64123bd4a8a31163956b3d1765eb61e4426</td><td>Customer satisfaction measuring based on the most +<br/>significant facial emotion +<br/>To cite this version: +<br/>most significant facial emotion. 15th IEEE International Multi-Conference on Systems, Signals +<br/>Devices (SSD 2018), Mar 2018, Hammamet, Tunisia. <hal-01790317> +<br/>HAL Id: hal-01790317 +<br/>https://hal-upec-upem.archives-ouvertes.fr/hal-01790317 +<br/>Submitted on 11 May 2018 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('50101862', 'Rostom Kachouri', 'rostom kachouri')<br/>('50101862', 'Rostom Kachouri', 'rostom kachouri')</td><td></td></tr><tr><td>922838dd98d599d1d229cc73896d55e7a769aa7c</td><td>Learning Hierarchical Representations for Face Verification +<br/>with Convolutional Deep Belief Networks +<br/>Erik Learned-Miller +<br/><b>University of Massachusetts</b><br/><b>University of Michigan</b><br/><b>University of Massachusetts</b><br/>Amherst, MA +<br/>Ann Arbor, MI +<br/>Amherst, MA +</td><td>('3219900', 'Gary B. Huang', 'gary b. huang')<br/>('1697141', 'Honglak Lee', 'honglak lee')</td><td>gbhuang@cs.umass.edu +<br/>honglak@eecs.umich.edu +<br/>elm@cs.umass.edu +</td></tr><tr><td>9294739e24e1929794330067b84f7eafd286e1c8</td><td>Expression Recognition using Elastic Graph Matching +<br/>21， +<br/>21， +<br/>21， +<br/>, Cairong Zhou 2 +<br/><b>Research Center for Learning Science, Southeast University, Nanjing 210096, China</b><br/><b>Southeast University, Nanjing 210096, China</b></td><td>('40622743', 'Yujia Cao', 'yujia cao')<br/>('40608983', 'Wenming Zheng', 'wenming zheng')<br/>('1718117', 'Li Zhao', 'li zhao')</td><td>Email: yujia_cao@seu.edu.cn +</td></tr><tr><td>92fada7564d572b72fd3be09ea3c39373df3e27c</td><td></td><td></td><td></td></tr><tr><td>927ad0dceacce2bb482b96f42f2fe2ad1873f37a</td><td>Interest-Point based Face Recognition System +<br/>87 +<br/>X +<br/>Interest-Point based Face Recognition System +<br/>Spain +<br/>1. Introduction +<br/>Among all applications of face recognition systems, surveillance is one of the most +<br/>challenging ones. In such an application, the goal is to detect known criminals in crowded +<br/>environments, like airports or train stations. Some attempts have been made, like those of +<br/>Tokio (Engadget, 2006) or Mainz (Deutsche Welle, 2006), with limited success. +<br/>The first task to be carried out in an automatic surveillance system involves the detection of +<br/>all the faces in the images taken by the video cameras. Current face detection algorithms are +<br/>highly reliable and thus, they will not be the focus of our work. Some of the best performing +<br/>examples are the Viola-Jones algorithm (Viola & Jones, 2004) or the Schneiderman-Kanade +<br/>algorithm (Schneiderman & Kanade, 2000). +<br/>The second task to be carried out involves the comparison of all detected faces among the +<br/>database of known criminals. The ideal behaviour of an automatic system performing this +<br/>task would be to get a 100% correct identification rate, but this behaviour is far from the +<br/>capabilities of current face recognition algorithms. Assuming that there will be false +<br/>identifications, supervised surveillance systems seem to be the most realistic option: the +<br/>automatic system issues an alarm whenever it detects a possible match with a criminal, and +<br/>a human decides whether it is a false alarm or not. Figure 1 shows an example. +<br/>However, even in a supervised scenario the requirements for the face recognition algorithm +<br/>are extremely high: the false alarm rate must be low enough as to allow the human operator +<br/>to cope with it; and the percentage of undetected criminals must be kept to a minimum in +<br/>order to ensure security. Fulfilling both requirements at the same time is the main challenge, +<br/>as a reduction in false alarm rate usually implies an increase of the percentage of undetected +<br/>criminals. +<br/>We propose a novel face recognition system based in the use of interest point detectors and +<br/>local descriptors. In order to check the performances of our system, and particularly its +<br/>performances in a surveillance application, we present experimental results in terms of +<br/>Receiver Operating Characteristic curves or ROC curves. From the experimental results, it +<br/>becomes clear that our system outperforms classical appearance based approaches. +<br/>www.intechopen.com +</td><td>('35178717', 'Cesar Fernandez', 'cesar fernandez')<br/>('3686544', 'Maria Asuncion Vicente', 'maria asuncion vicente')<br/>('2422580', 'Miguel Hernandez', 'miguel hernandez')</td><td></td></tr><tr><td>929bd1d11d4f9cbc638779fbaf958f0efb82e603</td><td>This is the author’s version of a work that was submitted/accepted for pub- +<br/>lication in the following source: +<br/>Zhang, Ligang & Tjondronegoro, Dian W. (2010) Improving the perfor- +<br/>mance of facial expression recognition using dynamic, subtle and regional +<br/>features. +<br/>In Kok, WaiWong, B. Sumudu, U. Mendis, & Abdesselam , +<br/>Bouzerdoum (Eds.) Neural Information Processing. Models and Applica- +<br/>tions, Lecture Notes in Computer Science, Sydney, N.S.W, pp. 582-589. +<br/>This file was downloaded from: http://eprints.qut.edu.au/43788/ +<br/>c(cid:13) Copyright 2010 Springer-Verlag +<br/>Conference proceedings published, by Springer Verlag, will be available +<br/>via Lecture Notes in Computer Science http://www.springer.de/comp/lncs/ +<br/>Notice: Changes introduced as a result of publishing processes such as +<br/>copy-editing and formatting may not be reflected in this document. For a +<br/>definitive version of this work, please refer to the published source: +<br/>http://dx.doi.org/10.1007/978-3-642-17534-3_72 +</td><td></td><td></td></tr><tr><td>923ec0da8327847910e8dd71e9d801abcbc93b08</td><td>Hide-and-Seek: Forcing a Network to be Meticulous for +<br/>Weakly-supervised Object and Action Localization +<br/><b>University of California, Davis</b></td><td>('19553871', 'Krishna Kumar Singh', 'krishna kumar singh')<br/>('1883898', 'Yong Jae Lee', 'yong jae lee')</td><td></td></tr><tr><td>0c741fa0966ba3ee4fc326e919bf2f9456d0cd74</td><td>Facial Age Estimation by Learning from Label Distributions +<br/><b>School of Mathematical Sciences, Monash University, VIC 3800, Australia</b><br/><b>School of Computer Science and Engineering, Southeast University, Nanjing 210096, China</b><br/><b>National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing 210093, China</b></td><td>('1735299', 'Xin Geng', 'xin geng')<br/>('2848275', 'Kate Smith-Miles', 'kate smith-miles')<br/>('1692625', 'Zhi-Hua Zhou', 'zhi-hua zhou')</td><td></td></tr><tr><td>0c435e7f49f3e1534af0829b7461deb891cf540a</td><td>Capturing Global Semantic Relationships for Facial Action Unit Recognition +<br/><b>Rensselaer Polytechnic Institute</b><br/><b>School of Electrical Engineering and Automation, Harbin Institute of Technology</b><br/><b>School of Computer Science and Technology, University of Science and Technology of China</b></td><td>('2860279', 'Ziheng Wang', 'ziheng wang')<br/>('1830523', 'Yongqiang Li', 'yongqiang li')<br/>('1791319', 'Shangfei Wang', 'shangfei wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>{wangz10,liy23,jiq}@rpi.edu +<br/>sfwang@ustc.edu.cn +</td></tr><tr><td>0cb7e4c2f6355c73bfc8e6d5cdfad26f3fde0baf</td><td>International Journal of Artificial Intelligence & Applications (IJAIA), Vol. 5, No. 3, May 2014 +<br/>FACIAL EXPRESSION RECOGNITION BASED ON +<br/><b>Computer Science, Engineering and Mathematics School, Flinders University, Australia</b><br/><b>Computer Science, Engineering and Mathematics School, Flinders University, Australia</b></td><td>('3105876', 'Humayra Binte Ali', 'humayra binte ali')<br/>('1739260', 'David M W Powers', 'david m w powers')</td><td></td></tr><tr><td>0c30f6303dc1ff6d05c7cee4f8952b74b9533928</td><td>Pareto Discriminant Analysis +<br/>Karim T. Abou–Moustafa +<br/>Centre of Intelligent Machines +<br/><b>The Robotics Institute</b><br/>Centre of Intelligent Machines +<br/><b>McGill University</b><br/><b>Carnegie Mellon University</b><br/><b>McGill University</b></td><td>('1707876', 'Fernando De la Torre', 'fernando de la torre')<br/>('1701344', 'Frank P. Ferrie', 'frank p. ferrie')</td><td>karimt@cim.mcgill.ca +<br/>ftorre@cs.cmu.edu +<br/>ferrie@cim.mcgill.ca +</td></tr><tr><td>0ccc535d12ad2142a8310d957cc468bbe4c63647</td><td>Better Exploiting OS-CNNs for Better Event Recognition in Images +<br/><b>Shenzhen Key Lab of CVPR, Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('1915826', 'Zhe Wang', 'zhe wang')<br/>('2072196', 'Sheng Guo', 'sheng guo')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td>{07wanglimin, buptwangzhe2012, guosheng1001}@gmail.com, yu.qiao@siat.ac.cn +</td></tr><tr><td>0c8a0a81481ceb304bd7796e12f5d5fa869ee448</td><td>International Journal of Fuzzy Logic and Intelligent Systems, vol. 10, no. 2, June 2010, pp. 95-100 +<br/>A Spatial Regularization of LDA for Face Recognition +<br/><b>Gangnung-Wonju National University</b><br/>123 Chibyun-Dong, Kangnung, 210-702, Korea +</td><td>('39845108', 'Lae-Jeong Park', 'lae-jeong park')</td><td>Tel : +82-33-640-2389, Fax : +82-33-646-0740, E-mail : ljpark@gwnu.ac.kr +</td></tr><tr><td>0c36c988acc9ec239953ff1b3931799af388ef70</td><td>Face Detection Using Improved Faster RCNN +<br/>Huawei Cloud BU, China +<br/>Figure1.Face detection results of FDNet1.0 +</td><td>('2568329', 'Changzheng Zhang', 'changzheng zhang')<br/>('5084124', 'Xiang Xu', 'xiang xu')<br/>('2929196', 'Dandan Tu', 'dandan tu')</td><td>{zhangzhangzheng, xuxiang12, tudandan}@huawei.com +</td></tr><tr><td>0c5ddfa02982dcad47704888b271997c4de0674b</td><td></td><td></td><td></td></tr><tr><td>0c79a39a870d9b56dc00d5252d2a1bfeb4c295f1</td><td>Face Recognition in Videos by Label Propagation +<br/><b>International Institute of Information Technology, Hyderabad, India</b></td><td>('37956314', 'Vijay Kumar', 'vijay kumar')<br/>('3185334', 'Anoop M. Namboodiri', 'anoop m. namboodiri')</td><td>{vijaykumar.r@research., anoop@, jawahar@}iiit.ac.in +</td></tr><tr><td>0cccf576050f493c8b8fec9ee0238277c0cfd69a</td><td></td><td></td><td></td></tr><tr><td>0cdb49142f742f5edb293eb9261f8243aee36e12</td><td>Combined Learning of Salient Local Descriptors and Distance Metrics +<br/>for Image Set Face Verification +<br/>NICTA, PO Box 6020, St Lucia, QLD 4067, Australia +<br/><b>University of Queensland, School of ITEE, QLD 4072, Australia</b></td><td>('1781182', 'Conrad Sanderson', 'conrad sanderson')<br/>('3026404', 'Yongkang Wong', 'yongkang wong')<br/>('2270092', 'Brian C. Lovell', 'brian c. lovell')</td><td></td></tr><tr><td>0c069a870367b54dd06d0da63b1e3a900a257298</td><td>Author manuscript, published in "ICANN 2011 - International Conference on Artificial Neural Networks (2011)" +</td><td></td><td></td></tr><tr><td>0c75c7c54eec85e962b1720755381cdca3f57dfb</td><td>2212 +<br/>Face Landmark Fitting via Optimized Part +<br/>Mixtures and Cascaded Deformable Model +</td><td>('39960064', 'Xiang Yu', 'xiang yu')<br/>('1768190', 'Junzhou Huang', 'junzhou huang')<br/>('1753384', 'Shaoting Zhang', 'shaoting zhang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td></td></tr><tr><td>0cf2eecf20cfbcb7f153713479e3206670ea0e9c</td><td>Privacy-Protective-GAN for Face De-identification +<br/><b>Temple University</b></td><td>('50117915', 'Yifan Wu', 'yifan wu')<br/>('46319628', 'Fan Yang', 'fan yang')<br/>('1805398', 'Haibin Ling', 'haibin ling')</td><td>{yifan.wu, fyang, hbling} @temple.edu +</td></tr><tr><td>0ca36ecaf4015ca4095e07f0302d28a5d9424254</td><td>Improving Bag-of-Visual-Words Towards Effective Facial Expressive +<br/>Image Classification +<br/>1Univ. Grenoble Alpes, CNRS, Grenoble INP∗ , GIPSA-lab, 38000 Grenoble, France +<br/>Keywords: +<br/>BoVW, k-means++, Relative Conjunction Matrix, SIFT, Spatial Pyramids, TF.IDF. +</td><td>('10762131', 'Dawood Al Chanti', 'dawood al chanti')<br/>('1788869', 'Alice Caplier', 'alice caplier')</td><td>dawood.alchanti@gmail.com +</td></tr><tr><td>0c1d85a197a1f5b7376652a485523e616a406273</td><td>Joint Registration and Representation Learning for Unconstrained Face +<br/>Identification +<br/><b>University of Canberra, Australia, Data61 - CSIRO and ANU, Australia</b><br/><b>Khalifa University, Abu Dhabi, United Arab Emirates</b></td><td>('2008898', 'Munawar Hayat', 'munawar hayat')<br/>('1802072', 'Naoufel Werghi', 'naoufel werghi')</td><td>{munawar.hayat,roland.goecke}@canberra.edu.au, salman.khan@csiro.au, naoufel.werghi@kustar.ac.ae +</td></tr><tr><td>0ca66283f4fb7dbc682f789fcf6d6732006befd5</td><td>Active Dictionary Learning for Image Representation +<br/>Department of Electrical and Computer Engineering +<br/><b>Rutgers, The State University of New Jersey, Piscataway, NJ</b></td><td>('37799945', 'Tong Wu', 'tong wu')<br/>('9208982', 'Anand D. Sarwate', 'anand d. sarwate')<br/>('2138101', 'Waheed U. Bajwa', 'waheed u. bajwa')</td><td></td></tr><tr><td>0c7f27d23a162d4f3896325d147f412c40160b52</td><td>Models and Algorithms for +<br/>Vision through the Atmosphere +<br/>Submitted in partial fulfillment of the +<br/>requirements for the degree +<br/>of Doctor of Philosophy +<br/>in the Graduate School of Arts and Sciences +<br/><b>COLUMBIA UNIVERSITY</b><br/>2003 +</td><td>('1779052', 'Srinivasa G. Narasimhan', 'srinivasa g. narasimhan')</td><td></td></tr><tr><td>0cfca73806f443188632266513bac6aaf6923fa8</td><td>Predictive Uncertainty in Large Scale Classification +<br/>using Dropout - Stochastic Gradient Hamiltonian +<br/>Monte Carlo. +<br/>Vergara, Diego∗1, Hern´andez, Sergio∗2, Valdenegro-Toro, Mat´ıas∗∗3 and Jorquera, Felipe∗4. +<br/>∗Laboratorio de Procesamiento de Informaci´on Geoespacial, Universidad Cat´olica del Maule, Chile. +<br/>∗∗German Research Centre for Artificial Intelligence, Bremen, Germany. +</td><td></td><td>Email: 1diego.vergara@alu.ucm.cl, 2shernandez@ucm.cl,3matias.valdenegro@dfki.de, +<br/>4f.jorquera.uribe@gmail.com +</td></tr><tr><td>0c20fd90d867fe1be2459223a3cb1a69fa3d44bf</td><td>A Monte Carlo Strategy to Integrate Detection +<br/>and Model-Based Face Analysis +<br/>Department for Mathematics and Computer Science +<br/><b>University of Basel, Switzerland</b></td><td>('2591294', 'Andreas Forster', 'andreas forster')<br/>('34460642', 'Bernhard Egger', 'bernhard egger')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td>sandro.schoenborn,andreas.forster,bernhard.egger,thomas.vetter@unibas.ch +</td></tr><tr><td>0c2875bb47db3698dbbb3304aca47066978897a4</td><td>Recurrent Models for Situation Recognition +<br/><b>University of Illinois at Urbana-Champaign</b></td><td>('36508529', 'Arun Mallya', 'arun mallya')<br/>('1749609', 'Svetlana Lazebnik', 'svetlana lazebnik')</td><td>{amallya2,slazebni}@illinois.edu +</td></tr><tr><td>0c3f7272a68c8e0aa6b92d132d1bf8541c062141</td><td>Hindawi Publishing Corporation +<br/>e Scientific World Journal +<br/>Volume 2014, Article ID 672630, 6 pages +<br/>http://dx.doi.org/10.1155/2014/672630 +<br/>Research Article +<br/>Kruskal-Wallis-Based Computationally Efficient Feature +<br/>Selection for Face Recognition +<br/><b>Foundation University, Rawalpindi 46000, Pakistan</b><br/><b>Shaheed Zulfikar Ali Bhutto Institute of Science and Technology Islamabad</b><br/>Islamabad 44000, Pakistan +<br/><b>International Islamic University, Islamabad 44000, Pakistan</b><br/>Received 5 December 2013; Accepted 10 February 2014; Published 21 May 2014 +<br/>Academic Editors: S. Balochian, V. Bhatnagar, and Y. Zhang +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Face recognition in today’s technological world, and face recognition applications attain much more importance. Most of the +<br/>existing work used frontal face images to classify face image. However these techniques fail when applied on real world face images. +<br/>The proposed technique effectively extracts the prominent facial features. Most of the features are redundant and do not contribute +<br/>to representing face. In order to eliminate those redundant features, computationally efficient algorithm is used to select the more +<br/>discriminative face features. Extracted features are then passed to classification step. In the classification step, different classifiers +<br/>are ensemble to enhance the recognition accuracy rate as single classifier is unable to achieve the high accuracy. Experiments are +<br/>performed on standard face database images and results are compared with existing techniques. +<br/>1. Introduction +<br/>Face recognition is becoming more acceptable in the domain +<br/>of computer vision and pattern recognition. The authenti- +<br/>cation systems based on the traditional ID card and pass- +<br/>word are nowadays replaced by the techniques which are +<br/>more preferable in order to handle the security issues. The +<br/>authentication systems based on biometrics are one of the +<br/>substitutes which are independent of the user’s memory and +<br/>not subjected to loss. Among those systems, face recognition +<br/>gains special attention because of the security it provides and +<br/>because it is independent of the high accuracy equipment +<br/>unlike iris and recognition based on the fingerprints. +<br/>Feature selection in pattern recognition is specifying the +<br/>subset of significant features to decrease the data dimensions +<br/>and at the same time it provides the set of selective features. +<br/>Image is represented by set of features in methods used for +<br/>feature extraction and each feature plays a vital role in the +<br/>process of recognition. The feature selection algorithm drops +<br/>all the unrelated features with the highly acceptable precision +<br/>rate as compared to some other pattern classification problem +<br/>in which higher precision rate cannot be obtained by greater +<br/>number of feature sets [1]. +<br/>The feature selected by the classifiers plays a vital role +<br/>in producing the best features that are vigorous to the +<br/>inconsistent environment, for example, change in expressions +<br/>and other barriers. Local (texture-based) and global (holistic) +<br/>approaches are the two approaches used for face recognition +<br/>[2]. Local approaches characterized the face in the form of +<br/>geometric measurements which matches the unfamiliar face +<br/>with the closest face from database. Geometric measurements +<br/>contain angles and the distance of different facial points, +<br/>for example, mouth position, nose length, and eyes. Global +<br/>features are extracted by the use of algebraic methods like +<br/>PCA (principle component analysis) and ICA (independent +<br/>component analysis) [3]. PCA shows a quick response to +<br/>light and variation as it serves inner and outer classes +<br/>fairly. In face recognition, LDA (linear discriminate analysis) +<br/>usually performs better than PCA but separable creation is +<br/>not precise in classification. Good recognition rates can be +<br/>produced by transformation techniques like DCT (discrete +<br/>cosine transform) and DWT (discrete wavelet transform) [4]. +</td><td>('8652075', 'Sajid Ali Khan', 'sajid ali khan')<br/>('9955306', 'Ayyaz Hussain', 'ayyaz hussain')<br/>('1959869', 'Abdul Basit', 'abdul basit')<br/>('2388005', 'Sheeraz Akram', 'sheeraz akram')<br/>('8652075', 'Sajid Ali Khan', 'sajid ali khan')</td><td>Correspondence should be addressed to Sajid Ali Khan; sajidalibn@gmail.com +</td></tr><tr><td>0cbc4dcf2aa76191bbf641358d6cecf38f644325</td><td>Visage: A Face Interpretation Engine for +<br/>Smartphone Applications +<br/><b>Dartmouth College, 6211 Sudiko Lab, Hanover, NH 03755, USA</b><br/><b>Intel Lab, 2200 Mission College Blvd, Santa Clara, CA 95054, USA</b><br/>3 Microsoft Research Asia, No. 5 Dan Ling St., Haidian District, Beijing, China +</td><td>('1840450', 'Xiaochao Yang', 'xiaochao yang')<br/>('1702472', 'Chuang-Wen You', 'chuang-wen you')<br/>('1884089', 'Hong Lu', 'hong lu')<br/>('1816301', 'Mu Lin', 'mu lin')<br/>('2772904', 'Nicholas D. Lane', 'nicholas d. lane')<br/>('1690035', 'Andrew T. Campbell', 'andrew t. campbell')</td><td>{Xiaochao.Yang,chuang-wen.you}@dartmouth.edu,hong.lu@intel.com, +<br/>mu.lin@dartmouth.edu,niclane@microsoft.com,campbell@cs.dartmouth.edu +</td></tr><tr><td>0ce8a45a77e797e9d52604c29f4c1e227f604080</td><td>International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol.3,No. 6,December 2013 +<br/>ZERNIKE MOMENT-BASED FEATURE EXTRACTION +<br/>FOR FACIAL RECOGNITION OF IDENTICAL TWINS +<br/>1Department of Electrical,Computer and Biomedical Engineering, Qazvin branch, +<br/><b>Amirkabir University of Technology, Tehran</b><br/><b>IslamicAzad University, Qazvin, Iran</b><br/>Iran +</td><td>('13302047', 'Hoda Marouf', 'hoda marouf')<br/>('1692435', 'Karim Faez', 'karim faez')</td><td></td></tr><tr><td>0ce3a786aed896d128f5efdf78733cc675970854</td><td>Learning the Face Prior +<br/>for Bayesian Face Recognition +<br/>Department of Information Engineering, +<br/><b>The Chinese University of Hong Kong, China</b></td><td>('2312486', 'Chaochao Lu', 'chaochao lu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>0c54e9ac43d2d3bab1543c43ee137fc47b77276e</td><td></td><td></td><td></td></tr><tr><td>0c5afb209b647456e99ce42a6d9d177764f9a0dd</td><td>97 +<br/>Recognizing Action Units for +<br/>Facial Expression Analysis +</td><td>('40383812', 'Ying-li Tian', 'ying-li tian')<br/>('1733113', 'Takeo Kanade', 'takeo kanade')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td></td></tr><tr><td>0c59071ddd33849bd431165bc2d21bbe165a81e0</td><td>Person Recognition in Personal Photo Collections +<br/><b>Max Planck Institute for Informatics</b><br/>Saarbrücken, Germany +</td><td>('2390510', 'Seong Joon Oh', 'seong joon oh')<br/>('1798000', 'Rodrigo Benenson', 'rodrigo benenson')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td>{joon,benenson,mfritz,schiele}@mpi-inf.mpg.de +</td></tr><tr><td>0c377fcbc3bbd35386b6ed4768beda7b5111eec6</td><td>258 +<br/>A Unified Probabilistic Framework +<br/>for Spontaneous Facial Action Modeling +<br/>and Understanding +</td><td>('1686235', 'Yan Tong', 'yan tong')<br/>('1713712', 'Jixu Chen', 'jixu chen')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td></td></tr><tr><td>0c12cbb9b9740dfa2816b8e5cde69c2f5a715c58</td><td>Memory-Augmented Attribute Manipulation Networks for +<br/>Interactive Fashion Search +<br/><b>Southwest Jiaotong University</b><br/><b>National University of Singapore</b><br/><b>AI Institute</b></td><td>('33901950', 'Bo Zhao', 'bo zhao')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')<br/>('1814091', 'Xiao Wu', 'xiao wu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>zhaobo@my.swjtu.edu.cn, elezhf@nus.edu.sg, wuxiaohk@swjtu.edu.cn, yanshuicheng@360.cn +</td></tr><tr><td>0cb2dd5f178e3a297a0c33068961018659d0f443</td><td></td><td>('2964917', 'Cameron Whitelam', 'cameron whitelam')<br/>('1885566', 'Emma Taborsky', 'emma taborsky')<br/>('1917247', 'Austin Blanton', 'austin blanton')<br/>('8033275', 'Brianna Maze', 'brianna maze')<br/>('15282121', 'Tim Miller', 'tim miller')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')<br/>('40205896', 'James A. Duncan', 'james a. duncan')<br/>('2040584', 'Kristen Allen', 'kristen allen')<br/>('39403529', 'Jordan Cheney', 'jordan cheney')<br/>('2136478', 'Patrick Grother', 'patrick grother')</td><td></td></tr><tr><td>0cd8895b4a8f16618686f622522726991ca2a324</td><td>Discrete Choice Models for Static Facial Expression +<br/>Recognition +<br/><b>Ecole Polytechnique Federale de Lausanne, Signal Processing Institute</b><br/>2 Ecole Polytechnique Federale de Lausanne, Operation Research Group +<br/>Ecublens, 1015 Lausanne, Switzerland +<br/>Ecublens, 1015 Lausanne, Switzerland +</td><td>('1794461', 'Gianluca Antonini', 'gianluca antonini')<br/>('2916630', 'Matteo Sorci', 'matteo sorci')<br/>('1690395', 'Michel Bierlaire', 'michel bierlaire')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')</td><td>{Matteo.Sorci,Gianluca.Antonini,JP.Thiran}@epfl.ch +<br/>Michel.Bierlaire@epfl.ch +</td></tr><tr><td>0cf7da0df64557a4774100f6fde898bc4a3c4840</td><td>Shape Matching and Object Recognition using Low Distortion Correspondences +<br/>Department of Electrical Engineering and Computer Science +<br/>U.C. Berkeley +</td><td>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('1689212', 'Jitendra Malik', 'jitendra malik')</td><td>faberg,millert,malikg@eecs.berkeley.edu +</td></tr><tr><td>0cbe059c181278a373292a6af1667c54911e7925</td><td>Owl and Lizard: Patterns of Head Pose and Eye +<br/>Pose in Driver Gaze Classification +<br/><b>Massachusetts Institute of Technology (MIT</b><br/><b>Chalmers University of Technology, SAFER</b></td><td>('7137846', 'Joonbum Lee', 'joonbum lee')<br/>('1901227', 'Bryan Reimer', 'bryan reimer')<br/>('35816778', 'Trent Victor', 'trent victor')</td><td></td></tr><tr><td>0c4659b35ec2518914da924e692deb37e96d6206</td><td>1236 +<br/>Registering a MultiSensor Ensemble of Images +</td><td>('1822837', 'Jeff Orchard', 'jeff orchard')<br/>('6056877', 'Richard Mann', 'richard mann')</td><td></td></tr><tr><td>0c6e29d82a5a080dc1db9eeabbd7d1529e78a3dc</td><td>Learning Bayesian Network Classifiers for Facial Expression Recognition using +<br/>both Labeled and Unlabeled Data +<br/><b>Beckman Institute, University of Illinois at Urbana-Champaign, IL, USA</b><br/>iracohen, huang +<br/> Escola Polit´ecnica, Universidade de S˜ao Paulo, S˜ao Paulo, Brazil +<br/>fgcozman, marcelo.cirelo +</td><td>('1774778', 'Ira Cohen', 'ira cohen')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>@ifp.uiuc.edu +<br/> Leiden Institute of Advanced Computer Science, Leiden University, The Netherlands, nicu@liacs.nl +<br/>@usp.br +</td></tr><tr><td>0ced7b814ec3bb9aebe0fcf0cac3d78f36361eae</td><td>Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/> A Monthly Journal of Computer Science and Information Technology +<br/>ISSN 2320–088X +<br/>IMPACT FACTOR: 6.017 +<br/> +<br/> IJCSMC, Vol. 6, Issue. 1, January 2017, pg.221 – 227 +<br/>Central Local Directional Pattern Value +<br/>Flooding Co-occurrence Matrix based +<br/>Features for Face Recognition +<br/><b>Gokaraju Rangaraju Institute of Engineering and Technology, Hyderabad</b></td><td>('40221166', 'Chandra Sekhar Reddy', 'chandra sekhar reddy')<br/>('40221166', 'Chandra Sekhar Reddy', 'chandra sekhar reddy')</td><td></td></tr><tr><td>0c53ef79bb8e5ba4e6a8ebad6d453ecf3672926d</td><td>SUBMITTED TO JOURNAL +<br/>Weakly Supervised PatchNets: Describing and +<br/>Aggregating Local Patches for Scene Recognition +</td><td>('40184588', 'Zhe Wang', 'zhe wang')<br/>('39709927', 'Limin Wang', 'limin wang')<br/>('40457196', 'Yali Wang', 'yali wang')<br/>('3047890', 'Bowen Zhang', 'bowen zhang')<br/>('40285012', 'Yu Qiao', 'yu qiao')</td><td></td></tr><tr><td>0c60eebe10b56dbffe66bb3812793dd514865935</td><td></td><td></td><td></td></tr><tr><td>0c05f60998628884a9ac60116453f1a91bcd9dda</td><td>Optimizing Open-Ended Crowdsourcing: The Next Frontier in +<br/>Crowdsourced Data Management +<br/><b>University of Illinois</b><br/><b>cid:63)Stanford University</b></td><td>('32953042', 'Akash Das Sarma', 'akash das sarma')<br/>('8336538', 'Vipul Venkataraman', 'vipul venkataraman')</td><td></td></tr><tr><td>6601a0906e503a6221d2e0f2ca8c3f544a4adab7</td><td>SRTM-2 2/9/06 3:27 PM Page 321 +<br/>Detection of Ancient Settlement Mounds: +<br/>Archaeological Survey Based on the +<br/>SRTM Terrain Model +<br/>B.H. Menze, J.A. Ur, and A.G. Sherratt +</td><td></td><td></td></tr><tr><td>660b73b0f39d4e644bf13a1745d6ee74424d4a16</td><td></td><td></td><td>3,250+OPEN ACCESS BOOKS106,000+INTERNATIONALAUTHORS AND EDITORS113+ MILLIONDOWNLOADSBOOKSDELIVERED TO151 COUNTRIESAUTHORS AMONGTOP 1%MOST CITED SCIENTIST12.2%AUTHORS AND EDITORSFROM TOP 500 UNIVERSITIESSelection of our books indexed in theBook Citation Index in Web of Science™Core Collection (BKCI)Chapter from the book Reviews, Refinements and New Ideas in Face RecognitionDownloaded from: http://www.intechopen.com/books/reviews-refinements-and-new-ideas-in-face-recognitionPUBLISHED BYWorld's largest Science,Technology & Medicine Open Access book publisherInterested in publishing with InTechOpen?Contact us at book.department@intechopen.com </td></tr><tr><td>66d512342355fb77a4450decc89977efe7e55fa2</td><td>Under review as a conference paper at ICLR 2018 +<br/>LEARNING NON-LINEAR TRANSFORM WITH DISCRIM- +<br/>INATIVE AND MINIMUM INFORMATION LOSS PRIORS +<br/>Anonymous authors +<br/>Paper under double-blind review +</td><td></td><td></td></tr><tr><td>66aad5b42b7dda077a492e5b2c7837a2a808c2fa</td><td>A Novel PCA-Based Bayes Classifier +<br/>and Face Analysis +<br/>1 Centre de Visi´o per Computador, +<br/>Universitat Aut`onoma de Barcelona, Barcelona, Spain +<br/>2 Department of Computer Science, +<br/><b>Nanjing University of Science and Technology</b><br/>Nanjing, People’s Republic of China +<br/>3 HEUDIASYC - CNRS Mixed Research Unit, +<br/><b>Compi`egne University of Technology</b><br/>60205 Compi`egne cedex, France +</td><td>('1761329', 'Zhong Jin', 'zhong jin')<br/>('1742818', 'Franck Davoine', 'franck davoine')<br/>('35428318', 'Zhen Lou', 'zhen lou')</td><td>zhong.jin@cvc.uab.es +<br/>jyyang@mail.njust.edu.cn +<br/>franck.davoine@hds.utc.fr +</td></tr><tr><td>66b9d954dd8204c3a970d86d91dd4ea0eb12db47</td><td>Evaluation of Gabor-Wavelet-Based Facial Action Unit Recognition +<br/>in Image Sequences of Increasing Complexity +<br/><b>IBM T. J. Watson Research Center, PO Box 704, Yorktown Heights, NY</b><br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA</b><br/><b>University of Pittsburgh, Pittsburgh, PA</b></td><td>('40383812', 'Ying-li Tian', 'ying-li tian')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>Email: yltian@us.ibm.com, +<br/>tk@cs.cmu.edu +<br/>jeffcohn@pitt.edu +</td></tr><tr><td>6643a7feebd0479916d94fb9186e403a4e5f7cbf</td><td>Chapter 8 +<br/>3D Face Recognition +</td><td>('1737428', 'Nick Pears', 'nick pears')</td><td></td></tr><tr><td>661ca4bbb49bb496f56311e9d4263dfac8eb96e9</td><td>Datasheets for Datasets +</td><td>('2076288', 'Timnit Gebru', 'timnit gebru')<br/>('1722360', 'Hal Daumé', 'hal daumé')</td><td></td></tr><tr><td>66dcd855a6772d2731b45cfdd75f084327b055c2</td><td>Quality Classified Image Analysis with Application +<br/>to Face Detection and Recognition +<br/>International Doctoral Innovation Centre +<br/><b>University of Nottingham Ningbo China</b><br/>School of Computer Science +<br/><b>University of Nottingham Ningbo China</b><br/><b>College of Information Engineering</b><br/><b>Shenzhen University, Shenzhen, China</b></td><td>('1684164', 'Fei Yang', 'fei yang')<br/>('1737486', 'Qian Zhang', 'qian zhang')<br/>('2155597', 'Miaohui Wang', 'miaohui wang')<br/>('1698461', 'Guoping Qiu', 'guoping qiu')</td><td></td></tr><tr><td>666939690c564641b864eed0d60a410b31e49f80</td><td>What Visual Attributes Characterize an Object Class ? +<br/><b>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of</b><br/>Sciences, No.95, Zhongguancun East Road, Beijing, 100190, China +<br/>2Microsoft Research, No.5, Dan Ling Street, Haidian District, Beijing 10080, China +</td><td>('3247966', 'Jianlong Fu', 'jianlong fu')<br/>('1783122', 'Jinqiao Wang', 'jinqiao wang')<br/>('3349534', 'Xin-Jing Wang', 'xin-jing wang')<br/>('3663422', 'Yong Rui', 'yong rui')<br/>('1694235', 'Hanqing Lu', 'hanqing lu')</td><td>1fjlfu, jqwang, luhqg@nlpr.ia.ac.cn, 2fxjwang, yongruig@microsoft.com +</td></tr><tr><td>66330846a03dcc10f36b6db9adf3b4d32e7a3127</td><td>Polylingual Multimodal Learning +<br/><b>Institute AIFB, Karlsruhe Institute of Technology, Germany</b></td><td>('3219864', 'Aditya Mogadala', 'aditya mogadala')</td><td>{aditya.mogadala}@kit.edu +</td></tr><tr><td>66d087f3dd2e19ffe340c26ef17efe0062a59290</td><td>Dog Breed Identification +<br/>Brian Mittl +<br/>Vijay Singh +</td><td></td><td>wlarow@stanford.edu +<br/>bmittl@stanford.edu +<br/>vpsingh@stanford.edu +</td></tr><tr><td>6618cff7f2ed440a0d2fa9e74ad5469df5cdbe4c</td><td>Ordinal Regression with Multiple Output CNN for Age Estimation +<br/><b>Xidian University 2Xi an Jiaotong University 3Microsoft Research Asia</b></td><td>('1786361', 'Zhenxing Niu', 'zhenxing niu')<br/>('1745420', 'Gang Hua', 'gang hua')<br/>('10699750', 'Xinbo Gao', 'xinbo gao')<br/>('36497527', 'Mo Zhou', 'mo zhou')<br/>('40367806', 'Le Wang', 'le wang')</td><td>{zhenxingniu,cdluminate}@gmail.com, lewang@mail.xjtu.edu.cn, xinbogao@mail.xidian.edu.cn +<br/>ganghua@gmail.com +</td></tr><tr><td>666300af8ffb8c903223f32f1fcc5c4674e2430b</td><td>Changing Fashion Cultures +<br/><b>National Institute of Advanced Industrial Science and Technology (AIST</b><br/>Tsukuba, Ibaraki, Japan +<br/><b>Tokyo Denki University</b><br/>Adachi, Tokyo, Japan +</td><td>('3408038', 'Kaori Abe', 'kaori abe')<br/>('5014206', 'Teppei Suzuki', 'teppei suzuki')<br/>('9935341', 'Shunya Ueta', 'shunya ueta')<br/>('1732705', 'Yutaka Satoh', 'yutaka satoh')<br/>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')<br/>('2462801', 'Akio Nakamura', 'akio nakamura')</td><td>{abe.keroko, suzuki-teppei, shunya.ueta, yu.satou, hirokatsu.kataoka}@aist.go.jp +<br/>nkmr-a@cck.dendai.ac.jp +</td></tr><tr><td>66029f1be1a5cee9a4e3e24ed8fcb65d5d293720</td><td>HWANG AND GRAUMAN: ACCOUNTING FOR IMPORTANCE IN IMAGE RETRIEVAL +<br/>Accounting for the Relative Importance of +<br/>Objects in Image Retrieval +<br/><b>The University of Texas</b><br/>Austin, TX, USA +</td><td>('35788904', 'Sung Ju Hwang', 'sung ju hwang')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>sjhwang@cs.utexas.edu +<br/>grauman@cs.utexas.edu +</td></tr><tr><td>6691dfa1a83a04fdc0177d8d70e3df79f606b10f</td><td>Illumination Modeling and Normalization for Face Recognition +<br/><b>Institute of Automation</b><br/>Chinese Academy of Sciences +<br/>Beijing, 100080, China +</td><td>('29948255', 'Haitao Wang', 'haitao wang')<br/>('34679741', 'Stan Z. Li', 'stan z. li')<br/>('1744302', 'Yangsheng Wang', 'yangsheng wang')<br/>('38248052', 'Weiwei Zhang', 'weiwei zhang')</td><td>{htwang, wys, wwzhang}@nlpr.ia.ac.cn +</td></tr><tr><td>66a2c229ac82e38f1b7c77a786d8cf0d7e369598</td><td>Proceedings of the 2016 Industrial and Systems Engineering Research Conference +<br/>H. Yang, Z. Kong, and MD Sarder, eds. +<br/>A Probabilistic Adaptive Search System +<br/>for Exploring the Face Space +<br/>Escuela Superior Politecnica del Litoral (ESPOL) +<br/>Guayaquil-Ecuador +</td><td>('3123974', 'Andres G. Abad', 'andres g. abad')<br/>('3044670', 'Luis I. Reyes Castro', 'luis i. reyes castro')</td><td></td></tr><tr><td>66886997988358847615375ba7d6e9eb0f1bb27f</td><td></td><td></td><td></td></tr><tr><td>66837add89caffd9c91430820f49adb5d3f40930</td><td></td><td></td><td></td></tr><tr><td>66a9935e958a779a3a2267c85ecb69fbbb75b8dc</td><td>FAST AND ROBUST FIXED-RANK MATRIX RECOVERY +<br/>Fast and Robust Fixed-Rank Matrix +<br/>Recovery +<br/>Antonio Lopez +</td><td>('34210410', 'Julio Guerrero', 'julio guerrero')</td><td></td></tr><tr><td>66533107f9abdc7d1cb8f8795025fc7e78eb1122</td><td>Vi a +<br/>i a Whee +<br/>W y g Sgy Dae i iy g S g iz ad Ze ga Biey +<br/>y EECS AST 373 1 g Dg Y g G Taej 305 701 REA +<br/>z VR Cee ETR 161 ajg Dg Y g G Taej 305 350 REA +<br/>Abac +<br/>Thee exi he c eaive aciviy bewee a h +<br/>a beig ad ehabi +<br/>a eae ehabi +<br/>e ad ha he bee(cid:12) f ehabi +<br/> ch a ai +<br/>eadig i e f he eeia +<br/>fied +<br/>cf ad afey f a +<br/>a +<br/>bic a ye ARES ad i h a b +<br/>ieaci ech +<br/>ech +<br/>a +<br/>vi a +<br/>ecgizig he iive ad egaive eaig f he +<br/> e i efed he bai f chage f he facia +<br/>exei a d +<br/> e iei whi +<br/> e wih a beveage. F he eÆcie vi a +<br/>i ceig +<br/>c +<br/>ed e(cid:11)ec f he bic a. The vi a +<br/>wih e(cid:11)ecive iei eadig i ccef +<br/> eve a beveage f he e. +<br/>d ci +<br/>Whee +<br/>ai he e +<br/>ca i ey ad f ci i +<br/>ye ci f a weed whee +<br/>a ad ha +<br/>he whee +<br/>he bic a ad h ake ib +<br/>exiece f a e ad a b i he ae evi +<br/>e. +<br/> hi cae he e eed ieac wih +<br/>he bic a i cfab +<br/>Fig e 1: The whee +<br/>h a b ieaci ech +<br/>eve i ha bee eed ha ay diÆc +<br/>i h a bf ieaci i exiig ehabi +<br/>b. F exa +<br/>a ake a high cgiive +<br/>hyica +<br/>eaig jyick dexe +<br/>de +<br/>ai e eed ha he diÆc +<br/>ig ehabi +<br/>a +<br/>id a he begiig [4]. Theefe h a fied +<br/>h a b ieaci i e f eeia +<br/>i a whee +<br/> hi ae we cide he whee +<br/>bic ye ARES AST Rehabi +<br/>gieeig Sevice ye which we ae deve +<br/>a a evice bic ye f he diab +<br/>e +<br/>i e Fig. 1. Ag h a b ieaci ech +<br/>i e vi a +</td><td></td><td>zbie@ee.kai.ac.k +</td></tr><tr><td>66810438bfb52367e3f6f62c24f5bc127cf92e56</td><td>Face Recognition of Illumination Tolerance in 2D +<br/>Subspace Based on the Optimum Correlation +<br/>Filter +<br/>Xu Yi +<br/>Department of Information Engineering, Hunan Industry Polytechnic, Changsha, China +<br/>images will be tested to project +</td><td></td><td></td></tr><tr><td>66af2afd4c598c2841dbfd1053bf0c386579234e</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Context Assisted Face Clustering Framework with +<br/>Human-in-the-Loop +<br/>Received: date / Accepted: date +</td><td>('3338094', 'Liyan Zhang', 'liyan zhang')<br/>('1686199', 'Sharad Mehrotra', 'sharad mehrotra')</td><td></td></tr><tr><td>66f02fbcad13c6ee5b421be2fc72485aaaf6fcb5</td><td>The AAAI-17 Workshop on +<br/>Human-Aware Artificial Intelligence +<br/>WS-17-10 +<br/>Using Co-Captured Face, Gaze and Verbal Reactions to Images of +<br/>Varying Emotional Content for Analysis and Semantic Alignment +<br/><b>Muhlenberg College</b><br/><b>Rochester Institute of Technology</b><br/><b>Rochester Institute of Technology</b></td><td>('40114708', 'Trevor Walden', 'trevor walden')<br/>('2459642', 'Preethi Vaidyanathan', 'preethi vaidyanathan')<br/>('37459359', 'Reynold Bailey', 'reynold bailey')<br/>('1695716', 'Cecilia O. Alm', 'cecilia o. alm')</td><td>ag249083@muhlenberg.edu +<br/>tjw5866@rit.edu +<br/>{pxv1621, emilypx, rjbvcs, coagla}@rit.edu +</td></tr><tr><td>66e9fb4c2860eb4a15f713096020962553696e12</td><td>A New Urban Objects Detection Framework +<br/>Using Weakly Annotated Sets +<br/><b>University of S ao Paulo - USP, S ao Paulo - Brazil</b><br/><b>New York University</b></td><td>('40014199', 'Claudio Silva', 'claudio silva')<br/>('1748049', 'Roberto M. Cesar', 'roberto m. cesar')</td><td>{keiji, gabriel.augusto.ferreira, rmcesar}@usp.br +<br/>csilva@nyu.edu +</td></tr><tr><td>66e6f08873325d37e0ec20a4769ce881e04e964e</td><td>Int J Comput Vis (2014) 108:59–81 +<br/>DOI 10.1007/s11263-013-0695-z +<br/>The SUN Attribute Database: Beyond Categories for Deeper Scene +<br/>Understanding +<br/>Received: 27 February 2013 / Accepted: 28 December 2013 / Published online: 18 January 2014 +<br/>© Springer Science+Business Media New York 2014 +</td><td>('40541456', 'Genevieve Patterson', 'genevieve patterson')<br/>('12532254', 'James Hays', 'james hays')</td><td></td></tr><tr><td>661da40b838806a7effcb42d63a9624fcd684976</td><td>53 +<br/>An Illumination Invariant Accurate +<br/>Face Recognition with Down Scaling +<br/>of DCT Coefficients +<br/>Department of Computer Science and Engineering, Amity School of Engineering and Technology, New Delhi, India +<br/>In this paper, a novel approach for illumination normal- +<br/>ization under varying lighting conditions is presented. +<br/>Our approach utilizes the fact that discrete cosine trans- +<br/>form (DCT) low-frequency coefficients correspond to +<br/>illumination variations in a digital image. Under varying +<br/>illuminations, the images captured may have low con- +<br/>trast; initially we apply histogram equalization on these +<br/>for contrast stretching. Then the low-frequency DCT +<br/>coefficients are scaled down to compensate the illumi- +<br/>nation variations. The value of scaling down factor and +<br/>the number of low-frequency DCT coefficients, which +<br/>are to be rescaled, are obtained experimentally. The +<br/>classification is done using k−nearest neighbor classi- +<br/>fication and nearest mean classification on the images +<br/>obtained by inverse DCT on the processed coefficients. +<br/>The correlation coefficient and Euclidean distance ob- +<br/>tained using principal component analysis are used as +<br/>distance metrics in classification. We have tested our +<br/>face recognition method using Yale Face Database B. +<br/>The results show that our method performs without any +<br/>error (100% face recognition performance), even on the +<br/>most extreme illumination variations. There are different +<br/>schemes in the literature for illumination normalization +<br/>under varying lighting conditions, but no one is claimed +<br/>to give 100% recognition rate under all illumination +<br/>variations for this database. The proposed technique is +<br/>computationally efficient and can easily be implemented +<br/>for real time face recognition system. +<br/>Keywords: discrete cosine transform, correlation co- +<br/>efficient, face recognition, illumination normalization, +<br/>nearest neighbor classification +<br/>1. Introduction +<br/>Two-dimensional pattern classification plays a +<br/>crucial role in real-world applications. To build +<br/>high-performance surveillance or information +<br/>security systems, face recognition has been +<br/>known as the key application attracting enor- +<br/>mous researchers highlighting on related topics +<br/>[1,2]. Even though current machine recognition +<br/>systems have reached a certain level of matu- +<br/>rity, their success is limited by the real appli- +<br/>cations constraints, like pose, illumination and +<br/>expression. The FERET evaluation shows that +<br/>the performance of a face recognition system +<br/>decline seriously with the change of pose and +<br/>illumination conditions [31]. +<br/>To solve the variable illumination problem a +<br/>variety of approaches have been proposed [3, 7- +<br/>11, 26-29]. Early work in illumination invariant +<br/>face recognition focused on image representa- +<br/>tions that are mostly insensitive to changes in +<br/>illumination. There were approaches in which +<br/>the image representations and distance mea- +<br/>sures were evaluated on a tightly controlled face +<br/>database that varied the face pose, illumination, +<br/>and expression. The image representations in- +<br/>clude edge maps, 2D Gabor-like filters, first and +<br/>second derivatives of the gray-level image, and +<br/>the logarithmic transformations of the intensity +<br/>image along with these representations [4]. +<br/>The different approaches to solve the prob- +<br/>lem of illumination invariant face recognition +<br/>can be broadly classified into two main cate- +<br/>gories. The first category is named as passive +<br/>approach in which the visual spectrum images +<br/>are analyzed to overcome this problem. The +<br/>approaches belonging to other category named +<br/>active, attempt to overcome this problem by +<br/>employing active imaging techniques to obtain +<br/>face images captured in consistent illumina- +<br/>tion condition, or images of illumination invari- +<br/>ant modalities. There is a hierarchical catego- +<br/>rization of these two approaches. An exten- +<br/>sive review of both approaches is given in [5]. +</td><td>('2650871', 'Virendra P. Vishwakarma', 'virendra p. vishwakarma')<br/>('2100294', 'Sujata Pandey', 'sujata pandey')<br/>('11690561', 'M. N. Gupta', 'm. n. gupta')</td><td></td></tr><tr><td>66886f5af67b22d14177119520bd9c9f39cdd2e6</td><td>T. KOBAYASHI: LEARNING ADDITIVE KERNEL +<br/>Learning Additive Kernel For Feature +<br/>Transformation and Its Application to CNN +<br/>Features +<br/><b>National Institute of Advanced Industrial</b><br/>Science and Technology +<br/>Tsukuba, Japan +</td><td>('1800592', 'Takumi Kobayashi', 'takumi kobayashi')</td><td>takumi.kobayashi@aist.go.jp +</td></tr><tr><td>3edb0fa2d6b0f1984e8e2c523c558cb026b2a983</td><td>Automatic Age Estimation Based on +<br/>Facial Aging Patterns +</td><td>('1735299', 'Xin Geng', 'xin geng')<br/>('1692625', 'Zhi-Hua Zhou', 'zhi-hua zhou')<br/>('2848275', 'Kate Smith-Miles', 'kate smith-miles')</td><td></td></tr><tr><td>3e69ed088f588f6ecb30969bc6e4dbfacb35133e</td><td>ACEEE Int. J. on Information Technology, Vol. 01, No. 02, Sep 2011 +<br/>Improving Performance of Texture Based Face +<br/>Recognition Systems by Segmenting Face Region +<br/><b>St. Xavier s Catholic College of Engineering, Nagercoil, India</b><br/><b>Manonmaniam Sundaranar University, Tirunelveli, India</b></td><td>('9375880', 'R. Reena Rose', 'r. reena rose')<br/>('3311251', 'A. Suruliandi', 'a. suruliandi')</td><td>mailtoreenarose@yahoo.in +<br/>suruliandi@yahoo.com +</td></tr><tr><td>3e0a1884448bfd7f416c6a45dfcdfc9f2e617268</td><td>Understanding and Controlling User Linkability in +<br/>Decentralized Learning +<br/><b>Max Planck Institute for Informatics</b><br/>Saarland Informatics Campus +<br/>Saarbrücken, Germany +</td><td>('9517443', 'Tribhuvanesh Orekondy', 'tribhuvanesh orekondy')<br/>('2390510', 'Seong Joon Oh', 'seong joon oh')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td>{orekondy,joon,schiele,mfritz}@mpi-inf.mpg.de +</td></tr><tr><td>3e4b38b0574e740dcbd8f8c5dfe05dbfb2a92c07</td><td>FACIAL EXPRESSION RECOGNITION WITH LOCAL BINARY PATTERNS +<br/>AND LINEAR PROGRAMMING +<br/>Xiaoyi Feng1, 2, Matti Pietikäinen1, Abdenour Hadid1 +<br/>1 Machine Vision Group, Infotech Oulu and Dept. of Electrical and Information Engineering +<br/><b>P. O. Box 4500 Fin-90014 University of Oulu, Finland</b><br/><b>College of Electronics and Information, Northwestern Polytechnic University</b><br/>710072 Xi’an, China +<br/>In this work, we propose a novel approach to recognize facial expressions from static +<br/>images. First, the Local Binary Patterns (LBP) are used to efficiently represent the facial +<br/>images and then the Linear Programming (LP) technique is adopted to classify the seven +<br/>facial expressions anger, disgust, fear, happiness, sadness, surprise and neutral. +<br/>Experimental results demonstrate an average recognition accuracy of 93.8% on the JAFFE +<br/>database, which outperforms the rates of all other reported methods on the same database. +<br/>Introduction +<br/>Facial expression recognition from static +<br/>images is a more challenging problem +<br/>than from image sequences because less +<br/>information for expression actions +<br/>is +<br/>available. However, information in a +<br/>single image is sometimes enough for +<br/>expression recognition, and +<br/>in many +<br/>applications it is also useful to recognize +<br/>single image’s facial expression. +<br/>In the recent years, numerous approaches +<br/>to facial expression analysis from static +<br/>images have been proposed [1] [2]. These +<br/>methods +<br/>face +<br/>representation and similarity measure. +<br/>For instance, Zhang [3] used two types of +<br/>features: the geometric position of 34 +<br/>manually selected fiducial points and a +<br/>set of Gabor wavelet coefficients at these +<br/>points. These two types of features were +<br/>used both independently and jointly with +<br/>a multi-layer perceptron for classification. +<br/>Guo and Dyer [4] also adopted a similar +<br/>face representation, combined with linear +<br/>to carry out +<br/>programming +<br/>selection +<br/>simultaneous +<br/>and +<br/>classifier +<br/>they reported +<br/>technique +<br/>feature +<br/>training, and +<br/>differ +<br/>generally +<br/>in +<br/>a +<br/>simple +<br/>imperative question +<br/>better result. Lyons et al. used a similar face +<br/>representation with +<br/>LDA-based +<br/>classification scheme [5]. All the above methods +<br/>required the manual selection of fiducial points. +<br/>Buciu et al. used ICA and Gabor representation for +<br/>facial expression recognition and reported good result +<br/>on the same database [6]. However, a suitable +<br/>combination of feature extraction and classification is +<br/>still one +<br/>for expression +<br/>recognition. +<br/>In this paper, we propose a novel method for facial +<br/>expression recognition. In the feature extraction step, +<br/>the Local Binary Pattern (LBP) operator is used to +<br/>describe facial expressions. In the classification step, +<br/>seven expressions (anger, disgust, fear, happiness, +<br/>sadness, surprise and neutral) are decomposed into 21 +<br/>expression pairs such as anger-fear, happiness- +<br/>sadness etc. 21 classifiers are produced by the Linear +<br/>Programming (LP) technique, each corresponding to +<br/>one of the 21 expression pairs. A simple binary tree +<br/>tournament scheme with pairwise comparisons is +<br/>used for classifying unknown expressions. +<br/>Face Representation with Local Binary Patterns +<br/> +<br/>Fig.1 shows the basic LBP operator [7], in which the +<br/>original 3×3 neighbourhood at the left is thresholded +<br/>by the value of the centre pixel, and a binary pattern +</td><td></td><td>{xiaoyi,mkp,hadid}@ee.oulu.fi +<br/>fengxiao@nwpu.edu.cn +</td></tr><tr><td>3ee7a8107a805370b296a53e355d111118e96b7c</td><td></td><td></td><td></td></tr><tr><td>3ebce6710135d1f9b652815e59323858a7c60025</td><td>Component-based Face Detection +<br/>(cid:1)Center for Biological and Computational Learning, M.I.T., Cambridge, MA, USA +<br/><b>cid:2)Honda RandD Americas, Inc., Boston, MA, USA</b><br/><b>University of Siena, Siena, Italy</b></td><td>('1684626', 'Bernd Heisele', 'bernd heisele')</td><td>(cid:1)heisele, serre, tp(cid:2) @ai.mit.edu pontil@dii.unisi.it +</td></tr><tr><td>3e4acf3f2d112fc6516abcdddbe9e17d839f5d9b</td><td>Deep Value Networks Learn to +<br/>Evaluate and Iteratively Refine Structured Outputs +</td><td>('3037160', 'Michael Gygli', 'michael gygli')</td><td></td></tr><tr><td>3e3f305dac4fbb813e60ac778d6929012b4b745a</td><td>Feature sampling and partitioning for visual vocabulary +<br/>generation on large action classification datasets. +<br/><b>Oxford Brookes University</b><br/><b>University of Oxford</b></td><td>('3019396', 'Michael Sapienza', 'michael sapienza')<br/>('1754181', 'Fabio Cuzzolin', 'fabio cuzzolin')</td><td></td></tr><tr><td>3ea8a6dc79d79319f7ad90d663558c664cf298d4</td><td></td><td>('40253814', 'IRA COHEN', 'ira cohen')</td><td></td></tr><tr><td>3e4f84ce00027723bdfdb21156c9003168bc1c80</td><td>1979 +<br/>© EURASIP, 2011 - ISSN 2076-1465 +<br/>19th European Signal Processing Conference (EUSIPCO 2011) +<br/>INTRODUCTION +</td><td></td><td></td></tr><tr><td>3e04feb0b6392f94554f6d18e24fadba1a28b65f</td><td>14 +<br/>Subspace Image Representation for Facial +<br/>Expression Analysis and Face Recognition +<br/>and its Relation to the Human Visual System +<br/><b>Aristotle University of Thessaloniki GR</b><br/>Thessaloniki, Box 451, Greece. +<br/>2 Electronics Department, Faculty of Electrical Engineering and Information +<br/><b>Technology, University of Oradea 410087, Universitatii 1, Romania</b><br/>Summary. Two main theories exist with respect to face encoding and representa- +<br/>tion in the human visual system (HVS). The first one refers to the dense (holistic) +<br/>representation of the face, where faces have “holon”-like appearance. The second one +<br/>claims that a more appropriate face representation is given by a sparse code, where +<br/>only a small fraction of the neural cells corresponding to face encoding is activated. +<br/>Theoretical and experimental evidence suggest that the HVS performs face analysis +<br/>(encoding, storing, face recognition, facial expression recognition) in a structured +<br/>and hierarchical way, where both representations have their own contribution and +<br/>goal. According to neuropsychological experiments, it seems that encoding for face +<br/>recognition, relies on holistic image representation, while a sparse image represen- +<br/>tation is used for facial expression analysis and classification. From the computer +<br/>vision perspective, the techniques developed for automatic face and facial expres- +<br/>sion recognition fall into the same two representation types. Like in Neuroscience, +<br/>the techniques which perform better for face recognition yield a holistic image rep- +<br/>resentation, while those techniques suitable for facial expression recognition use a +<br/>sparse or local image representation. The proposed mathematical models of image +<br/>formation and encoding try to simulate the efficient storing, organization and coding +<br/>of data in the human cortex. This is equivalent with embedding constraints in the +<br/>model design regarding dimensionality reduction, redundant information minimiza- +<br/>tion, mutual information minimization, non-negativity constraints, class informa- +<br/>tion, etc. The presented techniques are applied as a feature extraction step followed +<br/>by a classification method, which also heavily influences the recognition results. +<br/>Key words: Human Visual System; Dense, Sparse and Local Image Repre- +<br/>sentation and Encoding, Face and Facial Expression Analysis and Recogni- +<br/>tion. +<br/>R.P. W¨urtz (ed.), Organic Computing. Understanding Complex Systems, +<br/>doi: 10.1007/978-3-540-77657-4 14, © Springer-Verlag Berlin Heidelberg 2008 +</td><td>('2336758', 'Ioan Buciu', 'ioan buciu')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>pitas@zeus.csd.auth.gr +<br/>ibuciu@uoradea.ro +</td></tr><tr><td>3e685704b140180d48142d1727080d2fb9e52163</td><td>Single Image Action Recognition by Predicting +<br/>Space-Time Saliency +</td><td>('32998919', 'Marjaneh Safaei', 'marjaneh safaei')<br/>('1691260', 'Hassan Foroosh', 'hassan foroosh')</td><td></td></tr><tr><td>3e51d634faacf58e7903750f17111d0d172a0bf1</td><td>A COMPRESSIBLE TEMPLATE PROTECTION SCHEME +<br/>FOR FACE RECOGNITION BASED ON SPARSE REPRESENTATION +<br/><b>Tokyo Metropolitan University</b><br/>6–6 Asahigaoka, Hino-shi, Tokyo 191–0065, Japan +<br/>† NTT Network Innovation Laboratories, Japan +</td><td>('32403098', 'Yuichi Muraki', 'yuichi muraki')<br/>('11129971', 'Masakazu Furukawa', 'masakazu furukawa')<br/>('1728060', 'Masaaki Fujiyoshi', 'masaaki fujiyoshi')<br/>('34638424', 'Yoshihide Tonomura', 'yoshihide tonomura')<br/>('1737217', 'Hitoshi Kiya', 'hitoshi kiya')</td><td></td></tr><tr><td>3e40991ab1daa2a4906eb85a5d6a01a958b6e674</td><td>LIPNET: END-TO-END SENTENCE-LEVEL LIPREADING +<br/><b>University of Oxford, Oxford, UK</b><br/>Google DeepMind, London, UK 2 +<br/>CIFAR, Canada 3 +<br/>{yannis.assael,brendan.shillingford, +</td><td>('3365565', 'Yannis M. Assael', 'yannis m. assael')<br/>('3144580', 'Brendan Shillingford', 'brendan shillingford')<br/>('1766767', 'Shimon Whiteson', 'shimon whiteson')</td><td>shimon.whiteson,nando.de.freitas}@cs.ox.ac.uk +</td></tr><tr><td>3e687d5ace90c407186602de1a7727167461194a</td><td>Photo Tagging by Collection-Aware People Recognition +<br/>UFF +<br/>UFF +<br/>Asla S´a +<br/>FGV +<br/>IMPA +</td><td>('2901520', 'Cristina Nader Vasconcelos', 'cristina nader vasconcelos')<br/>('19264449', 'Vinicius Jardim', 'vinicius jardim')<br/>('1746637', 'Paulo Cezar Carvalho', 'paulo cezar carvalho')</td><td>crisnv@ic.uff.br +<br/>vinicius@id.uff.br +<br/>asla.sa@fgv.br +<br/>pcezar@impa.br +</td></tr><tr><td>3e3a87eb24628ab075a3d2bde3abfd185591aa4c</td><td>Effects of sparseness and randomness of +<br/>pairwise distance matrix on t-SNE results +<br/><b>BECS, Aalto University, Helsinki, Finland</b></td><td>('32430508', 'Eli Parviainen', 'eli parviainen')</td><td></td></tr><tr><td>3e207c05f438a8cef7dd30b62d9e2c997ddc0d3f</td><td>Objects as context for detecting their semantic parts +<br/><b>University of Edinburgh</b></td><td>('20758701', 'Abel Gonzalez-Garcia', 'abel gonzalez-garcia')<br/>('1996209', 'Davide Modolo', 'davide modolo')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')</td><td>a.gonzalez-garcia@sms.ed.ac.uk +<br/>davide.modolo@gmail.com +<br/>vferrari@staffmail.ed.ac.uk +</td></tr><tr><td>5040f7f261872a30eec88788f98326395a44db03</td><td>PAPAMAKARIOS, PANAGAKIS, ZAFEIRIOU: GENERALISED SCALABLE ROBUST PCA +<br/>Generalised Scalable Robust Principal +<br/>Component Analysis +<br/>Department of Computing +<br/><b>Imperial College London</b><br/>London, UK +</td><td>('2369138', 'Georgios Papamakarios', 'georgios papamakarios')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>georgios.papamakarios13@imperial.ac.uk +<br/>i.panagakis@imperial.ac.uk +<br/>s.zafeiriou@imperial.ac.uk +</td></tr><tr><td>50f0c495a214b8d57892d43110728e54e413d47d</td><td>Submitted 8/11; Revised 3/12; Published 8/12 +<br/>Pairwise Support Vector Machines and their Application to Large +<br/>Scale Problems +<br/><b>Institute for Numerical Mathematics</b><br/>Technische Universit¨at Dresden +<br/>01062 Dresden, Germany +<br/>Cognitec Systems GmbH +<br/>Grossenhainer Str. 101 +<br/>01127 Dresden, Germany +<br/>Editor: Corinna Cortes +</td><td>('25796572', 'Carl Brunner', 'carl brunner')<br/>('1833903', 'Andreas Fischer', 'andreas fischer')<br/>('2201239', 'Klaus Luig', 'klaus luig')<br/>('2439730', 'Thorsten Thies', 'thorsten thies')</td><td>C.BRUNNER@GMX.NET +<br/>ANDREAS.FISCHER@TU-DRESDEN.DE +<br/>LUIG@COGNITEC.COM +<br/>THIES@COGNITEC.COM +</td></tr><tr><td>501096cca4d0b3d1ef407844642e39cd2ff86b37</td><td>Illumination Invariant Face Image +<br/>Representation using Quaternions +<br/>Dayron Rizo-Rodr´ıguez, Heydi M´endez-V´azquez, and Edel Garc´ıa-Reyes +<br/>Advanced Technologies Application Center. 7a # 21812 b/ 218 and 222, +<br/>Rpto. Siboney, Playa, P.C. 12200, La Habana, Cuba. +</td><td></td><td>{drizo,hmendez,egarcia}@cenatav.co.cu +</td></tr><tr><td>500fbe18afd44312738cab91b4689c12b4e0eeee</td><td>ChaLearn Looking at People 2015 new competitions: +<br/>Age Estimation and Cultural Event Recognition +<br/><b>University of Barcelona</b><br/>Computer Vision Center, UAB +<br/>Jordi Gonz`alez +<br/>Xavier Bar´o +<br/>Univ. Aut`onoma de Barcelona +<br/>Computer Vision Center, UAB +<br/>Universitat Oberta de Catalunya +<br/>Computer Vision Center, UAB +<br/><b>University of Barcelona</b><br/>Univ. Aut`onoma de Barcelona +<br/>Computer Vision Center, UAB +<br/><b>University of Barcelona</b><br/>Computer Vision Center, UAB +<br/>INAOE +<br/>Ivan Huerta +<br/><b>University of Venezia</b><br/>Clopinet, Berkeley +</td><td>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('40378482', 'Pablo Pardo', 'pablo pardo')<br/>('37811966', 'Junior Fabian', 'junior fabian')<br/>('3305641', 'Marc Oliu', 'marc oliu')<br/>('1742688', 'Hugo Jair Escalante', 'hugo jair escalante')<br/>('1743797', 'Isabelle Guyon', 'isabelle guyon')</td><td>Email: sergio@maia.ub.es +<br/>Email: ppardoga7@gmail.com +<br/>Email: poal@cvc.uab.es +<br/>Email: xbaro@uoc.edu +<br/>Email: jfabian@cvc.uab.es +<br/>Email: moliusimon@gmail.com +<br/>Email: hugo.jair@gmail.com +<br/>Email: huertacasado@iuav.it +<br/>Email: guyon@chalearn.org +</td></tr><tr><td>501eda2d04b1db717b7834800d74dacb7df58f91</td><td></td><td>('3846862', 'Pedro Miguel Neves Marques', 'pedro miguel neves marques')</td><td></td></tr><tr><td>5083c6be0f8c85815ead5368882b584e4dfab4d1</td><td> Please do not quote. In press, Handbook of affective computing. New York, NY: Oxford +<br/>Automated Face Analysis for Affective Computing +</td><td>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td></td></tr><tr><td>506c2fbfa9d16037d50d650547ad3366bb1e1cde</td><td>Convolutional Channel Features: Tailoring CNN to Diverse Tasks +<br/>Junjie Yan +<br/>Zhen Lei +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences, China</b></td><td>('1716231', 'Bin Yang', 'bin yang')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{zlei, szli}@nlpr.ia.ac.cn +<br/>{yb.derek, yanjjie}@gmail.com +</td></tr><tr><td>500b92578e4deff98ce20e6017124e6d2053b451</td><td></td><td></td><td></td></tr><tr><td>504028218290d68859f45ec686f435f473aa326c</td><td>Multi-Fiber Networks for Video Recognition +<br/><b>National University of Singapore</b><br/>2 Facebook Research +<br/><b>Qihoo 360 AI Institute</b></td><td>('1713312', 'Yunpeng Chen', 'yunpeng chen')<br/>('1944225', 'Yannis Kalantidis', 'yannis kalantidis')<br/>('2757639', 'Jianshu Li', 'jianshu li')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')</td><td>{chenyunpeng, jianshu}@u.nus.edu, yannisk@fb.com, +<br/>{eleyans, elefjia}@nus.edu.sg +</td></tr><tr><td>5058a7ec68c32984c33f357ebaee96c59e269425</td><td>A Comparative Evaluation of Regression Learning +<br/>Algorithms for Facial Age Estimation +<br/>1 Herta Security +<br/>Pau Claris 165 4-B, 08037 Barcelona, Spain +<br/><b>DPDCE, University IUAV</b><br/>Santa Croce 1957, 30135 Venice, Italy +</td><td>('1733945', 'Andrea Prati', 'andrea prati')</td><td>carles.fernandez@hertasecurity.com +<br/>huertacasado@iuav.it, aprati@iuav.it +</td></tr><tr><td>50ff21e595e0ebe51ae808a2da3b7940549f4035</td><td>IEEE TRANSACTIONS ON LATEX CLASS FILES, VOL. XX, NO. X, AUGUST 2017 +<br/>Age Group and Gender Estimation in the Wild with +<br/>Deep RoR Architecture +</td><td>('32164792', 'Ke Zhang', 'ke zhang')<br/>('35038034', 'Ce Gao', 'ce gao')<br/>('3451321', 'Liru Guo', 'liru guo')<br/>('2598874', 'Miao Sun', 'miao sun')<br/>('3451660', 'Xingfang Yuan', 'xingfang yuan')<br/>('3244463', 'Tony X. Han', 'tony x. han')<br/>('2626320', 'Zhenbing Zhao', 'zhenbing zhao')<br/>('2047712', 'Baogang Li', 'baogang li')</td><td></td></tr><tr><td>5042b358705e8d8e8b0655d07f751be6a1565482</td><td>International Journal of +<br/>Emerging Research in Management &Technology +<br/>ISSN: 2278-9359 (Volume-4, Issue-8) +<br/>Research Article +<br/> August +<br/> 2015 +<br/>Review on Emotion Detection in Image +<br/>CSE & PCET, PTU HOD, CSE & PCET, PTU +<br/> Punjab, India Punj ab, India +</td><td></td><td></td></tr><tr><td>50e47857b11bfd3d420f6eafb155199f4b41f6d7</td><td>International Journal of Computer, Consumer and Control (IJ3C), Vol. 2, No.1 (2013) +<br/>3D Human Face Reconstruction Using a Hybrid of Photometric +<br/>Stereo and Independent Component Analysis +</td><td>('1734467', 'Cheng-Jian Lin', 'cheng-jian lin')<br/>('3318507', 'Shyi-Shiun Kuo', 'shyi-shiun kuo')<br/>('18305737', 'Hsueh-Yi Lin', 'hsueh-yi lin')<br/>('2911354', 'Cheng-Yi Yu', 'cheng-yi yu')</td><td></td></tr><tr><td>50eb75dfece76ed9119ec543e04386dfc95dfd13</td><td>Learning Visual Entities and their Visual Attributes from Text Corpora +<br/>Dept. of Computer Science +<br/>K.U.Leuven, Belgium +<br/>Dept. of Computer Science +<br/>K.U.Leuven, Belgium +<br/>Dept. of Computer Science +<br/>K.U.Leuven, Belgium +</td><td>('2955093', 'Erik Boiy', 'erik boiy')<br/>('1797588', 'Koen Deschacht', 'koen deschacht')<br/>('1802161', 'Marie-Francine Moens', 'marie-francine moens')</td><td>erik.boiy@cs.kuleuven.be +<br/>koen.deschacht@cs.kuleuven.be +<br/>sien.moens@cs.kuleuven.be +</td></tr><tr><td>5050807e90a925120cbc3a9cd13431b98965f4b9</td><td>To appear in the ECCV Workshop on Parts and Attributes, Oct. 2012. +<br/>Unsupervised Learning of Discriminative +<br/>Relative Visual Attributes +<br/><b>Boston University</b><br/><b>Hacettepe University</b></td><td>('2863531', 'Shugao Ma', 'shugao ma')<br/>('2011587', 'Nazli Ikizler-Cinbis', 'nazli ikizler-cinbis')</td><td></td></tr><tr><td>50a0930cb8cc353e15a5cb4d2f41b365675b5ebf</td><td></td><td></td><td></td></tr><tr><td>508702ed2bf7d1b0655ea7857dd8e52d6537e765</td><td>ZUO, ORGANISCIAK, SHUM, YANG: SST-VLAD AND SST-FV FOR VAR +<br/>Saliency-Informed Spatio-Temporal Vector +<br/>of Locally Aggregated Descriptors and +<br/>Fisher Vectors for Visual Action Recognition +<br/>Department of Computer and +<br/>Information Sciences +<br/><b>Northumbria University</b><br/>Newcastle upon Tyne, NE1 8ST, UK +</td><td>('40760781', 'Zheming Zuo', 'zheming zuo')<br/>('34975328', 'Daniel Organisciak', 'daniel organisciak')<br/>('2840036', 'Hubert P. H. Shum', 'hubert p. h. shum')<br/>('1706028', 'Longzhi Yang', 'longzhi yang')</td><td>zheming.zuo@northumbria.ac.uk +<br/>daniel.organisciak@northumbria.ac.uk +<br/>hubert.shum@northumbria.ac.uk +<br/>longzhi.yang@northumbria.ac.uk +</td></tr><tr><td>50eb2ee977f0f53ab4b39edc4be6b760a2b05f96</td><td>Australian Journal of Basic and Applied Sciences, 11(5) April 2017, Pages: 1-11 +<br/>AUSTRALIAN JOURNAL OF BASIC AND +<br/>APPLIED SCIENCES +<br/>ISSN:1991-8178 EISSN: 2309-8414 +<br/>Journal home page: www.ajbasweb.com +<br/>Emotion Recognition Based on Texture Analysis of Facial Expressions +<br/>Using Wavelets Transform +<br/>1Suhaila N. Mohammed and 2Loay E. George +<br/><b>Assistant Lecturer, College of Science, Baghdad University, Baghdad, Iraq</b><br/><b>College of Science, Baghdad University, Baghdad, Iraq</b><br/>Address For Correspondence: +<br/><b>Suhaila N. Mohammed, Baghdad University, College of Science, Baghdad, Iraq</b><br/>A R T I C L E I N F O +<br/>Article history: +<br/>Received 18 January 2017 +<br/>Accepted 28 March 2017 +<br/>Available online 15 April 2017 +<br/>Keywords: +<br/>Facial Emotion, Face Detection, +<br/>Template Based Methods, Texture +<br/>Based Features, Haar Wavelets +<br/>Transform, Image Blocking, Neural +<br/>Network. +<br/>A B S T R A C T +<br/>Background: The interests toward developing accurate automatic facial emotion +<br/>recognition methodologies are growing vastly and still an ever growing research field in +<br/>the region of computer vision, artificial intelligent and automation. Auto emotion +<br/>detection systems are demanded in various fields such as medicine, education, driver +<br/>safety, games, etc. Despite the importance of this issue it still remains an unsolved +<br/>problem Objective: In this paper a facial based emotion recognition system is +<br/>introduced. Template based method is used for face region extraction by exploiting +<br/>human knowledge about face components and the corresponding symmetry property. +<br/>The system is based on texture features to work as identical feature vector. These +<br/>features are extracted from face region through using Haar wavelets transform and +<br/>blocking idea by calculating the energy of each block The feed forward neural network +<br/>classifier is used for classification task. The network is trained using a training set of +<br/>samples, and then the generated weights are used to test the recognition ability of the +<br/>system. Results: JAFFE public dataset is used for system evaluation purpose; it holds +<br/>213 facial samples for seven basic emotions. The conducted tests on the developed +<br/>system gave accuracy around 90.05% when the number of blocks is set 4x4. +<br/>Conclusion: This result is considered the highest when compared with the results of +<br/>other newly published works, especially those based on texture features in which +<br/>blocking idea allows the extraction of statistical features according to local energy of +<br/>each block; this gave chance for more features to work more effectively. +<br/>INTRODUCTION +<br/>Due to the rapid development of technologies, it is being required to build a smart system for understanding +<br/>human emotion (Ruivo et al., 2016). There are different ways to distinguish person emotions such as facial +<br/>image, voice, shape of body and others. Mehrabian explained that person impression can be expressed through +<br/>words (verbal part) by 7%, and 38% through tone of voice (vocal part) while the facial image can give the +<br/>largest rate which reaches to 55% (Rani and Garg, 2014). Also, he indicated that one of the most important ways +<br/>to display emotions is through facial expressions; where facial image contains much information (such as, +<br/>person's identification and also about mood and state of mind) which can be used to distinguish human +<br/>inspiration (Saini and Rana, 2014). +<br/>Facial emotion recognition is an active area of research with several fields of applications. Some of the +<br/>significant applications are: feedback system for e-learning, alert system for driving, social robot emotion +<br/>recognition system, medical practices...etc (Dubey and Singh, 2016). +<br/>Human emotion is composed of thousands of expressions but in the last decade the focus on analyzing only +<br/>seven basic facial expressions such as happiness, sadness, surprise, disgust, fear, natural, and anger (Singh and +<br/>Open Access Journal +<br/>Published BY AENSI Publication +<br/>© 2017 AENSI Publisher All rights reserved +<br/>This work is licensed under the Creative Commons Attribution International License (CC BY). +<br/>http://creativecommons.org/licenses/by/4.0/ +<br/>To Cite This Article: Suhaila N. Mohammed and Loay E. George., Emotion Recognition Based on Texture Analysis of Facial Expressions +<br/>Using Wavelets Transform. Aust. J. Basic & Appl. Sci., 11(5): 1-11, 2017 +</td><td></td><td></td></tr><tr><td>50e45e9c55c9e79aaae43aff7d9e2f079a2d787b</td><td>Hindawi Publishing Corporation +<br/>e Scientific World Journal +<br/>Volume 2015, Article ID 471371, 18 pages +<br/>http://dx.doi.org/10.1155/2015/471371 +<br/>Research Article +<br/>Unbiased Feature Selection in Learning Random Forests for +<br/>High-Dimensional Data +<br/><b>Shenzhen Key Laboratory of High Performance Data Mining, Shenzhen Institutes of Advanced Technology</b><br/>Chinese Academy of Sciences, Shenzhen 518055, China +<br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b><br/><b>School of Computer Science and Engineering, Water Resources University, Hanoi 10000, Vietnam</b><br/><b>College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, China</b><br/><b>Faculty of Information Technology, Vietnam National University of Agriculture, Hanoi 10000, Vietnam</b><br/>Received 20 June 2014; Accepted 20 August 2014 +<br/>Academic Editor: Shifei Ding +<br/>License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly +<br/>cited. +<br/>Random forests (RFs) have been widely used as a powerful classification method. However, with the randomization in both bagging +<br/>samples and feature selection, the trees in the forest tend to select uninformative features for node splitting. This makes RFs +<br/>have poor accuracy when working with high-dimensional data. Besides that, RFs have bias in the feature selection process where +<br/>multivalued features are favored. Aiming at debiasing feature selection in RFs, we propose a new RF algorithm, called xRF, to select +<br/>good features in learning RFs for high-dimensional data. We first remove the uninformative features using 𝑝-value assessment, +<br/>and the subset of unbiased features is then selected based on some statistical measures. This feature subset is then partitioned into +<br/>two subsets. A feature weighting sampling technique is used to sample features from these two subsets for building trees. This +<br/>approach enables one to generate more accurate trees, while allowing one to reduce dimensionality and the amount of data needed +<br/>for learning RFs. An extensive set of experiments has been conducted on 47 high-dimensional real-world datasets including image +<br/>datasets. The experimental results have shown that RFs with the proposed approach outperformed the existing random forests in +<br/>increasing the accuracy and the AUC measures. +<br/>1. Introduction +<br/>Random forests (RFs) [1] are a nonparametric method that +<br/>builds an ensemble model of decision trees from random +<br/>subsets of features and bagged samples of the training data. +<br/>RFs have shown excellent performance for both clas- +<br/>sification and regression problems. RF model works well +<br/>even when predictive features contain irrelevant features +<br/>(or noise); it can be used when the number of features is +<br/>much larger than the number of samples. However, with +<br/>randomizing mechanism in both bagging samples and feature +<br/>selection, RFs could give poor accuracy when applied to high +<br/>dimensional data. The main cause is that, in the process of +<br/>growing a tree from the bagged sample data, the subspace +<br/>of features randomly sampled from thousands of features to +<br/>split a node of the tree is often dominated by uninformative +<br/>features (or noise), and the tree grown from such bagged +<br/>subspace of features will have a low accuracy in prediction +<br/>which affects the final prediction of the RFs. Furthermore, +<br/>Breiman et al. noted that feature selection is biased in the +<br/>classification and regression tree (CART) model because it is +<br/>based on an information criteria, called multivalue problem +<br/>[2]. It tends in favor of features containing more values, even if +<br/>these features have lower importance than other ones or have +<br/>no relationship with the response feature (i.e., containing +<br/>less missing values, many categorical or distinct numerical +<br/>values) [3, 4]. +<br/>In this paper, we propose a new random forests algo- +<br/>rithm using an unbiased feature sampling method to build +<br/>a good subspace of unbiased features for growing trees. +</td><td>('40538635', 'Thanh-Tung Nguyen', 'thanh-tung nguyen')<br/>('8192216', 'Joshua Zhexue Huang', 'joshua zhexue huang')<br/>('39340373', 'Thuy Thi Nguyen', 'thuy thi nguyen')<br/>('40538635', 'Thanh-Tung Nguyen', 'thanh-tung nguyen')</td><td>Correspondence should be addressed to Thanh-Tung Nguyen; tungnt@wru.vn +</td></tr><tr><td>5003754070f3a87ab94a2abb077c899fcaf936a6</td><td>Evaluation of LC-KSVD on UCF101 Action Dataset +<br/><b>University of Maryland, College Park</b><br/>2Noah’s Ark Lab, Huawei Technologies +</td><td>('3146162', 'Hyunjong Cho', 'hyunjong cho')<br/>('2445131', 'Hyungtae Lee', 'hyungtae lee')<br/>('34145947', 'Zhuolin Jiang', 'zhuolin jiang')</td><td>cho@cs.umd.edu, htlee@umd.edu, zhuolin.jiang@huawei.com +</td></tr><tr><td>503db524b9a99220d430e741c44cd9c91ce1ddf8</td><td>Who’s Better, Who’s Best: Skill Determination in Video using Deep Ranking +<br/><b>University of Bristol, Bristol, UK</b><br/>Walterio Mayol-Cuevas +</td><td>('28798386', 'Hazel Doughty', 'hazel doughty')<br/>('1728459', 'Dima Damen', 'dima damen')</td><td><Firstname>.<Surname>@bristol.ac.uk +</td></tr><tr><td>50d15cb17144344bb1879c0a5de7207471b9ff74</td><td>Divide, Share, and Conquer: Multi-task +<br/>Attribute Learning with Selective Sharing +</td><td>('3197570', 'Chao-Yeh Chen', 'chao-yeh chen')<br/>('2228235', 'Dinesh Jayaraman', 'dinesh jayaraman')<br/>('1693054', 'Fei Sha', 'fei sha')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>50d961508ec192197f78b898ff5d44dc004ef26d</td><td>International Journal of Computer science & Information Technology (IJCSIT), Vol 1, No 2, November 2009 +<br/>A LOW INDEXED CONTENT BASED +<br/>NEURAL NETWORK APPROACH FOR +<br/>NATURAL OBJECTS RECOGNITION +<br/>1Research Scholar, JNTUH, Hyderabad, AP. India +<br/><b>Principal, JNTUH College of Engineering, jagitial, Karimnagar, AP, India</b><br/><b>Principal, Chaithanya Institute of Engineering and Technology, Kakinada, AP, India</b></td><td></td><td> shyam_gunda2002@yahoo.co.in +<br/>govardhan_cse@yahoo.co.in +<br/>tv_venkat@yahoo.com +</td></tr><tr><td>50ccc98d9ce06160cdf92aaf470b8f4edbd8b899</td><td>Towards Robust Cascaded Regression for Face Alignment in the Wild +<br/>J¨urgen Beyerer2,1 +<br/><b>Vision and Fusion Laboratory (IES), Karlsruhe Institute of Technology (KIT</b><br/><b>Fraunhofer Institute of Optronics, System Technologies and Image Exploitation (Fraunhofer IOSB</b><br/>3Signal Processing Laboratory (LTS5), ´Ecole Polytechnique F´ed´erale de Lausanne (EPFL) +</td><td>('1797975', 'Chengchao Qu', 'chengchao qu')<br/>('1697965', 'Hua Gao', 'hua gao')<br/>('2233872', 'Eduardo Monari', 'eduardo monari')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')</td><td>firstname.lastname@iosb.fraunhofer.de +<br/>firstname.lastname@epfl.ch +</td></tr><tr><td>5028c0decfc8dd623c50b102424b93a8e9f2e390</td><td>Published as a conference paper at ICLR 2017 +<br/>REVISITING CLASSIFIER TWO-SAMPLE TESTS +<br/>1Facebook AI Research, 2WILLOW project team, Inria / ENS / CNRS +</td><td>('3016461', 'David Lopez-Paz', 'david lopez-paz')<br/>('2093491', 'Maxime Oquab', 'maxime oquab')</td><td>dlp@fb.com, maxime.oquab@inria.fr +</td></tr><tr><td>505e55d0be8e48b30067fb132f05a91650666c41</td><td>A Model of Illumination Variation for Robust Face Recognition +<br/>Institut Eur´ecom +<br/>Multimedia Communications Department +<br/>BP 193, 06904 Sophia Antipolis Cedex, France +</td><td>('1723883', 'Florent Perronnin', 'florent perronnin')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td>fflorent.perronnin, jean-luc.dugelayg@eurecom.fr +</td></tr><tr><td>507c9672e3673ed419075848b4b85899623ea4b0</td><td>Faculty of Informatics +<br/><b>Institute for Anthropomatics</b><br/>Chair Prof. Dr.-Ing. R. Stiefelhagen +<br/>Facial Image Processing and Analysis Group +<br/>Multi-View Facial Expression +<br/>Classification +<br/>ADVISORS +<br/>MARCH 2011 +<br/><b>KIT University of the State of Baden-W rttemberg and National Laboratory of the Helmholtz Association</b><br/>www.kit.edu +</td><td>('33357889', 'Nikolas Hesse', 'nikolas hesse')<br/>('38113750', 'Hua Gao', 'hua gao')<br/>('40303076', 'Tobias Gehrig', 'tobias gehrig')</td><td></td></tr><tr><td>50c0de2cccf7084a81debad5fdb34a9139496da0</td><td>ORIGINAL RESEARCH +<br/>published: 30 November 2016 +<br/>doi: 10.3389/fict.2016.00027 +<br/>The Influence of Annotation, Corpus +<br/>Design, and Evaluation on the +<br/>Outcome of Automatic Classification +<br/>of Human Emotions +<br/><b>Institute of Neural Information Processing, Ulm University, Ulm, Germany</b><br/>The integration of emotions into human–computer interaction applications promises a +<br/>more natural dialog between the user and the technical system operators. In order +<br/>to construct such machinery, continuous measuring of the affective state of the user +<br/>becomes essential. While basic research that is aimed to capture and classify affective +<br/>signals has progressed, many issues are still prevailing that hinder easy integration +<br/>of affective signals into human–computer interaction. In this paper, we identify and +<br/>investigate pitfalls in three steps of the work-flow of affective classification studies. It starts +<br/>with the process of collecting affective data for the purpose of training suitable classifiers. +<br/>Emotional data have to be created in which the target emotions are present. Therefore, +<br/>human participants have to be stimulated suitably. We discuss the nature of these stimuli, +<br/>their relevance to human–computer interaction, and the repeatability of the data recording +<br/>setting. Second, aspects of annotation procedures are investigated, which include the +<br/>variances of +<br/>individual raters, annotation delay, the impact of the used annotation +<br/>tool, and how individual ratings are combined to a unified label. Finally, the evaluation +<br/>protocol +<br/>is examined, which includes, among others, the impact of the performance +<br/>measure on the accuracy of a classification model. We hereby focus especially on the +<br/>evaluation of classifier outputs against continuously annotated dimensions. Together with +<br/>the discussed problems and pitfalls and the ways how they affect the outcome, we +<br/>provide solutions and alternatives to overcome these issues. As the final part of the paper, +<br/>we sketch a recording scenario and a set of supporting technologies that can contribute +<br/>to solve many of the issues mentioned above. +<br/>Keywords: affective computing, affective labeling, human–computer interaction, performance measures, machine +<br/>guided labeling +<br/>1. INTRODUCTION +<br/>The integration of affective signals into human–computer interaction (HCI) is generally considered +<br/>beneficial to improve the interaction process (Picard, 2000). The analysis of affective data in HCI +<br/>can be considered both cumbersome and prone to errors. The main reason for this is that the +<br/>important steps in affective classification are particularly difficult. This includes difficulties that arise +<br/>in the recording of suitable data collections comprising episodes of affective HCI, in the uncertainty +<br/>and subjectivity of the annotations of these data, and finally in the evaluation protocol that should +<br/>account for the continuous nature of the application. +<br/>Edited by: +<br/>Anna Esposito, +<br/>Seconda Università degli Studi di +<br/>Napoli, Italy +<br/>Reviewed by: +<br/>Anna Pribilova, +<br/><b>Slovak University of Technology in</b><br/>Bratislava, Slovakia +<br/>Alda Troncone, +<br/>Seconda Università degli Studi di +<br/>Napoli, Italy +<br/>*Correspondence: +<br/>contributed equally to this work. +<br/>Specialty section: +<br/>This article was submitted to +<br/>Human-Media Interaction, a section +<br/>of the journal Frontiers in ICT +<br/>Received: 15 May 2016 +<br/>Accepted: 26 October 2016 +<br/>Published: 30 November 2016 +<br/>Citation: +<br/>Kächele M, Schels M and +<br/>Schwenker F (2016) The Influence of +<br/>Annotation, Corpus Design, and +<br/>Evaluation on the Outcome of +<br/>Automatic Classification of Human +<br/>Emotions. +<br/>doi: 10.3389/fict.2016.00027 +<br/>Frontiers in ICT | www.frontiersin.org +<br/>November 2016 | Volume 3 | Article 27 +</td><td>('2144395', 'Markus Kächele', 'markus kächele')<br/>('3037635', 'Martin Schels', 'martin schels')<br/>('1685857', 'Friedhelm Schwenker', 'friedhelm schwenker')<br/>('2144395', 'Markus Kächele', 'markus kächele')<br/>('2144395', 'Markus Kächele', 'markus kächele')<br/>('3037635', 'Martin Schels', 'martin schels')</td><td>markus.kaechele@uni-ulm.de +</td></tr><tr><td>680d662c30739521f5c4b76845cb341dce010735</td><td>Int J Comput Vis (2014) 108:82–96 +<br/>DOI 10.1007/s11263-014-0716-6 +<br/>Part and Attribute Discovery from Relative Annotations +<br/>Received: 25 February 2013 / Accepted: 14 March 2014 / Published online: 26 April 2014 +<br/>© Springer Science+Business Media New York 2014 +</td><td>('35208858', 'Subhransu Maji', 'subhransu maji')</td><td></td></tr><tr><td>68f89c1ee75a018c8eff86e15b1d2383c250529b</td><td>Final Report for Project Localizing Objects and +<br/>Actions in Videos Using Accompanying Text +<br/><b>Johns Hopkins University, Center for Speech and Language Processing</b><br/>Summer Workshop 2010 +<br/>J. Neumann, StreamSage/Comcast +<br/><b>F.Ferraro, University of Rochester</b><br/><b>H. He, Honkong Polytechnic University</b><br/><b>Y. Li, University of Maryland</b><br/><b>C.L. Teo, University of Maryland</b><br/>November 4, 2010 +</td><td>('3167986', 'C. Fermueller', 'c. fermueller')<br/>('1743020', 'J. Kosecka', 'j. kosecka')<br/>('2601166', 'E. Tzoukermann', 'e. tzoukermann')<br/>('2995090', 'R. Chaudhry', 'r. chaudhry')<br/>('1937619', 'I. Perera', 'i. perera')<br/>('9133363', 'B. Sapp', 'b. sapp')<br/>('38873583', 'G. Singh', 'g. singh')<br/>('1870728', 'X. Yi', 'x. yi')</td><td></td></tr><tr><td>68a2ee5c5b76b6feeb3170aaff09b1566ec2cdf5</td><td>AGE CLASSIFICATION BASED ON +<br/>SIMPLE LBP TRANSITIONS +<br/><b>Aditya institute of Technology and Management, Tekkalli-532 201, A.P</b><br/>2Dr. V.Vijaya Kumar +<br/>3A. Obulesu +<br/>2Dean-Computer Sciences (CSE & IT), Anurag Group of Institutions, Hyderabad – 500088, A.P., India., +<br/> 3Asst. Professor, Dept. Of CSE, Anurag Group of Institutions, Hyderabad – 500088, A.P., India. +</td><td>('34964075', 'Satyanarayana Murty', 'satyanarayana murty')</td><td>India, 1gsn_73@yahoo.co.in +<br/>2drvvk144@gmail.com +<br/>3obulesh.a@gmail.com +</td></tr><tr><td>68d2afd8c5c1c3a9bbda3dd209184e368e4376b9</td><td>Representation Learning by Rotating Your Faces +</td><td>('1849929', 'Luan Tran', 'luan tran')<br/>('2399004', 'Xi Yin', 'xi yin')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td></td></tr><tr><td>68a3f12382003bc714c51c85fb6d0557dcb15467</td><td></td><td></td><td></td></tr><tr><td>6859b891a079a30ef16f01ba8b85dc45bd22c352</td><td>International Journal of Emerging Technology and Advanced Engineering +<br/>Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 4, Issue 10, October 2014) +<br/>2D Face Recognition Based on PCA & Comparison of +<br/>Manhattan Distance, Euclidean Distance & Chebychev +<br/>Distance +<br/><b>RCC Institute of Information Technology, Kolkata, India</b></td><td>('2467416', 'Rajib Saha', 'rajib saha')<br/>('2144187', 'Sayan Barman', 'sayan barman')</td><td></td></tr><tr><td>68d08ed9470d973a54ef7806318d8894d87ba610</td><td>Drive Video Analysis for the Detection of Traffic Near-Miss Incidents +</td><td>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')<br/>('5014206', 'Teppei Suzuki', 'teppei suzuki')<br/>('6881850', 'Shoko Oikawa', 'shoko oikawa')<br/>('1720770', 'Yasuhiro Matsui', 'yasuhiro matsui')<br/>('1732705', 'Yutaka Satoh', 'yutaka satoh')</td><td></td></tr><tr><td>68caf5d8ef325d7ea669f3fb76eac58e0170fff0</td><td></td><td></td><td></td></tr><tr><td>68003e92a41d12647806d477dd7d20e4dcde1354</td><td>ISSN: 0976-9102 (ONLINE) +<br/>DOI: 10.21917/ijivp.2013.0101 +<br/> ICTACT JOURNAL ON IMAGE AND VIDEO PROCESSING, NOVEMBER 2013, VOLUME: 04, ISSUE: 02 +<br/>FUZZY BASED IMAGE DIMENSIONALITY REDUCTION USING SHAPE +<br/>PRIMITIVES FOR EFFICIENT FACE RECOGNITION +<br/>1Deprtment of Computer Science and Engineering, Nalla Narasimha Reddy Education Society’s Group of Institutions, India +<br/><b>Deprtment of Computer Science and Engineering, JNTUA College of Engineering, India</b><br/>3Deprtment of Computer Science and Engineering, Anurag Group of Institutions, India +</td><td>('2086540', 'P. Chandra', 'p. chandra')<br/>('2803943', 'B. Eswara Reddy', 'b. eswara reddy')<br/>('36754879', 'Vijaya Kumar', 'vijaya kumar')</td><td>E-Mail: pchandureddy@yahoo.com +<br/>E-mail: eswarcsejntu@gmail.com +<br/>E-mail: vijayvakula@yahoo.com +</td></tr><tr><td>68d4056765c27fbcac233794857b7f5b8a6a82bf</td><td>Example-Based Face Shape Recovery Using the +<br/>Zenith Angle of the Surface Normal +<br/>Mario Castel´an1, Ana J. Almaz´an-Delf´ın2, Marco I. Ram´ırez-Sosa-Mor´an3, +<br/>and Luz A. Torres-M´endez1 +<br/>1 CINVESTAV Campus Saltillo, Ramos Arizpe 25900, Coahuila, M´exico +<br/>2 Universidad Veracruzana, Facultad de F´ısica e Inteligencia Artificial, Xalapa 91000, +<br/>3 ITESM, Campus Saltillo, Saltillo 25270, Coahuila, M´exico +<br/>Veracruz, M´exico +</td><td></td><td>mario.castelan@cinvestav.edu.mx +</td></tr><tr><td>684f5166d8147b59d9e0938d627beff8c9d208dd</td><td>IEEE TRANS. NNLS, JUNE 2017 +<br/>Discriminative Block-Diagonal Representation +<br/>Learning for Image Recognition +</td><td>('38448016', 'Zheng Zhang', 'zheng zhang')<br/>('40065614', 'Yong Xu', 'yong xu')<br/>('40799321', 'Ling Shao', 'ling shao')<br/>('49500178', 'Jian Yang', 'jian yang')</td><td></td></tr><tr><td>68c5238994e3f654adea0ccd8bca29f2a24087fc</td><td>PLSA-BASED ZERO-SHOT LEARNING +<br/>Centre of Image and Signal Processing +<br/>Faculty of Computer Science & Information Technology +<br/><b>University of Malaya, 50603 Kuala Lumpur, Malaysia</b></td><td>('2800072', 'Wai Lam Hoo', 'wai lam hoo')<br/>('2863960', 'Chee Seng Chan', 'chee seng chan')</td><td>{wailam88@siswa.um.edu.my; cs.chan@um.edu.my} +</td></tr><tr><td>68cf263a17862e4dd3547f7ecc863b2dc53320d8</td><td></td><td></td><td></td></tr><tr><td>68e9c837431f2ba59741b55004df60235e50994d</td><td>Detecting Faces Using Region-based Fully +<br/>Convolutional Networks +<br/>Tencent AI Lab, China +</td><td>('1996677', 'Yitong Wang', 'yitong wang')</td><td>{yitongwang,denisji,encorezhou,hawelwang,michaelzfli}@tencent.com +</td></tr><tr><td>685f8df14776457c1c324b0619c39b3872df617b</td><td>Master of Science Thesis in Electrical Engineering +<br/><b>Link ping University</b><br/>Face Recognition with +<br/>Preprocessing and Neural +<br/>Networks +</td><td></td><td></td></tr><tr><td>68484ae8a042904a95a8d284a7f85a4e28e37513</td><td>Spoofing Deep Face Recognition with Custom Silicone Masks +<br/>S´ebastien Marcel +<br/><b>Idiap Research Institute. Centre du Parc, Rue Marconi 19, Martigny (VS), Switzerland</b></td><td>('1952348', 'Sushil Bhattacharjee', 'sushil bhattacharjee')</td><td>{sushil.bhattacharjee; amir.mohammadi; sebastien.marcel}@idiap.ch +</td></tr><tr><td>687e17db5043661f8921fb86f215e9ca2264d4d2</td><td>A Robust Elastic and Partial Matching Metric for Face Recognition +<br/>Microsoft Corporate +<br/>One Microsoft Way, Redmond, WA 98052 +</td><td>('1745420', 'Gang Hua', 'gang hua')<br/>('33474090', 'Amir Akbarzadeh', 'amir akbarzadeh')</td><td>{ganghua, amir}@microsoft.com +</td></tr><tr><td>688754568623f62032820546ae3b9ca458ed0870</td><td>bioRxiv preprint first posted online Sep. 27, 2016; +<br/>doi: +<br/>http://dx.doi.org/10.1101/077784 +<br/>. +<br/>The copyright holder for this preprint (which was not +<br/>peer-reviewed) is the author/funder. It is made available under a +<br/>CC-BY-NC-ND 4.0 International license +<br/>. +<br/>Resting high frequency heart rate variability is not associated with the +<br/>recognition of emotional facial expressions in healthy human adults. +<br/>1 Univ. Grenoble Alpes, LPNC, F-38040, Grenoble, France +<br/>2 CNRS, LPNC UMR 5105, F-38040, Grenoble, France +<br/>3 IPSY, Université Catholique de Louvain, Louvain-la-Neuve, Belgium +<br/>4 Fund for Scientific Research (FRS-FNRS), Brussels, Belgium +<br/>Correspondence concerning this article should be addressed to Brice Beffara, Office E250, Institut +<br/>de Recherches en Sciences Psychologiques, IPSY - Place du Cardinal Mercier, 10 bte L3.05.01 B-1348 +<br/>Author note +<br/>This study explores whether the myelinated vagal connection between the heart and the brain +<br/>is involved in emotion recognition. The Polyvagal theory postulates that the activity of the +<br/>myelinated vagus nerve underlies socio-emotional skills. It has been proposed that the perception +<br/>of emotions could be one of this skills dependent on heart-brain interactions. However, this +<br/>assumption was differently supported by diverging results suggesting that it could be related to +<br/>confounded factors. In the current study, we recorded the resting state vagal activity (reflected by +<br/>High Frequency Heart Rate Variability, HF-HRV) of 77 (68 suitable for analysis) healthy human +<br/>adults and measured their ability to identify dynamic emotional facial expressions. Results show +<br/>that HF-HRV is not related to the recognition of emotional facial expressions in healthy human +<br/>adults. We discuss this result in the frameworks of the polyvagal theory and the neurovisceral +<br/>integration model. +<br/>Keywords: HF-HRV; autonomic flexibility; emotion identification; dynamic EFEs; Polyvagal +<br/>theory; Neurovisceral integration model +<br/>Word count: 9810 +<br/>10 +<br/>11 +<br/>12 +<br/>13 +<br/>14 +<br/>15 +<br/>16 +<br/>17 +<br/>Introduction +<br/>The behavior of an animal is said social when involved in in- +<br/>teractions with other animals (Ward & Webster, 2016). These +<br/>interactions imply an exchange of information, signals, be- +<br/>tween at least two animals. In humans, the face is an efficient +<br/>communication channel, rapidly providing a high quantity of +<br/>information. Facial expressions thus play an important role +<br/>in the transmission of emotional information during social +<br/>interactions. The result of the communication is the combina- +<br/>tion of transmission from the sender and decoding from the +<br/>receiver (Jack & Schyns, 2015). As a consequence, the quality +<br/>of the interaction depends on the ability to both produce and +<br/>identify facial expressions. Emotions are therefore a core +<br/>feature of social bonding (Spoor & Kelly, 2004). Health +<br/>of individuals and groups depend on the quality of social +<br/>bonds in many animals (Boyer, Firat, & Leeuwen, 2015; S. L. +<br/>Brown & Brown, 2015; Neuberg, Kenrick, & Schaller, 2011), +<br/>18 +<br/>19 +<br/>20 +<br/>21 +<br/>22 +<br/>23 +<br/>24 +<br/>25 +<br/>26 +<br/>27 +<br/>28 +<br/>29 +<br/>30 +<br/>31 +<br/>32 +<br/>33 +<br/>34 +<br/>35 +<br/>especially in highly social species such as humans (Singer & +<br/>Klimecki, 2014). +<br/>The recognition of emotional signals produced by others is +<br/>not independent from its production by oneself (Niedenthal, +<br/>2007). The muscles of the face involved in the production of +<br/>a facial expressions are also activated during the perception of +<br/>the same facial expressions (Dimberg, Thunberg, & Elmehed, +<br/>2000). In other terms, the facial mimicry of the perceived +<br/>emotional facial expression (EFE) triggers its sensorimotor +<br/>simulation in the brain, which improves the recognition abili- +<br/>ties (Wood, Rychlowska, Korb, & Niedenthal, 2016). Beyond +<br/>that, the emotion can be seen as the body -including brain- +<br/>dynamic itself (Gallese & Caruana, 2016) which helps to un- +<br/>derstand why behavioral simulation is necessary to understand +<br/>the emotion. +<br/>The interplay between emotion production, emotion percep- +<br/>tion, social communication and body dynamics has been sum- +<br/>marized in the framework of the polyvagal theory (Porges, +</td><td>('37799937', 'Nicolas Vermeulen', 'nicolas vermeulen')<br/>('2634712', 'Martial Mermillod', 'martial mermillod')</td><td>Louvain-la-Neuve, Belgium. E-mail: brice.beffara@univ-grenoble-alpes.fr +</td></tr><tr><td>68f9cb5ee129e2b9477faf01181cd7e3099d1824</td><td>ALDA Algorithms for Online Feature Extraction +</td><td>('2784763', 'Youness Aliyari Ghassabeh', 'youness aliyari ghassabeh')<br/>('2060085', 'Hamid Abrishami Moghaddam', 'hamid abrishami moghaddam')</td><td></td></tr><tr><td>68bf34e383092eb827dd6a61e9b362fcba36a83a</td><td></td><td></td><td></td></tr><tr><td>68d40176e878ebffbc01ffb0556e8cb2756dd9e9</td><td>International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 +<br/>International Conference on Humming Bird ( 01st March 2014) +<br/>RESEARCH ARTICLE +<br/> OPEN ACCESS +<br/>Locality Repulsion Projection and Minutia Extraction Based +<br/>Similarity Measure for Face Recognition +<br/><b>AgnelAnushya P. is currently pursuing M.E (Computer Science and engineering) at Vins Christian college of</b><br/>2Ramya P. is currently working as an Asst. Professor in the dept. of Information Technology at Vins Christian +<br/><b>college of Engineering</b></td><td></td><td>Engineering. e-mail:anushyase@gmail.com. +</td></tr><tr><td>68c4a1d438ea1c6dfba92e3aee08d48f8e7f7090</td><td>AgeNet: Deeply Learned Regressor and Classifier for +<br/>Robust Apparent Age Estimation +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/>2Tencent BestImage Team, Shanghai, 100080, China +</td><td>('1731144', 'Xin Liu', 'xin liu')<br/>('1688086', 'Shaoxin Li', 'shaoxin li')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('1698586', 'Jie Zhang', 'jie zhang')<br/>('3126238', 'Shuzhe Wu', 'shuzhe wu')<br/>('13323391', 'Wenxian Liu', 'wenxian liu')<br/>('34393045', 'Hu Han', 'hu han')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{xin.liu, meina.kan, jie.zhang, shuzhe.wu, wenxian.liu, hu.han}@vipl.ict.ac.cn +<br/>{darwinli}@tencent.com, {sgshan, xlchen}@ict.ac.cn +</td></tr><tr><td>6889d649c6bbd9c0042fadec6c813f8e894ac6cc</td><td>Analysis of Robust Soft Learning Vector +<br/>Quantization and an application to Facial +<br/>Expression Recognition +</td><td></td><td></td></tr><tr><td>68f69e6c6c66cfde3d02237a6918c9d1ee678e1b</td><td>Enhancing Concept Detection by Pruning Data with MCA-based Transaction +<br/>Weights +<br/>Department of Electrical and +<br/>Computer Engineering +<br/><b>University of Miami</b><br/>Coral Gables, FL 33124, USA +<br/>School of Computing and +<br/>Information Sciences +<br/><b>Florida International University</b><br/>Miami, FL 33199, USA +</td><td>('1685202', 'Lin Lin', 'lin lin')<br/>('1693826', 'Mei-Ling Shyu', 'mei-ling shyu')<br/>('1705664', 'Shu-Ching Chen', 'shu-ching chen')</td><td>Email: l.lin2@umiami.edu, shyu@miami.edu +<br/>Email: chens@cs.fiu.edu +</td></tr><tr><td>682760f2f767fb47e1e2ca35db3becbb6153756f</td><td>The Effect of Pets on Happiness: A Large-scale Multi-Factor +<br/>Analysis using Social Multimedia +<br/>From reducing stress and loneliness, to boosting productivity and overall well-being, pets are believed to play +<br/>a significant role in people’s daily lives. Many traditional studies have identified that frequent interactions +<br/>with pets could make individuals become healthier and more optimistic, and ultimately enjoy a happier life. +<br/>However, most of those studies are not only restricted in scale, but also may carry biases by using subjective +<br/>self-reports, interviews, and questionnaires as the major approaches. In this paper, we leverage large-scale +<br/>data collected from social media and the state-of-the-art deep learning technologies to study this phenomenon +<br/>in depth and breadth. Our study includes four major steps: 1) collecting timeline posts from around 20,000 +<br/>Instagram users; 2) using face detection and recognition on 2-million photos to infer users’ demographics, +<br/>relationship status, and whether having children, 3) analyzing a user’s degree of happiness based on images +<br/>and captions via smiling classification and textual sentiment analysis; 3) applying transfer learning techniques +<br/>to retrain the final layer of the Inception v3 model for pet classification; and 4) analyzing the effects of pets +<br/>on happiness in terms of multiple factors of user demographics. Our main results have demonstrated the +<br/>efficacy of our proposed method with many new insights. We believe this method is also applicable to other +<br/>domains as a scalable, efficient, and effective methodology for modeling and analyzing social behaviors and +<br/>psychological well-being. In addition, to facilitate the research involving human faces, we also release our +<br/>dataset of 700K analyzed faces. +<br/>CCS Concepts: • Human-centered computing → Social media; +<br/>Additional Key Words and Phrases: Happiness analysis, happiness, user demographics, pet and happiness, +<br/>social multimedia, social media. +<br/>ACM Reference format: +<br/>Analysis using Social Multimedia. ACM Trans. Intell. Syst. Technol. 9, 4, Article 39 (June 2017), 15 pages. +<br/>https://doi.org/0000001.0000001 +<br/>1 INTRODUCTION +<br/>Happiness has always been a subjective and multidimensional matter; its definition varies individu- +<br/>ally, and the factors impacting our feeling of happiness are diverse. A study in [21] has constructed +<br/><b>We thank the support of New York State through the Goergen Institute for Data Science, our corporate research sponsors</b><br/>Xerox and VisualDX, and NSF Award #1704309. +<br/><b>Author s addresses: X. Peng, University of Rochester; L. Chi</b><br/><b>University of Rochester and J. Luo, University of Rochester</b><br/>Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee +<br/>provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the +<br/>full citation on the first page. Copyrights for components of this work owned by others than the author(s) must be honored. +</td><td>('1901094', 'Xuefeng Peng', 'xuefeng peng')<br/>('35678395', 'Li-Kai Chi', 'li-kai chi')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')<br/>('1901094', 'Xuefeng Peng', 'xuefeng peng')<br/>('35678395', 'Li-Kai Chi', 'li-kai chi')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')</td><td></td></tr><tr><td>683ec608442617d11200cfbcd816e86ce9ec0899</td><td>Dual Linear Regression Based Classification for Face Cluster Recognition +<br/><b>University of Northern British Columbia</b><br/>Prince George, BC, Canada V2N 4Z9 +</td><td>('1692551', 'Liang Chen', 'liang chen')</td><td>chen.liang.97@gmail.com +</td></tr><tr><td>68c17aa1ecbff0787709be74d1d98d9efd78f410</td><td>International Journal of Optomechatronics, 6: 92–119, 2012 +<br/>Copyright # Taylor & Francis Group, LLC +<br/>ISSN: 1559-9612 print=1559-9620 online +<br/>DOI: 10.1080/15599612.2012.663463 +<br/>GENDER CLASSIFICATION FROM FACE IMAGES +<br/>USING MUTUAL INFORMATION AND FEATURE +<br/>FUSION +<br/>Department of Electrical Engineering and Advanced Mining Technology +<br/>Center, Universidad de Chile, Santiago, Chile +<br/>In this article we report a new method for gender classification from frontal face images +<br/>using feature selection based on mutual information and fusion of features extracted from +<br/>intensity, shape, texture, and from three different spatial scales. We compare the results of +<br/>three different mutual information measures: minimum redundancy and maximal relevance +<br/>(mRMR), normalized mutual information feature selection (NMIFS), and conditional +<br/>mutual information feature selection (CMIFS). We also show that by fusing features +<br/>extracted from six different methods we significantly improve the gender classification +<br/>results relative to those previously published, yielding 99.13% of the gender classification +<br/>rate on the FERET database. +<br/>Keywords: Feature fusion, feature selection, gender classification, mutual information, real-time gender +<br/>classification +<br/>1. INTRODUCTION +<br/>During the 90’s, one of the main issues addressed in the area of computer +<br/>vision was face detection. Many methods and applications were developed including +<br/>the face detection used in many digital cameras nowadays. Gender classification is +<br/>important in many possible applications including electronic marketing. Displays +<br/>at retail stores could show products and offers according to the person gender as +<br/>the person passes in front of a camera at the store. This is not a simple task since +<br/>faces are not rigid and depend on illumination, pose, gestures, facial expressions, +<br/>occlusions (glasses), and other facial features (makeup, beard). The high variability +<br/>in the appearance of the face directly affects their detection and classification. Auto- +<br/>matic classification of gender from face images has a wide range of possible applica- +<br/>tions, ranging from human-computer interaction to applications in real-time +<br/>electronic marketing in retail stores (Shan 2012; Bekios-Calfa et al. 2011; Chu +<br/>et al. 2010; Perez et al. 2010a). +<br/>Automatic gender classification has a wide range of possible applications for +<br/>improving human-machine interaction and face identification methods (Irick et al. +<br/>ing.uchile.cl +<br/>92 +</td><td>('32271973', 'Claudio Perez', 'claudio perez')<br/>('40333310', 'Juan Tapia', 'juan tapia')<br/>('32723983', 'Claudio Held', 'claudio held')<br/>('32271973', 'Claudio Perez', 'claudio perez')<br/>('32271973', 'Claudio Perez', 'claudio perez')</td><td>Engineering, Universidad de Chile Casilla 412-3, Av. Tupper 2007, Santiago, Chile. E-mail: clperez@ +</td></tr><tr><td>68f61154a0080c4aae9322110c8827978f01ac2e</td><td>Research Article +<br/>Journal of the Optical Society of America A +<br/>Recognizing blurred, non-frontal, illumination and +<br/>expression variant partially occluded faces +<br/><b>Indian Institute of Technology Madras, Chennai 600036, India</b><br/>Compiled June 26, 2016 +<br/>The focus of this paper is on the problem of recognizing faces across space-varying motion blur, changes +<br/>in pose, illumination, and expression, as well as partial occlusion, when only a single image per subject +<br/>is available in the gallery. We show how the blur incurred due to relative motion between the camera and +<br/>the subject during exposure can be estimated from the alpha matte of pixels that straddle the boundary +<br/>between the face and the background. We also devise a strategy to automatically generate the trimap re- +<br/>quired for matte estimation. Having computed the motion via the matte of the probe, we account for pose +<br/>variations by synthesizing from the intensity image of the frontal gallery, a face image that matches the +<br/>pose of the probe. To handle illumination and expression variations, and partial occlusion, we model the +<br/>probe as a linear combination of nine blurred illumination basis images in the synthesized non-frontal +<br/>pose, plus a sparse occlusion. We also advocate a recognition metric that capitalizes on the sparsity of the +<br/>occluded pixels. The performance of our method is extensively validated on synthetic as well as real face +<br/>data. © 2016 Optical Society of America +<br/>OCIS codes: +<br/>(150.0150) Machine vision. +<br/>http://dx.doi.org/10.1364/ao.XX.XXXXXX +<br/>(100.0100) Image processing; (100.5010) Pattern recognition; (100.3008) Image recognition, algorithms and filters; +<br/>1. INTRODUCTION +<br/>State-of-the-art face recognition (FR) systems can outperform +<br/>even humans when presented with images captured under con- +<br/>trolled environments. However, their performance drops quite +<br/>rapidly in unconstrained settings due to image degradations +<br/>arising from blur, variations in pose, illumination, and expres- +<br/>sion, partial occlusion etc. Motion blur is commonplace today +<br/>owing to the exponential rise in the use and popularity of light- +<br/>weight and cheap hand-held imaging devices, and the ubiquity +<br/>of mobile phones equipped with cameras. Photographs cap- +<br/>tured using a hand-held device usually contain blur when the +<br/>illumination is poor because larger exposure times are needed +<br/>to compensate for the lack of light, and this increases the possi- +<br/>bility of camera shake. On the other hand, reducing the shutter +<br/>speed results in noisy images while tripods inevitably restrict +<br/>mobility. Even for a well-lit scene, the face might be blurred if +<br/>the subject is in motion. The problem is further compounded +<br/>in the case of poorly-lit dynamic scenes since the blur observed +<br/>on the face is due to the combined effects of the blur induced +<br/>by the motion of the camera and the independent motion of +<br/>the subject. In addition to blur and illumination, practical face +<br/>recognition algorithms must also possess the ability to recognize +<br/>faces across reasonable variations in pose. Partial occlusion and +<br/>facial expression changes, common in real-world applications, +<br/>escalate the challenges further. Yet another factor that governs +<br/>the performance of face recognition algorithms is the number +<br/>of images per subject available for training. In many practical +<br/>application scenarios such as law enforcement, driver license or +<br/>passport identification, where there is usually only one training +<br/>sample per subject in the database, techniques that rely on the +<br/>size and representation of the training set suffer a serious perfor- +<br/>mance drop or even fail to work. Face recognition algorithms +<br/>can broadly be classified into either discriminative or genera- +<br/>tive approaches. While the availability of large labeled datasets +<br/>and greater computing power has boosted the performance of +<br/>discriminative methods [1, 2] recently, generative approaches +<br/>continue to remain very popular [3, 4], and there is concurrent +<br/>research in both directions. The model we present in this paper +<br/>falls into the latter category. In fact, generative models are even +<br/>useful for producing training samples for learning algorithms. +<br/>Literature on face recognition from blurred images can be +<br/>broadly classified into four categories. It is important to note +<br/>that all of them (except our own earlier work in [4]) are restricted +<br/>to the convolution model for uniform blur. In the first approach +<br/>[5, 6], the blurred probe image is first deblurred using standard +<br/>deconvolution algorithms before performing recognition. How- +</td><td></td><td>*Corresponding author: jithuthatswho@gmail.com +</td></tr><tr><td>6821113166b030d2123c3cd793dd63d2c909a110</td><td>STUDIA INFORMATICA +<br/>Volume 36 +<br/>2015 +<br/>Number 1 (119) +<br/><b>Gdansk University of Technology, Faculty of Electronics, Telecommunication</b><br/>and Informatics +<br/>ACQUISITION AND INDEXING OF RGB-D RECORDINGS FOR +<br/>FACIAL EXPRESSIONS AND EMOTION RECOGNITION1 +<br/>Summary. In this paper KinectRecorder comprehensive tool is described which +<br/>provides for convenient and fast acquisition, indexing and storing of RGB-D video +<br/>streams from Microsoft Kinect sensor. The application is especially useful as a sup- +<br/>porting tool for creation of fully indexed databases of facial expressions and emotions +<br/>that can be further used for learning and testing of emotion recognition algorithms for +<br/>affect-aware applications. KinectRecorder was successfully exploited for creation of +<br/>Facial Expression and Emotion Database (FEEDB) significantly reducing the time of +<br/>the whole project consisting of data acquisition, indexing and validation. FEEDB has +<br/>already been used as a learning and testing dataset for a few emotion recognition al- +<br/>gorithms which proved utility of the database, and the KinectRecorder tool. +<br/>Keywords: RGB-D data acquisition and indexing, facial expression recognition, +<br/>emotion recognition +<br/>AKWIZYCJA ORAZ INDEKSACJA NAGRAŃ RGB-D DO +<br/>Streszczenie. W pracy przedstawiono kompleksowe narzędzie, które pozwala na +<br/>wygodną i szybką akwizycję, indeksowanie i przechowywanie nagrań strumieni +<br/>RGB-D z czujnika Microsoft Kinect. Aplikacja jest szczególnie przydatna jako na- +<br/>mogą być następnie wykorzystywane do nauki i testowania algorytmów rozpoznawa- +<br/>nia emocji użytkownika dla aplikacji je uwzględniających. KinectRecorder został +<br/>skracając czas całego procesu, obejmującego akwizycję, indeksowanie i walidację +<br/>nagrań. Baza FEEDB została już z powodzeniem wykorzystana jako uczący i testują- +<br/> +<br/>1 The research leading to these results has received funding from the Polish-Norwegian Research Programme +<br/>operated by the National Centre for Research and Development under the Norwegian Financial Mechanism +<br/>2009-2014 in the frame of Project Contract No Pol-Nor/210629/51/2013. +</td><td>('3271448', 'Mariusz SZWOCH', 'mariusz szwoch')</td><td></td></tr><tr><td>68a04a3ae2086986877fee2c82ae68e3631d0356</td><td>THERMAL & REFLECTANCE BASED IDENTIFICATION IN CHALLENGING VARIABLE ILLUMINATIONS +<br/>Thermal and Reflectance Based Personal +<br/>Identification Methodology in Challenging +<br/>Variable Illuminations +<br/>†Department of Engineering +<br/><b>University of Cambridge</b><br/>‡Delphi Corporation, +<br/>Delphi Electronics and Safety +<br/>Cambridge, CB2 1PZ, UK +<br/>Kokomo, IN 46901-9005, USA +<br/>February 15, 2007 +<br/>DRAFT +</td><td>('2214319', 'Riad Hammoud', 'riad hammoud')</td><td>{oa214,cipolla}@eng.cam.ac.uk +<br/>riad.hammoud@delphi.com +</td></tr><tr><td>6888f3402039a36028d0a7e2c3df6db94f5cb9bb</td><td>Under review as a conference paper at ICLR 2018 +<br/>CLASSIFIER-TO-GENERATOR ATTACK: ESTIMATION +<br/>OF TRAINING DATA DISTRIBUTION FROM CLASSIFIER +<br/>Anonymous authors +<br/>Paper under double-blind review +</td><td></td><td></td></tr><tr><td>57f5711ca7ee5c7110b7d6d12c611d27af37875f</td><td>Illumination Invariance for Face Verification +<br/>Submitted for the Degree of +<br/>Doctor of Philosophy +<br/>from the +<br/><b>University of Surrey</b><br/>Centre for Vision, Speech and Signal Processing +<br/>School of Electronics and Physical Sciences +<br/><b>University of Surrey</b><br/>Guildford, Surrey GU2 7XH, U.K. +<br/>August 2006 +</td><td>('28467739', 'J. Short', 'j. short')<br/>('28467739', 'J. Short', 'j. short')</td><td></td></tr><tr><td>570308801ff9614191cfbfd7da88d41fb441b423</td><td>Unsupervised Synchrony Discovery in Human Interaction +<br/><b>Robotics Institute, Carnegie Mellon University 3University of Pittsburgh, USA</b><br/><b>Beihang University, Beijing, China</b><br/><b>University of Miami, USA</b></td><td>('39336289', 'Wen-Sheng Chu', 'wen-sheng chu')<br/>('1874236', 'Daniel S. Messinger', 'daniel s. messinger')</td><td></td></tr><tr><td>57bf9888f0dfcc41c5ed5d4b1c2787afab72145a</td><td>Robust Facial Expression Recognition Based on +<br/>Local Directional Pattern +<br/>Automatic facial expression recognition has many +<br/>potential applications +<br/>in different areas of human +<br/>computer interaction. However, they are not yet fully +<br/>realized due to the lack of an effective facial feature +<br/>descriptor. In this paper, we present a new appearance- +<br/>based feature descriptor, the local directional pattern +<br/>(LDP), to represent facial geometry and analyze its +<br/>performance in expression recognition. An LDP feature is +<br/>obtained by computing the edge response values in 8 +<br/>directions at each pixel and encoding them into an 8 bit +<br/>binary number using the relative strength of these edge +<br/>responses. The LDP descriptor, a distribution of LDP +<br/>codes within an image or image patch, is used to describe +<br/>each expression image. The effectiveness of dimensionality +<br/>reduction techniques, such as principal component +<br/>analysis and AdaBoost, is also analyzed in terms of +<br/>computational cost saving and classification accuracy. Two +<br/>well-known machine +<br/>template +<br/>matching and support vector machine, are used for +<br/>classification using the Cohn-Kanade and Japanese +<br/>female facial expression databases. Better classification +<br/>accuracy shows the superiority of LDP descriptor against +<br/>other appearance-based feature descriptors. +<br/>learning methods, +<br/>Keywords: Image representation, facial expression +<br/>recognition, local directional pattern, features extraction, +<br/>principal component analysis, support vector machine. +<br/> +<br/>Manuscript received Mar. 15, 2010; revised July 15, 2010; accepted Aug. 2, 2010. +<br/>This work was supported by the Korea Research Foundation Grant funded by the Korean +<br/>Government (KRF-2010-0015908). +<br/><b>Kyung Hee University, Yongin, Rep. of Korea</b><br/>doi:10.4218/etrij.10.1510.0132 +<br/>I. Introduction +<br/>Facial expression provides the most natural and immediate +<br/>indication about a person’s emotions and intentions [1], [2]. +<br/>Therefore, automatic facial expression analysis is an important +<br/>and challenging task that has had great impact in such areas as +<br/>human-computer +<br/>interaction and data-driven animation. +<br/>Furthermore, video cameras have recently become an integral +<br/>part of many consumer devices [3] and can be used for +<br/>capturing facial images for recognition of people and their +<br/>emotions. This ability to recognize emotions can enable +<br/>customized applications [4], [5]. Even though much work has +<br/>already been done on automatic facial expression recognition +<br/>[6], [7], higher accuracy with reasonable speed still remains a +<br/>great challenge [8]. Consequently, a fast but robust facial +<br/>expression recognition system is very much needed to support +<br/>these applications. +<br/>The most critical aspect for any successful facial expression +<br/>recognition system is to find an efficient facial feature +<br/>representation [9]. An extracted facial feature can be considered +<br/>an efficient representation if it can fulfill three criteria: first, it +<br/>minimizes within-class variations of expressions while +<br/>maximizes between-class variations; second, it can be easily +<br/>extracted from the raw face image; and third, it can be +<br/>described in a low-dimensional feature space to ensure +<br/>computational speed during the classification step [10], [11]. +<br/>The goal of the facial feature extraction is thus to find an +<br/>efficient and effective representation of the facial images which +<br/>would provide robustness during recognition process. Two +<br/>types of approaches have been proposed to extract facial +<br/>features for expression recognition: a geometric feature-based +<br/>system and an appearance-based system [12]. +<br/>In the geometric feature extraction system, the shape and +<br/>© 2010 +<br/> ETRI Journal, Volume 32, Number 5, October 2010 +</td><td>('3182680', 'Taskeed Jabid', 'taskeed jabid')<br/>('9408912', 'Hasanul Kabir', 'hasanul kabir')<br/>('1685505', 'Oksam Chae', 'oksam chae')<br/>('3182680', 'Taskeed Jabid', 'taskeed jabid')</td><td>Taskeed Jabid (phone: +82 31 201 2948, email: taskeed@khu.ac.kr), Md. Hasanul Kabir +<br/>(email: hasanul@khu.ac.kr), and Oksam Chae (email: oschae@khu.ac.kr) are with the +</td></tr><tr><td>57ebeff9273dea933e2a75c306849baf43081a8c</td><td>Deep Convolutional Network Cascade for Facial Point Detection +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('1681656', 'Yi Sun', 'yi sun')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>sy011@ie.cuhk.edu.hk +<br/>xgwang@ee.cuhk.edu.hk +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>574751dbb53777101502419127ba8209562c4758</td><td></td><td></td><td></td></tr><tr><td>5778d49c8d8d127351eee35047b8d0dc90defe85</td><td>Probabilistic Subpixel Temporal Registration +<br/>for Facial Expression Analysis +<br/><b>Queen Mary University of London</b><br/>Centre for Intelligent Sensing +</td><td>('1781916', 'Hatice Gunes', 'hatice gunes')<br/>('1713138', 'Andrea Cavallaro', 'andrea cavallaro')</td><td>fe.sariyanidi, h.gunes, a.cavallarog@qmul.ac.uk +</td></tr><tr><td>57ee3a8b0cafe211d1e9b477d210bb78b9d43bc1</td><td>Modeling the joint density of two images under a variety of transformations +<br/>Joshua Susskind +<br/><b>Institute for Neural Computation</b><br/><b>University of California, San Diego</b><br/>United States +<br/>Department of Computer Science +<br/><b>University of Frankfurt</b><br/>Germany +<br/>Department of Computer Science +<br/>Department of Computer Science +<br/>ETH Zurich +<br/>Switzerland +<br/>Geoffrey Hinton +<br/><b>University of Toronto</b><br/>Canada +</td><td>('1710604', 'Roland Memisevic', 'roland memisevic')<br/>('1742208', 'Marc Pollefeys', 'marc pollefeys')</td><td>josh@mplab.ucsd.edu +<br/>ro@cs.uni-frankfurt.de +<br/>hinton@cs.toronto.edu +<br/>marc.pollefeys@inf.ethz.ch +</td></tr><tr><td>57fd229097e4822292d19329a17ceb013b2cb648</td><td>Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-16) +<br/>Fast Structural Binary Coding +<br/><b>University of California, San Diego</b><br/><b>University of California, San Diego</b></td><td>('2451800', 'Dongjin Song', 'dongjin song')<br/>('1722649', 'Wei Liu', 'wei liu')<br/>('3520515', 'David A. Meyer', 'david a. meyer')</td><td>La Jolla, USA, 92093-0409. Email: dosong@ucsd.edu +<br/>] Didi Research, Didi Kuaidi, Beijing, China. Email: wliu@ee.columbia.edu +<br/>La Jolla, USA, 92093-0112. Email: dmeyer@math.ucsd.edu +</td></tr><tr><td>57c59011614c43f51a509e10717e47505c776389</td><td>Unsupervised Human Action Detection by Action Matching +<br/><b>The Australian National University Queensland University of Technology</b></td><td>('1688071', 'Basura Fernando', 'basura fernando')</td><td>firstname.lastname@anu.edu.au +<br/>s.shirazi@qut.edu.au +</td></tr><tr><td>57b8b28f8748d998951b5a863ff1bfd7ca4ae6a5</td><td></td><td></td><td></td></tr><tr><td>57101b29680208cfedf041d13198299e2d396314</td><td></td><td></td><td></td></tr><tr><td>57893403f543db75d1f4e7355283bdca11f3ab1b</td><td></td><td></td><td></td></tr><tr><td>571f493c0ade12bbe960cfefc04b0e4607d8d4b2</td><td>International Journal of Research Studies in Science, Engineering and Technology +<br/>Volume 3, Issue 2, February 2016, PP 18-41 +<br/>ISSN 2349-4751 (Print) & ISSN 2349-476X (Online) +<br/>Review on Content Based Image Retrieval: From Its Origin to the +<br/>New Age +<br/>Assistant Professor, ECE +<br/>Dr. B. L. Malleswari +<br/>Principal +<br/><b>Mahatma Gandhi Institute of Technology</b><br/><b>Sridevi Women's Engineering College</b><br/>Hyderabad, India +<br/>Hyderabad, India +</td><td></td><td>pasumarthinalini@gmil.com +<br/>blmalleswari@gmail.com +</td></tr><tr><td>57f8e1f461ab25614f5fe51a83601710142f8e88</td><td>Region Selection for Robust Face Verification using UMACE Filters +<br/>Department of Electrical, Electronic and Systems Engineering, Faculty of Engineering, +<br/>Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia. +<br/>In this paper, we investigate the verification performances of four subdivided face images with varying expressions. The +<br/>objective of this study is to evaluate which part of the face image is more tolerant to facial expression and still retains its personal +<br/>characteristics due to the variations of the image. The Unconstrained Minimum Average Correlation Energy (UMACE) filter is +<br/>implemented to perform the verification process because of its advantages such as shift–invariance, ability to trade-off between +<br/>discrimination and distortion tolerance, e.g. variations in pose, illumination and facial expression. The database obtained from the +<br/>facial expression database of Advanced Multimedia Processing (AMP) Lab at CMU is used in this study. Four equal +<br/>sizes of face regions i.e. bottom, top, left and right halves are used for the purpose of this study. The results show that the bottom +<br/>half of the face region gives the best performance in terms of the PSR values with zero false accepted rate (FAR) and zero false +<br/>rejection rate (FRR) compared to the other three regions. +<br/>1. Introduction +<br/>Face recognition is a well established field of research, +<br/>and a large number of algorithms have been proposed in the +<br/>literature. Various classifiers have been explored to improve +<br/>the accuracy of face classification. The basic approach is to +<br/>use distance-base methods which measure Euclidean distance +<br/>between any two vectors and then compare it with the preset +<br/>threshold. Neural Networks are often used as classifiers due +<br/>to their powerful generation ability [1]. Support Vector +<br/>Machines (SVM) have been applied with encouraging results +<br/>[2]. +<br/>In biometric applications, one of the important tasks is the +<br/>matching process between an individual biometrics against +<br/>the database that has been prepared during the enrolment +<br/>stage. For biometrics systems such as face authentication that +<br/>use images as personal characteristics, biometrics sensor +<br/>output and image pre-processing play an important role since +<br/>the quality of a biometric input can change significantly due +<br/>to illumination, noise and pose variations. Over the years, +<br/>researchers have studied the role of illumination variation, +<br/>pose variation, facial expression, and occlusions in affecting +<br/>the performance of face verification systems [3]. +<br/>The Minimum Average Correlation Energy (MACE) +<br/>filters have been reported to be an alternative solution to these +<br/>problems because of the advantages such as shift-invariance, +<br/>close-form expressions and distortion-tolerance. MACE +<br/>filters have been successfully applied in the field of automatic +<br/>target recognition as well as in biometric verification [3][4]. +<br/>Face and fingerprint verification using correlation filters have +<br/>been investigated in [5] and [6], respectively. Savvides et.al +<br/>performed face authentication and identification using +<br/>correlation filters based on illumination variation [7]. In the +<br/>process of implementing correlation filters, the number of +<br/>training images used depends on the level of distortions +<br/>applied to the images [5], [6]. +<br/>In this study, we investigate which part of a face image is +<br/>more tolerant to facial expression and retains its personal +<br/>characteristics for the verification process. Four subdivided +<br/>face images, i.e. bottom, top, left and right halves, with +<br/>varying expressions are investigated. By identifying only the +<br/>region of the face that gives the highest verification +<br/>performance, that region can be used instead of the full-face +<br/>to reduce storage requirements. +<br/>2. Unconstrained Minimum Average Correlation +<br/>Energy (UMACE) Filter +<br/>Correlation filter theory and the descriptions of the design +<br/>of the correlation filter can be found in a tutorial survey paper +<br/>[8]. According to [4][6], correlation filter evolves from +<br/>matched filters which are optimal for detecting a known +<br/>reference image in the presence of additive white Gaussian +<br/>noise. However, the detection rate of matched filters +<br/>decreases significantly due to even the small changes of scale, +<br/>rotation and pose of the reference image. +<br/>the pre-specified peak values +<br/>In an effort to solve this problem, the Synthetic +<br/>Discriminant Function (SDF) filter and the Equal Correlation +<br/>Peak SDF (ECP SDF) filter ware introduced which allowed +<br/>several training images to be represented by a single +<br/>correlation filter. SDF filter produces pre-specified values +<br/>called peak constraints. These peak values correspond to the +<br/>authentic class or impostor class when an image is tested. +<br/>However, +<br/>to +<br/>misclassifications when the sidelobes are larger than the +<br/>controlled values at the origin. +<br/>Savvides et.al developed +<br/>the Minimum Average +<br/>Correlation Energy (MACE) filters [5]. This filter reduces the +<br/>large sidelobes and produces a sharp peak when the test +<br/>image is from the same class as the images that have been +<br/>used to design the filter. There are two kinds of variants that +<br/>can be used in order to obtain a sharp peak when the test +<br/>image belongs to the authentic class. The first MACE filter +<br/>variant minimizes the average correlation energy of the +<br/>training images while constraining the correlation output at +<br/>the origin to a specific value for each of the training images. +<br/>The second MACE filter variant is the Unconstrained +<br/>Minimum Average Correlation Energy (UMACE) filter +<br/>which also minimizes the average correlation output while +<br/>maximizing the correlation output at the origin [4]. +<br/>lead +<br/>Proceedings of the International Conference onElectrical Engineering and InformaticsInstitut Teknologi Bandung, Indonesia June 17-19, 2007B-67ISBN 978-979-16338-0-2611 </td><td>('5461819', 'Salina Abdul Samad', 'salina abdul samad')<br/>('2864147', 'Dzati Athiar Ramli', 'dzati athiar ramli')<br/>('2573778', 'Aini Hussain', 'aini hussain')</td><td>* E-mail: salina@vlsi.eng.ukm.my +</td></tr><tr><td>57a1466c5985fe7594a91d46588d969007210581</td><td>A Taxonomy of Face-models for System Evaluation +<br/>Motivation and Data Types +<br/>Synthetic Data Types +<br/>Unverified – Have no underlying physical or +<br/>statistical basis +<br/>Physics -Based – Based on structure and +<br/>materials combined with the properties +<br/>formally modeled in physics. +<br/>Statistical – Use statistics from real +<br/>data/experiments to estimate/learn model +<br/>parameters. Generally have measurements +<br/>of accuracy +<br/>Guided Synthetic – Individual models based +<br/>on individual people. No attempt to capture +<br/>properties of large groups, a unique model +<br/>per person. For faces, guided models are +<br/>composed of 3D structure models and skin +<br/>textures, capturing many artifacts not +<br/>easily parameterized. Can be combined with +<br/>physics-based rendering to generate samples +<br/>under different conditions. +<br/>Semi–Synethetic – Use measured data such +<br/>as 2D images or 3D facial scans. These are +<br/>not truly synthetic as they are re-rendering’s +<br/>of real measured data. +<br/>Semi and Guided Synthetic data provide +<br/>higher operational relevance while +<br/>maintaining a high degree of control. +<br/>Generating statistically significant size +<br/>datasets for face matching system +<br/>evaluation is both a laborious and +<br/>expensive process. +<br/>There is a gap in datasets that allow for +<br/>evaluation of system issues including: +<br/> Long distance recognition +<br/> Blur caused by atmospherics +<br/> Various weather conditions +<br/> End to end systems evaluation +<br/>Our contributions: +<br/> Define a taxonomy of face-models +<br/>for controlled experimentations +<br/> Show how Synthetic addresses gaps +<br/>in system evaluation +<br/> Show a process for generating and +<br/>validating synthetic models +<br/> Use these models in long distance +<br/>face recognition system evaluation +<br/>Experimental Setup +<br/>Results and Conclusions +<br/>Example Models +<br/>Original Pie +<br/>Semi- +<br/>Synthetic +<br/>FaceGen +<br/>Animetrics +<br/>http://www.facegen.com +<br/>http://www.animetrics.com/products/Forensica.php +<br/>Guided- +<br/>Synthetic +<br/>Models +<br/> Models generated using the well +<br/>known CMU PIE [18] dataset. Each of +<br/>the 68 subjects of PIE were modeled +<br/>using a right profile and frontal +<br/>image from the lights subset. +<br/> Two modeling programs were used, +<br/>Facegen and Animetrics. Both +<br/>programs create OBJ files and +<br/>textures +<br/> Models are re-rendered using +<br/>custom display software built with +<br/>OpenGL, GLUT and DevIL libraries +<br/> Custom Display Box housing a BENQ SP820 high +<br/>powered projector rated at 4000 ANSI Lumens +<br/> Canon EOS 7D withd a Sigma 800mm F5.6 EX APO +<br/>DG HSM lens a 2x adapter imaging the display +<br/>from 214 meters +<br/>Normalized Example Captures +<br/>Real PIE 1 Animetrics +<br/>FaceGen +<br/>81M inside 214M outside +<br/>Real PIE 2 +<br/> Pre-cropped images were used for the +<br/>commercial core +<br/> Ground truth eye points + geometric/lighting +<br/>normalization pre processing before running +<br/>through the implementation of the V1 +<br/>recognition algorithm found in [1]. +<br/> Geo normalization highlights how the feature +<br/>region of the models looks very similar to +<br/>that of the real person. +<br/>Each test consisted of using 3 approximately frontal gallery images NOT used to +<br/>make the 3D model used as the probe, best score over 3 images determined score. +<br/>Even though the PIE-3D-20100224A–D sets were imaged on the same day, the V1 +<br/>core scored differently on each highlighting the synthetic data’s ability to help +<br/>evaluate data capture methods and effects of varying atmospherics. The ISO setting +<br/>varied which effects the shutter speed, with higher ISO generally yielding less blur. +<br/>Dataset +<br/>Range(m) +<br/>Iso +<br/>V1 +<br/>Comm. +<br/>Original PIE Images +<br/>FaceGen ScreenShots +<br/>Animetrics Screenshots +<br/>PIE-3D-20100210B +<br/>PIE-3D-20100224A +<br/>PIE-3D-20100224B +<br/>PIE-3D-20100224C +<br/>PIE-3D-20100224D +<br/>N/A +<br/>N/A +<br/>N/A +<br/>81m +<br/>214m +<br/>214m +<br/>214m +<br/>214m +<br/>N/A +<br/>N/A +<br/>N/A +<br/>500 +<br/>125 +<br/>125 +<br/>250 +<br/>400 +<br/>100 +<br/>47.76 +<br/>100 +<br/>100 +<br/>58.82 +<br/>45.59 +<br/>81.82 +<br/>79.1 +<br/>100 +<br/>100 +<br/>100 +<br/>100 +<br/>100 +<br/>100 +<br/> The same (100 percent) recognition rate on screenshots as original images +<br/>validate the Anmetrics guided synthetic models and fails FaceGen Models. +<br/> 100% recognition means dataset is too small/easy; exapanding pose and models +<br/>underway. +<br/> Expanded the photohead methodology into 3D +<br/> Developed a robust modeling system allowing for multiple configurations of a +<br/>single real life data set. +<br/> Gabor+SVM based V1[15] significantly more impacted by atmospheric blur than +<br/>the commercial algorithm +<br/>Key References: +<br/>[6 of 21] R. Bevridge, D. Bolme, M Teixeira, and B. Draper. The CSU Face Identification Evaluation System Users Guide: Version 5.0. Technical report, CSU 2003 +<br/>[8 of 21] T. Boult and W. Scheirer. Long range facial image acquisition and quality. In M. Tisarelli, S. Li, and R. Chellappa. +<br/>[15 of 21] N. Pinto, J. J. DiCarlo, and D. D. Cox. How far can you get with a modern face recognition test set using only simple features? In IEEE CVPR, 2009. +<br/>[18 of 21] T. Sim, S. Baker, and M. Bsat. The CMU Pose, Illumination and Expression (PIE) Database. In Proceedings of the IEEE F&G, May 2002. +</td><td>('31552290', 'Brian C. Parks', 'brian c. parks')<br/>('2613438', 'Walter J. Scheirer', 'walter j. scheirer')</td><td>{viyer,skirkbride,bparks,wscheirer,tboult}@vast.uccs.edu +</td></tr><tr><td>574b62c845809fd54cc168492424c5fac145bc83</td><td>Learning Warped Guidance for Blind Face +<br/>Restoration +<br/><b>School of Computer Science and Technology, Harbin Institute of Technology, China</b><br/><b>School of Data and Computer Science, Sun Yat-sen University, China</b><br/><b>University of Kentucky, USA</b></td><td>('21515518', 'Xiaoming Li', 'xiaoming li')<br/>('40508248', 'Yuting Ye', 'yuting ye')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('1737218', 'Liang Lin', 'liang lin')<br/>('38958903', 'Ruigang Yang', 'ruigang yang')</td><td>csxmli@hit.edu.cn, csmliu@outlook.com, yeyuting.jlu@gmail.com, +<br/>wmzuo@hit.edu.cn +<br/>linliang@ieee.org +<br/>ryang@cs.uky.edu +</td></tr><tr><td>57246142814d7010d3592e3a39a1ed819dd01f3b</td><td><b>MITSUBISHI ELECTRIC RESEARCH LABORATORIES</b><br/>http://www.merl.com +<br/>Verification of Very Low-Resolution Faces Using An +<br/>Identity-Preserving Deep Face Super-resolution Network +<br/>TR2018-116 August 24, 2018 +</td><td></td><td></td></tr><tr><td>5721216f2163d026e90d7cd9942aeb4bebc92334</td><td></td><td></td><td></td></tr><tr><td>575141e42740564f64d9be8ab88d495192f5b3bc</td><td>Age Estimation based on Multi-Region +<br/>Convolutional Neural Network +<br/>1Center for Biometrics and Security Research & National Laboratory of Pattern +<br/><b>Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, China</b><br/><b>University of Chinese Academy of Sciences</b></td><td>('40282288', 'Ting Liu', 'ting liu')<br/>('1756538', 'Jun Wan', 'jun wan')<br/>('39974958', 'Tingzhao Yu', 'tingzhao yu')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{ting.liu,jun.wan,zlei,szli}@nlpr.ia.ac.cn,yutingzhao2013@ia.ac.cn +</td></tr><tr><td>5789f8420d8f15e7772580ec373112f864627c4b</td><td>Efficient Global Illumination for Morphable Models +<br/><b>University of Basel, Switzerland</b></td><td>('1801001', 'Andreas Schneider', 'andreas schneider')<br/>('34460642', 'Bernhard Egger', 'bernhard egger')<br/>('32013053', 'Lavrenti Frobeen', 'lavrenti frobeen')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td>{andreas.schneider,sandro.schoenborn,bernhard.egger,l.frobeen,thomas.vetter}@unibas.ch +</td></tr><tr><td>574705812f7c0e776ad5006ae5e61d9b071eebdb</td><td>Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/>A Monthly Journal of Computer Science and Information Technology +<br/>ISSN 2320–088X +<br/>IJCSMC, Vol. 3, Issue. 5, May 2014, pg.780 – 787 +<br/> RESEARCH ARTICLE +<br/>A Novel Approach for Face Recognition +<br/>Using PCA and Artificial Neural Network +<br/><b>Dayananda Sagar College of Engg., India</b><br/><b>Dayananda Sagar College of Engg., India</b></td><td>('9856026', 'Karthik G', 'karthik g')<br/>('9856026', 'Karthik G', 'karthik g')</td><td>1 email : karthik.knocks@gmail.com; 2 email : hcsateesh@gmail.com +</td></tr><tr><td>5753b2b5e442eaa3be066daa4a2ca8d8a0bb1725</td><td></td><td></td><td></td></tr><tr><td>571b83f7fc01163383e6ca6a9791aea79cafa7dd</td><td>SeqFace: Make full use of sequence information for face recognition +<br/><b>College of Information Science and Technology</b><br/><b>Beijing University of Chemical Technology, China</b><br/>YUNSHITU Corp., China +</td><td>('48594708', 'Wei Hu', 'wei hu')<br/>('7524887', 'Yangyu Huang', 'yangyu huang')<br/>('8451319', 'Guodong Yuan', 'guodong yuan')<br/>('47191084', 'Fan Zhang', 'fan zhang')<br/>('50391855', 'Ruirui Li', 'ruirui li')<br/>('47113208', 'Wei Li', 'wei li')</td><td></td></tr><tr><td>574ad7ef015995efb7338829a021776bf9daaa08</td><td>AdaScan: Adaptive Scan Pooling in Deep Convolutional Neural Networks +<br/>for Human Action Recognition in Videos +<br/>1IIT Kanpur‡ +<br/>2SRI International +<br/>3UCSD +</td><td>('24899770', 'Amlan Kar', 'amlan kar')<br/>('12692625', 'Nishant Rai', 'nishant rai')<br/>('39707211', 'Karan Sikka', 'karan sikka')<br/>('39396475', 'Gaurav Sharma', 'gaurav sharma')</td><td></td></tr><tr><td>57a14a65e8ae15176c9afae874854e8b0f23dca7</td><td>UvA-DARE (Digital Academic Repository) +<br/>Seeing mixed emotions: The specificity of emotion perception from static and dynamic +<br/>facial expressions across cultures +<br/>Fang, X.; Sauter, D.A.; van Kleef, G.A. +<br/>Published in: +<br/>Journal of Cross-Cultural Psychology +<br/>DOI: +<br/>10.1177/0022022117736270 +<br/>Link to publication +<br/>Citation for published version (APA): +<br/>Fang, X., Sauter, D. A., & van Kleef, G. A. (2018). Seeing mixed emotions: The specificity of emotion perception +<br/>from static and dynamic facial expressions across cultures. Journal of Cross-Cultural Psychology, 49(1), 130- +<br/>148. DOI: 10.1177/0022022117736270 +<br/>General rights +<br/>It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), +<br/>other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). +<br/>Disclaimer/Complaints regulations +<br/>If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating +<br/>your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask +<br/><b>the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam</b><br/>The Netherlands. You will be contacted as soon as possible. +<br/>Download date: 08 Aug 2018 +<br/><b>UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl</b></td><td></td><td></td></tr><tr><td>57b052cf826b24739cd7749b632f85f4b7bcf90b</td><td>Fast Fashion Guided Clothing Image Retrieval: +<br/>Delving Deeper into What Feature Makes +<br/>Fashion +<br/><b>School of Data and Computer Science, Sun Yat-sen University</b><br/>Guangzhou, P.R China +</td><td>('3079146', 'Yuhang He', 'yuhang he')<br/>('40451106', 'Long Chen', 'long chen')</td><td>*Corresponding Author: chenl46@mail.sysu.edu.cn +</td></tr><tr><td>57d37ad025b5796457eee7392d2038910988655a</td><td>GEERATVEEETATF +<br/> +<br/>by +<br/>DagaEha +<br/>UdeheS eviif +<br/>f.DahaWeiha +<br/>ATheiS biediaia +<br/>Re ieefheDegeef +<br/>aefSciece +<br/>a +<br/>TheSch +<br/> +<br/>Decebe2009 +</td><td></td><td></td></tr><tr><td>57f7d8c6ec690bd436e70d7761bc5f46e993be4c</td><td>Facial Expression Recognition Using Histogram Variances Faces +<br/><b>University of Technology, Sydney, 15 Broadway, Ultimo, NSW 2007, Australia</b><br/><b>University of Aizu, Japan</b></td><td>('32796151', 'Ruo Du', 'ruo du')<br/>('37046680', 'Qiang Wu', 'qiang wu')<br/>('1706670', 'Xiangjian He', 'xiangjian he')<br/>('1714410', 'Wenjing Jia', 'wenjing jia')<br/>('40394300', 'Daming Wei', 'daming wei')</td><td>{ruodu, wuq, sean, wejia}@it.uts.edu.au +<br/>dm-wei@u-aizu.ac.jp +</td></tr><tr><td>3b1260d78885e872cf2223f2c6f3d6f6ea254204</td><td></td><td></td><td></td></tr><tr><td>3b1aaac41fc7847dd8a6a66d29d8881f75c91ad5</td><td>Sparse Representation-based Open Set Recognition +</td><td>('2310707', 'He Zhang', 'he zhang')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')</td><td></td></tr><tr><td>3b092733f428b12f1f920638f868ed1e8663fe57</td><td>On the Size of Convolutional Neural Networks and +<br/>Generalization Performance +<br/>Center for Automation Research, UMIACS* +<br/>Department of Electrical and Computer Engineering† +<br/><b>University of Maryland, College Park</b></td><td>('2747758', 'Maya Kabkab', 'maya kabkab')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>Email: {mayak, emhand, rama}@umiacs.umd.edu +</td></tr><tr><td>3b73f8a2b39751efb7d7b396bf825af2aaadee24</td><td>Connecting Pixels to Privacy and Utility: +<br/>Automatic Redaction of Private Information in Images +<br/><b>Max Planck Institute for Informatics</b><br/>Saarland Informatics Campus +<br/>Saabr¨ucken, Germany +</td><td>('9517443', 'Tribhuvanesh Orekondy', 'tribhuvanesh orekondy')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td>{orekondy,mfritz,schiele}@mpi-inf.mpg.de +</td></tr><tr><td>3b2d5585af59480531616fe970cb265bbdf63f5b</td><td>Robust Face Recognition under Varying Light +<br/>Based on 3D Recovery +<br/>Center of Computer Vision, School of +<br/>Mathematics and Computing, Sun Yat-sen +<br/><b>University, Guangzhou, China</b><br/>Ching Y Suen +<br/>Centre for Pattern Recognition and Machine +<br/><b>Intelligence, Concordia University, Montreal</b><br/>Canada, H3G 1M8 +</td><td>('3246510', 'Guan Yang', 'guan yang')</td><td>mcsfgc@mail.sysu.edu.cn +<br/>parmidir@cenparmi.concordia.ca +</td></tr><tr><td>3b64efa817fd609d525c7244a0e00f98feacc8b4</td><td>A Comprehensive Survey on Pose-Invariant +<br/>Face Recognition +<br/>Centre for Quantum Computation and Intelligent Systems +<br/>Faculty of Engineering and Information Technology +<br/><b>University of Technology, Sydney</b><br/>81-115 Broadway, Ultimo, NSW +<br/>Australia +<br/>15 March 2016 +</td><td>('37990555', 'Changxing Ding', 'changxing ding')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td>Emails: chx.ding@gmail.com, dacheng.tao@uts.edu.au +</td></tr><tr><td>3bc776eb1f4e2776f98189e17f0d5a78bb755ef4</td><td></td><td></td><td></td></tr><tr><td>3b7f6035a113b560760c5e8000540fc46f91fed5</td><td>COUPLING ALIGNMENTS WITH RECOGNITION FOR STILL-TO-VIDEO +<br/><b>Institute of Computing Technology, Chinese Academy of Sciences, Beijing, 100190, China</b><br/>FACE RECOGNITION +<br/>MOTIVATION +<br/>Problem: Still-to-Video face recognition +<br/>1. Gallery: high quality still face images (e.g., sharp and +<br/>high face resolution ones) +<br/>2. Probe: low quality video face frames (e.g., blur and low +<br/>face resolution ones) +<br/>Solution: Couple alignments with recognition +<br/>1. Quality Alignment (QA): select the frames of ‘best +<br/>quality’ from videos +<br/>2. Geometric Alignment (GA): jointly align the selected +<br/>frames to the still faces +<br/>3. Sparse Representation (SR): sparsely represent the +<br/>frames on the still faces +<br/>Frame 1 +<br/>20 +<br/>220 +<br/>301 +<br/>333 +<br/>image +<br/>OVERVIEW +<br/>GA: Geometric Alignment +<br/>SR: Sparse Representation +<br/>QA: Quality Alignment +<br/>T : Alignment parameters +<br/>L: Identity labels +<br/>C: Selecting confidences +<br/>FORMULATION +<br/>{ ˆT , ˆL} = arg minT,L (cid:107)Z(cid:107)1 +(cid:80)c +<br/>s.t. Y ◦ T = B + E, B = AZ, Si = {j|Lj = i}. +<br/>i=1 (cid:107)BSi(cid:107)∗ + (cid:107)E(cid:107)1, +<br/>• Couple GA with SR: Y ◦T = B+E, B = AZ, (cid:107)Z(cid:107)1 ≤ t +<br/>DATASETS +<br/>1. YouTube-S2V dataset: 100 subjects, privately +<br/>collected from YouTube Face DB [Wolf et al., CVPR’ 11] +<br/>2. COX-S2V dataset: 1,000 subjects, publicly released +<br/>in our prior work [Huang et al., ACCV ’12] +<br/>– Y : Video faces, A: dictionary (still faces) +<br/>– ◦ and T : Alignment operator and parameters +<br/>– B: Sparse representations, E: residual errors +<br/>Examples of still faces +<br/>Examples of still faces +<br/>• Couple SR with QA: Si = {j|Lj = i},(cid:80) (cid:107)BSi(cid:107)∗ ≤ k +<br/>– Identity label: Lj = arg mink (cid:107)yj ◦τj −Akzjk(cid:107)2 +<br/>– Confidence: Ci =(cid:80) +<br/>(cid:16) −(cid:107)ej(cid:107)1 +<br/>(cid:17) +<br/>j∈Si +<br/>exp +<br/>σ2 +<br/>Frame 1 +<br/>RESULTS +<br/>Frame 1 +<br/>31 +<br/>45 +<br/>72 +<br/>84 +<br/>Frame 1 +<br/>14 +<br/>25 +<br/>35 +<br/>46 +<br/>89 +<br/>Examples of video faces +<br/>118 +<br/>Frame 1 +<br/>Examples of video faces +<br/>14 +<br/>25 +<br/>OPTIMIZATION +<br/>{ ˆT , ˆL} = arg minT,L (cid:107)Z(cid:107)1 +(cid:80)c +<br/>Linearization: +<br/>i=1 (cid:107)BSi(cid:107)∗ + (cid:107)E(cid:107)1, +<br/>s.t.Y ◦ T + J∆T = B + E, B = AZ, Si = {j|Lj = i}. +<br/>∂T Y ◦ T : Jacobian matrices w.r.t transformations +<br/>J = ∂ +<br/>Main algorithm: +<br/>Comparative methods: +<br/>1. Baseline: SRC[1], CRC[2] +<br/>2. Blind Geometric Alignment: RASL[3] +<br/>3. Joint Geometric Alignment and Recognition: MRR[4] +<br/>4. Our method: Couping Alignments with Recognition +<br/>(CAR) +<br/>Evaluation terms: +<br/>1. Face Alignments (QA and GA) +<br/>2. Sparse Reprentation (SR) for Face Recognition +<br/>INPUT: Gallery data matrix A, probe video sequence +<br/>data matrix Y and initial transformation T of Y +<br/>1. WHILE not converged DO +<br/>2. Compute Jacobian matrices w.r.t transformations +<br/>3. Warp and normalize the images: +<br/>(cid:20) vec(Y1 ◦ τ1) +<br/>Y ◦ T = +<br/>Set the segments at coarse search stage: +<br/>vec((cid:107)Y1 ◦ τ1(cid:107)2) +<br/>S1 = {1, . . . , n}, Si = φ, i = 2, . . . , c +<br/>, . . . , +<br/>vec(Yn ◦ τn) +<br/>vec((cid:107)Yn ◦ τn(cid:107)2) +<br/>5. Apply Augmented Lagrange Multiplier to solve: +<br/>(cid:107)BSi(cid:107)∗ + (cid:107)E(cid:107)1, +<br/>{ ˆT , ˆZ} = arg min +<br/>(cid:107)Z(cid:107)1 + +<br/>c(cid:88) +<br/>4. +<br/>(cid:21) +<br/>T,Z +<br/>i=1 +<br/>s.t. Y ◦ T + J∆T = B + E, B = AZ; +<br/>6. Update transformations: T = T + ∆T ∗ +<br/>7. Update segments at fine search stage: +<br/>Si = {j|i = arg min +<br/>(cid:107)yj ◦ τj − Akzjk(cid:107)2}. +<br/>8. END WHILE +<br/>9. Compute Ci of Si, i = 1, . . . , n for voting class label. +<br/>OUTPUT: Class label of the probe video sequence. +<br/>QA, GA, SR results. +<br/>: correctly identified, (cid:3): finally selected +<br/>CONCLUSION +<br/>• The proposed method jointly performs GA, QA and SR +<br/>in a unified optimization. +<br/>• We employ an iterative EM-like algorithm to jointly op- +<br/>timize the three tasks. +<br/>• Experimental results demonstrate that GA, QA and SR +<br/>benefit from each other. +<br/>QA and GA results. Average faces of video frames finally +<br/>selected for face recognition +<br/>Methods +<br/>SRC[1] +<br/>CRC[2] +<br/>RASL[3] -SRC +<br/>RASL[3]-CRC +<br/>MRR[4] +<br/>CAR +<br/>10.78 +<br/>10.34 +<br/>26.29 +<br/>29.74 +<br/>28.45 +<br/>36.21 +<br/>C1 +<br/>15.57 +<br/>14.43 +<br/>22.14 +<br/>19.43 +<br/>26.43 +<br/>43.42 +<br/>C2 +<br/>42.29 +<br/>43.57 +<br/>39.00 +<br/>41.29 +<br/>44.14 +<br/>55.00 +<br/>C3 +<br/>2.86 +<br/>4.14 +<br/>4.57 +<br/>4.00 +<br/>3.57 +<br/>10.71 +<br/>C4 +<br/>18.71 +<br/>19.71 +<br/>18.29 +<br/>19.43 +<br/>13.57 +<br/>28.86 +<br/>Face recognition results. Intensity feature, Y: YouTube-S2V, Ci: +<br/>the i-the testing scenario of COX-S2V +<br/>REFERENCES +<br/>[1] +<br/>J. Wright, A. Yang, A. Ganesh, S. Sastry, Y. Ma. Robust face recognition via sparse representa- +<br/>tion. In TPAMI ’09 +<br/>[2] L. Zhang, M. Yang, X. Feng. Sparse representation or collaborative representation which helps +<br/>face recognition? In ICCV ’11 +<br/>[3] Y. Peng, A. Ganesh, J. Wright, W. Xu, Y. Ma. RASL: Robust alignement by sparse and low-rank +<br/>decomposition for linearly correlated images. In CVPR ’10 +<br/>[4] M. Yang, L. Zhang, D. Zhang. Efficient misalignment-robust representaion for real-time face +<br/>recognition. In ECCV ’12 +</td><td>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('1874505', 'Xiaowei Zhao', 'xiaowei zhao')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>3b2a2357b12cf0a5c99c8bc06ef7b46e40dd888e</td><td>Learning Person Trajectory Representations for Team Activity Analysis +<br/><b>Simon Fraser University</b></td><td>('10386960', 'Nazanin Mehrasa', 'nazanin mehrasa')<br/>('19198359', 'Yatao Zhong', 'yatao zhong')<br/>('2123865', 'Frederick Tung', 'frederick tung')<br/>('3004771', 'Luke Bornn', 'luke bornn')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td>{nmehrasa, yataoz, ftung, lbornn}@sfu.ca, mori@cs.sfu.ca +</td></tr><tr><td>3bd1d41a656c8159305ba2aa395f68f41ab84f31</td><td>Entity-based Opinion Mining from Text and +<br/>Multimedia +<br/>1 Introduction +<br/>Social web analysis is all about the users who are actively engaged and generate +<br/>content. This content is dynamic, reflecting the societal and sentimental fluctuations +<br/>of the authors as well as the ever-changing use of language. Social networks are +<br/>pools of a wide range of articulation methods, from simple ”Like” buttons to com- +<br/>plete articles, their content representing the diversity of opinions of the public. User +<br/>activities on social networking sites are often triggered by specific events and re- +<br/>lated entities (e.g. sports events, celebrations, crises, news articles) and topics (e.g. +<br/>global warming, financial crisis, swine flu). +<br/>With the rapidly growing volume of resources on the Web, archiving this material +<br/>becomes an important challenge. The notion of community memories extends tradi- +<br/>tional Web archives with related data from a variety of sources. In order to include +<br/>this information, a semantically-aware and socially-driven preservation model is a +<br/>natural way to go: the exploitation of Web 2.0 and the wisdom of crowds can make +<br/>web archiving a more selective and meaning-based process. The analysis of social +<br/>media can help archivists select material for inclusion, while social media mining +<br/>can enrich archives, moving towards structured preservation around semantic cat- +<br/>egories. In this paper, we focus on the challenges in the development of opinion +<br/>mining tools from both textual and multimedia content. +<br/>We focus on two very different domains: socially aware federated political +<br/>archiving (realised by the national parliaments of Greece and Austria), and socially +<br/>contextualized broadcaster web archiving (realised by two large multimedia broad- +<br/><b>University of Shef eld, Regent Court, 211 Portobello, Shef eld</b><br/>Jonathon Hare +<br/><b>Electronics and Computer Science, University of Southampton, Southampton, Hampshire</b></td><td>('2144272', 'Diana Maynard', 'diana maynard')<br/>('2144272', 'Diana Maynard', 'diana maynard')</td><td>S1 4DP, UK e-mail: diana@dcs.shef.ac.uk +<br/>SO17 1BJ, UK e-mail: jsh2@ecs.soton.ac.uk +</td></tr><tr><td>3bcd72be6fbc1a11492df3d36f6d51696fd6bdad</td><td>Multi-Task Zero-Shot Action Recognition with +<br/>Prioritised Data Augmentation +<br/>School of Electronic Engineering and Computer Science, +<br/><b>Queen Mary University of London</b></td><td>('1735328', 'Xun Xu', 'xun xu')<br/>('1697755', 'Timothy M. Hospedales', 'timothy m. hospedales')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')</td><td>{xun.xu,t.hospedales,s.gong}@qmul.ac.uk +</td></tr><tr><td>3b9c08381282e65649cd87dfae6a01fe6abea79b</td><td>CUHK & ETHZ & SIAT Submission to ActivityNet Challenge 2016 +<br/><b>Multimedia Laboratory, The Chinese University of Hong Kong, Hong Kong</b><br/>2Computer Vision Lab, ETH Zurich, Switzerland +<br/><b>Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('1915826', 'Zhe Wang', 'zhe wang')<br/>('3047890', 'Bowen Zhang', 'bowen zhang')<br/>('2313919', 'Hang Song', 'hang song')<br/>('1688012', 'Wei Li', 'wei li')<br/>('1807606', 'Dahua Lin', 'dahua lin')<br/>('1681236', 'Luc Van Gool', 'luc van gool')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>3b84d074b8622fac125f85ab55b63e876fed4628</td><td>End-to-End Localization and Ranking for +<br/>Relative Attributes +<br/><b>University of California, Davis</b></td><td>('19553871', 'Krishna Kumar Singh', 'krishna kumar singh')<br/>('1883898', 'Yong Jae Lee', 'yong jae lee')</td><td></td></tr><tr><td>3b4fd2aec3e721742f11d1ed4fa3f0a86d988a10</td><td>Glimpse: Continuous, Real-Time Object Recognition on +<br/>Mobile Devices +<br/>MIT CSAIL +<br/>Microsoft Research +<br/>MIT CSAIL +<br/>Microsoft Research +<br/>MIT CSAIL +</td><td>('32214366', 'Tiffany Yu-Han Chen', 'tiffany yu-han chen')<br/>('40125198', 'Lenin Ravindranath', 'lenin ravindranath')<br/>('1904357', 'Shuo Deng', 'shuo deng')<br/>('2292948', 'Paramvir Bahl', 'paramvir bahl')<br/>('1712771', 'Hari Balakrishnan', 'hari balakrishnan')</td><td>yuhan@csail.mit.edu +<br/>lenin@microsoft.com +<br/>shuodeng@csail.mit.edu +<br/>bahl@microsoft.com +<br/>hari@csail.mit.edu +</td></tr><tr><td>3be8f1f7501978287af8d7ebfac5963216698249</td><td>Deep Cascaded Regression for Face Alignment +<br/><b>School of Data and Computer Science, Sun Yat-Sen University, China</b><br/><b>National University of Singapore, Singapore</b><br/>algorithm refines the shape by estimating a shape increment +<br/>∆S. In particular, a shape increment at stage k is calculated +<br/>as: +</td><td>('3124720', 'Shengtao Xiao', 'shengtao xiao')<br/>('10338111', 'Zhen Cui', 'zhen cui')<br/>('48815683', 'Yan Pan', 'yan pan')<br/>('48258938', 'Chunyan Xu', 'chunyan xu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>3bc376f29bc169279105d33f59642568de36f17f</td><td>Active Shape Models with SIFT Descriptors and MARS +<br/><b>University of Cape Town, South Africa</b><br/>Keywords: +<br/>Facial Landmark, Active Shape Model, Multivariate Adaptive Regression Splines +</td><td>('2822258', 'Stephen Milborrow', 'stephen milborrow')<br/>('2537623', 'Fred Nicolls', 'fred nicolls')</td><td>milbo@sonic.net +</td></tr><tr><td>3b38c06caf54f301847db0dd622a6622c3843957</td><td>RESEARCH ARTICLE +<br/>Gender differences in emotion perception +<br/>and self-reported emotional intelligence: A +<br/>test of the emotion sensitivity hypothesis +<br/><b>University of Amsterdam, Amsterdam, the Netherlands, 2 Leiden University</b><br/><b>Leiden, the Netherlands, 3 Delft University of Technology</b><br/>Intelligent Systems, Delft, the Netherlands +</td><td>('1735303', 'Joost Broekens', 'joost broekens')</td><td>* a.h.fischer@uva.nl +</td></tr><tr><td>3b15a48ffe3c6b3f2518a7c395280a11a5f58ab0</td><td>On Knowledge Transfer in +<br/>Object Class Recognition +<br/>A dissertation approved by +<br/>TECHNISCHE UNIVERSITÄT DARMSTADT +<br/>Fachbereich Informatik +<br/>for the degree of +<br/>Doktor-Ingenieur (Dr.-Ing.) +<br/>presented by +<br/>Dipl.-Inform. +<br/>born in Mainz, Germany +<br/>Prof. Dr.-Ing. Michael Goesele, examiner +<br/>Prof. Martial Hebert, Ph.D., co-examiner +<br/>Prof. Dr. Bernt Schiele, co-examiner +<br/>Date of Submission: 12th of August, 2010 +<br/>Date of Defense: 23rd of September, 2010 +<br/>Darmstadt, 2010 +<br/>D17 +</td><td>('37718254', 'Michael Stark', 'michael stark')</td><td></td></tr><tr><td>3baa3d5325f00c7edc1f1427fcd5bdc6a420a63f</td><td>Enhancing Convolutional Neural Networks for Face Recognition with +<br/>Occlusion Maps and Batch Triplet Loss +<br/><b>aSchool of Engineering and Technology, University of Hertfordshire, Hat eld AL10 9AB, UK</b><br/>bIDscan Biometrics (a GBG company), London E14 9QD, UK +</td><td>('2133352', 'Li Meng', 'li meng')<br/>('46301106', 'Margaret Hartnett', 'margaret hartnett')</td><td></td></tr><tr><td>3b9b200e76a35178da940279d566bbb7dfebb787</td><td>Learning Channel Inter-dependencies at Multiple Scales on Dense +<br/>Networks for Face Recognition +<br/>109 Research Way — PO Box 6109 Morgantown, West Virginia +<br/><b>West Virginia University</b><br/>November 29, 2017 +</td><td>('16145333', 'Qiangchang Wang', 'qiangchang wang')<br/>('1822413', 'Guodong Guo', 'guodong guo')<br/>('23981570', 'Mohammad Iqbal Nouyed', 'mohammad iqbal nouyed')</td><td>qw0007@mix.wvu.edu, guodong.guo@mail.wvu.edu, monouyed@mix.wvu.edu +</td></tr><tr><td>3b408a3ca6fb39b0fda4d77e6a9679003b2dc9ab</td><td>Improving Classification by Improving Labelling: +<br/>Introducing Probabilistic Multi-Label Object Interaction Recognition +<br/>Walterio Mayol-Cuevas +<br/><b>University of Bristol</b></td><td>('2052236', 'Michael Wray', 'michael wray')<br/>('3420479', 'Davide Moltisanti', 'davide moltisanti')<br/>('1728459', 'Dima Damen', 'dima damen')</td><td><FirstName>.<LastName>@bristol.ac.uk +</td></tr><tr><td>3b02aaccc9f063ae696c9d28bb06a8cd84b2abb8</td><td>Who Leads the Clothing Fashion: Style, Color, or Texture? +<br/>A Computational Study +<br/><b>School of Computer Science, Wuhan University, P.R. China</b><br/><b>Shenzhen Key Laboratory of Spatial Smart Sensing and Service, Shenzhen University, P.R. China</b><br/><b>School of Data of Computer Science, Sun Yat-sen University, P.R. China</b><br/><b>University of South Carolina, USA</b></td><td>('4793870', 'Qin Zou', 'qin zou')<br/>('37361540', 'Zheng Zhang', 'zheng zhang')<br/>('40102806', 'Qian Wang', 'qian wang')<br/>('1720431', 'Qingquan Li', 'qingquan li')<br/>('40451106', 'Long Chen', 'long chen')<br/>('10829233', 'Song Wang', 'song wang')</td><td></td></tr><tr><td>3ba8f8b6bfb36465018430ffaef10d2caf3cfa7e</td><td>Local Directional Number Pattern for Face +<br/>Analysis: Face and Expression Recognition +</td><td>('2525887', 'Adin Ramirez Rivera', 'adin ramirez rivera')<br/>('1685505', 'Oksam Chae', 'oksam chae')</td><td></td></tr><tr><td>3b80bf5a69a1b0089192d73fa3ace2fbb52a4ad5</td><td></td><td></td><td></td></tr><tr><td>3b9d94752f8488106b2c007e11c193f35d941e92</td><td>CVPR +<br/>#2052 +<br/>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>045 +<br/>046 +<br/>047 +<br/>048 +<br/>049 +<br/>050 +<br/>051 +<br/>052 +<br/>053 +<br/>CVPR 2013 Submission #2052. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +<br/>CVPR +<br/>#2052 +<br/>Appearance, Visual and Social Ensembles for +<br/>Face Recognition in Personal Photo Collections +<br/>Anonymous CVPR submission +<br/>Paper ID 2052 +</td><td></td><td></td></tr><tr><td>3bb6570d81685b769dc9e74b6e4958894087f3f1</td><td>Hu-Fu: Hardware and Software Collaborative +<br/>Attack Framework against Neural Networks +<br/><b>Beijing National Research Center for Information Science and Technology</b><br/><b>Tsinghua University</b></td><td>('3493074', 'Wenshuo Li', 'wenshuo li')<br/>('1909938', 'Jincheng Yu', 'jincheng yu')<br/>('6636914', 'Xuefei Ning', 'xuefei ning')<br/>('2892980', 'Pengjun Wang', 'pengjun wang')<br/>('49988678', 'Qi Wei', 'qi wei')<br/>('47904166', 'Yu Wang', 'yu wang')<br/>('39150998', 'Huazhong Yang', 'huazhong yang')</td><td>{lws17@mails.tsinghua.edu.cn, yu-wang@tsinghua.edu.cn} +</td></tr><tr><td>3b557c4fd6775afc80c2cf7c8b16edde125b270e</td><td>Face Recognition: Perspectives from the +<br/>Real-World +<br/><b>Institute for Infocomm Research, A*STAR</b><br/>1 Fusionopolis Way, #21-01 Connexis (South Tower), Singapore 138632. +<br/>Phone: +65 6408 2071; Fax: +65 6776 1378; +</td><td>('1709001', 'Bappaditya Mandal', 'bappaditya mandal')</td><td>E-mail: bmandal@i2r.a-star.edu.sg +</td></tr><tr><td>3b3482e735698819a6a28dcac84912ec01a9eb8a</td><td>Individual Recognition Using Gait Energy Image +<br/>Center for Research in Intelligent Systems +<br/><b>University of California, Riverside, California 92521, USA</b><br/> +</td><td>('1699904', 'Ju Han', 'ju han')<br/>('1707159', 'Bir Bhanu', 'bir bhanu')</td><td>@cris.ucr.edu +</td></tr><tr><td>3b37d95d2855c8db64bd6b1ee5659f87fce36881</td><td>ADA: A Game-Theoretic Perspective on Data Augmentation for Object Detection +<br/><b>University of Illinois at Chicago</b><br/><b>Carnegie Mellon University</b><br/><b>University of Illinois at Chicago</b></td><td>('2761655', 'Sima Behpour', 'sima behpour')<br/>('37991449', 'Kris M. Kitani', 'kris m. kitani')<br/>('1753269', 'Brian D. Ziebart', 'brian d. ziebart')</td><td>sbehpo2@uic.edu +<br/>kkitani@cs.cmu.edu +<br/>bziebart@uic.edu +</td></tr><tr><td>3be7b7eb11714e6191dd301a696c734e8d07435f</td><td></td><td></td><td></td></tr><tr><td>3be027448ad49a79816cd21dcfcce5f4e1cec8a8</td><td>Actively Selecting Annotations Among Objects and Attributes +<br/><b>University of Texas at Austin</b></td><td>('1770205', 'Adriana Kovashka', 'adriana kovashka')<br/>('2259154', 'Sudheendra Vijayanarasimhan', 'sudheendra vijayanarasimhan')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>{adriana, svnaras, grauman}@cs.utexas.edu +</td></tr><tr><td>3bd56f4cf8a36dd2d754704bcb71415dcbc0a165</td><td>Robust Regression +<br/><b>Robotics Institute, Carnegie Mellon University</b></td><td>('39792229', 'Dong Huang', 'dong huang')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td></td></tr><tr><td>3b410ae97e4564bc19d6c37bc44ada2dcd608552</td><td>Scalability Analysis of Audio-Visual Person +<br/>Identity Verification +<br/>1 Communications Laboratory, +<br/>Universit´e catholique de Louvain, B-1348 Belgium, +<br/>2 IDIAP, CH-1920 Martigny, +<br/>Switzerland +</td><td>('34964585', 'Jacek Czyz', 'jacek czyz')<br/>('1751569', 'Samy Bengio', 'samy bengio')<br/>('2510802', 'Christine Marcel', 'christine marcel')<br/>('1698047', 'Luc Vandendorpe', 'luc vandendorpe')</td><td>czyz@tele.ucl.ac.be, +<br/>{Samy.Bengio,Christine.Marcel}@idiap.ch +</td></tr><tr><td>3b470b76045745c0ef5321e0f1e0e6a4b1821339</td><td>Consensus of Regression for Occlusion-Robust +<br/>Facial Feature Localization +<br/><b>Rutgers University, Piscataway, NJ 08854, USA</b><br/>2 Adobe Research, San Jose, CA 95110, USA +</td><td>('39960064', 'Xiang Yu', 'xiang yu')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td></td></tr><tr><td>6f5ce5570dc2960b8b0e4a0a50eab84b7f6af5cb</td><td>Low Resolution Face Recognition Using a +<br/>Two-Branch Deep Convolutional Neural Network +<br/>Architecture +</td><td>('19189138', 'Erfan Zangeneh', 'erfan zangeneh')<br/>('1772623', 'Mohammad Rahmati', 'mohammad rahmati')<br/>('3071758', 'Yalda Mohsenzadeh', 'yalda mohsenzadeh')</td><td></td></tr><tr><td>6f288a12033fa895fb0e9ec3219f3115904f24de</td><td>Learning Expressionlets via Universal Manifold +<br/>Model for Dynamic Facial Expression Recognition +</td><td>('1730228', 'Mengyi Liu', 'mengyi liu')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>6fa0c206873dcc5812f7ea74a48bb4bf4b273494</td><td>Real-time Mobile Facial Expression Recognition System – A Case Study +<br/>Department of Computer Engineering +<br/><b>The University of Texas at Dallas, Richardson, TX</b></td><td>('2774175', 'Myunghoon Suk', 'myunghoon suk')</td><td>{mhsuk, praba}@utdallas.edu +</td></tr><tr><td>6f9824c5cb5ac08760b08e374031cbdabc953bae</td><td>Unconstrained Human Identification Using Comparative Facial Soft Biometrics +<br/>Nawaf Y. Almudhahka +<br/><b>University of Southampton</b><br/>Southampton, United Kingdom +</td><td>('1727698', 'Mark S. Nixon', 'mark s. nixon')<br/>('31534955', 'Jonathon S. Hare', 'jonathon s. hare')</td><td>{nya1g14,msn,jsh2}@ecs.soton.ac.uk +</td></tr><tr><td>6f2dc51d607f491dbe6338711c073620c85351ac</td><td></td><td></td><td></td></tr><tr><td>6fed504da4e192fe4c2d452754d23d3db4a4e5e3</td><td>Learning Deep Features via Congenerous Cosine Loss for Person Recognition +<br/>1 SenseTime Group Ltd., Beijing, China +<br/><b>The Chinese University of Hong Kong, New Territories, Hong Kong</b></td><td>('1715752', 'Yu Liu', 'yu liu')<br/>('1929886', 'Hongyang Li', 'hongyang li')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')</td><td>liuyu@sensetime.com, {yangli, xgwang}@ee.cuhk.edu.hk +</td></tr><tr><td>6f957df9a7d3fc4eeba53086d3d154fc61ae88df</td><td>Mod´elisation et suivi des d´eformations faciales : +<br/>applications `a la description des expressions du visage +<br/>dans le contexte de la langue des signes +<br/>To cite this version: +<br/>des expressions du visage dans le contexte de la langue des signes. Interface homme-machine +<br/>[cs.HC]. Universit´e Paul Sabatier - Toulouse III, 2007. Fran¸cais. <tel-00185084> +<br/>HAL Id: tel-00185084 +<br/>https://tel.archives-ouvertes.fr/tel-00185084 +<br/>Submitted on 5 Nov 2007 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('3029015', 'Hugo Mercier', 'hugo mercier')<br/>('3029015', 'Hugo Mercier', 'hugo mercier')</td><td></td></tr><tr><td>6f26ab7edd971148723d9b4dc8ddf71b36be9bf7</td><td>Differences in Abundances of Cell-Signalling Proteins in +<br/>Blood Reveal Novel Biomarkers for Early Detection Of +<br/>Clinical Alzheimer’s Disease +<br/><b>Centre for Bioinformatics, Biomarker Discovery and Information-Based Medicine, The University of Newcastle, Callaghan, Australia, 2 Departamento de Engenharia de</b><br/>Produc¸a˜o, Universidade Federal de Minas Gerais (UFMG), Belo Horizonte, Brazil +</td><td>('8423987', 'Mateus Rocha de Paula', 'mateus rocha de paula')<br/>('34861417', 'Regina Berretta', 'regina berretta')<br/>('1738680', 'Pablo Moscato', 'pablo moscato')</td><td></td></tr><tr><td>6f75697a86d23d12a14be5466a41e5a7ffb79fad</td><td></td><td></td><td></td></tr><tr><td>6f7d06ced04ead3b9a5da86b37e7c27bfcedbbdd</td><td>Pages 51.1-51.12 +<br/>DOI: https://dx.doi.org/10.5244/C.30.51 +</td><td></td><td></td></tr><tr><td>6f7a8b3e8f212d80f0fb18860b2495be4c363eac</td><td>Creating Capsule Wardrobes from Fashion Images +<br/>UT-Austin +<br/>UT-Austin +</td><td>('22211024', 'Wei-Lin Hsiao', 'wei-lin hsiao')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>kimhsiao@cs.utexas.edu +<br/>grauman@cs.utexas.edu +</td></tr><tr><td>6f6b4e2885ea1d9bea1bb2ed388b099a5a6d9b81</td><td>Structured Output SVM Prediction of Apparent Age, +<br/>Gender and Smile From Deep Features +<br/>Michal Uˇriˇc´aˇr +<br/>CMP, Dept. of Cybernetics +<br/>FEE, CTU in Prague +<br/>Computer Vision Lab +<br/>D-ITET, ETH Zurich +<br/>Computer Vision Lab +<br/>D-ITET, ETH Zurich +<br/>PSI, ESAT, KU Leuven +<br/>CVL, D-ITET, ETH Zurich +<br/>Jiˇr´ı Matas +<br/>CMP, Dept. of Cybernetics +<br/>FEE, CTU in Prague +</td><td>('1732855', 'Radu Timofte', 'radu timofte')<br/>('2173683', 'Rasmus Rothe', 'rasmus rothe')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>uricamic@cmp.felk.cvut.cz +<br/>radu.timofte@vision.ee.ethz.ch +<br/>rrothe@vision.ee.ethz.ch +<br/>vangool@vision.ee.ethz.ch +<br/>matas@cmp.felk.cvut.cz +</td></tr><tr><td>6f08885b980049be95a991f6213ee49bbf05c48d</td><td>This article appeared in a journal published by Elsevier. The attached +<br/>copy is furnished to the author for internal non-commercial research +<br/><b>and education use, including for instruction at the authors institution</b><br/>and sharing with colleagues. +<br/><b>Other uses, including reproduction and distribution, or selling or</b><br/>licensing copies, or posting to personal, institutional or third party +<br/>websites are prohibited. +<br/>In most cases authors are permitted to post their version of the +<br/>article (e.g. in Word or Tex form) to their personal website or +<br/>institutional repository. Authors requiring further information +<br/>regarding Elsevier’s archiving and manuscript policies are +<br/>encouraged to visit: +<br/>http://www.elsevier.com/authorsrights +</td><td></td><td></td></tr><tr><td>6f0900a7fe8a774a1977c5f0a500b2898bcbe149</td><td>1 +<br/>Quotient Based Multiresolution Image Fusion of Thermal +<br/>and Visual Images Using Daubechies Wavelet Transform +<br/>for Human Face Recognition +<br/><b>Tripura University (A Central University</b><br/>Suryamaninagar, Tripura 799130, India +<br/><b>Jadavpur University</b><br/>Kolkata, West Bengal 700032, India +<br/>*AICTE Emeritus Fellow +</td><td>('1694317', 'Mrinal Kanti Bhowmik', 'mrinal kanti bhowmik')<br/>('1721942', 'Debotosh Bhattacharjee', 'debotosh bhattacharjee')<br/>('1729425', 'Mita Nasipuri', 'mita nasipuri')<br/>('1679476', 'Dipak Kumar Basu', 'dipak kumar basu')<br/>('1727663', 'Mahantapas Kundu', 'mahantapas kundu')</td><td>mkb_cse@yahoo.co.in +<br/>debotosh@indiatimes.com, mitanasipuri@gmail.com, dipakkbasu@gmail.com, mkundu@cse.jdvu.ac.in +</td></tr><tr><td>6fea198a41d2f6f73e47f056692f365c8e6b04ce</td><td>Video Captioning with Boundary-aware Hierarchical Language +<br/>Decoding and Joint Video Prediction +<br/><b>Nanyang Technological University</b><br/><b>Nanyang Technological University</b><br/>Singapore, Singapore +<br/>Singapore, Singapore +<br/><b>Nanyang Technological University</b><br/>Singapore, Singapore +<br/>Shafiq Joty +<br/><b>Nanyang Technological University</b><br/>Singapore, Singapore +</td><td>('8668622', 'Xiangxi Shi', 'xiangxi shi')<br/>('1688642', 'Jianfei Cai', 'jianfei cai')<br/>('2174964', 'Jiuxiang Gu', 'jiuxiang gu')</td><td>xxshi@ntu.edu.sg +<br/>JGU004@e.ntu.edu.sg +<br/>asjfcai@ntu.edu.sg +<br/>srjoty@ntu.edu.sg +</td></tr><tr><td>6fbb179a4ad39790f4558dd32316b9f2818cd106</td><td>Input Aggregated Network for Face Video Representation +<br/><b>Beijing Laboratory of IIT, School of Computer Science, Beijing Institute of Technology, Beijing, China</b><br/><b>Stony Brook University, Stony Brook, USA</b></td><td>('40061483', 'Zhen Dong', 'zhen dong')<br/>('3306427', 'Su Jia', 'su jia')<br/>('1690083', 'Chi Zhang', 'chi zhang')<br/>('35371203', 'Mingtao Pei', 'mingtao pei')</td><td></td></tr><tr><td>6f84e61f33564e5188136474f9570b1652a0606f</td><td>Dual Motion GAN for Future-Flow Embedded Video Prediction +<br/><b>Carnegie Mellon University</b></td><td>('40250403', 'Xiaodan Liang', 'xiaodan liang')<br/>('3682478', 'Lisa Lee', 'lisa lee')</td><td>{xiaodan1,lslee}@cs.cmu.edu +</td></tr><tr><td>6f35b6e2fa54a3e7aaff8eaf37019244a2d39ed3</td><td>DOI 10.1007/s00530-005-0177-4 +<br/>R E G U L A R PA P E R +<br/>Learning probabilistic classifiers for human–computer +<br/>interaction applications +<br/>Published online: 10 May 2005 +<br/>c(cid:1) Springer-Verlag 2005 +<br/>intelligent +<br/>interaction, +</td><td>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td></td></tr><tr><td>6f3054f182c34ace890a32fdf1656b583fbc7445</td><td>Article +<br/>Age Estimation Robust to Optical and Motion +<br/>Blurring by Deep Residual CNN +<br/><b>Division of Electronics and Electrical Engineering, Dongguk University, 30 Pildong-ro, 1-gil, Jung-gu</b><br/>Received: 9 March 2018; Accepted: 10 April 2018; Published: 13 April 2018 +</td><td>('31515471', 'Jeon Seong Kang', 'jeon seong kang')<br/>('31864414', 'Chan Sik Kim', 'chan sik kim')<br/>('29944844', 'Se Woon Cho', 'se woon cho')<br/>('4634733', 'Kang Ryoung Park', 'kang ryoung park')</td><td>Seoul 100-715, Korea; kjs2605@dgu.edu (J.S.K.); kimchsi9004@naver.com (C.S.K.); +<br/>lyw941021@dongguk.edu (Y.W.L.); jsu319@naver.com (S.W.C.) +<br/>* Correspondence: parkgr@dongguk.edu; Tel.: +82-10-3111-7022; Fax: +82-2-2277-8735 +</td></tr><tr><td>6fa3857faba887ed048a9e355b3b8642c6aab1d8</td><td>Face Recognition in Challenging Environments: +<br/>An Experimental and Reproducible Research +<br/>Survey +</td><td>('2121764', 'Laurent El Shafey', 'laurent el shafey')</td><td></td></tr><tr><td>6fda12c43b53c679629473806c2510d84358478f</td><td>Journal of Academic and Applied Studies +<br/>Vol. 1(1), June 2011, pp. 29-38 +<br/>A Training Model for Fuzzy Classification +<br/>System +<br/> +<br/><b>Islamic Azad University</b><br/>Iran +</td><td></td><td>Available online @ www.academians.org +<br/>Email:a.jamshidnejad@yahoo.com +</td></tr><tr><td>6fef65bd7287b57f0c3b36bf8e6bc987fd161b7d</td><td>Deep Discriminative Model for Video +<br/>Classification +<br/>Center for Machine Vision and Signal Analysis (CMVS) +<br/><b>University of Oulu, Finland</b></td><td>('2014145', 'Mohammad Tavakolian', 'mohammad tavakolian')<br/>('1751372', 'Abdenour Hadid', 'abdenour hadid')</td><td>firstname.lastname@oulu.fi +</td></tr><tr><td>6f7ce89aa3e01045fcd7f1c1635af7a09811a1fe</td><td>978-1-4673-0046-9/12/$26.00 ©2012 IEEE +<br/>937 +<br/>ICASSP 2012 +</td><td></td><td></td></tr><tr><td>6fe2efbcb860767f6bb271edbb48640adbd806c3</td><td>SOFT BIOMETRICS: HUMAN IDENTIFICATION USING COMPARATIVE DESCRIPTIONS +<br/>Soft Biometrics; Human Identification using +<br/>Comparative Descriptions +</td><td>('34386180', 'Daniel A. Reid', 'daniel a. reid')<br/>('1727698', 'Mark S. Nixon', 'mark s. nixon')<br/>('2093843', 'Sarah V. Stevenage', 'sarah v. stevenage')</td><td></td></tr><tr><td>6fdc0bc13f2517061eaa1364dcf853f36e1ea5ae</td><td>DAISEE: Dataset for Affective States in +<br/>E-Learning Environments +<br/>1 Microsoft India R&D Pvt. Ltd. +<br/>2 Department of Computer Science, IIT Hyderabad +</td><td>('50178849', 'Abhay Gupta', 'abhay gupta')<br/>('3468123', 'Richik Jaiswal', 'richik jaiswal')<br/>('3468212', 'Sagar Adhikari', 'sagar adhikari')<br/>('1973980', 'Vineeth Balasubramanian', 'vineeth balasubramanian')</td><td>abhgup@microsoft.com +<br/>{cs12b1032, cs12b1034, vineethnb}@iith.ac.in +</td></tr><tr><td>6f5151c7446552fd6a611bf6263f14e729805ec7</td><td>5KHHAO /7 %:0 7 +<br/>)>IJH=?J 9EJDE JDA ?JANJ B=?A ANFHAIIE ?=IIE?=JE KIEC JDA +<br/>FH>=>EEJEAI JD=J A=?D A B IALAH= ?O ??KHHEC )7 CHKFI EI +<br/>?=IIIAF=H=>EEJO MAECDJEC +<br/>/=>H M=LAAJI H FHE?EF= ?FAJI ==OIEI 2+) ! 1 JDEI F=FAH MA +</td><td></td><td>.=?E= )?JE 7EJ 4A?CEJE KIEC .EJAHA@ +<br/>?= *E=HO 2=JJAH .A=JKHAI MEJD *JIJH=FFA@ +<br/>=@ 9AECDJA@ -++ +=IIEAHI +<br/>455EJD =@ 69E@A=JJ +<br/>+AJHA BH 8EIE 5FAA?D =@ 5EC= 2H?AIIEC 7ELAHIEJO B 5KHHAO /KE@BH@ +<br/>4=O@5EJD 69E@A=JJ(IKHHAO=?K +<br/>B=?E= =?JE ?@EC IOIJA .)+5 MA =@@HAII JDA FH>A B @AJA?J +<br/>EC B=?E= =?JE KEJI )7I 6DA AJD@ =@FJA@ EI J JH=E = IECA +<br/>AHHH?HHA?JEC KJFKJ ?@A -++ KJE?=II ?=IIEAH J AIJE=JA JDA +<br/>FHAIAJ E JDA FH>A E=CA 2=JJ I?=EC EI KIA@ J ?=E>H=JA JDA -++ +<br/>KJFKJI J FH>=>EEJEAI =@ =FFHFHE=JA IKI B JDAIA FH>=>EEJEAI =HA +<br/>J=A J >J=E = IAF=H=JA FH>=>EEJO BH A=?D )7 E@ELE@K=O .A=JKHA +<br/>ANJH=?JE EI FAHBHA@ >O CAAH=JEC = =HCA K>AH B ?= >E=HO F=J +<br/>JAH *2 BA=JKHAI =@ JDA IAA?JEC BH JDAIA KIEC B=IJ ?HHA=JE +<br/>>=IA@ JAHEC .+*. 6DA >E=I =@ L=HE=?A FHFAHJEAI B JDA ?=IIEAH +<br/>=HA A=IKHA@ =@ MA IDM JD=J >JD JDAIA IKH?AI B AHHH ?= >A HA +<br/>@K?A@ >O AD=?EC -++ JDHKCD JDA =FFE?=JE B >JIJH=FFEC =@ +<br/>1JH@K?JE +<br/>6DA B=?E==?JE ?@EC IOIJA .)+5 B -= =@ .HEAIA EI ?O +<br/>AFOA@ E =FFE?=JEI MDE?D FAHBH =KJ=JE? B=?E= ANFHAIIE HA?CEJE +<br/>1 JDEI AJD@ E@ELE@K= B=?E= LAAJI =HA ?D=H=?JAHEIA@ =I A B "" JOFAI +<br/>M =I =?JE KEJI )7I /HKFI B )7I =O JDA >A =FFA@ J AJEI +<br/>KIEC = IJ=@=H@ ?@A > JA DMALAH JD=J )7I =HA J A?AII=HEO E@A +<br/>FA@AJ =I JDA FHAIA?A B A )7 =O =A?J JDA =FFA=H=?A B =JDAH 6DAO +<br/>=O =I ??KH =J @EAHAJ EJAIEJEAI =@ =O ??KH O A IE@A B JDA +<br/>B=?A 1 JDEI F=FAH MA B?KI HA?CEIEC IEN )7I BH JDA HACE =HK@ JDA +<br/>AOAI =I EKIJH=JA@ E .EC +<br/>1EJE= HAFHAIAJ=JE AJD@I BH )7 ?=IIE?=JE MAHA >=IA@ A=IKHEC +<br/>JDA HA=JELA FIEJE B = =HCA K>AH B =@=H FEJI JDA B=?A 1J +<br/>D=I >AA BK@ DMALAH JD=J ?F=H=>A H >AJJAH HAIKJI ?= >A >J=EA@ >O +<br/>J=EC = HA DEIJE? =FFH=?D J BA=JKHA ANJH=?JE KIEC AJD@I IK?D =I +<br/>?F=HA JM IK?D AJD@I =AO 2+) " =@ ?= >E=HO F=JJAH *2 +</td></tr><tr><td>03c56c176ec6377dddb6a96c7b2e95408db65a7a</td><td>A Novel Geometric Framework on Gram Matrix +<br/>Trajectories for Human Behavior Understanding +</td><td>('46243486', 'Anis Kacem', 'anis kacem')<br/>('2909056', 'Mohamed Daoudi', 'mohamed daoudi')<br/>('2125606', 'Boulbaba Ben Amor', 'boulbaba ben amor')<br/>('2507859', 'Stefano Berretti', 'stefano berretti')</td><td></td></tr><tr><td>03d9ccce3e1b4d42d234dba1856a9e1b28977640</td><td></td><td></td><td></td></tr><tr><td>0322e69172f54b95ae6a90eb3af91d3daa5e36ea</td><td>Face Classification using Adjusted Histogram in +<br/>Grayscale +</td><td></td><td></td></tr><tr><td>036c41d67b49e5b0a578a401eb31e5f46b3624e0</td><td>The Tower Game Dataset: A Multimodal Dataset +<br/>for Analyzing Social Interaction Predicates +<br/>∗ SRI International +<br/><b>University of California, Santa Cruz</b><br/><b>University of California, Berkeley</b></td><td>('1955011', 'David A. Salter', 'david a. salter')<br/>('1860011', 'Amir Tamrakar', 'amir tamrakar')<br/>('1832513', 'Behjat Siddiquie', 'behjat siddiquie')<br/>('4599641', 'Mohamed R. Amer', 'mohamed r. amer')<br/>('1696401', 'Ajay Divakaran', 'ajay divakaran')<br/>('40530418', 'Brian Lande', 'brian lande')<br/>('2108704', 'Darius Mehri', 'darius mehri')</td><td>Email: {david.salter, amir.tamrakar, behjat.siddiquie, mohamed.amer, ajay.divakaran}@sri.com +<br/>Email: brianlande@soe.ucsc.edu +<br/>Email: darius mehri@berkeley.edu +</td></tr><tr><td>03b03f5a301b2ff88ab3bb4969f54fd9a35c7271</td><td>Multi-kernel learning of deep convolutional features for action recognition +<br/><b>Imperial College London</b><br/>Noah’s Ark Lab (Huawei Technologies UK) +<br/>Cortexica Vision Systems Limited +</td><td>('39599054', 'Biswa Sengupta', 'biswa sengupta')<br/>('29742002', 'Yu Qian', 'yu qian')</td><td>b.sengupta@imperial.ac.uk +</td></tr><tr><td>03f7041515d8a6dcb9170763d4f6debd50202c2b</td><td>Clustering Millions of Faces by Identity +</td><td>('40653304', 'Charles Otto', 'charles otto')<br/>('7496032', 'Dayong Wang', 'dayong wang')<br/>('40217643', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>03ce2ff688f9b588b6f264ca79c6857f0d80ceae</td><td>Attention Clusters: Purely Attention Based +<br/>Local Feature Integration for Video Classification +<br/><b>Tsinghua University, 2Rutgers University, 3Massachusetts Institute of Technology, 4Baidu IDL</b></td><td>('1716690', 'Xiang Long', 'xiang long')<br/>('2551285', 'Chuang Gan', 'chuang gan')<br/>('1732213', 'Gerard de Melo', 'gerard de melo')<br/>('3045089', 'Jiajun Wu', 'jiajun wu')<br/>('48033101', 'Xiao Liu', 'xiao liu')<br/>('35247507', 'Shilei Wen', 'shilei wen')</td><td></td></tr><tr><td>03b99f5abe0e977ff4c902412c5cb832977cf18e</td><td>CROWLEY AND ZISSERMAN: OF GODS AND GOATS +<br/>Of Gods and Goats: Weakly Supervised +<br/>Learning of Figurative Art +<br/>Elliot J. Crowley +<br/>Department of Engineering Science +<br/><b>University of Oxford</b></td><td>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>elliot@robots.ox.ac.uk +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>038ce930a02d38fb30d15aac654ec95640fe5cb0</td><td>Approximate Structured Output Learning for Constrained Local +<br/>Models with Application to Real-time Facial Feature Detection and +<br/>Tracking on Low-power Devices +</td><td>('40474289', 'Shuai Zheng', 'shuai zheng')<br/>('3274976', 'Paul Sturgess', 'paul sturgess')<br/>('1730268', 'Philip H. S. Torr', 'philip h. s. torr')</td><td></td></tr><tr><td>03167776e17bde31b50f294403f97ee068515578</td><td>Chapter 11. Facial Expression Analysis +<br/><b>University of Pittsburgh, Pittsburgh, PA 15260, USA</b><br/>1 Principles of Facial Expression Analysis +<br/>1.1 What Is Facial Expression Analysis? +<br/>Facial expressions are the facial changes in response to a person’s internal emotional states, +<br/>intentions, or social communications. Facial expression analysis has been an active research +<br/>topic for behavioral scientists since the work of Darwin in 1872 [18, 22, 25, 71]. Suwa et +<br/>al. [76] presented an early attempt to automatically analyze facial expressions by tracking the +<br/>motion of 20 identified spots on an image sequence in 1978. After that, much progress has +<br/>been made to build computer systems to help us understand and use this natural form of human +<br/>communication [6, 7, 17, 20, 28, 39, 51, 55, 65, 78, 81, 92, 93, 94, 96]. +<br/>In this chapter, facial expression analysis refers to computer systems that attempt to auto- +<br/>matically analyze and recognize facial motions and facial feature changes from visual informa- +<br/>tion. Sometimes the facial expression analysis has been confused with emotion analysis in the +<br/>computer vision domain. For emotion analysis, higher level knowledge is required. For exam- +<br/>ple, although facial expressions can convey emotion, they can also express intention, cognitive +<br/>processes, physical effort, or other intra- or interpersonal meanings. Interpretation is aided by +<br/>context, body gesture, voice, individual differences, and cultural factors as well as by facial +<br/>configuration and timing [10, 67, 68]. Computer facial expression analysis systems need to +<br/>analyze the facial actions regardless of context, culture, gender, and so on. +<br/>The accomplishments in the related areas such as psychological studies, human movement +<br/>analysis, face detection, face tracking, and recognition make the automatic facial expression +<br/>analysis possible. Automatic facial expression analysis can be applied in many areas such as +<br/>emotion and paralinguistic communication, clinical psychology, psychiatry, neurology, pain +<br/>assessment, lie detection, intelligent environments, and multimodal human computer interface +<br/>(HCI). +<br/>1.2 Basic Structure of Facial Expression Analysis Systems +<br/>Facial expression analysis includes both measurement of facial motion and recognition of ex- +<br/>pression. The general approach to automatic facial expression analysis (AFEA) consists of +</td><td>('40383812', 'Ying-Li Tian', 'ying-li tian')<br/>('1733113', 'Takeo Kanade', 'takeo kanade')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>1 IBM T. J. Watson Research Center, Hawthorne, NY 10532, USA. yltian@us.ibm.com +<br/>2 Robotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213, USA. tk@cs.cmu.edu +<br/>jeffcohn@pitt.edu +</td></tr><tr><td>0334a8862634988cc684dacd4279c5c0d03704da</td><td>FaceNet2ExpNet: Regularizing a Deep Face Recognition Net for +<br/>Expression Recognition +<br/><b>University of Maryland, College Park</b><br/>2 Siemens Healthcare Technology Center, Princeton, New Jersey +</td><td>('1700765', 'Hui Ding', 'hui ding')<br/>('1682187', 'Shaohua Kevin Zhou', 'shaohua kevin zhou')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>03c1fc9c3339813ed81ad0de540132f9f695a0f8</td><td>Proceedings of Machine Learning Research 81:1–15, 2018 +<br/>Conference on Fairness, Accountability, and Transparency +<br/>Gender Shades: Intersectional Accuracy Disparities in +<br/>Commercial Gender Classification∗ +<br/>MIT Media Lab 75 Amherst St. Cambridge, MA 02139 +<br/>Microsoft Research 641 Avenue of the Americas, New York, NY 10011 +<br/>Editors: Sorelle A. Friedler and Christo Wilson +</td><td>('38222513', 'Joy Buolamwini', 'joy buolamwini')<br/>('2076288', 'Timnit Gebru', 'timnit gebru')</td><td>joyab@mit.edu +<br/>timnit.gebru@microsoft.com +</td></tr><tr><td>0339459a5b5439d38acd9c40a0c5fea178ba52fb</td><td>D|C|I&I 2009 Prague +<br/>Multimodal recognition of emotions in car +<br/>environments +</td><td></td><td></td></tr><tr><td>030ef31b51bd4c8d0d8f4a9a32b80b9192fe4c3f</td><td>11936 • The Journal of Neuroscience, August 26, 2015 • 35(34):11936 –11945 +<br/>Behavioral/Cognitive +<br/>Inhibition-Induced Forgetting Results from Resource +<br/>Competition between Response Inhibition and Memory +<br/>Encoding Processes +<br/><b>Center for Cognitive Neuroscience, Duke University, Durham, North Carolina</b><br/>Response inhibition is a key component of executive control, but its relation to other cognitive processes is not well understood. We +<br/>recently documented the “inhibition-induced forgetting effect”: no-go cues are remembered more poorly than go cues. We attributed this +<br/>effect to central-resource competition, whereby response inhibition saps attention away from memory encoding. However, this proposal +<br/>is difficult to test with behavioral means alone. We therefore used fMRI in humans to test two neural predictions of the “common resource +<br/>hypothesis”: (1) brain regions associated with response inhibition should exhibit greater resource demands during encoding of subse- +<br/>quently forgotten than remembered no-go cues; and (2) this higher inhibitory resource demand should lead to memory encoding regions +<br/>having less resources available during encoding of subsequently forgotten no-go cues. Participants categorized face stimuli by gender in +<br/>a go/no-go task and, following a delay, performed a surprise recognition memory test for those faces. Replicating previous findings, +<br/>memory was worse for no-go than for go stimuli. Crucially, forgetting of no-go cues was predicted by high inhibitory resource demand, as +<br/>quantified by the trial-by-trial ratio of activity in neural “no-go” versus “go” networks. Moreover, this index of inhibitory demand +<br/>exhibited an inverse trial-by-trial relationship with activity in brain regions responsible for the encoding of no-go cues into memory, +<br/>notably the ventrolateral prefrontal cortex. This seesaw pattern between the neural resource demand of response inhibition and activity +<br/>related to memory encoding directly supports the hypothesis that response inhibition temporarily saps attentional resources away from +<br/>stimulus processing. +<br/>Key words: attention; cognitive control; memory; response inhibition +<br/>Significance Statement +<br/>Recent behavioral experiments showed that inhibiting a motor response to a stimulus (a “no-go cue”) impairs subsequent +<br/>memory for that cue. Here, we used fMRI to test whether this “inhibition-induced forgetting effect” is caused by competition for +<br/>neural resources between the processes of response inhibition and memory encoding. We found that trial-by-trial variations in +<br/>neural inhibitory resource demand predicted subsequent forgetting of no-go cues and that higher inhibitory demand was further- +<br/>more associated with lower concurrent activation in brain regions responsible for successful memory encoding of no-go cues. +<br/>Thus, motor inhibition and stimulus encoding appear to compete with each other: when more resources have to be devoted to +<br/>inhibiting action, less are available for encoding sensory stimuli. +<br/>Introduction +<br/>Response inhibition, the ability to preempt or cancel goal- +<br/>inappropriate actions, is considered a core cognitive control +<br/>Received Feb. 6, 2015; revised July 22, 2015; accepted July 24, 2015. +<br/>Author contributions: Y.-C.C. and T.E. designed research; Y.-C.C. performed research; Y.-C.C. analyzed data; +<br/>Y.-C.C. and T.E. wrote the paper. +<br/><b>This work was supported in part by National Institute of Mental Health Award R01 MH 087610 to T.E</b><br/>The authors declare no competing financial interests. +<br/>DOI:10.1523/JNEUROSCI.0519-15.2015 +<br/>Copyright © 2015 the authors +<br/>0270-6474/15/3511936-10$15.00/0 +<br/>function (Logan and Cowan, 1984; Aron, 2007), an impairment +<br/>that contributes to impulsive symptoms of multiple psychiatric +<br/>diseases, +<br/>including obsessive-compulsive disorder, substance +<br/>abuse, and attention-deficit/hyperactivity disorder (Horn et al., +<br/>2003; de Wit, 2009). However, the relation of response inhibition +<br/>to other cognitive control functions, and to traditional cognitive +<br/>domains, such as perception, memory, and attention, remains +<br/>poorly understood (Jurado and Rosselli, 2007; Miyake and Fried- +<br/>man, 2012). +<br/>A recent behavioral study has shed new light on this issue by +<br/>documenting an “inhibition-induced forgetting” effect, whereby +<br/>inhibiting responses to no-go or stop cues impaired subsequent +</td><td>('2846298', 'Yu-Chin Chiu', 'yu-chin chiu')<br/>('1900710', 'Tobias Egner', 'tobias egner')<br/>('2846298', 'Yu-Chin Chiu', 'yu-chin chiu')</td><td>LSRC, Box 90999, Durham, NC 27708. E-mail: chiu.yuchin@duke.edu. +</td></tr><tr><td>03f98c175b4230960ac347b1100fbfc10c100d0c</td><td>Supervised Descent Method and its Applications to Face Alignment +<br/><b>The Robotics Institute, Carnegie Mellon University, Pittsburgh PA</b></td><td>('3182065', 'Xuehan Xiong', 'xuehan xiong')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td>xxiong@andrew.cmu.edu +<br/>ftorre@cs.cmu.edu +</td></tr><tr><td>032825000c03b8ab4c207e1af4daeb1f225eb025</td><td>J. Appl. Environ. Biol. Sci., 7(10)159-164, 2017 +<br/>ISSN: 2090-4274 +<br/>© 2017, TextRoad Publication +<br/>Journal of Applied Environmental +<br/>and Biological Sciences +<br/>www.textroad.com +<br/>A Novel Approach for Human Face Detection in Color Images Using Skin +<br/>Color and Golden Ratio +<br/><b>Bacha Khan University, Charsadda, KPK, Pakistan</b><br/><b>Abdul WaliKhan University, Mardan, KPK, Pakistan</b><br/>Received: May 9, 2017 +<br/>Accepted: August 2, 2017 +</td><td>('12144785', 'Faizan Ullah', 'faizan ullah')<br/>('49669073', 'Dilawar Shah', 'dilawar shah')<br/>('46463663', 'Sabir Shah', 'sabir shah')<br/>('47160013', 'Abdus Salam', 'abdus salam')<br/>('12579194', 'Shujaat Ali', 'shujaat ali')</td><td></td></tr><tr><td>03264e2e2709d06059dd79582a5cc791cbef94b1</td><td>Convolutional Neural Networks for Facial Attribute-based Active Authentication +<br/>On Mobile Devices +<br/><b>University of Maryland, College Park</b><br/><b>University of Maryland, College Park</b><br/>MD, USA +<br/>MD, USA +</td><td>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('3383048', 'Pouya Samangouei', 'pouya samangouei')</td><td>pouya@umiacs.umd.org +<br/>rama@umiacs.umd.edu +</td></tr><tr><td>03a8f53058127798bc2bc0245d21e78354f6c93b</td><td>Max-Margin Additive Classifiers for Detection +<br/>Sam Hare +<br/>VGG Reading Group +<br/>October 30, 2009 +</td><td>('35208858', 'Subhransu Maji', 'subhransu maji')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')</td><td></td></tr><tr><td>03fc466fdbc8a2efb6e3046fcc80e7cb7e86dc20</td><td>A Real Time System for Model-based Interpretation of +<br/>the Dynamics of Facial Expressions +<br/>Technische Universit¨at M¨unchen +<br/>Boltzmannstr. 3, 85748 Garching +<br/>1. Motivation +<br/>Recent progress in the field of Computer Vision allows +<br/>intuitive interaction via speech, gesture or facial expressions +<br/>between humans and technical systems.Model-based tech- +<br/>niques facilitate accurately interpreting images with faces +<br/>by exploiting a priori knowledge, such as shape and texture +<br/>information. This renders them an inevitable component +<br/>to realize the paradigm of intuitive human-machine interac- +<br/>tion. +<br/>Our demonstration shows model-based recognition of +<br/>facial expressions in real-time via the state-of-the-art +<br/>Candide-3 face model [1] as visible in Figure 1. This three- +<br/>dimensional and deformable model is highly appropriate +<br/>for real-world face interpretation applications. However, +<br/>its complexity challenges the task of model fitting and we +<br/>tackle this challenge with an algorithm that has been auto- +<br/>matically learned from a large set of images. This solution +<br/>provides both, high accuracy and runtime. Note, that our +<br/>system is not limited to facial expression estimation. Gaze +<br/>direction, gender and age are also estimated. +<br/>2. Face Model Fitting +<br/>Models reduce the large amount of image data to a +<br/>small number of model parameters to describe the im- +<br/>age content, which facilitates and accelerates the subse- +<br/>quent interpretation task. Cootes et al. [3] introduced mod- +<br/>elling shapes with Active Contours. Further enhancements +<br/>emerged the idea of expanding shape models with texture +<br/>information [2]. Recent research considers modelling faces +<br/>in 3D space [1, 10]. +<br/>Fitting the face model is the computational challenge of +<br/>finding the parameters that best describe the face within a +<br/>given image. This task is often addressed by minimizing +<br/>an objective function, such as the pixel error between the +<br/>model’s rendered surface and the underlying image content. +<br/>This section describes the four main components of model- +<br/>based techniques, see [9]. +<br/>The face model contains a parameter vector p that repre- +<br/>sents its configurations. We integrate the complex and de- +<br/>formable 3D wire frame Candide-3 face model [1]. The +<br/>model consists of 116 anatomical landmarks and its param- +<br/>eter vector p = (rx, ry, rz, s, tx, ty, σ, α)T describes the +<br/>affine transformation (rx, ry, rz, s, tx, ty) and the deforma- +<br/>tion (σ, α). The 79 deformation parameters indicate the +<br/>shape of facial components such as the mouth, the eyes, or +<br/>the eye brows, etc., see Figure 2. +<br/>The localization algorithm computes an initial estimate of +<br/>the model parameters that is further refined by the subse- +<br/>quent fitting algorithm. Our system integrates the approach +<br/>of [8], which detects the model’s affine transformation in +<br/>case the image shows a frontal view face. +<br/>The objective function yields a comparable value that +<br/>specifies how accurately a parameterized model matches an +<br/>image. Traditional approaches manually specify the objec- +<br/>tive function in a laborious and erroneous task. In contrast, +<br/>we automatically learn the objective function from a large +<br/>set of training data based on objective information theoretic +<br/>measures [9]. This approach does not require expert knowl- +<br/>edge and it is domain-independently applicable. As a re- +<br/>sult, this approach yields more robust and accurate objective +<br/>functions, which greatly facilitate the task of the associated +<br/>fitting algorithms. Accurately estimated model parameters +<br/>in turn are required to infer correct high-level information, +<br/>such as facial expression or gaze direction. +<br/>Figure 1. Interpreting expressions with the Candide-3 face model. +</td><td>('1685773', 'Christoph Mayer', 'christoph mayer')<br/>('32131501', 'Matthias Wimmer', 'matthias wimmer')<br/>('1704997', 'Freek Stulp', 'freek stulp')<br/>('1725709', 'Zahid Riaz', 'zahid riaz')<br/>('36401753', 'Anton Roth', 'anton roth')<br/>('34667371', 'Martin Eggers', 'martin eggers')<br/>('1699132', 'Bernd Radig', 'bernd radig')</td><td>{mayerc,wimmerm,stulp,riaz,roth,eggers,radig}@in.tum.de +</td></tr><tr><td>03b98b4a2c0b7cc7dae7724b5fe623a43eaf877b</td><td>Acume: A Novel Visualization Tool for Understanding Facial +<br/>Expression and Gesture Data +</td><td></td><td></td></tr><tr><td>03adcf58d947a412f3904a79f2ab51cfdf0e838a</td><td>World Journal of Science and Technology 2012, 2(4):136-139 +<br/>ISSN: 2231 – 2587 +<br/>Available Online: www.worldjournalofscience.com +<br/>_________________________________________________________________ +<br/> Proceedings of "Conference on Advances in Communication and Computing (NCACC'12)” +<br/><b>Held at R.C.Patel Institute of Technology, Shirpur, Dist. Dhule, Maharastra, India</b><br/> April 21, 2012 +<br/>Video-based face recognition: a survey +<br/><b>R.C.Patel Institute of Technology, Shirpur, Dist.Dhule.Maharashtra, India</b></td><td>('40628915', 'Shailaja A Patil', 'shailaja a patil')<br/>('30751046', 'Pramod J Deore', 'pramod j deore')</td><td></td></tr><tr><td>03104f9e0586e43611f648af1132064cadc5cc07</td><td></td><td></td><td></td></tr><tr><td>03f14159718cb495ca50786f278f8518c0d8c8c9</td><td>2015 IEEE International Conference on Control System, Computing and Engineering, Nov 27 – Nov 29, 2015 Penang, Malaysia +<br/>2015 IEEE International Conference on Control System, +<br/>Computing and Engineering (ICCSCE2015) +<br/>Technical Session 1A – DAY 1 – 27th Nov 2015 +<br/>Time: 3.00 pm – 4.30 pm +<br/>Venue: Jintan +<br/>Topic: Signal and Image Processing +<br/>3.00 pm – 3.15pm +<br/>3.15 pm – 3.30pm +<br/>3.30 pm – 3.45pm +<br/>3.45 pm – 4.00pm +<br/>4.00 pm – 4.15pm +<br/>4.15 pm – 4.30pm +<br/>4.30 pm – 4.45pm +<br/>1A 01 ID3 +<br/>Can Subspace Based Learning Approach Perform on Makeup Face +<br/>Recognition? +<br/>Khor Ean Yee, Pang Ying Han, Ooi Shih Yin and Wee Kuok Kwee +<br/>1A 02 ID35 +<br/>Performance Evaluation of HOG and Gabor Features for Vision-based +<br/>Vehicle Detection +<br/>1A 03 ID23 +<br/>Experimental Method to Pre-Process Fuzzy Bit Planes before Low-Level +<br/>Feature Extraction in Thermal Images +<br/>Chan Wai Ti and Sim Kok Swee +<br/>1A 04 ID84 +<br/>Fractal-based Texture and HSV Color Features for Fabric Image Retrieval +<br/>Nanik Suciati, Darlis Herumurti and Arya Yudhi Wijaya +<br/>1A 05 ID168 +<br/>Study of Automatic Melody Extraction Methods for Philippine Indigenous +<br/>Music +<br/>Jason Disuanco, Vanessa Tan, Franz de Leon +<br/>1A 06 ID211 +<br/>Acoustical Comparison between Voiced and Voiceless Arabic Phonemes of +<br/>Malay +<br/>Speakers +<br/>Ali Abd Almisreb, Ahmad Farid Abidin, Nooritawati Md Tahir +<br/>*shaded cell is the proposed session chair +<br/>viii +<br/>©Faculty of Electrical Engineering, Universiti Teknologi MARA +</td><td>('2715116', 'Soo Siang Teoh', 'soo siang teoh')</td><td>Tea Break @ Foyer +</td></tr><tr><td>0394040749195937e535af4dda134206aa830258</td><td>Geodesic Entropic Graphs for Dimension and +<br/>Entropy Estimation in Manifold Learning +<br/>December 16, 2003 +</td><td>('1759109', 'Jose A. Costa', 'jose a. costa')<br/>('1699402', 'Alfred O. Hero', 'alfred o. hero')</td><td></td></tr><tr><td>0334cc0374d9ead3dc69db4816d08c917316c6c4</td><td></td><td></td><td></td></tr><tr><td>03c48d8376990cff9f541d542ef834728a2fcda2</td><td>Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs +<br/><b>Columbia University</b><br/>New York, NY, USA +</td><td>('2195345', 'Zheng Shou', 'zheng shou')<br/>('2704179', 'Dongang Wang', 'dongang wang')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>{zs2262,dw2648,sc250}@columbia.edu +</td></tr><tr><td>0319332ded894bf1afe43f174f5aa405b49305f0</td><td>Shearlet Network-based Sparse Coding Augmented by +<br/>Facial Texture Features for Face Recognition +<br/>Ben Amar1 +<br/><b>Research Groups on Intelligent Machines, University of Sfax, Sfax 3038, Tunisia</b><br/><b>University of Houston, Houston, TX 77204, USA</b></td><td>('2791150', 'Mohamed Anouar Borgi', 'mohamed anouar borgi')<br/>('8847309', 'Demetrio Labate', 'demetrio labate')</td><td>{anoir.borgi@ieee.org ; dlabate@math.uh.edu ; +<br/>maher.elarbi@gmail.com; chokri.benamar@ieee.org} +</td></tr><tr><td>03ac1c694bc84a27621da6bfe73ea9f7210c6d45</td><td>Chapter 1 +<br/>Introduction to information security +<br/>foundations and applications +<br/>1.1 Background +<br/>Information security has extended to include several research directions like user +<br/>authentication and authorization, network security, hardware security, software secu- +<br/>rity, and data cryptography. Information security has become a crucial need for +<br/>protecting almost all information transaction applications. Security is considered as +<br/>an important science discipline whose many multifaceted complexities deserve the +<br/>synergy of the computer science and engineering communities. +<br/>Recently, due to the proliferation of Information and Communication Tech- +<br/>nologies, information security has started to cover emerging topics such as cloud +<br/>computing security, smart cities’ security and privacy, healthcare and telemedicine, +<br/>the Internet-of-Things (IoT) security [1], the Internet-of-Vehicles security, and sev- +<br/>eral types of wireless sensor networks security [2,3]. In addition, information security +<br/>has extended further to cover not only technical security problems but also social and +<br/>organizational security challenges [4,5]. +<br/>Traditional systems’ development approaches were focusing on the system’s +<br/>usability where security was left to the last stage with less priority. However, the +<br/>new design approaches consider security-in-design process where security is consid- +<br/>ered at the early phase of the design process. The new designed systems should be +<br/>well protected against the available security attacks. Having new systems such as IoT +<br/>or healthcare without enough security may lead to a leakage of sensitive data and, in +<br/>some cases, life threatening situations. +<br/>Taking the social aspect into account, security education is a vital need for both +<br/>practitioners and system users [6]. Users’ misbehaviour due to a lack of security +<br/>knowledge is the weakest point in the system security chain. The users’ misbehaviour +<br/>is considered as a security vulnerability that may be exploited for launching security +<br/>attacks. A successful security attack such as distributed denial-of-service attack will +<br/>impose incident recovery cost in addition to the downtime cost. +<br/><b>Electrical and Space Engineering, Lule University of Technology</b><br/>Sweden +<br/><b>Faculty of Engineering, Al Azhar University, Qena, Egypt</b></td><td>('4073409', 'Ali Ismail Awad', 'ali ismail awad')</td><td></td></tr><tr><td>03baf00a3d00887dd7c828c333d4a29f3aacd5f5</td><td>Entropy Based Feature Selection for 3D Facial +<br/>Expression Recognition +<br/>Submitted to the +<br/><b>Institute of Graduate Studies and Research</b><br/>in partial fulfillment of the requirements for the Degree of +<br/>Doctor of Philosophy +<br/>in +<br/> Electrical and Electronic Engineering +<br/><b>Eastern Mediterranean University</b><br/> September 2014 +<br/>Gazimağusa, North Cyprus +</td><td>('1974278', 'Kamil Yurtkan', 'kamil yurtkan')</td><td></td></tr><tr><td>0359f7357ea8191206b9da45298902de9f054c92</td><td>Going Deeper in Facial Expression Recognition using Deep Neural Networks +<br/>1 Department of Electrical and Computer Engineering +<br/>2 Department of Computer Science +<br/><b>University of Denver, Denver, CO</b></td><td>('2314025', 'Ali Mollahosseini', 'ali mollahosseini')<br/>('38461715', 'David Chan', 'david chan')<br/>('3093835', 'Mohammad H. Mahoor', 'mohammad h. mahoor')</td><td>ali.mollahosseini@du.edu, davidchan@cs.du.edu, and mmahoor@du.edu ∗ † +</td></tr><tr><td>0394e684bd0a94fc2ff09d2baef8059c2652ffb0</td><td>Median Robust Extended Local Binary Pattern +<br/>for Texture Classification +<br/>Index Terms— Texture descriptors, rotation invariance, local +<br/>binary pattern (LBP), feature extraction, texture analysis. +<br/>how the texture recognition process works in humans as +<br/>well as in the important role it plays in the wide variety of +<br/>applications of computer vision and image analysis [1], [2]. +<br/>The many applications of texture classification include medical +<br/>image analysis and understanding, object recognition, biomet- +<br/>rics, content-based image retrieval, remote sensing, industrial +<br/>inspection, and document classification. +<br/>As a classical pattern recognition problem, texture classifi- +<br/>cation primarily consists of two critical subproblems: feature +<br/>extraction and classifier designation [1], [2]. It is generally +<br/>agreed that the extraction of powerful texture features plays a +<br/>relatively more important role, since if poor features are used +<br/>even the best classifier will fail to achieve good recognition +<br/>results. Consequently, most research in texture classification +<br/>focuses on the feature extraction part and numerous texture +<br/>feature extraction methods have been developed, with excellent +<br/>surveys given in [1]–[5]. Most existing methods have not, +<br/>however, been capable of performing sufficiently well for +<br/>real-world applications, which have demanding requirements +<br/>including database size, nonideal environmental conditions, +<br/>and running in real-time. +</td><td>('39695518', 'Li Liu', 'li liu')<br/>('1716428', 'Songyang Lao', 'songyang lao')<br/>('1731709', 'Paul W. Fieguth', 'paul w. fieguth')<br/>('1714724', 'Matti Pietikäinen', 'matti pietikäinen')</td><td></td></tr><tr><td>03e88bf3c5ddd44ebf0e580d4bd63072566613ad</td><td></td><td></td><td></td></tr><tr><td>03f4c0fe190e5e451d51310bca61c704b39dcac8</td><td>J Ambient Intell Human Comput +<br/>DOI 10.1007/s12652-016-0406-z +<br/>O R I G I N A L R E S E A R C H +<br/>CHEAVD: a Chinese natural emotional audio–visual database +<br/>Received: 30 March 2016 / Accepted: 22 August 2016 +<br/>Ó Springer-Verlag Berlin Heidelberg 2016 +</td><td>('1704841', 'Ya Li', 'ya li')<br/>('37670752', 'Jianhua Tao', 'jianhua tao')<br/>('1850313', 'Linlin Chao', 'linlin chao')<br/>('1694779', 'Wei Bao', 'wei bao')<br/>('3095820', 'Yazhu Liu', 'yazhu liu')</td><td></td></tr><tr><td>03bd58a96f635059d4bf1a3c0755213a51478f12</td><td>Smoothed Low Rank and Sparse Matrix Recovery by +<br/>Iteratively Reweighted Least Squares Minimization +<br/>This work presents a general framework for solving the low +<br/>rank and/or sparse matrix minimization problems, which may +<br/>involve multiple non-smooth terms. The Iteratively Reweighted +<br/>Least Squares (IRLS) method is a fast solver, which smooths the +<br/>objective function and minimizes it by alternately updating the +<br/>variables and their weights. However, the traditional IRLS can +<br/>only solve a sparse only or low rank only minimization problem +<br/>with squared loss or an affine constraint. This work generalizes +<br/>IRLS to solve joint/mixed low rank and sparse minimization +<br/>problems, which are essential formulations for many tasks. As a +<br/>concrete example, we solve the Schatten-p norm and (cid:96)2,q-norm +<br/>regularized Low-Rank Representation (LRR) problem by IRLS, +<br/>and theoretically prove that the derived solution is a stationary +<br/>point (globally optimal if p, q ≥ 1). Our convergence proof of +<br/>IRLS is more general than previous one which depends on +<br/>the special properties of the Schatten-p norm and (cid:96)2,q-norm. +<br/>Extensive experiments on both synthetic and real data sets +<br/>demonstrate that our IRLS is much more efficient. +<br/>Index Terms—Low-rank and sparse minimization, Iteratively +<br/>Reweighted Least Squares. +<br/>I. INTRODUCTION +<br/>I N recent years, the low rank and sparse matrix learning +<br/>problems have been hot research topics and lead to broad +<br/>applications in computer vision and machine learning, such +<br/>as face recognition [1], collaborative filtering [2], background +<br/>modeling [3], and subspace segmentation [4], [5]. The (cid:96)1- +<br/>norm and nuclear norm are popular choices for sparse and +<br/>low rank matrix minimizations with theoretical guarantees +<br/>and competitive performance in practice. The models can be +<br/>formulated as a joint low rank and sparse matrix minimization +<br/>problem as follow: +<br/>T(cid:88) +<br/>nuclear norm ||M||∗ = (cid:80) +<br/>min +<br/>i=1 +<br/>where x and bi can be either vectors or matrices, Fi is a +<br/>convex function (the Frobenius norm ||M||2 +<br/>ij; +<br/>ij M 2 +<br/>i σi(M ), the sum of all singular +<br/>F = (cid:80) +<br/>Fi(Ai(x) + bi), +<br/>(1) +<br/>Copyright (c) 2014 IEEE. Personal use of this material +<br/>is permitted. +<br/>However, permission to use this material for any other purposes must be +<br/>This research is supported by the Singapore National Research Foundation +<br/>administered by the IDM Programme Office. Z. Lin is supported by NSF +<br/>China (grant nos. 61272341 and 61231002), 973 Program of China (grant no. +<br/>2015CB3525) and MSRA Collaborative Research Program. +<br/>C. Lu and S. Yan are with the Department of Electrical and Com- +<br/><b>puter Engineering, National University of Singapore, Singapore (e-mails</b><br/>Z. Lin is with the Key Laboratory of Machine Perception (MOE), School +<br/>values of a matrix; (cid:96)1-norm ||M||1 = (cid:80) +<br/>norm ||M||2,1 =(cid:80) +<br/>= (cid:80) +<br/>ij |Mij|; and (cid:96)2,1- +<br/>j ||Mj||2, the sum of the (cid:96)2-norm of each +<br/>column of a matrix) and Ai : Rd → Rm is a linear mapping. +<br/>In this work, we further consider the nonconvex Schatten-p +<br/>norm ||M||p +<br/>ij |Mij|p +<br/>and (cid:96)2,p-norm ||M||p +<br/>j ||Mj||p +<br/>2 with 0 < p < 1 for +<br/>pursuing lower rank or sparser solutions. +<br/>i σp(M ), (cid:96)p-norm ||M||p +<br/>2,p = (cid:80) +<br/>p = (cid:80) +<br/>Sp +<br/>Problem (1) is general which involves a wide range of +<br/>problems, such as Lasso [6], group Lasso [7], trace Lasso [4], +<br/>matrix completion [8], Robust Principle Component Analysis +<br/>(RPCA) [3] and Low-Rank Representation (LRR) [5]. In this +<br/>work, we aim to propose a general solver for (1). For the ease +<br/>of discussion, we focus on the following two representative +<br/>problems, +<br/>RPCA: +<br/>s.t. X = Z + E, +<br/>(2) +<br/>||Z||∗ + λ||E||1, +<br/>min +<br/>Z,E +<br/>||Z||∗ + λ||E||2,1, +<br/>min +<br/>Z,E +<br/>s.t. X = XZ + E, +<br/>LRR: +<br/>(3) +<br/>where X ∈ Rd×n is a given data matrix, Z and E are with +<br/>compatible dimensions and λ > 0 is the model parameter. No- +<br/>tice that these problems can be reformulated as unconstrained +<br/>problems (by representing E by Z) as that in problem (1). +<br/>A. Related Works +<br/>The sparse and low rank minimization problems can be +<br/>solved by various methods, such as Semi-Definite Program- +<br/>ming (SDP) [9], Accelerated Proximal Gradient (APG) [10], +<br/>and Alternating Direction Method (ADM) [11]. However, SDP +<br/>has a complexity of O(n6) for an n × n sized matrix, which +<br/>is unbearable for large scale applications. APG requires that +<br/>at +<br/>least one term of the objective function has Lipschitz +<br/>continuous gradient. Such an assumption is violated in many +<br/>problems, e.g., problem (2) and (3). Compared with SDP +<br/>and APG, ADM is the most widely used one. But it usually +<br/>requires introducing several auxiliary variables corresponding +<br/>to non-smooth terms. The auxiliary variables may slow down +<br/>the convergence, or even lead to divergence when there are +<br/>too many variables. Linearized ADM (LADM) [12] may +<br/>reduce the number of auxiliary variables, but suffer the same +<br/>convergence issue. The work [12] proposes an accelerated +<br/>LADM with Adaptive Penalty (LADMAP) with lower per- +<br/>iteration cost. However, the accelerating trick is special for the +<br/>LRR problem. And thus are not general for other problems. +<br/>Another drawback for many low rank minimization solvers is +<br/>that they have to perform the soft singular value thresholding: +<br/>λ||Z||∗ + +<br/>||Z − Y ||2 +<br/>F , +<br/>min +<br/>(4) +</td><td>('33224509', 'Canyi Lu', 'canyi lu')<br/>('33383055', 'Zhouchen Lin', 'zhouchen lin')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>obtained from the IEEE by sending a request to pubs-permissions@ieee.org. +<br/>under its International Research Centre @Singapore Funding Initiative and +<br/>canyilu@gmail.com; eleyans@nus.edu.sg). +<br/>of EECS, Peking University, China (e-mail: zlin@pku.edu.cn). +</td></tr><tr><td>031055c241b92d66b6984643eb9e05fd605f24e2</td><td>Multi-fold MIL Training for Weakly Supervised Object Localization +<br/>Inria∗ +</td><td>('1939006', 'Ramazan Gokberk Cinbis', 'ramazan gokberk cinbis')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>0332ae32aeaf8fdd8cae59a608dc8ea14c6e3136</td><td>Int J Comput Vis +<br/>DOI 10.1007/s11263-017-1009-7 +<br/>Large Scale 3D Morphable Models +<br/>Received: 15 March 2016 / Accepted: 24 March 2017 +<br/>© The Author(s) 2017. This article is an open access publication +</td><td>('1848903', 'James Booth', 'james booth')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('2931390', 'Anastasios Roussos', 'anastasios roussos')<br/>('5137183', 'Allan Ponniah', 'allan ponniah')</td><td></td></tr><tr><td>034addac4637121e953511301ef3a3226a9e75fd</td><td>Implied Feedback: Learning Nuances of User Behavior in Image Search +<br/>Virginia Tech +</td><td>('1713589', 'Devi Parikh', 'devi parikh')</td><td>parikh@vt.edu +</td></tr><tr><td>03701e66eda54d5ab1dc36a3a6d165389be0ce79</td><td>179 +<br/>Improved Principal Component Regression for Face +<br/>Recognition Under Illumination Variations +</td><td>('1776127', 'Shih-Ming Huang', 'shih-ming huang')<br/>('1749263', 'Jar-Ferr Yang', 'jar-ferr yang')</td><td></td></tr><tr><td>03fe3d031afdcddf38e5cc0d908b734884542eeb</td><td>DOI: http://dx.doi.org/10.14236/ewic/EVA2017.60 +<br/>Engagement with Artificial Intelligence +<br/>through Natural Interaction Models +<br/>Sara (Salevati) Feldman +<br/><b>Simon Fraser University</b><br/>Vancouver, Canada +<br/><b>Simon Fraser University</b><br/>Vancouver, Canada +<br/><b>Simon Fraser University</b><br/>Vancouver, Canada +<br/>As Artificial Intelligence (AI) systems become more ubiquitous, what user experience design +<br/>paradigms will be used by humans to impart their needs and intents to an AI system, in order to +<br/>engage in a more social interaction? In our work, we look mainly at expression and creativity +<br/>based systems, where the AI both attempts to model or understand/assist in processes of human +<br/>expression and creativity. We therefore have designed and implemented a prototype system with +<br/>more natural interaction modes for engagement with AI as well as other human computer +<br/>interaction (HCI) where a more open natural communication stream is beneficial. Our proposed +<br/>conversational agent system makes use of the affective signals from the gestural behaviour of the +<br/>user and the semantic information from the speech input in order to generate a personalised, +<br/>human-like conversation that is expressed in the visual and conversational output of the 3D virtual +<br/>avatar system. We describe our system and two application spaces we are using it in – a care +<br/>advisor / assistant for the elderly and an interactive creative assistant for uses to produce art +<br/>forms. +<br/>Artificial Intelligence. Natural user interfaces. Voice systems. Expression systems. ChatBots. +<br/>1. INTRODUCTION +<br/>is +<br/>for +<br/>way +<br/>there +<br/>sensor +<br/>natural +<br/>devices, +<br/>understand +<br/>requirement +<br/>to +<br/>the human +<br/>Due to the increase of natural user interfaces and +<br/>untethered +<br/>a +<br/>corresponding +<br/>for computational +<br/>models that can utilise interactive and affective +<br/>user data in order to understand and emulate a +<br/>more +<br/>conversational +<br/>communication. From an emulation standpoint, it is +<br/>the mechanisms +<br/>important +<br/>underlying +<br/>to human multilayered +<br/>semantic communication to achieve a more natural +<br/>user experience. Humans tend to make use of +<br/>gestures and expressions +<br/>in a conversational +<br/>setting in addition to the linguistic components that +<br/>allow them to express more than the semantics of +<br/>is usually +<br/>the utterances. This phenomenon +<br/>automated +<br/>current +<br/>disregarded +<br/>to +<br/>conversational +<br/>due +<br/>being +<br/>computationally demanding and +<br/>requiring a +<br/>cognitive component to be able to model the +<br/>complexity of the additional signals. With the +<br/>advances in the current technology we are now +<br/>closer to achieve more natural-like conversational +<br/>systems. Gesture capture and recognition systems +<br/>for video and sound input can be combined with +<br/>output systems such as Artificial Intelligence (AI) +<br/>based conversational +<br/>tools and 3D modelling +<br/>systems +<br/>the +<br/>in +<br/>© Feldman et al. Published by +<br/>BCS Learning and Development Ltd. +<br/>Proceedings of Proceedings of EVA London 2017, UK +<br/>296 +<br/>to +<br/>include +<br/>in order +<br/>systems +<br/>to achieve human-level +<br/>meaningful communication. This may allow the +<br/>interaction to be more intuitive, open and fluent that +<br/>can be more helpful in certain situations. In this +<br/>work, we attempt +<br/>the affective +<br/>components from these input signals in order to +<br/>generate a compatible and personalised character +<br/>that can reflect some human-like qualities. +<br/>Given +<br/>these goals, we overview our 3D +<br/>conversational avatar system and describe its use +<br/>in our two application spaces, stressing its use +<br/>where AI systems are involved. Our first application +<br/>space is CareAdvisor, for maintaining active and +<br/>healthy aging in older adults through a multi- +<br/>modular Personalised Virtual Coaching system. +<br/>Here the natural communication system is better +<br/>suited for the elderly, who are technologically less +<br/>experienced, +<br/>non- +<br/>confrontationally and as an assistant conduit to +<br/>health data from other less conversational devices. +<br/>Our second application space is in the interactive +<br/>art exhibition area, where our avatar system is able +<br/>to converse with users in a more open way, +<br/>compared to say forms and input systems, on +<br/>issues of art and creativity. This allows for more +<br/>open, +<br/>to an +<br/>intuitive conversation +<br/>especially when +<br/>leading +<br/>used +</td><td>('22588208', 'Ozge Nilay Yalcin', 'ozge nilay yalcin')<br/>('1700040', 'Steve DiPaola', 'steve dipaola')</td><td>sara_salevati@sfu.ca +<br/>oyalcin@sfu.ca +<br/>sdipaola@sfu.ca +</td></tr><tr><td>9b318098f3660b453fbdb7a579778ab5e9118c4c</td><td>3931 +<br/>Joint Patch and Multi-label Learning for Facial +<br/>Action Unit and Holistic Expression Recognition +<br/>classifiers without +</td><td>('2393320', 'Kaili Zhao', 'kaili zhao')<br/>('39336289', 'Wen-Sheng Chu', 'wen-sheng chu')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')<br/>('1720776', 'Honggang Zhang', 'honggang zhang')</td><td></td></tr><tr><td>9be94fa0330dd493f127d51e4ef7f9fd64613cfc</td><td>Research Article +<br/>Effects of pose and image resolution on +<br/>automatic face recognition +<br/>ISSN 2047-4938 +<br/>Received on 5th February 2015 +<br/>Revised on 16th May 2015 +<br/>Accepted on 14th September 2015 +<br/>doi: 10.1049/iet-bmt.2015.0008 +<br/>www.ietdl.org +<br/><b>North Dakota State University, Fargo, ND 58108-6050, USA</b><br/><b>Faculty of Computer Science, Mathematics, and Engineering, University of Twente, Enschede, Netherlands</b></td><td>('3001880', 'Zahid Mahmood', 'zahid mahmood')<br/>('1798087', 'Tauseef Ali', 'tauseef ali')</td><td>✉ E-mail: zahid.mahmood@ndsu.edu +</td></tr><tr><td>9bd35145c48ce172b80da80130ba310811a44051</td><td>Face Detection with End-to-End Integration of a +<br/>ConvNet and a 3D Model +<br/>1Nat’l Engineering Laboratory for Video Technology, +<br/>Key Laboratory of Machine Perception (MoE), +<br/>Cooperative Medianet Innovation Center, Shanghai +<br/><b>Sch l of EECS, Peking University, Beijing, 100871, China</b><br/>2Department of ECE and the Visual Narrative Cluster, +<br/><b>North Carolina State University, Raleigh, USA</b></td><td>('3422021', 'Yunzhu Li', 'yunzhu li')<br/>('3423002', 'Benyuan Sun', 'benyuan sun')<br/>('47353858', 'Tianfu Wu', 'tianfu wu')<br/>('1717863', 'Yizhou Wang', 'yizhou wang')</td><td>{leo.liyunzhu, sunbenyuan, Yizhou.Wang}@pku.edu.cn, tianfu wu@ncsu.edu +</td></tr><tr><td>9b000ccc04a2605f6aab867097ebf7001a52b459</td><td></td><td></td><td></td></tr><tr><td>9b0489f2d5739213ef8c3e2e18739c4353c3a3b7</td><td>Visual Data Augmentation through Learning +<br/><b>Imperial College London, UK</b><br/><b>Middlesex University London, UK</b></td><td>('34586458', 'Grigorios G. Chrysos', 'grigorios g. chrysos')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>{g.chrysos, i.panagakis, s.zafeiriou}@imperial.ac.uk +</td></tr><tr><td>9b474d6e81e3b94e0c7881210e249689139b3e04</td><td>VG-RAM Weightless Neural Networks for +<br/>Face Recognition +<br/>Departamento de Inform´atica +<br/>Universidade Federal do Esp´ırito Santo +<br/>Av. Fernando Ferrari, 514, 29075-910 - Vit´oria-ES +<br/>Brazil +<br/>1. Introduction +<br/>Computerized human face recognition has many practical applications, such as access control, +<br/>security monitoring, and surveillance systems, and has been one of the most challenging and +<br/>active research areas in computer vision for many decades (Zhao et al.; 2003). Even though +<br/>current machine recognition systems have reached a certain level of maturity, the recognition +<br/>of faces with different facial expressions, occlusions, and changes in illumination and/or pose +<br/>is still a hard problem. +<br/>A general statement of the problem of machine recognition of faces can be formulated as fol- +<br/>lows: given an image of a scene, (i) identify or (ii) verify one or more persons in the scene +<br/>using a database of faces. In identification problems, given a face as input, the system reports +<br/>back the identity of an individual based on a database of known individuals; whereas in veri- +<br/>fication problems, the system confirms or rejects the claimed identity of the input face. In both +<br/>cases, the solution typically involves segmentation of faces from scenes (face detection), fea- +<br/>ture extraction from the face regions, recognition, or verification. In this chapter, we examine +<br/>the recognition of frontal face images required in the context of identification problems. +<br/>Many approaches have been proposed to tackle the problem of face recognition. One can +<br/>roughly divide these into (i) holistic approaches, (ii) feature-based approaches, and (iii) hybrid +<br/>approaches (Zhao et al.; 2003). Holistic approaches use the whole face region as the raw input +<br/>to a recognition system (a classifier). In feature-based approaches, local features, such as the +<br/>eyes, nose, and mouth, are first extracted and their locations and local statistics (geometric +<br/>and/or appearance based) are fed into a classifier. Hybrid approaches use both local features +<br/>and the whole face region to recognize a face. +<br/>Among +<br/>fisher- +<br/>faces (Belhumeur et al.; 1997; Etemad and Chellappa; 1997) have proved to be effective +<br/>(Turk and Pentland; +<br/>eigenfaces +<br/>holistic +<br/>approaches, +<br/>1991) +<br/>and +</td><td>('1699216', 'Alberto F. De Souza', 'alberto f. de souza')<br/>('3015563', 'Claudine Badue', 'claudine badue')<br/>('3158075', 'Felipe Pedroni', 'felipe pedroni')<br/>('3169286', 'Hallysson Oliveira', 'hallysson oliveira')</td><td></td></tr><tr><td>9b928c0c7f5e47b4480cb9bfdf3d5b7a29dfd493</td><td>Close the Loop: Joint Blind Image Restoration and Recognition +<br/>with Sparse Representation Prior +<br/><b>School of Computer Science, Northwestern Polytechnical University, Xi an China</b><br/><b>Beckman Institute, University of Illinois at Urbana-Champaign, IL USA</b><br/><b>U.S. Army Research Laboratory, 2800 Powder Mill Road, Adelphi, MD USA</b></td><td>('40479011', 'Haichao Zhang', 'haichao zhang')<br/>('1706007', 'Jianchao Yang', 'jianchao yang')<br/>('1801395', 'Yanning Zhang', 'yanning zhang')<br/>('8147588', 'Nasser M. Nasrabadi', 'nasser m. nasrabadi')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>‡{hczhang,jyang29,huang}@ifp.uiuc.edu †ynzhang@nwpu.edu.cn §nasser.m.nasrabadi.civ@mail.mil +</td></tr><tr><td>9bc01fa9400c231e41e6a72ec509d76ca797207c</td><td></td><td></td><td></td></tr><tr><td>9b2c359c36c38c289c5bacaeb5b1dd06b464f301</td><td>Dense Face Alignment +<br/><b>Michigan State University, MI</b><br/>2Monta Vista High School, Cupertino, CA +</td><td>('6797891', 'Yaojie Liu', 'yaojie liu')<br/>('2357264', 'Amin Jourabloo', 'amin jourabloo')<br/>('26365310', 'William Ren', 'william ren')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td>1{liuyaoj1,jourablo,liuxm}@msu.edu, 2williamyren@gmail.com +</td></tr><tr><td>9bcfadd22b2c84a717c56a2725971b6d49d3a804</td><td>How to Detect a Loss of Attention in a Tutoring System +<br/>using Facial Expressions and Gaze Direction +</td><td>('2975858', 'Mark ter Maat', 'mark ter maat')</td><td></td></tr><tr><td>9b1bcef8bfef0fb5eb5ea9af0b699aa0534fceca</td><td>Position-Squeeze and Excitation Module +<br/>for Facial Attribute Analysis +<br/>Shanghai Key Laboratory of +<br/>Multidimensional Information +<br/>Processing, +<br/><b>East China Normal University</b><br/>200241 Shanghai, China +</td><td>('36124320', 'Yan Zhang', 'yan zhang')<br/>('7962836', 'Wanxia Shen', 'wanxia shen')<br/>('49755228', 'Li Sun', 'li sun')<br/>('12493943', 'Qingli Li', 'qingli li')<br/>('36124320', 'Yan Zhang', 'yan zhang')<br/>('7962836', 'Wanxia Shen', 'wanxia shen')<br/>('49755228', 'Li Sun', 'li sun')<br/>('12493943', 'Qingli Li', 'qingli li')</td><td>452642781@qq.com +<br/>51151214005@ecnu.cn +<br/>sunli@ee.ecnu.edu.cn +<br/>qlli@cs.ecnu.edu.cn +</td></tr><tr><td>9b07084c074ba3710fee59ed749c001ae70aa408</td><td>698535 CDPXXX10.1177/0963721417698535MartinezComputational Models of Face Perception +<br/>research-article2017 +<br/>Computational Models of Face Perception +<br/>Aleix M. Martinez +<br/>Department of Electrical and Computer Engineering, Center for Cognitive and Brain Sciences, +<br/><b>and Mathematical Biosciences Institute, The Ohio State University</b><br/>Current Directions in Psychological +<br/>Science +<br/> 1 –7 +<br/>© The Author(s) 2017 +<br/>Reprints and permissions: +<br/>sagepub.com/journalsPermissions.nav +<br/>DOI: 10.1177/0963721417698535 +<br/>https://doi.org/10.1177/0963721417698535 +<br/>www.psychologicalscience.org/CDPS +</td><td></td><td></td></tr><tr><td>9be653e1bc15ef487d7f93aad02f3c9552f3ee4a</td><td>Computer Vision for Head Pose Estimation: +<br/>Review of a Competition +<br/><b>Tampere University of Technology, Finland</b><br/><b>University of Paderborn, Germany</b><br/>3 Zorgon, The Netherlands +</td><td>('1847889', 'Heikki Huttunen', 'heikki huttunen')<br/>('40394658', 'Ke Chen', 'ke chen')<br/>('2364638', 'Abhishek Thakur', 'abhishek thakur')<br/>('2558923', 'Artus Krohn-Grimberghe', 'artus krohn-grimberghe')<br/>('2300445', 'Oguzhan Gencoglu', 'oguzhan gencoglu')<br/>('3328835', 'Xingyang Ni', 'xingyang ni')<br/>('2067035', 'Mohammed Al-Musawi', 'mohammed al-musawi')<br/>('40448210', 'Lei Xu', 'lei xu')<br/>('3152947', 'Hendrik Jacob van Veen', 'hendrik jacob van veen')</td><td></td></tr><tr><td>9b246c88a0435fd9f6d10dc88f47a1944dd8f89e</td><td>PICODES: Learning a Compact Code for +<br/>Novel-Category Recognition +<br/><b>Dartmouth College</b><br/>Hanover, NH, U.S.A. +<br/>Andrew Fitzgibbon +<br/>Microsoft Research +<br/>Cambridge, United Kingdom +</td><td>('34338883', 'Alessandro Bergamo', 'alessandro bergamo')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')</td><td>{aleb, lorenzo}@cs.dartmouth.edu +<br/>awf@microsoft.com +</td></tr><tr><td>9b164cef4b4ad93e89f7c1aada81ae7af802f3a4</td><td> Research Journal of Recent Sciences _________________________________________________ ISSN 2277-2502 +<br/> Vol. 2(1), 17-20, January (2013) +<br/>Res.J.Recent Sci. +<br/>A Fully Automatic and Haar like Feature Extraction-Based Method for Lip +<br/>Contour Detection +<br/><b>School of Computer Engineering, Shahrood University of Technology, Shahrood, IRAN</b><br/>Received 26th September 2012, revised 27th October 2012, accepted 6th November 2012 +<br/>Available online at: www.isca.in +</td><td></td><td></td></tr><tr><td>9bac481dc4171aa2d847feac546c9f7299cc5aa0</td><td>Matrix Product State for Higher-Order Tensor +<br/>Compression and Classification +</td><td>('2852180', 'Johann A. Bengua', 'johann a. bengua')<br/>('2839912', 'Ho N. Phien', 'ho n. phien')<br/>('1834451', 'Minh N. Do', 'minh n. do')</td><td></td></tr><tr><td>9b93406f3678cf0f16451140ea18be04784faeee</td><td>A Bayesian Approach to Alignment-Based +<br/>Image Hallucination +<br/><b>University of Central Florida</b><br/>2 Microsoft Research New England +</td><td>('1802944', 'Marshall F. Tappen', 'marshall f. tappen')<br/>('1681442', 'Ce Liu', 'ce liu')</td><td>mtappen@eecs.ucf.edu +<br/>celiu@microsoft.com +</td></tr><tr><td>9b7974d9ad19bb4ba1ea147c55e629ad7927c5d7</td><td>Faical Expression Recognition by Combining +<br/>Texture and Geometrical Features +</td><td>('3057167', 'Renjie Liu', 'renjie liu')<br/>('36485086', 'Ruofei Du', 'ruofei du')<br/>('40371477', 'Bao-Liang Lu', 'bao-liang lu')</td><td></td></tr><tr><td>9b6d0b3fbf7d07a7bb0d86290f97058aa6153179</td><td>NII, Japan at the first THUMOS Workshop 2013 +<br/><b>National Institute of Informatics</b><br/>2-1-2 Hitotsubashi, Chiyoda-ku, Tokyo, Japan 101-8430 +</td><td>('39814149', 'Sang Phan', 'sang phan')<br/>('1802416', 'Duy-Dinh Le', 'duy-dinh le')<br/>('40693818', 'Shin’ichi Satoh', 'shin’ichi satoh')</td><td>{plsang,ledduy,satoh}@nii.ac.jp +</td></tr><tr><td>9b684e2e2bb43862f69b12c6be94db0e7a756187</td><td>Differentiating Objects by Motion: +<br/>Joint Detection and Tracking of Small Flying Objects +<br/><b>The University of Tokyo</b><br/>CSIRO-Data61 +<br/><b>Australian National University</b><br/><b>The University of Tokyo</b><br/>Figure 1: Importance of multi-frame information for recognizing apparently small flying objects (birds in these examples). +<br/><b>While visual features in single frames are vague and limited, multi-frame information, including deformation and pose</b><br/>changes, provides better clues with which to recognize birds. To extract such useful motion patterns, tracking is necessary for +<br/>compensating translation of objects, but the tracking itself is a challenge due to the limited visual information. The blue boxes +<br/>are birds tracked by our method that utilizes multi-frame representation for detection, while the red boxes are the results of a +<br/>single-frame handcrafted-feature-based tracker [11] , which tends to fail when tracking small objects. +</td><td>('1890560', 'Ryota Yoshihashi', 'ryota yoshihashi')<br/>('38621343', 'Tu Tuan Trinh', 'tu tuan trinh')<br/>('48727803', 'Rei Kawakami', 'rei kawakami')<br/>('2941564', 'Shaodi You', 'shaodi you')<br/>('33313329', 'Makoto Iida', 'makoto iida')<br/>('48795689', 'Takeshi Naemura', 'takeshi naemura')</td><td>{yoshi, tu, rei, naemura}@hc.ic.i.u-tokyo.ac.jp +<br/>iida@ilab.eco.rcast.u-tokyo.ac.jp +<br/>shaodi.you@data61.csiro.au +</td></tr><tr><td>9e8637a5419fec97f162153569ec4fc53579c21e</td><td>Segmentation and Normalization of Human Ears +<br/>using Cascaded Pose Regression +<br/><b>University of Applied Sciences Darmstadt - CASED</b><br/>Haardtring 100, +<br/>64295 Darmstadt, Germany +<br/>http://www.h-da.de +</td><td>('1742085', 'Christoph Busch', 'christoph busch')</td><td>anika.pflug@cased.de +<br/>christoph.busch@hig.no +</td></tr><tr><td>9ea223c070ec9a00f4cb5ca0de35d098eb9a8e32</td><td>Exploring Temporal Preservation Networks for Precise Temporal Action +<br/>Localization +<br/>National Laboratory for Parallel and Distributed Processing, +<br/><b>National University of Defense Technology</b><br/>Changsha, China +</td><td>('2352864', 'Ke Yang', 'ke yang')<br/>('2292038', 'Peng Qiao', 'peng qiao')<br/>('1718853', 'Dongsheng Li', 'dongsheng li')<br/>('1893776', 'Shaohe Lv', 'shaohe lv')<br/>('1791001', 'Yong Dou', 'yong dou')</td><td>{yangke13,pengqiao,dongshengli,yongdou,shaohelv}@nudt.edu.cn +</td></tr><tr><td>9e4b052844d154c3431120ec27e78813b637b4fc</td><td>Journal of AI and Data Mining +<br/>Vol. 2, No .1, 2014, 33-38. +<br/>Local gradient pattern - A novel feature representation for facial +<br/>expression recognition +<br/><b>School of Applied Statistics, National Institute of Development Administration, Bangkok, Thailand</b><br/>Received 23 April 2013; accepted 16 June 2013 +</td><td>('31914125', 'M. Shahidul Islam', 'm. shahidul islam')</td><td>*Corresponding author: suva.93@grads.nida.ac.th (M.Shahidul Islam) +</td></tr><tr><td>9e42d44c07fbd800f830b4e83d81bdb9d106ed6b</td><td>Learning Discriminative Aggregation Network for Video-based Face Recognition +<br/><b>Tsinghua University, Beijing, China</b><br/>2State Key Lab of Intelligent Technologies and Systems, Beijing, China +<br/>3Tsinghua National Laboratory for Information Science and Technology (TNList), Beijing, China +</td><td>('39358728', 'Yongming Rao', 'yongming rao')<br/>('2772283', 'Ji Lin', 'ji lin')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('39491387', 'Jie Zhou', 'jie zhou')</td><td>raoyongming95@gmail.com; lin-j14@mails.tsinghua.edu.cn; {lujiwen,jzhou}@tsinghua.edu.cn +</td></tr><tr><td>9eb86327c82b76d77fee3fd72e2d9eff03bbe5e0</td><td>Max-Margin Invariant Features from Transformed +<br/>Unlabeled Data +<br/>Department of Electrical and Computer Engineering +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +</td><td>('2628116', 'Dipan K. Pal', 'dipan k. pal')<br/>('27756148', 'Ashwin A. Kannan', 'ashwin a. kannan')<br/>('27693929', 'Gautam Arakalgud', 'gautam arakalgud')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td>{dipanp,aalapakk,garakalgud,marioss}@cmu.edu +</td></tr><tr><td>9ea73660fccc4da51c7bc6eb6eedabcce7b5cead</td><td>Talking Head Detection by Likelihood-Ratio Test† +<br/>MIT Lincoln Laboratory, +<br/>Lexington MA 02420, USA +</td><td>('2877010', 'Carl Quillen', 'carl quillen')</td><td>wcampbell@ll.mit.edu +</td></tr><tr><td>9e9052256442f4e254663ea55c87303c85310df9</td><td>International Journal of Advanced Research in Computer Engineering & Technology (IJARCET) +<br/>Volume 4 Issue 10, October 2015 +<br/>Review On Attribute-assisted Reranking for +<br/>Image Search +<br/> +</td><td></td><td></td></tr><tr><td>9eeada49fc2cba846b4dad1012ba8a7ee78a8bb7</td><td>A New Facial Expression Recognition Method Based on Local Gabor Filter Bank and PCA plus LDA +<br/>A New Facial Expression Recognition Method Based on +<br/>Local Gabor Filter Bank and PCA plus LDA +<br/>1 School of Electronic and Information Engineering, South China +<br/><b>University of Technology, Guangzhou, 510640, P.R.China</b><br/><b>Motorola China Research Center, Shanghai, 210000, P.R.China</b></td><td>('15414934', 'Hong-Bo Deng', 'hong-bo deng')<br/>('2949795', 'Lian-Wen Jin', 'lian-wen jin')<br/>('1751744', 'Li-Xin Zhen', 'li-xin zhen')<br/>('34824270', 'Jian-Cheng Huang', 'jian-cheng huang')<br/>('15414934', 'Hong-Bo Deng', 'hong-bo deng')<br/>('2949795', 'Lian-Wen Jin', 'lian-wen jin')<br/>('1751744', 'Li-Xin Zhen', 'li-xin zhen')<br/>('34824270', 'Jian-Cheng Huang', 'jian-cheng huang')</td><td>{hbdeng, eelwjin}@scut.edu.cn +<br/>{Li-Xin.Zhen, Jian-Cheng.Huang}@motorola.com +</td></tr><tr><td>9ef2b2db11ed117521424c275c3ce1b5c696b9b3</td><td>Robust Face Alignment Using a Mixture of Invariant Experts +<br/>‡Intel Corporation +<br/><b>Mitsubishi Electric Research Labs (MERL</b></td><td>('2577513', 'Oncel Tuzel', 'oncel tuzel')<br/>('14939251', 'Salil Tambe', 'salil tambe')<br/>('34749896', 'Tim K. Marks', 'tim k. marks')</td><td>{oncel, tmarks}@merl.com, +<br/>salil.tambe@intel.com +</td></tr><tr><td>9e5acdda54481104aaf19974dca6382ed5ff21ed</td><td>Yulia Gizatdinova and Veikko Surakka +<br/>Automatic localization of facial +<br/>landmarks from expressive images +<br/>of high complexity +<br/>DEPARTMENT OF COMPUTER SCIENCES +<br/><b>UNIVERSITY OF TAMPERE</b><br/>D‐2008‐9 +<br/>TAMPERE 2008 +</td><td></td><td></td></tr><tr><td>9ed943f143d2deaac2efc9cf414b3092ed482610</td><td>Independent subspace of dynamic Gabor features for facial expression classification +<br/>School of Information Science +<br/><b>Japan Advanced Institute of Science and Technology</b><br/>Asahidai 1-1, Nomi-city, Ishikawa, Japan +</td><td>('2847306', 'Prarinya Siritanawan', 'prarinya siritanawan')<br/>('1791753', 'Kazunori Kotani', 'kazunori kotani')<br/>('1753878', 'Fan Chen', 'fan chen')</td><td>Email: {p.siritanawan, ikko, chen-fan}@jaist.ac.jp +</td></tr><tr><td>9e1c3b8b1653337094c1b9dba389e8533bc885b0</td><td>Demographic Classification with Local Binary +<br/>Patterns +<br/>Department of Computer Science and Technology, +<br/><b>Tsinghua University, Beijing 100084, China</b></td><td>('4381671', 'Zhiguang Yang', 'zhiguang yang')<br/>('1679380', 'Haizhou Ai', 'haizhou ai')</td><td>ahz@mail.tsinghua.edu.cn +</td></tr><tr><td>9e0285debd4b0ba7769b389181bd3e0fd7a02af6</td><td>From face images and attributes to attributes +<br/>Computer Vision Laboratory, ETH Zurich, Switzerland +</td><td>('9664434', 'Robert Torfason', 'robert torfason')<br/>('2794259', 'Eirikur Agustsson', 'eirikur agustsson')<br/>('2173683', 'Rasmus Rothe', 'rasmus rothe')<br/>('1732855', 'Radu Timofte', 'radu timofte')</td><td></td></tr><tr><td>9ed4ad41cbad645e7109e146ef6df73f774cd75d</td><td>SARFRAZ, SIDDIQUE, STIEFELHAGEN: RPM FOR PAIR-WISE FACE-SIMILARITY +<br/>RPM: Random Points Matching for Pair-wise +<br/>Face-Similarity +<br/><b>Institute for Anthropomatics</b><br/><b>Karlsruhe Institute of Technology</b><br/>Karlsruhe, Germany +<br/><b>Swiss Federal Institute of Technology</b><br/>(ETH) Zurich +<br/>Zurich, Switzerland +</td><td>('4241648', 'M. Saquib Sarfraz', 'm. saquib sarfraz')<br/>('6262445', 'Muhammad Adnan Siddique', 'muhammad adnan siddique')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>saquib.sarfraz@kit.edu +<br/>siddique@ifu.baug.ethz.ch +<br/>rainer.stiefelhagen@kit.edu +</td></tr><tr><td>9e182e0cd9d70f876f1be7652c69373bcdf37fb4</td><td>Talking Face Generation by Adversarially +<br/>Disentangled Audio-Visual Representation +<br/><b>The Chinese University of Hong Kong</b></td><td>('40576774', 'Hang Zhou', 'hang zhou')<br/>('1715752', 'Yu Liu', 'yu liu')<br/>('3243969', 'Ziwei Liu', 'ziwei liu')<br/>('47571885', 'Ping Luo', 'ping luo')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')</td><td></td></tr><tr><td>9e8d87dc5d8a6dd832716a3f358c1cdbfa97074c</td><td>What Makes an Image Popular? +<br/><b>Massachusetts Institute</b><br/>of Technology +<br/><b>eBay Research Labs</b><br/>DigitalGlobe +</td><td>('2556428', 'Aditya Khosla', 'aditya khosla')<br/>('2541992', 'Atish Das Sarma', 'atish das sarma')<br/>('37164887', 'Raffay Hamid', 'raffay hamid')</td><td>khosla@csail.mit.edu +<br/>atish.dassarma@gmail.com +<br/>raffay@gmail.com +</td></tr><tr><td>9e5c2d85a1caed701b68ddf6f239f3ff941bb707</td><td></td><td></td><td></td></tr><tr><td>044d9a8c61383312cdafbcc44b9d00d650b21c70</td><td>300 Faces in-the-Wild Challenge: The first facial landmark localization +<br/>Challenge +<br/><b>Imperial College London, UK</b><br/><b>School of Computer Science, University of Lincoln, U.K</b><br/><b>EEMCS, University of Twente, The Netherlands</b></td><td>('3320415', 'Christos Sagonas', 'christos sagonas')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{c.sagonas, gt204, s.zafeiriou, m.pantic}@imperial.ac.uk +</td></tr><tr><td>04bb3fa0824d255b01e9db4946ead9f856cc0b59</td><td></td><td></td><td></td></tr><tr><td>040dc119d5ca9ea3d5fc39953a91ec507ed8cc5d</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Large-scale Bisample Learning on ID vs. Spot Face Recognition +<br/>Received: date / Accepted: date +</td><td>('8362374', 'Xiangyu Zhu', 'xiangyu zhu')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td></td></tr><tr><td>04f0292d9a062634623516edd01d92595f03bd3f</td><td>Distribution-based Iterative Pairwise Classification of +<br/>Emotions in the Wild Using LGBP-TOP +<br/><b>The University of Nottingham</b><br/>Mised Reality Lab +<br/>Anıl Yüce +<br/>Signal Processing +<br/>Laboratory(LTS5) +<br/>École Polytechnique Fédérale +<br/>de Lausanne, Switzerland +<br/><b>The University of Nottingham</b><br/>Mixed Reality Lab +<br/><b>The University of Nottingham</b><br/>Mixed Reality Lab +</td><td>('2449665', 'Timur R. Almaev', 'timur r. almaev')<br/>('1795528', 'Michel F. Valstar', 'michel f. valstar')<br/>('2321668', 'Alexandru Ghitulescu', 'alexandru ghitulescu')</td><td>psxta4@nottingham.ac.uk +<br/>anil.yuce@epfl.ch +<br/>psyadg@nottingham.ac.uk +<br/>michel.valstar@nottingham.ac.uk +</td></tr><tr><td>047f6afa87f48de7e32e14229844d1587185ce45</td><td>An Improvement of Energy-Transfer Features +<br/>Using DCT for Face Detection +<br/><b>Technical University of Ostrava, FEECS</b><br/>17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic +</td><td>('2467747', 'Radovan Fusek', 'radovan fusek')<br/>('2557877', 'Eduard Sojka', 'eduard sojka')</td><td>{radovan.fusek,eduard.sojka,karel.mozdren,milan.surkala}@vsb.cz +</td></tr><tr><td>04b851f25d6d49e61a528606953e11cfac7df2b2</td><td>Optical Flow Guided Feature: A Fast and Robust Motion Representation for +<br/>Video Action Recognition +<br/><b>The University of Sydney 2SenseTime Research 3The Chinese University of Hong Kong</b></td><td>('1837024', 'Shuyang Sun', 'shuyang sun')<br/>('1874900', 'Zhanghui Kuang', 'zhanghui kuang')<br/>('37145669', 'Lu Sheng', 'lu sheng')<br/>('3001348', 'Wanli Ouyang', 'wanli ouyang')<br/>('1726357', 'Wei Zhang', 'wei zhang')</td><td>{shuyang.sun wanli.ouyang}@sydney.edu.au +<br/>{wayne.zhang kuangzhanghui}@sensetime.com +<br/>lsheng@ee.cuhk.edu.hk +</td></tr><tr><td>04522dc16114c88dfb0ebd3b95050fdbd4193b90</td><td>Appears in 2nd Canadian Conference on Computer and Robot Vision, Victoria, Canada, 2005. +<br/>Minimum Bayes Error Features for Visual Recognition by Sequential Feature +<br/>Selection and Extraction +<br/>Department of Computer Science +<br/><b>University of British Columbia</b><br/>Department of Electrical and Computer engineering +<br/><b>University of California San Diego</b></td><td>('3265767', 'Gustavo Carneiro', 'gustavo carneiro')<br/>('1699559', 'Nuno Vasconcelos', 'nuno vasconcelos')</td><td>carneiro@cs.ubc.ca +<br/>nuno@ece.ucsd.edu +</td></tr><tr><td>04470861408d14cc860f24e73d93b3bb476492d0</td><td></td><td></td><td></td></tr><tr><td>0486214fb58ee9a04edfe7d6a74c6d0f661a7668</td><td>Patch-based Probabilistic Image Quality Assessment for +<br/>Face Selection and Improved Video-based Face Recognition +<br/>NICTA, PO Box 6020, St Lucia, QLD 4067, Australia ∗ +<br/><b>The University of Queensland, School of ITEE, QLD 4072, Australia</b></td><td>('3026404', 'Yongkang Wong', 'yongkang wong')<br/>('3104113', 'Shaokang Chen', 'shaokang chen')<br/>('40080354', 'Sandra Mau', 'sandra mau')<br/>('1781182', 'Conrad Sanderson', 'conrad sanderson')<br/>('2270092', 'Brian C. Lovell', 'brian c. lovell')</td><td></td></tr><tr><td>0447bdb71490c24dd9c865e187824dee5813a676</td><td>Manifold Estimation in View-based Feature +<br/>Space for Face Synthesis Across Pose +<br/>Paper 27 +</td><td></td><td></td></tr><tr><td>0435a34e93b8dda459de49b499dd71dbb478dc18</td><td>VEGAC: Visual Saliency-based Age, Gender, and Facial Expression Classification +<br/>Using Convolutional Neural Networks +<br/>Department of Electronics and Communication Engineering and +<br/><b>Computer Vision Group, L. D. College of Engineering, Ahmedabad, India</b><br/>the need for handcrafted facial descriptors and data +<br/>preprocessing. D-CNN models have been not only +<br/>successfully applied to human face analysis, but also for +<br/>the visual saliency detection [21, 22, 23]. Visual Saliency +<br/>is fundamentally an intensity map where higher intensity +<br/>signifies regions, where a general human being would +<br/>look, and lower intensities mean decreasing level of visual +<br/>attention. It’s a measure of visual attention of humans +<br/>based on the content of the image. It has numerous +<br/>applications in computer vision and image processing +<br/>tasks. It is still an open problem when considering the MIT +<br/>Saliency Benchmark [24]. +<br/>In previous five years, considering age estimation, +<br/>gender classification and facial expression classification +<br/>accuracies +<br/>increased rapidly on several benchmarks. +<br/>However, in unconstrained environments, i.e. low to high +<br/>occluded face and +<br/>this +<br/>classification tasks are still facing challenges to achieve +<br/>competitive results. Some of the sample images are shown +<br/>in the Fig. 1. +<br/>low-resolution facial +<br/>image, +<br/>Figure 1: Sample images having unconstrained environments i.e. +<br/>occlusion, low resolution. +<br/>In this paper, we tackle the age, gender, and facial +<br/>expression classification problem from different angle. We +<br/>are inspired by the recent progress in the domain of image +<br/>classification and visual saliency prediction using deep +<br/>learning to achieve the competitive results. Based on the +<br/>above motivation our work +<br/>this multi-task +<br/>classification of the facial image is as follows: +<br/> Our VEGAC method uses off-the-shelf face detector +<br/>proposed by Mathias et al. [2] to obtain the location of the +<br/>face in the test image. Then, we increase the margin of +<br/>detected face by 30% and crop the face. After getting the +<br/>cropped face, we pass the cropped face on the Deep Multi- +<br/>for +</td><td>('27343041', 'Ayesha Gurnani', 'ayesha gurnani')<br/>('23922616', 'Vandit Gajjar', 'vandit gajjar')<br/>('22239413', 'Viraj Mavani', 'viraj mavani')<br/>('26425477', 'Yash Khandhediya', 'yash khandhediya')</td><td>{gurnani.ayesha.52, gajjar.vandit.381, mavani.viraj.604, khandhediya.yash.364}@ldce.ac.in +</td></tr><tr><td>043efe5f465704ced8d71a067d2b9d5aa5b59c29</td><td>EGGER ET AL.: OCCLUSION-AWARE 3D MORPHABLE FACE MODELS +<br/>Occlusion-aware 3D Morphable Face Models +<br/>Department of Mathematics and +<br/>Computer Science +<br/><b>University of Basel</b><br/>Basel Switzerland +<br/>http://gravis.cs.unibas.ch +<br/>Andreas Morel-Forster +</td><td>('34460642', 'Bernhard Egger', 'bernhard egger')<br/>('49462138', 'Andreas Schneider', 'andreas schneider')<br/>('39550224', 'Clemens Blumer', 'clemens blumer')<br/>('1987368', 'Sandro Schönborn', 'sandro schönborn')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td>bernhard.egger@unibas.ch +<br/>andreas.schneider@unibas.ch +<br/>clemens.blumer@unibas.ch +<br/>andreas.forster@unibas.ch +<br/>sandro.schoenborn@unibas.ch +<br/>thomas.vetter@unibas.ch +</td></tr><tr><td>044ba70e6744e80c6a09fa63ed6822ae241386f2</td><td>TO APPEAR IN AUTONOMOUS ROBOTS, SPECIAL ISSUE IN LEARNING FOR HUMAN-ROBOT COLLABORATION +<br/>Early Prediction for Physical Human Robot +<br/>Collaboration in the Operating Room +</td><td>('2641330', 'Tian Zhou', 'tian zhou')</td><td></td></tr><tr><td>04661729f0ff6afe4b4d6223f18d0da1d479accf</td><td>From Facial Parts Responses to Face Detection: A Deep Learning Approach +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Key Lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('1692609', 'Shuo Yang', 'shuo yang')<br/>('1693209', 'Ping Luo', 'ping luo')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{ys014, pluo, ccloy, xtang}@ie.cuhk,edu.hk +</td></tr><tr><td>04dcdb7cb0d3c462bdefdd05508edfcff5a6d315</td><td>Assisting the training of deep neural networks +<br/>with applications to computer vision +<br/>tesi doctoral està subjecta a +<br/>la +<br/>Aquesta +<br/>CompartirIgual 4.0. Espanya de Creative Commons. +<br/>Esta tesis doctoral está sujeta a la licencia Reconocimiento - NoComercial – CompartirIgual +<br/>4.0. España de Creative Commons. +<br/>This doctoral thesis is licensed under the Creative Commons Attribution-NonCommercial- +<br/>ShareAlike 4.0. Spain License. +<br/>llicència Reconeixement- NoComercial – +</td><td>('3995639', 'Adriana Romero', 'adriana romero')</td><td></td></tr><tr><td>044fdb693a8d96a61a9b2622dd1737ce8e5ff4fa</td><td>Dynamic Texture Recognition Using Local Binary +<br/>Patterns with an Application to Facial Expressions +</td><td>('1757287', 'Guoying Zhao', 'guoying zhao')</td><td></td></tr><tr><td>04f55f81bbd879773e2b8df9c6b7c1d324bc72d8</td><td>Multi-view Face Analysis Based on Gabor Features +<br/><b>College of Information and Control Engineering in China University of Petroleum</b><br/>Qingdao 266580, China +<br/> +</td><td>('1707922', 'Hongli Liu', 'hongli liu')</td><td></td></tr><tr><td>04250e037dce3a438d8f49a4400566457190f4e2</td><td></td><td></td><td></td></tr><tr><td>0431e8a01bae556c0d8b2b431e334f7395dd803a</td><td>Learning Localized Perceptual Similarity Metrics for Interactive Categorization +<br/>Google Inc. +<br/>google.com +</td><td>('2367820', 'Catherine Wah', 'catherine wah')</td><td></td></tr><tr><td>04b4c779b43b830220bf938223f685d1057368e9</td><td>Video retrieval based on deep convolutional +<br/>neural network +<br/>Yajiao Dong +<br/>School of Information and Electronics, +<br/>Beijing Institution of Technology, Beijing, China +<br/>Jianguo Li +<br/>School of Information and Electronics, +<br/>Beijing Institution of Technology, Beijing, China +</td><td></td><td>yajiaodong@bit.edu.cn +<br/>jianguoli@bit.edu.cn +</td></tr><tr><td>04616814f1aabe3799f8ab67101fbaf9fd115ae4</td><td><b>UNIVERSIT´EDECAENBASSENORMANDIEU.F.R.deSciences´ECOLEDOCTORALESIMEMTH`ESEPr´esent´eeparM.GauravSHARMAsoutenuele17D´ecembre2012envuedel’obtentionduDOCTORATdel’UNIVERSIT´EdeCAENSp´ecialit´e:InformatiqueetapplicationsArrˆet´edu07aoˆut2006Titre:DescriptionS´emantiquedesHumainsPr´esentsdansdesImagesVid´eo(SemanticDescriptionofHumansinImages)TheworkpresentedinthisthesiswascarriedoutatGREYC-UniversityofCaenandLEAR–INRIAGrenobleJuryM.PatrickPEREZDirecteurdeRechercheINRIA/Technicolor,RennesRapporteurM.FlorentPERRONNINPrincipalScientistXeroxRCE,GrenobleRapporteurM.JeanPONCEProfesseurdesUniversit´esENS,ParisExaminateurMme.CordeliaSCHMIDDirectricedeRechercheINRIA,GrenobleDirectricedeth`eseM.Fr´ed´ericJURIEProfesseurdesUniversit´esUniversit´edeCaenDirecteurdeth`ese</b></td><td></td><td></td></tr><tr><td>04c2cda00e5536f4b1508cbd80041e9552880e67</td><td>Hipster Wars: Discovering Elements +<br/>of Fashion Styles +<br/><b>University of North Carolina at Chapel Hill, NC, USA</b><br/><b>Tohoku University, Japan</b></td><td>('1772294', 'M. Hadi Kiapour', 'm. hadi kiapour')<br/>('1721910', 'Kota Yamaguchi', 'kota yamaguchi')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('1685538', 'Tamara L. Berg', 'tamara l. berg')</td><td>{hadi,aberg,tlberg}@cs.unc.edu +<br/>kyamagu@vision.is.tohoku.ac.jp +</td></tr><tr><td>04ff69aa20da4eeccdabbe127e3641b8e6502ec0</td><td>Sequential Face Alignment via Person-Specific Modeling in the Wild +<br/><b>Rutgers University</b><br/><b>University of Texas at Arlington</b><br/>Piscataway, NJ 08854 +<br/>Arlington, TX 76019 +<br/><b>Rutgers University</b><br/>Piscataway, NJ 08854 +</td><td>('4340744', 'Xi Peng', 'xi peng')<br/>('1768190', 'Junzhou Huang', 'junzhou huang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>xpeng.nb@cs.rutgers.edu +<br/>jzhuang@uta.edu +<br/>dnm@cs.rutgers.edu +</td></tr><tr><td>046a694bbb3669f2ff705c6c706ca3af95db798c</td><td>Conditional Convolutional Neural Network for Modality-aware Face Recognition +<br/><b>Imperial College London</b><br/><b>National University of Singapore</b><br/>3Panasonic R&D Center Singapore +</td><td>('34336393', 'Chao Xiong', 'chao xiong')<br/>('1874505', 'Xiaowei Zhao', 'xiaowei zhao')<br/>('40245930', 'Danhang Tang', 'danhang tang')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('1700968', 'Tae-Kyun Kim', 'tae-kyun kim')</td><td>{chao.xiong10, x.zhao, d.tang11}@imperial.ac.uk, Karlekar.Jayashree@sg.panasonic.com, eleyans@nus.edu.sg, tk.kim@imperial.ac.uk +</td></tr><tr><td>047d7cf4301cae3d318468fe03a1c4ce43b086ed</td><td>Co-Localization of Audio Sources in Images Using +<br/>Binaural Features and Locally-Linear Regression +<br/>To cite this version: +<br/>Sources in Images Using Binaural Features and Locally-Linear Regression. IEEE Transactions +<br/>on Audio Speech and Language Processing, 2015, 15p. <hal-01112834> +<br/>HAL Id: hal-01112834 +<br/>https://hal.inria.fr/hal-01112834 +<br/>Submitted on 3 Feb 2015 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('3307172', 'Antoine Deleforge', 'antoine deleforge')<br/>('1794229', 'Radu Horaud', 'radu horaud')<br/>('2159538', 'Yoav Y. Schechner', 'yoav y. schechner')<br/>('1780746', 'Laurent Girin', 'laurent girin')<br/>('3307172', 'Antoine Deleforge', 'antoine deleforge')<br/>('1794229', 'Radu Horaud', 'radu horaud')<br/>('2159538', 'Yoav Y. Schechner', 'yoav y. schechner')<br/>('1780746', 'Laurent Girin', 'laurent girin')</td><td></td></tr><tr><td>04317e63c08e7888cef480fe79f12d3c255c5b00</td><td>Face Recognition Using a Unified 3D Morphable Model +<br/>Hu, G., Yan, F., Chan, C-H., Deng, W., Christmas, W., Kittler, J., & Robertson, N. M. (2016). Face Recognition +<br/>Using a Unified 3D Morphable Model. In Computer Vision – ECCV 2016: 14th European Conference, +<br/>Amsterdam, The Netherlands, October 11-14, 2016, Proceedings, Part VIII (pp. 73-89). (Lecture Notes in +<br/>Computer Science; Vol. 9912). Springer Verlag. DOI: 10.1007/978-3-319-46484-8_5 +<br/>Published in: +<br/>Computer Vision – ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11-14, +<br/>2016, Proceedings, Part VIII +<br/>Document Version: +<br/>Peer reviewed version +<br/><b>Queen's University Belfast - Research Portal</b><br/><b>Link to publication record in Queen's University Belfast Research Portal</b><br/>Publisher rights +<br/>The final publication is available at Springer via http://dx.doi.org/10.1007/978-3-319-46484-8_5 +<br/>General rights +<br/><b>Copyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or other</b><br/>copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated +<br/>with these rights. +<br/>Take down policy +<br/>The Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made to +<br/>ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the +<br/>Download date:12. Sep. 2018 +</td><td></td><td>Research Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk. +</td></tr><tr><td>046865a5f822346c77e2865668ec014ec3282033</td><td>Discovering Informative Social Subgraphs and Predicting +<br/>Pairwise Relationships from Group Photos +<br/><b>National Taiwan University, Taipei, Taiwan</b><br/>†Academia Sinica, Taipei, Taiwan +</td><td>('35081710', 'Yan-Ying Chen', 'yan-ying chen')<br/>('1716836', 'Winston H. Hsu', 'winston h. hsu')<br/>('1704678', 'Hong-Yuan Mark Liao', 'hong-yuan mark liao')</td><td>yanying@cmlab.csie.ntu.edu.tw, winston@csie.ntu.edu.tw, liao@iis.sinica.edu.tw +</td></tr><tr><td>047bb1b1bd1f19b6c8d7ee7d0324d5ecd1a3efff</td><td>Unsupervised Training for 3D Morphable Model Regression +<br/><b>Princeton University</b><br/>2Google Research +<br/>3MIT CSAIL +</td><td>('32627314', 'Kyle Genova', 'kyle genova')<br/>('39578349', 'Forrester Cole', 'forrester cole')</td><td></td></tr><tr><td>0470b0ab569fac5bbe385fa5565036739d4c37f8</td><td>Automatic Face Naming with Caption-based Supervision +<br/>To cite this version: +<br/>with Caption-based Supervision. CVPR 2008 - IEEE Conference on Computer Vision +<br/>Pattern Recognition, +<br/>ciety, +<br/><10.1109/CVPR.2008.4587603>. <inria-00321048v2> +<br/>Jun +<br/>2008, +<br/>pp.1-8, +<br/>2008, Anchorage, United +<br/>so- +<br/><http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4587603>. +<br/>IEEE Computer +<br/>States. +<br/>HAL Id: inria-00321048 +<br/>https://hal.inria.fr/inria-00321048v2 +<br/>Submitted on 11 Apr 2011 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('2737253', 'Matthieu Guillaumin', 'matthieu guillaumin')<br/>('1722052', 'Thomas Mensink', 'thomas mensink')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')<br/>('2737253', 'Matthieu Guillaumin', 'matthieu guillaumin')<br/>('1722052', 'Thomas Mensink', 'thomas mensink')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>6a3a07deadcaaab42a0689fbe5879b5dfc3ede52</td><td>Learning to Estimate Pose by Watching Videos +<br/>Department of Computer Science and Engineering +<br/>IIT Kanpur +</td><td>('36668573', 'Prabuddha Chakraborty', 'prabuddha chakraborty')<br/>('1744135', 'Vinay P. Namboodiri', 'vinay p. namboodiri')</td><td>{prabudc, vinaypn} @iitk.ac.in +</td></tr><tr><td>6a67e6fbbd9bcd3f724fe9e6cecc9d48d1b6ad4d</td><td>Cooperative Learning with Visual Attributes +<br/><b>Carnegie Mellon University</b><br/>Georgia Tech +</td><td>('32519394', 'Tanmay Batra', 'tanmay batra')<br/>('1713589', 'Devi Parikh', 'devi parikh')</td><td>tbatra@cmu.edu +<br/>parikh@gatech.edu +</td></tr><tr><td>6afed8dc29bc568b58778f066dc44146cad5366c</td><td>Kernel Hebbian Algorithm for Single-Frame +<br/>Super-Resolution +<br/><b>Max Planck Institute f ur biologische Kybernetik</b><br/>Spemannstr. 38, D-72076 T¨ubingen, Germany +<br/>http://www.kyb.tuebingen.mpg.de/ +</td><td>('1808255', 'Kwang In Kim', 'kwang in kim')<br/>('30541601', 'Matthias O. Franz', 'matthias o. franz')</td><td>{kimki, mof, bs}@tuebingen.mpg.de +</td></tr><tr><td>6ad107c08ac018bfc6ab31ec92c8a4b234f67d49</td><td></td><td></td><td></td></tr><tr><td>6a184f111d26787703f05ce1507eef5705fdda83</td><td></td><td></td><td></td></tr><tr><td>6a16b91b2db0a3164f62bfd956530a4206b23fea</td><td>A Method for Real-Time Eye Blink Detection and Its Application +<br/>Mahidol Wittayanusorn School +<br/>Puttamonton, Nakornpatom 73170, Thailand +</td><td></td><td>Chinnawat.Deva@gmail.com +</td></tr><tr><td>6a806978ca5cd593d0ccd8b3711b6ef2a163d810</td><td>Facial feature tracking for Emotional Dynamic +<br/>Analysis +<br/>1ISIR, CNRS UMR 7222 +<br/>Univ. Pierre et Marie Curie, Paris +<br/>2LAMIA, EA 4540 +<br/>Univ. of Fr. West Indies & Guyana +</td><td>('3093849', 'Thibaud Senechal', 'thibaud senechal')<br/>('3074790', 'Vincent Rapp', 'vincent rapp')<br/>('2554802', 'Lionel Prevost', 'lionel prevost')</td><td>{rapp, senechal}@isir.upmc.fr +<br/>lionel.prevost@univ-ag.fr +</td></tr><tr><td>6a8a3c604591e7dd4346611c14dbef0c8ce9ba54</td><td>ENTERFACE’10, JULY 12TH - AUGUST 6TH, AMSTERDAM, THE NETHERLANDS. +<br/>58 +<br/>An Affect-Responsive Interactive Photo Frame +</td><td>('1713360', 'Ilkka Kosunen', 'ilkka kosunen')<br/>('32062164', 'Marcos Ortega Hortas', 'marcos ortega hortas')<br/>('1764521', 'Albert Ali Salah', 'albert ali salah')</td><td></td></tr><tr><td>6aa43f673cc42ed2fa351cbc188408b724cb8d50</td><td></td><td></td><td></td></tr><tr><td>6a2b83c4ae18651f1a3496e48a35b0cd7a2196df</td><td>Top Rank Supervised Binary Coding for Visual Search +<br/>Department of ECE +<br/>School of Electronic Engineering +<br/>School of Information Science +<br/>UC San Diego +<br/><b>Xidian University</b><br/>and Engineering +<br/><b>Xiamen University</b><br/>Department of Mathematics +<br/>UC San Diego +<br/>IBM T. J. Watson +<br/><b>Research Center</b></td><td>('2451800', 'Dongjin Song', 'dongjin song')<br/>('39059457', 'Wei Liu', 'wei liu')<br/>('1725599', 'Rongrong Ji', 'rongrong ji')<br/>('3520515', 'David A. Meyer', 'david a. meyer')<br/>('1732563', 'John R. Smith', 'john r. smith')</td><td>dosong@ucsd.edu +<br/>wliu@ee.columbia.edu +<br/>rrji@xmu.edu.cn +<br/>dmeyer@math.ucsd.edu +<br/>jsmith@us.ibm.com +</td></tr><tr><td>6a52e6fce541126ff429f3c6d573bc774f5b8d89</td><td>Role of Facial Emotion in Social Correlation +<br/>Department of Computer Science and Engineering +<br/><b>Nagoya Institute of Technology, Gokiso, Showa-ku, Nagoya, 466-8555 Japan</b></td><td>('2159044', 'Pankaj Mishra', 'pankaj mishra')<br/>('47865262', 'Takayuki Ito', 'takayuki ito')</td><td>{pankaj.mishra, rafik}@itolab.nitech.ac.jp, +<br/>ito.takayuki@nitech.ac.jp +</td></tr><tr><td>6a5fe819d2b72b6ca6565a0de117c2b3be448b02</td><td>Supervised and Projected Sparse Coding for Image Classification +<br/>Computer Science and Engineering Department +<br/><b>University of Texas at Arlington</b><br/>Arlington,TX,76019 +</td><td>('39122448', 'Jin Huang', 'jin huang')<br/>('1688370', 'Feiping Nie', 'feiping nie')<br/>('1748032', 'Heng Huang', 'heng huang')</td><td>huangjinsuzhou@gmail.com, feipingnie@gmail.com, heng@uta.edu, chqding@uta.edu +</td></tr><tr><td>6afeb764ee97fbdedfa8f66810dfc22feae3fa1f</td><td>Robust Principal Component Analysis with Complex Noise +<br/><b>School of Mathematics and Statistics, Xi an Jiaotong University, Xi an, China</b><br/><b>School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China</b><br/><b>The Hong Kong Polytechnic University, Hong Kong, China</b></td><td>('40209122', 'Qian Zhao', 'qian zhao')<br/>('1803714', 'Deyu Meng', 'deyu meng')<br/>('7814629', 'Zongben Xu', 'zongben xu')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('36685537', 'Lei Zhang', 'lei zhang')</td><td>TIMMY.ZHAOQIAN@GMAIL.COM +<br/>DYMENG@MAIL.XJTU.EDU.CN +<br/>ZBXU@MAIL.XJTU.EDU.CN +<br/>CSWMZUO@GMAIL.COM +<br/>CSLZHANG@COMP.POLYU.EDU.HK +</td></tr><tr><td>6aa61d28750629febe257d1cb69379e14c66c67f</td><td>Max–Planck–Institut f¨ur biologische Kybernetik +<br/><b>Max Planck Institute for Biological Cybernetics</b><br/>Technical Report No. 109 +<br/>Kernel Hebbian Algorithm for +<br/>Iterative Kernel Principal +<br/>Component Analysis +<br/>Sch¨olkopf1 +<br/>June 2003 +<br/>This report is available in PDF–format via anonymous ftp at ftp://ftp.kyb.tuebingen.mpg.de/pub/mpi-memos/pdf/kha.pdf. The com- +<br/>plete series of Technical Reports is documented at: http://www.kyb.tuebingen.mpg.de/techreports.html +</td><td>('1808255', 'Kwang In Kim', 'kwang in kim')<br/>('30541601', 'Matthias O. Franz', 'matthias o. franz')</td><td>1 Department Sch¨olkopf, email: kimki;mof;bs@tuebingen.mpg.de +</td></tr><tr><td>6ae96f68187f1cdb9472104b5431ec66f4b2470f</td><td><b>Carnegie Mellon University</b><br/><b>Dietrich College Honors Theses</b><br/><b>Dietrich College of Humanities and Social Sciences</b><br/>4-30-2012 +<br/>Improving Task Performance in an Affect-mediated +<br/>Computing System +<br/>Follow this and additional works at: http://repository.cmu.edu/hsshonors +<br/>Part of the Databases and Information Systems Commons +</td><td>('29120285', 'Vivek Pai', 'vivek pai')</td><td>Research Showcase @ CMU +<br/>Carnegie Mellon University, vpai@cmu.edu +<br/>This Thesis is brought to you for free and open access by the Dietrich College of Humanities and Social Sciences at Research Showcase @ CMU. It has +<br/>been accepted for inclusion in Dietrich College Honors Theses by an authorized administrator of Research Showcase @ CMU. For more information, +<br/>please contact research-showcase@andrew.cmu.edu. +</td></tr><tr><td>6a4419ce2338ea30a570cf45624741b754fa52cb</td><td>Statistical transformer networks: learning shape +<br/>and appearance models via self supervision +<br/><b>University of York</b></td><td>('39180407', 'Anil Bas', 'anil bas')<br/>('1687021', 'William A. P. Smith', 'william a. p. smith')</td><td>{ab1792,william.smith}@york.ac.uk +</td></tr><tr><td>6af65e2a1eba6bd62843e7bf717b4ccc91bce2b8</td><td>A New Weighted Sparse Representation Based +<br/>on MSLBP and Its Application to Face Recognition +<br/><b>School of IoT Engineering, Jiangnan University, Wuxi 214122, China</b></td><td>('1823451', 'He-Feng Yin', 'he-feng yin')<br/>('37020604', 'Xiao-Jun Wu', 'xiao-jun wu')</td><td>yinhefeng@126.com, wu_xiaojun@yahoo.com.cn +</td></tr><tr><td>6a657995b02bc9dee130701138ea45183c18f4ae</td><td>THE TIMING OF FACIAL MOTION IN POSED AND SPONTANEOUS SMILES +<br/>J.F. COHN* and K.L.SCHMIDT +<br/><b>University of Pittsburgh</b><br/>Department of Psychology +<br/>4327 Sennott Square, 210 South Bouquet Street +<br/>Pittsburgh, PA 15260, USA +<br/>Revised 19 March 2004 +<br/>Almost all work in automatic facial expression analysis has focused on recognition of prototypic +<br/>expressions rather than dynamic changes in appearance over time. To investigate the relative +<br/>contribution of dynamic features to expression recognition, we used automatic feature tracking to +<br/>measure the relation between amplitude and duration of smile onsets in spontaneous and deliberate +<br/>smiles of 81 young adults of Euro- and African-American background. Spontaneous smiles were of +<br/>smaller amplitude and had a larger and more consistent relation between amplitude and duration than +<br/>deliberate smiles. A linear discriminant classifier using timing and amplitude measures of smile +<br/>onsets achieved a 93% recognition rate. Using timing measures alone, recognition rate declined only +<br/>marginally to 89%. These findings suggest that by extracting and representing dynamic as well as +<br/>morphological features, automatic facial expression analysis can begin to discriminate among the +<br/>message values of morphologically similar expressions. +<br/> Keywords: automatic facial expression analysis, timing, spontaneous facial behavior +<br/> AMS Subject Classification: +<br/>1. Introduction +<br/>Almost all work in automatic facial expression analysis has sought to recognize either +<br/>prototypic expressions of emotion (e.g., joy or anger) or more molecular appearance +<br/>prototypes such as FACS action units. This emphasis on prototypic expressions follows +<br/>from the work of Darwin10and more recently Ekman12 who proposed that basic emotions +<br/>have corresponding prototypic expressions and described their components, such as +<br/>crows-feet wrinkles lateral to the outer eye corners, in emotion-specified joy expressions. +<br/>Considerable evidence suggests that six prototypic expressions (joy, surprise, anger, +<br/>sadness, disgust, and fear) are universal in their performance and in their perception12 +<br/>and can communicate subjective emotion, communicative +<br/>intent, and action +<br/>tendencies.18, 19, 26 +</td><td></td><td>*jeffcohn@pitt.edu +<br/>kschmidt@pitt.edu +</td></tr><tr><td>6a0368b4e132f4aa3bbdeada8d894396f201358a</td><td>One-Class Multiple Instance Learning via +<br/>Robust PCA for Common Object Discovery +<br/><b>Huazhong University of Science and Technology</b><br/>2Visual Computing Group, Microsoft Research Asia +<br/>3Lab of Neuro Imaging and Department of Computer Science, UCLA +</td><td>('2443233', 'Xinggang Wang', 'xinggang wang')<br/>('2554701', 'Zhengdong Zhang', 'zhengdong zhang')<br/>('1700297', 'Yi Ma', 'yi ma')<br/>('1686737', 'Xiang Bai', 'xiang bai')<br/>('1743698', 'Wenyu Liu', 'wenyu liu')<br/>('1736745', 'Zhuowen Tu', 'zhuowen tu')</td><td>{wxghust,zhangzdfaint}@gmail.com, mayi@microsoft.com, +<br/>{xbai,liuwy}@hust.edu.cn, ztu@loni.ucla.edu +</td></tr><tr><td>6ab33fa51467595f18a7a22f1d356323876f8262</td><td>Ordinal Hyperplanes Ranker with Cost Sensitivities for Age Estimation +<br/><b>Institute of Information Science, Academia Sinica, Taipei, Taiwan</b><br/><b>Research Center for Information Technology Innovation, Academia Sinica, Taipei, Taiwan</b><br/><b>National Taiwan University, Taipei, Taiwan</b><br/><b>Graduate Institute of Networking and Multimedia, National Taiwan University, Taipei, Taiwan</b></td><td>('34692779', 'Kuang-Yu Chang', 'kuang-yu chang')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')<br/>('1732064', 'Yi-Ping Hung', 'yi-ping hung')</td><td>{kuangyu, song}@iis.sinica.edu.tw, hung@csie.ntu.edu.tw +</td></tr><tr><td>6aefe7460e1540438ffa63f7757c4750c844764d</td><td>Non-rigid Segmentation using Sparse Low Dimensional Manifolds and +<br/>Deep Belief Networks ∗ +<br/>Instituto de Sistemas e Rob´otica +<br/>Instituto Superior T´ecnico, Portugal +</td><td>('3259175', 'Jacinto C. Nascimento', 'jacinto c. nascimento')</td><td></td></tr><tr><td>6a2ac4f831bd0f67db45e7d3cdaeaaa075e7180a</td><td>Excitation Dropout: +<br/>Encouraging Plasticity in Deep Neural Networks +<br/>1Pattern Analysis & Computer Vision (PAVIS), Istituto Italiano di Tecnologia +<br/><b>Boston University</b><br/>3Adobe Research +<br/><b>University of Verona</b></td><td>('40063519', 'Andrea Zunino', 'andrea zunino')<br/>('3298267', 'Sarah Adel Bargal', 'sarah adel bargal')<br/>('2322579', 'Pietro Morerio', 'pietro morerio')<br/>('1701293', 'Jianming Zhang', 'jianming zhang')<br/>('1749590', 'Stan Sclaroff', 'stan sclaroff')<br/>('1727204', 'Vittorio Murino', 'vittorio murino')</td><td>{andrea.zunino,vittorio.murino}@iit.it, +<br/>{sbargal,sclaroff}@bu.edu, jianmzha@adobe.com +</td></tr><tr><td>6a4ebd91c4d380e21da0efb2dee276897f56467a</td><td>HOG ACTIVE APPEARANCE MODELS +<br/><b>cid:2)Imperial College London, U.K</b><br/><b>University of Lincoln, School of Computer Science, U.K</b></td><td>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('2575567', 'Joan Alabort-i-Medina', 'joan alabort-i-medina')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td></td></tr><tr><td>6a1beb34a2dfcdf36ae3c16811f1aef6e64abff2</td><td></td><td></td><td></td></tr><tr><td>6a7e464464f70afea78552c8386f4d2763ea1d9c</td><td>Review Article +<br/>International Journal of Current Engineering and Technology +<br/>E-ISSN 2277 – 4106, P-ISSN 2347 - 5161 +<br/>©2014 INPRESSCO +<br/>, All Rights Reserved +<br/>Available at http://inpressco.com/category/ijcet +<br/>Facial Landmark Localization – A Literature Survey +<br/><b>PES Institute of Technology, Bangalore, Karnataka, India</b><br/>Accepted 25 May 2014, Available online 01 June2014, Vol.4, No.3 (June 2014) +</td><td></td><td></td></tr><tr><td>32925200665a1bbb4fc8131cd192cb34c2d7d9e3</td><td>3-9 +<br/>MVA2009 IAPR Conference on Machine Vision Applications, May 20-22, 2009, Yokohama, JAPAN +<br/>An Active Appearance Model with a Derivative-Free +<br/>Optimization +<br/><b>CNRS , Institute of Automation of the Chinese Academy of Sciences</b><br/>95, Zhongguancun Dong Lu, PO Box 2728 − Beijing 100190 − PR China +<br/>LIAMA Sino-French IT Lab. +</td><td>('8214735', 'Jixia Zhang', 'jixia zhang')<br/>('1742818', 'Franck Davoine', 'franck davoine')<br/>('3364363', 'Chunhong Pan', 'chunhong pan')</td><td>Franck.Davoine@gmail.com +</td></tr><tr><td>322c063e97cd26f75191ae908f09a41c534eba90</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Improving Image Classification using Semantic Attributes +<br/>Received: date / Accepted: date +</td><td>('1758652', 'Yu Su', 'yu su')</td><td></td></tr><tr><td>325b048ecd5b4d14dce32f92bff093cd744aa7f8</td><td>CVPR +<br/>#2670 +<br/>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>045 +<br/>046 +<br/>047 +<br/>048 +<br/>049 +<br/>050 +<br/>051 +<br/>052 +<br/>053 +<br/>CVPR 2008 Submission #2670. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +<br/>CVPR +<br/>#2670 +<br/>Multi-Image Graph Cut Clothing Segmentation for Recognizing People +<br/>Anonymous CVPR submission +<br/>Paper ID 2670 +</td><td></td><td></td></tr><tr><td>32f7e1d7fa62b48bedc3fcfc9d18fccc4074d347</td><td>HIERARCHICAL SPARSE AND COLLABORATIVE LOW-RANK REPRESENTATION FOR +<br/>EMOTION RECOGNITION +<br/><b>Johns Hopkins University, 3400 N. Charles Street, Baltimore, MD 21218, USA</b></td><td>('40031188', 'Xiang Xiang', 'xiang xiang')<br/>('31507586', 'Minh Dao', 'minh dao')<br/>('1678633', 'Gregory D. Hager', 'gregory d. hager')<br/>('1709073', 'Trac D. Tran', 'trac d. tran')</td><td>{xxiang, minh.dao, ghager1, trac}@jhu.edu +</td></tr><tr><td>32d8e555441c47fc27249940991f80502cb70bd5</td><td>Machine Learning Models that Remember Too Much +<br/><b>Cornell University</b><br/>Cornell Tech +<br/>Cornell Tech +</td><td>('3469125', 'Congzheng Song', 'congzheng song')<br/>('1723945', 'Vitaly Shmatikov', 'vitaly shmatikov')<br/>('1707461', 'Thomas Ristenpart', 'thomas ristenpart')</td><td>cs2296@cornell.edu +<br/>ristenpart@cornell.edu +<br/>shmat@cs.cornell.edu +</td></tr><tr><td>3294e27356c3b1063595885a6d731d625b15505a</td><td>Illumination Face Spaces are Idiosyncratic +<br/>2, H. Kley1, C. Peterson1 ∗ +<br/><b>Colorado State University, Fort Collins, CO 80523, USA</b></td><td>('2640182', 'Jen-Mei Chang', 'jen-mei chang')</td><td></td></tr><tr><td>324f39fb5673ec2296d90142cf9a909e595d82cf</td><td>Hindawi Publishing Corporation +<br/>Mathematical Problems in Engineering +<br/>Volume 2011, Article ID 864540, 15 pages +<br/>doi:10.1155/2011/864540 +<br/>Research Article +<br/>Relationship Matrix Nonnegative +<br/>Decomposition for Clustering +<br/>Faculty of Science and State Key Laboratory for Manufacturing Systems Engineering, Xi’an Jiaotong +<br/><b>University, Xi an Shaanxi Province, Xi an 710049, China</b><br/>Received 18 January 2011; Revised 28 February 2011; Accepted 9 March 2011 +<br/>Copyright q 2011 J.-Y. Pan and J.-S. Zhang. This is an open access article distributed under +<br/>the Creative Commons Attribution License, which permits unrestricted use, distribution, and +<br/>reproduction in any medium, provided the original work is properly cited. +<br/>Nonnegative matrix factorization (cid:2)NMF(cid:3) is a popular tool for analyzing the latent structure of non- +<br/>negative data. For a positive pairwise similarity matrix, symmetric NMF (cid:2)SNMF(cid:3) and weighted +<br/>NMF (cid:2)WNMF(cid:3) can be used to cluster the data. However, both of them are not very efficient +<br/>for the ill-structured pairwise similarity matrix. In this paper, a novel model, called relationship +<br/>matrix nonnegative decomposition (cid:2)RMND(cid:3), is proposed to discover the latent clustering structure +<br/>from the pairwise similarity matrix. The RMND model is derived from the nonlinear NMF +<br/>algorithm. RMND decomposes a pairwise similarity matrix into a product of three low rank +<br/>nonnegative matrices. The pairwise similarity matrix is represented as a transformation of a +<br/>positive semidefinite matrix which pops out the latent clustering structure. We develop a learning +<br/>procedure based on multiplicative update rules and steepest descent method to calculate the +<br/>nonnegative solution of RMND. Experimental results in four different databases show that the +<br/>proposed RMND approach achieves higher clustering accuracy. +<br/>1. Introduction +<br/>Nonnegative matrix factorization (cid:2)NMF(cid:3) (cid:6)1(cid:7) has been introduced as an effective technique for +<br/>analyzing the latent structure of nonnegative data such as images and documents. A variety +<br/>of real-world applications of NMF has been found in many areas such as machine learning, +<br/>signal processing (cid:6)2–4(cid:7), data clustering (cid:6)5, 6(cid:7), and computer vision (cid:6)7(cid:7). +<br/>Most applications focus on the clustering aspect of NMF (cid:6)8, 9(cid:7). Each sample can be +<br/>represented as a linear combination of clustering centroids. Recently, a theoretic analysis +<br/>has shown the equivalence between NMF and K-means/spectral clustering (cid:6)10(cid:7). Symmetric +<br/>NMF (cid:2)SNMF(cid:3) (cid:6)10(cid:7) is an extension of NMF. It aims at learning clustering structure from +<br/>the kernel matrix or pairwise similarity matrix which is positive semidefinite. When the simi- +<br/>larity matrix is not positive semidefinite, SNMF is not able to capture the clustering structure +</td><td>('9416881', 'Ji-Yuan Pan', 'ji-yuan pan')<br/>('2265568', 'Jiang-She Zhang', 'jiang-she zhang')<br/>('14464924', 'Angelo Luongo', 'angelo luongo')</td><td>Correspondence should be addressed to Ji-Yuan Pan, panjiyuan@gmail.com +</td></tr><tr><td>321bd4d5d80abb1bae675a48583f872af3919172</td><td>Wang et al. EURASIP Journal on Image and Video Processing (2016) 2016:44 +<br/>DOI 10.1186/s13640-016-0152-3 +<br/>EURASIP Journal on Image +<br/>and Video Processing +<br/>R EV I E W +<br/>Entropy-weighted feature-fusion method +<br/>for head-pose estimation +<br/>Open Access +</td><td>('40579241', 'Kang Liu', 'kang liu')<br/>('2076553', 'Xu Qian', 'xu qian')</td><td></td></tr><tr><td>3240c9359061edf7a06bfeb7cc20c103a65904c2</td><td>PPR-FCN: Weakly Supervised Visual Relation Detection via Parallel Pairwise +<br/>R-FCN +<br/><b>Columbia University, National University of Singapore</b></td><td>('5462268', 'Hanwang Zhang', 'hanwang zhang')<br/>('26538630', 'Zawlin Kyaw', 'zawlin kyaw')<br/>('46380822', 'Jinyang Yu', 'jinyang yu')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>{hanwangzhang, kzl.zawlin, yjy941124}@gmail.com; shih.fu.chang@columbia.edu +</td></tr><tr><td>32b8c9fd4e3f44c371960eb0074b42515f318ee7</td><td></td><td></td><td></td></tr><tr><td>32575ffa69d85bbc6aef5b21d73e809b37bf376d</td><td>-)5741/ *1-641+ 5)2- 37)16; 1 6-45 . *1-641+ 1.4)61 +<br/>7ELAHIEJO B JJ=M= +<br/>)*564)+6 +<br/>IKHA L=HE=JEI E >EAJHE? I=FA GK=EJO 9A >ACE MEJD +<br/>IKHAAJI 9A JDA IDM JD=J JDA >EAJHE? EBH=JE BH +<br/>JA EI JDA A= D(p(cid:107)q) BH = FAHII E JDA FFK=JE 1 +<br/>BH I= ALAI B >KH MEJD = =IOFJJE? >AD=LEH =J =HCAH +<br/>>KH +<br/> 164,7+61 +<br/>*EAJHE? I=FA GK=EJO EI = A=IKHA B JDA KIABKAII B = +<br/>GK=EJO +<br/>F=FAH MA FHFIA = AM =FFH=?D J A=IKHA JDEI GK=JEJO +<br/>JDA EJKEJELA >IAHL=JE JD=J = DECD GK=EJO >EAJHE? E=CA +<br/>>EAJHE? EBH=JE +<br/>EIIKAI E >EAJHE? JA?DCO .H AN=FA A B JDA IJ +<br/>? >EAJHE? GKAIJEI EI JD=J B KEGKAAII AC J MD=J +<br/>ANJAJ =HA CAHFHEJI KEGKA .H JDA FEJ B LEAM B +<br/>=>A EBH=JE EI =L=E=>A BH = CELA JA?DCO IK?D +<br/> $ " +<br/>1 JDEI F=FAH MA A=>H=JA = =FFH=?D J +<br/>BMI +<br/>AJI +<br/> >ABHA = >EAJHE? A=IKHAAJ t0 =J MDE?D JEA MA O +<br/>M = FAHI p EI F=HJ B = FFK=JE q MDE?D =O >A JDA +</td><td></td><td>4E?D=H@ ;K=H= =@ )@O )@AH +<br/>5?D B 1BH=JE 6A?DCO =@ -CEAAHEC +<br/>J=HE +==@= +<br/>6DEI F=FAH @ALAFI = AM =FFH=?D J K@AHIJ=@ =@ A= +<br/>JDA EJKEJE JD=J @ACH=@=JEI J = >EAJHE? I=FA ME HA +<br/>@K?A JDA =KJ B E@AJE=>A EBH=JE =L=E=>A 1 H +<br/>@AH J A=IKHA JDA =KJ B E@AJE=>A EBH=JE MA +<br/>@AA >EAJHE? EBH=JE =I JDA @A?HA=IA E K?AHJ=EJO +<br/>=>KJ JDA E@AJEJO B = FAHI @KA J = IAJ B >EAJHE? A= +<br/>= FAHI =O >A ?=?K=JA@ >O JDA HA=JELA AJHFO D(p(cid:107)q) +<br/>>AJMAA JDA FFK=JE BA=JKHA @EIJHE>KJE q =@ JDA FAHII +<br/>BA=JKHA @EIJHE>KJE p 6DA >EAJHE? EBH=JE BH = IOI +<br/>H@AH J FH=?JE?=O A=IKHA D(p(cid:107)q) MEJD EEJA@ @=J= I= +<br/>FAI MA EJH@K?A = =CHEJD MDE?D HACK=HEAI = /=KIIE= +<br/>@A B JDA BA=JKHA ?L=HE=?AI ) AN=FA B JDEI AJD@ +<br/>EI IDM BH 2+) .EIDAH EA=H @EI?HEE=J ., =@ 1+) +<br/>>=IA@ B=?A HA?CEJE MEJD >EAJHE? EBH=JE ?=?K=JA@ +<br/>J >A 45.0 >EJI 2+) 37.0 >EJI ., 39.0 >EJI 1+) =@ +<br/>55.6 >EJI BKIE B 2+) =@ ., BA=JKHAI *=IA@ JDEI +<br/>@AEJE B >EAJHE? EBH=JE MA IEK=JA @ACH=@=JEI +<br/>B >EAJHE? E=CAI =@ ?=?K=JA JDA HAIKJEC @A?HA=IA E +<br/>>EAJHE? EBH=JE 4AIKJI IDM = GK=IEEA=H @A?HA=IA +<br/>>EAJHE? E=CA ' A HA?AJ @ALAFAJ EI JDA IECEB +<br/>E?=J ALA B EJAHAIJ E IJ=@=H@I BH A=IKHAAJ B >E +<br/>AJHE? GK=EJO .H AN=FA 15 D=I HA?AJO AIJ=>EIDA@ = +<br/>>EAJHE? I=FA GK=EJO @H=BJ IJ=@=H@ ' )??H@EC J ' +<br/>>EAJHE? I=FA GK=EJO =O >A ?IE@AHA@ BH JDA FEJ B +<br/>LEAM B ?D=H=?JAH EDAHAJ BA=JKHAI @AEJO =??KH=?O B BA= +<br/>JKHAI H KJEEJO FHA@E?JA@ >EAJHE?I FAHBH=?A ) CA +<br/>AH= ?IAIKI D=I @ALAFA@ JD=J JDA IJ EFHJ=J A=IKHA +<br/>B = GK=EJO AJHE? EI EJI KJEEJO ` E=CAI AL=K=JA@ =I DECDAH +<br/>GK=EJO KIJ >A JDIA JD=J HAIKJ E >AJJAH E@AJE?=JE B E +<br/>@ELE@K=I =I A=IKHA@ >O = E?HA=IA@ IAF=H=JE B CAKEA +<br/>=@ EFIJH =J?D I?HA @EIJHE>KJEI 6DA =JKHA B >E +<br/>AJHE? I=FA @AEJO D=I IAA EJJA ELAIJEC=JE =JDKCD +<br/>BH IFA?E? >EAJHE? @=EJEAI =CHEJDI J A=IKHA >E +<br/>AJHE? GK=EJO D=LA >AA FHFIA@ .H AN=FA JDA .13 +<br/>=CHEJD EI = ME@AO KIA@ A=IKHA BH CAHFHEJ E=CA +<br/>A ?KHHAJ @EB?KJO EI JD=J JDAHA EI ?IAIKI =I J MD=J +<br/>= A=IKHA B >EAJHE? I=FA @AEJO IDK@ CELA 1 JDEI +<br/>>=IA@ = EBH=JE JDAHAJE? BH=AMH 9A >ACE MEJD +<br/>EI HA KIABK J E@AJEBO JDA E@ELE@K= JD= = M GK=EJO +<br/>E=CA 6DEI IKCCAIJI JD=J JDA GK=JEJO B E@AJE=>A EBH +<br/>=JE @A?HA=IAI MEJD = HA@K?JE E GK=EJO /ELA = M=O J +<br/>A=IKHA JDA @A?HA=IA E EBH=JE ?=KIA@ >O = CELA E +<br/>=CA @ACH=@=JE A ?= A=IKHA JDA =II?E=JA@ @A?HA=IA E +<br/>A=IKHEC >EAJHE? EBH=JE ?JAJ EI HA=JA@ J =O +<br/>E@AJE=>EEJO A =O >A EJAHAIJA@ E DM K?D E@AJE +<br/>=I LE@A IKHLAE=?A 1 JDA ?JANJ B >EAJHE? BKIE +<br/>A MK@ EA J >A =>A J GK=JEBO JDA >EAJHE? EBH= +<br/>JE E A=?D IOIJA E@ELE@K=O =@ JDA FJAJE= C=E BH +<br/>BKIEC JDA IOIJAI )@@EJE=O IK?D = A=IKHA EI HAAL=J +<br/>J >EAJHE? ?HOFJIOIJAI =@ FHEL=?O A=IKHAI 5ALAH= +<br/>=KJDHI D=LA FHAIAJA@ =FFH=?DAI HAAL=J J JDEI GKAIJE +<br/>=@@HAII JDEI GKAIJE >=IA@ @AEJEI BH EBH=JE +<br/>JDAHO 9A @AA JDA JAH ]>EAJHE? EBH=JE^ =I +<br/>>EAJHE? EBH=JE *1 JDA @A?HA=IA E K?AHJ=EJO =>KJ +<br/>JDA E@AJEJO B = FAHI @KA J = IAJ B >EAJHE? A=IKHA +<br/>1 H@AH J EJAHFHAJ JDEI @AEJE MA HABAH J JM EIJ=JI +<br/>MDA F=AJ =@ =BJAH HA?AELEC = IAJ B A=IKHAAJI +<br/>t1 MA D=LA HA EBH=JE =@ AII K?AHJ=EJO =>KJ JDA +<br/>FAHII E@AJEJO +<br/>*=IA@ JDAIA A=IKHAI MA JDA @AA JDA EBH=JE II +<br/>@KA J = @ACH=@=JE E E=CA GK=EJO =I JDA HA=JELA ?D=CA +</td></tr><tr><td>32ecbbd76fdce249f9109594eee2d52a1cafdfc7</td><td>Object Specific Deep Learning Feature and Its Application to Face Detection +<br/><b>University of Nottingham, Ningbo, China</b><br/><b>University of Nottingham, Ningbo, China</b><br/><b>Shenzhen University, Shenzhen, China</b><br/><b>University of Nottingham, Ningbo, China</b></td><td>('3468964', 'Xianxu Hou', 'xianxu hou')<br/>('39508183', 'Ke Sun', 'ke sun')<br/>('1687690', 'LinLin Shen', 'linlin shen')<br/>('1698461', 'Guoping Qiu', 'guoping qiu')</td><td>xianxu.hou@nottingham.edu.cn +<br/>ke.sun@nottingham.edu.cn +<br/>llshen@szu.edu.cn +<br/>guoping.qiu@nottingham.edu.cn +</td></tr><tr><td>32c20afb5c91ed7cdbafb76408c3a62b38dd9160</td><td>Viewing Real-World Faces in 3D +<br/><b>The Open University of Israel, Israel</b></td><td>('1756099', 'Tal Hassner', 'tal hassner')</td><td>hassner@openu.ac.il +</td></tr><tr><td>32a40c43a9bc1f1c1ed10be3b9f10609d7e0cb6b</td><td>Lighting Aware Preprocessing for Face +<br/>Recognition across Varying Illumination +<br/>1 Key Lab of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>CAS), Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>Graduate University of Chinese Academy of Sciences, Beijing 100049, China</b><br/><b>Institute of Digital Media, Peking University, Beijing 100871, China</b></td><td>('34393045', 'Hu Han', 'hu han')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('2343895', 'Laiyun Qing', 'laiyun qing')<br/>('1710220', 'Xilin Chen', 'xilin chen')<br/>('1698902', 'Wen Gao', 'wen gao')</td><td>{hhan,sgshan,lyqing,xlchen,wgao}@jdl.ac.cn +</td></tr><tr><td>329394480fc5e9e96de4250cc1a2b060c3677c94</td><td>Improved Dense Trajectory with Cross Streams +<br/>Graduate School of +<br/>Information +<br/>Science and Technology +<br/><b>University of Tokyo</b><br/>tokyo.ac.jp +<br/>Graduate School of +<br/>Information +<br/>Science and Technology +<br/><b>University of Tokyo</b><br/>tokyo.ac.jp +<br/>Graduate School of +<br/>Information +<br/>Science and Technology +<br/><b>University of Tokyo</b><br/>tokyo.ac.jp +</td><td>('8197937', 'Katsunori Ohnishi', 'katsunori ohnishi')<br/>('2859204', 'Masatoshi Hidaka', 'masatoshi hidaka')<br/>('1790553', 'Tatsuya Harada', 'tatsuya harada')</td><td>ohnishi@mi.t.u- +<br/>hidaka@mi.t.u- +<br/>harada@mi.t.u- +</td></tr><tr><td>32728e1eb1da13686b69cc0bd7cce55a5c963cdd</td><td>Automatic Facial Emotion Recognition Method Based on Eye +<br/>Region Changes +<br/><b>Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran</b><br/><b>Faculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran</b><br/><b>Faculty of Electrical and Computer Engineering, Bu-Ali Sina University, Hamadan, Iran</b><br/>Received: 19/Apr/2015 Revised: 19/Mar/2016 Accepted: 19/Apr/2016 +</td><td>('35191740', 'Nasrollah Moghadam Charkari', 'nasrollah moghadam charkari')<br/>('2239524', 'Muharram Mansoorizadeh', 'muharram mansoorizadeh')</td><td>m.navran@modares.ac.ir +<br/>charkari@modares.ac.ir +<br/>mansoorm@basu.ac.ir +</td></tr><tr><td>32c9ebd2685f522821eddfc19c7c91fd6b3caf22</td><td>Finding Correspondence from Multiple Images +<br/>via Sparse and Low-Rank Decomposition +<br/><b>School of Computer Engineering, Nanyang Technological University, Singapore</b><br/>2 Advanced Digital Sciences Center, Singapore +</td><td>('1920683', 'Zinan Zeng', 'zinan zeng')<br/>('1926757', 'Tsung-Han Chan', 'tsung-han chan')<br/>('2370507', 'Kui Jia', 'kui jia')<br/>('1714390', 'Dong Xu', 'dong xu')</td><td>{znzeng,dongxu}@ntu.edu.sg, {Th.chan,Chris.jia}@adsc.com.sg +</td></tr><tr><td>3270b2672077cc345f188500902eaf7809799466</td><td>Multibiometric Systems: Fusion Strategies and +<br/>Template Security +<br/>By +<br/>A Dissertation +<br/>Submitted to +<br/><b>Michigan State University</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>Doctor of Philosophy +<br/>Department of Computer Science and Engineering +<br/>2008 +</td><td>('34633765', 'Karthik Nandakumar', 'karthik nandakumar')</td><td></td></tr><tr><td>321c8ba38db118d8b02c0ba209be709e6792a2c7</td><td>Learn to Combine Multiple Hypotheses for Accurate Face Alignment +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences, China</b></td><td>('1721677', 'Junjie Yan', 'junjie yan')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1716143', 'Dong Yi', 'dong yi')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jjyan,zlei,dyi,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>324b9369a1457213ec7a5a12fe77c0ee9aef1ad4</td><td>Dynamic Facial Analysis: From Bayesian Filtering to Recurrent Neural Network +<br/>NVIDIA +</td><td>('2931118', 'Jinwei Gu', 'jinwei gu')</td><td>{jinweig,xiaodongy,shalinig,jkautz}@nvidia.com +</td></tr><tr><td>329d58e8fb30f1bf09acb2f556c9c2f3e768b15c</td><td>Leveraging Intra and Inter-Dataset Variations for +<br/>Robust Face Alignment +<br/>Department of Computer Science and Technology +<br/><b>Tsinghua University</b><br/>Department of Information Engineering +<br/><b>The Chinese University of Hong Kong</b></td><td>('38766009', 'Wenyan Wu', 'wenyan wu')<br/>('1692609', 'Shuo Yang', 'shuo yang')</td><td>wwy15@mails.tsinghua.edu.cn +<br/>ys014@ie.cuhk.edu.hk +</td></tr><tr><td>32df63d395b5462a8a4a3c3574ae7916b0cd4d1d</td><td>978-1-4577-0539-7/11/$26.00 ©2011 IEEE +<br/>1489 +<br/>ICASSP 2011 +</td><td></td><td></td></tr><tr><td>35308a3fd49d4f33bdbd35fefee39e39fe6b30b7</td><td></td><td>('1799216', 'Jeong-Jik Seo', 'jeong-jik seo')<br/>('1780155', 'Jisoo Son', 'jisoo son')<br/>('7627712', 'Wesley De Neve', 'wesley de neve')<br/>('1692847', 'Yong Man Ro', 'yong man ro')</td><td></td></tr><tr><td>353b6c1f431feac6edde12b2dde7e6e702455abd</td><td>Multi-scale Patch based Collaborative +<br/>Representation for Face Recognition with +<br/>Margin Distribution Optimization +<br/><b>Biometric Research Center</b><br/><b>The Hong Kong Polytechnic University</b><br/><b>School of Computer Science and Technology, Tianjin University</b></td><td>('2873638', 'Pengfei Zhu', 'pengfei zhu')<br/>('36685537', 'Lei Zhang', 'lei zhang')<br/>('1688792', 'Qinghua Hu', 'qinghua hu')</td><td>{cspzhu,cslzhang}@comp.polyu.edu.hk +</td></tr><tr><td>352d61eb66b053ae5689bd194840fd5d33f0e9c0</td><td>Analysis Dictionary Learning based +<br/>Classification: Structure for Robustness +</td><td>('49501811', 'Wen Tang', 'wen tang')<br/>('1733181', 'Ashkan Panahi', 'ashkan panahi')<br/>('1769928', 'Hamid Krim', 'hamid krim')<br/>('2622498', 'Liyi Dai', 'liyi dai')</td><td></td></tr><tr><td>350da18d8f7455b0e2920bc4ac228764f8fac292</td><td>From: AAAI Technical Report SS-03-08. Compilation copyright © 2003, AAAI (www.aaai.org). All rights reserved. +<br/>Automatic Detecting Neutral Face for Face Authentication and +<br/>Facial Expression Analysis +<br/>Exploratory Computer Vision Group +<br/><b>IBM Thomas J. Watson Research Center</b><br/>PO Box 704, Yorktown Heights, NY 10598 +</td><td>('40383812', 'Ying-li Tian', 'ying-li tian')<br/>('1773140', 'Ruud M. Bolle', 'ruud m. bolle')</td><td>{yltian, bolle}@us.ibm.com +</td></tr><tr><td>3538d2b5f7ab393387ce138611ffa325b6400774</td><td>A DSP-BASED APPROACH FOR THE IMPLEMENTATION OF FACE RECOGNITION +<br/>ALGORITHMS +<br/>A. U. Batur +<br/>B. E. Flinchbaugh +<br/>M. H. Hayes IIl +<br/>Center for Signal and Image Proc. +<br/>Georgia Inst. Of Technology +<br/>Atlanta, GA +<br/>Imaging and Audio Lab. +<br/>Texas Instruments +<br/>Dallas, TX +<br/>Center for Signal and Image Proc. +<br/>Georgia Inst. Of Technology +<br/>Atlanta, CA +</td><td></td><td></td></tr><tr><td>3504907a2e3c81d78e9dfe71c93ac145b1318f9c</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Unconstrained Still/Video-Based Face Verification with Deep +<br/>Convolutional Neural Networks +<br/>Received: date / Accepted: date +</td><td>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('2682056', 'Ching-Hui Chen', 'ching-hui chen')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('26988560', 'Rajeev Ranjan', 'rajeev ranjan')</td><td></td></tr><tr><td>35b1c1f2851e9ac4381ef41b4d980f398f1aad68</td><td>Geometry Guided Convolutional Neural Networks for +<br/>Self-Supervised Video Representation Learning +</td><td>('2551285', 'Chuang Gan', 'chuang gan')<br/>('40206014', 'Boqing Gong', 'boqing gong')<br/>('2473509', 'Kun Liu', 'kun liu')<br/>('49466491', 'Hao Su', 'hao su')<br/>('1744254', 'Leonidas J. Guibas', 'leonidas j. guibas')</td><td></td></tr><tr><td>351c02d4775ae95e04ab1e5dd0c758d2d80c3ddd</td><td>ActionSnapping: Motion-based Video +<br/>Synchronization +<br/>Disney Research +</td><td>('2893744', 'Alexander Sorkine-Hornung', 'alexander sorkine-hornung')</td><td></td></tr><tr><td>35f03f5cbcc21a9c36c84e858eeb15c5d6722309</td><td>Placing Broadcast News Videos in their Social Media +<br/>Context using Hashtags +<br/><b>Columbia University</b></td><td>('2136860', 'Joseph G. Ellis', 'joseph g. ellis')<br/>('2602265', 'Svebor Karaman', 'svebor karaman')<br/>('1786871', 'Hongzhi Li', 'hongzhi li')<br/>('36009509', 'Hong Bin Shim', 'hong bin shim')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>{jge2105, svebor.karaman, hongzhi.li, h.shim, sc250}@columbia.edu +</td></tr><tr><td>35e4b6c20756cd6388a3c0012b58acee14ffa604</td><td>Gender Classification in Large Databases +<br/>E. Ram´on-Balmaseda, J. Lorenzo-Navarro, and M. Castrill´on-Santana (cid:63) +<br/>Universidad de Las Palmas de Gran Canaria +<br/>SIANI +<br/>Spain +</td><td></td><td>enrique.de101@.alu.ulpgc.es{jlorenzo,mcastrillon}@siani.es +</td></tr><tr><td>356b431d4f7a2a0a38cf971c84568207dcdbf189</td><td>Recognize Complex Events from Static Images by Fusing Deep Channels +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen key lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology</b><br/>CAS, China +</td><td>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')</td><td>xy012@ie.cuhk.edu.hk +<br/>zk013@ie.cuhk.edu.hk +<br/>dhlin@ie.cuhk.edu.hk +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>35f921def890210dda4b72247849ad7ba7d35250</td><td>Exemplar-based Graph Matching +<br/>for Robust Facial Landmark Localization +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>http://www.f-zhou.com +<br/>Adobe Research +<br/>San Jose, CA 95110 +</td><td>('1757386', 'Feng Zhou', 'feng zhou')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')</td><td>{jbrandt, zlin}@adobe.com +</td></tr><tr><td>357963a46dfc150670061dbc23da6ba7d6da786e</td><td></td><td></td><td></td></tr><tr><td>35ec9b8811f2d755c7ad377bdc29741b55b09356</td><td>Efficient, Robust and Accurate Fitting of a 3D Morphable Model +<br/><b>University of Basel</b><br/>Bernoullistrasse 16, CH - 4056 Basel, Switzerland +</td><td>('3293655', 'Sami Romdhani', 'sami romdhani')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td>fsami.romdhani, thomas.vetterg@unibas.ch +</td></tr><tr><td>35f1bcff4552632419742bbb6e1927ef5e998eb4</td><td></td><td></td><td></td></tr><tr><td>35c973dba6e1225196566200cfafa150dd231fa8</td><td></td><td></td><td></td></tr><tr><td>35f084ddee49072fdb6e0e2e6344ce50c02457ef</td><td>A Bilinear Illumination Model +<br/>for Robust Face Recognition +<br/>The Harvard community has made this +<br/>article openly available. Please share how +<br/>this access benefits you. Your story matters +<br/>Citation +<br/>Machiraju. 2005. A bilinear illumination model for robust face +<br/>recognition. Proceedings of the Tenth IEEE International Conference +<br/>on Computer Vision: October 17-21, 2005, Beijing, China. 1177-1184. +<br/>Los Almamitos, C.A.: IEEE Computer Society. +<br/>Published Version +<br/>doi:10.1109/ICCV.2005.5 +<br/>Citable link +<br/>http://nrs.harvard.edu/urn-3:HUL.InstRepos:4238979 +<br/>Terms of Use +<br/><b></b><br/>repository, and is made available under the terms and conditions +<br/>applicable to Other Posted Material, as set forth at http:// +<br/>nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- +<br/>use#LAA +</td><td>('1780935', 'Baback Moghaddam', 'baback moghaddam')<br/>('1701371', 'Hanspeter Pfister', 'hanspeter pfister')</td><td></td></tr><tr><td>3505c9b0a9631539e34663310aefe9b05ac02727</td><td>A Joint Discriminative Generative Model for Deformable Model +<br/>Construction and Classification +<br/><b>Imperial College London, UK</b><br/><b>Faculty of Electrical Engineering, Mathematics and Computer Science, University of Twente, The</b><br/>Netherlands +</td><td>('2000297', 'Ioannis Marras', 'ioannis marras')<br/>('1793625', 'Symeon Nikitidis', 'symeon nikitidis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>2 Yoti Ltd, London, UK, e-mail: symeon.nikitidis@yoti.com +</td></tr><tr><td>3506518d616343d3083f4fe257a5ee36b376b9e1</td><td>Unsupervised Domain Adaptation for +<br/>Personalized Facial Emotion Recognition +<br/><b>University of Trento</b><br/>Trento, Italy +<br/>FBK +<br/><b>University of Perugia</b><br/>Trento, Italy +<br/>Perugia, Italy +<br/><b>University of Trento</b><br/>Trento, Italy +</td><td>('2933565', 'Gloria Zen', 'gloria zen')<br/>('1716310', 'Enver Sangineto', 'enver sangineto')<br/>('40811261', 'Elisa Ricci', 'elisa ricci')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')</td><td></td></tr><tr><td>353a89c277cca3e3e4e8c6a199ae3442cdad59b5</td><td></td><td></td><td></td></tr><tr><td>35e0256b33212ddad2db548484c595334f15b4da</td><td>Attentive Fashion Grammar Network for +<br/>Fashion Landmark Detection and Clothing Category Classification +<br/><b>Beijing Lab of Intelligent Information Technology, School of Computer Science, Beijing Institute of Technology, China</b><br/><b>University of California, Los Angeles, USA</b></td><td>('2693875', 'Wenguan Wang', 'wenguan wang')<br/>('2762640', 'Yuanlu Xu', 'yuanlu xu')<br/>('34926055', 'Jianbing Shen', 'jianbing shen')<br/>('3133970', 'Song-Chun Zhu', 'song-chun zhu')</td><td></td></tr><tr><td>35e6f6e5f4f780508e5f58e87f9efe2b07d8a864</td><td>This paper is a preprint (IEEE accepted status). IEEE copyright notice. 2018 IEEE. +<br/>Personal use of this material is permitted. Permission from IEEE must be obtained for all +<br/><b>other uses, in any current or future media, including reprinting/republishing this material for</b><br/>advertising or promotional purposes, creating new collective works, for resale or redistribu- +<br/>tion to servers or lists, or reuse of any copyrighted. +<br/>A. Tejero-de-Pablos, Y. Nakashima, T. Sato, N. Yokoya, M. Linna and E. Rahtu, ”Sum- +<br/>marization of User-Generated Sports Video by Using Deep Action Recognition Features,” in +<br/>doi: 10.1109/TMM.2018.2794265 +<br/>keywords: Cameras; Feature extraction; Games; Hidden Markov models; Semantics; +<br/>Three-dimensional displays; 3D convolutional neural networks; Sports video summarization; +<br/>action recognition; deep learning; long short-term memory; user-generated video, +<br/>URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8259321&isnumber=4456689 +</td><td></td><td></td></tr><tr><td>35e87e06cf19908855a16ede8c79a0d3d7687b5c</td><td>Strategies for Multi-View Face Recognition for +<br/>Identification of Human Faces: A Review +<br/>Department of Computer Science +<br/>Mahatma Gandhi Shikshan Mandal’s, +<br/><b>Arts, Science and Commerce College, Chopda</b><br/>Dist: Jalgaon (M.S) +<br/>Dr. R.R.Manza +<br/>Department of Computer Science and IT +<br/><b>Dr. Babasaheb Ambedkar Marathwada University</b><br/>Aurangabad. +</td><td>('21182750', 'Pritesh G. Shah', 'pritesh g. shah')</td><td>pritshah143@gmail.com +<br/>manzaramesh@gmail.com +</td></tr><tr><td>352110778d2cc2e7110f0bf773398812fd905eb1</td><td>TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. X, NO. X, JUNE 2014 +<br/>Matrix Completion for Weakly-supervised +<br/>Multi-label Image Classification +</td><td>('31671904', 'Ricardo Cabral', 'ricardo cabral')<br/>('1683568', 'Fernando De la Torre', 'fernando de la torre')<br/>('2884203', 'Alexandre Bernardino', 'alexandre bernardino')</td><td></td></tr><tr><td>6964af90cf8ac336a2a55800d9c510eccc7ba8e1</td><td>Temporal Relational Reasoning in Videos +<br/>MIT CSAIL +</td><td>('1804424', 'Bolei Zhou', 'bolei zhou')<br/>('50112310', 'Alex Andonian', 'alex andonian')<br/>('1690178', 'Antonio Torralba', 'antonio torralba')</td><td>{bzhou,aandonia,oliva,torralba}@csail.mit.edu +</td></tr><tr><td>697b0b9630213ca08a1ae1d459fabc13325bdcbb</td><td></td><td></td><td></td></tr><tr><td>69ff40fd5ce7c3e6db95a2b63d763edd8db3a102</td><td>HUMAN AGE ESTIMATION VIA GEOMETRIC AND TEXTURAL +<br/>FEATURES +<br/>Merve KILINC1 and Yusuf Sinan AKGUL2 +<br/>1TUBITAK BILGEM UEKAE, Anibal Street, 41470, Gebze, Kocaeli, Turkey +<br/><b>GIT Vision Lab, http://vision.gyte.edu.tr/, Gebze Institute of Technology</b><br/>Kocaeli, Turkey +<br/>Keywords: +<br/>Age estimation:age classification:geometric features:LBP:Gabor:LGBP:cross ratio:FGNET:MORPH +</td><td></td><td>mkilinc@uekae.tubitak.gov.tr1, mkilinc@gyte.edu.tr2, akgul@bilmuh.gyte.edu.tr2 +</td></tr><tr><td>69adbfa7b0b886caac15ebe53b89adce390598a3</td><td>Face hallucination using cascaded +<br/>super-resolution and identity priors +<br/><b>University of Ljubljana, Faculty of Electrical Engineering</b><br/><b>University of Notre Dame</b><br/>Fig. 1. Sample face hallucination results generated with the proposed method. +</td><td>('3387470', 'Klemen Grm', 'klemen grm')<br/>('2613438', 'Walter J. Scheirer', 'walter j. scheirer')</td><td></td></tr><tr><td>69d29012d17cdf0a2e59546ccbbe46fa49afcd68</td><td>Subspace clustering of dimensionality-reduced data +<br/>ETH Zurich, Switzerland +</td><td>('1730683', 'Reinhard Heckel', 'reinhard heckel')<br/>('2208878', 'Michael Tschannen', 'michael tschannen')</td><td>Email: {heckel,boelcskei}@nari.ee.ethz.ch, michaelt@student.ethz.ch +</td></tr><tr><td>69a68f9cf874c69e2232f47808016c2736b90c35</td><td>Learning Deep Representation for Imbalanced Classification +<br/><b>The Chinese University of Hong Kong</b><br/>2SenseTime Group Limited +<br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('2000034', 'Chen Huang', 'chen huang')<br/>('9263285', 'Yining Li', 'yining li')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{chuang,ly015,ccloy,xtang}@ie.cuhk.edu.hk +</td></tr><tr><td>69de532d93ad8099f4d4902c4cad28db958adfea</td><td></td><td></td><td></td></tr><tr><td>69a55c30c085ad1b72dd2789b3f699b2f4d3169f</td><td>International Journal of Computer Trends and Technology (IJCTT) – Volume 34 Number 3 - April 2016 +<br/>Automatic Happiness Strength Analysis of a +<br/>Group of People using Facial Expressions +<br/>Sagiri Prasanthi#1, Maddali M.V.M. Kumar*2, +<br/>#1PG Student, #2Assistant Professor +<br/><b>St. Ann s College of Engineering and Technology, Andhra Pradesh, India</b><br/>is a collective concern +</td><td></td><td></td></tr><tr><td>69b18d62330711bfd7f01a45f97aaec71e9ea6a5</td><td>RESEARCH ARTICLE +<br/>M-Track: A New Software for Automated +<br/>Detection of Grooming Trajectories in Mice +<br/><b>State University of New York Polytechnic Institute, Utica, New York</b><br/><b>United States of America, State University of New York Albany, Albany, New York</b><br/><b>United States of America, State University of New York Albany, Albany</b><br/>New York, United States of America +<br/>☯ These authors contributed equally to this work. +<br/>a11111 +</td><td>('35820210', 'Sheldon L. Reeves', 'sheldon l. reeves')<br/>('8626210', 'Kelsey E. Fleming', 'kelsey e. fleming')<br/>('1708615', 'Lin Zhang', 'lin zhang')<br/>('3976998', 'Annalisa Scimemi', 'annalisa scimemi')</td><td>* scimemia@gmail.com, ascimemi@albany.edu +</td></tr><tr><td>69526cdf6abbfc4bcd39616acde544568326d856</td><td>636 +<br/>[17] B. Moghaddam, T. Jebara, and A. Pentland, “Bayesian face recogni- +<br/>tion,” Pattern Recognit., vol. 33, no. 11, pp. 1771–1782, Nov. 2000. +<br/>[18] A. Nefian, “A hidden Markov model-based approach for face detection +<br/>and recognition,” Ph.D. dissertation, Dept. Elect. Comput. Eng. Elect. +<br/>Eng., Georgia Inst. Technol., Atlanta, 1999. +<br/>[19] P. J. Phillips et al., “Overview of the face recognition grand challenge,” +<br/>presented at the IEEE CVPR, San Diego, CA, Jun. 2005. +<br/>[20] H. T. Tanaka, M. Ikeda, and H. Chiaki, “Curvature-based face surface +<br/>recognition using spherical correlation-principal direction for curved +<br/>object recognition,” in Proc. Int. Conf. Automatic Face and Gesture +<br/>Recognition, 1998, pp. 372–377. +<br/>[21] M. Turk and A. Pentland, “Eigenfaces for recognition,” J. Cognit. Sci., +<br/>pp. 71–86, 1991. +<br/>[22] V. N. Vapnik, Statistical Learning Theory. New York: Wiley, 1998. +<br/>[23] W. Zhao, R. Chellappa, A. Rosenfeld, and P. Phillips, “Face recogni- +<br/>tion: A literature survey,” ACM Comput. Surveys, vol. 35, no. 44, pp. +<br/>399–458, 2003. +<br/>[24] W. Zhao, R. Chellappa, and P. J. Phillips, “Subspace linear discrimi- +<br/>nant analysis for face recognition,” UMD TR4009, 1999. +<br/>Face Verification Using Template Matching +</td><td>('2627097', 'Anil Kumar Sao', 'anil kumar sao')</td><td></td></tr><tr><td>690d669115ad6fabd53e0562de95e35f1078dfbb</td><td>Progressive versus Random Projections for Compressive Capture of Images, +<br/>Lightfields and Higher Dimensional Visual Signals +<br/>MIT Media Lab +<br/>75 Amherst St, Cambridge, MA +<br/>MERL +<br/>201 Broadway, Cambridge MA +<br/>MIT Media Lab +<br/>75 Amherst St, Cambridge, MA +</td><td>('1912905', 'Rohit Pandharkar', 'rohit pandharkar')<br/>('1785066', 'Ashok Veeraraghavan', 'ashok veeraraghavan')<br/>('1717566', 'Ramesh Raskar', 'ramesh raskar')</td><td></td></tr><tr><td>6993bca2b3471f26f2c8a47adfe444bfc7852484</td><td>The Do’s and Don’ts for CNN-based Face Verification +<br/>Carlos Castillo +<br/><b>University of Maryland, College Park</b><br/>UMIACS +</td><td>('2068427', 'Ankan Bansal', 'ankan bansal')<br/>('48467498', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{ankan,carlos,rranjan1,rama}@umiacs.umd.edu +</td></tr><tr><td>69eb6c91788e7c359ddd3500d01fb73433ce2e65</td><td>CAMGRAPH: Distributed Graph Processing for +<br/>Camera Networks +<br/><b>College of Computing</b><br/><b>Georgia Institute of Technology</b><br/>Atlanta, GA, USA +</td><td>('3427189', 'Steffen Maass', 'steffen maass')<br/>('5540701', 'Kirak Hong', 'kirak hong')<br/>('1751741', 'Umakishore Ramachandran', 'umakishore ramachandran')</td><td>steffen.maass@gatech.edu,khong9@cc.gatech.edu,rama@cc.gatech.edu +</td></tr><tr><td>691964c43bfd282f6f4d00b8b0310c554b613e3b</td><td>Temporal Hallucinating for Action Recognition with Few Still Images +<br/>2† +<br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, China</b><br/><b>The Chinese University of Hong Kong 3 SenseTime Group Limited</b></td><td>('46696518', 'Lei Zhou', 'lei zhou')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td></td></tr><tr><td>69063f7e0a60ad6ce16a877bc8f11b59e5f7348e</td><td>Class-Specific Image Deblurring +<br/>2, Fatih Porikli1 +<br/><b>The Australian National University Canberra ACT 2601, Australia</b><br/>2NICTA, Locked Bag 8001, Canberra ACT 2601, Australia +</td><td>('33672969', 'Saeed Anwar', 'saeed anwar')<br/>('1774721', 'Cong Phuoc Huynh', 'cong phuoc huynh')</td><td></td></tr><tr><td>69a9da55bd20ce4b83e1680fbc6be2c976067631</td><td></td><td></td><td></td></tr><tr><td>69c2ac04693d53251500557316c854a625af84ee</td><td>JID: PATREC +<br/>ARTICLE IN PRESS +<br/>Contents lists available at ScienceDirect +<br/>Pattern Recognition Letters +<br/>journal homepage: www.elsevier.com/locate/patrec +<br/>[m5G; April 22, 2016;10:30 ] +<br/>50 years of biometric research: Accomplishments, challenges, +<br/>and opportunities +<br/>a , 1 , +<br/>a +<br/><b>Michigan State University, East Lansing, MI 48824, USA</b><br/>b IBM Research Singapore, 9 Changi Business Park Central 1, 486048 Singapore +<br/>a r t i c l e +<br/>i n f o +<br/>a b s t r a c t +<br/>Article history: +<br/>Received 4 February 2015 +<br/>Available online xxx +<br/>Keywords: +<br/>Biometrics +<br/>Fingerprints +<br/>Face +<br/>Iris +<br/>Security +<br/>Privacy +<br/>Forensics +<br/>Biometric recognition refers to the automated recognition of individuals based on their biological and +<br/>behavioral characteristics such as fingerprint, face, iris, and voice. The first scientific paper on automated +<br/>fingerprint matching was published by Mitchell Trauring in the journal Nature in 1963. The first objec- +<br/>tive of this paper is to document the significant progress that has been achieved in the field of biometric +<br/>recognition in the past 50 years since Trauring’s landmark paper. This progress has enabled current state- +<br/>of-the-art biometric systems to accurately recognize individuals based on biometric trait(s) acquired un- +<br/>der controlled environmental conditions from cooperative users. Despite this progress, a number of chal- +<br/>lenging issues continue to inhibit the full potential of biometrics to automatically recognize humans. The +<br/>second objective of this paper is to enlist such challenges, analyze the solutions proposed to overcome +<br/>them, and highlight the research opportunities in this field. One of the foremost challenges is the de- +<br/>sign of robust algorithms for representing and matching biometric samples obtained from uncooperative +<br/>subjects under unconstrained environmental conditions (e.g., recognizing faces in a crowd). In addition, +<br/>fundamental questions such as the distinctiveness and persistence of biometric traits need greater atten- +<br/>tion. Problems related to the security of biometric data and robustness of the biometric system against +<br/>spoofing and obfuscation attacks, also remain unsolved. Finally, larger system-level issues like usability, +<br/>user privacy concerns, integration with the end application, and return on investment have not been ad- +<br/>equately addressed. Unlocking the full potential of biometrics through inter-disciplinary research in the +<br/>above areas will not only lead to widespread adoption of this promising technology, but will also result +<br/>in wider user acceptance and societal impact. +<br/>© 2016 Published by Elsevier B.V. +<br/>1. Introduction +<br/>“It is the purpose of this article to present, together with some evi- +<br/>dence of its feasibility, a method by which decentralized automatic +<br/>identity verification, such as might be desired for credit, banking +<br/>or security purposes, can be accomplished through automatic com- +<br/>parison of the minutiae in finger-ridge patterns.” +<br/>– Mitchell Trauring, Nature, March 1963 +<br/>In modern society, the ability to reliably identify individu- +<br/>als in real-time is a fundamental requirement in many applica- +<br/>tions including forensics, international border crossing, financial +<br/>transactions, and computer security. Traditionally, an exclusive pos- +<br/> This paper has been recommended for acceptance by S. Sarkar. +<br/>Corresponding author. Tel.: +1 517 355 9282; fax: +1 517 432 1061. +<br/>1 IAPR Fellow. +<br/>http://dx.doi.org/10.1016/j.patrec.2015.12.013 +<br/>0167-8655/© 2016 Published by Elsevier B.V. +<br/>session of a token, such as a passport or an ID card, has been ex- +<br/>tensively used for identifying individuals. In the context of com- +<br/>puter systems and applications, knowledge-based schemes based +<br/>on passwords and PINs are commonly used for person authentica- +<br/>2 Since both token-based and knowledge-based mechanisms +<br/>tion. +<br/>have their own strengths and limitations, the use of two-factor +<br/>authentication schemes that combine both these authentication +<br/>mechanisms are also popular. +<br/>Biometric recognition, or simply biometrics, refers to the auto- +<br/>mated recognition of individuals based on their biological and be- +<br/>havioral characteristics [39] . Examples of biometric traits that have +<br/>been successfully used in practical applications include face, fin- +<br/>gerprint, palmprint, iris, palm/finger vein, and voice. The use of +<br/>DNA, in the context of biometrics (as opposed to just forensics), is +<br/>also beginning to gain traction. Since biometric traits are generally +<br/>inherent to an individual, there is a strong and reasonably +<br/>2 Authentication involves verifying the claimed identity of a person. +<br/>Please cite this article as: A.K. Jain et al., 50 years of biometric research: Accomplishments, challenges, and opportunities, Pattern Recog- +<br/>nition Letters (2016), http://dx.doi.org/10.1016/j.patrec.2015.12.013 +</td><td>('6680444', 'Anil K. Jain', 'anil k. jain')<br/>('34633765', 'Karthik Nandakumar', 'karthik nandakumar')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td>E-mail addresses: jain@cse.msu.edu (A.K. Jain), nkarthik@sg.ibm.com +<br/>(K. Nandakumar), rossarun@cse.msu.edu (A. Ross). +</td></tr><tr><td>6974449ce544dc208b8cc88b606b03d95c8fd368</td><td></td><td></td><td></td></tr><tr><td>69fb98e11df56b5d7ec7d45442af274889e4be52</td><td>Harnessing the Deep Net Object Models for +<br/>enhancing Human Action Recognition +<br/>O.V. Ramana Murthy1 and Roland Goecke1,2 +<br/><b>Vision and Sensing, HCC Lab, ESTeM, University of Canberra</b><br/><b>IHCC, RSCS, CECS, Australian National University</b></td><td></td><td>Email: O.V.RamanaMurthy@ieee.org, roland.goecke@ieee.org +</td></tr><tr><td>3cb2841302af1fb9656f144abc79d4f3d0b27380</td><td>See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/319928941 +<br/>When 3D-Aided 2D Face Recognition Meets Deep +<br/>Learning: An extended UR2D for Pose-Invariant +<br/>Face Recognition +<br/>Article · September 2017 +<br/>CITATIONS +<br/>4 authors: +<br/>READS +<br/>33 +<br/>Xiang Xu +<br/><b>University of Houston</b><br/>Pengfei Dou +<br/><b>University of Houston</b><br/>8 PUBLICATIONS 10 CITATIONS +<br/>9 PUBLICATIONS 29 CITATIONS +<br/>SEE PROFILE +<br/>SEE PROFILE +<br/>Ha Le +<br/><b>University of Houston</b><br/>7 PUBLICATIONS 2 CITATIONS +<br/>Ioannis A Kakadiaris +<br/><b>University of Houston</b><br/>468 PUBLICATIONS 5,233 CITATIONS +<br/>SEE PROFILE +<br/>SEE PROFILE +<br/>Some of the authors of this publication are also working on these related projects: +<br/>3D-Aided 2D Face Recognition View project +<br/>iRay: mobile medical AR View project +<br/>All content following this page was uploaded by Xiang Xu on 27 September 2017. +<br/>The user has requested enhancement of the downloaded file. +</td><td></td><td></td></tr><tr><td>3c78b642289d6a15b0fb8a7010a1fb829beceee2</td><td>Analysis of Facial Dynamics +<br/>Using a Tensor Framework +<br/><b>University of Bristol</b><br/>Department of Computer Science +<br/>Bristol, United Kingdom +<br/><b>University of Bristol</b><br/>Department of Experimental Psychology +<br/>Bristol, United Kingdom +</td><td>('2903159', 'Lisa Gralewski', 'lisa gralewski')<br/>('23725787', 'Edward Morrison', 'edward morrison')<br/>('2022210', 'Ian Penton-Voak', 'ian penton-voak')</td><td>gralewsk@cs.bris.ac.uk +</td></tr><tr><td>3cc3cf57326eceb5f20a02aefae17108e8c8ab57</td><td>BENCHMARK FOR EVALUATING BIOLOGICAL IMAGE ANALYSIS TOOLS +<br/>Center for Bio-Image Informatics, Electrical and Computer Engineering Department, +<br/><b>University of California, Santa Barbara</b><br/>http://www.bioimage.ucsb.edu +<br/>Biological images are critical components for a detailed understanding of the structure and functioning of cells and proteins. +<br/>Image processing and analysis tools increasingly play a significant role in better harvesting this vast amount of data, most of +<br/>which is currently analyzed manually and qualitatively. A number of image analysis tools have been proposed to automatically +<br/>extract the image information. As the studies relying on image analysis tools have become widespread, the validation of +<br/>these methods, in particular, segmentation methods, has become more critical. There have been very few efforts at creating +<br/>benchmark datasets in the context of cell and tissue imaging, while, there have been successful benchmarks in other fields, such +<br/>as the Berkeley segmentation dataset [1], the handwritten digit recognition dataset MNIST [2] and face recognition dataset [3, 4]. +<br/>In the field of biomedical image processing, most of standardized benchmark data sets concentrates on macrobiological images +<br/>such as mammograms and magnet resonance imaging (MRI) images [5], however, there is still a lack of a standardized dataset +<br/>for microbiological structures (e.g. cells and tissues) and it is well known in biomedical imaging [5]. +<br/>We propose a benchmark for biological images to: 1) provide image collections with well defined ground truth; 2) provide +<br/>image analysis tools and evaluation methods to compare and validate analysis tools. We include a representative dataset of +<br/>microbiological structures whose scales range from a subcellular level (nm) to a tissue level (µm), inheriting intrinsic challenges +<br/>in the domain of biomedical image analysis (Fig. 1). The dataset is acquired through two of the main microscopic imaging +<br/>techniques: transmitted light microscopy and confocal laser scanning microscopy. The analysis tools1in the benchmark are +<br/>designed to obtain different quantitative measures from the dataset including microtubule tracing, cell segmentation, and retinal +<br/>layer segmentation. +<br/>Fig. 1. Example dataset provided in the benchmark. +<br/>This research is supported by NSF ITR-0331697. +<br/>1All analysis tools mentioned in this work can be found at http://www.bioimage.ucsb.edu/publications/. +<br/>ScaleConfocal microscopyLight microscopymicrotubulehorizontal cellSubcellular(< 1 µm)photoreceptorsbreast cancer cellsCOS1 cellsCellularTissue(< 10 µm)(< 30 µm)(< 350 µm)(≈10-50 µm in width)retinal layers </td><td>('8451780', 'Elisa Drelie Gelasca', 'elisa drelie gelasca')<br/>('3045933', 'Jiyun Byun', 'jiyun byun')<br/>('3064236', 'Boguslaw Obara', 'boguslaw obara')</td><td></td></tr><tr><td>3cb488a3b71f221a8616716a1fc2b951dd0de549</td><td>Facial Age Estimation by +<br/>Adaptive Label Distribution Learning +<br/>School of Computer Science and Engineering +<br/>Key Lab of Computer Network and Information Integration, Ministry of Education +<br/><b>Southeast University, Nanjing 211189, China</b></td><td>('1735299', 'Xin Geng', 'xin geng')<br/>('1794816', 'Qin Wang', 'qin wang')<br/>('40228279', 'Yu Xia', 'yu xia')</td><td>Email: {xgeng, qinwang, xiayu}@seu.edu.cn +</td></tr><tr><td>3cfbe1f100619a932ba7e2f068cd4c41505c9f58</td><td>A Realistic Simulation Tool for Testing Face Recognition +<br/>Systems under Real-World Conditions∗ +<br/>M. Correa, J. Ruiz-del-Solar, S. Parra-Tsunekawa, R. Verschae +<br/>Department of Electrical Engineering, Universidad de Chile +<br/>Advanced Mining Technology Center, Universidad de Chile +</td><td></td><td></td></tr><tr><td>3c563542db664321aa77a9567c1601f425500f94</td><td>TV-GAN: Generative Adversarial Network Based Thermal to Visible Face +<br/>Recognition +<br/><b>The University of Queensland, School of ITEE, QLD 4072, Australia</b></td><td>('50615828', 'Teng Zhang', 'teng zhang')<br/>('2331880', 'Arnold Wiliem', 'arnold wiliem')<br/>('1973322', 'Siqi Yang', 'siqi yang')<br/>('2270092', 'Brian C. Lovell', 'brian c. lovell')</td><td>[patrick.zhang, a.williem, siqi.yang]@uq.edu.au, lovell@itee.uq.edu.au +</td></tr><tr><td>3c03d95084ccbe7bf44b6d54151625c68f6e74d0</td><td></td><td></td><td></td></tr><tr><td>3cd7b15f5647e650db66fbe2ce1852e00c05b2e4</td><td></td><td></td><td></td></tr><tr><td>3c6cac7ecf546556d7c6050f7b693a99cc8a57b3</td><td>Robust Facial Landmark Detection in the Wild +<br/>Submitted for the Degree of +<br/>Doctor of Philosophy +<br/>from the +<br/><b>University of Surrey</b><br/>Centre for Vision, Speech and Signal Processing +<br/>Faculty of Engineering and Physical Sciences +<br/><b>University of Surrey</b><br/>Guildford, Surrey GU2 7XH, U.K. +<br/>January 2016 +</td><td>('37705062', 'Zhenhua Feng', 'zhenhua feng')<br/>('37705062', 'Zhenhua Feng', 'zhenhua feng')</td><td></td></tr><tr><td>3c57e28a4eb463d532ea2b0b1ba4b426ead8d9a0</td><td>Defeating Image Obfuscation with Deep Learning +<br/><b>The University of Texas at</b><br/>Austin +<br/>Cornell Tech +<br/>Cornell Tech +</td><td>('34861228', 'Richard McPherson', 'richard mcpherson')<br/>('2520493', 'Reza Shokri', 'reza shokri')<br/>('1723945', 'Vitaly Shmatikov', 'vitaly shmatikov')</td><td>richard@cs.utexas.edu +<br/>shokri@cornell.edu +<br/>shmat@cs.cornell.edu +</td></tr><tr><td>3cd9b0a61bdfa1bb8a0a1bf0369515a76ecd06e3</td><td>Submitted 2/11; Revised 10/11; Published ??/11 +<br/>Distance Metric Learning with Eigenvalue Optimization +<br/><b>College of Engineering, Mathematics and Physical Sciences</b><br/><b>University of Exeter</b><br/>Harrison Building, North Park Road +<br/>Exeter, EX4 4QF, UK +<br/>Department of Engineering Mathematics +<br/><b>University of Bristol</b><br/>Merchant Venturers Building, Woodland Road +<br/>Bristol, BS8 1UB, UK +<br/>Editor: +</td><td>('38954213', 'Yiming Ying', 'yiming ying')<br/>('1695363', 'Peng Li', 'peng li')</td><td>y.ying@exeter.ac.uk +<br/>lipeng@ieee.org +</td></tr><tr><td>3c97c32ff575989ef2869f86d89c63005fc11ba9</td><td>Face Detection with the Faster R-CNN +<br/>Erik Learned-Miller +<br/><b>University of Massachusetts Amherst</b><br/><b>University of Massachusetts Amherst</b><br/>Amherst MA 01003 +<br/>Amherst MA 01003 +</td><td>('40175280', 'Huaizu Jiang', 'huaizu jiang')</td><td>hzjiang@cs.umass.edu +<br/>elm@cs.umass.edu +</td></tr><tr><td>3ce2ecf3d6ace8d80303daf67345be6ec33b3a93</td><td></td><td></td><td></td></tr><tr><td>3c1aef7c2d32a219bdbc89a44d158bc2695e360a</td><td>Adversarial Attack Type I: Generating False Positives +<br/><b>Shanghai Jiao Tong University</b><br/>Shanghai, P.R. China 200240 +<br/><b>Shanghai Jiao Tong University</b><br/>Shanghai, P.R. China 200240 +<br/><b>Shanghai Jiao Tong University</b><br/>Shanghai, P.R. China 200240 +<br/><b>Shanghai Jiao Tong University</b><br/>Shanghai, P.R. China 200240 +</td><td>('51428687', 'Sanli Tang', 'sanli tang')<br/>('13858459', 'Mingjian Chen', 'mingjian chen')<br/>('2182657', 'Xiaolin Huang', 'xiaolin huang')<br/>('1688428', 'Jie Yang', 'jie yang')</td><td>tangsanli@sjtu.edu.cn +<br/>w179261466@sjtu.edu.cn +<br/>xiaolinhuang@sjtu.edu.cn +<br/>jieyang@sjtu.edu.cn +</td></tr><tr><td>3c374cb8e730b64dacb9fbf6eb67f5987c7de3c8</td><td>Measuring Gaze Orientation for Human-Robot +<br/>Interaction +<br/>∗ CNRS; LAAS; 7 avenue du Colonel Roche, 31077 Toulouse Cedex, France +<br/>† Universit´e de Toulouse; UPS; LAAS-CNRS : F-31077 Toulouse, France +<br/>Introduction +<br/>In the context of Human-Robot interaction estimating gaze orientation brings +<br/>useful information about human focus of attention. This is a contextual infor- +<br/>mation : when you point something you usually look at it. Estimating gaze +<br/>orientation requires head pose estimation. There are several techniques to esti- +<br/>mate head pose from images, they are mainly based on training [3, 4] or on local +<br/>face features tracking [6]. The approach described here is based on local face +<br/>features tracking in image space using online learning, it is a mixed approach +<br/>since we track face features using some learning at feature level. It uses SURF +<br/>features [2] to guide detection and tracking. Such key features can be matched +<br/>between images, used for object detection or object tracking [10]. Several ap- +<br/>proaches work on fixed size images like training techniques which mainly work +<br/>on low resolution images because of computation costs whereas approaches based +<br/>on local features tracking work on high resolution images. Tracking face features +<br/>such as eyes, nose and mouth is a common problem in many applications such as +<br/>detection of facial expression or video conferencing [8] but most of those appli- +<br/>cations focus on front face images [9]. We developed an algorithm based on face +<br/>features tracking using a parametric model. First we need face detection, then +<br/>we detect face features in following order: eyes, mouth, nose. In order to achieve +<br/>full profile detection we use sets of SURF to learn what eyes, mouth and nose +<br/>look like once tracking is initialized. Once those sets of SURF are known they +<br/>are used to detect and track face features. SURF have a descriptor which is often +<br/>used to identify a key point and here we add some global geometry information +<br/>by using the relative position between key points. Then we use a particle filter to +<br/>track face features using those SURF based detectors, we compute the head pose +<br/>angles from features position and pass the results through a median filter. This +<br/>paper is organized as follows. Section 2 describes our modeling of visual features, +<br/>section 3 presents our tracking implementation. Section 4 presents results we get +<br/>with our implementation and future works in section 5. +<br/>2 Visual features +<br/>We use some basic properties of facial features to initialize our algorithm : eyes +<br/>are dark and circular, mouth is an horizontal dark line with a specific color,... +</td><td>('5253126', 'R. Brochard', 'r. brochard')<br/>('2667229', 'B. Burger', 'b. burger')<br/>('2325221', 'A. Herbulot', 'a. herbulot')<br/>('1797260', 'F. Lerasle', 'f. lerasle')</td><td></td></tr><tr><td>3c0bbfe664fb083644301c67c04a7f1331d9515f</td><td>The Role of Color and Contrast in Facial Age Estimation +<br/>Paper ID: 7 +<br/><b>No Institute Given</b></td><td></td><td></td></tr><tr><td>3c4f6d24b55b1fd3c5b85c70308d544faef3f69a</td><td>A Hybrid Deep Learning Architecture for +<br/>Privacy-Preserving Mobile Analytics +<br/><b>cid:63)Sharif University of Technology, University College London, Queen Mary University of London</b></td><td>('8201306', 'Seyed Ali Ossia', 'seyed ali ossia')<br/>('9920557', 'Ali Shahin Shamsabadi', 'ali shahin shamsabadi')<br/>('2251846', 'Ali Taheri', 'ali taheri')<br/>('1688652', 'Hamid R. Rabiee', 'hamid r. rabiee')<br/>('1763096', 'Hamed Haddadi', 'hamed haddadi')</td><td></td></tr><tr><td>3cb0ef5aabc7eb4dd8d32a129cb12b3081ef264f</td><td>Absolute Head Pose Estimation From Overhead Wide-Angle Cameras +<br/><b>IBM T.J. Watson Research Center</b><br/>19 Skyline Drive, Hawthorne, NY 10532 USA +</td><td>('40383812', 'Ying-li Tian', 'ying-li tian')<br/>('1690709', 'Arun Hampapur', 'arun hampapur')</td><td>{ yltian,lisabr,jconnell,sharat,arunh,aws,bolle }@us.ibm.com +</td></tr><tr><td>3cb64217ca2127445270000141cfa2959c84d9e7</td><td></td><td></td><td></td></tr><tr><td>3c11a1f2bd4b9ce70f699fb6ad6398171a8ad3bd</td><td>International Journal of Computer Information Systems and Industrial Management Applications (IJCISIM) +<br/>ISSN: 2150-7988 Vol.2 (2010), pp.262-278 +<br/>http://www.mirlabs.org/ijcisim +<br/>Simulating Pareidolia of Faces for Architectural Image Analysis +<br/>Newcastle Robotics Laboratory +<br/>School of Electrical Engineering and Computer Science +<br/><b>The University of Newcastle, Callaghan 2308, Australia</b><br/>School of Architecture and Built Environment +<br/><b>The University of Newcastle</b><br/>Callaghan 2308, Australia +</td><td>('1716539', 'Stephan K. Chalup', 'stephan k. chalup')<br/>('40211094', 'Michael J. Ostwald', 'michael j. ostwald')</td><td>Stephan.Chalup@newcastle.edu.au, Kenny.Hong@uon.edu.au +<br/>Michael.Ostwald@newcastle.edu.au +</td></tr><tr><td>3cd8ab6bb4b038454861a36d5396f4787a21cc68</td><td> Video‐Based Facial Expression Recognition Using Hough Forest +<br/><b>National Tsing Hua University, Hsin-Chu, Taiwan</b><br/><b>Asian University, Taichung, Taiwan</b></td><td>('2790846', 'Shih-Chung Hsu', 'shih-chung hsu')<br/>('1793389', 'Chung-Lin Huang', 'chung-lin huang')</td><td>E-mail: d9761817@oz.nthu.edu.tw, clhuang@asia.edu.tw +</td></tr><tr><td>3cd5da596060819e2b156e8b3a28331ef633036b</td><td></td><td></td><td></td></tr><tr><td>3ca5d3b8f5f071148cb50f22955fd8c1c1992719</td><td>EVALUATING RACE AND SEX DIVERSITY IN THE WORLD’S LARGEST +<br/>COMPANIES USING DEEP NEURAL NETWORKS +<br/>1 Youth Laboratories, Ltd, Diversity AI Group, Skolkovo Innovation Center, Nobel Street 5, +<br/>143026, Moscow, Russia +<br/>2 Insilico Medicine, Emerging Technology Centers, JHU, 1101 33rd Street, Baltimore, MD, +<br/>21218, USA +<br/><b>University of Oxford, Oxford, United Kingdom</b><br/><b>Computer Engineering and Computer Science, Duthie Center for Engineering, University of</b><br/>Louisville, Louisville, KY 40292, USA +<br/>5 Computer Vision Lab, Department of Information Technology and Electrical Engineering, ETH +<br/>Zürich, Switzerland +<br/><b>Center for Healthy Aging, University of</b><br/>Copenhagen, Denmark +<br/>7 The Biogerontology Research Foundation, 2354 Chynoweth House, Trevissome Park, Truro, +<br/>TR4 8UN, UK. +<br/><b>Moscow Institute of Physics and Technology, Institutskiy per., 9, Dolgoprudny, 141701, Russia</b></td><td>('3888942', 'Konstantin Chekanov', 'konstantin chekanov')<br/>('4017984', 'Polina Mamoshina', 'polina mamoshina')<br/>('1976753', 'Roman V. Yampolskiy', 'roman v. yampolskiy')<br/>('1732855', 'Radu Timofte', 'radu timofte')<br/>('40336662', 'Alex Zhavoronkov', 'alex zhavoronkov')</td><td>Morten Scheibye-Knudsen: mscheibye@sund.ku.dk +<br/>Alex Zhavoronkov: alex@biogerontology.org +</td></tr><tr><td>3c56acaa819f4e2263638b67cea1ec37a226691d</td><td>Body Joint guided 3D Deep Convolutional +<br/>Descriptors for Action Recognition +</td><td>('3201156', 'Congqi Cao', 'congqi cao')<br/>('46867228', 'Yifan Zhang', 'yifan zhang')<br/>('1713887', 'Chunjie Zhang', 'chunjie zhang')<br/>('1694235', 'Hanqing Lu', 'hanqing lu')</td><td></td></tr><tr><td>3cc46bf79fb9225cf308815c7d41c8dd5625cc29</td><td>AGE INTERVAL AND GENDER PREDICTION USING PARAFAC2 APPLIED TO SPEECH +<br/>UTTERANCES +<br/><b>Aristotle University of Thessaloniki</b><br/>Thessaloniki 54124, GREECE +<br/><b>Cyprus University of Technology</b><br/>3040 Limassol, Cyprus +</td><td>('3352401', 'Evangelia Pantraki', 'evangelia pantraki')<br/>('1736143', 'Constantine Kotropoulos', 'constantine kotropoulos')<br/>('1830709', 'Andreas Lanitis', 'andreas lanitis')</td><td>{pantraki@|costas@aiia}.csd.auth.gr +<br/>andreas.lanitis@cut.ac.cy +</td></tr><tr><td>3c8da376576938160cbed956ece838682fa50e9f</td><td>Chapter 4 +<br/>Aiding Face Recognition with +<br/>Social Context Association Rule +<br/>based Re-Ranking +<br/>Humans are very efficient at recognizing familiar face images even in challenging condi- +<br/>tions. One reason for such capabilities is the ability to understand social context between +<br/>individuals. Sometimes the identity of the person in a photo can be inferred based on the +<br/>identity of other persons in the same photo, when some social context between them is +<br/>known. This chapter presents an algorithm to utilize the co-occurrence of individuals as +<br/>the social context to improve face recognition. Association rule mining is utilized to infer +<br/>multi-level social context among subjects from a large repository of social transactions. +<br/>The results are demonstrated on the G-album and on the SN-collection pertaining to 4675 +<br/>identities prepared by the authors from a social networking website. The results show that +<br/>association rules extracted from social context can be used to augment face recognition and +<br/>improve the identification performance. +<br/>4.1 +<br/>Introduction +<br/>Face recognition capabilities of humans have inspired several researchers to understand +<br/>the science behind it and use it in developing automated algorithms. Recently, it is also +<br/>argued that encoding social context among individuals can be leveraged for improved +<br/>automatic face recognition [175]. As shown in Figure 4.1, often times a person’s identity +<br/>can be inferred based on the identity of other persons in the same photo, when some social +<br/>context between them is known. A subject’s face in consumer photos generally co-occur +<br/>along with their socially relevant people. With the advent of social networking services, +<br/>the social context between individuals is readily available. Face recognition performance +<br/>105 +</td><td></td><td></td></tr><tr><td>56e4dead93a63490e6c8402a3c7adc493c230da5</td><td>World Journal of Computer Application and Technology 1(2): 41-50, 2013 +<br/>DOI: 10.13189/wjcat.2013.010204 +<br/> http://www.hrpub.org +<br/>Face Recognition Techniques: A Survey +<br/>V.Vijayakumari +<br/><b>Sri krishna College of Technology, Coimbatore, India</b><br/>Copyright © 2013 Horizon Research Publishing All rights reserved. +</td><td></td><td>*Corresponding Author: ebinviji@rediffmail.com +</td></tr><tr><td>56e885b9094391f7d55023a71a09822b38b26447</td><td>FREQUENCY DECODED LOCAL BINARY PATTERN +<br/>Face Retrieval using Frequency Decoded Local +<br/>Descriptor +</td><td>('34992579', 'Shiv Ram Dubey', 'shiv ram dubey')</td><td></td></tr><tr><td>56c700693b63e3da3b985777da6d9256e2e0dc21</td><td>Global Refinement of Random Forest +<br/><b>University of Science and Technology of China</b><br/>Microsoft Research +</td><td>('3080683', 'Shaoqing Ren', 'shaoqing ren')<br/>('2032273', 'Xudong Cao', 'xudong cao')<br/>('1732264', 'Yichen Wei', 'yichen wei')<br/>('40055995', 'Jian Sun', 'jian sun')</td><td>sqren@mail.ustc.edu.cn +<br/>{xudongca,yichenw,jiansun}@microsoft.com +</td></tr><tr><td>56359d2b4508cc267d185c1d6d310a1c4c2cc8c2</td><td>Shape Driven Kernel Adaptation in +<br/>Convolutional Neural Network for Robust Facial Trait Recognition +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>National Laboratory of Pattern Recognition, Institute of Automation, CAS, Beijing, 100190, China</b><br/><b>National University of Singapore, Singapore</b></td><td>('1688086', 'Shaoxin Li', 'shaoxin li')<br/>('1757173', 'Junliang Xing', 'junliang xing')<br/>('1773437', 'Zhiheng Niu', 'zhiheng niu')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>56e079f4eb40744728fd1d7665938b06426338e5</td><td>Bayesian Approaches to Distribution Regression +<br/><b>University of Oxford</b><br/><b>University College London</b><br/><b>University of Oxford</b><br/><b>Imperial College London</b></td><td>('35142231', 'Ho Chung Leon Law', 'ho chung leon law')<br/>('36326783', 'Dougal J. Sutherland', 'dougal j. sutherland')<br/>('1698032', 'Dino Sejdinovic', 'dino sejdinovic')<br/>('2127497', 'Seth Flaxman', 'seth flaxman')</td><td>ho.law@spc.ox.ac.uk +<br/>dougal@gmail.com +<br/>dino.sejdinovic@stats.ox.ac.uk +<br/>s.flaxman@imperial.ac.uk +</td></tr><tr><td>56e6f472090030a6f172a3e2f46ef9daf6cad757</td><td>Asian Face Image Database PF01 +<br/>Intelligent Multimedia Lab. +<br/>†Department of Computer Science and Engineering +<br/><b>Pohang University of Science and Technology</b><br/>San 31, Hyoja-Dong, Nam-Gu, Pohang, 790-784, Korea +</td><td></td><td></td></tr><tr><td>56a653fea5c2a7e45246613049fb16b1d204fc96</td><td>3287 +<br/>Quaternion Collaborative and Sparse Representation +<br/>With Application to Color Face Recognition +<br/>representation-based +</td><td>('2888882', 'Cuiming Zou', 'cuiming zou')<br/>('3369665', 'Kit Ian Kou', 'kit ian kou')<br/>('3154834', 'Yulong Wang', 'yulong wang')</td><td></td></tr><tr><td>56f86bef26209c85f2ef66ec23b6803d12ca6cd6</td><td>Pyramidal RoR for Image Classification +<br/><b>North China Electric Power University, Baoding, China</b></td><td>('32164792', 'Ke Zhang', 'ke zhang')<br/>('3451321', 'Liru Guo', 'liru guo')<br/>('35038034', 'Ce Gao', 'ce gao')<br/>('2626320', 'Zhenbing Zhao', 'zhenbing zhao')</td><td>Eail：zhangke41616@126.com +</td></tr><tr><td>5666ed763698295e41564efda627767ee55cc943</td><td>Manuscript +<br/>Click here to download Manuscript: template.tex +<br/>Click here to view linked References +<br/>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Relatively-Paired Space Analysis: Learning a Latent Common +<br/>Space from Relatively-Paired Observations +<br/>Received: date / Accepted: date +</td><td>('1874900', 'Zhanghui Kuang', 'zhanghui kuang')</td><td></td></tr><tr><td>566a39d753c494f57b4464d6bde61bf3593f7ceb</td><td>A Critical Review of Action Recognition Benchmarks +<br/><b>The Open University of Israel</b></td><td>('1756099', 'Tal Hassner', 'tal hassner')</td><td>hassner@openu.ac.il +</td></tr><tr><td>56c2fb2438f32529aec604e6fc3b06a595ddbfcc</td><td>MAICS 2016 +<br/>pp. 97–102 +<br/>Comparison of Recent Machine Learning Techniques for Gender Recognition +<br/>from Facial Images +<br/>Computer Science Department +<br/><b>Central Washington University</b><br/>Ellensburg, WA, USA +<br/>Computer Science Department +<br/><b>Central Washington University</b><br/>Ellensburg, WA, USA +<br/>R˘azvan Andonie +<br/>Computer Science Department +<br/><b>Central Washington University</b><br/>Computer Science Department +<br/><b>Central Washington University</b><br/>Ellensburg, WA, USA +<br/>Ellensburg, WA, USA +<br/>and +<br/>Electronics and Computers Department +<br/><b>Transilvania University</b><br/>Bras¸ov, Romania +</td><td>('9770023', 'Joseph Lemley', 'joseph lemley')<br/>('9770023', 'Joseph Lemley', 'joseph lemley')<br/>('40470929', 'Sami Abdul-Wahid', 'sami abdul-wahid')<br/>('35877118', 'Dipayan Banik', 'dipayan banik')</td><td></td></tr><tr><td>56f231fc40424ed9a7c93cbc9f5a99d022e1d242</td><td>Age Estimation Based on A Single Network with +<br/>Soft Softmax of Aging Modeling +<br/>1Center for Biometrics and Security Research & National Laboratory of Pattern +<br/><b>Recognition, Institute of Automation, Chinese Academy of Sciences, Beijing, China</b><br/><b>University of Chinese Academy of Sciences</b><br/>3Faculty of Information Technology, +<br/><b>Macau University of Science and Technology, Macau</b></td><td>('9645431', 'Zichang Tan', 'zichang tan')<br/>('2950852', 'Shuai Zhou', 'shuai zhou')<br/>('1756538', 'Jun Wan', 'jun wan')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td></td></tr><tr><td>5615d6045301ecbc5be35e46cab711f676aadf3a</td><td>Discriminatively Learned Hierarchical Rank Pooling Networks +<br/>Received: date / Accepted: date +</td><td>('1688071', 'Basura Fernando', 'basura fernando')</td><td></td></tr><tr><td>561ae67de137e75e9642ab3512d3749b34484310</td><td>December 2017 +<br/>DeepGestalt - Identifying Rare Genetic Syndromes +<br/>Using Deep Learning +<br/>1FDNA Inc., Boston, Massachusetts, USA +<br/><b>Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel</b><br/><b>Recanati Genetic Institute, Rabin Medical Center and Schneider Children s Medical Center, Petah Tikva, Israel</b><br/><b>Institute for Genomic Statistic and Bioinformatics, University Hospital Bonn</b><br/><b>Rheinische-Friedrich-Wilhelms University, Bonn, Germany</b><br/><b>Institute of Human Genetics, University Hospital Magdeburg, Magdeburg, Germany</b><br/><b>University of California, San Diego, California, USA</b><br/>7Division of Genetics/Dysmorphology, Rady Children’s Hospital San Diego, San Diego, California, USA +<br/>8Division of Medical Genetics, A. I. du Pont Hospital for Children/Nemours, Wilmington, Delaware,USA +<br/>Boston 186 South St. 5th Floor, Boston, MA 02111 U.S.A., Tel: +1 (617) 412-7000 +<br/>Conflict of interest: YG, YH, OB, NF, DG are employees of FDNA; LBS is an advisor of FDNA; +<br/>LBS, PK, LMB, KWG are members of the scientific advisory board of FDNA +</td><td>('2916582', 'Yaron Gurovich', 'yaron gurovich')<br/>('1917486', 'Yair Hanani', 'yair hanani')<br/>('40142952', 'Omri Bar', 'omri bar')<br/>('40443403', 'Nicole Fleischer', 'nicole fleischer')<br/>('35487552', 'Dekel Gelbman', 'dekel gelbman')<br/>('20717247', 'Lina Basel-Salmon', 'lina basel-salmon')<br/>('4346029', 'Martin Zenker', 'martin zenker')<br/>('6335877', 'Lynne M. Bird', 'lynne m. bird')<br/>('5404116', 'Karen W. Gripp', 'karen w. gripp')</td><td></td></tr><tr><td>568cff415e7e1bebd4769c4a628b90db293c1717</td><td>Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence (AAAI-16) +<br/>Concepts Not Alone: Exploring Pairwise Relationships +<br/>for Zero-Shot Video Activity Recognition +<br/><b>IIIS, Tsinghua University, Beijing, China</b><br/><b>QCIS, University of Technology Sydney, Sydney, Australia</b><br/><b>DCMandB, University of Michigan, Ann Arbor, USA 4 SCS, Carnegie Mellon University, Pittsburgh, USA</b></td><td>('2551285', 'Chuang Gan', 'chuang gan')<br/>('2735055', 'Ming Lin', 'ming lin')<br/>('39033919', 'Yi Yang', 'yi yang')<br/>('1732213', 'Gerard de Melo', 'gerard de melo')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td></td></tr><tr><td>560e0e58d0059259ddf86fcec1fa7975dee6a868</td><td>Face Recognition in Unconstrained Videos with Matched Background Similarity +<br/><b>The Blavatnik School of Computer Science, Tel-Aviv University, Israel</b><br/><b>Computer Science Division, The Open University of Israel</b></td><td>('1776343', 'Lior Wolf', 'lior wolf')<br/>('3352629', 'Itay Maoz', 'itay maoz')</td><td></td></tr><tr><td>56a677c889e0e2c9f68ab8ca42a7e63acf986229</td><td>Mining Spatial and Spatio-Temporal ROIs for Action Recognition +<br/>Jiang Wang2 Alan Yuille1,3 +<br/><b>University of California, Los Angeles</b><br/><b>Baidu Research, USA 3John Hopkins University</b></td><td>('5964529', 'Xiaochen Lian', 'xiaochen lian')</td><td>{lianxiaochen@,yuille@stat.}ucla.edu +<br/>{chenzhuoyuan,yangyi05,wangjiang03}@baidu.com +</td></tr><tr><td>566038a3c2867894a08125efe41ef0a40824a090</td><td>978-1-4244-2354-5/09/$25.00 ©2009 IEEE +<br/>1945 +<br/>ICASSP 2009 +</td><td></td><td></td></tr><tr><td>56dca23481de9119aa21f9044efd7db09f618704</td><td>Riemannian Dictionary Learning and Sparse +<br/>Coding for Positive Definite Matrices +</td><td>('2691929', 'Anoop Cherian', 'anoop cherian')<br/>('3072326', 'Suvrit Sra', 'suvrit sra')</td><td></td></tr><tr><td>56ae6d94fc6097ec4ca861f0daa87941d1c10b70</td><td>Distance Estimation of an Unknown Person +<br/>from a Portrait +<br/>1 Technicolor - Cesson S´evign´e, France +<br/><b>California Institute of Technology, Pasadena, CA, USA</b></td><td>('2232848', 'Xavier P. Burgos-Artizzu', 'xavier p. burgos-artizzu')<br/>('3339867', 'Matteo Ruggero Ronchi', 'matteo ruggero ronchi')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>xavier.burgos@technicolor.com, {mronchi,perona}@caltech.edu +</td></tr><tr><td>56f812661c3248ed28859d3b2b39e033b04ae6ae</td><td>Multiple Feature Fusion by Subspace Learning +<br/><b>Beckman Institute</b><br/><b>University of Illinois at</b><br/>Urbana-Champaign +<br/>Urbana, IL 61801, USA +<br/>Durham, NC 27707, USA +<br/>Computer Science +<br/>North Carolina Central +<br/><b>University</b><br/><b>Beckman Institute</b><br/><b>University of Illinois at</b><br/>Urbana-Champaign +<br/>Urbana, IL 61801, USA +</td><td>('1708679', 'Yun Fu', 'yun fu')<br/>('37575012', 'Liangliang Cao', 'liangliang cao')<br/>('1822413', 'Guodong Guo', 'guodong guo')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>{yunfu2,cao4}@uiuc.edu +<br/>gdguo@nccu.edu +<br/>huang@ifp.uiuc.edu +</td></tr><tr><td>516a27d5dd06622f872f5ef334313350745eadc3</td><td>> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < +<br/>1 +<br/>Fine-Grained Facial Expression Analysis Us- +<br/>ing Dimensional Emotion Model +<br/> +</td><td>('41179750', 'Feng Zhou', 'feng zhou')<br/>('34362536', 'Shu Kong', 'shu kong')<br/>('3157443', 'Charless C. Fowlkes', 'charless c. fowlkes')<br/>('29889388', 'Tao Chen', 'tao chen')<br/>('40216538', 'Baiying Lei', 'baiying lei')</td><td></td></tr><tr><td>512befa10b9b704c9368c2fbffe0dc3efb1ba1bf</td><td>Evidence and a Computational Explanation of Cultural Differences in +<br/>Facial Expression Recognition +<br/>Matthew N. Dailey +<br/>Computer Science and Information Management +<br/><b>Asian Institute of Technology, Pathumthani, Thailand</b><br/>Computer Science and Engineering +<br/><b>University of California, San Diego, USA</b><br/>Michael J. Lyons +<br/><b>College of Image Arts and Sciences</b><br/><b>Ritsumeikan University, Kyoto, Japan</b><br/>Faculty of Informatics +<br/><b>Kogakuin University, Tokyo, Japan</b><br/>Department of Design and Computer Applications +<br/><b>Sendai National College of Technology, Natori, Japan</b><br/>Department of Psychology +<br/><b>Tohoku University, Sendai, Japan</b><br/>Garrison W. Cottrell +<br/>Computer Science and Engineering +<br/><b>University of California, San Diego, USA</b><br/>Facial expressions are crucial to human social communication, but the extent to which they are +<br/>innate and universal versus learned and culture dependent is a subject of debate. Two studies +<br/>explored the effect of culture and learning on facial expression understanding. In Experiment +<br/>better than the other at classifying facial expressions posed by members of the same culture. +<br/>In Experiment 2, this reciprocal in-group advantage was reproduced by a neurocomputational +<br/>model trained in either a Japanese cultural context or an American cultural context. The model +<br/>demonstrates how each of us, interacting with others in a particular cultural context, learns to +<br/>recognize a culture-specific facial expression dialect. +<br/>The scientific literature on innate versus culture-specific +<br/>years ago, Darwin (1872/1998) argued for innate production +<br/>of facial expressions based on cross-cultural comparisons. +<br/>Landis (1924), however, found little agreement between par- +<br/>ticipants. Woodworth (1938) and Schlosberg (1952) found +<br/>structure in the disagreement in interpretation, proposing a +<br/>low-dimensional similarity space characterizing affective fa- +<br/>cial expressions. +<br/>Starting in the 1960’s, researchers found more support for +<br/>facial expressions as innate, universal indicators of particular +<br/>sions (Tomkins, 1962–1963; Tomkins & McCarter, 1964). +<br/>Ekman and colleagues found cross-cultural consistency in +<br/>pressions in both literate and preliterate cultures (Ekman, +<br/>1972; Ekman, Friesen, O’Sullivan, et al., 1987; Ekman, +<br/>Sorensen, & Friesen, 1969). +<br/>Today, researchers disagree on the precise degree to which +<br/>sal versus culture-specific (Ekman, 1994, 1999b; Fridlund, +<br/>1994; Izard, 1994; Russell, 1994, 1995), but there appears +<br/>to be consensus that universal factors interact to some extent +<br/>with culture-specific learning to produce differences between +<br/>cultures. A number of modern theories (Ekman, 1999a; Rus- +<br/>sell & Bullock, 1986; Scherer, 1992; Russell, 1994) attempt +<br/>to account for these universals and culture-specific varia- +<br/>tions. +<br/>Cultural differences in facial expression interpre- +<br/>tation +<br/>The early cross-cultural studies on facial expression +<br/>recognition focused mainly on the question of universality +<br/>sought to analyze and interpret the cultural differences that +<br/>came up in those studies. However, a steadily increasing +<br/>number of studies have focused on the factors underlying +<br/>cultural differences. These studies either compare the fa- +<br/>cial expression judgments made by participants from differ- +<br/>ent cultures or attempt to find the relevant dimensions of +<br/>culture predicting observed cultural differences. Much of +<br/>the research was framed by Ekman’s “neuro-cultural” theory +<br/>elicitors, display rules, and/or consequences due to culture- +<br/>specific learning. +<br/>Ekman (1972) and Friesen (1972) proposed display rules +</td><td>('33597747', 'Carrie Joyce', 'carrie joyce')<br/>('40533190', 'Miyuki Kamachi', 'miyuki kamachi')<br/>('12030857', 'Hanae Ishi', 'hanae ishi')<br/>('8365437', 'Jiro Gyoba', 'jiro gyoba')</td><td></td></tr><tr><td>51c3050fb509ca685de3d9ac2e965f0de1fb21cc</td><td>Fantope Regularization in Metric Learning +<br/>Marc T. Law +<br/>Sorbonne Universit´es, UPMC Univ Paris 06, UMR 7606, LIP6, F-75005, Paris, France +</td><td>('1728523', 'Nicolas Thome', 'nicolas thome')<br/>('1702233', 'Matthieu Cord', 'matthieu cord')</td><td></td></tr><tr><td>516d0d9eb08825809e4618ca73a0697137ebabd5</td><td>Regularizing Long Short Term +<br/>Memory with 3D Human-Skeleton +<br/>Sequences for Action Recognition +<br/><b>Oregon State University</b><br/>CVPR 2016 +</td><td>('3112334', 'Behrooz Mahasseni', 'behrooz mahasseni')<br/>('34917793', 'Sinisa Todorovic', 'sinisa todorovic')</td><td></td></tr><tr><td>519a724426b5d9ad384d38aaf2a4632d3824f243</td><td>WANG et al.: LEARNING OBJECT RECOGNITION FROM DESCRIPTIONS +<br/>Learning Models for Object Recognition +<br/>from Natural Language Descriptions +<br/>School of Computing +<br/><b>University of Leeds</b><br/>Leeds, UK +</td><td>('2635321', 'Josiah Wang', 'josiah wang')<br/>('1686341', 'Katja Markert', 'katja markert')<br/>('3056091', 'Mark Everingham', 'mark everingham')</td><td>scs6jwks@comp.leeds.ac.uk +<br/>markert@comp.leeds.ac.uk +<br/>me@comp.leeds.ac.uk +</td></tr><tr><td>5180df9d5eb26283fb737f491623395304d57497</td><td>Scalable Angular Discriminative Deep Metric Learning +<br/>for Face Recognition +<br/><b>aCenter for Combinatorics, Nankai University, Tianjin 300071, China</b><br/><b>bCenter for Applied Mathematics, Tianjin University, Tianjin 300072, China</b></td><td>('2143751', 'Bowen Wu', 'bowen wu')</td><td></td></tr><tr><td>51c7c5dfda47647aef2797ac3103cf0e108fdfb4</td><td>CS 395T: Celebrity Look-Alikes ∗ +</td><td>('2362854', 'Adrian Quark', 'adrian quark')</td><td>quark@mail.utexas.edu +</td></tr><tr><td>519f4eb5fe15a25a46f1a49e2632b12a3b18c94d</td><td>Non-Lambertian Reflectance Modeling and +<br/>Shape Recovery of Faces using Tensor Splines +</td><td>('9432255', 'Ritwik Kumar', 'ritwik kumar')<br/>('1765280', 'Angelos Barmpoutis', 'angelos barmpoutis')<br/>('3163927', 'Arunava Banerjee', 'arunava banerjee')<br/>('1733005', 'Baba C. Vemuri', 'baba c. vemuri')</td><td></td></tr><tr><td>518edcd112991a1717856841c1a03dd94a250090</td><td><b>Rice University</b><br/>Endogenous Sparse Recovery +<br/>by +<br/>A Thesis Submitted +<br/>in Partial Fulfillment of the +<br/>Requirements for the Degree +<br/>Masters of Science +<br/>Approved, Thesis Committee: +<br/>Dr. Richard G. Baraniuk, Chair +<br/>Victor E. Cameron Professor of Electrical +<br/>and Computer Engineering +<br/>Dr. Don H. Johnson +<br/>J.S. Abercrombie Professor Emeritus of +<br/>Electrical and Computer Engineering +<br/>Dr. Wotao Yin +<br/>Assistant Professor of Computational and +<br/>Applied Mathematics +<br/>Houston, Texas +<br/>December 2011 +</td><td>('1746363', 'Eva L. Dyer', 'eva l. dyer')</td><td></td></tr><tr><td>51683eac8bbcd2944f811d9074a74d09d395c7f3</td><td>Automatic Analysis of Facial Actions: +<br/>Learning from Transductive, Supervised and +<br/>Unsupervised Frameworks +<br/>CMU-RI-TR-17-01 +<br/>January 2017 +<br/><b>The Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>Thesis Committee: +<br/>Fernando De la Torre, Co-chair +<br/>Submitted in partial fulfillment of the requirements +<br/>for the degree of Doctor of Philosophy in Robotics. +</td><td>('39336289', 'Wen-Sheng Chu', 'wen-sheng chu')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')<br/>('1820249', 'Simon Lucey', 'simon lucey')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')<br/>('1736042', 'Vladimir Pavlovic', 'vladimir pavlovic')<br/>('39336289', 'Wen-Sheng Chu', 'wen-sheng chu')</td><td></td></tr><tr><td>51faacfa4fb1e6aa252c6970e85ff35c5719f4ff</td><td>Zoom-Net: Mining Deep Feature Interactions for +<br/>Visual Relationship Recognition +<br/><b>University of Science and Technology of China, Key Laboratory of Electromagnetic</b><br/>Space Information, the Chinese Academy of Sciences, 2SenseTime Group Limited, +<br/><b>CUHK-SenseTime Joint Lab, The Chinese University of Hong Kong</b><br/><b>SenseTime-NTU Joint AI Research Centre, Nanyang Technological University</b></td><td>('4332039', 'Guojun Yin', 'guojun yin')<br/>('37145669', 'Lu Sheng', 'lu sheng')<br/>('50677886', 'Bin Liu', 'bin liu')<br/>('1708598', 'Nenghai Yu', 'nenghai yu')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('49895575', 'Jing Shao', 'jing shao')<br/>('1717179', 'Chen Change Loy', 'chen change loy')</td><td>gjyin@mail.ustc.edu.cn, {flowice,ynh}@ustc.edu.cn, ccloy@ieee.org, +<br/>{lsheng,xgwang}@ee.cuhk.edu.hk, shaojing@sensetime.com +</td></tr><tr><td>51cc78bc719d7ff2956b645e2fb61bab59843d2b</td><td>Face and Facial Expression Recognition with an +<br/>Embedded System for Human-Robot Interaction +<br/><b>School of Computer Engineering, Sejong University, Seoul, Korea</b></td><td>('2241562', 'Yang-Bok Lee', 'yang-bok lee')<br/>('2706430', 'Yong-Guk Kim', 'yong-guk kim')</td><td>*ykim@sejong.ac.kr +</td></tr><tr><td>511b06c26b0628175c66ab70dd4c1a4c0c19aee9</td><td>International Journal of Engineering Research and General ScienceVolume 2, Issue 5, August – September 2014 +<br/>ISSN 2091-2730 +<br/>Face Recognition using Laplace Beltrami Operator by Optimal Linear +<br/>Approximations +<br/><b>Institute of Engineering and Technology, Alwar, Rajasthan Technical University, Kota(Raj</b><br/><b>Research Scholar (M.Tech, IT), Institute of Engineering and Technology</b></td><td></td><td></td></tr><tr><td>51528cdce7a92835657c0a616c0806594de7513b</td><td></td><td></td><td></td></tr><tr><td>51cb09ee04831b95ae02e1bee9b451f8ac4526e3</td><td>Beyond Short Snippets: Deep Networks for Video Classification +<br/>Matthew Hausknecht2 +<br/><b>University of Maryland, College Park</b><br/><b>University of Texas at Austin</b><br/><b>Google, Inc</b></td><td>('2340579', 'Joe Yue-Hei Ng', 'joe yue-hei ng')<br/>('1689108', 'Oriol Vinyals', 'oriol vinyals')<br/>('3089272', 'Rajat Monga', 'rajat monga')<br/>('2259154', 'Sudheendra Vijayanarasimhan', 'sudheendra vijayanarasimhan')<br/>('1805076', 'George Toderici', 'george toderici')</td><td>yhng@umiacs.umd.edu +<br/>mhauskn@cs.utexas.edu +<br/>svnaras@google.com +<br/>vinyals@google.com +<br/>rajatmonga@google.com +<br/>gtoderici@google.com +</td></tr><tr><td>514a74aefb0b6a71933013155bcde7308cad2b46</td><td><b>CARNEGIE MELLON UNIVERSITY</b><br/>OPTIMAL CLASSIFIER ENSEMBLES +<br/>FOR IMPROVED BIOMETRIC VERIFICATION +<br/>A Dissertation +<br/>Submitted to the Faculty of Graduate School +<br/>In Partial Fulfillment of the Requirements +<br/>for The Degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>in +<br/>ELECTRICAL AND COMPUTER ENGINEERING +<br/>by +<br/>COMMITTEE: +<br/>Advisor: Prof. Vijayakumar Bhagavatula +<br/>Prof. Tsuhan Chen +<br/>Prof. David Casasent +<br/>Prof. Arun Ross +<br/>Pittsburgh, Pennsylvania +<br/>January, 2007 +</td><td>('2202489', 'Krithika Venkataramani', 'krithika venkataramani')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td></td></tr><tr><td>51a8dabe4dae157aeffa5e1790702d31368b9161</td><td>SPI-J068 00418 +<br/>International Journal of Pattern Recognition +<br/>and Artificial Intelligence +<br/>Vol. 19, No. 4 (2005) 513–531 +<br/>c(cid:1) World Scientific Publishing Company +<br/>FACE RECOGNITION UNDER GENERIC ILLUMINATION +<br/>BASED ON HARMONIC RELIGHTING +<br/>Graduate School of Chinese Academy Sciences +<br/>No. 19, Yuquan Road, Beijing, 100039, P.R. China +<br/><b>Institute of Computing Technology, CAS</b><br/>No. 6 Kexueyuan South Road, Beijing, 100080, P.R. China +<br/>The performances of the current face recognition systems suffer heavily from the vari- +<br/>ations in lighting. To deal with this problem, this paper presents an illumination nor- +<br/>malization approach by relighting face images to a canonical illumination based on the +<br/>harmonic images model. Benefiting from the observations that human faces share sim- +<br/>ilar shape, and the albedos of the face surfaces are quasi-constant, we first estimate +<br/>the nine low-frequency components of the illumination from the input facial image. The +<br/>facial image is then normalized to the canonical illumination by re-rendering it using +<br/>the illumination ratio image technique. For the purpose of face recognition, two kinds of +<br/>canonical illuminations, the uniform illumination and a frontal flash with the ambient +<br/>lights, are considered, among which the former encodes merely the texture information, +<br/>while the latter encodes both the texture and shading information. Our experiments on +<br/>the CMU-PIE face database and the Yale B face database have shown that the proposed +<br/>relighting normalization can significantly improve the performance of a face recognition +<br/>system when the probes are collected under varying lighting conditions. +<br/>Keywords: Face recognition; varying lighting; harmonic images; lighting estimation; +<br/>illumination normalization. +<br/>1. Introduction +<br/>Face recognition has various potential applications in public security, law enforce- +<br/>ment and commerce such as mug-shot database matching, identity authentication +<br/>for credit card or driver license, access control, information security, and video +<br/>surveillance. In addition, there are many emerging fields that can benefit from face +<br/><b>recognition, such as human computer interfaces and e-services, including e-home</b><br/>online-shopping and online-banking. Related research activities have significantly +<br/>increased over the past few years.5,26 +<br/>513 +</td><td>('2343895', 'Laiyun Qing', 'laiyun qing')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1698902', 'Wen Gao', 'wen gao')<br/>('1691233', 'Bo Du', 'bo du')</td><td>lyqing@jdl.ac.cn +<br/>sgshan@jdl.ac.cn +<br/>wgao@jdl.ac.cn +<br/>bdu@jdl.ac.cn +</td></tr><tr><td>512b4c8f0f3fb23445c0c2dab768bcd848fa8392</td><td> Analysis and Synthesis of Facial Expressions by Feature- +<br/>Points Tracking and Deformable Model +<br/> 1- Faculty of Electrical and Computer Eng., +<br/><b>University of Tabriz, Tabriz, Iran</b><br/>2- Department of Electrical Eng., +<br/><b>Tarbiat Modarres University, Tehran, Iran</b><br/>in +<br/>an +<br/>role +<br/>essential +<br/>facial expressions +<br/> +</td><td>('3210269', 'H. Seyedarabi', 'h. seyedarabi')<br/>('31092101', 'A. Aghagolzadeh', 'a. aghagolzadeh')<br/>('2052255', 'S. Khanmohammadi', 's. khanmohammadi')<br/>('2922912', 'E. Kabir', 'e. kabir')</td><td>seyedarabi@tahoo.com, aghagol@tabrizu.ac.ir, khan@tabrizu.ac.ir +</td></tr><tr><td>51eba481dac6b229a7490f650dff7b17ce05df73</td><td>Situation Recognition: +<br/>Visual Semantic Role Labeling for Image Understanding +<br/><b>Computer Science and Engineering, University of Washington, Seattle, WA</b><br/><b>Allen Institute for Arti cial Intelligence (AI2), Seattle, WA</b><br/>Figure 1. Six images that depict situations where actors, objects, substances, and locations play roles in an activity. Below each image is a +<br/>realized frame that summarizes the situation: the left columns (blue) list activity-specific roles (derived from FrameNet, a broad coverage +<br/>verb lexicon) while the right columns (green) list values (from ImageNet) for each role. Three different activities are shown, highlighting +<br/>that visual properties can vary widely between role values (e.g., clipping a sheep’s wool looks very different from clipping a dog’s nails). +</td><td>('2064210', 'Mark Yatskar', 'mark yatskar')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')</td><td>[my89, lsz, ali]@cs.washington.edu +</td></tr><tr><td>5173a20304ea7baa6bfe97944a5c7a69ea72530f</td><td>Sensors 2013, 13, 12830-12851; doi:10.3390/s131012830 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>Best Basis Selection Method Using Learning Weights for +<br/>Face Recognition +<br/><b>The School of Electrical and Electronic Engineering, Yonsei University, 134 Shinchon-Dong</b><br/><b>The School of Electrical Electronic and Control Engineering, Kongju National University</b><br/>275 Budae-Dong, Seobuk-Gu, Cheonan, Chungnam 331-717, Korea +<br/>Tel.: +82-41-521-9168; Fax: +82-41-563-3689. +<br/>Received: 24 July 2013; in revised form: 26 August 2013 / Accepted: 16 September 2013/ +<br/>Published: 25 September 2013 +</td><td>('1801849', 'Wonju Lee', 'wonju lee')<br/>('2840643', 'Minkyu Cheon', 'minkyu cheon')<br/>('2638048', 'Chang-Ho Hyun', 'chang-ho hyun')<br/>('1718637', 'Mignon Park', 'mignon park')</td><td>Seodaemun-Gu, Seoul 120-749, Korea; E-Mails: delicado@yonsei.ac.kr (W.L.); +<br/>1000minkyu@gmail.com (M.C.); mignpark@yonsei.ac.kr (M.P.) +<br/>* Author to whom correspondence should be addressed; E-Mail: hyunch@kongju.ac.kr; +</td></tr><tr><td>51ed4c92cab9336a2ac41fa8e0293c2f5f9bf3b6</td><td>Computing and Informatics, Vol. 22, 2003, ??–?? +<br/>A SURVEY OF FACE DETECTION, EXTRACTION +<br/>AND RECOGNITION +<br/>National Storage System Laboratory +<br/>School of Software Engineering +<br/><b>Huazhong University of Science and Technology</b><br/>Wuhan, 430074, P. R. China +<br/>Manuscript received 23 June 2002; revised 27 January 2003 +<br/>Communicated by Ladislav Hluch´y +</td><td>('2366162', 'Yongzhong Lu', 'yongzhong lu')<br/>('1711876', 'Jingli Zhou', 'jingli zhou')<br/>('1714618', 'Shengsheng Yu', 'shengsheng yu')</td><td>e-mail: luyongz0@sohu.com +</td></tr><tr><td>5161e38e4ea716dcfb554ccb88901b3d97778f64</td><td>SSPP-DAN: DEEP DOMAIN ADAPTATION NETWORK FOR +<br/>FACE RECOGNITION WITH SINGLE SAMPLE PER PERSON +<br/>School of Computing, KAIST, Republic of Korea +</td><td>('2487892', 'Sungeun Hong', 'sungeun hong')<br/>('40506942', 'Woobin Im', 'woobin im')</td><td></td></tr><tr><td>5121f42de7cb9e41f93646e087df82b573b23311</td><td>CLASSIFYING ONLINE DATING PROFILES ON TINDER USING FACENET FACIAL +<br/>EMBEDDINGS +<br/><b>FL</b></td><td></td><td>Charles F. Jekel (cjekel@ufl.edu; cj@jekel.me) and Raphael T. Haftka +</td></tr><tr><td>51d1a6e15936727e8dd487ac7b7fd39bd2baf5ee</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>A Fast and Accurate System for Face Detection, +<br/>Identification, and Verification +</td><td>('48467498', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('2068427', 'Ankan Bansal', 'ankan bansal')<br/>('7674316', 'Jingxiao Zheng', 'jingxiao zheng')<br/>('2680836', 'Hongyu Xu', 'hongyu xu')<br/>('35199438', 'Joshua Gleason', 'joshua gleason')<br/>('2927406', 'Boyu Lu', 'boyu lu')<br/>('8435884', 'Anirudh Nanduri', 'anirudh nanduri')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('38171682', 'Carlos D. Castillo', 'carlos d. castillo')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>5141cf2e59fb2ec9bb489b9c1832447d3cd93110</td><td>Learning Person Trajectory Representations for Team Activity Analysis +<br/><b>Simon Fraser University</b></td><td>('10386960', 'Nazanin Mehrasa', 'nazanin mehrasa')<br/>('19198359', 'Yatao Zhong', 'yatao zhong')<br/>('2123865', 'Frederick Tung', 'frederick tung')<br/>('3004771', 'Luke Bornn', 'luke bornn')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td>{nmehrasa, yataoz, ftung, lbornn}@sfu.ca, mori@cs.sfu.ca +</td></tr><tr><td>5185f2a40836a754baaa7419a1abdd1e7ffaf2ad</td><td>A Multimodality Framework for Creating Speaker/Non-Speaker Profile +<br/>Databases for Real-World Video +<br/><b>Beckman Institute</b><br/><b>University of Illinois</b><br/>Urbana, IL 61801 +<br/><b>Beckman Institute</b><br/><b>University of Illinois</b><br/>Urbana, IL 61801 +<br/><b>Beckman Institute</b><br/><b>University of Illinois</b><br/>Urbana, IL 61801 +</td><td>('3082579', 'Jehanzeb Abbas', 'jehanzeb abbas')<br/>('1804874', 'Charlie K. Dagli', 'charlie k. dagli')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>jabbas2@ifp.uiuc.edu +<br/>dagli@ifp.uiuc.edu +<br/>huang@ifp.uiuc.edu +</td></tr><tr><td>511a8cdf2127ef8aa07cbdf9660fe9e0e2dfbde7</td><td>Hindawi +<br/>Computational Intelligence and Neuroscience +<br/>Volume 2018, Article ID 4512473, 10 pages +<br/>https://doi.org/10.1155/2018/4512473 +<br/>Research Article +<br/>A Community Detection Approach to Cleaning Extremely +<br/>Large Face Database +<br/><b>Computer School, University of South China, Hengyang, China</b><br/><b>National Laboratory for Parallel and Distributed Processing, National University of Defense Technology, Changsha, China</b><br/>Received 11 December 2017; Accepted 12 March 2018; Published 22 April 2018 +<br/>Academic Editor: Amparo Alonso-Betanzos +<br/>permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Though it has been easier to build large face datasets by collecting images from the Internet in this Big Data era, the time-consuming +<br/>manual annotation process prevents researchers from constructing larger ones, which makes the automatic cleaning of noisy labels +<br/>highly desirable. However, identifying mislabeled faces by machine is quite challenging because the diversity of a person’s face +<br/>images that are captured wildly at all ages is extraordinarily rich. In view of this, we propose a graph-based cleaning method that +<br/>mainly employs the community detection algorithm and deep CNN models to delete mislabeled images. As the diversity of faces is +<br/>preserved in multiple large communities, our cleaning results have both high cleanness and rich data diversity. With our method, we +<br/>clean the extremely large MS-Celeb-1M face dataset (approximately 10 million images with noisy labels) and obtain a clean version +<br/>of it called C-MS-Celeb (6,464,018 images of 94,682 celebrities). By training a single-net model using our C-MS-Celeb dataset, +<br/>without fine-tuning, we achieve 99.67% at Equal Error Rate on the LFW face recognition benchmark, which is comparable to other +<br/>state-of-the-art results. This demonstrates the data cleaning positive effects on the model training. To the best of our knowledge, +<br/>our C-MS-Celeb is the largest clean face dataset that is publicly available so far, which will benefit face recognition researchers. +<br/>1. Introduction +<br/>In the last few years, researchers have witnessed the remark- +<br/>able progress in face recognition due to the significant success +<br/>of deep convolutional neural networks [1] and the emergence +<br/>of large scale face datasets [2]. Although the data explosion +<br/>has made it easier to build datasets by collecting real world +<br/>images from the Internet [3], constructing a large scale face +<br/>dataset remains a highly time-consuming and costly task +<br/>because the mislabeled images returned by search engines +<br/>need to be manually removed [4]. Thus, automatic cleaning +<br/>of noisy labels in the raw dataset is strongly desirable. +<br/>However, identifying mislabeled faces automatically by +<br/>machine is by no means easy. The main reason for this is that, +<br/>for faces that are captured wildly, the variation of a man’s faces +<br/>can be so large that some of his images may easily be identified +<br/>as someone else’s [5]. Thus, a machine may be misled by this +<br/>rich data diversity within one person and delete correctly +<br/>labeled images. For example, if old faces of a man are the +<br/>majority in the dataset, a young face of him may be regarded +<br/>as someone else and removed. Another challenge is that, due +<br/>to the ambiguity of people’s names, searching for someone’s +<br/>pictures online usually returns images from multiple people +<br/>[2], which requires the cleaning method to be tolerant to the +<br/>high proportion of noisy labels in the raw dataset constructed +<br/>by online searching. +<br/>In order to clean noisy labels and meanwhile preserve +<br/>the rich data diversity of various faces, we propose a three- +<br/>stage graph-based method to clean large face datasets using +<br/>the community detection algorithm. For each image in the +<br/>raw dataset, we firstly use pretrained deep CNN models to +<br/>align the face and extract a feature vector to represent each +<br/>face. Secondly, for features of the same identity, based on the +<br/>cosine similarity between different features, we construct an +<br/>undirected graph, named “face similarity graph,” to quantify +<br/>the similarity between different images. After deleting weak +<br/>edges and applying the community detection algorithm, we +<br/>delete mislabeled images by removing minor communities. In +<br/>the last stage, we try to relabel each previously deleted image +</td><td>('3335298', 'Chi Jin', 'chi jin')<br/>('9856301', 'Ruochun Jin', 'ruochun jin')<br/>('38536592', 'Kai Chen', 'kai chen')<br/>('1791001', 'Yong Dou', 'yong dou')<br/>('3335298', 'Chi Jin', 'chi jin')</td><td>Correspondence should be addressed to Ruochun Jin; sczjrc@163.com +</td></tr><tr><td>51d048b92f6680aca4a8adf07deb380c0916c808</td><td>This is the accepted version of the following article: "State of the Art on Monocular 3D Face Reconstruction, Tracking, and Applications", +<br/>which has been published in final form at http://onlinelibrary.wiley.com. This article may be used for non-commercial purposes in accordance +<br/>with the Wiley Self-Archiving Policy [http://olabout.wiley.com/WileyCDA/Section/id-820227.html]. +<br/>EUROGRAPHICS 2018 +<br/>K. Hildebrandt and C. Theobalt +<br/>(Guest Editors) +<br/>Volume 37 (2018), Number 2 +<br/>STAR – State of The Art Report +<br/>State of the Art on Monocular 3D Face +<br/>Reconstruction, Tracking, and Applications +<br/>M. Zollhöfer1,2 +<br/>J. Thies3 P. Garrido1,5 D. Bradley4 T. Beeler4 P. Pérez5 M. Stamminger6 M. Nießner3 C. Theobalt1 +<br/><b>Max Planck Institute for Informatics</b><br/><b>Stanford University</b><br/><b>Technical University of Munich</b><br/>4Disney Research +<br/>5Technicolor +<br/><b>University of Erlangen-Nuremberg</b><br/>Figure 1: This state-of-the-art report provides an overview of monocular 3D face reconstruction and tracking, and highlights applications. +</td><td></td><td></td></tr><tr><td>5134353bd01c4ea36bd007c460e8972b1541d0ad</td><td>Face Recognition with Multi-Resolution Spectral Feature +<br/>Images +<br/><b>School of Electrical Engineering and Automation, Anhui University, Hefei, China, Hong Kong Polytechnic</b><br/><b>University, Hong Kong, China, 3 Center for Intelligent Electricity Networks, University of Newcastle, Newcastle, Australia, 4 School of Electrical and Electronic Engineering</b><br/><b>Nanyang Technological University, Singapore, Singapore</b></td><td>('31443079', 'Zhan-Li Sun', 'zhan-li sun')<br/>('1703078', 'Kin-Man Lam', 'kin-man lam')<br/>('50067626', 'Zhao-yang Dong', 'zhao-yang dong')<br/>('40465036', 'Han Wang', 'han wang')<br/>('29927490', 'Qing-wei Gao', 'qing-wei gao')</td><td></td></tr><tr><td>5160569ca88171d5fa257582d161e9063c8f898d</td><td>Local Binary Patterns as an Image Preprocessing for Face Authentication +<br/><b>IDIAP Research Institute, Martigny, Switzerland</b><br/>Ecole Polytechnique F´ed´erale de Lausanne (EPFL), Switzerland +</td><td>('16602458', 'Guillaume Heusch', 'guillaume heusch')<br/>('2820403', 'Yann Rodriguez', 'yann rodriguez')</td><td>fheusch, rodrig, marcelg@idiap.ch +</td></tr><tr><td>5157dde17a69f12c51186ffc20a0a6c6847f1a29</td><td>Evolutionary Cost-sensitive Extreme Learning +<br/>Machine +<br/>1 +</td><td>('40613723', 'Lei Zhang', 'lei zhang')<br/>('1698371', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>51dc127f29d1bb076d97f515dca4cc42dda3d25b</td><td></td><td></td><td></td></tr><tr><td>3d18ce183b5a5b4dcaa1216e30b774ef49eaa46f</td><td>Face Alignment Across Large Poses: A 3D Solution +<br/>Hailin Shi1 +<br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/><b>Michigan State University</b></td><td>('8362374', 'Xiangyu Zhu', 'xiangyu zhu')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{xiangyu.zhu,zlei,hailin.shi,szli}@nlpr.ia.ac.cn +<br/>liuxm@msu.edu +</td></tr><tr><td>3d143cfab13ecd9c485f19d988242e7240660c86</td><td>Discriminative Collaborative Representation for +<br/>Classification +<br/><b>Academic Center for Computing and Media Studies, Kyoto University, Kyoto 606-8501, Japan</b><br/><b>Institute of Scienti c and Industrial Research, Osaka University, Ibaraki-shi 567-0047, Japan</b><br/>3 OMRON Social Solutions Co., LTD, Kyoto 619-0283, Japan +</td><td>('2549020', 'Yang Wu', 'yang wu')<br/>('40400215', 'Wei Li', 'wei li')<br/>('1707934', 'Masayuki Mukunoki', 'masayuki mukunoki')<br/>('1681266', 'Michihiko Minoh', 'michihiko minoh')<br/>('1710195', 'Shihong Lao', 'shihong lao')</td><td>yangwu@mm.media.kyoto-u.ac.jp,seuliwei@126.com, +<br/>{minoh,mukunoki}@media.kyoto-u.ac.jp,lao_shihong@oss.omron.co.jp +</td></tr><tr><td>3daafe6389d877fe15d8823cdf5ac15fd919676f</td><td>Human Action Localization +<br/>with Sparse Spatial Supervision +</td><td>('2492127', 'Philippe Weinzaepfel', 'philippe weinzaepfel')<br/>('49142153', 'Xavier Martin', 'xavier martin')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>3dabf7d853769cfc4986aec443cc8b6699136ed0</td><td>In A. Esposito, N. Bourbakis, N. Avouris, and I. Hatzilygeroudis. (Eds.) Lecture Notes in +<br/>Computer Science, Vol 5042: Verbal and Nonverbal Features of Human-human and Human- +<br/>machine Interaction, Springer Verlag, p. 1-21. +<br/>Data mining spontaneous facial behavior with +<br/>automatic expression coding +<br/><b>Institute for Neural Computation, University of California, San Diego, La Jolla, CA</b><br/><b>Human Development and Applied Psychology, University of Toronto, Ontario, Canada</b><br/>0445, USA +<br/><b>Engineering and Natural Science, Sabanci University, Istanbul, Turkey</b></td><td>('2724380', 'Gwen Littlewort', 'gwen littlewort')<br/>('40322754', 'Esra Vural', 'esra vural')<br/>('2855884', 'Kang Lee', 'kang lee')</td><td>mbartlett@ucsd.edu; gwen@mpmlab.ucsd.edu, movellan@mplab.ucsd.edu, +<br/>vesra@ucsd.edu, kang.lee@utoronto.ca +</td></tr><tr><td>3db75962857a602cae65f60f202d311eb4627b41</td><td></td><td></td><td></td></tr><tr><td>3daf1191d43e21a8302d98567630b0e2025913b0</td><td>Can Autism be Catered with Artificial Intelligence-Assisted Intervention +<br/>Technology? A Literature Review +<br/><b>Faculty of Information Technology, Barrett Hodgson University, Karachi, Pakistan</b><br/>†Universit´e Claude Bernard Lyon 1, France +</td><td>('38817141', 'Muhammad Shoaib Jaliawala', 'muhammad shoaib jaliawala')<br/>('1943666', 'Rizwan Ahmed Khan', 'rizwan ahmed khan')</td><td></td></tr><tr><td>3d36f941d8ec613bb25e80fb8f4c160c1a2848df</td><td>Out-of-sample generalizations for supervised +<br/>manifold learning for classification +</td><td>('12636684', 'Elif Vural', 'elif vural')<br/>('1780587', 'Christine Guillemot', 'christine guillemot')</td><td></td></tr><tr><td>3d5a1be4c1595b4805a35414dfb55716e3bf80d8</td><td>Hidden Two-Stream Convolutional Networks for +<br/>Action Recognition +</td><td>('1749901', 'Yi Zhu', 'yi zhu')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td></td></tr><tr><td>3d62b2f9cef997fc37099305dabff356d39ed477</td><td>Joint Face Alignment and 3D Face +<br/>Reconstruction with Application to Face +<br/>Recognition +</td><td>('33320460', 'Feng Liu', 'feng liu')<br/>('7345195', 'Qijun Zhao', 'qijun zhao')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('39422721', 'Dan Zeng', 'dan zeng')</td><td></td></tr><tr><td>3dc522a6576c3475e4a166377cbbf4ba389c041f</td><td></td><td></td><td></td></tr><tr><td>3dd4d719b2185f7c7f92cc97f3b5a65990fcd5dd</td><td>Ensemble of Hankel Matrices for +<br/>Face Emotion Recognition +<br/>DICGIM, Universit´a degli Studi di Palermo, +<br/>V.le delle Scienze, Ed. 6, 90128 Palermo, Italy, +<br/>DRAFT +<br/>To appear in ICIAP 2015 +</td><td>('1711610', 'Liliana Lo Presti', 'liliana lo presti')<br/>('9127836', 'Marco La Cascia', 'marco la cascia')</td><td>liliana.lopresti@unipa.it +</td></tr><tr><td>3d1a6a5fd5915e0efb953ede5af0b23debd1fc7f</td><td>Proceedings of the Pakistan Academy of Sciences 52 (1): 27–38 (2015) +<br/>Copyright © Pakistan Academy of Sciences +<br/>ISSN: 0377 - 2969 (print), 2306 - 1448 (online) +<br/> Pakistan Academy of Sciences +<br/>Research Article +<br/>Bimodal Human Emotion Classification in the +<br/>Speaker-Dependent Scenario +<br/><b>University of Peshawar, Peshawar, Pakistan</b><br/><b>University of Engineering and Technology</b><br/><b>Sarhad University of Science and Information Technology</b><br/><b>University of Peshawar, Peshawar, Pakistan</b><br/>Peshawar, Pakistan +<br/>Peshawar, Pakistan +<br/> +</td><td>('34267835', 'Sanaul Haq', 'sanaul haq')<br/>('3124216', 'Tariqullah Jan', 'tariqullah jan')<br/>('1766329', 'Muhammad Asif', 'muhammad asif')<br/>('1710701', 'Amjad Ali', 'amjad ali')<br/>('40332145', 'Naveed Ahmad', 'naveed ahmad')</td><td></td></tr><tr><td>3d0379688518cc0e8f896e30815d0b5e8452d4cd</td><td>Autotagging Facebook: +<br/>Social Network Context Improves Photo Annotation +<br/><b>Harvard University</b><br/>Todd Zickler +<br/><b>Harvard University</b><br/>UC Berkeley EECS & ICSI +</td><td>('2201347', 'Zak Stone', 'zak stone')<br/>('1753210', 'Trevor Darrell', 'trevor darrell')</td><td>zstone@fas.harvard.edu +<br/>zickler@seas.harvard.edu +<br/>trevor@eecs.berkeley.edu +</td></tr><tr><td>3dda181be266950ba1280b61eb63ac11777029f9</td><td></td><td></td><td></td></tr><tr><td>3d24b386d003bee176a942c26336dbe8f427aadd</td><td>Sequential Person Recognition in Photo Albums with a Recurrent Network∗ +<br/><b>The University of Adelaide, Australia</b></td><td>('39948681', 'Yao Li', 'yao li')<br/>('2604251', 'Guosheng Lin', 'guosheng lin')<br/>('3194022', 'Bohan Zhuang', 'bohan zhuang')<br/>('2161037', 'Lingqiao Liu', 'lingqiao liu')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')<br/>('5546141', 'Anton van den Hengel', 'anton van den hengel')</td><td></td></tr><tr><td>3dcebd4a1d66313dcd043f71162d677761b07a0d</td><td> Yerel Đkili Örüntü Ortamında Yerel Görünüme Dayalı Yüz Tanıma +<br/>Local Binary Pattern Domain Local Appearance Face Recognition +<br/>Hazım K. Ekenel1, Mika Fischer1, Erkin Tekeli2, Rainer Stiefelhagen1, Aytül Erçil2 +<br/>1Institut für Theorestische Informatik, Universität Karlsruhe (TH), Karlsruhe, Germany +<br/><b>Faculty of Engineering and Natural Sciences, Sabanc University, stanbul, Turkey</b><br/>Özetçe +<br/>Bu bildiride, ayrık kosinüs dönüşümü tabanlı yerel görünüme +<br/>dayalı yüz tanıma algoritması ile yüz imgelerinin yerel ikili +<br/>örüntüye (YĐÖ) dayalı betimlemesini birleştiren hızlı bir yüz +<br/>tanıma algoritması sunulmuştur. Bu tümleştirmedeki amaç, +<br/>yerel ikili örüntünün dayanıklı imge betimleme yeteneği ile +<br/>ayrık kosinüs dönüşümünün derli-toplu veri betimleme +<br/>yeteneğinden yararlanmaktır. Önerilen yaklaşımda, yerel +<br/>görünümün modellenmesinden önce girdi yüz imgesi yerel +<br/>ikili örüntü ile betimlenmiştir. Elde edilen YĐÖ betimlemesi, +<br/>birbirleri ile örtüşmeyen bloklara ayrılmış ve her blok +<br/>üzerinde yerel özniteliklerin çıkartımı için ayrık kosinüs +<br/>dönüşümü uygulanmıştır. Çıkartımı yapılan yerel öznitelikler +<br/>daha sonra arka arkaya eklenerek global öznitelik vektörü +<br/>oluşturulmuştur. Önerilen algoritma, CMU PIE ve FRGC +<br/>versiyon 2 veritabanlarından seçilen yüz imgeleri üzerinde +<br/>sınanmıştır. Deney sonuçları, tümleşik yöntemin başarımı +<br/>önemli ölçüde arttırdığını göstermiştir. +</td><td></td><td>{ekenel,mika.fischer,stiefel}@ira.uka.de, {erkintekeli,aytulercil}@sabanciuniv.edu +</td></tr><tr><td>3d0f9a3031bee4b89fab703ff1f1d6170493dc01</td><td>SVDD-Based Illumination Compensation +<br/>for Face Recognition +<br/><b>The Robotics Institute, Carnegie Mellon University</b><br/>5000 Forbes Ave., Pittsburgh, PA 15213, USA +<br/><b>Center for Arti cial Vision Research, Korea University</b><br/>Anam-dong, Seongbuk-ku, Seoul 136-713, Korea +</td><td>('2348968', 'Sang-Woong Lee', 'sang-woong lee')<br/>('1703007', 'Seong-Whan Lee', 'seong-whan lee')</td><td>rhiephil@cs.cmu.edu +<br/>swlee@image.korea.ac.kr +</td></tr><tr><td>3d6ee995bc2f3e0f217c053368df659a5d14d5b5</td><td></td><td></td><td></td></tr><tr><td>3d0c21d4780489bd624a74b07e28c16175df6355</td><td>Deep or Shallow Facial Descriptors? A Case for +<br/>Facial Attribute Classification and Face Retrieval +<br/>1 Faculty of Engineering, +<br/><b>Multimedia University, Cyberjaya, Malaysia</b><br/>2 Faculty of Computing & Informatics, +<br/><b>Multimedia University, Cyberjaya, Malaysia</b></td><td>('3366793', 'Rasoul Banaeeyan', 'rasoul banaeeyan')<br/>('31612015', 'Mohd Haris Lye', 'mohd haris lye')<br/>('4759494', 'Mohammad Faizal Ahmad Fauzi', 'mohammad faizal ahmad fauzi')<br/>('2339975', 'John See', 'john see')</td><td>banaeeyan@gmail.com, {haris.lye, faizal1, hezerul, johnsee}@mmu.edu.my +</td></tr><tr><td>3df8cc0384814c3fb05c44e494ced947a7d43f36</td><td>The Pose Knows: Video Forecasting by Generating Pose Futures +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Avenue, Pittsburgh, PA 15213 +</td><td>('14192361', 'Jacob Walker', 'jacob walker')<br/>('35789996', 'Kenneth Marino', 'kenneth marino')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')<br/>('1709305', 'Martial Hebert', 'martial hebert')</td><td>{jcwalker, kdmarino, abhinavg, hebert}@cs.cmu.edu +</td></tr><tr><td>3d42e17266475e5d34a32103d879b13de2366561</td><td>Proc.4thIEEEInt’lConf.AutomaticFace&GestureRecognition,Grenoble,France,pp264–270 +<br/>The Global Dimensionality of Face Space +<br/>(cid:3) +<br/>http://venezia.rockefeller.edu/ +<br/><b>The Rockefeller University</b><br/>Laboratory of Computational Neuroscience +<br/>Laboratory for Applied Mathematics +<br/>Mount Sinai School of Medicine +<br/>c(cid:13) IEEE2000 +<br/>1230 York Avenue, New York, NY 10021 +<br/>One Gustave L. Levy Place, New York, NY 10029 +</td><td>('2939761', 'Penio S. Penev', 'penio s. penev')<br/>('3266322', 'Lawrence Sirovich', 'lawrence sirovich')</td><td>PenevPS@IEEE.org +<br/>chico@camelot.mssm.edu +</td></tr><tr><td>3dd906bc0947e56d2b7bf9530b11351bbdff2358</td><td></td><td></td><td></td></tr><tr><td>3dfd94d3fad7e17f52a8ae815eb9cc5471172bc0</td><td>Face2Text: Collecting an Annotated Image Description Corpus for the +<br/>Generation of Rich Face Descriptions +<br/><b>University of Malta</b><br/><b>University of Copenhagen</b></td><td>('1700894', 'Albert Gatt', 'albert gatt')<br/>('32227979', 'Marc Tanti', 'marc tanti')<br/>('35347012', 'Adrian Muscat', 'adrian muscat')<br/>('1782032', 'Patrizia Paggio', 'patrizia paggio')<br/>('2870709', 'Claudia Borg', 'claudia borg')<br/>('3356545', 'Lonneke van der Plas', 'lonneke van der plas')</td><td>{albert.gatt, marc.tanti.06, adrian.muscat, patrizia.paggio, reuben.farrugia}@um.edu.mt +<br/>{claudia.borg, kenneth.camilleri, mike.rosner, lonneke.vanderplas}@um.edu.mt +<br/>paggio@hum.ku.dk +</td></tr><tr><td>3dbfd2fdbd28e4518e2ae05de8374057307e97b3</td><td>Improving Face Detection +<br/><b>CISUC, University of Coimbra</b><br/><b>Faculty of Computer Science, University of A Coru na, Coru na, Spain</b></td><td>('2045142', 'Penousal Machado', 'penousal machado')<br/>('39583137', 'Juan Romero', 'juan romero')</td><td>3030 Coimbra, Portugal machado@dei.uc.pt, jncor@dei.uc.pt +<br/>jj@udc.pt +</td></tr><tr><td>3df7401906ae315e6aef3b4f13126de64b894a54</td><td>Robust Learning of Discriminative Projection for Multicategory Classification on +<br/>the Stiefel Manifold +<br/><b>Curtin University of Technology</b><br/>GPO Box U1987, Perth, WA 6845, Australia +</td><td>('1725024', 'Duc-Son Pham', 'duc-son pham')<br/>('1679520', 'Svetha Venkatesh', 'svetha venkatesh')</td><td>dspham@ieee.org, svetha@cs.curtin.edu.au +</td></tr><tr><td>3d68cedd80babfbb04ab197a0b69054e3c196cd9</td><td>Bimodal Information Analysis for Emotion Recognition +<br/>Master of Engineering +<br/>Department of Electrical and Computer Engineering +<br/><b>McGill University</b><br/>Montreal, Quebec +<br/>October 2009 +<br/>Revised: February 2010 +<br/><b>A Thesis submitted to McGill University in partial fulfillment of the requirements for the</b><br/>degree of Master of Engineering +<br/>i +</td><td>('2376514', 'Malika Meghjani', 'malika meghjani')<br/>('2376514', 'Malika Meghjani', 'malika meghjani')</td><td></td></tr><tr><td>3dfb822e16328e0f98a47209d7ecd242e4211f82</td><td>Cross-Age LFW: A Database for Studying Cross-Age Face Recognition in +<br/>Unconstrained Environments +<br/><b>Beijing University of Posts and Telecommunications</b><br/>Beijing 100876,China +</td><td>('15523767', 'Tianyue Zheng', 'tianyue zheng')<br/>('1774956', 'Weihong Deng', 'weihong deng')<br/>('23224233', 'Jiani Hu', 'jiani hu')</td><td>2231135739@qq.com, whdeng@bupt.edu.cn, 40902063@qq.com +</td></tr><tr><td>3d1af6c531ebcb4321607bcef8d9dc6aa9f0dc5a</td><td>1892 +<br/>Random Multispace Quantization as +<br/>an Analytic Mechanism for BioHashing +<br/>of Biometric and Random Identity Inputs +</td><td>('2124820', 'Alwyn Goh', 'alwyn goh')</td><td></td></tr><tr><td>3d6943f1573f992d6897489b73ec46df983d776c</td><td></td><td></td><td></td></tr><tr><td>3d948e4813a6856e5b8b54c20e50cc5050e66abe</td><td>A Smart Phone Image Database for Single +<br/>Image Recapture Detection +<br/><b>Institute for Infocomm Research, A*STAR, Singapore</b><br/>2 Department of Electrical and Computer Engineering +<br/><b>National University of Singapore, Singapore</b><br/>3 Department of Electrical and Computer Engineering +<br/><b>New Jersey Institute of Technology, USA</b></td><td>('2740420', 'Xinting Gao', 'xinting gao')<br/>('2821964', 'Bo Qiu', 'bo qiu')<br/>('3138499', 'JingJing Shen', 'jingjing shen')<br/>('2475944', 'Tian-Tsong Ng', 'tian-tsong ng')</td><td>{xgao, qiubo, ttng}@i2r.a-star.eud.sg +<br/>shenjingjing89@gmail.com +<br/>shi@njit.edu +</td></tr><tr><td>3d94f81cf4c3a7307e1a976dc6cb7bf38068a381</td><td>3846 +<br/>Data-Dependent Label Distribution Learning +<br/>for Age Estimation +</td><td>('3276410', 'Zhouzhou He', 'zhouzhou he')<br/>('40613648', 'Xi Li', 'xi li')<br/>('1720488', 'Zhongfei Zhang', 'zhongfei zhang')<br/>('28342797', 'Fei Wu', 'fei wu')<br/>('1735299', 'Xin Geng', 'xin geng')<br/>('2998634', 'Yaqing Zhang', 'yaqing zhang')<br/>('37144787', 'Ming-Hsuan Yang', 'ming-hsuan yang')<br/>('1755711', 'Yueting Zhuang', 'yueting zhuang')</td><td></td></tr><tr><td>3d9db1cacf9c3bb7af57b8112787b59f45927355</td><td>Original research +<br/>published: 20 June 2016 +<br/>doi: 10.3389/fict.2016.00011 +<br/>improving Medical students’ +<br/>awareness of Their non-Verbal +<br/>communication through automated +<br/>non-Verbal Behavior Feedback +<br/><b>School of Electrical and Information Engineering, The University of Sydney, Sydney, NSW, Australia, 2 Sydney Medical</b><br/><b>School, The University of Sydney, Sydney, NSW, Australia</b><br/>The non-verbal communication of clinicians has an impact on patients’ satisfaction and +<br/>health outcomes. Yet medical students are not receiving enough training on the appropri- +<br/>ate non-verbal behaviors in clinical consultations. Computer vision techniques have been +<br/>used for detecting different kinds of non-verbal behaviors, and they can be incorporated +<br/>in educational systems that help medical students to develop communication skills. +<br/>We describe EQClinic, a system that combines a tele-health platform with automated +<br/>non-verbal behavior recognition. The system aims to help medical students improve +<br/>their communication skills through a combination of human and automatically generated +<br/>feedback. EQClinic provides fully automated calendaring and video conferencing features +<br/>for doctors or medical students to interview patients. We describe a pilot (18 dyadic +<br/>interactions) in which standardized patients (SPs) (i.e., someone acting as a real patient) +<br/>were interviewed by medical students and provided assessments and comments about +<br/>their performance. After the interview, computer vision and audio processing algorithms +<br/>were used to recognize students’ non-verbal behaviors known to influence the quality of +<br/>a medical consultation: including turn taking, speaking ratio, sound volume, sound pitch, +<br/>smiling, frowning, head leaning, head tilting, nodding, shaking, face-touch gestures and +<br/>overall body movements. The results showed that students’ awareness of non-verbal +<br/>communication was enhanced by the feedback information, which was both provided +<br/>by the SPs and generated by the machines. +<br/>Keywords: non-verbal communication, non-verbal behavior, clinical consultation, medical education, +<br/>communication skills, non-verbal behavior detection, automated feedback +<br/>inTrODUcTiOn +<br/>Edited by: +<br/>Leman Figen Gul, +<br/><b>Istanbul Technical University, Turkey</b><br/>Reviewed by: +<br/>Marc Aurel Schnabel, +<br/><b>Victoria University of Wellington</b><br/>New Zealand +<br/>Antonella Lotti, +<br/><b>University of Genoa, Italy</b><br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted +<br/>to Digital Education, +<br/>a section of the journal +<br/>Frontiers in ICT +<br/>Received: 28 April 2016 +<br/>Accepted: 07 June 2016 +<br/>Published: 20 June 2016 +<br/>Citation: +<br/>Liu C, Calvo RA and Lim R (2016) +<br/>Improving Medical Students’ +<br/>Awareness of Their Non-Verbal +<br/>Communication through Automated +<br/>Non-Verbal Behavior Feedback. +<br/>doi: 10.3389/fict.2016.00011 +<br/>Over the last 10 years, we have witnessed a dramatic improvement in affective computing (Picard, +<br/>2000; Calvo et al., 2015) and behavior recognition techniques (Vinciarelli et al., 2012). These +<br/>techniques have progressed from the recognition of person-specific posed behavior to the more +<br/>difficult person-independent recognition of behavior in “the-wild” (Vinciarelli et al., 2009). They +<br/>are considered robust enough that they are being incorporated into new applications. For example, +<br/>new learning technologies have been developed that detect a student’s emotions and use this to guide +<br/>the learning experience (Calvo and D’Mello, 2011). They can also be used to support reflection by +<br/>Frontiers in ICT | www.frontiersin.org +<br/>June 2016 | Volume 3 | Article 11 +</td><td>('30772945', 'Chunfeng Liu', 'chunfeng liu')<br/>('1742162', 'Rafael A. Calvo', 'rafael a. calvo')<br/>('36807976', 'Renee Lim', 'renee lim')<br/>('1742162', 'Rafael A. Calvo', 'rafael a. calvo')</td><td>rafael.calvo@sydney.edu.au +</td></tr><tr><td>580f86f1ace1feed16b592d05c2b07f26c429b4b</td><td>Dense-Captioning Events in Videos +<br/><b>Stanford University</b></td><td>('2580593', 'Ranjay Krishna', 'ranjay krishna')<br/>('35163655', 'Kenji Hata', 'kenji hata')<br/>('3260219', 'Frederic Ren', 'frederic ren')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')<br/>('9200530', 'Juan Carlos Niebles', 'juan carlos niebles')</td><td>{ranjaykrishna, kenjihata, fren, feifeili, jniebles}@cs.stanford.edu +</td></tr><tr><td>58d47c187b38b8a2bad319c789a09781073d052d</td><td>Factorizable Net: An Efficient Subgraph-based +<br/>Framework for Scene Graph Generation +<br/><b>The Chinese University of Hong Kong, Hong Kong SAR, China</b><br/><b>The University of Sydney, SenseTime Computer Vision Research Group</b><br/>3 MIT CSAIL, USA +<br/>4 Sensetime Ltd, Beijing, China +<br/><b>Samsung Telecommunication Research Institute, Beijing, China</b></td><td>('2180892', 'Yikang Li', 'yikang li')<br/>('3001348', 'Wanli Ouyang', 'wanli ouyang')<br/>('1804424', 'Bolei Zhou', 'bolei zhou')<br/>('1788070', 'Jianping Shi', 'jianping shi')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')</td><td>{ykli, xgwang}@ee.cuhk.edu.hk, wanli.ouyang@sydney.edu.au, +<br/>bzhou@csail.mit.edu, shijianping@sensetime.com, c0502.zhang@samsung.com +</td></tr><tr><td>582edc19f2b1ab2ac6883426f147196c8306685a</td><td>Do We Really Need to Collect Millions of Faces +<br/>for Effective Face Recognition? +<br/><b>Institute for Robotics and Intelligent Systems, USC, CA, USA</b><br/><b>Information Sciences Institute, USC, CA, USA</b><br/><b>The Open University of Israel, Israel</b></td><td>('11269472', 'Iacopo Masi', 'iacopo masi')<br/>('2955822', 'Jatuporn Toy Leksut', 'jatuporn toy leksut')<br/>('1756099', 'Tal Hassner', 'tal hassner')</td><td></td></tr><tr><td>5859774103306113707db02fe2dd3ac9f91f1b9e</td><td></td><td></td><td></td></tr><tr><td>5892f8367639e9c1e3cf27fdf6c09bb3247651ed</td><td>Estimating Missing Features to Improve Multimedia Information Retrieval +</td><td>('2666918', 'Abraham Bagherjeiran', 'abraham bagherjeiran')<br/>('35089151', 'Nicole S. Love', 'nicole s. love')<br/>('1696815', 'Chandrika Kamath', 'chandrika kamath')</td><td></td></tr><tr><td>5850aab97e1709b45ac26bb7d205e2accc798a87</td><td></td><td></td><td></td></tr><tr><td>587f81ae87b42c18c565694c694439c65557d6d5</td><td>DeepFace: Face Generation using Deep Learning +</td><td>('31560532', 'Hardie Cate', 'hardie cate')<br/>('6415321', 'Fahim Dalvi', 'fahim dalvi')<br/>('8815003', 'Zeshan Hussain', 'zeshan hussain')</td><td>ccate@stanford.edu +<br/>fdalvi@cs.stanford.edu +<br/>zeshanmh@stanford.edu +</td></tr><tr><td>580054294ca761500ada71f7d5a78acb0e622f19</td><td>1331 +<br/>A Subspace Model-Based Approach to Face +<br/>Relighting Under Unknown Lighting and Poses +</td><td>('2081318', 'Hyunjung Shim', 'hyunjung shim')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td></td></tr><tr><td>587c48ec417be8b0334fa39075b3bfd66cc29dbe</td><td>Journal of Vision (2016) 16(15):28, 1–8 +<br/>Serial dependence in the perception of attractiveness +<br/><b>University of California</b><br/>Berkeley, CA, USA +<br/><b>University of California</b><br/>Berkeley, CA, USA +<br/><b>University of California</b><br/>Berkeley, CA, USA +<br/><b>Helen Wills Neuroscience Institute, University of</b><br/>California, Berkeley, CA, USA +<br/><b>Vision Science Group, University of California</b><br/>Berkeley, CA, USA +<br/>The perception of attractiveness is essential for choices +<br/>of food, object, and mate preference. Like perception of +<br/>other visual features, perception of attractiveness is +<br/>stable despite constant changes of image properties due +<br/>to factors like occlusion, visual noise, and eye +<br/>movements. Recent results demonstrate that perception +<br/>of low-level stimulus features and even more complex +<br/>attributes like human identity are biased towards recent +<br/>percepts. This effect is often called serial dependence. +<br/>Some recent studies have suggested that serial +<br/>dependence also exists for perceived facial +<br/>attractiveness, though there is also concern that the +<br/>reported effects are due to response bias. Here we used +<br/>an attractiveness-rating task to test the existence of +<br/>serial dependence in perceived facial attractiveness. Our +<br/>results demonstrate that perceived face attractiveness +<br/>was pulled by the attractiveness level of facial images +<br/>encountered up to 6 s prior. This effect was not due to +<br/>response bias and did not rely on the previous motor +<br/>response. This perceptual pull increased as the difference +<br/>in attractiveness between previous and current stimuli +<br/>increased. Our results reconcile previously conflicting +<br/>findings and extend previous work, demonstrating that +<br/>sequential dependence in perception operates across +<br/>different levels of visual analysis, even at the highest +<br/>levels of perceptual interpretation. +<br/>Introduction +<br/>Humans make aesthetic judgments all the time about +<br/>the attractiveness or desirability of objects and scenes. +<br/>Aesthetic judgments are not merely about judging +<br/>works of art; they are constantly involved in our daily +<br/>activity, influencing or determining our choices of food, +<br/>object (Creusen & Schoormans, 2005), and mate +<br/>preference (Rhodes, Simmons, & Peters, 2005). +<br/>Aesthetic judgments are based on perceptual pro- +<br/>cessing (Arnheim, 1954; Livingstone & Hubel, 2002; +<br/>Solso, 1996). These judgments, like other perceptual +<br/>experiences, are thought to be relatively stable in spite +<br/>of fluctuations in the raw visual input we receive due to +<br/>factors like occlusion, visual noise, and eye movements. +<br/>One mechanism that allows the visual system to achieve +<br/>this stability is serial dependence. Recent results have +<br/>revealed that the perception of visual features such as +<br/>orientation (Fischer & Whitney, 2014), numerosity +<br/>(Cicchini, Anobile, & Burr, 2014), and facial identity +<br/>(Liberman, Fischer, & Whitney, 2014) are systemati- +<br/>cally assimilated toward visual input from the recent +<br/>past. This perceptual pull has been distinguished from +<br/>hysteresis in motor responses or decision processes, and +<br/>has been shown to be tuned by the magnitude of the +<br/>difference between previous and current visual inputs +<br/>(Fischer & Whitney, 2014; Liberman, Fischer, & +<br/>Whitney, 2014). +<br/>Is aesthetics perception similarly stable like feature +<br/>perception? Some previous studies have suggested that +<br/>the answer is yes. It has been shown that there is a +<br/>positive correlation between observers’ successive +<br/>attractiveness ratings of facial images (Kondo, Taka- +<br/>hashi, & Watanabe, 2012; Taubert, Van der Burg, & +<br/>Alais, 2016). This suggests that there is an assimilative +<br/>sequential dependence in attractiveness judgments. +<br/>Citation: Xia, Y., Leib, A. Y., & Whitney, D. (2016). Serial dependence in the perception of attractiveness. Journal of Vision, +<br/>16(15):28, 1–8, doi:10.1167/16.15.28. +<br/>doi: 10 .116 7 /1 6. 15 . 28 +<br/>Received July 13, 2016; published December 22, 2016 +<br/>ISSN 1534-7362 +<br/>This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. +</td><td>('37397364', 'Ye Xia', 'ye xia')<br/>('6931574', 'Allison Yamanashi Leib', 'allison yamanashi leib')<br/>('1821337', 'David Whitney', 'david whitney')</td><td></td></tr><tr><td>58081cb20d397ce80f638d38ed80b3384af76869</td><td>Embedded Real-Time Fall Detection Using Deep +<br/>Learning For Elderly Care +<br/>Samsung Research, Samsung Electronics +</td><td>('1729858', 'Hyunwoo Lee', 'hyunwoo lee')<br/>('1784186', 'Jooyoung Kim', 'jooyoung kim')<br/>('32671800', 'Dojun Yang', 'dojun yang')<br/>('3443235', 'Joon-Ho Kim', 'joon-ho kim')</td><td>{hyun0772.lee, joody.kim, dojun.yang, mythos.kim}@samsung.com +</td></tr><tr><td>581e920ddb6ecfc2a313a3aa6fed3d933b917ab0</td><td>Automatic Mapping of Remote Crowd Gaze to +<br/>Stimuli in the Classroom +<br/><b>University of T ubingen, T ubingen, Germany</b><br/>2 Leibniz-Institut f¨ur Wissensmedien, T¨ubingen, Germany +<br/><b>Hector Research Institute of Education Sciences and Psychology, T ubingen</b><br/>Germany +</td><td>('2445102', 'Thiago Santini', 'thiago santini')<br/>('24003697', 'Lucas Draghetti', 'lucas draghetti')<br/>('3286609', 'Peter Gerjets', 'peter gerjets')<br/>('2446461', 'Ulrich Trautwein', 'ulrich trautwein')<br/>('1884159', 'Enkelejda Kasneci', 'enkelejda kasneci')</td><td></td></tr><tr><td>58fa85ed57e661df93ca4cdb27d210afe5d2cdcd</td><td>Cancún Center, Cancún, México, December 4-8, 2016 +<br/>978-1-5090-4847-2/16/$31.00 ©2016 IEEE +<br/>4118 +</td><td></td><td></td></tr><tr><td>5860cf0f24f2ec3f8cbc39292976eed52ba2eafd</td><td>International Journal of Automated Identification Technology, 3(2), July-December 2011, pp. 51-60 +<br/>COMPUTATION EvaBio: A TOOL FOR PERFORMANCE +<br/>EVALUATION IN BIOMETRICS +<br/><b>GREYC Laboratory, ENSICAEN - University of Caen Basse Normandie - CNRS</b><br/> 6 Boulevard Maréchal Juin, 14000 Caen Cedex - France +</td><td>('2774452', 'Julien Mahier', 'julien mahier')<br/>('3356614', 'Baptiste Hemery', 'baptiste hemery')<br/>('2174941', 'Mohamad El-Abed', 'mohamad el-abed')<br/>('1793765', 'Christophe Rosenberger', 'christophe rosenberger')</td><td></td></tr><tr><td>584909d2220b52c0d037e8761d80cb22f516773f</td><td>OCR-Free Transcript Alignment +<br/>Dept. of Mathematics and Computer Science +<br/>School of Computer Science +<br/>School of Computer Science +<br/><b>The Open University</b><br/>Israel +<br/><b>Tel Aviv University</b><br/>Tel-Aviv, Israel +<br/><b>Tel Aviv University</b><br/>Tel-Aviv, Israel +</td><td>('1756099', 'Tal Hassner', 'tal hassner')<br/>('1776343', 'Lior Wolf', 'lior wolf')<br/>('1759551', 'Nachum Dershowitz', 'nachum dershowitz')</td><td>Email: hassner@openu.ac.il +<br/>Email: wolf@cs.tau.ac.il +<br/>Email: nachumd@tau.ac.il +</td></tr><tr><td>58bf72750a8f5100e0c01e55fd1b959b31e7dbce</td><td>PyramidBox: A Context-assisted Single Shot +<br/>Face Detector. +<br/>Baidu Inc. +</td><td>('48785141', 'Xu Tang', 'xu tang')<br/>('14931829', 'Daniel K. Du', 'daniel k. du')<br/>('31239588', 'Zeqiang He', 'zeqiang he')<br/>('2272123', 'Jingtuo Liu', 'jingtuo liu')</td><td>tangxu02@baidu.com,daniel.kang.du@gmail.com,{hezeqiang,liujingtuo}@baidu.com +</td></tr><tr><td>58542eeef9317ffab9b155579256d11efb4610f2</td><td>International Journal of Science and Research (IJSR) +<br/>ISSN (Online): 2319-7064 +<br/>Index Copernicus Value (2013): 6.14 | Impact Factor (2014): 5.611 +<br/>Face Recognition Revisited on Pose, Alignment, +<br/>Color, Illumination and Expression-PyTen +<br/>Computer Science, BIT Noida, India +</td><td></td><td></td></tr><tr><td>58823377757e7dc92f3b70a973be697651089756</td><td>Technical Report +<br/>UCAM-CL-TR-861 +<br/>ISSN 1476-2986 +<br/>Number 861 +<br/>Computer Laboratory +<br/>Automatic facial expression analysis +<br/>October 2014 +<br/>15 JJ Thomson Avenue +<br/>Cambridge CB3 0FD +<br/>United Kingdom +<br/>phone +44 1223 763500 +<br/>http://www.cl.cam.ac.uk/ +</td><td>('1756344', 'Tadas Baltrusaitis', 'tadas baltrusaitis')</td><td></td></tr><tr><td>580e48d3e7fe1ae0ceed2137976139852b1755df</td><td>THE EFFECTS OF MOTION AND ORIENTATION ON PERCEPTION OF +<br/>FACIAL EXPRESSIONS AND FACE RECOGNITION +<br/>by +<br/><b>B.S. University of Indonesia</b><br/><b>M.S. Brunel University of West London</b><br/>Submitted to the Graduate Faculty of +<br/>Arts and Sciences in partial fulfillment +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/><b>University of Pittsburgh</b><br/>2002 +</td><td>('2059653', 'Zara Ambadar', 'zara ambadar')</td><td></td></tr><tr><td>5865e824e3d8560e07840dd5f75cfe9bf68f9d96</td><td>RESEARCH ARTICLE +<br/>Embodied conversational agents for +<br/>multimodal automated social skills training in +<br/>people with autism spectrum disorders +<br/><b>Graduate School of Information Science, Nara Institute of Science and Technology, Ikoma-shi, Nara</b><br/><b>Japan, 2 Center for Special Needs Education, Nara University of Education, Nara-shi, Nara</b><br/>Japan, 3 Developmental Center for Child and Adult, Shigisan Hospital, Ikoma-gun, Nara, 636-0815, Japan +</td><td>('3162048', 'Hiroki Tanaka', 'hiroki tanaka')<br/>('1867578', 'Hideki Negoro', 'hideki negoro')<br/>('35238212', 'Hidemi Iwasaka', 'hidemi iwasaka')<br/>('40285672', 'Satoshi Nakamura', 'satoshi nakamura')</td><td>* hiroki-tan@is.naist.jp +</td></tr><tr><td>58bb77dff5f6ee0fb5ab7f5079a5e788276184cc</td><td>Facial Expression Recognition with PCA and LBP +<br/>Features Extracting from Active Facial Patches +<br/> +</td><td>('7895427', 'Yanpeng Liu', 'yanpeng liu')<br/>('16879896', 'Yuwen Cao', 'yuwen cao')<br/>('29275442', 'Yibin Li', 'yibin li')<br/>('1686211', 'Ming Liu', 'ming liu')<br/>('1772484', 'Rui Song', 'rui song')<br/>('1706513', 'Yafang Wang', 'yafang wang')<br/>('40395865', 'Zhigang Xu', 'zhigang xu')<br/>('1708045', 'Xin Ma', 'xin ma')</td><td></td></tr><tr><td>585260468d023ffc95f0e539c3fa87254c28510b</td><td>Cardea: Context–Aware Visual Privacy Protection +<br/>from Pervasive Cameras +<br/>HKUST-DT System and Media Laboratory +<br/><b>Hong Kong University of Science and Technology, Hong Kong</b></td><td>('3432205', 'Jiayu Shu', 'jiayu shu')<br/>('2844817', 'Rui Zheng', 'rui zheng')<br/>('2119751', 'Pan Hui', 'pan hui')</td><td>Email: ∗jshuaa@ust.hk, †rzhengac@ust.hk, ‡panhui@ust.hk +</td></tr><tr><td>58cb1414095f5eb6a8c6843326a6653403a0ee17</td><td></td><td></td><td></td></tr><tr><td>58db008b204d0c3c6744f280e8367b4057173259</td><td>International Journal of Current Engineering and Technology +<br/>ISSN 2277 - 4106 +<br/> © 2012 INPRESSCO. All Rights Reserved. +<br/>Available at http://inpressco.com/category/ijcet +<br/>Research Article +<br/>Facial Expression Recognition +<br/><b>Jaipur, Rajasthan, India</b><br/>Accepted 3June 2012, Available online 8 June 2012 +</td><td>('40621542', 'Riti Kushwaha', 'riti kushwaha')<br/>('2117075', 'Neeta Nain', 'neeta nain')</td><td></td></tr><tr><td>58628e64e61bd2776a2a7258012eabe3c79ca90c</td><td>Active Grounding of Visual Situations +<br/><b>Portland State University</b><br/><b>Santa Fe Institute</b><br/>Unpublished Draft +</td><td>('3438473', 'Max H. Quinn', 'max h. quinn')<br/>('27572284', 'Erik Conser', 'erik conser')<br/>('38388831', 'Jordan M. Witte', 'jordan m. witte')<br/>('4421478', 'Melanie Mitchell', 'melanie mitchell')</td><td></td></tr><tr><td>676a136f5978783f75b5edbb38e8bb588e8efbbe</td><td>Matrix Completion for Resolving Label Ambiguity +<br/><b>UMIACS, University of Maryland, College Park, USA</b><br/>Learning a visual classifier requires a large amount of labeled images +<br/>and videos. However, labeling images is expensive and time-consuming +<br/>due to the significant amount of human efforts involved. As a result, brief +<br/>descriptions such as tags, captions and screenplays accompanying the im- +<br/>ages and videos become important for training classifiers. Although such +<br/>information is publicly available, it is not as explicitly labeled as human +<br/>annotation. For instance, names in the caption of a news photo provide +<br/>possible candidates for faces appearing in the image [1]. The names in the +<br/>screenplays are only weakly associated with faces in the shots [4]. The prob- +<br/>lem in which instead of a single label per instance, one is given a candidate +<br/>set of labels, of which only one is correct is known as ambiguously labeled +<br/>learning [2, 6]. +<br/>Ambiguous Labels +<br/>Disambiguated Labels +<br/>Class 2 +<br/>MCar +<br/>Class 1 +<br/>L={1} +<br/>L={2} +<br/>L={3} +<br/>L={1, 2} +<br/>L={2, 3} +<br/>L={1, 3} +<br/>Class 3 +<br/>The ambiguously labeled data is denoted as L = {(x j , L j), j = 1, 2, . . . , N}, +<br/>Figure 1: MCar reassigns the labels for those ambiguously labeled in- +<br/>stances such that instances of the same subjects cohesively form potentially- +<br/>separable convex hulls. +<br/>where N is the number of instances. There are c classes, and the class labels +<br/>are denoted as Y = {1, 2, . . . , c}. Note that x j is the feature vector of the jth +<br/>instance, and its ambiguous labeling set L j ⊆ Y consists of the candidate +<br/>labels associated with the jth instance. The true label of the jth instance is +<br/>l j ∈ L j. In other words, one of the labels in L j is the true label of x j. The +<br/>objective is to resolve the ambiguity in L such that each predicted label ˆl j +<br/>of x j matches its true label l j. +<br/>We interpret the ambiguous labeling set L j with soft labeling vector p j, +<br/>where pi, j indicates the probability that instance j belongs to class i. This +<br/>allows us to quantitatively assign the likelihood of each class the instance +<br/>belongs to if such information is provided. Without any prior knowledge, +<br/>we assume equal probability for each candidate label. Let P ∈ Rc×N denotes +<br/>the ambiguous labeling matrix with p j in its jth column. With this, one can +<br/>model the ambiguous labeling as P = P0 + EP, where P0 and EP denote the +<br/>true labeling matrix and the labeling noise, respectively. The jth column +<br/>vector of P0 is p0 +<br/>j = el j , where el j is the canonical vector corresponding to +<br/>the 1-of-K coding of its true label l j. Similarly, assuming that the feature +<br/>vectors are corrupted by some noise or occlusion, the feature matrix X with +<br/>x j in its jth column can be modeled as X = X0 + EX , where X ∈ Rm×N con- +<br/>sists of N feature vectors of dimension m, X0 represents the feature matrix +<br/>in the absence of noise and EX accounts for the noise. +<br/>Figure 1 shows the geometric interpretation of our proposed method, +<br/>Matrix Completion for Ambiguity Resolving (MCar). When each element +<br/>in the ambiguous labeling set is trivially treated as the true label, the convex +<br/>hulls of each class are erroneously expanded. MCar reassigns the ambiguous +<br/>labels such that each over-expanded convex hull shrinks to its actual contour, +<br/>and the convex hulls becomes potentially separable. +<br/>In the paper, we show that the heterogeneous feature matrix, which is +<br/>the concatenation of the labeling matrix P and feature matrix X, is ideally +<br/>low-rank in the absence of noise (Figure 2), which allows us to convert the +<br/>aforementioned label reassignment problem as a matrix completion prob- +<br/>lem [5]. The proposed MCar takes the heterogeneous feature matrix as in- +<br/>put, and returns the predicted labeling matrix Y by solving the following +<br/>optimization problem +<br/>= +<br/>= +<br/>+ +<br/>+ +<br/> ۾ +<br/> ܆ +<br/> ۾ +<br/> ܆ +<br/> +<br/>۳ +<br/>۳ +<br/>Figure 2: Ideal decomposition of heterogeneous feature matrix using MCar. +<br/>The underlying low-rank structure and the ambiguous labeling are recovered +<br/>simultaneously. +<br/>The proposed method inherits the benefit of low-rank recovery and pos- +<br/>sesses the capability to resolve the label ambiguity via low-rank approxima- +<br/>tion of the heterogeneous matrix. As a result, our method is more robust +<br/>compared to some of the existing discriminative ambiguous learning meth- +<br/>ods [3, 7], sparsity/dictionary-based method [2], and low-rank representation- +<br/>based method [8]. Moreover, we generalize MCar to include the labeling +<br/>constraints between the instances for practical applications. Compared to +<br/>the state of the arts, our proposed framework achieves 2.9% improvement +<br/>on the labeling accuracy of the Lost dataset and performs comparably on the +<br/>Labeled Yahoo! News dataset. +<br/>[1] T. L. Berg, A. C. Berg, J. Edwards, M. Maire, R. White, Y.-W. Teh, +<br/>E. Learned-Miller, and D. A. Forsyth. Names and faces in the news. In +<br/>CVPR, 2004. +<br/>[2] Y.-C. Chen, V. M. Patel, J. K. Pillai, R. Chellappa, and P. J. Phillips. +<br/>Dictionary learning from ambiguously labeled data. In CVPR, 2013. +<br/>[3] T. Cour, B. Sapp, C. Jordan, and B. Taskar. Learning from ambiguously +<br/>labeled images. In CVPR, 2009. +<br/>[4] M. Everingham, J. Sivic, and A. Zisserman. Hello! My name is... Buffy +<br/>(1) +<br/>- Automatic naming of characters in TV video. In BMVC, 2006. +<br/>[5] A. B. Goldberg, X. Zhu, B. Recht, J.-M. Xu, and R. D. Nowak. Trans- +<br/>duction with matrix completion: Three birds with one stone. In NIPS, +<br/>2010. +<br/>[6] E. Hüllermeier and J. Beringer. Learning from ambiguously labeled +<br/>examples. In Intell. Data Anal., 2006. +<br/>[7] J. Luo and F. Orabona. Learning from candidate labeling sets. In NIPS, +<br/>2010. +<br/>min +<br/>Y,EX +<br/>rank(H) + λ kEX k0 + γkYk0 +<br/>Z(cid:21) =(cid:20)P +<br/>X(cid:21) −(cid:20)EP +<br/>EX(cid:21) , +<br/>N , Y ∈ Rc×N +<br/>+ , +<br/>s.t. H =(cid:20)Y +<br/>1T +<br/>c Y = 1T +<br/>yi, j = 0 if pi, j = 0, +<br/>where λ ∈ R+ and γ ∈ R+ control the sparsity of data noise and predicted +<br/>labeling matrix, respectively. Consequently, the predicted label of instance +<br/>j can be obtained as +<br/>ˆl j = arg max +<br/>i∈Y +<br/>yi, j . +<br/>(2) +</td><td>('2682056', 'Ching-Hui Chen', 'ching-hui chen')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>677585ccf8619ec2330b7f2d2b589a37146ffad7</td><td>A flexible model for training action localization +<br/>with varying levels of supervision +</td><td>('1902524', 'Guilhem Chéron', 'guilhem chéron')<br/>('2285263', 'Jean-Baptiste Alayrac', 'jean-baptiste alayrac')<br/>('1785596', 'Ivan Laptev', 'ivan laptev')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>676f9eabf4cfc1fd625228c83ff72f6499c67926</td><td>FACE IDENTIFICATION AND CLUSTERING +<br/>A thesis submitted to the +<br/>Graduate School—New Brunswick +<br/><b>Rutgers, The State University of New Jersey</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>Master of Science +<br/>Graduate Program in Computer Science +<br/>Written under the direction of +<br/>Dr. Vishal Patel, Dr. Ahmed Elgammal +<br/>and approved by +<br/>New Brunswick, New Jersey +<br/>May, 2017 +</td><td>('34805991', 'Atul Dhingra', 'atul dhingra')</td><td></td></tr><tr><td>677477e6d2ba5b99633aee3d60e77026fb0b9306</td><td></td><td></td><td></td></tr><tr><td>6789bddbabf234f31df992a3356b36a47451efc7</td><td>Unsupervised Generation of Free-Form and +<br/>Parameterized Avatars +</td><td>('33964593', 'Adam Polyak', 'adam polyak')<br/>('2188620', 'Yaniv Taigman', 'yaniv taigman')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td></td></tr><tr><td>679b7fa9e74b2aa7892eaea580def6ed4332a228</td><td>Communication and automatic +<br/>interpretation of affect from facial +<br/>expressions1 +<br/><b>University of Amsterdam, the Netherlands</b><br/><b>University of Trento, Italy</b><br/><b>University of Amsterdam, the Netherlands</b></td><td>('1764521', 'Albert Ali Salah', 'albert ali salah')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td></td></tr><tr><td>675b2caee111cb6aa7404b4d6aa371314bf0e647</td><td>AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions +<br/>Carl Vondrick∗ +</td><td>('39599498', 'Chunhui Gu', 'chunhui gu')<br/>('1758054', 'Yeqing Li', 'yeqing li')<br/>('1726241', 'Chen Sun', 'chen sun')<br/>('48536531', 'David A. Ross', 'david a. ross')<br/>('2259154', 'Sudheendra Vijayanarasimhan', 'sudheendra vijayanarasimhan')<br/>('1805076', 'George Toderici', 'george toderici')<br/>('2997956', 'Caroline Pantofaru', 'caroline pantofaru')<br/>('2262946', 'Susanna Ricco', 'susanna ricco')<br/>('1694199', 'Rahul Sukthankar', 'rahul sukthankar')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')<br/>('1689212', 'Jitendra Malik', 'jitendra malik')</td><td></td></tr><tr><td>679b72d23a9cfca8a7fe14f1d488363f2139265f</td><td></td><td></td><td></td></tr><tr><td>67484723e0c2cbeb936b2e863710385bdc7d5368</td><td>Anchor Cascade for Efficient Face Detection +</td><td>('2425630', 'Baosheng Yu', 'baosheng yu')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td></td></tr><tr><td>670637d0303a863c1548d5b19f705860a23e285c</td><td>Face Swapping: Automatically Replacing Faces in Photographs +<br/><b>Columbia University</b><br/>Peter Belhumeur +<br/>Figure 1: We have developed a system that automatically replaces faces in an input image with ones selected from a large collection of +<br/>face images, obtained by applying face detection to publicly available photographs on the internet. In this example, the faces of (a) two +<br/>people are shown after (b) automatic replacement with the top three ranked candidates. Our system for face replacement can be used for face +<br/>de-identification, personalized face replacement, and creating an appealing group photograph from a set of “burst” mode images. Original +<br/>images in (a) used with permission from Retna Ltd. (top) and Getty Images Inc. (bottom). +<br/>Rendering, Computational Photography +<br/>1 Introduction +<br/>it +<br/>Advances in digital photography have made it possible to cap- +<br/>ture large collections of high-resolution images and share them +<br/>on the internet. While the size and availability of these col- +<br/>lections is leading to many exciting new applications, +<br/>is +<br/>also creating new problems. One of the most +<br/>important of +<br/>these problems is privacy. Online systems such as Google +<br/>Street View (http://maps.google.com/help/maps/streetview) and +<br/>EveryScape (http://everyscape.com) allow users to interactively +<br/>navigate through panoramic images of public places created using +<br/>thousands of photographs. Many of the images contain people who +<br/>have not consented to be photographed, much less to have these +<br/>photographs publicly viewable. Identity protection by obfuscating +<br/>the face regions in the acquired photographs using blurring, pixela- +<br/>tion, or simply covering them with black pixels is often undesirable +<br/>as it diminishes the visual appeal of the image. Furthermore, many +</td><td>('2085183', 'Dmitri Bitouk', 'dmitri bitouk')<br/>('40631426', 'Neeraj Kumar', 'neeraj kumar')<br/>('2057606', 'Samreen Dhillon', 'samreen dhillon')<br/>('1750470', 'Shree K. Nayar', 'shree k. nayar')</td><td></td></tr><tr><td>6742c0a26315d7354ab6b1fa62a5fffaea06da14</td><td>BAS AND SMITH: WHAT DOES 2D GEOMETRIC INFORMATION REALLY TELL US ABOUT 3D FACE SHAPE? +<br/>What does 2D geometric information +<br/>really tell us about 3D face shape? +</td><td>('39180407', 'Anil Bas', 'anil bas')<br/>('1687021', 'William A. P. Smith', 'william a. p. smith')</td><td></td></tr><tr><td>67a50752358d5d287c2b55e7a45cc39be47bf7d0</td><td></td><td></td><td></td></tr><tr><td>67c3c1194ee72c54bc011b5768e153a035068c43</td><td>StreetScenes: Towards Scene Understanding in +<br/>Still Images +<br/>by +<br/>Stanley Michael Bileschi +<br/>Submitted to the Department of Electrical Engineering and Computer +<br/>Science +<br/>in partial fulflllment of the requirements for the degree of +<br/>Doctor of Philosophy in Computer Science and Engineering +<br/>at the +<br/><b>MASSACHUSETTS INSTITUTE OF TECHNOLOGY</b><br/>May 2006 +<br/><b>c(cid:176) Massachusetts Institute of Technology 2006. All rights reserved</b><br/>Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Department of Electrical Engineering and Computer Science +<br/>May 5, 2006 +<br/>Certifled by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Tomaso A. Poggio +<br/>McDermott Professor +<br/>Thesis Supervisor +<br/>Accepted by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Arthur C. Smith +<br/>Chairman, Department Committee on Graduate Students +</td><td></td><td></td></tr><tr><td>673d4885370b27c863e11a4ece9189a6a45931cc</td><td>Recurrent Residual Module for Fast Inference in Videos +<br/><b>Shanghai Jiao Tong University, 2Zhejiang University, 3Massachusetts Institute of Technology</b><br/>networks for video recognition are more challenging. For +<br/>example, for Youtube-8M dataset [1] with over 8 million +<br/>video clips, it will take 50 years for a CPU to extract the +<br/>deep features using a standard CNN model. +</td><td>('35654996', 'Bowen Pan', 'bowen pan')<br/>('35992009', 'Wuwei Lin', 'wuwei lin')<br/>('2126444', 'Xiaolin Fang', 'xiaolin fang')<br/>('35933894', 'Chaoqin Huang', 'chaoqin huang')<br/>('1804424', 'Bolei Zhou', 'bolei zhou')<br/>('1830034', 'Cewu Lu', 'cewu lu')</td><td>†{googletornado,linwuwei13, huangchaoqin}@sjtu.edu.cn, ¶fxlfang@gmail.com +<br/>§bzhou@csail.mit.edu; ‡lu-cw@cs.sjtu.edu.cn +</td></tr><tr><td>67c703a864aab47eba80b94d1935e6d244e00bcb</td><td> (IJACSA) International Journal of Advanced Computer Science and Applications +<br/>Vol. 7, No. 6, 2016 +<br/>Face Retrieval Based On Local Binary Pattern and Its +<br/>Variants: A Comprehensive Study +<br/><b>University of Science, VNU-HCM, Viet Nam</b><br/>face searching, +</td><td>('3911040', 'Phan Khoi', 'phan khoi')</td><td></td></tr><tr><td>6754c98ba73651f69525c770fb0705a1fae78eb5</td><td>Joint Cascade Face Detection and Alignment +<br/><b>University of Science and Technology of China</b><br/>2 Microsoft Research +</td><td>('39447786', 'Dong Chen', 'dong chen')<br/>('3080683', 'Shaoqing Ren', 'shaoqing ren')<br/>('1732264', 'Yichen Wei', 'yichen wei')<br/>('47300766', 'Xudong Cao', 'xudong cao')<br/>('40055995', 'Jian Sun', 'jian sun')</td><td>{chendong,sqren}@mail.ustc.edu.cn +<br/>{yichenw,xudongca,jiansun}@microsoft.com +</td></tr><tr><td>672fae3da801b2a0d2bad65afdbbbf1b2320623e</td><td>Pose-Selective Max Pooling for Measuring Similarity +<br/>1Dept. of Computer Science +<br/>2Dept. of Electrical & Computer Engineering +<br/><b>Johns Hopkins University, 3400 N. Charles St, Baltimore, MD 21218, USA</b></td><td>('40031188', 'Xiang Xiang', 'xiang xiang')<br/>('1709073', 'Trac D. Tran', 'trac d. tran')</td><td>xxiang@cs.jhu.edu +</td></tr><tr><td>677ebde61ba3936b805357e27fce06c44513a455</td><td>Facial Expression Recognition Based on Facial +<br/>Components Detection and HOG Features +<br/><b>The Hong Kong Polytechnic University, Hong Kong</b><br/><b>Chu Hai College of Higher Education, Hong Kong</b></td><td>('2366262', 'Junkai Chen', 'junkai chen')<br/>('1715231', 'Zenghai Chen', 'zenghai chen')<br/>('8590720', 'Zheru Chi', 'zheru chi')<br/>('1965426', 'Hong Fu', 'hong fu')</td><td>Email: Junkai.Chen@connect.polyu.hk +</td></tr><tr><td>67ba3524e135c1375c74fe53ebb03684754aae56</td><td>978-1-5090-4117-6/17/$31.00 ©2017 IEEE +<br/>1767 +<br/>ICASSP 2017 +</td><td></td><td></td></tr><tr><td>6769cfbd85329e4815bb1332b118b01119975a95</td><td>Tied factor analysis for face recognition across +<br/>large pose changes +</td><td></td><td></td></tr><tr><td>0be43cf4299ce2067a0435798ef4ca2fbd255901</td><td>Title +<br/>A temporal latent topic model for facial expression recognition +<br/>Author(s) +<br/>Shang, L; Chan, KP +<br/>Citation +<br/>The 10th Asian Conference on Computer Vision (ACCV 2010), +<br/>Queenstown, New Zealand, 8-12 November 2010. In Lecture +<br/>Notes in Computer Science, 2010, v. 6495, p. 51-63 +<br/>Issued Date +<br/>2011 +<br/>URL +<br/>http://hdl.handle.net/10722/142604 +<br/>Rights +<br/>Creative Commons: Attribution 3.0 Hong Kong License +</td><td></td><td></td></tr><tr><td>0bc53b338c52fc635687b7a6c1e7c2b7191f42e5</td><td>ZHANG, BHALERAO: LOGLET SIFT FOR PART DESCRIPTION +<br/>Loglet SIFT for Part Description in +<br/>Deformable Part Models: Application to Face +<br/>Alignment +<br/>Department of Computer Science +<br/><b>University of Warwick</b><br/>Coventry, UK +</td><td>('39900385', 'Qiang Zhang', 'qiang zhang')<br/>('2227351', 'Abhir Bhalerao', 'abhir bhalerao')</td><td>q.zhang.13@warwick.ac.uk +<br/>abhir.bhalerao@warwick.ac.uk +</td></tr><tr><td>0b2277a0609565c30a8ee3e7e193ce7f79ab48b0</td><td>944 +<br/>Cost-Sensitive Semi-Supervised Discriminant +<br/>Analysis for Face Recognition +</td><td>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('3353607', 'Xiuzhuang Zhou', 'xiuzhuang zhou')<br/>('1689805', 'Yap-Peng Tan', 'yap-peng tan')<br/>('38152390', 'Yuanyuan Shang', 'yuanyuan shang')<br/>('39491387', 'Jie Zhou', 'jie zhou')</td><td></td></tr><tr><td>0b9ce839b3c77762fff947e60a0eb7ebbf261e84</td><td>Proceedings of the IASTED International Conference +<br/>Computer Vision (CV 2011) +<br/>June 1 - 3, 2011 Vancouver, BC, Canada +<br/>LOGARITHMIC FOURIER PCA: A NEW APPROACH TO FACE +<br/>RECOGNITION +<br/>1 Lakshmiprabha Nattamai Sekar, +<br/>omjyoti +<br/>Majumder +<br/>Surface Robotics Lab +<br/><b>Central Mechanical Engineering Research Institute</b><br/>Mahatma Gandhi Avenue, +<br/>Durgapur - 713209, West Bengal, India. +</td><td>('9155672', 'Jhilik Bhattacharya', 'jhilik bhattacharya')</td><td>email: 1 n prabha mech@cmeri.res.in, 2 bjhilik@cmeri.res.in, 3 sjm@cmeri.res.in +</td></tr><tr><td>0b8b8776684009e537b9e2c0d87dbd56708ddcb4</td><td>Adversarial Discriminative Heterogeneous Face Recognition +<br/>National Laboratory of Pattern Recognition, CASIA +<br/>Center for Research on Intelligent Perception and Computing, CASIA +<br/>Center for Excellence in Brain Science and Intelligence Technology, CAS +<br/><b>University of Chinese Academy of Sciences, Beijing 100190, China</b></td><td>('3051419', 'Lingxiao Song', 'lingxiao song')<br/>('2567523', 'Man Zhang', 'man zhang')<br/>('2225749', 'Xiang Wu', 'xiang wu')<br/>('1705643', 'Ran He', 'ran he')</td><td></td></tr><tr><td>0ba64f4157d80720883a96a73e8d6a5f5b9f1d9b</td><td></td><td></td><td></td></tr><tr><td>0b6a5200c33434cbfa9bf24ba482f6e06bf5fff7</td><td>1 +<br/>The Use of Deep Learning in Image +<br/>Segmentation, Classification and Detection +<br/><b>The Image Processing and Analysis Lab (LAPI), Politehnica University of Bucharest, Romania</b><br/> +</td><td>('33789881', 'Mihai-Sorin Badea', 'mihai-sorin badea')<br/>('3407753', 'Laura Maria Florea', 'laura maria florea')<br/>('2905899', 'Constantin Vertan', 'constantin vertan')</td><td></td></tr><tr><td>0b605b40d4fef23baa5d21ead11f522d7af1df06</td><td>Label-Embedding for Attribute-Based Classification +<br/>a Computer Vision Group∗, XRCE, France +<br/>b LEAR†, INRIA, France +</td><td>('2893664', 'Zeynep Akata', 'zeynep akata')<br/>('1723883', 'Florent Perronnin', 'florent perronnin')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>0b0eb562d7341231c3f82a65cf51943194add0bb</td><td>> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < +<br/>Facial Image Analysis Based on Local Binary +<br/>Patterns: A Survey +<br/> +</td><td>('40451093', 'Di Huang', 'di huang')<br/>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('40703561', 'Mohsen Ardebilian', 'mohsen ardebilian')<br/>('40231048', 'Liming Chen', 'liming chen')</td><td></td></tr><tr><td>0b3a146c474166bba71e645452b3a8276ac05998</td><td>Who’s in the Picture? +<br/>Berkeley, CA 94720 +<br/>Computer Science Division +<br/>U.C. Berkeley +</td><td>('1685538', 'Tamara L. Berg', 'tamara l. berg')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('34497462', 'Jaety Edwards', 'jaety edwards')</td><td>millert@cs.berkeley.edu +</td></tr><tr><td>0b78fd881d0f402fd9b773249af65819e48ad36d</td><td>ANALYSIS AND MODELING OF AFFECTIVE AUDIO VISUAL SPEECH +<br/>BASED ON PAD EMOTION SPACE +<br/><b>Tsinghua University</b></td><td>('2180849', 'Shen Zhang', 'shen zhang')<br/>('1856341', 'Yingjin Xu', 'yingjin xu')<br/>('25714033', 'Jia Jia', 'jia jia')<br/>('7239047', 'Lianhong Cai', 'lianhong cai')</td><td>{zhangshen05, xuyj03, jiajia}@mails.tsinghua.edu.cn, clh-dcs@tsinghua.edu.cn +</td></tr><tr><td>0b835284b8f1f45f87b0ce004a4ad2aca1d9e153</td><td>Cartooning for Enhanced Privacy in Lifelogging and Streaming Videos +<br/>David Crandall +<br/>School of Informatics and Computing +<br/><b>Indiana University Bloomington</b></td><td>('3053390', 'Eman T. Hassan', 'eman t. hassan')<br/>('2221434', 'Rakibul Hasan', 'rakibul hasan')<br/>('34507388', 'Patrick Shaffer', 'patrick shaffer')<br/>('1996617', 'Apu Kapadia', 'apu kapadia')</td><td>{emhassan, rakhasan, patshaff, djcran, kapadia}@indiana.edu +</td></tr><tr><td>0b5bd3ce90bf732801642b9f55a781e7de7fdde0</td><td></td><td></td><td></td></tr><tr><td>0b0958493e43ca9c131315bcfb9a171d52ecbb8a</td><td>A Unified Neural Based Model for Structured Output Problems +<br/>Soufiane Belharbi∗1, Cl´ement Chatelain∗1, Romain H´erault∗1, and S´ebastien Adam∗2 +<br/>1LITIS EA 4108, INSA de Rouen, Saint ´Etienne du Rouvray 76800, France +<br/>2LITIS EA 4108, UFR des Sciences, Universit´e de Rouen, France. +<br/>April 13, 2015 +</td><td></td><td></td></tr><tr><td>0b51197109813d921835cb9c4153b9d1e12a9b34</td><td><b>THE UNIVERSITY OF CHICAGO</b><br/>JOINTLY LEARNING MULTIPLE SIMILARITY METRICS FROM TRIPLET +<br/>CONSTRAINTS +<br/>A DISSERTATION SUBMITTED TO +<br/>THE FACULTY OF THE DIVISION OF THE PHYSICAL SCIENCES +<br/>IN CANDIDACY FOR THE DEGREE OF +<br/>MASTER OF SCIENCE +<br/>DEPARTMENT OF COMPUTER SCIENCE +<br/>BY +<br/>CHICAGO, ILLINOIS +<br/>WINTER, 2015 +</td><td>('40504838', 'LIWEN ZHANG', 'liwen zhang')</td><td></td></tr><tr><td>0bf3513d18ec37efb1d2c7934a837dabafe9d091</td><td>Robust Subspace Clustering via Thresholding Ridge Regression +<br/><b>Institute for Infocomm Research, Agency for Science, Technology and Research (A*STAR), Singapore</b><br/><b>College of Computer Science, Sichuan University, Chengdu 610065, P.R. China</b></td><td>('8249791', 'Xi Peng', 'xi peng')<br/>('9276020', 'Zhang Yi', 'zhang yi')<br/>('3134548', 'Huajin Tang', 'huajin tang')</td><td>pangsaai@gmail.com, zhangyi@scu.edu.cn, htang@i2r.a-star.edu.sg. +</td></tr><tr><td>0b20f75dbb0823766d8c7b04030670ef7147ccdd</td><td>1 +<br/>Feature selection using nearest attributes +</td><td>('1744784', 'Alex Pappachen James', 'alex pappachen james')<br/>('1697594', 'Sima Dimitrijev', 'sima dimitrijev')</td><td></td></tr><tr><td>0b5a82f8c0ee3640503ba24ef73e672d93aeebbf</td><td>On Learning 3D Face Morphable Model +<br/>from In-the-wild Images +</td><td>('1849929', 'Luan Tran', 'luan tran')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td></td></tr><tr><td>0b174d4a67805b8796bfe86cd69a967d357ba9b6</td><td> Research Journal of Recent Sciences _________________________________________________ ISSN 2277-2502 +<br/> Vol. 3(4), 56-62, April (2014) +<br/>Res.J.Recent Sci. +</td><td></td><td></td></tr><tr><td>0ba449e312894bca0d16348f3aef41ca01872383</td><td></td><td></td><td></td></tr><tr><td>0b87d91fbda61cdea79a4b4dcdcb6d579f063884</td><td>The Open Automation and Control Systems Journal, 2015, 7, 569-579 +<br/>569 +<br/>Open Access +<br/>Research on Theory and Method for Facial Expression Recognition Sys- +<br/>tem Based on Dynamic Image Sequence +<br/><b>School of Computer and Information Engineering, Nanyang Institute of Technology, Henan, Nanyang, 473000, P.R</b><br/>China +<br/><b>Henan University of Traditional Chinese Medicine, Henan, Zhengzhou, 450000, P.R. China</b></td><td>('9296838', 'Yang Xinfeng', 'yang xinfeng')<br/>('2083303', 'Jiang Shan', 'jiang shan')</td><td>Send Orders for Reprints to reprints@benthamscience.ae +</td></tr><tr><td>0be2245b2b016de1dcce75ffb3371a5e4b1e731b</td><td>On the Variants of the Self-Organizing Map That Are +<br/>Based on Order Statistics +<br/><b>Aristotle University of Thessaloniki</b><br/>Box 451, Thessaloniki 54124, Greece +</td><td>('1762248', 'Vassiliki Moschou', 'vassiliki moschou')<br/>('1711062', 'Dimitrios Ververidis', 'dimitrios ververidis')<br/>('1736143', 'Constantine Kotropoulos', 'constantine kotropoulos')</td><td>{vmoshou, jimver, costas}@aiia.csd.auth.gr +</td></tr><tr><td>0b79356e58a0df1d0efcf428d0c7c4651afa140d</td><td>Appears In: Advances in Neural Information Processing Systems , MIT Press, . +<br/>Bayesian Modeling of Facial Similarity +<br/><b>Mitsubishi Electric Research Laboratory</b><br/> +<br/>Cambridge, MA +<br/><b>Massachusettes Institute of Technology</b><br/> +<br/>Cambridge, MA +</td><td>('1780935', 'Baback Moghaddam', 'baback moghaddam')<br/>('1768120', 'Tony Jebara', 'tony jebara')<br/>('1682773', 'Alex Pentland', 'alex pentland')</td><td>baback@merl.com +<br/>fjebara,sandyg@media.mit.edu +</td></tr><tr><td>0b572a2b7052b15c8599dbb17d59ff4f02838ff7</td><td>Automatic Subspace Learning via Principal +<br/>Coefficients Embedding +</td><td>('8249791', 'Xi Peng', 'xi peng')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('1709367', 'Zhang Yi', 'zhang yi')<br/>('1680126', 'Rui Yan', 'rui yan')</td><td></td></tr><tr><td>0b85b50b6ff03a7886c702ceabad9ab8c8748fdc</td><td>http://www.journalofvision.org/content/11/3/17 +<br/>Is there a dynamic advantage for facial expressions? +<br/><b>Institute of Child Health, University College London, UK</b><br/>Laboratory of Neuromotor Physiology, Santa Lucia +<br/>Foundation, Rome, Italy +<br/>Some evidence suggests that it is easier to identify facial expressions (FEs) shown as dynamic displays than as photographs +<br/>(dynamic advantage hypothesis). Previously, this has been tested by using dynamic FEs simulated either by morphing a +<br/>neutral face into an emotional one or by computer animations. For the first time, we tested the dynamic advantage hypothesis +<br/>by using high-speed recordings of actors’ FEs. In the dynamic condition, stimuli were graded blends of two recordings +<br/>(duration: 4.18 s), each describing the unfolding of an expression from neutral to apex. In the static condition, stimuli (duration: +<br/>3 s) were blends of just the apex of the same recordings. Stimuli for both conditions were generated by linearly morphing one +<br/>expression into the other. Performance was estimated by a forced-choice task asking participants to identify which prototype +<br/>the morphed stimulus was more similar to. Identification accuracy was not different between conditions. Response times (RTs) +<br/>measured from stimulus onset were shorter for static than for dynamic stimuli. Yet, most responses to dynamic stimuli were +<br/>given before expressions reached their apex. Thus, with a threshold model, we tested whether discriminative information is +<br/>integrated more effectively in dynamic than in static conditions. We did not find any systematic difference. In short, neither +<br/>identification accuracy nor RTs supported the dynamic advantage hypothesis. +<br/>Keywords: facial expressions, dynamic advantage, emotion, identification +<br/>1–15, http://www.journalofvision.org/content/11/3/17, doi:10.1167/11.3.17. +<br/>Introduction +<br/>Research on emotion recognition has relied primarily on +<br/>static images of intense facial expressions (FEs), which— +<br/>despite being accurately identified (Ekman & Friesen, +<br/>1982)—are fairly impoverished representations of real-life +<br/>FEs. As a motor behavior determined by facial muscle +<br/>actions, expressions are intrinsically dynamic. Insofar as +<br/>detecting moment-to-moment changes in others’ affective +<br/>states is fundamental for regulating social +<br/>interactions +<br/>(Yoshikawa & Sato, 2008), visual sensitivity to the +<br/>dynamic properties of FEs might be an important aspect +<br/>of our emotion recognition abilities. +<br/>There is considerable evidence that dynamic information +<br/>is not redundant and may be beneficial for various aspect of +<br/><b>face processing, including age (Berry, 1990), sex (Hill</b><br/>Johnston, 2001; Mather & Murdoch, 1994), and identity +<br/>(Hill & Johnston, 2001; Lander, Christie, & Bruce, 1999; +<br/>see O’Toole, Roark, & Abdi, 2002 for a review) recogni- +<br/>tion. In real life, static information—such as the invariant +<br/>geometrical parameters of +<br/>features—and +<br/>dynamic information describing the contraction of the +<br/>expressive muscles are closely intertwined and contribute +<br/>jointly to the overall perception. The relative contribution +<br/>of either type of cues, which is likely to depend on the +<br/>meaning that one is asked to extract from the stimulus, is +<br/>still poorly understood. Pure motion information is suffi- +<br/>cient to recognize a person’s identity and sex (Hill & +<br/>the facial +<br/>Johnston, 2001). Other studies have shown that face +<br/>identity is better recognized from dynamic than static +<br/>displays when the stimuli are degraded (e.g., shown as +<br/>negatives, upside down, thresholded, pixilated, or blurred). +<br/>However, +<br/>the advantage disappears with unmodified +<br/>stimuli (Knight & Johnston, 1997; Lander et al., 1999). In +<br/>short, insofar as recognition of identity from complete +<br/>static images is already close to perfect, motion appears to +<br/>be beneficial only when static information is insufficient or +<br/>has been manipulated (Katsiri, 2006; O’Toole et al., 2002). +<br/>In comparison to face identity, fewer studies have +<br/>investigated the role of dynamic information in FE recog- +<br/>nition (see Katsiri, 2006, for a review). Taken together, +<br/>they seem to suggest that the process of emotion identi- +<br/>fication is facilitated when expressions are dynamic rather +<br/>than static. However, because of various methodological +<br/>issues and conceptual inconsistencies across studies, this +<br/>suggestion needs to be qualified. We can divide the avail- +<br/>able studies in three main groups. +<br/>First, there are studies showing that dynamic information +<br/>improves expression recognition in a variety of suboptimal +<br/>conditions, i.e., when static information is either unavail- +<br/>able or is only partially accessible. As in the case of +<br/>identity recognition, emotions can be inferred from +<br/>animated point-light descriptions of the faces that neglect +<br/>facial features (Bassili, 1978, 1979; see also Bruce & +<br/>Valentine, 1988). Furthermore, in various neuropsycholog- +<br/>ical and developmental conditions, there is evidence that +<br/>dynamic presentation improves emotion recognition with +<br/>doi: 10.1167/11.3.17 +<br/>Received November 18, 2010; published March 22, 2011 +<br/>ISSN 1534-7362 * ARVO +<br/>Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/jov/933483/ on 03/30/2017 </td><td>('34569930', 'Chiara Fiorentini', 'chiara fiorentini')<br/>('32709245', 'Paolo Viviani', 'paolo viviani')</td><td></td></tr><tr><td>0b84f07af44f964817675ad961def8a51406dd2e</td><td>Person Re-identification in the Wild +<br/>3USTC +<br/>4UCSD +<br/><b>University of Technology Sydney</b><br/>2UTSA +</td><td>('14904242', 'Liang Zheng', 'liang zheng')<br/>('1983351', 'Hengheng Zhang', 'hengheng zhang')<br/>('3141359', 'Shaoyan Sun', 'shaoyan sun')<br/>('1698559', 'Yi Yang', 'yi yang')<br/>('1713616', 'Qi Tian', 'qi tian')</td><td>{liangzheng06,manu.chandraker,yee.i.yang,wywqtian}@gmail.com +</td></tr><tr><td>0b242d5123f79defd5f775d49d8a7047ad3153bc</td><td>CBMM Memo No. 36 +<br/>September 15, 2015 +<br/>How Important is Weight Symmetry in +<br/>Backpropagation? +<br/>by +<br/><b>Center for Brains, Minds and Machines, McGovern Institute, MIT</b></td><td>('1694846', 'Qianli Liao', 'qianli liao')<br/>('1700356', 'Joel Z. Leibo', 'joel z. leibo')</td><td></td></tr><tr><td>0ba1d855cd38b6a2c52860ae4d1a85198b304be4</td><td>Variable-state Latent Conditional Random Fields +<br/>for Facial Expression Recognition and Action Unit Detection +<br/><b>Imperial College London, UK</b><br/><b>Rutgers University, USA</b></td><td>('2616466', 'Robert Walecki', 'robert walecki')<br/>('1729713', 'Ognjen Rudovic', 'ognjen rudovic')<br/>('1736042', 'Vladimir Pavlovic', 'vladimir pavlovic')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>0b50e223ad4d9465bb92dbf17a7b79eccdb997fb</td><td>Implicit Elastic Matching with Random Projections for Pose-Variant Face +<br/>Recognition +<br/>Electrical and Computer Engineering +<br/><b>University of Illinois at Urbana-Champaign</b><br/>Microsoft Live Labs Research +</td><td>('1738310', 'John Wright', 'john wright')<br/>('1745420', 'Gang Hua', 'gang hua')</td><td>ganghua@microsoft.com +<br/>jnwright@uiuc.edu +</td></tr><tr><td>0badf61e8d3b26a0d8b60fe94ba5c606718daf0b</td><td>Rev. Téc. Ing. Univ. Zulia. Vol. 39, Nº 2, 384 - 392, 2016 +<br/>Facial Expression Recognition Using Deep Belief Network +<br/><b>School of Information Science and Technology, Northwestern University, Xi an710127, Shanxi, China</b><br/><b>Teaching Affairs Office, Chongqing Normal University, Chongqing 401331, China</b><br/><b>School of Information Science and Technology, Northwestern University, Xi an710127, Shanxi, China</b><br/><b>School of Computer and Information Science, Chongqing Normal University 401331, China</b><br/>Deli Zhu +</td><td>('3439338', 'Yunong Yang', 'yunong yang')<br/>('2068791', 'Dingyi Fang', 'dingyi fang')</td><td></td></tr><tr><td>0b02bfa5f3a238716a83aebceb0e75d22c549975</td><td>Learning Probabilistic Models for Recognizing Faces +<br/>under Pose Variations +<br/><b>Computer vision and Remote Sensing, Berlin university of Technology</b><br/>Sekr. FR-3-1, Franklinstr. 28/29, Berlin, Germany +</td><td>('2326207', 'M. Saquib', 'm. saquib')<br/>('2962236', 'Olaf Hellwich', 'olaf hellwich')</td><td>{saquib;hellwich}@fpk.tu-berlin.de +</td></tr><tr><td>0bce54bfbd8119c73eb431559fc6ffbba741e6aa</td><td>Published as a conference paper at ICLR 2018 +<br/>SKIP RNN: LEARNING TO SKIP STATE UPDATES IN +<br/>RECURRENT NEURAL NETWORKS +<br/>†Barcelona Supercomputing Center, ‡Google Inc, +<br/><b>Universitat Polit`ecnica de Catalunya, Columbia University</b></td><td>('2447185', 'Brendan Jou', 'brendan jou')<br/>('1711068', 'Jordi Torres', 'jordi torres')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>{victor.campos, jordi.torres}@bsc.es, bjou@google.com, +<br/>xavier.giro@upc.edu, shih.fu.chang@columbia.edu +</td></tr><tr><td>0b2966101fa617b90510e145ed52226e79351072</td><td>Beyond Verbs: Understanding Actions in Videos +<br/>with Text +<br/>Department of Computer Science +<br/><b>University of Manitoba</b><br/>Winnipeg, MB, Canada +<br/>Department of Computer Science +<br/><b>University of Manitoba</b><br/>Winnipeg, MB, Canada +</td><td>('3056962', 'Shujon Naha', 'shujon naha')<br/>('2295608', 'Yang Wang', 'yang wang')</td><td>Email: shujon@cs.umanitoba.ca +<br/>Email: ywang@cs.umanitoba.ca +</td></tr><tr><td>0ba0f000baf877bc00a9e144b88fa6d373db2708</td><td>Facial Expression Recognition Based on Local +<br/>Directional Pattern Using SVM Decision-level Fusion +<br/>1. Key Laboratory of Education Informalization for Nationalities, Ministry of +<br/><b>Education, Yunnan NormalUniversity, Kunming, China2. College of Information, Yunnan</b><br/><b>Normal University, Kunming, China</b></td><td>('2535958', 'Juxiang Zhou', 'juxiang zhou')<br/>('3305175', 'Tianwei Xu', 'tianwei xu')<br/>('2411704', 'Jianhou Gan', 'jianhou gan')</td><td>{zjuxiang@126.com,xutianwei@ynnu.edu.cn,kmganjh@yahoo.com.cn} +</td></tr><tr><td>0be80da851a17dd33f1e6ffdd7d90a1dc7475b96</td><td>Hindawi Publishing Corporation +<br/>Computational Intelligence and Neuroscience +<br/>Volume 2016, Article ID 7696035, 7 pages +<br/>http://dx.doi.org/10.1155/2016/7696035 +<br/>Research Article +<br/>Weighted Feature Gaussian Kernel SVM for +<br/>Emotion Recognition +<br/><b>School of Automation, Beijing University of Posts and Telecommunications, Beijing 100876, China</b><br/>Received 26 June 2016; Revised 14 August 2016; Accepted 14 September 2016 +<br/>Academic Editor: Francesco Camastra +<br/>Copyright © 2016 W. Wei and Q. Jia. This is an open access article distributed under the Creative Commons Attribution License, +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Emotion recognition with weighted feature based on facial expression is a challenging research topic and has attracted great +<br/>attention in the past few years. This paper presents a novel method, utilizing subregion recognition rate to weight kernel function. +<br/>First, we divide the facial expression image into some uniform subregions and calculate corresponding recognition rate and weight. +<br/>Then, we get a weighted feature Gaussian kernel function and construct a classifier based on Support Vector Machine (SVM). At +<br/>last, the experimental results suggest that the approach based on weighted feature Gaussian kernel function has good performance +<br/>on the correct rate in emotion recognition. The experiments on the extended Cohn-Kanade (CK+) dataset show that our method +<br/>has achieved encouraging recognition results compared to the state-of-the-art methods. +<br/>1. Introduction +<br/>Emotion recognition has necessary applications in the real +<br/>world. Its applications include but are not limited to artificial +<br/>intelligence and human computer interaction. It remains a +<br/>challenging and attractive topic. There are many methods +<br/>which have been proposed for handling problems in emotion +<br/>recognition. Speech [1, 2], physiological [3–5], and visual +<br/>signals have been explored for emotion recognition. Speech +<br/>signals are discontinuous signals, since they can be captured +<br/>only when people are talking. Acquirement of physiological +<br/>signal needs some special physiological sensors. Visual signal +<br/>is the best choice for emotion recognition based on the above +<br/>reasons. Although the visual information provided is useful, +<br/>there are challenges regarding how to utilize this information +<br/>reliably and robustly. According to Albert Mehrabian’s 7%– +<br/>38%–55% rule, facial expression is an important mean of +<br/>detecting emotions [6]. +<br/>Further studies have been carried out on emotion recog- +<br/>nition problems in facial expression images during the last +<br/>decade [7, 8]. Given a facial expression image, estimate the +<br/>correct emotional state, such as anger, happiness, sadness, and +<br/>surprise. The general process has two steps: feature extraction +<br/>and classification. For feature extraction, geometric feature, +<br/>texture feature, motion feature, and statistical feature are in +<br/>common use. For classification, methods based on machine +<br/>learning algorithm are frequently used. According to special- +<br/>ity of features, applying weighted features to machine learning +<br/>algorithm has become an active research topic. +<br/>In recent years, emotion recognition with weighted fea- +<br/>ture based on facial expression has become a new research +<br/>topic and received more and more attention [9, 10]. The +<br/>aim is to estimate emotion type from a facial expression +<br/>image captured during physical facial expression process of +<br/>a subject. But the emotion features captured from the facial +<br/>expression image are strongly linked to not the whole face +<br/>but some specific regions in the face. For instance, features +<br/>of eyebrow, eye, nose, and mouth areas are closely related +<br/>to facial expression [11]. Besides, the effect of each feature +<br/>on recognition result is different. In order to make the best +<br/>of feature, using feature weighting technique can further +<br/>enhance recognition performance. While there are several +<br/>approaches of confirming weight, it remains an open issue +<br/>on how to select feature and calculate corresponding weight +<br/>effectively. +<br/>In this paper, a new emotion recognition method based +<br/>on weighted feature facial expression is presented. It is +<br/>motivated by the fact that emotion can be described by facial +<br/>expression and each facial expression feature has different +<br/>impact on recognition results. Different from previous works +</td><td>('39248132', 'Wei Wei', 'wei wei')<br/>('2301733', 'Qingxuan Jia', 'qingxuan jia')</td><td>Correspondence should be addressed to Wei Wei; wei wei@bupt.edu.cn +</td></tr><tr><td>0b183f5260667c16ef6f640e5da50272c36d599b</td><td>Spatio-temporal Event Classification Using +<br/>Time-Series Kernel Based Structured Sparsity(cid:2) +<br/>L´aszl´o A. Jeni1, Andr´as L˝orincz2, Zolt´an Szab´o3, +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, USA</b><br/><b>Faculty of Informatics, E otv os Lor and University, Budapest, Hungary</b><br/><b>Gatsby Computational Neuroscience Unit, University College London, London, UK</b><br/><b>University of Pittsburgh, Pittsburgh, PA, USA</b></td><td>('1733113', 'Takeo Kanade', 'takeo kanade')</td><td>laszlo.jeni@ieee.org, andras.lorincz@elte.hu, +<br/>zoltan.szabo@gatsby.ucl.ac.uk, {jeffcohn,tk}@cs.cmu.edu +</td></tr><tr><td>0b4c4ea4a133b9eab46b217e22bda4d9d13559e6</td><td>MORF: Multi-Objective Random Forests for Face Characteristic Estimation +<br/><b>MICC - University of Florence</b><br/>2CVC - Universitat Autonoma de Barcelona +<br/><b>DVMM Lab - Columbia University</b></td><td>('37822746', 'Dario Di Fina', 'dario di fina')<br/>('2602265', 'Svebor Karaman', 'svebor karaman')<br/>('1749498', 'Andrew D. Bagdanov', 'andrew d. bagdanov')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')</td><td>{dario.difina, alberto.delbimbo}@unifi.it, svebor.karaman@columbia.edu, bagdanov@cvc.uab.es +</td></tr><tr><td>0ba99a709cd34654ac296418a4f41a9543928149</td><td></td><td></td><td></td></tr><tr><td>0be764800507d2e683b3fb6576086e37e56059d1</td><td>Learning from Geometry +<br/>by +<br/>Department of Electrical and Computer Engineering +<br/><b>Duke University</b><br/>Date: +<br/>Approved: +<br/>Robert Calderbank, Supervisor +<br/>Lawrence Carin +<br/>Ingrid Daubechies +<br/>Gallen Reeves +<br/>Guillermo Sapiro +<br/>Dissertation submitted in partial fulfillment of the requirements for the degree of +<br/>Doctor of Philosophy in the Department of Electrical and Computer Engineering +<br/><b>in the Graduate School of Duke University</b><br/>2016 +</td><td>('34060310', 'Jiaji Huang', 'jiaji huang')</td><td></td></tr><tr><td>0b642f6d48a51df64502462372a38c50df2051b1</td><td>A Domain Adaptation Approach to Improve +<br/>Speaker Turn Embedding Using Face Representation +<br/><b>Idiap Research Institute, Martigny, Switzerland</b><br/>École Polytechnique Fédéral de Lausanne, Switzerland +</td><td>('39560344', 'Nam Le', 'nam le')<br/>('1719610', 'Jean-Marc Odobez', 'jean-marc odobez')</td><td>nle@idiap.ch,odobez@idiap.ch +</td></tr><tr><td>0b7d1386df0cf957690f0fe330160723633d2305</td><td>Learning American English Accents Using Ensemble Learning with GMMs +<br/>Department of Computer Science +<br/><b>Rensselaer Polytechnic Institute</b><br/>Troy, NY 12180 +<br/>Department of Computer Science +<br/><b>Rensselaer Polytechnic Institute</b><br/>Troy, NY 12180 +</td><td>('38769302', 'Jonathan T. Purnell', 'jonathan t. purnell')<br/>('1705107', 'Malik Magdon-Ismail', 'malik magdon-ismail')</td><td>purnej@cs.rpi.edu +<br/>magdon@cs.rpi.edu +</td></tr><tr><td>0b6616f3ebff461e4b6c68205fcef1dae43e2a1a</td><td>Rectifying Self Organizing Maps +<br/>for Automatic Concept Learning from Web Images +<br/><b>Bilkent University</b><br/>06800 Ankara/Turkey +<br/>Pinar Duygulu +<br/><b>Bilkent University</b><br/>06800 Ankara/Turkey +</td><td>('2540074', 'Eren Golge', 'eren golge')</td><td>eren.golge@bilkent.edu.tr +<br/>pinar.duygulu@gmail.com +</td></tr><tr><td>0b8c92463f8f5087696681fb62dad003c308ebe2</td><td>On Matching Sketches with Digital Face Images +<br/>in local +</td><td>('2559473', 'Himanshu S. Bhatt', 'himanshu s. bhatt')<br/>('34173298', 'Samarth Bharadwaj', 'samarth bharadwaj')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')</td><td></td></tr><tr><td>0bc0f9178999e5c2f23a45325fa50300961e0226</td><td>Recognizing facial expressions from videos using Deep +<br/>Belief Networks +<br/>CS 229 Project +</td><td>('34699434', 'Andrew Ng', 'andrew ng')</td><td>Adithya Rao (adithyar@stanford.edu), Narendran Thiagarajan (naren@stanford.edu) +</td></tr><tr><td>0ba402af3b8682e2aa89f76bd823ddffdf89fa0a</td><td>Squared Earth Mover’s Distance-based Loss for Training Deep Neural Networks +<br/>Computer Science Department +<br/><b>Stony Brook University</b><br/>Cognitive Neuroscience Lab +<br/>Computer Science Department +<br/><b>Harvard University</b><br/><b>Stony Brook University</b></td><td>('2321406', 'Le Hou', 'le hou')<br/>('2576295', 'Chen-Ping Yu', 'chen-ping yu')<br/>('1686020', 'Dimitris Samaras', 'dimitris samaras')</td><td>lehhou@cs.stonybrook.edu +<br/>chenpingyu@fas.harvard.edu +<br/>samaras@cs.stonybrook.edu +</td></tr><tr><td>0bf0029c9bdb0ac61fda35c075deb1086c116956</td><td>Article +<br/>Modelling of Orthogonal Craniofacial Profiles +<br/><b>University of York, Heslington, York YO10 5GH, UK</b><br/>Received: 20 October 2017; Accepted: 23 November 2017; Published: 30 November 2017 +</td><td>('1694260', 'Hang Dai', 'hang dai')<br/>('1737428', 'Nick Pears', 'nick pears')<br/>('1678859', 'Christian Duncan', 'christian duncan')</td><td>nick.pears@york.ac.uk +<br/>2 Alder Hey Children’s Hospital, Liverpool L12 2AP, UK; Christian.Duncan@alderhey.nhs.uk +<br/>* Correspondence: hd816@york.ac.uk; Tel.: +44-1904-325-643 +</td></tr><tr><td>0b3f354e6796ef7416bf6dde9e0779b2fcfabed2</td><td></td><td></td><td></td></tr><tr><td>9391618c09a51f72a1c30b2e890f4fac1f595ebd</td><td>Globally Tuned Cascade Pose Regression via +<br/>Back Propagation with Application in 2D Face +<br/>Pose Estimation and Heart Segmentation in 3D +<br/>CT Images +<br/><b>Dalio Institute of Cardiovascular Imaging, Weill Cornell Medical College</b><br/>April 1, 2015 +<br/>This work was submitted to ICML 2015 but got rejected. We put the initial +<br/>submission ”as is” in Page 2 - 11 and add updated contents at the tail. The +<br/>code of this work is available at https://github.com/pengsun/bpcpr5. +</td><td></td><td>Peng Sun pes2021@med.cornell.edu +<br/>James K Min jkm2001@med.cornell.edu +<br/>Guanglei Xiong gux2003@med.cornell.edu +</td></tr><tr><td>93675f86d03256f9a010033d3c4c842a732bf661</td><td>Universit´edesSciencesetTechnologiesdeLilleEcoleDoctoraleSciencesPourl’ing´enieurUniversit´eLilleNord-de-FranceTHESEPr´esent´ee`al’Universit´edesSciencesetTechnologiesdeLillePourobtenirletitredeDOCTEURDEL’UNIVERSIT´ESp´ecialit´e:MicroetNanotechnologieParTaoXULocalizedgrowthandcharacterizationofsiliconnanowiresSoutenuele25Septembre2009Compositiondujury:Pr´esident:TuamiLASRIRapporteurs:ThierryBARONHenriMARIETTEExaminateurs:EricBAKKERSXavierWALLARTDirecteurdeth`ese:BrunoGRANDIDIER </td><td></td><td></td></tr><tr><td>935a7793cbb8f102924fa34fce1049727de865c2</td><td>AGE ESTIMATION UNDER CHANGES IN IMAGE QUALITY: AN EXPERIMENTAL STUDY +<br/><b>ISLA Lab, Informatics Institute, University of Amsterdam</b></td><td>('1765602', 'Fares Alnajar', 'fares alnajar')<br/>('1695527', 'Theo Gevers', 'theo gevers')<br/>('1968574', 'Sezer Karaoglu', 'sezer karaoglu')</td><td></td></tr><tr><td>9326d1390e8601e2efc3c4032152844483038f3f</td><td>Landmark Based Facial Component Reconstruction +<br/>for Recognition Across Pose +<br/>Department of Mechanical Engineering +<br/><b>National Taiwan University of Science and Technology</b><br/>Taipei, Taiwan +</td><td>('38801529', 'Gee-Sern Hsu', 'gee-sern hsu')<br/>('3329222', 'Hsiao-Chia Peng', 'hsiao-chia peng')<br/>('2329565', 'Kai-Hsiang Chang', 'kai-hsiang chang')</td><td>Email: ∗jison@mail.ntust.edu.tw +</td></tr><tr><td>93747de3d40376761d1ef83ffa72ec38cd385833</td><td>COGNITION AND EMOTION, 2015 +<br/>http://dx.doi.org/10.1080/02699931.2015.1039494 +<br/>Team members’ emotional displays as indicators +<br/>of team functioning +<br/><b>University of Amsterdam, Amsterdam, The</b><br/>Netherlands +<br/><b>University of Amsterdam, Amsterdam, The Netherlands</b><br/><b>Ross School of Business, University of Michigan, Ann Arbor, MI, USA</b><br/>(Received 18 August 2014; accepted 6 April 2015) +<br/>Emotions are inherent to team life, yet it is unclear how observers use team members’ emotional +<br/>expressions to make sense of team processes. Drawing on Emotions as Social Information theory, we +<br/>propose that observers use team members’ emotional displays as a source of information to predict the +<br/>team’s trajectory. We argue and show that displays of sadness elicit more pessimistic inferences +<br/>regarding team dynamics (e.g., trust, satisfaction, team effectiveness, conflict) compared to displays of +<br/>happiness. Moreover, we find that this effect is strengthened when the future interaction between the +<br/>team members is more ambiguous (i.e., under ethnic dissimilarity; Study 1) and when emotional +<br/>displays can be clearly linked to the team members’ collective experience (Study 2). These studies shed +<br/>light on when and how people use others’ emotional expressions to form impressions of teams. +<br/>Keywords: Emotions as social information; Impression formation; Team functioning; Sense-making. +<br/>How do people make sense of social collectives? This +<br/>question has a long-standing interest in the social +<br/>sciences (Hamilton & Sherman, 1996), because +<br/>observers’ understanding of what goes on between +<br/>other individuals informs their behavioural responses +<br/>(Abelson, Dasgupta, Park, & Banaji, 1998; Magee & +<br/>Tiedens, 2006). A special type of social collective is +<br/>the team, in which individuals work together on a +<br/>joint task (Ilgen, 1999). There are many reasons why +<br/>outside observers may want to develop an under- +<br/>standing of a team’s functioning and future trajectory, +<br/>for instance because their task is to supervise the team +<br/>or because they are considering sponsoring or poten- +<br/>tially joining the team as a member. However, +<br/>making sense of a team’s trajectory is an uncertain +<br/>endeavour because explicit information about team +<br/>functioning is often not available. This problem is +<br/>further exacerbated by the fact that team ventures are +<br/>simultaneously potent and precarious. When indivi- +<br/>duals join forces in teams, great achievements can be +<br/>obtained (Guzzo & Dickson, 1996), but teams are +<br/>also a potential breeding ground for myriad negative +<br/>outcomes such as intra-team conflicts, social inhibi- +<br/>tion, decision-making biases and productivity losses +<br/>(Jehn, 1995; Kerr & Tindale, 2004). We propose +<br/>that, in their sense-making efforts, observers there- +<br/>fore make use of dynamic signals that provide up-to- +<br/>date diagnostic information about the likely trajectory +<br/><b>Correspondence should be addressed to: Astrid C. Homan, University of Amsterdam, Weesperplein</b><br/>© 2015 Taylor & Francis +</td><td>('2863272', 'Jeffrey Sanchez-Burks', 'jeffrey sanchez-burks')</td><td>1018 XA Amsterdam, The Netherlands. E-mail: ac.homan@uva.nl +</td></tr><tr><td>936c7406de1dfdd22493785fc5d1e5614c6c2882</td><td>2012 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 762–772, +<br/>Montr´eal, Canada, June 3-8, 2012. c(cid:13)2012 Association for Computational Linguistics +<br/>762 +</td><td></td><td></td></tr><tr><td>93721023dd6423ab06ff7a491d01bdfe83db7754</td><td>ROBUST FACE ALIGNMENT USING CONVOLUTIONAL NEURAL +<br/>NETWORKS +<br/>Orange Labs, 4, Rue du Clos Courtel, 35512 Cesson-S´evign´e, France +<br/>Keywords: +<br/>Face alignment, Face registration, Convolutional Neural Networks. +</td><td>('1762557', 'Stefan Duffner', 'stefan duffner')<br/>('34798028', 'Christophe Garcia', 'christophe garcia')</td><td>{stefan.duffner, christophe.garcia}@orange-ftgroup.com +</td></tr><tr><td>93971a49ef6cc88a139420349a1dfd85fb5d3f5c</td><td>Scalable Probabilistic Models: +<br/>Applied to Face Identification in the Wild +<br/>Biometric Person Recognition Group +<br/><b>Idiap Research Institute</b><br/>Rue Marconi 19 PO Box 592 +<br/>1920 Martigny +</td><td>('2121764', 'Laurent El Shafey', 'laurent el shafey')</td><td>laurent.el-shafey@idiap.ch +<br/>sebastien.marcel@idiap.ch +</td></tr><tr><td>93420d9212dd15b3ef37f566e4d57e76bb2fab2f</td><td>An All-In-One Convolutional Neural Network for Face Analysis +<br/><b>Center for Automation Research, UMIACS, University of Maryland, College Park, MD</b></td><td>('48467498', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('2716670', 'Swami Sankaranarayanan', 'swami sankaranarayanan')<br/>('38171682', 'Carlos D. Castillo', 'carlos d. castillo')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{rranjan1,swamiviv,carlos,rama}@umiacs.umd.edu +</td></tr><tr><td>93af36da08bf99e68c9b0d36e141ed8154455ac2</td><td>Workshop track - ICLR 2018 +<br/>ADDITIVE MARGIN SOFTMAX +<br/>FOR FACE VERIFICATION +<br/>Department of Information and Communication Engineering +<br/><b>University of Electronic Science and Technology of China</b><br/>Chengdu, Sichuan 611731 China +<br/><b>College of Computing</b><br/><b>Georgia Institute of Technology</b><br/>Atlanta, United States. +<br/>Department of Information and Communication Engineering +<br/><b>University of Electronic Science and Technology of China</b><br/>Chengdu, Sichuan 611731 China +</td><td>('47939378', 'Feng Wang', 'feng wang')<br/>('51094998', 'Weiyang Liu', 'weiyang liu')<br/>('8424682', 'Haijun Liu', 'haijun liu')</td><td>feng.wff@gmail.com +<br/>{wyliu, hanjundai}@gatech.edu +<br/>haijun liu@126.com chengjian@uestc.edu.cn +</td></tr><tr><td>93cbb3b3e40321c4990c36f89a63534b506b6daf</td><td>IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 35, NO. 3, JUNE 2005 +<br/>477 +<br/>Learning From Examples in the Small Sample Case: +<br/>Face Expression Recognition +</td><td>('1822413', 'Guodong Guo', 'guodong guo')<br/>('1724754', 'Charles R. Dyer', 'charles r. dyer')</td><td></td></tr><tr><td>937ffb1c303e0595317873eda5ce85b1a17f9943</td><td>Eyes Do Not Lie: Spontaneous versus Posed Smiles +<br/><b>Intelligent Systems Lab Amsterdam, University of Amsterdam</b><br/>Science Park 107, Amsterdam, The Netherlands +</td><td>('9301018', 'Roberto Valenti', 'roberto valenti')<br/>('1764521', 'Albert Ali Salah', 'albert ali salah')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td>h.dibeklioglu@uva.nl, r.valenti@uva.nl, a.a.salah@uva.nl, th.gevers@uva.nl +</td></tr><tr><td>93f37c69dd92c4e038710cdeef302c261d3a4f92</td><td>Compressed Video Action Recognition +<br/>Philipp Kr¨ahenb¨uhl1 +<br/><b>The University of Texas at Austin, 2Carnegie Mellon University</b><br/><b>University of Southern California, 4A9, 5Amazon</b></td><td>('2978413', 'Chao-Yuan Wu', 'chao-yuan wu')<br/>('1771307', 'Manzil Zaheer', 'manzil zaheer')<br/>('2804000', 'Hexiang Hu', 'hexiang hu')<br/>('1691629', 'Alexander J. Smola', 'alexander j. smola')<br/>('1758550', 'R. Manmatha', 'r. manmatha')</td><td>cywu@cs.utexas.edu +<br/>manzil@cmu.edu +<br/>smola@amazon.com +<br/>hexiangh@usc.edu +<br/>philkr@cs.utexas.edu +<br/>manmatha@a9.com +</td></tr><tr><td>936227f7483938097cc1cdd3032016df54dbd5b6</td><td>Learning to generalize to new compositions in image understanding +<br/><b>Gonda Brain Research Center, Bar Ilan University, Israel</b><br/>3Google Research, Mountain View CA, USA +<br/><b>Tel Aviv University, Israel</b></td><td>('34815079', 'Yuval Atzmon', 'yuval atzmon')<br/>('1750652', 'Jonathan Berant', 'jonathan berant')<br/>('3451674', 'Vahid Kezami', 'vahid kezami')<br/>('1786843', 'Amir Globerson', 'amir globerson')<br/>('1732280', 'Gal Chechik', 'gal chechik')</td><td>yuval.atzmon@biu.ac.il +</td></tr><tr><td>939123cf21dc9189a03671484c734091b240183e</td><td>Within- and Cross- Database Evaluations for Gender +<br/>Classification via BeFIT Protocols +<br/><b>Idiap Research Institute</b><br/>Centre du Parc, Rue Marconi 19, CH-1920, Martigny, Switzerland +</td><td>('2128163', 'Nesli Erdogmus', 'nesli erdogmus')<br/>('2059725', 'Matthias Vanoni', 'matthias vanoni')</td><td>Email: nesli.erdogmus, matthias.vanoni, marcel@idiap.ch +</td></tr><tr><td>938ae9597f71a21f2e47287cca318d4a2113feb2</td><td>Classifier Learning with Prior Probabilities +<br/>for Facial Action Unit Recognition +<br/>1National Laboratory of Pattern Recognition, CASIA +<br/><b>University of Chinese Academy of Sciences</b><br/><b>Rensselaer Polytechnic Institute</b></td><td>('49889545', 'Yong Zhang', 'yong zhang')<br/>('38690089', 'Weiming Dong', 'weiming dong')<br/>('39495638', 'Bao-Gang Hu', 'bao-gang hu')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>zhangyong201303@gmail.com, weiming.dong@ia.ac.cn, hubg@nlpr.ia.ac.cn, qji@ecse.rpi.edu +</td></tr><tr><td>94b9c0a6515913bad345f0940ee233cdf82fffe1</td><td>International Journal of Science and Research (IJSR) +<br/>ISSN (Online): 2319-7064 +<br/>Impact Factor (2012): 3.358 +<br/>Face Recognition using Local Ternary Pattern for +<br/>Low Resolution Image +<br/><b>Research Scholar, CGC Group of Colleges, Gharuan, Punjab, India</b><br/><b>Chandigarh University, Gharuan, Punjab, India</b></td><td>('40440964', 'Amanpreet Kaur', 'amanpreet kaur')</td><td></td></tr><tr><td>946017d5f11aa582854ac4c0e0f1b18b06127ef1</td><td>Tracking Persons-of-Interest +<br/>via Adaptive Discriminative Features +<br/><b>Xi an Jiaotong University</b><br/><b>Hanyang University</b><br/><b>University of Illinois, Urbana-Champaign</b><br/><b>University of California, Merced</b><br/>http://shunzhang.me.pn/papers/eccv2016/ +</td><td>('2481388', 'Shun Zhang', 'shun zhang')<br/>('1698965', 'Yihong Gong', 'yihong gong')<br/>('3068086', 'Jia-Bin Huang', 'jia-bin huang')<br/>('33047058', 'Jongwoo Lim', 'jongwoo lim')<br/>('32014778', 'Jinjun Wang', 'jinjun wang')<br/>('1752333', 'Narendra Ahuja', 'narendra ahuja')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>94eeae23786e128c0635f305ba7eebbb89af0023</td><td>Journal of Machine Learning Research 18 (2018) 1-34 +<br/>Submitted 01/17; Revised 4/18; Published 6/18 +<br/>Emergence of Invariance and Disentanglement +<br/>in Deep Representations∗ +<br/>Department of Computer Science +<br/><b>University of California</b><br/>Los Angeles, CA 90095, USA +<br/>Department of Computer Science +<br/><b>University of California</b><br/>Los Angeles, CA 90095, USA +<br/>Editor: Yoshua Bengio +</td><td>('16163297', 'Alessandro Achille', 'alessandro achille')<br/>('1715959', 'Stefano Soatto', 'stefano soatto')</td><td>achille@cs.ucla.edu +<br/>soatto@cs.ucla.edu +</td></tr><tr><td>944faf7f14f1bead911aeec30cc80c861442b610</td><td>Action Tubelet Detector for Spatio-Temporal Action Localization +</td><td>('1881509', 'Vicky Kalogeiton', 'vicky kalogeiton')<br/>('2492127', 'Philippe Weinzaepfel', 'philippe weinzaepfel')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>9458c518a6e2d40fb1d6ca1066d6a0c73e1d6b73</td><td>5967 +<br/>A Benchmark and Comparative Study of +<br/>Video-Based Face Recognition +<br/>on COX Face Database +</td><td>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('1705483', 'Haihong Zhang', 'haihong zhang')<br/>('1710195', 'Shihong Lao', 'shihong lao')<br/>('2378840', 'Alifu Kuerban', 'alifu kuerban')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>948af4b04b4a9ae4bff2777ffbcb29d5bfeeb494</td><td>Available online at www.sciencedirect.com +<br/> Procedia Engineering 41 ( 2012 ) 465 – 472 +<br/>International Symposium on Robotics and Intelligent Sensors 2012 (IRIS 2012) +<br/>Face Recognition From Single Sample Per Person by Learning of +<br/>Generic Discriminant Vectors +<br/><b>aFaculty of Electrical Engineering, University of Technology MARA, Shah Alam, 40450 Selangor, Malaysia</b><br/><b>bFaculty of Engineering, International Islamic University, Jalan Gombak, 53100 Kuala Lumpur, Malaysia</b></td><td>('7453141', 'Fadhlan Hafiz', 'fadhlan hafiz')<br/>('2412523', 'Amir A. Shafie', 'amir a. shafie')<br/>('9146253', 'Yasir Mohd Mustafah', 'yasir mohd mustafah')</td><td></td></tr><tr><td>94aa8a3787385b13ee7c4fdd2b2b2a574ffcbd81</td><td></td><td></td><td></td></tr><tr><td>94325522c9be8224970f810554611d6a73877c13</td><td></td><td></td><td></td></tr><tr><td>9487cea80f23afe9bccc94deebaa3eefa6affa99</td><td>Fast, Dense Feature SDM on an iPhone +<br/><b>Queensland University of Technology, Brisbane, Queensland, Australia</b><br/><b>Carnegie Mellon University, Pittsburgh, PA, USA</b></td><td>('3231493', 'Ashton Fagg', 'ashton fagg')<br/>('1820249', 'Simon Lucey', 'simon lucey')<br/>('1729760', 'Sridha Sridharan', 'sridha sridharan')</td><td></td></tr><tr><td>9441253b638373a0027a5b4324b4ee5f0dffd670</td><td>A Novel Scheme for Generating Secure Face +<br/>Templates Using BDA +<br/>P.G. Student, Department of Computer Engineering, +<br/>Associate Professor, Department of Computer +<br/>MCERC, +<br/>Nashik (M.S.), India +</td><td>('40075681', 'Shraddha S. Shinde', 'shraddha s. shinde')<br/>('2590072', 'Anagha P. Khedkar', 'anagha p. khedkar')</td><td>e-mail: shraddhashinde@gmail.com +</td></tr><tr><td>949699d0b865ef35b36f11564f9a4396f5c9cddb</td><td>Anders, Ende, Junghofer, Kissler & Wildgruber (Eds.) +<br/>ISSN 0079-6123 +<br/>CHAPTER 18 +<br/>Processing of facial identity and expression: a +<br/>psychophysical, physiological and computational +<br/>perspective +<br/>Sarah D. Chiller-Glaus2 +<br/><b>Max Planck Institute for Biological Cybernetics, Spemannstr. 38, 72076 T bingen, Germany</b><br/><b>University of Zurich, Zurich, Switzerland</b></td><td>('2388249', 'Adrian Schwaninger', 'adrian schwaninger')<br/>('1793750', 'Christian Wallraven', 'christian wallraven')</td><td></td></tr><tr><td>94ac3008bf6be6be6b0f5140a0bea738d4c75579</td><td></td><td></td><td></td></tr><tr><td>94e259345e82fa3015a381d6e91ec6cded3971b4</td><td>Classiflcation of Photometric Factors +<br/>Based on Photometric Linearization +<br/><b>The Institute of Scienti c and Industrial Research, Osaka University</b><br/>8-1 Mihogaoka, Ibaraki-shi, Osaka 567-0047, JAPAN +<br/>2 Matsushita Electric Industrial Co., Ltd. +<br/><b>Okayama University</b><br/>Okayama-shi, Okayama 700-8530, JAPAN +</td><td>('3155610', 'Yasuhiro Mukaigawa', 'yasuhiro mukaigawa')<br/>('2740479', 'Yasunori Ishii', 'yasunori ishii')<br/>('1695509', 'Takeshi Shakunaga', 'takeshi shakunaga')</td><td>mukaigaw@am.sanken.osaka-u.ac.jp +</td></tr><tr><td>94a11b601af77f0ad46338afd0fa4ccbab909e82</td><td></td><td></td><td></td></tr><tr><td>0efdd82a4753a8309ff0a3c22106c570d8a84c20</td><td>LDA WITH SUBGROUP PCA METHOD FOR FACIAL IMAGE RETRIEVAL +<br/><b>Human Computer Interaction Lab., Samsung Advanced Institute of Technology, Korea</b></td><td>('34600044', 'Wonjun Hwang', 'wonjun hwang')<br/>('1700968', 'Tae-Kyun Kim', 'tae-kyun kim')<br/>('37980373', 'Seokcheol Kee', 'seokcheol kee')</td><td>wjhwang@sait.samsung.co.kr +</td></tr><tr><td>0e5dcc6ae52625fd0637c6bba46a973e46d58b9c</td><td>Pareto Models for Multiclass Discriminative Linear +<br/>Dimensionality Reduction +<br/><b>University of Alberta, Edmonton, AB T6G 2E8, Canada</b><br/><b>bRobotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213, U.S.A</b><br/><b>cCentre of Intelligent Machines, McGill University, Montr eal, QC H3A 0E9, Canada</b></td><td>('3141839', 'Fernando De La Torre', 'fernando de la torre')<br/>('1701344', 'Frank P. Ferrie', 'frank p. ferrie')</td><td></td></tr><tr><td>0e73d2b0f943cf8559da7f5002414ccc26bc77cd</td><td>Similarity Comparisons for Interactive Fine-Grained Categorization +<br/><b>California Institute of Technology</b><br/>vision.caltech.edu +<br/>Serge Belongie4 +<br/><b>Toyota Technological Institute at Chicago</b><br/>ttic.edu +<br/>4 Cornell Tech +<br/>tech.cornell.edu +<br/>Approach +<br/>1) Image +<br/>Database w/ +<br/>Class Labels +<br/>2) Collect Similarity +<br/>Comparisons +<br/>3) Learn Perceptual +<br/>Embedding +<br/>A +<br/>Mallard +<br/>Cardinal +<br/>? +<br/>1) Query Image +<br/>2) Computer +<br/>Vision +<br/>B +<br/>3) Human-in-the-Loop Categorization +<br/>C +<br/>(cid:1876) +<br/>(cid:1855) +<br/>(cid:1868) +<br/>C +<br/>D +<br/>perceptual space +<br/>where +<br/>(cid:1826) True location of (cid:1876) in +<br/>(cid:1872) Time step +<br/>(cid:1847)(cid:3047) User responses at (cid:1872) +<br/>(cid:1876) Query image +<br/>(cid:1855) Class +<br/>INTERACTIVE +<br/>CATEGORIZATION +<br/>• Compute per-class probabilities as: +<br/>(cid:1826) +<br/>(cid:1868)(cid:1855),|(cid:1876),(cid:1847)(cid:3047) (cid:1503)(cid:1868)(cid:1855),(cid:1847)(cid:3047)|(cid:1876) = (cid:3505) (cid:1868)(cid:1855),(cid:1826),(cid:1847)(cid:3047)|(cid:1876)(cid:1856)(cid:1826) +<br/>(cid:1875)(cid:3047)=(cid:1868)(cid:1855),(cid:1826),(cid:1847)(cid:3047)|(cid:1876) =(cid:1868)(cid:1847)(cid:3047)| (cid:1855),(cid:1826),(cid:1876) (cid:1868)(cid:1855),(cid:1826)(cid:1876) +<br/>(cid:1868)(cid:1855),|(cid:1876),(cid:1847)(cid:3047) (cid:3406)(cid:963) +<br/>(cid:1875)(cid:3038)(cid:3047) +<br/>(cid:3038),(cid:3030)(cid:3286)(cid:2880)(cid:3030)(cid:963) (cid:1875)(cid:3038)(cid:3047) +<br/>i.e. sum of weights of examples of class (cid:1855) +<br/>(cid:3038) +<br/>where (cid:1863) enumerates training examples +<br/>• Weight (cid:1875)(cid:3038) represents how likely (cid:1826)(cid:3038) is +<br/>true location (cid:1826): +<br/>(cid:1875)(cid:3038)(cid:3047)=(cid:1868)(cid:1855)(cid:3038),(cid:1826)(cid:3038),(cid:1847)(cid:3047)|(cid:1876) =(cid:1868)(cid:1847)(cid:3047)| (cid:1855)(cid:3038),(cid:1826)(cid:3038),(cid:1876) (cid:1868)(cid:1855)(cid:3038),(cid:1826)(cid:3038)(cid:1876) +<br/>Efficient computation +<br/>• Approximate per-class probabilities as: +<br/>such that +<br/>(cid:1875)(cid:3038)(cid:3047)(cid:2878)(cid:2869)=(cid:1868)(cid:1873)(cid:3047)(cid:2878)(cid:2869)(cid:1826)(cid:3038)(cid:1875)(cid:3038)(cid:3047) +<br/>= (cid:2038)(cid:1845)(cid:3036)(cid:3038) +<br/>(cid:1875)(cid:3038)(cid:3047) +<br/>(cid:963) +<br/>(cid:2038)(cid:1845)(cid:3037)(cid:3038) +<br/>(cid:3037)(cid:1488)(cid:3005) +<br/>(cid:3513) Initialize weights (cid:1875)(cid:3038)(cid:2868)= (cid:1868)(cid:1855)(cid:3038),(cid:1826)(cid:3038)(cid:1876) +<br/>(cid:3514) Update weights (cid:1875)(cid:3038)(cid:3047)(cid:2878)(cid:2869) when user answers +<br/>Efficient update rule: +<br/>a similarity question +<br/>(cid:3515) Update per-class probabilities +<br/>? +<br/>(cid:3047) +<br/>(cid:1847) +<br/>(cid:1876) +<br/>(cid:1855) +<br/>(cid:1868) +<br/>D +<br/>A +<br/>Learning a Metric +<br/>• Given set of triplet comparisons (cid:2286), learn +<br/>embedding (cid:1800) of (cid:1840) training images with +<br/>From (cid:1800), generate similarity matrix +<br/>(cid:1845)(cid:1488)(cid:1840)×(cid:1840) +<br/>stochastic triplet embedding [van der Maaten +<br/>& Weinberger 2012] +<br/>B +<br/>D +<br/>D +<br/>Computer Vision +<br/>• Easy to map off-the-shelf CV +<br/>algorithms into framework, e.g., +<br/>multiclass classification scores +<br/>(cid:1868)(cid:1855),(cid:1826)(cid:1876) (cid:1503)(cid:1868)(cid:1855)|(cid:1876) +<br/>Incorporate independent user +<br/>response as: +<br/>Incorporating Users +<br/>• (cid:1830) is grid of images for each question +<br/>(cid:1868)(cid:1873)(cid:1826) = (cid:2038)(cid:1871)((cid:1826),(cid:1826)(cid:3036)) +<br/>(cid:963) +<br/>(cid:2038)(cid:1871)((cid:1826),(cid:1826)(cid:3037)) +<br/>(cid:3037)(cid:1488)(cid:3005) +<br/>entropy of (cid:1868)(cid:1855),(cid:1826)(cid:3038),(cid:1847)(cid:3047)|(cid:1876) +<br/>largest (cid:1875)(cid:3038)(cid:3047) +<br/>Selecting the Display +<br/>• Approximate solution: maximizes +<br/>[Fang & Geman 2005] +<br/>From each cluster, select image with +<br/>expected information gain in terms of +<br/>• Group images into equal-weight clusters +<br/>Results +<br/>Learned Embedding +<br/>Learn category-level embedding of +<br/>• Category-level embedding requires +<br/>(cid:1840)=200 nodes +<br/>Simulated noisy users +<br/>With computer vision +<br/>Deterministic users +<br/>No computer vision +<br/>Deterministic users +<br/>With computer vision +<br/>Interactive Categorization +<br/>• Using computer vision reduces the burden on the user +<br/>• The system is robust to user noise +<br/>much fewer comparisons compared to +<br/>at the instance-level +<br/>Similarity comparisons are advantageous compared to part/attribute questions +<br/>Intelligently selecting image displays reduces effort +<br/>System supports multiple similarity +<br/>metrics as different types of +<br/>questions +<br/>Simulate perceptual spaces using +<br/>CUB-200-2011 attribute +<br/>annotations +<br/>Multiple Metrics +<br/>CV, Color Similarity +<br/>CV, Shape Similarity +<br/>CV, Pattern Similarity +<br/>No CV, Color/Shape/Pattern Similarity +<br/>CV, Color/Shape/Pattern Similarity +<br/>Method +<br/>Avg. #Qs +<br/>2.70 +<br/>2.67 +<br/>2.67 +<br/>2.64 +<br/>4.21 +<br/>Qualitative Results +<br/>Vermilion +<br/>Fly- +<br/>catcher +<br/>Query Image +<br/>Q1: Most Similar? +<br/>Q2: Most Similar? +<br/>Query Image +<br/>Q1: Most Similar By Color? +<br/>Q2: Most Similar By Pattern? +<br/>Hooded +<br/>Merganser +<br/><b>University of California, San Diego</b><br/>vision.ucsd.edu +<br/>Overview +<br/>Problem +<br/>• Parts and attributes exhibit weaknesses +<br/>(cid:190) Scalability issues; costly; reliance on experts, but experts are scarce +<br/>Proposed Solution +<br/>• Use relative similarity comparisons to reduce dependence on expert- +<br/>derived part and attribute vocabularies +<br/>Contributions +<br/>• We present an efficient, flexible, and scalable system for interactive +<br/>fine-grained visual categorization +<br/>(cid:190) Based on perceptual similarity +<br/>(cid:190) Combines similarity metrics and computer vision methods in a +<br/>unified framework +<br/>• Outperforms state-of-the-art relevance feedback-based and +<br/>part/attribute-based approaches +<br/>Similarity Comparisons +<br/>A +<br/>A. Collect grid-based +<br/>similarity comp- +<br/>arisons that do not +<br/>require prior expertise +<br/>B. Broadcast grid-based +<br/>comparisons to triplet +<br/>comparisons +<br/>B +<br/>(cid:2286)= (cid:1861),(cid:1862),(cid:1864) (cid:1876)(cid:3036) more similar to (cid:1876)(cid:3037) than (cid:1876)(cid:3039) +<br/>Is this more similar to… (cid:1876)(cid:3036) +<br/>(cid:1876)(cid:3037) +<br/>This one? +<br/>(cid:1876)(cid:3039) +<br/>Or this one? +<br/>(cid:1871) , > (cid:1871) , +<br/>(cid:1871) , > (cid:1871) , +<br/>(cid:1871) , > (cid:1871) , +<br/>(cid:1871) , > (cid:1871) , +<br/>(cid:1871) , > (cid:1871) , +<br/>(cid:1871) , > (cid:1871) , +<br/>(cid:1871) , > (cid:1871) , +<br/>(cid:1871) , > (cid:1871) , +<br/>? +<br/>(cid:1871)((cid:1861),(cid:1862)): perceptual similarity +<br/>between images (cid:1876)(cid:3036) and (cid:1876)(cid:3037) +</td><td>('2367820', 'Catherine Wah', 'catherine wah')<br/>('2996914', 'Grant Van Horn', 'grant van horn')<br/>('3251767', 'Steve Branson', 'steve branson')<br/>('35208858', 'Subhransu Maji', 'subhransu maji')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>{sbranson,perona}@caltech.edu +<br/>smaji@ttic.edu +<br/>sjb@cs.cornell.edu +<br/>{cwah@cs,gvanhorn@}ucsd.edu +</td></tr><tr><td>0ed0e48b245f2d459baa3d2779bfc18fee04145b</td><td>Semi-Supervised Dimensionality Reduction∗ +<br/>1National Laboratory for Novel Software Technology +<br/><b>Nanjing University, Nanjing 210093, China</b><br/>2Department of Computer Science and Engineering +<br/><b>Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China</b></td><td>('1772283', 'Daoqiang Zhang', 'daoqiang zhang')<br/>('1692625', 'Zhi-Hua Zhou', 'zhi-hua zhou')<br/>('1680768', 'Songcan Chen', 'songcan chen')</td><td>dqzhang@nuaa.edu.cn +<br/>zhouzh@nju.edu.cn +<br/>s.chen@nuaa.edu.cn +</td></tr><tr><td>0eac652139f7ab44ff1051584b59f2dc1757f53b</td><td>Efficient Branching Cascaded Regression +<br/>for Face Alignment under Significant Head Rotation +<br/><b>University of Wisconsin Madison</b></td><td>('2721523', 'Brandon M. Smith', 'brandon m. smith')<br/>('1724754', 'Charles R. Dyer', 'charles r. dyer')</td><td>bmsmith@cs.wisc.edu +<br/>dyer@cs.wisc.edu +</td></tr><tr><td>0ef96d97365899af797628e80f8d1020c4c7e431</td><td>Improving the Speed of Kernel PCA on Large Scale Datasets +<br/><b>Institute for Vision Systems Engineering</b><br/><b>Monash University, Victoria, Australia</b></td><td>('2451050', 'Tat-Jun Chin', 'tat-jun chin')<br/>('2220700', 'David Suter', 'david suter')</td><td>{ tat.chin | d.suter }@eng.monash.edu.au +</td></tr><tr><td>0e7f277538142fb50ce2dd9179cffdc36b794054</td><td>Combining Image Captions and Visual Analysis +<br/>for Image Concept Classification +<br/>Department of Information and +<br/>Knowledge Engineering +<br/>Faculty of Informatics and +<br/><b>Statistics, University of</b><br/>Economics, Prague +<br/>Multimedia and Vision +<br/>Research Group +<br/><b>Queen Mary University</b><br/>Mile End Road, London +<br/>United Kingdom +<br/>Department of Information and +<br/>Knowledge Engineering +<br/>Faculty of Informatics and +<br/><b>Statistics, University of</b><br/>Economics, Prague +<br/>Department of Information and +<br/>Knowledge Engineering +<br/>Faculty of Informatics and +<br/><b>Statistics, University of</b><br/>Economics, Prague +<br/>Multimedia and Vision +<br/>Research Group +<br/><b>Queen Mary University</b><br/>Mile End Road, London +<br/>United Kingdom +</td><td>('2005670', 'Tomas Kliegr', 'tomas kliegr')<br/>('3183509', 'Krishna Chandramouli', 'krishna chandramouli')<br/>('2073485', 'Jan Nemrava', 'jan nemrava')<br/>('1740821', 'Vojtech Svatek', 'vojtech svatek')<br/>('1732655', 'Ebroul Izquierdo', 'ebroul izquierdo')</td><td>tomas.kliegr@vse.cz +<br/>krishna.c@ieee.org +<br/>nemrava@vse.cz +<br/>svatek@vse.cz +<br/>ebroul.izquierdo@elec.qmul.ac.uk +</td></tr><tr><td>0e8760fc198a7e7c9f4193478c0e0700950a86cd</td><td></td><td></td><td></td></tr><tr><td>0ec0fc9ed165c40b1ef4a99e944abd8aa4e38056</td><td>HHS Public Access +<br/>Author manuscript +<br/>Curr Res Psychol. Author manuscript; available in PMC 2017 January 17. +<br/>Published in final edited form as: +<br/>Curr Res Psychol. 2016 ; 6(2): 22–30. doi:10.3844/crpsp.2015.22.30. +<br/>The Role of Perspective-Taking on Ability to Recognize Fear +<br/><b>Virginia Polytechnic Institute and State University, Blacksburg</b><br/>Virginia, USA +<br/><b>Virginia Polytechnic Institute and State University, Blacksburg, Virginia</b><br/>USA +<br/><b>Virginia Tech Carilion Research Institute</b><br/>Roanoke, Virginia, USA +<br/><b>Virginia Polytechnic Institute and State University, Blacksburg</b><br/>Virginia, USA +</td><td>('2974674', 'Andrea Trubanova', 'andrea trubanova')<br/>('2359365', 'Inyoung Kim', 'inyoung kim')<br/>('3712207', 'Marika C. Coffman', 'marika c. coffman')<br/>('6057482', 'Martha Ann Bell', 'martha ann bell')<br/>('2294952', 'Stephen M. LaConte', 'stephen m. laconte')<br/>('1709677', 'Denis Gracanin', 'denis gracanin')<br/>('2197231', 'Susan W. White', 'susan w. white')</td><td></td></tr><tr><td>0e652a99761d2664f28f8931fee5b1d6b78c2a82</td><td>BERGSTRA, YAMINS, AND COX: MAKING A SCIENCE OF MODEL SEARCH +<br/>Making a Science of Model Search +<br/>J. Bergstra1 +<br/>D. Yamins2 +<br/>D. D. Cox1 +<br/><b>Rowland Institute at Harvard</b><br/>100 Edwin H. Land Boulevard +<br/>Cambridge, MA 02142, USA +<br/>2 Department of Brain and Cognitive +<br/>Sciences +<br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, MA 02139, USA +</td><td></td><td>bergstra@rowland.harvard.edu +<br/>yamins@mit.edu +<br/>davidcox@fas.harvard.edu +</td></tr><tr><td>0e50fe28229fea45527000b876eb4068abd6ed8c</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>2936 +</td><td></td><td></td></tr><tr><td>0eff410cd6a93d0e37048e236f62e209bc4383d1</td><td>Anchorage Convention District +<br/>May 3-8, 2010, Anchorage, Alaska, USA +<br/>978-1-4244-5040-4/10/$26.00 ©2010 IEEE +<br/>4803 +</td><td></td><td></td></tr><tr><td>0ea7b7fff090c707684fd4dc13e0a8f39b300a97</td><td>Integrated Face Analytics Networks through +<br/>Cross-Dataset Hybrid Training +<br/><b>School of Computing, National University of Singapore, Singapore</b><br/><b>Electrical and Computer Engineering, National University of Singapore, Singapore</b><br/><b>Beijing Institute of Technology University, P. R. China</b><br/>4 SAP Innovation Center Network Singapore, Singapore +</td><td>('2757639', 'Jianshu Li', 'jianshu li')<br/>('2052311', 'Jian Zhao', 'jian zhao')<br/>('1715286', 'Terence Sim', 'terence sim')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('3124720', 'Shengtao Xiao', 'shengtao xiao')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')<br/>('40345914', 'Fang Zhao', 'fang zhao')<br/>('1943724', 'Jianan Li', 'jianan li')</td><td>{jianshu,xiao_shengtao,zhaojian90}@u.nus.edu,lijianan15@gmail.com +<br/>{elezhf,elefjia,eleyans}@nus.edu.sg,tsim@comp.nus.edu.sg +</td></tr><tr><td>0ee737085af468f264f57f052ea9b9b1f58d7222</td><td>SiGAN: Siamese Generative Adversarial Network +<br/>for Identity-Preserving Face Hallucination +</td><td>('3192517', 'Chih-Chung Hsu', 'chih-chung hsu')<br/>('1685088', 'Chia-Wen Lin', 'chia-wen lin')<br/>('3404171', 'Weng-Tai Su', 'weng-tai su')<br/>('1705205', 'Gene Cheung', 'gene cheung')</td><td></td></tr><tr><td>0ee661a1b6bbfadb5a482ec643573de53a9adf5e</td><td>JOURNAL OF LATEX CLASS FILES, VOL. X, NO. X, MONTH YEAR +<br/>On the Use of Discriminative Cohort Score +<br/>Normalization for Unconstrained Face Recognition +</td><td>('1725688', 'Massimo Tistarelli', 'massimo tistarelli')<br/>('2384894', 'Yunlian Sun', 'yunlian sun')<br/>('2404207', 'Norman Poh', 'norman poh')</td><td></td></tr><tr><td>0e36ada8cb9c91f07c9dcaf196d036564e117536</td><td>Much Ado About Time: Exhaustive Annotation of Temporal Data +<br/><b>Carnegie Mellon University</b><br/>2Inria +<br/><b>University of Washington 4The Allen Institute for AI</b><br/>http://allenai.org/plato/charades/ +</td><td>('34280810', 'Gunnar A. Sigurdsson', 'gunnar a. sigurdsson')<br/>('2192178', 'Olga Russakovsky', 'olga russakovsky')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('1785596', 'Ivan Laptev', 'ivan laptev')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')</td><td></td></tr><tr><td>0e986f51fe45b00633de9fd0c94d082d2be51406</td><td>Face Detection, Pose Estimation, and Landmark Localization in the Wild +<br/><b>University of California, Irvine</b></td><td>('32542103', 'Xiangxin Zhu', 'xiangxin zhu')</td><td>{xzhu,dramanan}@ics.uci.edu +</td></tr><tr><td>0ebc50b6e4b01eb5eba5279ce547c838890b1418</td><td>Similarity-Preserving Binary Signature for Linear Subspaces +<br/>∗State Key Laboratory of Intelligent Technology and Systems, +<br/>Tsinghua National Laboratory for Information Science and Technology (TNList), +<br/><b>Tsinghua University, Beijing 100084, China</b><br/><b>National University of Singapore, Singapore</b></td><td>('1901939', 'Jianqiu Ji', 'jianqiu ji')<br/>('38376468', 'Jianmin Li', 'jianmin li')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('1713616', 'Qi Tian', 'qi tian')<br/>('34997537', 'Bo Zhang', 'bo zhang')</td><td>jijq10@mails.tsinghua.edu.cn, {lijianmin, dcszb}@mail.tsinghua.edu.cn +<br/>‡Department of Computer Science, University of Texas at San Antonio, qi.tian@utsa.edu +<br/>eleyans@nus.edu.sg +</td></tr><tr><td>0e49a23fafa4b2e2ac097292acf00298458932b4</td><td>Theory and Applications of Mathematics & Computer Science 3 (1) (2013) 13–31 +<br/>Unsupervised Detection of Outlier Images Using Multi-Order +<br/>Image Transforms +<br/><b>aLawrence Technological University, 21000 W Ten Mile Rd., South eld, MI 48075, United States</b></td><td></td><td></td></tr><tr><td>0ec1673609256b1e457f41ede5f21f05de0c054f</td><td>Blessing of Dimensionality: High-dimensional Feature and Its Efficient +<br/>Compression for Face Verification +<br/><b>University of Science and Technology of China</b><br/>Microsoft Research Asia +</td><td>('39447786', 'Dong Chen', 'dong chen')<br/>('2032273', 'Xudong Cao', 'xudong cao')<br/>('1716835', 'Fang Wen', 'fang wen')<br/>('40055995', 'Jian Sun', 'jian sun')</td><td>chendong@mail.ustc.edu.cn +<br/>{xudongca,fangwen,jiansun}@microsoft.com +</td></tr><tr><td>0e3840ea3227851aaf4633133dd3cbf9bbe89e5b</td><td></td><td></td><td></td></tr><tr><td>0e5dad0fe99aed6978c6c6c95dc49c6dca601e6a</td><td></td><td></td><td></td></tr><tr><td>0ea38a5ba0c8739d1196da5d20efb13406bb6550</td><td>Relative Attributes +<br/><b>Toyota Technological Institute Chicago (TTIC</b><br/><b>University of Texas at Austin</b></td><td>('1713589', 'Devi Parikh', 'devi parikh')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>dparikh@ttic.edu +<br/>grauman@cs.utexas.edu +</td></tr><tr><td>0e21c9e5755c3dab6d8079d738d1188b03128a31</td><td>Constrained Clustering and Its Application to Face Clustering in Videos +<br/>1NLPR, CASIA, Beijing 100190, China +<br/><b>Rensselaer Polytechnic Institute, Troy, NY 12180, USA</b></td><td>('2040015', 'Baoyuan Wu', 'baoyuan wu')<br/>('40382978', 'Yifan Zhang', 'yifan zhang')<br/>('39495638', 'Bao-Gang Hu', 'bao-gang hu')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td></td></tr><tr><td>0e78af9bd0f9a0ce4ceb5f09f24bc4e4823bd698</td><td>Spontaneous Subtle Expression Recognition: +<br/>Imbalanced Databases & Solutions (cid:63) +<br/>1 Faculty of Engineering, +<br/><b>Multimedia University (MMU), Cyberjaya, Malaysia</b><br/>2 Faculty of Computing & Informatics, +<br/><b>Multimedia University (MMU), Cyberjaya, Malaysia</b></td><td>('2339975', 'John See', 'john see')</td><td>lengoanhcat@gmail.com, raphael@mmu.edu.my +<br/>johnsee@mmu.edu.my +</td></tr><tr><td>0e93a5a7f6dbdb3802173dca05717d27d72bfec0</td><td>Attribute Recognition by Joint Recurrent Learning of Context and Correlation +<br/><b>Queen Mary University of London</b><br/>Vision Semantics Ltd.2 +</td><td>('48093957', 'Jingya Wang', 'jingya wang')<br/>('2171228', 'Xiatian Zhu', 'xiatian zhu')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('47113208', 'Wei Li', 'wei li')</td><td>{jingya.wang, s.gong, wei.li}@qmul.ac.uk +<br/>eddy@visionsemantics.com +</td></tr><tr><td>0e2ea7af369dbcaeb5e334b02dd9ba5271b10265</td><td></td><td></td><td></td></tr><tr><td>0ed1c1589ed284f0314ed2aeb3a9bbc760dcdeb5</td><td>Max-Margin Early Event Detectors +<br/>Minh Hoai +<br/><b>Robotics Institute, Carnegie Mellon University</b></td><td>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td></td></tr><tr><td>0e7c70321462694757511a1776f53d629a1b38f3</td><td>NIST Special Publication 1136 +<br/>2012 Proceedings of the +<br/>Performance Metrics for Intelligent +<br/>Systems (PerMI ‘12) Workshop +<br/> +<br/>http://dx.doi.org/10.6028/NIST.SP.1136 +</td><td>('39737545', 'Rajmohan Madhavan', 'rajmohan madhavan')<br/>('2105056', 'Elena R. Messina', 'elena r. messina')<br/>('31797581', 'Brian A. Weiss', 'brian a. weiss')</td><td></td></tr><tr><td>0ec2049a1dd7ae14c7a4c22c5bcd38472214f44d</td><td>Fast Subspace Search via Grassmannian Based Hashing +<br/><b>University of Minnesota</b><br/><b>Proto Labs, Inc</b><br/><b>Columbia University</b><br/><b>University of Minnesota</b></td><td>('1712593', 'Xu Wang', 'xu wang')<br/>('1734862', 'Stefan Atev', 'stefan atev')<br/>('1738310', 'John Wright', 'john wright')<br/>('1919996', 'Gilad Lerman', 'gilad lerman')</td><td>wang1591@umn.edu +<br/>stefan.atev@gmail.com +<br/>johnwright@ee.columbia.edu +<br/>lerman@umn.edu +</td></tr><tr><td>0ec67c69e0975cfcbd8ba787cc0889aec4cc5399</td><td>Locating Salient Object Features +<br/>K.N.Walker, T.F.Cootes and C.J.Taylor +<br/>Dept. Medical Biophysics, +<br/><b>Manchester University, UK</b></td><td></td><td>knw@sv1.smb.man.ac.uk +</td></tr><tr><td>0e1983e9d0e8cb4cbffef7af06f6bc8e3f191a64</td><td>Estimating Illumination Parameters In Real Space +<br/>With Application To Image Relighting +<br/><b>Key Laboratory of Pervasive Computing (Tsinghua University), Ministry of Education</b><br/>Guangyou Xu +<br/><b>Tsinghua University, Beijing 100084, P.R.China</b><br/>Categories and Subject Descriptors +<br/>I.4.8 [Image Processing and Computer Vision]: Scene Analysis +<br/>– photometry, shading, shape. +<br/>General Terms +<br/>Algorithms +<br/>Keywords +<br/>Illumination parameters estimation, spherical harmonic, image +<br/>relighting. +<br/>1. INTRODUCTION +<br/>Illumination condition is a fundamental problem in both computer +<br/>vision and graphics. For instance, the estimation of lighting +<br/>condition is important in face relighting and recognition, since +<br/>synthesized realistic images can alleviate the small sample +<br/>problem in face recognition applications. +<br/>Recently Basri [2] and Ramamoorthi [3] independently apply the +<br/>spherical harmonics techniques to explain the low dimensionality +<br/>of differently illuminated images for convex Lambertian object. +<br/>Ramamoorthi even derives analytically the principal components +<br/>of this low dimensional image subspace. This method have +<br/>already been widely applied to the areas of inverse rendering, +<br/>image relighting, face recognition, etc. +<br/>One of the limitations of this method is that the cast shadows are +<br/>ignored. In the experiment results of [1], the cast shadows +<br/>improve the face recognition result on the most extreme light +<br/>directions. How to overcome this limitation is one of the +<br/>motivations of our work. Furthermore, rendering realistic image +<br/>need the real light direction. Although the spherical harmonics +<br/>coefficient of illumination could be easily estimated, how to +<br/>recover the real light direction from these coefficients is still a +<br/>problem. +<br/>We propose a novel algorithm for estimating the illumination +<br/>parameters including the direction and strength of point light with +<br/>the strength of ambient illumination. Images are projected into the +<br/>analytical subspace derived in [3] according to a known 3D +<br/>geometry, then the illumination parameters are estimated from +<br/>these projected coefficients. Our primary experiments proved the +<br/>stability and effectiveness of this method. +<br/>Copyright is held by the author/owner(s). +<br/>MM'05, November 6-11, 2005, Singapore. +<br/>ACM 1-59593-044-2/05/0011. +<br/>2. METHODOLOGY +<br/>Consider a convex Lambertian object of known geometry with +<br/>uniform albedo illuminated by distant isotropic light sources, the +<br/>irradiance could be expressed as a linear combination of the +<br/>spherical harmonic basis functions. In fact, 99% of the energy of +<br/>the Lambertian BRDF filter is constrained by the first 9 basis [3]. +<br/>In this paper we consider a simple illumination model consisting +<br/>of one distant directional point light source and ambient +<br/>illumination. We could write the illumination coefficients as +<br/>formula of four illumination parameters (Azimuth and Elevation +<br/>angel for point light direction, Sp for point light strength and Sa +<br/>for ambient illumination strength). +<br/>One problem is that, although the spherical harmonic basis +<br/>functions are orthogonal in the sphere coordinates, they are not +<br/>orthogonal in the image space. This property causes the algorithm +<br/>unstable in some case. We choose the analytical subspace +<br/>constructed in [3], which requires no training data. The image is +<br/>projected to this subspace and the PCA coefficients are computed. +<br/>Then the illumination parameters could be estimated from these +<br/>PCA coefficients by solving a nonlinear least-square problem. +<br/>Finding a global extreme of nonlinear problem is very difficult. +<br/>We choose the popular Gauss-Newton method to solve this +<br/>minimal problem, which might stay on local minima. The +<br/>experimental results show that if we choose enough PCA +<br/>coefficients, the energy surface guarantee the local minima is +<br/>same as the global minima.(Note that we can use only a part of +<br/>the PCA coefficients to solve this nonlinear minimal problem.) +<br/>Actually the first five PCA coefficients are enough for estimate +<br/>these parameters stably. (For limited length of this paper, the +<br/>equations and stability analysis of the result is omitted.) +<br/>3. RESULTS +<br/>We experimented on both synthesized sphere images and real face +<br/>images in CMU PIE database [4] and Yale Database B [1]. +<br/>3.1 Synthesized sphere images result +<br/>First, we randomly select the four illumination parameters and +<br/>synthesize 600 sphere images under the different illumination, in +<br/>which the incident directions are limited to the upper hemisphere +<br/>and the light strength parameters are normalized to sum to unity. +<br/>Then we test our algorithm on these synthesized sphere images. +<br/>Similar to the Yale Database B, we divide the images into 5 +<br/>subsets (12°, 25°, 55°, 77°, 90°) according to the angle which the +<br/>light source direction makes with the camera's axis. +<br/>1039 </td><td>('13801076', 'Feng Xie', 'feng xie')<br/>('3265275', 'Linmi Tao', 'linmi tao')</td><td>xiefeng97@mails.tsinghua.edu.cn +<br/>{linmi, xgy-dcs}@tsinghua.edu.cn +</td></tr><tr><td>0ee5c4112208995bf2bb0fb8a87efba933a94579</td><td>Understanding Clothing Preference Based on Body Shape From Online Sources +<br/>Fashion is Taking Shape: +<br/>1Scalable Learning and Perception Group, 2Real Virtual Humans +<br/><b>Max Planck Institute for Informatics, Saarbr ucken, Germany</b></td><td>('26879574', 'Hosnieh Sattar', 'hosnieh sattar')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('2635816', 'Gerard Pons-Moll', 'gerard pons-moll')</td><td>{sattar,mfritz,gpons}@mpi-inf.mpg.de +</td></tr><tr><td>0e1a18576a7d3b40fe961ef42885101f4e2630f8</td><td>Automated Detection and Identification of +<br/>Persons in Video +<br/>Visual Geometry Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b><br/>September 24, 2004 +</td><td>('3056091', 'Mark Everingham', 'mark everingham')</td><td>{me|az}@robots.ox.ac.uk +</td></tr><tr><td>6080f26675e44f692dd722b61905af71c5260af8</td><td></td><td></td><td></td></tr><tr><td>60a006bdfe5b8bf3243404fae8a5f4a9d58fa892</td><td>A Reference-Based Framework for +<br/>Pose Invariant Face Recognition +<br/>1 HP Labs, Palo Alto, CA 94304, USA +<br/>2 Google Inc., Mountain View, CA 94043, USA +<br/><b>BRIC, University of North Carolina at Chapel Hill, NC 27599, USA</b><br/><b>Center for Research in Intelligent Systems, University of California, Riverside, CA 92521, USA</b></td><td>('1784929', 'Mehran Kafai', 'mehran kafai')<br/>('1745657', 'Kave Eshghi', 'kave eshghi')<br/>('39776603', 'Le An', 'le an')<br/>('1707159', 'Bir Bhanu', 'bir bhanu')</td><td>mehran.kafai@hp.com, kave@google.com, lan004@unc.edu, bhanu@cris.ucr.edu +</td></tr><tr><td>6043006467fb3fd1e9783928d8040ee1f1db1f3a</td><td>Face Recognition with Learning-based Descriptor +<br/><b>The Chinese University of Hong Kong</b><br/><b>ITCS, Tsinghua University</b><br/><b>Shenzhen Institutes of Advanced Technology</b><br/>4Microsoft Research Asia +<br/>Chinese Academy of Sciences, China +</td><td>('2695115', 'Zhimin Cao', 'zhimin cao')<br/>('2274228', 'Qi Yin', 'qi yin')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('40055995', 'Jian Sun', 'jian sun')</td><td></td></tr><tr><td>600025c9a13ff09c6d8b606a286a79c823d89db8</td><td>Machine Learning and Applications: An International Journal (MLAIJ) Vol.1, No.1, September 2014 +<br/>A REVIEW ON LINEAR AND NON-LINEAR +<br/>DIMENSIONALITY REDUCTION +<br/>TECHNIQUES +<br/>1Arunasakthi. K, 2KamatchiPriya. L +<br/>1 Assistant Professor +<br/>Department of Computer Science and Engineering +<br/><b>Ultra College of Engineering and Technology for Women, India</b><br/>2Assistant Professor +<br/>Department of Computer Science and Engineering +<br/><b>Vickram College of Engineering, Enathi, Tamil Nadu, India</b></td><td></td><td></td></tr><tr><td>60d765f2c0a1a674b68bee845f6c02741a49b44e</td><td></td><td></td><td></td></tr><tr><td>60c24e44fce158c217d25c1bae9f880a8bd19fc3</td><td>Controllable Image-to-Video Translation: +<br/>A Case Study on Facial Expression Generation +<br/>MIT CSAIL +<br/>Wenbing Huang +<br/>Tencent AI Lab +<br/>MIT-Waston Lab +<br/>Tencent AI Lab +<br/>Tencent AI Lab +</td><td>('2548303', 'Lijie Fan', 'lijie fan')<br/>('2551285', 'Chuang Gan', 'chuang gan')<br/>('1768190', 'Junzhou Huang', 'junzhou huang')<br/>('40206014', 'Boqing Gong', 'boqing gong')</td><td></td></tr><tr><td>60e2b9b2e0db3089237d0208f57b22a3aac932c1</td><td>Frankenstein: Learning Deep Face Representations +<br/>using Small Data +</td><td>('38819702', 'Guosheng Hu', 'guosheng hu')<br/>('1766837', 'Xiaojiang Peng', 'xiaojiang peng')<br/>('2653152', 'Yongxin Yang', 'yongxin yang')<br/>('1697755', 'Timothy M. Hospedales', 'timothy m. hospedales')<br/>('34602236', 'Jakob Verbeek', 'jakob verbeek')</td><td></td></tr><tr><td>60542b1a857024c79db8b5b03db6e79f74ec8f9f</td><td>Learning to Detect Human-Object Interactions +<br/><b>University of Michigan, Ann Arbor</b><br/><b>Washington University in St. Louis</b></td><td>('2820136', 'Yu-Wei Chao', 'yu-wei chao')<br/>('1860829', 'Yunfan Liu', 'yunfan liu')<br/>('9539636', 'Xieyang Liu', 'xieyang liu')<br/>('9344937', 'Huayi Zeng', 'huayi zeng')<br/>('8342699', 'Jia Deng', 'jia deng')</td><td>{ywchao,yunfan,lxieyang,jiadeng}@umich.edu +<br/>{zengh}@wustl.edu +</td></tr><tr><td>60d4cef56efd2f5452362d4d9ac1ae05afa970d1</td><td>Learning End-to-end Video Classification with Rank-Pooling +<br/><b>Research School of Engineering, The Australian National University, ACT 2601, Australia</b><br/><b>Research School of Computer Science, The Australian National University, ACT 2601, Australia</b></td><td>('1688071', 'Basura Fernando', 'basura fernando')<br/>('2377076', 'Stephen Gould', 'stephen gould')</td><td>BASURA.FERNANDO@ANU.EDU.AU +<br/>STEPHEN.GOULD@ANU.EDU.AU +</td></tr><tr><td>60ce4a9602c27ad17a1366165033fe5e0cf68078</td><td>TECHNICAL NOTE +<br/>DIGITAL & MULTIMEDIA SCIENCES +<br/>J Forensic Sci, 2015 +<br/>doi: 10.1111/1556-4029.12800 +<br/>Available online at: onlinelibrary.wiley.com +<br/>Ph.D. +<br/>Combination of Face Regions in Forensic +<br/>Scenarios* +</td><td>('1808344', 'Pedro Tome', 'pedro tome')<br/>('1701431', 'Julian Fierrez', 'julian fierrez')<br/>('1692626', 'Ruben Vera-Rodriguez', 'ruben vera-rodriguez')<br/>('1732220', 'Javier Ortega-Garcia', 'javier ortega-garcia')</td><td></td></tr><tr><td>6097ea6fd21a5f86a10a52e6e4dd5b78a436d5bf</td><td></td><td></td><td></td></tr><tr><td>60c699b9ec71f7dcbc06fa4fd98eeb08e915eb09</td><td>Long-Term Video Interpolation with Bidirectional +<br/>Predictive Network +<br/><b>Peking University</b></td><td>('8082703', 'Xiongtao Chen', 'xiongtao chen')<br/>('1788029', 'Wenmin Wang', 'wenmin wang')<br/>('3258842', 'Jinzhuo Wang', 'jinzhuo wang')</td><td></td></tr><tr><td>60970e124aa5fb964c9a2a5d48cd6eee769c73ef</td><td>Subspace Clustering for Sequential Data +<br/>School of Computing and Mathematics +<br/><b>Charles Sturt University</b><br/>Bathurst, NSW 2795, Australia +<br/>Division of Computational Informatics +<br/>CSIRO +<br/>North Ryde, NSW 2113, Australia +</td><td>('40635684', 'Stephen Tierney', 'stephen tierney')<br/>('1750488', 'Junbin Gao', 'junbin gao')<br/>('1767638', 'Yi Guo', 'yi guo')</td><td>{stierney, jbgao}@csu.edu.au +<br/>yi.guo@csiro.au +</td></tr><tr><td>60efdb2e204b2be6701a8e168983fa666feac1be</td><td>Int J Comput Vis +<br/>DOI 10.1007/s11263-017-1043-5 +<br/>Transferring Deep Object and Scene Representations for Event +<br/>Recognition in Still Images +<br/>Received: 31 March 2016 / Accepted: 1 September 2017 +<br/>© Springer Science+Business Media, LLC 2017 +</td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('1915826', 'Zhe Wang', 'zhe wang')</td><td></td></tr><tr><td>60824ee635777b4ee30fcc2485ef1e103b8e7af9</td><td>Cascaded Collaborative Regression for Robust Facial +<br/>Landmark Detection Trained using a Mixture of Synthetic and +<br/>Real Images with Dynamic Weighting +<br/>Life Member, IEEE, William Christmas, and Xiao-Jun Wu +</td><td>('2976854', 'Zhen-Hua Feng', 'zhen-hua feng')<br/>('38819702', 'Guosheng Hu', 'guosheng hu')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td></td></tr><tr><td>60643bdab1c6261576e6610ea64ea0c0b200a28d</td><td></td><td></td><td></td></tr><tr><td>60a20d5023f2bcc241eb9e187b4ddece695c2b9b</td><td>Invertible Nonlinear Dimensionality Reduction +<br/>via Joint Dictionary Learning +<br/>Department of Electrical and Computer Engineering +<br/>Technische Universit¨at M¨unchen, Germany +</td><td>('30013158', 'Xian Wei', 'xian wei')<br/>('1744239', 'Martin Kleinsteuber', 'martin kleinsteuber')<br/>('36559760', 'Hao Shen', 'hao shen')</td><td>{xian.wei, kleinsteuber, hao.shen}@tum.de. +</td></tr><tr><td>60cdcf75e97e88638ec973f468598ae7f75c59b4</td><td>86 +<br/>Face Annotation Using Transductive +<br/>Kernel Fisher Discriminant +</td><td>('1704030', 'Jianke Zhu', 'jianke zhu')<br/>('1681775', 'Michael R. Lyu', 'michael r. lyu')</td><td></td></tr><tr><td>60040e4eae81ab6974ce12f1c789e0c05be00303</td><td>Center for Energy Harvesting +<br/>Materials and Systems (CEHMS), +<br/>Bio-Inspired Materials and +<br/>Devices Laboratory (BMDL), +<br/>Center for Intelligent Material +<br/>Systems and Structure (CIMSS), +<br/>Department of Mechanical Engineering, +<br/>Virginia Tech, +<br/>Blacksburg, VA 24061 +<br/>Graphical Facial Expression +<br/>Analysis and Design Method: +<br/>An Approach to Determine +<br/>Humanoid Skin Deformation +<br/>The architecture of human face is complex consisting of 268 voluntary muscles that perform +<br/>coordinated action to create real-time facial expression. In order to replicate facial expres- +<br/>sion on humanoid face by utilizing discrete actuators, the first and foremost step is the identi- +<br/>fication of a pair of origin and sinking points (SPs). In this paper, we address this issue and +<br/>present a graphical analysis technique that could be used to design expressive robotic faces. +<br/>The underlying criterion in the design of faces being deformation of a soft elastomeric skin +<br/>through tension in anchoring wires attached on one end to the skin through the sinking point +<br/>and the other end to the actuator. The paper also addresses the singularity problem of facial +<br/>control points and important phenomena such as slacking of actuators. Experimental charac- +<br/>terization on a prototype humanoid face was performed to validate the model and demon- +<br/>strate the applicability on a generic platform. [DOI: 10.1115/1.4006519] +<br/>Keywords: humanoid prototype, facial expression, artificial skin, contractile actuator, +<br/>graphical analysis +<br/>Introduction +<br/>Facial expression of humanoid is becoming a key research topic +<br/>in recent years in the areas of social robotics. The embodiment of +<br/>robotic head akin to that of human being promotes a more friendly +<br/>communication between the humanoid and the user. There are +<br/>many challenges in realizing human-like face such as material +<br/>suitable for artificial skin, muscles, sensors, supporting structures, +<br/>machine elements, vision, and audio systems. In addition to mate- +<br/>rials and their integration, computational tools, static and dynamic +<br/>analysis are required to fully understand the effect of each param- +<br/>eter on the overall performance of a prototype humanoid face and +<br/>provide optimum condition. +<br/>This paper is organized in eight sections. First, we introduce the +<br/>background and methodology for creating facial expression in +<br/>robotic heads. A thorough description of the overall problem asso- +<br/>ciated with expression analysis is presented along with pictorial +<br/>representation of the muscle arrangement on a prototype face. +<br/>Second, a literature survey is presented on facial expression analy- +<br/>sis techniques applied to humanoid head. Third, the description of +<br/>graphical facial expression analysis and design (GFEAD) method +<br/>is presented focusing on two generic cases. Fourth, application +<br/>of the GFEAD method on a prototype skull is presented and +<br/>important manifestations that could not be obtained with other +<br/>techniques are discussed. Fifth, results from experimental charac- +<br/>terization of facial movement with a skin layer are discussed. +<br/>Sixth, the effect of the skin properties and associated issues will +<br/>be discussed. Section 7 discusses the significance of GFEAD +<br/>method on practical platforms. Finally, the summary of this study +<br/>is presented in Sec. 8. +<br/>In the last few years, we have demonstrated humanoid heads +<br/>using a variety of actuation technologies including: piezoelectric +<br/>ultrasonic motors for actuation and macrofiber composite for sens- +<br/>ing [1]; electromagnetic RC servo motor for actuation and embed- +<br/><b>University of Texas at</b><br/>Dallas 800 West Campbell Rd., Richardson, TX 75080. +<br/>2Corresponding author. +<br/>Contributed by the Mechanisms and Robotics Committee of ASME for publica- +<br/>tion in the JOURNAL OF MECHANISMS AND ROBOTICS. Manuscript received October 10, +<br/>2010; final manuscript received February 23, 2012; published online April 25, 2012. +<br/>Assoc. Editor: Qiaode Jeffrey Ge. +<br/>ded unimorph for sensing [2,3], and recently shape memory alloy +<br/>(SMA) based actuation for baby humanoid robot focusing on the +<br/>face and jaw movement [4]. We have also reported facial muscles +<br/>based on conducting polymer actuators to overcome the high +<br/>power requirement of current actuation technologies including +<br/>polypyrrole–polyvinylidene difluoride composite stripe and zig- +<br/>zag actuators [5] and axial type helically wounded polypyrrole– +<br/>platinum composite actuators [6]. All these studies have identified +<br/>the issues related to the design of facial structure and artificial +<br/>muscle requirements. Other types of actuators such as dielectric +<br/>elastomer were also studied for general robotics application [7]. +<br/>There are several other studies reported in literature related to +<br/>humanoid facial expression. Facial expression generation and ges- +<br/>ture synthesis from sign language has been applied in the animation +<br/>of an avatar [8], expressive humanoid robot Albert-HUBO with 31 +<br/>Degree of Freedom (DOF) head and 35 DOF body motions based +<br/>on servo motors [9], facial expression imitation system for face rec- +<br/>ognition and implementation on mascot type robotic system [10], +<br/>facial expressive humanoid robot SAYA based on McKibben pneu- +<br/>matic actuators [11], and android robot Repliee for studying psy- +<br/>chological aspects [12]. However, none of these studies address the +<br/>design strategy for humanoid head based on discrete actuators. +<br/>Computational tools for precise analysis of the effect of actuator +<br/>arrangement on the facial expression are missing. +<br/>Even though significant efforts have been made, there is little +<br/>fundamental understanding of the structural design questions. +<br/>How these facial expressions can be precisely designed? How are +<br/>the terminating points on the skull determined? What will be the +<br/>effect of variation in arrangement of actuators? The answer to +<br/>these questions requires the development of an accurate mathe- +<br/>matical model that can be easily coded and visualized. For this +<br/>purpose, we present a GFEAD method for application in human- +<br/>oid head development. This method will be briefly discussed for +<br/>generic cases to illustrate all the computational steps. +<br/>The prime motivation behind using the graphical approach is +<br/>that it provides both visual information as well as quantitative +<br/>data required for the design and analysis of humanoid face. The +<br/>deformation analysis and design is performed directly on the skull +<br/>surface, which ultimately forms the platform for actuation. The +<br/>graphical approach is simple to implement as it is conducted in +<br/>2D. Generally, the skull is created from a scanned model; thus, +<br/>Journal of Mechanisms and Robotics +<br/>Copyright VC 2012 by ASME +<br/>MAY 2012, Vol. 4 / 021010-1 +</td><td>('2248772', 'Yonas Tadesse', 'yonas tadesse')<br/>('25310631', 'Shashank Priya', 'shashank priya')</td><td>e-mail: yonas@vt.edu; +<br/>yonas.tadesse@utdallas.edu +<br/>e-mail: spriya@vt.edu +</td></tr><tr><td>60b3601d70f5cdcfef9934b24bcb3cc4dde663e7</td><td>SUBMITTED TO IEEE TRANS. ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +<br/>Binary Gradient Correlation Patterns +<br/>for Robust Face Recognition +</td><td>('1739171', 'Weilin Huang', 'weilin huang')<br/>('1709042', 'Hujun Yin', 'hujun yin')</td><td></td></tr><tr><td>60737db62fb5fab742371709485e4b2ddf64b7b2</td><td>Crowdsourced Selection on Multi-Attribute Data +<br/><b>Tsinghua University</b></td><td>('39163188', 'Xueping Weng', 'xueping weng')<br/>('23492509', 'Guoliang Li', 'guoliang li')<br/>('1802748', 'Huiqi Hu', 'huiqi hu')<br/>('33091680', 'Jianhua Feng', 'jianhua feng')</td><td>wxp15@mails.tsinghua.edu.cn, liguoliang@tsinghua.edu.cn, hqhu@sei.ecnu.edu.cn, fengjh@tsinghua.edu.cn +</td></tr><tr><td>60496b400e70acfbbf5f2f35b4a49de2a90701b5</td><td>Avoiding Boosting Overfitting by Removing Confusing +<br/>Samples +<br/><b>Moscow State University, dept. of Computational Mathematics and Cybernetics</b><br/>Graphics and Media Lab +<br/>119992 Moscow, Russia +</td><td>('2918740', 'Alexander Vezhnevets', 'alexander vezhnevets')<br/>('3319972', 'Olga Barinova', 'olga barinova')</td><td>{avezhnevets, obarinova}@graphics.cs.msu.ru +</td></tr><tr><td>60bffecd79193d05742e5ab8550a5f89accd8488</td><td>PhD Thesis Proposal +<br/>Classification using sparse representation and applications to skin +<br/>lesion diagnosis +<br/>In only a few decades, sparse representation modeling has undergone a tremendous expansion with +<br/>successful applications in many fields including signal and image processing, computer science, +<br/>machine learning, statistics. Mathematically, it can be considered as the problem of finding the +<br/>sparsest solution (the one with the fewest non-zeros entries) to an underdetermined linear system +<br/>of equations [1]. Based on the observation for natural images (or images rich in textures) that small +<br/>scale structures tend to repeat themselves in an image or in a group of similar images, a signal +<br/>source can be sparsely represented over some well-chosen redundant basis (a dictionary). In other +<br/>words, it can be approximately representable by a linear combination of a few elements (also called +<br/>atoms or basis vectors) of a redundant/over-complete dictionary. +<br/>Such models have been proven successful in many tasks including denoising [2]-[5], compression +<br/>[6],[7], super-resolution [8],[9], classification and pattern recognition [10]-[16]. In the context of +<br/>classification, the objective is to find the class to which a test signal belongs, given training data +<br/>from multiple classes. Sparse representation has become a powerful technique in classification and +<br/><b>applications, including texture classification [16], face recognition [12], object detection [10], and</b><br/>segmentation of medical images [17], [18]. In conventional Sparse Representation Classification +<br/>(SRC) schemes, learned dictionaries and sparse representation are involved to classify image pixels +<br/>(the image is divided into patches surrounding each image pixel). The performance of a SRC relies +<br/>on a good dictionary, and on the sparse representation optimization model. Typically, a dictionary +<br/>is learned for each signal class using training data, and classification of a new signal is achieved +<br/>by associating it with the class whose dictionary allows the best approximation of the signal via an +<br/>optimization problem that minimize the reconstruction error under some constraints including the +<br/>sparsity one. It is important to note that the dictionary may not be a trained one [12]. In [12], the +<br/>dictionary used for the face recognition is composed of many face images. Generally, the +<br/>classification methods consider sparse modeling of natural high-dimensional signals and assume +<br/>that the data belonging to the same class lie in the same subspace of a much lower dimension. Thus, +<br/>the data can be modeled as a union of low dimensional linear subspaces. Then a union of a small +<br/>subset of these linear subspaces is found to be a model of each class [19]. More advanced methods +<br/>take into account the multi-subspace structure of the data of a high dimensional space. That is the +<br/>case when data in multiple classes lie in multiple low-dimensional subspaces. Then, the +<br/>classification problem can be formulated via a structured sparsity-based model, or group sparsity +<br/>one [13, 20]. Other approach proposed to increase the performance of classification by using +<br/>multiple disjoint sparse representation for the dictionary of each class instead of a single signal +<br/>representation [21]. +<br/>II. Objective +<br/>In this study, we focus on a highly accurate classification methods by sparse representation in order +<br/>to improve existing methods. More specifically, we aim to improve the result of classification for +<br/>-1- +</td><td></td><td></td></tr><tr><td>601834a4150e9af028df90535ab61d812c45082c</td><td>A short review and primer on using video for +<br/>psychophysiological observations in +<br/>human-computer interaction applications +<br/><b>Quanti ed Employee unit, Finnish Institute of Occupational Health</b><br/>POBox 40, 00250, Helsinki, Finland +</td><td>('2612057', 'Teppo Valtonen', 'teppo valtonen')</td><td>teppo. valtonen @ttl. fi, +</td></tr><tr><td>346dbc7484a1d930e7cc44276c29d134ad76dc3f</td><td><b></b><br/>On: 21 November 2007 +<br/>Access Details: [subscription number 785020433] +<br/>Publisher: Informa Healthcare +<br/>Informa Ltd Registered in England and Wales Registered Number: 1072954 +<br/>Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK +<br/>Systems +<br/><b>Publication details, including instructions for authors and subscription information</b><br/>http://www.informaworld.com/smpp/title~content=t713663148 +<br/>Artists portray human faces with the Fourier statistics of +<br/>complex natural scenes +<br/><b>a Institute of Anatomy I, School of Medicine, Friedrich Schiller University, Germany</b><br/><b>Friedrich Schiller University, D-07740 Jena</b><br/>Germany +<br/>First Published on: 28 August 2007 +<br/>To cite this Article: Redies, Christoph, Hänisch, Jan, Blickhan, Marko and Denzler, +<br/>Joachim (2007) 'Artists portray human faces with the Fourier statistics of complex +<br/>To link to this article: DOI: 10.1080/09548980701574496 +<br/>URL: http://dx.doi.org/10.1080/09548980701574496 +<br/>PLEASE SCROLL DOWN FOR ARTICLE +<br/>Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf +<br/>This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction, +<br/>re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly +<br/>forbidden. +<br/>The publisher does not give any warranty express or implied or make any representation that the contents will be +<br/>complete or accurate or up to date. The accuracy of any instructions, +<br/>formulae and drug doses should be +<br/>independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, +<br/>demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or +<br/>arising out of the use of this material. +</td><td>('2485437', 'Christoph Redies', 'christoph redies')</td><td></td></tr><tr><td>34a41ec648d082270697b9ee264f0baf4ffb5c8d</td><td></td><td></td><td></td></tr><tr><td>34b3b14b4b7bfd149a0bd63749f416e1f2fc0c4c</td><td>The AXES submissions at TrecVid 2013 +<br/><b>University of Twente 2Dublin City University 3Oxford University</b><br/>4KU Leuven 5Fraunhofer Sankt Augustin 6INRIA Grenoble +</td><td>('3157479', 'Robin Aly', 'robin aly')<br/>('3271933', 'Matthijs Douze', 'matthijs douze')<br/>('1688071', 'Basura Fernando', 'basura fernando')<br/>('9401491', 'Zaid Harchaoui', 'zaid harchaoui')<br/>('1767756', 'Kevin McGuinness', 'kevin mcguinness')<br/>('3095774', 'Dan Oneata', 'dan oneata')<br/>('3188342', 'Omkar M. Parkhi', 'omkar m. parkhi')<br/>('2319574', 'Danila Potapov', 'danila potapov')<br/>('3428663', 'Jérôme Revaud', 'jérôme revaud')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')<br/>('1809436', 'Jochen Schwenninger', 'jochen schwenninger')<br/>('1783430', 'David Scott', 'david scott')<br/>('1704728', 'Tinne Tuytelaars', 'tinne tuytelaars')<br/>('34602236', 'Jakob Verbeek', 'jakob verbeek')<br/>('40465030', 'Heng Wang', 'heng wang')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>34bb11bad04c13efd575224a5b4e58b9249370f3</td><td>Towards Good Practices for Action Video Encoding +<br/>National Key Laboratory for Novel Software Technology +<br/><b>Nanyang Technological University</b><br/><b>Shanghai Jiao Tong University</b><br/><b>Nanjing University, China</b><br/>Singapore +<br/>China +</td><td>('1808816', 'Jianxin Wu', 'jianxin wu')<br/>('22183596', 'Yu Zhang', 'yu zhang')<br/>('8131625', 'Weiyao Lin', 'weiyao lin')</td><td>wujx2001@nju.edu.cn +<br/>roykimbly@hotmail.com +<br/>wylin@sjtu.edu.cn +</td></tr><tr><td>3411ef1ff5ad11e45106f7863e8c7faf563f4ee1</td><td>Image Retrieval and Ranking via Consistently +<br/>Reconstructing Multi-attribute Queries +<br/><b>School of Computer Science and Technology, Tianjin University, Tianjin, China</b><br/>2 State Key Laboratory of Information Security, IIE, Chinese Academy of Sciences, China +<br/><b>National University of Singapore</b><br/>4 State Key Laboratory of Virtual Reality Technology and Systems School of Computer Science +<br/><b>and Engineering, Beihang University, Beijing, China</b></td><td>('1719250', 'Xiaochun Cao', 'xiaochun cao')<br/>('38188331', 'Hua Zhang', 'hua zhang')<br/>('33465926', 'Xiaojie Guo', 'xiaojie guo')<br/>('2705801', 'Si Liu', 'si liu')<br/>('33610144', 'Xiaowu Chen', 'xiaowu chen')</td><td>caoxiaochun@iie.ac.cn, huazhang@tju.edu.cn, xj.max.guo@gmail.com, +<br/>dcslius@nus.edu.sg, chen@buaa.edu.cn +</td></tr><tr><td>345cc31c85e19cea9f8b8521be6a37937efd41c2</td><td>Deep Manifold Traversal: Changing Labels with +<br/>Convolutional Features +<br/><b>Cornell University, Washington University in St. Louis</b><br/>*Authors contributing equally +</td><td>('31693738', 'Jacob R. Gardner', 'jacob r. gardner')<br/>('3222840', 'Paul Upchurch', 'paul upchurch')<br/>('1940272', 'Matt J. Kusner', 'matt j. kusner')<br/>('7769997', 'Yixuan Li', 'yixuan li')<br/>('1706504', 'John E. Hopcroft', 'john e. hopcroft')</td><td></td></tr><tr><td>34d484b47af705e303fc6987413dc0180f5f04a9</td><td>RI:Medium: Unsupervised and Weakly-Supervised +<br/>Discovery of Facial Events +<br/>1 Introduction +<br/>The face is one of the most powerful channels of nonverbal communication. Facial expression has been a +<br/>focus of emotion research for over a hundred years [11]. It is central to several leading theories of emotion +<br/>[16, 28, 44] and has been the focus of at times heated debate about issues in emotion science [17, 23, 40]. +<br/><b>Facial expression gures prominently in research on almost every aspect of emotion, including psychophys</b><br/>iology [30], neural correlates [18], development [31], perception [4], addiction [24], social processes [26], +<br/>depression [39] and other emotion disorders [46], to name a few. In general, facial expression provides cues +<br/>about emotional response, regulates interpersonal behavior, and communicates aspects of psychopathology. +<br/>While people have believed for centuries that facial expressions can reveal what people are thinking and +<br/>feeling, it is relatively recently that the face has been studied scientifically for what it can tell us about +<br/>internal states, social behavior, and psychopathology. +<br/>Faces possess their own language. Beginning with Darwin and his contemporaries, extensive efforts +<br/>have been made to manually describe this language. A leading approach, the Facial Action Coding System +<br/>(FACS) [19] , segments the visible effects of facial muscle activation into ”action units.” Because of its +<br/>descriptive power, FACS has become the state of the art in manual measurement of facial expression and is +<br/>widely used in studies of spontaneous facial behavior. The FACS taxonomy was develop by manually ob- +<br/>serving graylevel variation between expressions in images and to a lesser extent by recording the electrical +<br/>activity of underlying facial muscles [9]. Because of its importance to human social dynamics, person per- +<br/>ception, biological bases of behavior, extensive efforts have been made to automatically detect this language +<br/>(i.e., facial expression) using computer vision and machine learning. In part for these reasons, much effort +<br/>in automatic facial image analysis seeks to automatically recognize FACS action units [5, 45, 38, 42]. With +<br/>few exceptions, previous work on facial expression has been supervised in nature (i.e. event categories are +<br/>defined in advance in labeled training data, see [5, 45, 38, 42] for a review of state-of-the-art algorithms) +<br/>using either FACS or emotion labels (e.g. angry, surprise, sad). Because manual coding is highly labor +<br/>intensive, progress in automated facial image analysis has been limited by lack of sufficient training data +<br/>especially with respect to human behavior in naturally occurring settings (as opposed to posed facial be- +<br/>havior). Little attention has been paid to the problem of unsupervised or weakly-supervised discovery of +<br/>facial events prior to recognition. In this proposal we question whether the reliance on supervised learning +<br/>is necessary. Specifically, Can unsupervised or weakly-supervised learning algorithms discover useful and +<br/>meaningful facial events in video sequences of natural occurring behavior?. Three are the main contributions +<br/>of this proposal: +<br/>• We ask whether unsupervised or weakly-supervised learning algorithms can discover useful and +<br/>meaningful facial events in video sequences of one or more persons with natural occurring behavior. +<br/>Several issues contribute to the challenge of discovery of facial events; these include the large vari- +<br/>ability in the temporal scale and periodicity of facial expressions, illumination and fast pose changes, +<br/>the complexity of decoupling rigid and non-rigid motion from video, the exponential nature of all +<br/>possible facial movement combinations, and characterization of subtle facial behavior. +<br/>• We propose two novel non-parametric algorithms for unsupervised and weakly-supervised time-series +<br/>analysis. In preliminary experiments these algorithms were able to discover meaningful facial events +</td><td></td><td></td></tr><tr><td>341002fac5ae6c193b78018a164d3c7295a495e4</td><td>von Mises-Fisher Mixture Model-based Deep +<br/>learning: Application to Face Verification +</td><td>('1773090', 'Md. Abul Hasnat', 'md. abul hasnat')<br/>('34767162', 'Jonathan Milgram', 'jonathan milgram')<br/>('34086868', 'Liming Chen', 'liming chen')</td><td></td></tr><tr><td>34ce703b7e79e3072eed7f92239a4c08517b0c55</td><td>What impacts skin color in digital photos? +<br/><b>Advanced Digital Sciences Center, University of Illinois at Urbana-Champaign, Singapore</b></td><td>('3213946', 'Albrecht Lindner', 'albrecht lindner')<br/>('1702224', 'Stefan Winkler', 'stefan winkler')</td><td></td></tr><tr><td>345bea5f7d42926f857f395c371118a00382447f</td><td>Transfiguring Portraits +<br/><b>Computer Science and Engineering, University of Washington</b><br/>Figure 1: Our system’s goal is to let people imagine and explore how they may look like in a different country, era, hair style, hair color, age, +<br/>and anything else that can be queried in an image search engine. The examples above show a single input photo (left) and automatically +<br/>synthesized appearances of the input person with ”curly hair” (top row), in ”india” (2nd row), and at ”1930” (3rd row). +</td><td>('2419955', 'Ira Kemelmacher-Shlizerman', 'ira kemelmacher-shlizerman')</td><td></td></tr><tr><td>34ec83c8ff214128e7a4a4763059eebac59268a6</td><td>Action Anticipation By Predicting Future +<br/>Dynamic Images +<br/>Australian Centre for Robotic Vision, ANU, Canberra, Australia +</td><td>('46771280', 'Cristian Rodriguez', 'cristian rodriguez')<br/>('1688071', 'Basura Fernando', 'basura fernando')<br/>('40124570', 'Hongdong Li', 'hongdong li')</td><td>{cristian.rodriguez, basura.fernando, hongdong.li}@.anu.edu.au +</td></tr><tr><td>3463f12ad434d256cd5f94c1c1bfd2dd6df36947</td><td>Article +<br/>Facial Expression Recognition with Fusion Features +<br/>Extracted from Salient Facial Areas +<br/><b>School of Control Science and Engineering, Shandong University, Jinan 250061, China</b><br/>Academic Editors: Xue-Bo Jin; Shuli Sun; Hong Wei and Feng-Bao Yang +<br/>Received: 23 January 2017; Accepted: 24 March 2017; Published: 29 March 2017 +</td><td>('7895427', 'Yanpeng Liu', 'yanpeng liu')<br/>('29275442', 'Yibin Li', 'yibin li')<br/>('1708045', 'Xin Ma', 'xin ma')<br/>('1772484', 'Rui Song', 'rui song')</td><td>liuyanpeng@sucro.org (Y.L.); liyb@sdu.edu.cn (Y.L.); maxin@sdu.edu.cn (X.M.) +<br/>* Correspondence: rsong@sdu.edu.cn +</td></tr><tr><td>346c9100b2fab35b162d7779002c974da5f069ee</td><td>Photo Search by Face Positions and Facial Attributes +<br/>on Touch Devices +<br/><b>National Taiwan University, Taipei, Taiwan</b></td><td>('2476032', 'Yu-Heng Lei', 'yu-heng lei')<br/>('35081710', 'Yan-Ying Chen', 'yan-ying chen')<br/>('2817570', 'Lime Iida', 'lime iida')<br/>('33970300', 'Bor-Chun Chen', 'bor-chun chen')<br/>('1776110', 'Hsiao-Hang Su', 'hsiao-hang su')<br/>('1716836', 'Winston H. Hsu', 'winston h. hsu')</td><td>{limeiida, siriushpa}@gmail.com, b95901019@ntu.edu.tw, winston@csie.ntu.edu.tw +<br/>{ryanlei, yanying}@cmlab.csie.ntu.edu.tw, +</td></tr><tr><td>34863ecc50722f0972e23ec117f80afcfe1411a9</td><td>An Efficient Face Recognition Algorithm Based +<br/>on Robust Principal Component Analysis +<br/>TNLIST and Department of Automation +<br/><b>Tsinghua University</b><br/>Beijing, China +</td><td>('2860279', 'Ziheng Wang', 'ziheng wang')<br/>('2842970', 'Xudong Xie', 'xudong xie')</td><td> zihengwang.thu@gmail.com, xdxie@tsinghua.edu.cn +</td></tr><tr><td>34b7e826db49a16773e8747bc8dfa48e344e425d</td><td></td><td></td><td></td></tr><tr><td>34c594abba9bb7e5813cfae830e2c4db78cf138c</td><td>Transport-Based Single Frame Super Resolution of Very Low Resolution Face Images +<br/><b>Carnegie Mellon University</b><br/>We describe a single-frame super-resolution method for reconstructing high- +<br/>resolution (abbr. high-res) faces from very low-resolution (abbr. low-res) +<br/>face images (e.g. smaller than 16× 16 pixels) by learning a nonlinear La- +<br/>grangian model for the high-res face images. Our technique is based on the +<br/>mathematics of optimal transport, and hence we denote it as transport-based +<br/>SFSR (TB-SFSR). In the training phase, a nonlinear model of high-res fa- +<br/>cial images is constructed based on transport maps that morph a reference +<br/>image into the training face images. In the testing phase, the resolution of +<br/>a degraded image is enhanced by finding the model parameters that best fit +<br/>the given low resolution data. +<br/>Generally speaking, most SFSR methods [2, 3, 4, 5] are based on a +<br/>linear model for the high-res images. Hence, ultimately, the majority of +<br/>SFSR models in the literature can be written as, Ih(x) = ∑i wiψi(x), where +<br/>Ih is a high-res image or a high-res image patch, w’s are weight coefficients, +<br/>and ψ’s are high-res images (or image patches), which are learned from the +<br/>training images using a specific model. Here we propose a fundamentally +<br/>different approach toward modeling high-res images. In our approach the +<br/>high-res image is modeled as a mass preserving mapping of a high-res tem- +<br/>plate image, I0, as follows +<br/>Ih(x) = det(I +∑ +<br/>αiDvi(x))I0(x +∑ +<br/>αivi(x)), +<br/>(1) +<br/>where I is the identity matrix, αi is the weight coefficient of displacement +<br/>field vi (i.e. a smooth vector field), and Dvi(x) is the Jacobian matrix of the +<br/>displacement field vi, evaluated at x. The proposed method can be viewed +<br/>as a linear modeling in the space of mass-preserving mappings, which cor- +<br/>responds to a non-linear model in the image space. Thus (through the use of +<br/>the optimal mapping function f(x) = x +∑i αivi(x)) our modeling approach +<br/>can also displace pixels, in addition to changing their intensities. +<br/>Given a training set of high-res face images, I1, ...,IN : Ω → R with +<br/>Ω = [0,1]2 the image intensities are first normalized to integrate to 1. This +<br/>is done so the images can be treated as distributions of a fixed amount of in- +<br/>tensity values (i.e. fixed amount of mass). Next, the reference face is defined +<br/>to be the average image, I0 = 1 +<br/>i=1 Ii, and the optimal transport distance +<br/>between the reference image and the i’th training image, Ii, is defined to be, +<br/>N ∑N +<br/>(cid:90) +<br/>dOT (I0,Ii) = minui +<br/>|ui(x)|2Ii(x)dx +<br/>s.t. det(I + Dui(x))I0(x + ui(x)) = Ii(x) +<br/>(2) +<br/>where (f(x) = x + u(x)) : Ω → Ω is a mass preserving transform from Ii to +<br/>I0, u is the optimal displacement field, and Dui is the Jacobian matrix of +<br/>u. The optimization problem above is well posed and has a unique min- +<br/>imizer [1]. Having optimal displacement fields ui for i = 1, . . . ,N a sub- +<br/>space, V , is learned for these displacement fields. Let v j for j = 1, ...,M +<br/>be a basis for subspace V. Then, any combination of the basis displacement +<br/>fields can be used to construct an arbitrary deformation field, fα (x) = x + +<br/>∑M +<br/>j=1 α jv j(x), which can then be used to construct a given image Iα (x) = +<br/>det(Dfα (x))I0(fα (x)). Hence, subspace V provides a generative model for +<br/>the high-res face image. +<br/>In the testing phase, we constrain the space of +<br/>possible high-res solutions to those, which are representable as Iα for some +<br/>α ∈ RM. Hence, for a degraded input image, Il, and assuming that φ (.) is +<br/>known and following the MAP criteria we can write, +<br/>α∗ = argminα +<br/>(cid:107)Il − φ (Iα )(cid:107)2 +<br/>(3) +<br/>where a gradient descent approach is used to obtain a local optima α∗. Note +<br/>that, images of faces (and other deformable objects) differ from each other +<br/>s.t Iα (x) = det(Dfα (x))I0(fα (x)) +</td><td>('2062432', 'Soheil Kolouri', 'soheil kolouri')<br/>('1818350', 'Gustavo K. Rohde', 'gustavo k. rohde')</td><td></td></tr><tr><td>34108098e1a378bc15a5824812bdf2229b938678</td><td>Reconstructive Sparse Code Transfer for +<br/>Contour Detection and Semantic Labeling +<br/>1TTI Chicago +<br/><b>California Institute of Technology</b><br/><b>University of California at Berkeley / ICSI</b></td><td>('1965929', 'Michael Maire', 'michael maire')<br/>('2251428', 'Stella X. Yu', 'stella x. yu')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>mmaire@ttic.edu, stellayu@berkeley.edu, perona@caltech.edu +</td></tr><tr><td>341ed69a6e5d7a89ff897c72c1456f50cfb23c96</td><td>DAGER: Deep Age, Gender and Emotion +<br/>Recognition Using Convolutional Neural +<br/>Networks +<br/>Computer Vision Lab, Sighthound Inc., Winter Park, FL +</td><td>('1707795', 'Afshin Dehghan', 'afshin dehghan')<br/>('16131262', 'Enrique G. Ortiz', 'enrique g. ortiz')<br/>('37574860', 'Guang Shu', 'guang shu')<br/>('2234898', 'Syed Zain Masood', 'syed zain masood')</td><td>{afshindehghan, egortiz, guangshu, zainmasood}@sighthound.com +</td></tr><tr><td>348a16b10d140861ece327886b85d96cce95711e</td><td>Finding Good Features for Object Recognition +<br/>by +<br/><b>B.S. (Cornell University</b><br/><b>M.S. (University of California, Berkeley</b><br/>A dissertation submitted in partial satisfaction +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>in +<br/>Computer Science +<br/>in the +<br/>GRADUATE DIVISION +<br/>of the +<br/><b>UNIVERSITY OF CALIFORNIA, BERKELEY</b><br/>Committee in charge: +<br/>Professor Jitendra Malik, Chair +<br/>Spring 2005 +</td><td>('3236352', 'Andras David Ferencz', 'andras david ferencz')<br/>('1744452', 'David A. Forsyth', 'david a. forsyth')<br/>('1678771', 'Peter J. Bickel', 'peter j. bickel')</td><td></td></tr><tr><td>3419af6331e4099504255a38de6f6b7b3b1e5c14</td><td>Modified Eigenimage Algorithm for Painting +<br/>Image Retrieval +<br/><b>Stanford University</b><br/> +</td><td>('12833413', 'Qun Feng Tan', 'qun feng tan')</td><td></td></tr><tr><td>34c8de02a5064e27760d33b861b7e47161592e65</td><td>Video Action Recognition based on Deeper Convolution Networks with +<br/>Pair-Wise Frame Motion Concatenation +<br/><b>School of Computer Science, Northwestern Polytechnical University, China</b><br/><b>Sensor-enhanced Social Media (SeSaMe) Centre, National University of Singapore, Singapore</b><br/><b>School of Information Engineering, Nanchang University, China</b></td><td>('9229148', 'Yamin Han', 'yamin han')<br/>('40188000', 'Peng Zhang', 'peng zhang')<br/>('2628886', 'Tao Zhuo', 'tao zhuo')<br/>('1730584', 'Wei Huang', 'wei huang')<br/>('1801395', 'Yanning Zhang', 'yanning zhang')</td><td></td></tr><tr><td>340d1a9852747b03061e5358a8d12055136599b0</td><td>Audio-Visual Recognition System Insusceptible +<br/>to Illumination Variation over Internet Protocol +<br/> +</td><td>('1968167', 'Yee Wan Wong', 'yee wan wong')</td><td></td></tr><tr><td>34ccdec6c3f1edeeecae6a8f92e8bdb290ce40fd</td><td>Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-16) +<br/>A Virtual Assistant to Help Dysphagia Patients Eat Safely at Home +<br/><b>SRI International, Menlo Park California / *Brooklyn College, Brooklyn New York</b><br/> +</td><td>('6647218', 'Michael Freed', 'michael freed')<br/>('1936842', 'Brian Burns', 'brian burns')<br/>('39451362', 'Aaron Heller', 'aaron heller')<br/>('3431324', 'Sharon Beaumont-Bowman', 'sharon beaumont-bowman')</td><td>{first name, last name}@sri.com, sharonb@brooklyn.cuny.edu +</td></tr><tr><td>34b42bcf84d79e30e26413f1589a9cf4b37076f9</td><td>Learning Sparse Representations of High +<br/>Dimensional Data on Large Scale Dictionaries +<br/><b>Princeton University</b><br/>Princeton, NJ 08544, USA +</td><td>('1730249', 'Zhen James Xiang', 'zhen james xiang')<br/>('1693135', 'Peter J. Ramadge', 'peter j. ramadge')</td><td>{zxiang,haoxu,ramadge}@princeton.edu +</td></tr><tr><td>5a3da29970d0c3c75ef4cb372b336fc8b10381d7</td><td>CNN-based Real-time Dense Face Reconstruction +<br/>with Inverse-rendered Photo-realistic Face Images +</td><td>('8280113', 'Yudong Guo', 'yudong guo')<br/>('2938279', 'Juyong Zhang', 'juyong zhang')<br/>('1688642', 'Jianfei Cai', 'jianfei cai')<br/>('15679675', 'Boyi Jiang', 'boyi jiang')<br/>('48510441', 'Jianmin Zheng', 'jianmin zheng')</td><td></td></tr><tr><td>5a93f9084e59cb9730a498ff602a8c8703e5d8a5</td><td>HUSSAIN ET. AL: FACE RECOGNITION USING LOCAL QUANTIZED PATTERNS +<br/>Face Recognition using Local Quantized +<br/>Patterns +<br/>Fréderic Jurie +<br/>GREYC — CNRS UMR 6072, +<br/><b>University of Caen Basse-Normandie</b><br/>Caen, France +</td><td>('2695106', 'Sibt ul Hussain', 'sibt ul hussain')<br/>('3423479', 'Thibault Napoléon', 'thibault napoléon')</td><td>Sibt.ul.Hussain@gmail.com +<br/>Thibault.Napoleon@unicaen.fr +<br/>Frederic.Jurie@unicaen.fr +</td></tr><tr><td>5a87bc1eae2ec715a67db4603be3d1bb8e53ace2</td><td>A Novel Convergence Scheme for Active Appearance Models +<br/>School of Electrical and Computer Engineering +<br/><b>Georgia Institute of Technology</b><br/>Atlanta, GA 30332 +</td><td>('38410822', 'Aziz Umit Batur', 'aziz umit batur')<br/>('2583044', 'Monson H. Hayes', 'monson h. hayes')</td><td>{batur, mhh3}@ece.gatech.edu +</td></tr><tr><td>5aad56cfa2bac5d6635df4184047e809f8fecca2</td><td>A VISUAL DICTIONARY ATTACK ON PICTURE PASSWORDS +<br/><b>Cornell University</b></td><td>('1803066', 'Amir Sadovnik', 'amir sadovnik')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td></td></tr><tr><td>5a8ca0cfad32f04449099e2e3f3e3a1c8f6541c0</td><td>Available online at www.sciencedirect.com +<br/>ScienceDirect +<br/> Procedia Computer Science 87 ( 2016 ) 300 – 305 +<br/>4th International Conference on Recent Trends in Computer Science &Engineering +<br/>Automatic Frontal Face Reconstruction Approach for Pose Invariant Face +<br/>Recognition +<br/>Kavitha.Ja,Mirnalinee.T.Tb +<br/><b>aResearch Scholar, Anna University, Chennai, Inida</b><br/><b>SSN College of Engineering, Kalavakkam, Tamil Nadu, India</b></td><td></td><td></td></tr><tr><td>5ac80e0b94200ee3ecd58a618fe6afd077be0a00</td><td>Unifying Geometric Features and Facial Action Units for Improved +<br/>Performance of Facial Expression Analysis +<br/><b>Kent State University</b><br/>Keywords: +<br/>Facial Action Unit, Facial Expression, Geometric features. +</td><td>('1688430', 'Mehdi Ghayoumi', 'mehdi ghayoumi')</td><td>{mghayoum,akbansal}@kent.edu +</td></tr><tr><td>5aadd85e2a77e482d44ac2a215c1f21e4a30d91b</td><td>Face Recognition using Principle Components and Linear +<br/>Discriminant Analysis +<br/>HATIM A. +<br/>ABOALSAMH 1,2 +<br/>HASSAN I. +<br/>MATHKOUR 1,2 +<br/>GHAZY M.R. +<br/>ASSASSA 1,2 +<br/>MONA F.M. +<br/>MURSI 1,3 +<br/>1 Center of Excellence in Information Assurance (CoEIA), +<br/>2 Department of Computer Science +<br/>3 Department of Information Technology +<br/><b>College of Computer and Information Sciences</b><br/><b>King Saud University, Riyadh</b><br/>SAUDI ARABIA +</td><td></td><td>hatim@ksu.edu.sa +<br/>mathkour@ksu.edu.sa +<br/>gassassa@coeia.edu.sa +<br/>monmursi@coeia.edu.sa +</td></tr><tr><td>5a34a9bb264a2594c02b5f46b038aa1ec3389072</td><td>Label-Embedding for Image Classification +</td><td>('2893664', 'Zeynep Akata', 'zeynep akata')<br/>('1723883', 'Florent Perronnin', 'florent perronnin')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>5a5f9e0ed220ce51b80cd7b7ede22e473a62062c</td><td>Videos as Space-Time Region Graphs +<br/><b>Robotics Institute, Carnegie Mellon University</b><br/>Figure 1. How do you recognize simple actions such as opening book? We argue action +<br/>understanding requires appearance modeling but also capturing temporal dynamics +<br/>(how shape of book changes) and functional relationships. We propose to represent +<br/>videos as space-time region graphs followed by graph convolutions for inference. +</td><td>('39849136', 'Xiaolong Wang', 'xiaolong wang')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')</td><td></td></tr><tr><td>5ac946fc6543a445dd1ee6d5d35afd3783a31353</td><td>FEATURELESS: BYPASSING FEATURE EXTRACTION IN ACTION CATEGORIZATION +<br/>S. L. Pinteaa, P. S. Mettesa +<br/>J. C. van Gemerta,b, A. W. M. Smeuldersa +<br/>aIntelligent Sensory Information Systems, +<br/><b>University of Amsterdam</b><br/>Amsterdam, Netherlands +</td><td></td><td></td></tr><tr><td>5a4c6246758c522f68e75491eb65eafda375b701</td><td>978-1-4244-4296-6/10/$25.00 ©2010 IEEE +<br/>1118 +<br/>ICASSP 2010 +</td><td></td><td></td></tr><tr><td>5aad5e7390211267f3511ffa75c69febe3b84cc7</td><td>Driver Gaze Estimation +<br/>Without Using Eye Movement +<br/>MIT AgeLab +</td><td>('2145054', 'Lex Fridman', 'lex fridman')<br/>('2180983', 'Philipp Langhans', 'philipp langhans')<br/>('7137846', 'Joonbum Lee', 'joonbum lee')<br/>('1901227', 'Bryan Reimer', 'bryan reimer')</td><td>fridman@mit.edu, philippl@mit.edu, joonbum@mit.edu, reimer@mit.edu +</td></tr><tr><td>5a029a0b0ae8ae7fc9043f0711b7c0d442bfd372</td><td></td><td></td><td></td></tr><tr><td>5ae970294aaba5e0225122552c019eb56f20af74</td><td>International Journal of Computer and Electrical Engineering +<br/>Establishing Dense Correspondence of High Resolution 3D +<br/>Faces via Möbius Transformations +<br/><b>College of Electronic Science and Engineering, National University of Defense Technology, Changsha, China</b><br/>Manuscript submitted July 14, 2014; accepted November 2, 2014. +<br/>doi: 10. 17706/ijcee.2014.v6.866 +</td><td>('30373915', 'Jian Liu', 'jian liu')<br/>('37509862', 'Quan Zhang', 'quan zhang')<br/>('3224964', 'Chaojing Tang', 'chaojing tang')</td><td>* Corresponding author. Email: cjtang@263.net +</td></tr><tr><td>5a86842ab586de9d62d5badb2ad8f4f01eada885</td><td>International Journal of Engineering Research and General Science Volume 3, Issue 3, May-June, 2015 +<br/>ISSN 2091-2730 +<br/>Facial Emotion Recognition and Classification Using Hybridization +<br/>Method +<br/><b>Chandigarh Engg. College, Mohali, Punjab, India</b></td><td>('6010530', 'Anchal Garg', 'anchal garg')<br/>('9744572', 'Rohit Bajaj', 'rohit bajaj')</td><td>anchalgarg949@gmail.com, 07696449500 +</td></tr><tr><td>5a4ec5c79f3699ba037a5f06d8ad309fb4ee682c</td><td>Downloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Electronic-Imaging on 12/17/2017 Terms of Use: https://www.spiedigitallibrary.org/terms-of-use +<br/>AutomaticageandgenderclassificationusingsupervisedappearancemodelAliMainaBukarHassanUgailDavidConnahAliMainaBukar,HassanUgail,DavidConnah,“Automaticageandgenderclassificationusingsupervisedappearancemodel,”J.Electron.Imaging25(6),061605(2016),doi:10.1117/1.JEI.25.6.061605. </td><td></td><td></td></tr><tr><td>5aa57a12444dbde0f5645bd9bcec8cb2f573c6a0</td><td>The International Arab Journal of Information Technology, Vol. 11, No. 2, March 2014 +<br/> +<br/>149 +<br/> +<br/>Face Recognition using Adaptive Margin Fisher’s +<br/>Criterion and Linear Discriminant Analysis +<br/> +<br/>(AMFC-LDA) +<br/><b>COMSATS Institute of Information Technology, Pakistan</b></td><td>('2151799', 'Marryam Murtaza', 'marryam murtaza')<br/>('33088042', 'Muhammad Sharif', 'muhammad sharif')<br/>('36739230', 'Mudassar Raza', 'mudassar raza')<br/>('1814986', 'Jamal Hussain Shah', 'jamal hussain shah')</td><td></td></tr><tr><td>5aed0f26549c6e64c5199048c4fd5fdb3c5e69d6</td><td>International Journal of Computer Applications® (IJCA) (0975 – 8887) +<br/>International Conference on Knowledge Collaboration in Engineering, ICKCE-2014 +<br/>Human Expression Recognition using Facial Features +<br/>G.Saranya +<br/>Post graduate student, Dept. of ECE +<br/><b>Parisutham Institute of Technology and Science</b><br/>Thanjavur. +<br/><b>Affiliated to Anna university, Chennai</b><br/>recognition can be used +</td><td></td><td></td></tr><tr><td>5a7520380d9960ff3b4f5f0fe526a00f63791e99</td><td>The Indian Spontaneous Expression +<br/>Database for Emotion Recognition +</td><td>('38657440', 'Priyadarshi Patnaik', 'priyadarshi patnaik')<br/>('2680543', 'Aurobinda Routray', 'aurobinda routray')<br/>('2730256', 'Rajlakshmi Guha', 'rajlakshmi guha')</td><td></td></tr><tr><td>5a07945293c6b032e465d64f2ec076b82e113fa6</td><td>Pulling Actions out of Context: Explicit Separation for Effective Combination +<br/><b>Stony Brook University, Stony Brook, NY 11794, USA</b></td><td>('50874742', 'Yang Wang', 'yang wang')</td><td>{wang33, minhhoai}@cs.stonybrook.edu +</td></tr><tr><td>5fff61302adc65d554d5db3722b8a604e62a8377</td><td>Additive Margin Softmax for Face Verification +<br/>UESTC +<br/>Georgia Tech +<br/>UESTC +<br/>UESTC +</td><td>('47939378', 'Feng Wang', 'feng wang')<br/>('51094998', 'Weiyang Liu', 'weiyang liu')<br/>('8424682', 'Haijun Liu', 'haijun liu')<br/>('1709439', 'Jian Cheng', 'jian cheng')</td><td>feng.wff@gmail.com +<br/>wyliu@gatech.edu +<br/>haijun liu@126.com +<br/>chengjian@uestc.edu.cn +</td></tr><tr><td>5f771fed91c8e4b666489ba2384d0705bcf75030</td><td>Understanding Humans in Crowded Scenes: Deep Nested Adversarial Learning +<br/>and A New Benchmark for Multi-Human Parsing +<br/><b>National University of Singapore</b><br/><b>National University of Defense Technology</b><br/><b>Qihoo 360 AI Institute</b></td><td>('46509484', 'Jian Zhao', 'jian zhao')<br/>('2757639', 'Jianshu Li', 'jianshu li')<br/>('48207454', 'Li Zhou', 'li zhou')<br/>('1715286', 'Terence Sim', 'terence sim')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')</td><td>chengyu996@gmail.com zhouli2025@gmail.com +<br/>{eleyans, elefjia}@nus.edu.sg +<br/>{zhaojian90, jianshu}@u.nus.edu +<br/>tsim@comp.nus.edu.sg +</td></tr><tr><td>5fa04523ff13a82b8b6612250a39e1edb5066521</td><td>Dockerface: an Easy to Install and Use Faster R-CNN Face Detector in a Docker +<br/>Container +<br/>Center for Behavioral Imaging +<br/><b>College of Computing</b><br/><b>Georgia Institute of Technology</b></td><td>('31601235', 'Nataniel Ruiz', 'nataniel ruiz')<br/>('1692956', 'James M. Rehg', 'james m. rehg')</td><td>nataniel.ruiz@gatech.edu +<br/>rehg@gatech.edu +</td></tr><tr><td>5fa6e4a23da0b39e4b35ac73a15d55cee8608736</td><td>IJCV special issue (Best papers of ECCV 2016) manuscript No. +<br/>(will be inserted by the editor) +<br/>RED-Net: +<br/>A Recurrent Encoder-Decoder Network for Video-based Face Alignment +<br/>Submitted: April 19 2017 / Revised: December 12 2017 +</td><td>('4340744', 'Xi Peng', 'xi peng')</td><td></td></tr><tr><td>5f871838710a6b408cf647aacb3b198983719c31</td><td>1716 +<br/>Locally Linear Regression for Pose-Invariant +<br/>Face Recognition +</td><td>('1695600', 'Xiujuan Chai', 'xiujuan chai')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')<br/>('1698902', 'Wen Gao', 'wen gao')</td><td></td></tr><tr><td>5f64a2a9b6b3d410dd60dc2af4a58a428c5d85f9</td><td></td><td></td><td></td></tr><tr><td>5f344a4ef7edfd87c5c4bc531833774c3ed23542</td><td>c +</td><td></td><td></td></tr><tr><td>5f6ab4543cc38f23d0339e3037a952df7bcf696b</td><td>Video2Vec: Learning Semantic Spatial-Temporal +<br/>Embeddings for Video Representation +<br/>School of Computer Engineering +<br/>School of Electrical Engineering +<br/>School of Computer Science +<br/><b>Arizona State University</b><br/>Tempe, Arizona 85281 +<br/><b>Arizona State University</b><br/>Tempe, Arizona 85281 +<br/><b>Arizona State University</b><br/>Tempe, Arizona 85281 +</td><td>('8060096', 'Sheng-hung Hu', 'sheng-hung hu')<br/>('2180892', 'Yikang Li', 'yikang li')<br/>('2913552', 'Baoxin Li', 'baoxin li')</td><td>Email:shenghun@asu.edu +<br/>Email:yikangli@asu.edu +<br/>Email:Baoxin.Li@asu.edu +</td></tr><tr><td>5f7c4c20ae2731bfb650a96b69fd065bf0bb950e</td><td>Turk J Elec Eng & Comp Sci +<br/>(2016) 24: 1797 { 1814 +<br/>c⃝ T (cid:127)UB_ITAK +<br/>doi:10.3906/elk-1310-253 +<br/>A new fuzzy membership assignment and model selection approach based on +<br/>dynamic class centers for fuzzy SVM family using the (cid:12)re(cid:13)y algorithm +<br/><b>Young Researchers and Elite Club, Mashhad Branch, Islamic Azad University, Mashhad, Iran</b><br/><b>Faculty of Engineering, Ferdowsi University, Mashhad, Iran</b><br/>Received: 01.11.2013 +<br/>(cid:15) +<br/>Accepted/Published Online: 30.06.2014 +<br/>(cid:15) +<br/>Final Version: 23.03.2016 +</td><td>('9437627', 'Omid Naghash Almasi', 'omid naghash almasi')<br/>('4945660', 'Modjtaba Rouhani', 'modjtaba rouhani')</td><td></td></tr><tr><td>5f94969b9491db552ffebc5911a45def99026afe</td><td>Multimodal Learning and Reasoning for Visual +<br/>Question Answering +<br/>Integrative Sciences and Engineering +<br/><b>National University of Singapore</b><br/>Electrical and Computer Engineering +<br/><b>National University of Singapore</b></td><td>('3393294', 'Ilija Ilievski', 'ilija ilievski')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')</td><td>ilija.ilievski@u.nus.edu +<br/>elefjia@nus.edu.sg +</td></tr><tr><td>5f758a29dae102511576c0a5c6beda264060a401</td><td>Fine-grained Video Attractiveness Prediction Using Multimodal +<br/>Deep Learning on a Large Real-world Dataset +<br/><b>Wuhan University, Tencent AI Lab, National University of Singapore, University of Rochester</b></td><td>('3179887', 'Xinpeng Chen', 'xinpeng chen')<br/>('47740660', 'Jingyuan Chen', 'jingyuan chen')<br/>('34264361', 'Lin Ma', 'lin ma')<br/>('1849993', 'Jian Yao', 'jian yao')<br/>('46641573', 'Wei Liu', 'wei liu')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')<br/>('38144094', 'Tong Zhang', 'tong zhang')</td><td></td></tr><tr><td>5fa0e6da81acece7026ac1bc6dcdbd8b204a5f0a</td><td></td><td></td><td></td></tr><tr><td>5feb1341a49dd7a597f4195004fe9b59f67e6707</td><td>A Deep Ranking Model for Spatio-Temporal Highlight Detection +<br/>from a 360◦ Video +<br/><b>Seoul National University</b></td><td>('7877122', 'Youngjae Yu', 'youngjae yu')<br/>('1693291', 'Sangho Lee', 'sangho lee')<br/>('35272603', 'Joonil Na', 'joonil na')<br/>('35365676', 'Jaeyun Kang', 'jaeyun kang')<br/>('1743920', 'Gunhee Kim', 'gunhee kim')</td><td>{yj.yu, sangho.lee, joonil}@vision.snu.ac.kr, {kjy13411}@gmail.com, gunhee@snu.ac.kr +</td></tr><tr><td>5f0d4a0b5f72d8700cdf8cb179263a8fa866b59b</td><td>CBMM Memo No. 85 +<br/>06/2018 +<br/>Deep Regression Forests for Age Estimation +<br/><b>Key Laboratory of Specialty Fiber Optics and Optical Access Networks, Shanghai University</b><br/><b>Johns Hopkins University</b><br/><b>College of Computer and Control Engineering, Nankai University 4: Hikvision Research</b></td><td>('41187410', 'Wei Shen', 'wei shen')<br/>('9544564', 'Yilu Guo', 'yilu guo')<br/>('46394340', 'Yan Wang', 'yan wang')<br/>('1681247', 'Kai Zhao', 'kai zhao')<br/>('46172451', 'Bo Wang', 'bo wang')<br/>('35922327', 'Alan Yuille', 'alan yuille')</td><td></td></tr><tr><td>5f57a1a3a1e5364792b35e8f5f259f92ad561c1f</td><td>Implicit Sparse Code Hashing +<br/><b>Institute of Information Science</b><br/>Academia Sinica, Taiwan +</td><td>('2144284', 'Tsung-Yu Lin', 'tsung-yu lin')<br/>('2301765', 'Tsung-Wei Ke', 'tsung-wei ke')<br/>('1805102', 'Tyng-Luh Liu', 'tyng-luh liu')</td><td></td></tr><tr><td>5f27ed82c52339124aa368507d66b71d96862cb7</td><td>Semi-supervised Learning of Classifiers: Theory, Algorithms +<br/>and Their Application to Human-Computer Interaction +<br/>This work has been partially funded by NSF Grant IIS 00-85980. +<br/>DRAFT +</td><td>('1774778', 'Ira Cohen', 'ira cohen')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>Ira Cohen: Hewlett-Packard Labs, Palo Alto, CA, USA, ira.cohen@hp.com +<br/>Fabio G. Cozman and Marcelo C. Cirelo: Escola Polit´ecnica, Universidade de S˜ao Paulo, S˜ao Paulo,Brazil. fgcozman@usp.br, +<br/>marcelo.cirelo@poli.usp.br +<br/>Nicu Sebe: Faculty of Science, University of Amsterdam, The Netherlands. nicu@science.uva.nl +<br/>Thomas S. Huang: Beckman Institute, University of Illinois at Urbana-Champaign, USA. huang@ifp.uiuc.edu +</td></tr><tr><td>5fa932be4d30cad13ea3f3e863572372b915bec8</td><td></td><td></td><td></td></tr><tr><td>5fea26746f3140b12317fcf3bc1680f2746e172e</td><td>Dense Supervision for Visual Comparisons via Synthetic Images +<br/>Semantic Jitter: +<br/><b>University of Texas at Austin</b><br/><b>University of Texas at Austin</b><br/>Distinguishing subtle differences in attributes is valuable, yet +<br/>learning to make visual comparisons remains non-trivial. Not +<br/>only is the number of possible comparisons quadratic in the +<br/>number of training images, but also access to images adequately +<br/>spanning the space of fine-grained visual differences is limited. +<br/>We propose to overcome the sparsity of supervision problem +<br/>via synthetically generated images. Building on a state-of-the- +<br/>art image generation engine, we sample pairs of training images +<br/>exhibiting slight modifications of individual attributes. Augment- +<br/>ing real training image pairs with these examples, we then train +<br/>attribute ranking models to predict the relative strength of an +<br/>attribute in novel pairs of real images. Our results on datasets of +<br/>faces and fashion images show the great promise of bootstrapping +<br/>imperfect image generators to counteract sample sparsity for +<br/>learning to rank. +<br/>INTRODUCTION +<br/>Fine-grained analysis of images often entails making visual +<br/>comparisons. For example, given two products in a fashion +<br/>catalog, a shopper may judge which shoe appears more pointy +<br/>at the toe. Given two selfies, a teen may gauge in which one he +<br/>is smiling more. Given two photos of houses for sale on a real +<br/>estate website, a home buyer may analyze which facade looks +<br/>better maintained. Given a series of MRI scans, a radiologist +<br/>may judge which pair exhibits the most shape changes. +<br/>In these and many other such cases, we are interested in +<br/>inferring how a pair of images compares in terms of a par- +<br/>ticular property, or “attribute”. That is, which is more pointy, +<br/>smiling, well-maintained, etc. Importantly, the distinctions of +<br/>interest are often quite subtle. Subtle comparisons arise both +<br/>in image pairs that are very similar in almost every regard +<br/>(e.g., two photos of the same individual wearing the same +<br/>clothing, yet smiling more in one photo than the other), as +<br/>well as image pairs that are holistically different yet exhibit +<br/>only slight differences in the attribute in question (e.g., two +<br/>individuals different in appearance, and one is smiling slightly +<br/>more than the other). +<br/>A growing body of work explores computational models +<br/>for visual comparisons [1], [2], [3], [4], [5], [6], [7], [8], [9], +<br/>[10], [11], [12]. In particular, ranking models for “relative +<br/>attributes” [2], [3], [4], [5], [9], [11] use human-ordered pairs +<br/>of images to train a system to predict the relative ordering in +<br/>novel image pairs. +<br/>A major challenge in training a ranking model is the sparsity +<br/>of supervision. That sparsity stems from two factors: label +<br/>availability and image availability. Because training instances +<br/>consist of pairs of images—together with the ground truth +<br/>human judgment about which exhibits the property more +<br/>Fig. 1: Our method “densifies” supervision for training ranking functions to +<br/>make visual comparisons, by generating ordered pairs of synthetic images. +<br/>Here, when learning the attribute smiling, real training images need not be +<br/>representative of the entire attribute space (e.g., Web photos may cluster +<br/>around commonly photographed expressions, like toothy smiles). Our idea +<br/>“fills in” the sparsely sampled regions to enable fine-grained supervision. +<br/>Given a novel pair (top), the nearest synthetic pairs (right) may present better +<br/>training data than the nearest real pairs (left). +<br/>or less—the space of all possible comparisons is quadratic +<br/>in the number of potential +<br/>training images. This quickly +<br/>makes it intractable to label an image collection exhaustively +<br/>for its comparative properties. At the same time, attribute +<br/>comparisons entail a greater cognitive load than, for example, +<br/>object category labeling. Indeed, the largest existing relative +<br/>attribute datasets sample only less than 0.1% of all image pairs +<br/>for ground truth labels [11], and there is a major size gap +<br/>between standard datasets labeled for classification (now in +<br/>the millions [13]) and those for comparisons (at best in the +<br/>thousands [11]). A popular shortcut is to propagate category- +<br/>level comparisons down to image instances [4], [14]—e.g., +<br/>deem all ocean scenes as “more open” than all forest scenes— +<br/>but +<br/>label noise and in practice +<br/>underperforms training with instance-level comparisons [2]. +<br/>this introduces substantial +<br/>Perhaps more insidious than the annotation cost, however, +<br/>is the problem of even curating training images that suf- +<br/>ficiently illustrate fine-grained differences. Critically, sparse +<br/>supervision arises not simply because 1) we lack resources +<br/>to get enough image pairs labeled, but also because 2) we +<br/>lack a direct way to curate photos demonstrating all sorts +<br/>of subtle attribute changes. For example, how might we +<br/>gather unlabeled image pairs depicting all subtle differences +<br/>Novel PairReal PairsSynthetic Pairsvs. </td><td>('2206630', 'Aron Yu', 'aron yu')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>aron.yu@utexas.edu +<br/>grauman@cs.utexas.edu +</td></tr><tr><td>5f5906168235613c81ad2129e2431a0e5ef2b6e4</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>A Unified Framework for Compositional Fitting of +<br/>Active Appearance Models +<br/>Received: date / Accepted: date +</td><td>('2575567', 'Joan Alabort-i-Medina', 'joan alabort-i-medina')</td><td></td></tr><tr><td>5fb5d9389e2a2a4302c81bcfc068a4c8d4efe70c</td><td>Multiple Facial Attributes Estimation based on +<br/>Weighted Heterogeneous Learning +<br/>H.Fukui* T.Yamashita* Y.Kato* R.Matsui* +<br/><b>Chubu University</b><br/>**Abeja Inc. +<br/>1200, Matuoto-cho, Kasugai, +<br/>4-1-20, Toranomon, Minato-ku, +<br/>Aichi, Japan +<br/>Tokyo, Japan +</td><td>('2531207', 'T. Ogata', 't. ogata')</td><td></td></tr><tr><td>5f676d6eca4c72d1a3f3acf5a4081c29140650fb</td><td>To Skip or not to Skip? A Dataset of Spontaneous Affective Response +<br/>of Online Advertising (SARA) for Audience Behavior Analysis +<br/><b>College of Electronics and Information Engineering, Sichuan University, Chengdu 610064, China</b><br/><b>BRIC, University of North Carolina at Chapel Hill, NC 27599, USA</b><br/>3 HP Labs, Palo Alto, CA 94304, USA +<br/><b>Center for Research in Intelligent Systems, University of California, Riverside, CA 92521, USA</b></td><td>('1803478', 'Songfan Yang', 'songfan yang')<br/>('39776603', 'Le An', 'le an')<br/>('1784929', 'Mehran Kafai', 'mehran kafai')<br/>('1707159', 'Bir Bhanu', 'bir bhanu')</td><td>syang@scu.edu.cn, lan004@unc.edu, mehran.kafai@hp.com, bhanu@cris.ucr.edu +</td></tr><tr><td>5f453a35d312debfc993d687fd0b7c36c1704b16</td><td><b>Clemson University</b><br/>TigerPrints +<br/>All Theses +<br/>12-2015 +<br/>Theses +<br/>A Training Assistant Tool for the Automated Visual +<br/>Inspection System +<br/>Follow this and additional works at: http://tigerprints.clemson.edu/all_theses +<br/>Part of the Electrical and Computer Engineering Commons +<br/>Recommended Citation +<br/>Ramaraj, Mohan Karthik, "A Training Assistant Tool for the Automated Visual Inspection System" (2015). All Theses. Paper 2285. +<br/>This Thesis is brought to you for free and open access by the Theses at TigerPrints. It has been accepted for inclusion in All Theses by an authorized +</td><td>('4154752', 'Mohan Karthik Ramaraj', 'mohan karthik ramaraj')</td><td>Clemson University, rmohankarthik91@gmail.com +<br/>administrator of TigerPrints. For more information, please contact awesole@clemson.edu. +</td></tr><tr><td>5fc664202208aaf01c9b62da5dfdcd71fdadab29</td><td>arXiv:1504.05308v1 [cs.CV] 21 Apr 2015 +</td><td></td><td></td></tr><tr><td>5fac62a3de11125fc363877ba347122529b5aa50</td><td>AMTnet: Action-Micro-Tube Regression by +<br/>End-to-end Trainable Deep Architecture +<br/><b>Oxford Brookes University, Oxford, United Kingdom</b></td><td>('3017538', 'Suman Saha', 'suman saha')<br/>('1754181', 'Fabio Cuzzolin', 'fabio cuzzolin')<br/>('1931660', 'Gurkirt Singh', 'gurkirt singh')</td><td>{suman.saha-2014, gurkirt.singh-2015, fabio.cuzzolin}@brookes.ac.uk +</td></tr><tr><td>5fa1724a79a9f7090c54925f6ac52f1697d6b570</td><td>Proceedings of the Workshop on Grammar and Lexicon: Interactions and Interfaces, +<br/>pages 41–47, Osaka, Japan, December 11 2016. +<br/>41 +</td><td></td><td></td></tr><tr><td>5fba1b179ac80fee80548a0795d3f72b1b6e49cd</td><td>Virtual U: Defeating Face Liveness Detection by Building Virtual Models +<br/>From Your Public Photos +<br/><b>University of North Carolina at Chapel Hill</b></td><td>('1734114', 'Yi Xu', 'yi xu')<br/>('39310157', 'True Price', 'true price')<br/>('40454588', 'Jan-Michael Frahm', 'jan-michael frahm')<br/>('1792232', 'Fabian Monrose', 'fabian monrose')</td><td>{yix, jtprice, jmf, fabian}@cs.unc.edu +</td></tr><tr><td>33f7e78950455c37236b31a6318194cfb2c302a4</td><td>Parameterizing Object Detectors +<br/>in the Continuous Pose Space +<br/><b>Boston University, USA</b><br/>2 Disney Research Pittsburgh, USA +</td><td>('1702188', 'Kun He', 'kun he')<br/>('14517812', 'Leonid Sigal', 'leonid sigal')</td><td>{hekun,sclaroff}@cs.bu.edu, lsigal@disneyresearch.com +</td></tr><tr><td>33548531f9ed2ce6f87b3a1caad122c97f1fd2e9</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 104 – No.2, October 2014 +<br/>Facial Expression Recognition in Video using +<br/>Adaboost and SVM +<br/>Surabhi Prabhakar +<br/>Department of CSE +<br/><b>Amity University</b><br/>Noida, India +<br/>Jaya Sharma +<br/>Shilpi Gupta +<br/>Department of CSE +<br/>Department of CSE +<br/><b>Amity University</b><br/>Noida, India +<br/><b>Amity University</b><br/>Noida, India +</td><td></td><td></td></tr><tr><td>33ac7fd3a622da23308f21b0c4986ae8a86ecd2b</td><td>Building an On-Demand Avatar-Based Health Intervention for Behavior Change +<br/>School of Computing and Information Sciences +<br/><b>Florida International University</b><br/>Miami, FL, 33199, USA +<br/>Department of Computer Science +<br/><b>University of Miami</b><br/>Coral Gables, FL, 33146, USA +</td><td>('2671668', 'Ugan Yasavur', 'ugan yasavur')<br/>('2782570', 'Claudia de Leon', 'claudia de leon')<br/>('1809087', 'Reza Amini', 'reza amini')<br/>('1765935', 'Ubbo Visser', 'ubbo visser')</td><td></td></tr><tr><td>33030c23f6e25e30b140615bb190d5e1632c3d3b</td><td>Toward a General Framework for Words and +<br/>Pictures +<br/><b>Stony Brook University</b><br/><b>Stony Brook University</b><br/>Hal Daum´e III +<br/><b>University of Maryland</b><br/>Jesse Dodge +<br/><b>University of Washington</b><br/><b>University of Maryland</b><br/><b>Stony Brook University</b><br/>Alyssa Mensch +<br/>M.I.T. +<br/><b>University of Aberdeen</b><br/>Karl Stratos +<br/><b>Columbia University</b><br/><b>Stony Brook University</b></td><td>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('1685538', 'Tamara L. Berg', 'tamara l. berg')<br/>('2694557', 'Amit Goyal', 'amit goyal')<br/>('1682965', 'Xufeng Han', 'xufeng han')<br/>('38390487', 'Margaret Mitchell', 'margaret mitchell')<br/>('1721910', 'Kota Yamaguchi', 'kota yamaguchi')</td><td></td></tr><tr><td>33ba256d59aefe27735a30b51caf0554e5e3a1df</td><td>Early Active Learning via Robust +<br/>Representation and Structured Sparsity +<br/>†Department of Computer Science and Engineering +<br/><b>University of Texas at Arlington, Arlington, Texas 76019, USA</b><br/>‡Department of Electrical Engineering and Computer Science +<br/>Colorado School of Mines, Golden, Colorado 80401, USA +</td><td>('1688370', 'Feiping Nie', 'feiping nie')<br/>('1683402', 'Hua Wang', 'hua wang')<br/>('1748032', 'Heng Huang', 'heng huang')</td><td>feipingnie@gmail.com, huawangcs@gmail.com, heng@uta.edu, chqding@uta.edu +</td></tr><tr><td>33c3702b0eee6fc26fc49f79f9133f3dd7fa3f13</td><td><b>Imperial College London</b><br/>Department of Computing +<br/>Machine Learning Techniques +<br/>for Automated Analysis of Facial +<br/>Expressions +<br/>December, 2013 +<br/>Supervised by Prof. Maja Pantic +<br/>Submitted in part fulfilment of the requirements for the degree of PhD in Computing and +<br/><b>the Diploma of Imperial College London. This thesis is entirely my own work, and, except</b><br/>where otherwise indicated, describes my own research. +</td><td>('1729713', 'Ognjen Rudovic', 'ognjen rudovic')</td><td></td></tr><tr><td>33aff42530c2fd134553d397bf572c048db12c28</td><td>From Emotions to Action Units with Hidden and Semi-Hidden-Task Learning +<br/>Universitat Pompeu Fabra +<br/>Centre de Visio per Computador +<br/>Universitat Pompeu Fabra +<br/>Barcelona +<br/>Barcelona +<br/>Barcelona +</td><td>('40097226', 'Adria Ruiz', 'adria ruiz')<br/>('2820687', 'Joost van de Weijer', 'joost van de weijer')<br/>('1692494', 'Xavier Binefa', 'xavier binefa')</td><td>adria.ruiz@upf.es +<br/>joost@cvc.uab.es +<br/>xavier.binefa@upf.es +</td></tr><tr><td>33a1a049d15e22befc7ddefdd3ae719ced8394bf</td><td>FULL PAPER +<br/> International Journal of Recent Trends in Engineering, Vol 2, No. 1, November 2009 +<br/>An Efficient Approach to Facial Feature Detection +<br/>for Expression Recognition +<br/>S.P. Khandait1, P.D. Khandait2 and Dr.R.C.Thool2 +<br/>1Deptt. of Info.Tech., K.D.K.C.E., Nagpur, India +<br/> 2Deptt.of Electronics Engg., K.D.K.C.E., Nagpur, India, 2Deptt. of Info.Tech., SGGSIET, Nanded +</td><td></td><td>Prapti_khandait@yahoo.co.in +<br/>prabhakark_117@yahoo.co.in , rcthool@yahoo.com, +</td></tr><tr><td>334e65b31ad51b1c1f84ce12ef235096395f1ca7</td><td>Emotion in Human-Computer Interaction +<br/>Emotion in Human-Computer Interaction +<br/>Brave, S. & Nass, C. (2002). Emotion in human-computer interaction. In J. Jacko & A. +<br/>Sears (Eds.), Handbook of human-computer interaction (pp. 251-271). Hillsdale, NJ: +<br/>Lawrence Erlbaum Associates. +<br/>Scott Brave and Clifford Nass +<br/>Department of Communication +<br/><b>Stanford University</b><br/>Stanford, CA 94305-2050 +<br/>Phone: 650-428-1805,650-723-5499 +<br/>Fax: 650-725-2472 +</td><td></td><td>brave,nass@stanford.edu +</td></tr><tr><td>3328413ee9944de1cc7c9c1d1bf2fece79718ba1</td><td>Co-Training of Audio and Video Representations +<br/>from Self-Supervised Temporal Synchronization +<br/><b>Dartmouth College</b><br/>Facebook Research +<br/><b>Dartmouth College</b></td><td>('3443095', 'Bruno Korbar', 'bruno korbar')<br/>('1687325', 'Du Tran', 'du tran')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')</td><td>bruno.18@dartmouth.edu +<br/>trandu@fb.com +<br/>LT@dartmouth.edu +</td></tr><tr><td>3399f8f0dff8fcf001b711174d29c9d4fde89379</td><td>Face R-CNN +<br/>Tencent AI Lab, China +</td><td>('39049654', 'Hao Wang', 'hao wang')</td><td>{hawelwang,michaelzfli,denisji,yitongwang}@tencent.com +</td></tr><tr><td>333aa36e80f1a7fa29cf069d81d4d2e12679bc67</td><td>Suggesting Sounds for Images +<br/>from Video Collections +<br/>1Computer Science Department, ETH Z¨urich, Switzerland +<br/>2Disney Research, Switzerland +</td><td>('39231399', 'Oliver Wang', 'oliver wang')<br/>('1734448', 'Andreas Krause', 'andreas krause')<br/>('2893744', 'Alexander Sorkine-Hornung', 'alexander sorkine-hornung')</td><td>{msoler,krausea}@ethz.ch +<br/>{jean-charles.bazin,owang,alex}@disneyresearch.com +</td></tr><tr><td>3312eb79e025b885afe986be8189446ba356a507</td><td>This is a post-print of the original paper published in ECCV 2016 (SpringerLink). +<br/>MOON : A Mixed Objective Optimization +<br/>Network for the Recognition of Facial Attributes +<br/>Vision and Security Technology (VAST) Lab, +<br/><b>University of Colorado at Colorado Springs</b></td><td>('39886114', 'Ethan M. Rudd', 'ethan m. rudd')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')</td><td>{erudd,mgunther,tboult}@vast.uccs.edu +</td></tr><tr><td>33792bb27ef392973e951ca5a5a3be4a22a0d0c6</td><td>Two-dimensional Whitening Reconstruction for +<br/>Enhancing Robustness of Principal Component +<br/>Analysis +</td><td>('2766473', 'Xiaoshuang Shi', 'xiaoshuang shi')<br/>('1759643', 'Zhenhua Guo', 'zhenhua guo')<br/>('1688370', 'Feiping Nie', 'feiping nie')<br/>('1705066', 'Lin Yang', 'lin yang')<br/>('1748883', 'Jane You', 'jane you')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td></td></tr><tr><td>3328674d71a18ed649e828963a0edb54348ee598</td><td>IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 34, NO. 6, DECEMBER 2004 +<br/>2405 +<br/>A Face and Palmprint Recognition Approach Based +<br/>on Discriminant DCT Feature Extraction +</td><td>('15132338', 'Xiao-Yuan Jing', 'xiao-yuan jing')<br/>('1698371', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>339937141ffb547af8e746718fbf2365cc1570c8</td><td>Facial Emotion Recognition in Real Time +</td><td>('1849233', 'Dan Duncan', 'dan duncan')<br/>('3133285', 'Gautam Shine', 'gautam shine')<br/>('3158339', 'Chris English', 'chris english')</td><td>duncand@stanford.edu +<br/>gshine@stanford.edu +<br/>chriseng@stanford.edu +</td></tr><tr><td>33402ee078a61c7d019b1543bb11cc127c2462d2</td><td>Self-Supervised Video Representation Learning With Odd-One-Out Networks +<br/><b>ACRV, The Australian National University University of Oxford QUVA Lab, University of Amsterdam</b></td><td>('1688071', 'Basura Fernando', 'basura fernando')</td><td></td></tr><tr><td>33aa980544a9d627f305540059828597354b076c</td><td></td><td></td><td></td></tr><tr><td>33ae696546eed070717192d393f75a1583cd8e2c</td><td></td><td></td><td></td></tr><tr><td>33f2b44742cc828347ccc5ec488200c25838b664</td><td>Pooling the Convolutional Layers in Deep ConvNets for Action Recognition +<br/><b>School of Computer Science and Technology, Tianjin University, China</b><br/><b>School of Computer and Information, Hefei University of Technology, China</b></td><td>('2905510', 'Shichao Zhao', 'shichao zhao')<br/>('1732242', 'Yanbin Liu', 'yanbin liu')<br/>('2302512', 'Yahong Han', 'yahong han')<br/>('2248826', 'Richang Hong', 'richang hong')</td><td>{zhaoshichao, csyanbin, yahong}@tju.edu.cn, hongrc.hfut@gmail.com +</td></tr><tr><td>3393459600368be2c4c9878a3f65a57dcc0c2cfa</td><td>Eigen-PEP for Video Face Recognition +<br/><b>Stevens Institute of Technology Adobe Systems Inc</b></td><td>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('1745420', 'Gang Hua', 'gang hua')<br/>('1720987', 'Xiaohui Shen', 'xiaohui shen')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')</td><td></td></tr><tr><td>3352426a67eabe3516812cb66a77aeb8b4df4d1b</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 4, NO. 5, APRIL 2015 +<br/>Joint Multi-view Face Alignment in the Wild +</td><td>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('2814229', 'George Trigeorgis', 'george trigeorgis')<br/>('47943220', 'Yuxiang Zhou', 'yuxiang zhou')</td><td></td></tr><tr><td>334d6c71b6bce8dfbd376c4203004bd4464c2099</td><td>BICONVEX RELAXATION FOR SEMIDEFINITE PROGRAMMING IN +<br/>COMPUTER VISION +</td><td>('36861219', 'Sohil Shah', 'sohil shah')<br/>('1746575', 'Christoph Studer', 'christoph studer')<br/>('1962083', 'Tom Goldstein', 'tom goldstein')</td><td></td></tr><tr><td>33695e0779e67c7722449e9a3e2e55fde64cfd99</td><td>Riemannian Coding and Dictionary Learning: Kernels to the Rescue +<br/><b>Australian National University and NICTA</b><br/>While sparse coding on non-flat Riemannian manifolds has recently become +<br/>increasingly popular, existing solutions either are dedicated to specific man- +<br/>ifolds, or rely on optimization problems that are difficult to solve, especially +<br/>when it comes to dictionary learning. In this paper, we propose to make use +<br/>of kernels to perform coding and dictionary learning on Riemannian man- +<br/>ifolds. To this end, we introduce a general Riemannian coding framework +<br/>with its kernel-based counterpart. This lets us (i) generalize beyond the spe- +<br/>cial case of sparse coding; (ii) introduce efficient solutions to two coding +<br/>schemes; (iii) learn the kernel parameters; (iv) perform unsupervised and +<br/>supervised dictionary learning in a much simpler manner than previous Rie- +<br/>mannian coding approaches. +<br/>i=1, di ∈ M, be a dictionary on a Rie- +<br/>mannian manifold M, and x ∈ M be a query point on the manifold. We +<br/>(cid:17) +<br/>define a general Riemannian coding formulation as +<br/>More specifically, let D = {di}N +<br/>(cid:93)N +<br/>j=1 α jd j +<br/>min +<br/>s.t. α ∈ C, +<br/>+ λγ(α;x,D) +<br/>δ 2(cid:0)x, +<br/>(1) +<br/>on α. Moreover, (cid:85) : M×···×M× R× R···× R → M is an operator +<br/>where δ : M×M → R+ is a metric on M, α ∈ RN is the vector of Rie- +<br/>mannian codes, γ is a prior on the codes α and C is a set of constraints +<br/>that combines multiple dictionary atoms {d j ∈ M} with weights {α j} and +<br/>generates a point ˆx on M. This general formulation encapsulates intrinsic +<br/>sparse coding [2, 5], but also lets us derive and intrinsic version of Locality- +<br/>constrained Linear Coding [10]. Such intrinsic formulations, however, de- +<br/>pend on the logarithm map, which may be highly nonlinear, or not even have +<br/>an analytic solution. +<br/>To overcome these weaknesses and obtain a general formulation of Rie- +<br/>mannian coding, we propose to perform coding in RKHS. This has the +<br/>twofold advantage of yielding simple solutions to several popular coding +<br/>techniques and of resulting in a potentially better representation than stan- +<br/>dard coding techniques due to the nonlinearity of the approach. To this +<br/>end, let φ : M → H be a mapping to an RKHS induced by the kernel +<br/>k(x,y) = φ (x)T φ (y). Coding in H can then be formulated as +<br/>(cid:13)(cid:13)(cid:13)φ(cid:0)x)−∑N +<br/>(cid:13)(cid:13)(cid:13)2 +<br/>j=1 α jφ(cid:0)d j) +<br/>+ λγ(α;φ(cid:0)x),φ(cid:0)D)) +<br/>min +<br/>s.t. α ∈ C. +<br/>(2) +<br/>As shown in the paper, the reconstruction term in (2) can be kernelized. +<br/>More importantly, after kernelization, this term remains quadratic, convex +<br/>and similar to its counterpart in Euclidean space. This lets us derive efficient +<br/>solutions to two coding schemes: kernel Sparse Coding (kSC) and kernel +<br/>Locality Constrained Coding (kLCC). +<br/>In many cases, it is beneficial not only to compute the codes for a given +<br/>dictionary, but also to optimize the dictionary to best suit the problem at +<br/>hand. Given training data, and for fixed codes, we then show that, by relying +<br/>on the Representer theorem [8], the dictionary update has an analytic form. +<br/>Furthermore, we introduce an approach to supervised dictionary learning, +<br/>which, given labeled data, jointly learns the dictionary and a classifier acting +<br/>on the codes. The resulting supervised coding schemes are referred to as +<br/>kSSC and kSLCC. +<br/>We demonstrate the effectiveness of our approach on three different +<br/>types of non-flat manifolds, as well as illustrate its generality by also ap- +<br/>plying it to Euclidean space, which simply is a special type of Rieman- +<br/>nian manifold. In particular, we evaluated our different techniques on two +<br/>challenging classification datasets where the images are represented with +<br/>region covariance descriptors (RCovDs) [9], which lie on SPD manifolds. +</td><td>('2862871', 'Mathieu Salzmann', 'mathieu salzmann')</td><td></td></tr><tr><td>334ac2a459190b41923be57744aa6989f9a54a51</td><td>Apples to Oranges: Evaluating Image Annotations from Natural Language +<br/>Processing Systems +<br/>Brown Laboratory for Linguistic Information Processing (BLLIP) +<br/><b>Brown University, Providence, RI</b></td><td>('2139196', 'Rebecca Mason', 'rebecca mason')<br/>('1749837', 'Eugene Charniak', 'eugene charniak')</td><td>{rebecca,ec}@cs.brown.edu +</td></tr><tr><td>33e20449aa40488c6d4b430a48edf5c4b43afdab</td><td>TRANSACTIONS ON AFFECTIVE COMPUTING +<br/>The Faces of Engagement: Automatic +<br/>Recognition of Student Engagement from Facial +<br/>Expressions +</td><td>('1775637', 'Jacob Whitehill', 'jacob whitehill')<br/>('3089406', 'Zewelanji Serpell', 'zewelanji serpell')<br/>('3267606', 'Yi-Ching Lin', 'yi-ching lin')<br/>('39687351', 'Aysha Foster', 'aysha foster')<br/>('1741200', 'Javier R. Movellan', 'javier r. movellan')</td><td></td></tr><tr><td>333e7ad7f915d8ee3bb43a93ea167d6026aa3c22</td><td>This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. +<br/>The final version of record is available at http://dx.doi.org/10.1109/TIFS.2014.2309851 +<br/>DRAFT +<br/>3D Assisted Face Recognition: Dealing With +<br/>Expression Variations +<br/> +</td><td>('2128163', 'Nesli Erdogmus', 'nesli erdogmus')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td></td></tr><tr><td>334166a942acb15ccc4517cefde751a381512605</td><td> International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 +<br/> Volume: 04 Issue: 10 | Oct -2017 www.irjet.net p-ISSN: 2395-0072 +<br/>Facial Expression Analysis using Deep Learning +<br/><b>M.Tech Student, SSG Engineering College, Odisha, India</b><br/>---------------------------------------------------------------------***--------------------------------------------------------------------- +<br/>examination structures need to analyse the facial exercises +</td><td>('13518951', 'Raman Patel', 'raman patel')</td><td></td></tr><tr><td>33403e9b4bbd913ae9adafc6751b52debbd45b0e</td><td></td><td></td><td></td></tr><tr><td>33ef419dffef85443ec9fe89a93f928bafdc922e</td><td>SelfKin: Self Adjusted Deep Model For +<br/>Kinship Verification +<br/><b>Faculty of Engineering, Bar-Ilan University, Israel</b></td><td>('32450996', 'Eran Dahan', 'eran dahan')<br/>('1926432', 'Yosi Keller', 'yosi keller')</td><td></td></tr><tr><td>33ad23377eaead8955ed1c2b087a5e536fecf44e</td><td>Augmenting CRFs with Boltzmann Machine Shape Priors for Image Labeling +<br/>∗ indicates equal contribution +</td><td>('2177037', 'Andrew Kae', 'andrew kae')<br/>('1729571', 'Kihyuk Sohn', 'kihyuk sohn')<br/>('1697141', 'Honglak Lee', 'honglak lee')</td><td>1 University of Massachusetts, Amherst, MA, USA, {akae,elm}@cs.umass.edu +<br/>2 University of Michigan, Ann Arbor, MI, USA, {kihyuks,honglak}@umich.edu +</td></tr><tr><td>053b263b4a4ccc6f9097ad28ebf39c2957254dfb</td><td>Cost-Effective HITs for Relative Similarity Comparisons +<br/><b>Cornell University</b><br/><b>University of California, San Diego</b><br/><b>Cornell University</b></td><td>('3035230', 'Michael J. Wilber', 'michael j. wilber')<br/>('2064392', 'Iljung S. Kwak', 'iljung s. kwak')<br/>('1769406', 'Serge J. Belongie', 'serge j. belongie')</td><td></td></tr><tr><td>05b8673d810fadf888c62b7e6c7185355ffa4121</td><td>(will be inserted by the editor) +<br/>A Comprehensive Survey to Face Hallucination +<br/>Received: date / Accepted: date +</td><td>('2870173', 'Nannan Wang', 'nannan wang')</td><td></td></tr><tr><td>056d5d942084428e97c374bb188efc386791e36d</td><td>Temporally Robust Global Motion +<br/>Compensation by Keypoint-based Congealing +<br/><b>Michigan State University</b></td><td>('2447931', 'Yousef Atoum', 'yousef atoum')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td></td></tr><tr><td>05e658fed4a1ce877199a4ce1a8f8cf6f449a890</td><td></td><td></td><td></td></tr><tr><td>05ad478ca69b935c1bba755ac1a2a90be6679129</td><td>Attribute Dominance: What Pops Out? +<br/>Georgia Tech +</td><td>('3169410', 'Naman Turakhia', 'naman turakhia')</td><td>nturakhia@gatech.edu +</td></tr><tr><td>0595d18e8d8c9fb7689f636341d8a55cc15b3e6a</td><td>Discriminant Analysis on Riemannian Manifold of Gaussian Distributions +<br/>for Face Recognition with Image Sets +<br/>1Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b></td><td>('39792743', 'Wen Wang', 'wen wang')<br/>('39792743', 'Ruiping Wang', 'ruiping wang')<br/>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{wen.wang, zhiwu.huang}@vipl.ict.ac.cn, {wangruiping, sgshan, xlchen}@ict.ac.cn +</td></tr><tr><td>0573f3d2754df3a717368a6cbcd940e105d67f0b</td><td>Emotion Recognition In The Wild Challenge 2013∗ +<br/>Res. School of Computer +<br/>Science +<br/><b>Australian National University</b><br/>Roland Goecke +<br/>Vision & Sensing Group +<br/><b>University of Canberra</b><br/><b>Australian National University</b><br/>Vision & Sensing Group +<br/><b>University of Canberra</b><br/>HCC Lab +<br/><b>University of Canberra</b><br/><b>Australian National University</b></td><td>('1735697', 'Abhinav Dhall', 'abhinav dhall')<br/>('2942991', 'Jyoti Joshi', 'jyoti joshi')<br/>('1743035', 'Michael Wagner', 'michael wagner')</td><td>jyoti.joshi@canberra.edu.au +<br/>abhinav.dhall@anu.edu.au +<br/>roland.goecke@ieee.org +<br/>michael.wagner@canberra.edu.au +</td></tr><tr><td>05a0d04693b2a51a8131d195c68ad9f5818b2ce1</td><td>Dual-reference Face Retrieval +<br/><b>School of Computing Sciences, University of East Anglia, Norwich, UK</b><br/><b>University of Pittsburgh, Pittsburgh, USA</b><br/>3JD Artificial Intelligence Research (JDAIR), Beijing, China +</td><td>('19285980', 'BingZhang Hu', 'bingzhang hu')<br/>('40255667', 'Feng Zheng', 'feng zheng')<br/>('40799321', 'Ling Shao', 'ling shao')</td><td>bingzhang.hu@uea.ac.uk, feng.zheng@pitt.edu, ling.shao@ieee.org +</td></tr><tr><td>0562fc7eca23d47096472a1d42f5d4d086e21871</td><td></td><td></td><td></td></tr><tr><td>054738ce39920975b8dcc97e01b3b6cc0d0bdf32</td><td>Towards the Design of an End-to-End Automated +<br/>System for Image and Video-based Recognition +</td><td>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('26988560', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('2716670', 'Swami Sankaranarayanan', 'swami sankaranarayanan')<br/>('40080979', 'Amit Kumar', 'amit kumar')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('38171682', 'Carlos D. Castillo', 'carlos d. castillo')</td><td></td></tr><tr><td>05e03c48f32bd89c8a15ba82891f40f1cfdc7562</td><td>Scalable Robust Principal Component +<br/>Analysis using Grassmann Averages +</td><td>('2142792', 'Søren Hauberg', 'søren hauberg')<br/>('1808965', 'Aasa Feragen', 'aasa feragen')<br/>('2105795', 'Michael J. Black', 'michael j. black')</td><td></td></tr><tr><td>05a312478618418a2efb0a014b45acf3663562d7</td><td>Accelerated Sampling for the Indian Buffet Process +<br/><b>Cambridge University, Trumpington Street, Cambridge CB21PZ, UK</b></td><td>('2292194', 'Finale Doshi-Velez', 'finale doshi-velez')<br/>('1983575', 'Zoubin Ghahramani', 'zoubin ghahramani')</td><td>finale@alum.mit.edu +<br/>zoubin@eng.cam.ac.uk +</td></tr><tr><td>056ba488898a1a1b32daec7a45e0d550e0c51ae4</td><td>Cascaded Continuous Regression for Real-time +<br/>Incremental Face Tracking +<br/>Enrique S´anchez-Lozano, Brais Martinez, +<br/><b>Computer Vision Laboratory. University of Nottingham</b></td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>{psxes1,yorgos.tzimiropoulos,michel.valstar}@nottingham.ac.uk +</td></tr><tr><td>050fdbd2e1aa8b1a09ed42b2e5cc24d4fe8c7371</td><td>Contents +<br/>Scale Space and PDE Methods +<br/>Spatio-Temporal Scale Selection in Video Data . . . . . . . . . . . . . . . . . . . . . +<br/>Dynamic Texture Recognition Using Time-Causal Spatio-Temporal +<br/>Scale-Space Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Corner Detection Using the Affine Morphological Scale Space . . . . . . . . . . . +<br/>Luis Alvarez +<br/>Nonlinear Spectral Image Fusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Martin Benning, Michael Möller, Raz Z. Nossek, Martin Burger, +<br/>Daniel Cremers, Guy Gilboa, and Carola-Bibiane Schönlieb +<br/>16 +<br/>29 +<br/>41 +<br/>Tubular Structure Segmentation Based on Heat Diffusion. . . . . . . . . . . . . . . +<br/>54 +<br/>Fang Yang and Laurent D. Cohen +<br/>Analytic Existence and Uniqueness Results for PDE-Based Image +<br/>Reconstruction with the Laplacian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Laurent Hoeltgen, Isaac Harris, Michael Breuß, and Andreas Kleefeld +<br/>Combining Contrast Invariant L1 Data Fidelities with Nonlinear +<br/>Spectral Image Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Leonie Zeune, Stephan A. van Gils, Leon W.M.M. Terstappen, +<br/>and Christoph Brune +<br/>An Efficient and Stable Two-Pixel Scheme for 2D +<br/>Forward-and-Backward Diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Martin Welk and Joachim Weickert +<br/>66 +<br/>80 +<br/>94 +<br/>Restoration and Reconstruction +<br/>Blind Space-Variant Single-Image Restoration of Defocus Blur. . . . . . . . . . . +<br/>109 +<br/>Leah Bar, Nir Sochen, and Nahum Kiryati +<br/>Denoising by Inpainting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>121 +<br/>Robin Dirk Adam, Pascal Peter, and Joachim Weickert +<br/>Stochastic Image Reconstruction from Local Histograms +<br/>of Gradient Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Agnès Desolneux and Arthur Leclaire +<br/>133 +</td><td>('3205375', 'Tony Lindeberg', 'tony lindeberg')<br/>('3205375', 'Tony Lindeberg', 'tony lindeberg')</td><td></td></tr><tr><td>056294ff40584cdce81702b948f88cebd731a93e</td><td></td><td></td><td></td></tr><tr><td>052880031be0a760a5b606b2ad3d22f237e8af70</td><td>Datasets on object manipulation and interaction: a survey +</td><td>('3112203', 'Yongqiang Huang', 'yongqiang huang')<br/>('35760122', 'Yu Sun', 'yu sun')</td><td></td></tr><tr><td>055de0519da7fdf27add848e691087e0af166637</td><td>Joint Unsupervised Face Alignment +<br/>and Behaviour Analysis(cid:2) +<br/><b>Imperial College London, UK</b></td><td>('1786302', 'Lazaros Zafeiriou', 'lazaros zafeiriou')<br/>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{l.zafeiriou12,e.antonakos,s.zafeiriou,m.pantic}@imperial.ac.uk +</td></tr><tr><td>0515e43c92e4e52254a14660718a9e498bd61cf5</td><td>MACHINE LEARNING SYSTEMS FOR DETECTING DRIVER DROWSINESS +<br/><b>Sabanci University</b><br/>Faculty of +<br/>Engineering and Natural Sciences +<br/>Orhanli, Istanbul +<br/><b>University Of California San Diego</b><br/><b>Institute of</b><br/>Neural Computation +<br/>La Jolla, San Diego +</td><td>('40322754', 'Esra Vural', 'esra vural')<br/>('2724380', 'Gwen Littlewort', 'gwen littlewort')<br/>('1858421', 'Marian Bartlett', 'marian bartlett')<br/>('29794862', 'Javier Movellan', 'javier movellan')</td><td></td></tr><tr><td>053c2f592a7f153e5f3746aa5ab58b62f2cf1d21</td><td>International Journal of Research in +<br/>Engineering & Technology (IJRET) +<br/>ISSN 2321-8843 +<br/>Vol. 1, Issue 2, July 2013, 11-20 +<br/>© Impact Journals +<br/>PERFORMANCE EVALUATION OF ILLUMINATION NORMALIZATION TECHNIQUES +<br/>FOR FACE RECOGNITION +<br/><b>PSG College of Technology, Coimbatore, Tamil Nadu, India</b></td><td></td><td></td></tr><tr><td>05891725f5b27332836cf058f04f18d74053803f</td><td>One-shot Action Localization by Learning Sequence Matching Network +<br/><b>The Australian National University</b><br/><b>ShanghaiTech University</b><br/>Fatih Porikli +<br/><b>The Australian National University</b></td><td>('51050729', 'Hongtao Yang', 'hongtao yang')<br/>('33913193', 'Xuming He', 'xuming he')</td><td>u5226028@anu.edu.au +<br/>hexm@shanghaitech.edu.cn +<br/>fatih.porikli@anu.edu.au +</td></tr><tr><td>0568fc777081cbe6de95b653644fec7b766537b2</td><td>Learning Expressionlets on Spatio-Temporal Manifold for Dynamic Facial +<br/>Expression Recognition +<br/>1Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences (UCAS), Beijing, 100049, China</b><br/><b>University of Oulu, Finland</b></td><td>('1730228', 'Mengyi Liu', 'mengyi liu')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>mengyi.liu@vipl.ict.ac.cn, {sgshan, wangruiping, xlchen}@ict.ac.cn +</td></tr><tr><td>05d80c59c6fcc4652cfc38ed63d4c13e2211d944</td><td>On Sampling-based Approximate Spectral Decomposition +<br/>Google Research, New York, NY +<br/><b>Courant Institute of Mathematical Sciences and Google Research, New York, NY</b><br/><b>Courant Institute of Mathematical Sciences, New York, NY</b></td><td>('2794322', 'Sanjiv Kumar', 'sanjiv kumar')<br/>('1709415', 'Mehryar Mohri', 'mehryar mohri')<br/>('8395559', 'Ameet Talwalkar', 'ameet talwalkar')</td><td>sanjivk@google.com +<br/>mohri@cs.nyu.edu +<br/>ameet@cs.nyu.edu +</td></tr><tr><td>05ea7930ae26165e7e51ff11b91c7aa8d7722002</td><td>Learning And-Or Model to Represent Context and +<br/>Occlusion for Car Detection and Viewpoint Estimation +</td><td>('3198440', 'Tianfu Wu', 'tianfu wu')<br/>('40479452', 'Bo Li', 'bo li')<br/>('3133970', 'Song-Chun Zhu', 'song-chun zhu')</td><td></td></tr><tr><td>055530f7f771bb1d5f352e2758d1242408d34e4d</td><td>A Facial Expression Recognition System from +<br/>Depth Video +<br/>Department of Computer Education +<br/><b>Sungkyunkwan University</b><br/>Seoul, Republic of Korea +</td><td>('3241032', 'Md. Zia Uddin', 'md. zia uddin')</td><td>Email: ziauddin@skku.edu +</td></tr><tr><td>050eda213ce29da7212db4e85f948b812a215660</td><td>Combining Models and Exemplars for Face Recognition: +<br/>An Illuminating Example +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +</td><td>('1715286', 'Terence Sim', 'terence sim')<br/>('1733113', 'Takeo Kanade', 'takeo kanade')</td><td></td></tr><tr><td>051a84f0e39126c1ebeeb379a405816d5d06604d</td><td>Cogn Comput (2009) 1:257–267 +<br/>DOI 10.1007/s12559-009-9018-7 +<br/>Biometric Recognition Performing in a Bioinspired System +<br/>Joan Fa`bregas Æ Marcos Faundez-Zanuy +<br/>Published online: 20 May 2009 +<br/>Ó Springer Science+Business Media, LLC 2009 +</td><td></td><td></td></tr><tr><td>05e3acc8afabc86109d8da4594f3c059cf5d561f</td><td>Actor-Action Semantic Segmentation with Grouping Process Models +<br/>Department of Electrical Engineering and Computer Science +<br/><b>University of Michigan, Ann Arbor</b><br/>CVPR 2016 +<br/>OBJECTIVE +<br/>We seek to label each pixel in a video with a pair of actor (e.g. adult, baby and +<br/>dog) and action (e.g. eating, walking and jumping) labels. +<br/>Overview of the Grouping Process Model +<br/>Video Labeling +<br/>- We propose a novel grouping process model (GPM) that adaptively adds +<br/>long-ranging interactions of the supervoxel hierarchy to the labeling CRF. +<br/>Input Video +<br/>Segment-Level +<br/>s. +<br/>The Tree Slice Problem +<br/>slice +<br/>The Video Labeling Problem +<br/>Selected Nodes +<br/>Input Video +<br/>- We incorporate the video-level recognition into segment-level labeling by +<br/>the means of global labeling cost and the GPM. +<br/> - a set of random variables defined on the segments taking +<br/>Definition & Joint Modeling +<br/>Segment-Level: +<br/> - a video segmentation with n segments. +<br/>V = {q1, q2, . . . , qN} +<br/>L = {l1, l2, . . . , lN} +<br/>labels from both actor space and action space, e.g. adult-eating, dog-crawling. +<br/>Supervoxel Hierarchy: +<br/>T = {T1, T2, . . . , TS} +<br/>chy with S total supervoxels. +<br/>s = {s1, s2, . . . , sS} +<br/>voxels denoting its active or not. +<br/> - a segmentation tree extracted from a supervoxel hierar- +<br/>- a set of binary random variables defined on the super- +<br/>The Overall Objective Function: +<br/>(L∗, s∗) = arg min +<br/>E(L, s|V,T ) +<br/>E(L, s|V,T ) = Ev(L|V) +E h(s|T ) +<br/>L,s +<br/>+(cid:31)t∈T +<br/>(Eh(Lt|st) +E h(st|Lt)) +<br/>Grouping Cues from Segment Labeling. The GPM uses evidence directly from +<br/>the segment-level CRF to locate supervoxels across various scales that best cor- +<br/>respond to the actor and its action. +<br/>Eh(st|Lt) = (H(Lt)|Lt| + θh)st +<br/>The Tree Slice Constraint. We seek a single labeling over the video. Each node +<br/>in CRF is associated with one and only one supervoxel in the hierarchy. This con- +<br/>straint is the same as our previous work in Xu et al. ICCV 2013. +<br/>Eh(s|T ) = +<br/>P(cid:31)p=1 +<br/>δ(PT +<br/>p s (cid:31)= 1)θτ +<br/>Labeling Cues from Supervoxel Hierarchy. Once the supervoxels are selected, +<br/>they provide strong labeling cues to the segment-level CRF. The CRF nodes con- +<br/>nected to the same active supervoxel are encouraged to have the same label. +<br/>Eh(Lt|st) =(cid:31) (cid:30)i∈Lt(cid:30)j(cid:30)=i,j∈Lt +<br/>ij(li, lj) =(cid:31) θt +<br/>ψh +<br/>if li (cid:31)= lj +<br/>otherwise +<br/>ψh +<br/>ij(li, lj) +<br/>if st = 1 +<br/>otherwise +<br/>Segment-Level CRF +<br/>The segment-level CRF considers the interplay of actors and actions. +<br/>- denotes the set of actor labels (e.g. adult, baby and dog). +<br/>- denotes the set of action labels (e.g. eating, running and crawling). +<br/>Ev(L|V) =(cid:31)i∈V +<br/>ξv +<br/>i (li) +(cid:31)i∈V (cid:31)j∈E(i) +<br/>ξv +<br/>ij(li, lj) +<br/>ξv +<br/>i (li) = ψv +<br/>i (lXi ) +φ v +<br/>i (lYi ) +ϕ v +<br/>i (lXi , lYj ) +<br/>ij(lXi , lXj ) +<br/>ψv +<br/>ij(lYi , lYj ) +<br/>φv +<br/>ij(lXi , lXj ) +φ v +<br/>ψv +<br/>ij(lYi , lYj ) +<br/>(cid:31)= lXj ∧ lYi = lYj +<br/>if lXi +<br/>(cid:31)= lYj +<br/>if lXi = lXj ∧ lYi +<br/>(cid:31)= lXj ∧ lYi +<br/>(cid:31)= lYj +<br/>if lXi +<br/>if lXi = lXj ∧ lYi = lYj . +<br/>ξv +<br/>ij(li, lj) = +<br/>Iterative Inference +<br/>Directly solving the overall objective function is hard. We use an iterative inference +<br/>schema to efficiently solve it. +<br/>The Video Labeling Problem. Given a tree slice, we find the best labeling. +<br/>L∗ = arg min +<br/>= arg min +<br/>E(L|s,V,T ) +<br/>Ev(L|V) +(cid:31)t∈T +<br/>- Optimization depends on +<br/>- Solvable by graph-cuts multi-label inference. +<br/> +<br/>Eh(Lt|st) +<br/>The Tree Slice Problem. Given a labeling, we find the best tree slice. +<br/>- Rewrite as a binary linear program. +<br/>s∗ = arg min +<br/>E(s|L,V,T ) +<br/>= arg min +<br/>Eh(st|Lt) +<br/>Eh(s|T ) +(cid:31)t∈T +<br/>s.t. Ps = 1P and s ∈ {0, 1}S +<br/>min(cid:31)t∈T +<br/>αtst +<br/>Experiments: The Actor-Action Semantic Segmentation +<br/>- Dataset: the A2D large-scale video labeling dataset. +<br/>One-third of videos have more than one actor performing different actions. +<br/>- Two different hierarchies: TSP and GBH. +<br/>- Video-level recognition is added through both global labeling cost and the GPM. +<br/>It consists of 3782 YouTube videos with an average length of 136 frames. +<br/>100.0 +<br/>80.0 +<br/>60.0 +<br/>40.0 +<br/>20.0 +<br/>0.0 +<br/>!-./' +<br/>77.9 +<br/>74.6 +<br/>44.8 +<br/>45.7 +<br/>64.9 +<br/>38.0 +<br/>85.2 +<br/>84.9 +<br/>58.3 +<br/>59.4 +<br/>!"#$ +<br/>$%#$ &'()*+,' !"#$%& !"#$!& +<br/>100.0 +<br/>80.0 +<br/>60.0 +<br/>40.0 +<br/>20.0 +<br/>0.0 +<br/>!-.(/0 +<br/>77.6 +<br/>74.6 +<br/>45.5 +<br/>47.0 +<br/>85.3 +<br/>84.8 +<br/>60.5 +<br/>61.2 +<br/>63.9 +<br/>29.0 +<br/>!"#$ +<br/>$%#$ &'()*+,' !"#$%& !"#$!& +<br/>100.0 +<br/>80.0 +<br/>60.0 +<br/>40.0 +<br/>20.0 +<br/>0.0 +<br/>1!-./'23!-.(/04 +<br/>84.2 +<br/>76.2 +<br/>72.9 +<br/>63.0 +<br/>83.8 +<br/>43.3 +<br/>43.9 +<br/>25.4 +<br/>26.5 +<br/>13.9 +<br/>!"#$ +<br/>$%#$ &'()*+,' !"#$%& !"#$!& +<br/>5)6,73'()*+,)-")*./0+11-'223*+24839,))/:73 ;)/<*)3=(>,)3!--6'*-+ +<br/>Visual example of the actor-action video labelings for all methods. (a) - (c) are +<br/>videos where most methods get correct labelings; (d) - (e) are videos where GPM +<br/>models outperform; (h) - (i) are different videos with partially correct labelings. +<br/>(a) +<br/>(b) +<br/>(c) +<br/>(d) +<br/>(e) +<br/>(f) +<br/>(g) +<br/>(h) +<br/>(i) +<br/>Ground-Truth +<br/>AHRF +<br/>FCRF +<br/>adult-none +<br/>adult-eating +<br/>adult-eating +<br/>adult-eating +<br/>baby-crawling +<br/>Trilayer +<br/>GPM (TSP) +<br/>GPM (GBH) +<br/>adult-none +<br/>adult-eating +<br/>adult-eating +<br/>adult-eating +<br/>car-running +<br/>car-running +<br/>car-running +<br/>car-running +<br/>car-running +<br/>car-running +<br/>baby-rolling +<br/>baby-rolling +<br/>baby-rolling +<br/>baby-rolling +<br/>baby-rolling +<br/>baby-rolling +<br/>dog-eating +<br/>baby-crawling +<br/>dog-crawling +<br/>adult-none +<br/>car-rolling +<br/>car-rolling +<br/>dog-crawling +<br/>dog-crawling +<br/>bird-eating +<br/>cat-climbing +<br/>adult-walking +<br/>adult-walking +<br/>bird-eating +<br/>bird-eating +<br/>adult-walking +<br/>bird-walking +<br/>bird-flying +<br/>car-running +<br/>car-running +<br/>bird-walking +<br/>bird-walking +<br/>car-running +<br/>dog-walking +<br/>dog-walking +<br/>adult-walking +<br/>car-jumping +<br/>ball-flying +<br/>adult-walking +<br/>car-running +<br/>adult-walking +<br/>adult-walking +<br/>adult-walking +<br/>car-running +<br/>adult-walking +<br/>adult-running +<br/>ball-rolling +<br/>adult-none +<br/>adult-walking +<br/>adult-running +<br/>adult-walking +<br/>adult-walking +<br/>dog-walking +<br/>dog-rolling +<br/>ball-rolling +<br/>ball-rolling +<br/>adult-none +<br/>car-jumping +<br/>adult-none +<br/>bird-walking +<br/>adult-crawling +<br/>adult-jumping +<br/>adult-crawling +<br/>car-flying +<br/>adult-crawling +<br/>adult-crawling +<br/>αt = H(Lt)|Lt| + θh +<br/>Acknowledgements. This work has been supported in part by Google, Samsung, DARPA W32P4Q-15-C-0070 +<br/>and ARO W911NF-15-1-0354. +</td><td>('2026123', 'Chenliang Xu', 'chenliang xu')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')</td><td></td></tr><tr><td>05f4d907ee2102d4c63a3dc337db7244c570d067</td><td></td><td></td><td></td></tr><tr><td>0559fb9f5e8627fecc026c8ee6f7ad30e54ee929</td><td>4 +<br/>Facial Expression Recognition +<br/><b>ADSIP Research Centre, University of Central Lancashire</b><br/>UK +<br/>1. Introduction +<br/>Facial expressions are visible signs of a person’s affective state, cognitive activity and +<br/>personality. Humans can perform expression recognition with a remarkable robustness +<br/>without conscious effort even under a variety of adverse conditions such as partially +<br/>occluded faces, different appearances and poor illumination. Over the last two decades, the +<br/>advances in imaging technology and ever increasing computing power have opened up a +<br/>possibility of automatic facial expression recognition and this has led to significant research +<br/>efforts from the computer vision and pattern recognition communities. One reason for this +<br/>growing interest is due to a wide spectrum of possible applications in diverse areas, such as +<br/>more engaging human-computer interaction (HCI) systems, video conferencing, augmented +<br/>reality. Additionally from the biometric perspective, automatic recognition of facial +<br/>expressions has been investigated in the context of monitoring patients in the intensive care +<br/>and neonatal units for signs of pain and anxiety, behavioural research, identifying level of +<br/>concentration, and improving face recognition. +<br/>Automatic facial expression recognition is a difficult task due to its inherent subjective +<br/>nature, which is additionally hampered by usual difficulties encountered in pattern +<br/>recognition and computer vision research. The vast majority of the current state-of-the-art +<br/>facial expression recognition systems are based on 2-D facial images or videos, which offer +<br/>good performance only for the data captured under controlled conditions. As a result, there +<br/>is currently a shift towards the use of 3-D facial data to yield better recognition performance. +<br/>However, it requires more expensive data acquisition systems and sophisticated processing +<br/>algorithms. The aim of this chapter is to provide an overview of the existing methodologies +<br/>and recent advances in the facial expression recognition, as well as present a systematic +<br/>description of the authors’ work on the use of 3-D facial data for automatic recognition of +<br/>facial expressions, starting from data acquisition and database creation to data processing +<br/>algorithms and performance evaluation. +<br/>1.1 Facial expression +<br/>Facial expressions are generated ... skin texture” (Pantic & Rothkrantz, 2000)” should be +<br/>replaced by “Expressions shown on the face are produced by a combination of contraction +<br/>activities made by facial muscles, with most noticeable temporal deformation around nose, +<br/>lips, eyelids, and eyebrows as well as facial skin texture patterns (Pantic & Rothkrantz, +<br/>2000). Typical facial expressions last for a few seconds, normally between 250 milliseconds +<br/>and five seconds (Fasel & Luettin, 2003). According to psychologists Ekman and Friesen +</td><td>('2647218', 'Bogdan J. Matuszewski', 'bogdan j. matuszewski')<br/>('2343120', 'Wei Quan', 'wei quan')</td><td></td></tr><tr><td>052f994898c79529955917f3dfc5181586282cf8</td><td>Unsupervised Domain Adaptation for Face Recognition in Unlabeled Videos +<br/>1NEC Labs America +<br/>2UC Merced +<br/><b>Dalian University of Technology</b><br/>4UC San Diego +</td><td>('1729571', 'Kihyuk Sohn', 'kihyuk sohn')</td><td></td></tr><tr><td>05a7be10fa9af8fb33ae2b5b72d108415519a698</td><td>Multilayer and Multimodal Fusion of Deep Neural Networks +<br/>for Video Classification +<br/>NVIDIA +</td><td>('2214162', 'Xiaodong Yang', 'xiaodong yang')</td><td>{xiaodongy, pmolchanov, jkautz}@nvidia.com +</td></tr><tr><td>050a149051a5d268fcc5539e8b654c2240070c82</td><td>MAGISTERSKÉ A DOKTORSKÉSTUDIJNÍ PROGRAMY31. 5. 2018SBORNÍKSTUDENTSKÁ VĚDECKÁ KONFERENCE </td><td></td><td></td></tr><tr><td>05318a267226f6d855d83e9338eaa9e718b2a8dd</td><td>_______________________________________________________PROCEEDING OF THE 16TH CONFERENCE OF FRUCT ASSOCIATION +<br/>Age Estimation from Face Images: Challenging +<br/>Problem for Audience Measurement Systems +<br/><b>Yaroslavl State University</b><br/>Russia +</td><td>('1857299', 'Alexander Ganin', 'alexander ganin')<br/>('39942308', 'Olga Stepanova', 'olga stepanova')<br/>('39635716', 'Anton Lebedev', 'anton lebedev')</td><td>vhr@yandex.ru, angnn@mail.ru, dcslab@uniyar.ac.ru, lebedevdes@gmail.com +</td></tr><tr><td>057d5f66a873ec80f8ae2603f937b671030035e6</td><td>Newtonian Image Understanding: +<br/>Unfolding the Dynamics of Objects in Static Images +<br/><b>Allen Institute for Arti cial Intelligence (AI</b><br/><b>University of Washington</b></td><td>('3012475', 'Roozbeh Mottaghi', 'roozbeh mottaghi')<br/>('2456400', 'Hessam Bagherinezhad', 'hessam bagherinezhad')<br/>('2563325', 'Mohammad Rastegari', 'mohammad rastegari')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')</td><td></td></tr><tr><td>0580edbd7865414c62a36da9504d1169dea78d6f</td><td>Baseline CNN structure analysis for facial expression recognition +</td><td>('2448391', 'Minchul Shin', 'minchul shin')<br/>('1702520', 'Munsang Kim', 'munsang kim')<br/>('1750864', 'Dong-Soo Kwon', 'dong-soo kwon')</td><td></td></tr><tr><td>050a3346e44ca720a54afbf57d56b1ee45ffbe49</td><td>Multi-Cue Zero-Shot Learning with Strong Supervision +<br/><b>Max-Planck Institute for Informatics</b></td><td>('2893664', 'Zeynep Akata', 'zeynep akata')<br/>('34070834', 'Mateusz Malinowski', 'mateusz malinowski')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td></td></tr><tr><td>0517d08da7550241fb2afb283fc05d37fce5d7b7</td><td>Sensors & Transducers, Vol. 153, Issue 6, June 2013, pp. 92-99 +<br/> +<br/>SSSeeennnsssooorrrsss &&& TTTrrraaannnsssddduuuccceeerrrsss +<br/>© 2013 by IFSA +<br/>http://www.sensorsportal.com +<br/>Combination of Local Multiple Patterns and Exponential +<br/>Discriminant Analysis for Facial Recognition +<br/><b>College of Computer Science, Chongqing University, Chongqing, 400030, China</b><br/><b>College of software, Chongqing University of Posts and Telecommunications Chongqing</b><br/><b>Institute of Computer Science and Technology, Chongqing University of Posts and</b><br/>400065, China +<br/>Telecommunications, Chongqing 400065, China +<br/>1 Tel.: 023-65112784, fax: 023-65112784 +<br/>Received: 26 April 2013 /Accepted: 14 June 2013 /Published: 25 June 2013 +</td><td>('2623870', 'Lifang Zhou', 'lifang zhou')<br/>('1713814', 'Bin Fang', 'bin fang')<br/>('1964987', 'Weisheng Li', 'weisheng li')<br/>('2103166', 'Lidou Wang', 'lidou wang')</td><td>1 E-mail: zhoulf@cqupt.edu.cn +</td></tr><tr><td>053931267af79a89791479b18d1b9cde3edcb415</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Attributes for Improved Attributes: A Multi-Task Network +<br/>Utilizing Implicit and Explicit Relationships for Facial Attribute Classification +<br/><b>University of Maryland, College Park</b><br/><b>College Park, MD</b></td><td>('3351637', 'Emily M. Hand', 'emily m. hand')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{emhand, rama}@umiacs.umd.edu +</td></tr><tr><td>05f3d1e9fb254b275354ca69018e9ed321dd8755</td><td>Face Recognition using Optimal Representation +<br/>Ensemble +<br/><b>NICTA , Queensland Research Laboratory, QLD, Australia</b><br/><b>Grif th University, QLD, Australia</b><br/><b>University of Adelaide, SA, Australia</b><br/>29·4·2013 +</td><td>('1711119', 'Hanxi Li', 'hanxi li')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')<br/>('1744926', 'Yongsheng Gao', 'yongsheng gao')</td><td></td></tr><tr><td>05e96d76ed4a044d8e54ef44dac004f796572f1a</td><td></td><td></td><td></td></tr><tr><td>051f03bc25ec633592aa2ff5db1d416b705eac6c</td><td>To appear in the International Joint Conference on Biometrics (IJCB 2011), Washington D.C., October 2011 +<br/>Partial Face Recognition: An Alignment Free Approach +<br/>Department of Computer Science and Engineering +<br/><b>Michigan State University, East Lansing, MI 48824, U.S.A</b></td><td>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>{scliao,jain}@cse.msu.edu +</td></tr><tr><td>9d58e8ab656772d2c8a99a9fb876d5611fe2fe20</td><td>Beyond Temporal Pooling: Recurrence and Temporal +<br/>Convolutions for Gesture Recognition in Video +<br/>{lionel.pigou,aaron.vandenoord,sander.dieleman, +<br/><b>Ghent University</b><br/>February 11, 2016 +</td><td>('2660640', 'Lionel Pigou', 'lionel pigou')<br/>('48373216', 'Sander Dieleman', 'sander dieleman')<br/>('10182287', 'Mieke Van Herreweghe', 'mieke van herreweghe')</td><td>mieke.vanherreweghe, joni.dambre}@ugent.be +</td></tr><tr><td>9d8ff782f68547cf72b7f3f3beda9dc3e8ecfce6</td><td>International Journal of Pattern Recognition +<br/>and Arti¯cial Intelligence +<br/>Vol. 26, No. 1 (2012) 1250002 (9 pages) +<br/>#.c World Scienti¯c Publishing Company +<br/>DOI: 10.1142/S0218001412500024 +<br/>IMPROVED PSEUDOINVERSE LINEAR +<br/>DISCRIMINANT ANALYSIS METHOD FOR +<br/>DIMENSIONALITY REDUCTION +<br/>*Signal Processing Laboratory, School of Engineering +<br/><b>Gri th University, QLD-4111, Brisbane, Australia</b><br/><b>University of the South Paci c, Fiji</b><br/>‡Laboratory of DNA Information Analysis +<br/><b>Human Genome Center, Institute of Medical Science</b><br/><b>University of Tokyo, 4-6-1 Shirokanedai</b><br/>Minato-ku, Tokyo 108-8639, Japan +<br/>Received 4 November 2010 +<br/>Accepted 22 September 2011 +<br/>Published 11 May 2012 +<br/>Pseudoinverse linear discriminant analysis (PLDA) is a classical method for solving small +<br/>sample size problem. However, its performance is limited. In this paper, we propose an improved +<br/>PLDA method which is faster and produces better classi¯cation accuracy when experimented on +<br/>several datasets. +<br/>Keywords : Pseudoinverse; +<br/>tational complexity. +<br/>linear discriminant analysis; dimensionality reduction; compu- +<br/>1. Introduction +<br/>Dimensionality reduction is an important aspect of pattern classi¯cation. It helps in +<br/>improving the robustness (or generalization capability) of the pattern classi¯er and +<br/>in reducing its computational complexity. The linear discriminant analysis (LDA) +<br/>method5 is a well-known dimensionality reduction technique studied in the litera- +<br/>ture. The LDA technique ¯nds an orientation matrix W that transforms high- +<br/>dimensional feature vectors belonging to di®erent classes to lower dimensional +<br/>feature vectors such that the projected feature vectors of a class are well separated +<br/>from the feature vectors of other classes. The orientation W is obtained by max- +<br/>imizing the Fisher's criterion function J1ðWÞ ¼ jW TSBWj=jW TSW Wj, where SB is +<br/>between-class scatter matrix and SW is within-class scatter matrix. It has been shown +<br/>in the literature that modi¯ed version of Fisher's criterion J2ðWÞ ¼ jW TSBWj= +<br/>jW TST Wj produces similar results, where ST is total scatter matrix.6 +<br/>1250002-1 +<br/><b>Int. J. Patt. Recogn. Artif. Intell. 2012.26. Downloaded from www.worldscientific.comby GRIFFITH UNIVERSITY INFORMATION SERVICES on 09/05/12. For personal use only</b></td><td>('3150542', 'Kuldip K. Paliwal', 'kuldip k. paliwal')<br/>('40532633', 'Alok Sharma', 'alok sharma')</td><td>§aloks@ims.u-tokyo.ac.jp +<br/>¶sharma_al@usp.ac.fj +</td></tr><tr><td>9d42df42132c3d76e3447ea61e900d3a6271f5fe</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Advanced Computing and Communication Techniques for High Performance Applications (ICACCTHPA-2014) +<br/>AutoCAP: An Automatic Caption Generation System +<br/>based on the Text Knowledge Power Series +<br/>Representation Model +<br/>M.Tech Dept of CSE +<br/><b>NSS College of Engineering</b><br/>Palakkad, Kerala +</td><td>('24326432', 'Krishnapriya P S', 'krishnapriya p s')</td><td></td></tr><tr><td>9d55ec73cab779403cd933e6eb557fb04892b634</td><td>Kernel principal component analysis network for image classification1 +<br/><b>Key Laboratory of Computer Network and Information Integration of Ministry of Education, Southeast University, Nanjing</b><br/>210096, China) +<br/>(2 Institut National de la Santé et de la Recherche Médicale U 1099, Rennes 35000, France) +<br/>(3 Laboratoire Traitement du Signal et de l’Image, Université de Rennes 1, Rennes 35000, France) +<br/>(4Centre de Recherche en Information Biomédicale Sino-français, Nanjing 210096, China) +</td><td>('1684465', 'Lotfi Senhadji', 'lotfi senhadji')</td><td></td></tr><tr><td>9d8fd639a7aeab0dd1bc6eef9d11540199fd6fe2</td><td>Workshop track - ICLR 2018 +<br/>LEARNING TO CLUSTER +<br/><b>ZHAW Datalab, Zurich University of Applied Sciences</b><br/>Winterthur, Switzerland +</td><td>('40087403', 'Benjamin B. Meier', 'benjamin b. meier')<br/>('2793787', 'Thilo Stadelmann', 'thilo stadelmann')</td><td>benjamin.meier70@gmail.com, stdm@zhaw.ch, oliver.duerr@gmail.com +</td></tr><tr><td>9d357bbf014289fb5f64183c32aa64dc0bd9f454</td><td>Face Identification by Fitting a 3D Morphable Model +<br/>using Linear Shape and Texture Error Functions +<br/><b>University of Freiburg, Instit ut f ur Informatik</b><br/>Georges-K¨ohler-Allee 52, 79110 Freiburg, Germany, +</td><td>('3293655', 'Sami Romdhani', 'sami romdhani')<br/>('2880906', 'Volker Blanz', 'volker blanz')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td>fromdhani, volker, vetterg@informatik.uni-freiburg.de +</td></tr><tr><td>9d66de2a59ec20ca00a618481498a5320ad38481</td><td>POP: Privacy-preserving Outsourced Photo Sharing +<br/>and Searching for Mobile Devices +<br/><b>cid:3) School of Software, Tsinghua University</b><br/><b>Illinois Institute of Technology</b></td><td>('1718343', 'Lan Zhang', 'lan zhang')<br/>('8645024', 'Taeho Jung', 'taeho jung')<br/>('1773806', 'Cihang Liu', 'cihang liu')<br/>('1752660', 'Xuan Ding', 'xuan ding')<br/>('34569491', 'Xiang-Yang Li', 'xiang-yang li')<br/>('10258874', 'Yunhao Liu', 'yunhao liu')</td><td></td></tr><tr><td>9d839dfc9b6a274e7c193039dfa7166d3c07040b</td><td>Augmented Faces +<br/>1ETH Z¨urich +<br/>2Kooaba AG +<br/>3K.U. Leuven +</td><td>('1727791', 'Matthias Dantone', 'matthias dantone')<br/>('1696393', 'Lukas Bossard', 'lukas bossard')<br/>('1726249', 'Till Quack', 'till quack')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{dantone,bossard,tquack,vangool}@vision.ee.ethz.ch +</td></tr><tr><td>9dcc6dde8d9f132577290d92a1e76b5decc6d755</td><td>Journal of Trends in the Development of Machinery +<br/> and Associated Technology +<br/>Vol. 16, No. 1, 2012, ISSN 2303-4009 (online), p.p. 175-178 +<br/>FACIAL EXPRESSION ANALYSIS BASED +<br/>ON OPTIMIZED GABOR FEATURES +<br/><b>Istanbul University</b><br/>Avcilar, 34320 Istanbul +<br/>Turkey +<br/>Yalçın Çekiç +<br/><b>Bahcesehir University</b><br/>Besiktas, 34349 Istanbul +<br/>Turkey +</td><td>('40701205', 'Aydın Akan', 'aydın akan')</td><td></td></tr><tr><td>9d36c81b27e67c515df661913a54a797cd1260bb</td><td>Applications (IJERA) ISSN: 2248-9622 www.ijera.com +<br/>Vol. 2, Issue 1,Jan-Feb 2012, pp.787-793 +<br/> 3D FACE RECOGNITION TECHNIQUES - A REVIEW +<br/><b>Gujarat Technological University, India</b><br/><b>Gujarat Technological University, India</b><br/>security at many places +</td><td>('9318822', 'Mahesh M. Goyani', 'mahesh m. goyani')<br/>('9198701', 'Preeti B. Sharma', 'preeti b. sharma')<br/>('9318822', 'Mahesh M. Goyani', 'mahesh m. goyani')</td><td></td></tr><tr><td>9d757c0fede931b1c6ac344f67767533043cba14</td><td>Search Based Face Annotation Using PCA and +<br/>Unsupervised Label Refinement Algorithms +<br/><b>Savitribai Phule Pune University</b><br/><b>D.Y.Patil Institute of Engineering and Technology, Pimpri, Pune</b><br/>Mahatma Phulenagar, 120/2 Mahaganpati soc, Chinchwad, Pune-19, MH, India +<br/><b>D.Y.Patil Institute of Engineering and Technology, Pimpri, Pune</b><br/>Computer Department, D.Y.PIET, Pimpri, Pune-18, MH, India +<br/>presents +</td><td>('15731441', 'Shital Shinde', 'shital shinde')<br/>('3392505', 'Archana Chaugule', 'archana chaugule')</td><td></td></tr><tr><td>9d57c4036a0e5f1349cd11bc342ac515307b6720</td><td>Landmark Weighting for 3DMM Shape Fitting +<br/><b>aSchool of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China</b><br/><b>bCVSSP, University of Surrey, Guildford, GU2 7XH, UK</b><br/>A B S T R A C T +</td><td>('51232704', 'Yu Yanga', 'yu yanga')<br/>('37020604', 'Xiao-Jun Wu', 'xiao-jun wu')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td></td></tr><tr><td>9d941a99e6578b41e4e32d57ece580c10d578b22</td><td>Sensors 2015, 15, 4326-4352; doi:10.3390/s150204326 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>Illumination-Invariant and Deformation-Tolerant Inner Knuckle +<br/>Print Recognition Using Portable Devices +<br/><b>School of Computer Science and Engineering, South China University of Technology</b><br/>Higher Education Mega Center, Panyu, Guangzhou 510006, China; +<br/>2 National-Regional Key Technology Engineering Laboratory for Medical Ultrasound, +<br/><b>School of Medicine, Shenzhen University, Shenzhen 518060, China</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Research Institute, The Chinese University of Hong Kong, Shenzhen 518057, China</b><br/>Academic Editor: Vittorio M.N. Passaro +<br/>Received: 6 January 2015 / Accepted: 6 February 2015 / Published: 12 February 2015 +</td><td>('2884662', 'Xuemiao Xu', 'xuemiao xu')<br/>('35636977', 'Qiang Jin', 'qiang jin')<br/>('3041338', 'Le Zhou', 'le zhou')<br/>('38166238', 'Jing Qin', 'jing qin')<br/>('1720633', 'Tien-Tsin Wong', 'tien-tsin wong')<br/>('2513505', 'Guoqiang Han', 'guoqiang han')</td><td>E-Mails: jin.q@mail.scut.edu.cn (Q.J.); z.le02@mail.scut.edu.cn (L.Z.); csgqhan@scut.edu.cn (G.H.) +<br/>Hong Kong 999077, China; E-Mail: ttwong@cse.cuhk.edu.hk +<br/>* Authors to whom correspondence should be addressed; E-Mails: xuemx@scut.edu.cn (X.X.); +<br/>jqin@szu.edu.cn (J.Q.); Tel.:+86-20-39380285 (X.X.); +86-755-86392117 (J.Q.). +</td></tr><tr><td>9d60ad72bde7b62be3be0c30c09b7d03f9710c5f</td><td>A Survey: Face Recognition Techniques +<br/>Assistant Professor, ITM GOI +<br/>M Tech, ITM GOI +<br/>face +<br/>video +<br/>(Eigen +<br/>passport-verification, +</td><td>('4122158', 'Arun Agrawal', 'arun agrawal')<br/>('3731551', 'Ranjana Sikarwar', 'ranjana sikarwar')</td><td></td></tr><tr><td>9d896605fbf93315b68d4ee03be0770077f84e40</td><td>Baby Talk: Understanding and Generating Image Descriptions +<br/><b>Stony Brook University</b><br/><b>Stony Brook University, NY 11794, USA</b></td><td>('2170826', 'Girish Kulkarni', 'girish kulkarni')<br/>('1699545', 'Yejin Choi', 'yejin choi')<br/>('40305780', 'Siming Li', 'siming li')<br/>('1685538', 'Tamara L Berg', 'tamara l berg')<br/>('3128210', 'Visruth Premraj', 'visruth premraj')<br/>('2985883', 'Sagnik Dhar', 'sagnik dhar')<br/>('39668247', 'Alexander C Berg', 'alexander c berg')</td><td>{tlberg}@cs.stonybrook.edu +</td></tr><tr><td>9d61b0beb3c5903fc3032655dc0fd834ec0b2af3</td><td>Learning a Locality Preserving Subspace for Visual Recognition +<br/>Microsoft Research Asia, Beijing 100080, China +<br/><b>School of Mathematical Science, Peking University, China</b></td><td>('3945955', 'Xiaofei He', 'xiaofei he')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('1689532', 'Yuxiao Hu', 'yuxiao hu')</td><td>*Department of Computer Science, University of Chicago (xiaofei@cs.uchicago.edu) +</td></tr><tr><td>9d24179aa33a94c8c61f314203bf9e906d6b64de</td><td>Searching for People through +<br/>Textual and Visual Attributes +<br/><b>Institute of Computing</b><br/><b>University of Campinas (Unicamp</b><br/>Campinas-SP, Brazil +<br/>Fig. 1. The proposed approach aims at searching for people using textual and visual attributes. Given an image database of faces, we extract the points of +<br/>interest (PoIs) to construct a visual dictionary that allow us to obtain the feature vectors by a quantization process (top). Then we train attribute classifiers to +<br/>generate a score for each image (middle). Finally, given a textual query (e.g., male), we fusion obtained scores to return a unique final rank (bottom). +</td><td>('37811966', 'Junior Fabian', 'junior fabian')<br/>('1820089', 'Ramon Pires', 'ramon pires')<br/>('2145405', 'Anderson Rocha', 'anderson rocha')</td><td></td></tr><tr><td>9d3aa3b7d392fad596b067b13b9e42443bbc377c</td><td>Facial Biometric Templates and Aging: +<br/>Problems and Challenges for Artificial +<br/>Intelligence +<br/><b>Cyprus University of Technology</b><br/>P.O Box 50329, Lemesos, 3066, Cyprus +</td><td>('1830709', 'Andreas Lanitis', 'andreas lanitis')</td><td>andreas.lanitis@cut.ac.cy +</td></tr><tr><td>9db4b25df549555f9ffd05962b5adf2fd9c86543</td><td>Nonlinear 3D Face Morphable Model +<br/>Department of Computer Science and Engineering +<br/><b>Michigan State University, East Lansing MI</b></td><td>('1849929', 'Luan Tran', 'luan tran')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td>{tranluan, liuxm}@msu.edu +</td></tr><tr><td>9d06d43e883930ddb3aa6fe57c6a865425f28d44</td><td>Clustering Appearances of Objects Under Varying Illumination Conditions +<br/>Computer Science & Engineering +<br/><b>University of California at San Diego</b><br/><b>cid:1) Honda Research Institute</b><br/>David Kriegman +<br/>Computer Science +<br/>800 California Street +<br/><b>University of Illinois at Urbana-Champaign</b><br/>La Jolla, CA 92093 +<br/>Mountain View, CA 94041 +<br/>Urbana, IL 61801 +</td><td>('1788818', 'Jeffrey Ho', 'jeffrey ho')<br/>('33047058', 'Jongwoo Lim', 'jongwoo lim')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')<br/>('2457452', 'Kuang-chih Lee', 'kuang-chih lee')</td><td>jho@cs.ucsd.edu myang@honda-ri.com jlim1@uiuc.edu +<br/>klee10@uiuc.edu +<br/>kriegman@cs.ucsd.edu +</td></tr><tr><td>9c1305383ce2c108421e9f5e75f092eaa4a5aa3c</td><td>SPEAKER RETRIEVAL FOR TV SHOW VIDEOS BY ASSOCIATING AUDIO SPEAKER +<br/>RECOGNITION RESULT TO VISUAL FACES∗ +<br/><b>School of Electrical and Information Engineering, Xi an Jiaotong University, Xi an, China</b><br/>’CNRS-LTCI, TELECOM-ParisTech, Paris, France +</td><td>('1859487', 'Yina Han', 'yina han')<br/>('2485487', 'Joseph Razik', 'joseph razik')<br/>('1693574', 'Gerard Chollet', 'gerard chollet')<br/>('1774346', 'Guizhong Liu', 'guizhong liu')</td><td></td></tr><tr><td>9cfb3a68fb10a59ec2a6de1b24799bf9154a8fd1</td><td></td><td></td><td></td></tr><tr><td>9c1860de6d6e991a45325c997bf9651c8a9d716f</td><td>3D Reconstruction and Face Recognition Using Kernel-Based +<br/> ICA and Neural Networks +<br/> Chi-Yung Lee +<br/>Dept. of Electrical Dept. of CSIE Dept. of CSIE +<br/><b>Engineering Chaoyang University Nankai Institute of</b><br/><b>National University of Technology Technology</b></td><td>('1734467', 'Cheng-Jian Lin', 'cheng-jian lin')</td><td> of Kaohsiung s9527618@cyut.edu.tw cylee@nkc.edu.tw +<br/>cjlin@nuk.edu.tw +</td></tr><tr><td>9c9ef6a46fb6395702fad622f03ceeffbada06e5</td><td>EUROGRAPHICS 2004 / M.-P. Cani and M. Slater +<br/>(Guest Editors) +<br/>Volume 23 (2004), Number 3 +<br/>Exchanging Faces in Images +<br/>1 Max-Planck-Institut für Informatik, Saarbrücken, Germany +<br/><b>University of Basel, Departement Informatik, Basel, Switzerland</b></td><td>('2880906', 'Volker Blanz', 'volker blanz')<br/>('2658043', 'Kristina Scherbaum', 'kristina scherbaum')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')<br/>('1746884', 'Hans-Peter Seidel', 'hans-peter seidel')</td><td></td></tr><tr><td>9c1cdb795fd771003da4378f9a0585730d1c3784</td><td>Stacked Deformable Part Model with Shape +<br/>Regression for Object Part Localization +<br/>Center for Biometrics and Security Research & National Laboratory +<br/><b>of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences, China</b></td><td>('1721677', 'Junjie Yan', 'junjie yan')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1708973', 'Yang Yang', 'yang yang')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jjyan,zlei,yang.yang,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>9ca7899338129f4ba6744f801e722d53a44e4622</td><td>Deep Neural Networks Regularization for Structured +<br/>Output Prediction +<br/>Soufiane Belharbi∗ +<br/>INSA Rouen, LITIS +<br/>76000 Rouen, France +<br/>INSA Rouen, LITIS +<br/>76000 Rouen, France +<br/>INSA Rouen, LITIS +<br/>76000 Rouen, France +<br/>INSA Rouen, LITIS +<br/>76000 Rouen, France +<br/>Normandie Univ, UNIROUEN, UNIHAVRE, +<br/>Normandie Univ, UNIROUEN, UNIHAVRE, +<br/>Normandie Univ, UNIROUEN, UNIHAVRE, +<br/>Normandie Univ, UNIROUEN, UNIHAVRE, +</td><td>('1712446', 'Clément Chatelain', 'clément chatelain')<br/>('1782268', 'Romain Hérault', 'romain hérault')<br/>('37078795', 'Sébastien Adam', 'sébastien adam')</td><td>soufiane.belharbi@insa-rouen.fr +<br/>romain.herault@insa-rouen.fr +<br/>clement.chatelain@insa-rouen.fr +<br/>sebastien.adam@univ-rouen.fr +</td></tr><tr><td>9c1664f69d0d832e05759e8f2f001774fad354d6</td><td>Action representations in robotics: A +<br/>taxonomy and systematic classification +<br/>Journal Title +<br/>XX(X):1–32 +<br/>c(cid:13)The Author(s) 2016 +<br/>Reprints and permission: +<br/>sagepub.co.uk/journalsPermissions.nav +<br/>DOI: 10.1177/ToBeAssigned +<br/>www.sagepub.com/ +</td><td>('33237072', 'Philipp Zech', 'philipp zech')<br/>('2898615', 'Erwan Renaudo', 'erwan renaudo')<br/>('36081156', 'Simon Haller', 'simon haller')<br/>('46447747', 'Xiang Zhang', 'xiang zhang')</td><td></td></tr><tr><td>9c25e89c80b10919865b9c8c80aed98d223ca0c6</td><td>GENDER PREDICTION BY GAIT ANALYSIS BASED ON TIME SERIES VARIATION OF +<br/>JOINT POSITIONS +<br/>Dept. of Computer Science +<br/>School of Science and Technology +<br/><b>Meiji University</b><br/>Dept. of Fundamental Science and Technology +<br/>Graduate School of Science and Technology +<br/><b>Meiji University</b><br/>1-1-1 Higashimita Tama-ku +<br/>Kawasaki Kanagawa Japan +<br/>1-1-1 Higashimita Tama-ku +<br/>Kawasaki Kanagawa Japan +</td><td>('1800246', 'Ryusuke Miyamoto', 'ryusuke miyamoto')<br/>('8187964', 'Risako Aoki', 'risako aoki')</td><td>E-mail: miya@cs.meiji.ac.jp +<br/>E-mail: aori@cs.meiji.ac.jp +</td></tr><tr><td>9c7444c6949427994b430787a153d5cceff46d5c</td><td>Journal of Computer Science 5 (11): 801-810, 2009 +<br/>ISSN 1549-3636 +<br/>© 2009 Science Publications +<br/>Boosting Kernel Discriminative Common Vectors for Face Recognition +<br/>1Department of Computer Science and Engineering, +<br/><b>SRM University, Kattankulathur, Chennai-603 203, Tamilnadu, India</b><br/><b>Bharathidasan University, Trichy, India</b></td><td>('34608395', 'C. Lakshmi', 'c. lakshmi')<br/>('2594379', 'M. Ponnavaikko', 'm. ponnavaikko')</td><td></td></tr><tr><td>9c065dfb26ce280610a492c887b7f6beccf27319</td><td>Learning from Video and Text via Large-Scale Discriminative Clustering +<br/>1 ´Ecole Normale Sup´erieure +<br/>2Inria +<br/>3CIIRC +</td><td>('19200186', 'Antoine Miech', 'antoine miech')<br/>('2285263', 'Jean-Baptiste Alayrac', 'jean-baptiste alayrac')<br/>('2329288', 'Piotr Bojanowski', 'piotr bojanowski')<br/>('1785596', 'Ivan Laptev', 'ivan laptev')<br/>('1782755', 'Josef Sivic', 'josef sivic')</td><td></td></tr><tr><td>9c781f7fd5d8168ddae1ce5bb4a77e3ca12b40b6</td><td> International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056 +<br/> Volume: 03 Issue: 07 | July-2016 www.irjet.net p-ISSN: 2395-0072 +<br/>Attribute Based Face Classification Using Support Vector Machine +<br/><b>Research Scholar, PSGR Krishnammal College for Women, Coimbatore</b><br/><b>PSGR Krishnammal College for Women, Coimbatore</b></td><td></td><td></td></tr><tr><td>9c373438285101d47ab9332cdb0df6534e3b93d1</td><td>Occupancy Detection in Vehicles Using Fisher Vector +<br/>Image Representation +<br/><b>Xerox Research Center</b><br/>Webster, NY 14580 +<br/><b>Xerox Research Center</b><br/>Webster, NY 14580 +</td><td>('1762503', 'Yusuf Artan', 'yusuf artan')<br/>('5942563', 'Peter Paul', 'peter paul')</td><td>Yusuf.Artan@xerox.com +<br/>Peter.Paul@xerox.com +</td></tr><tr><td>9cbb6e42a35f26cf1d19f4875cd7f6953f10b95d</td><td>Expression Recognition with Ri-HOG Cascade +<br/><b>Graduate School of System Informatics, Kobe University, Kobe, 657-8501, Japan</b><br/><b>RIEB, Kobe University, Kobe, 657-8501, Japan</b></td><td>('2866465', 'Jinhui Chen', 'jinhui chen')<br/>('2834542', 'Zhaojie Luo', 'zhaojie luo')<br/>('1744026', 'Tetsuya Takiguchi', 'tetsuya takiguchi')<br/>('1678564', 'Yasuo Ariki', 'yasuo ariki')</td><td></td></tr><tr><td>9ce0d64125fbaf625c466d86221505ad2aced7b1</td><td>Saliency Based Framework for Facial Expression +<br/>Recognition +<br/>To cite this version: +<br/>Facial Expression Recognition. Frontiers of Computer Science, 2017, <10.1007/s11704-017-6114-9>. +<br/><hal-01546192> +<br/>HAL Id: hal-01546192 +<br/>https://hal.archives-ouvertes.fr/hal-01546192 +<br/>Submitted on 23 Jun 2017 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('1943666', 'Rizwan Ahmed Khan', 'rizwan ahmed khan')<br/>('39469581', 'Alexandre Meyer', 'alexandre meyer')<br/>('1971616', 'Hubert Konik', 'hubert konik')<br/>('1768560', 'Saïda Bouakaz', 'saïda bouakaz')<br/>('1943666', 'Rizwan Ahmed Khan', 'rizwan ahmed khan')<br/>('39469581', 'Alexandre Meyer', 'alexandre meyer')<br/>('1971616', 'Hubert Konik', 'hubert konik')<br/>('1768560', 'Saïda Bouakaz', 'saïda bouakaz')</td><td></td></tr><tr><td>9c4cc11d0df2de42d6593f5284cfdf3f05da402a</td><td>Appears in the 14th International Conference on Pattern Recognition, ICPR’98, Queensland, Australia, August 17-20, 1998. +<br/>Enhanced Fisher Linear Discriminant Models for Face Recognition +<br/><b>George Mason University</b><br/><b>University Drive, Fairfax, VA 22030-4444, USA</b><br/> +</td><td>('39664966', 'Chengjun Liu', 'chengjun liu')<br/>('1781577', 'Harry Wechsler', 'harry wechsler')</td><td>@cs.gmu.edu +</td></tr><tr><td>9cd6a81a519545bf8aa9023f6e879521f85d4cd1</td><td>Domain-invariant Face Recognition using Learned Low-rank +<br/>Transformation +<br/><b>Duke University</b><br/>Durham, NC, 27708 +<br/><b>Duke University</b><br/>Durham, NC, 27708 +<br/><b>University of Maryland</b><br/><b>College Park, MD</b><br/>May 11, 2014 +</td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('1699339', 'Guillermo Sapiro', 'guillermo sapiro')<br/>('2682056', 'Ching-Hui Chen', 'ching-hui chen')</td><td>qiang.qiu@duke.edu +<br/>guillermo.sapiro@duke.edu +<br/>ching@umd.edu +</td></tr><tr><td>9cadd166893f1b8aaecb27280a0915e6694441f5</td><td>Appl. Math. Inf. Sci. 7, No. 2, 455-462 (2013) +<br/>455 +<br/>Applied Mathematics & Information Sciences +<br/>An International Journal +<br/>c⃝ 2013 NSP +<br/>Natural Sciences Publishing Cor. +<br/>Multi-Modal Emotion Recognition Fusing Video and +<br/>Audio +<br/><b>School of Computer Software, Tianjin University, 300072 Tianjin, China</b><br/><b>School of Computer Science and Technology, Tianjin University, 300072 Tianjin, China</b><br/>Received: 7 Sep. 2012; Revised 15 Nov. 2012; Accepted 18 Nov. 2012 +<br/>Published online: 1 Mar. 2013 +</td><td>('29962190', 'Chao Xu', 'chao xu')<br/>('2531641', 'Pufeng Du', 'pufeng du')<br/>('38465490', 'Zhiyong Feng', 'zhiyong feng')<br/>('1889014', 'Zhaopeng Meng', 'zhaopeng meng')<br/>('2375971', 'Tianyi Cao', 'tianyi cao')<br/>('36675950', 'Caichao Dong', 'caichao dong')</td><td></td></tr><tr><td>02601d184d79742c7cd0c0ed80e846d95def052e</td><td>Graphical Representation for Heterogeneous +<br/>Face Recognition +</td><td>('2299758', 'Chunlei Peng', 'chunlei peng')<br/>('10699750', 'Xinbo Gao', 'xinbo gao')<br/>('2870173', 'Nannan Wang', 'nannan wang')<br/>('38158055', 'Jie Li', 'jie li')</td><td></td></tr><tr><td>02cc96ad997102b7c55e177ac876db3b91b4e72c</td><td>MuseumVisitors: a dataset for pedestrian and group detection, gaze estimation +<br/>and behavior understanding +</td><td>('36971654', 'Federico Bartoli', 'federico bartoli')<br/>('2973738', 'Giuseppe Lisanti', 'giuseppe lisanti')<br/>('2831602', 'Lorenzo Seidenari', 'lorenzo seidenari')<br/>('2602265', 'Svebor Karaman', 'svebor karaman')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')</td><td>1{firstname.lastname}@unifi.it, University of Florence +<br/>2sk4089@columbia.edu, Columbia University +</td></tr><tr><td>02e43d9ca736802d72824892c864e8cfde13718e</td><td>Transferring a Semantic Representation for Person Re-Identification and +<br/>Search +<br/>Shi, Z; Yang, Y; Hospedales, T; XIANG, T; IEEE Conference on Computer Vision and +<br/>Pattern Recognition +<br/>© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be +<br/><b>obtained for all other uses, in any current or future media, including reprinting/republishing</b><br/>this material for advertising or promotional purposes, creating new collective works, for resale +<br/>or redistribution to servers or lists, or reuse of any copyrighted component of this work in +<br/>other works. +<br/>For additional information about this publication click this link. +<br/>http://qmro.qmul.ac.uk/xmlui/handle/123456789/10075 +<br/>Information about this research object was correct at the time of download; we occasionally +<br/>make corrections to records, please therefore check the published record when citing. For +</td><td></td><td>more information contact scholarlycommunications@qmul.ac.uk +</td></tr><tr><td>02fda07735bdf84554c193811ba4267c24fe2e4a</td><td>Illumination Invariant Face Recognition +<br/>Using Near-Infrared Images +</td><td>('34679741', 'Stan Z. Li', 'stan z. li')<br/>('1724841', 'Rufeng Chu', 'rufeng chu')<br/>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('39306651', 'Lun Zhang', 'lun zhang')</td><td></td></tr><tr><td>023ed32ac3ea6029f09b8c582efbe3866de7d00a</td><td>CENTER FOR +<br/>MACHINE PERCEPTION +<br/>Discriminative learning from +<br/>partially annotated examples +<br/>CZECH TECHNICAL +<br/><b>UNIVERSITY IN PRAGUE</b><br/>Study Programme: Electrical Engineering and +<br/>Information Technology +<br/>Branch of Study: Artificial Intelligence and Biocybernetics +<br/>CTU–CMP–2016–07 +<br/>June 14, 2016 +<br/>ftp://cmp.felk.cvut.cz/pub/cvl/articles/antoniuk/Antoniuk-TR-2016-07.pdf +<br/>Available at +<br/>Thesis Advisors: Ing. Vojtˇech Franc, Ph.D. , +<br/>prof. Ing. V´aclav Hlav´aˇc, CSc. +<br/>Acknowledgements: SGS15/201/OHK3/3T/13, CAK/TE01020197, +<br/>UP-Driving/688652, GACR/P103/12/G084. +<br/><b>Research Reports of CMP, Czech Technical University in Prague, No</b><br/>Published by +<br/>Center for Machine Perception, Department of Cybernetics +<br/><b>Faculty of Electrical Engineering, Czech Technical University</b><br/>Technick´a 2, 166 27 Prague 6, Czech Republic +<br/>fax +420 2 2435 7385, phone +420 2 2435 7637, www: http://cmp.felk.cvut.cz +</td><td>('2742026', 'Kostiantyn Antoniuk', 'kostiantyn antoniuk')</td><td>antonkos@fel.cvut.cz +</td></tr><tr><td>0241513eeb4320d7848364e9a7ef134a69cbfd55</td><td>Supervised Translation-Invariant Sparse +<br/>Coding +<br/><b>University of Illinois at Urbana Champaign</b><br/>²NEC Laboratories America at Cupertino +</td><td>('1706007', 'Jianchao Yang', 'jianchao yang')<br/>('38701713', 'Kai Yu', 'kai yu')</td><td></td></tr><tr><td>02dd0af998c3473d85bdd1f77254ebd71e6158c6</td><td>PPP: Joint Pointwise and Pairwise Image Label Prediction +<br/>1Department of Computer Science, Arizona State Univerity +<br/>2Yahoo Research +</td><td>('33513248', 'Yilin Wang', 'yilin wang')<br/>('1736632', 'Jiliang Tang', 'jiliang tang')</td><td>{yilinwang,suhang.wang,huan.liu,baoxin.li}@asu.edu +<br/>jlt@yahoo-inc.com +</td></tr><tr><td>0290523cabea481e3e147b84dcaab1ef7a914612</td><td>Generated Motion Maps +<br/><b>Tokyo Denki University</b><br/><b>National Institute of Advanced Industrial Science and Technology (AIST</b></td><td>('20505300', 'Yuta Matsuzaki', 'yuta matsuzaki')<br/>('34935749', 'Kazushige Okayasu', 'kazushige okayasu')<br/>('2462801', 'Akio Nakamura', 'akio nakamura')<br/>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')</td><td>matsuzaki.y, okayasu.k@is.dendai.ac.jp, nkmr-a@cck.dendai.ac.jp +<br/>hirokatsu.kataoka@aist.go.jp +</td></tr><tr><td>0229829e9a1eed5769a2b5eccddcaa7cd9460b92</td><td>Pooled Motion Features for First-Person Videos +<br/><b>Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA</b><br/>Figure 1: Overall representation framework of our pooled time series (PoT). Given a sequence of per-frame feature descriptors (e.g., HOF or CNN +<br/>features) from a video, PoT represents motion information in the video by computing short-term/long-term changes in each descriptor value. +<br/>In this paper, we present a new feature representation for first-person videos. +<br/>In first-person video understanding (e.g., activity recognition [4]), it is very +<br/>important to capture both entire scene dynamics (i.e., egomotion) and salient +<br/>local motion observed in videos. We describe a representation framework +</td><td>('1904850', 'Brandon Rothrock', 'brandon rothrock')</td><td></td></tr><tr><td>025720574ef67672c44ba9e7065a83a5d6075c36</td><td>Unsupervised Learning of Video Representations using LSTMs +<br/><b>University of Toronto, 6 Kings College Road, Toronto, ON M5S 3G4 CANADA</b></td><td>('2897313', 'Nitish Srivastava', 'nitish srivastava')<br/>('2711409', 'Elman Mansimov', 'elman mansimov')<br/>('1776908', 'Ruslan Salakhutdinov', 'ruslan salakhutdinov')</td><td>NITISH@CS.TORONTO.EDU +<br/>EMANSIM@CS.TORONTO.EDU +<br/>RSALAKHU@CS.TORONTO.EDU +</td></tr><tr><td>029317f260b3303c20dd58e8404a665c7c5e7339</td><td>1276 +<br/>Character Identification in Feature-Length Films +<br/>Using Global Face-Name Matching +<br/>and Yeh-Min Huang, Member, IEEE +</td><td>('1688633', 'Changsheng Xu', 'changsheng xu')<br/>('1694235', 'Hanqing Lu', 'hanqing lu')</td><td></td></tr><tr><td>026e4ee480475e63ae68570d73388f8dfd4b4cde</td><td>Evaluating gender portrayal in Bangladeshi TV +<br/>Department of CSE +<br/><b>Eastern University</b><br/>Dhaka, Bangladesh +<br/>Department of Women and Gender Studies +<br/>Rawshan E Fatima +<br/><b>Dhaka University</b><br/>Dhaka, Bangladesh +<br/><b>Khulna University of Engineering and Technology</b><br/><b>Massachusetts Institute of Technology</b><br/>Department of EEE +<br/>Khulna, Bangladesh +<br/>Media Lab +<br/>Cambridge, MA, USA +</td><td>('34688479', 'Md. Naimul Hoque', 'md. naimul hoque')<br/>('40081015', 'Manash Kumar Mandal', 'manash kumar mandal')<br/>('1706468', 'Nazmus Saquib', 'nazmus saquib')</td><td>naimul.et@easternuni.edu.bd +<br/>rawshan.e.fatima@gmail.com +<br/>manashmndl@gmail.com +<br/>saquib@mit.edu +</td></tr><tr><td>02e628e99f9a1b295458cb453c09863ea1641b67</td><td>Two-stage Convolutional Part Heatmap +<br/>Regression for the 1st 3D Face Alignment in the +<br/>Wild (3DFAW) Challenge +<br/><b>Computer Vision Laboratory, University of Nottingham, Nottingham, UK</b></td><td>('3458121', 'Adrian Bulat', 'adrian bulat')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>{adrian.bulat,yorgos.tzimiropoulos}@nottingham.ac.uk +</td></tr><tr><td>0273414ba7d56ab9ff894959b9d46e4b2fef7fd0</td><td>Photographic home styles in Congress: a +<br/>computer vision approach∗ +<br/>December 1, 2016 +</td><td>('40845190', 'L. Jason Anastasopoulos', 'l. jason anastasopoulos')<br/>('2007721', 'Dhruvil Badani', 'dhruvil badani')<br/>('2647307', 'Crystal Lee', 'crystal lee')<br/>('2361255', 'Shiry Ginosar', 'shiry ginosar')<br/>('40411568', 'Jake Williams', 'jake williams')</td><td></td></tr><tr><td>02e133aacde6d0977bca01ffe971c79097097b7f</td><td></td><td></td><td></td></tr><tr><td>02567fd428a675ca91a0c6786f47f3e35881bcbd</td><td>ACCEPTED BY IEEE TIP +<br/>Deep Label Distribution Learning +<br/>With Label Ambiguity +</td><td>('2226422', 'Bin-Bin Gao', 'bin-bin gao')<br/>('1694501', 'Chao Xing', 'chao xing')<br/>('3407628', 'Chen-Wei Xie', 'chen-wei xie')<br/>('1808816', 'Jianxin Wu', 'jianxin wu')<br/>('1735299', 'Xin Geng', 'xin geng')</td><td></td></tr><tr><td>02f4b900deabbe7efa474f2815dc122a4ddb5b76</td><td>Local and Global Optimization Techniques in Graph-based Clustering +<br/><b>The University of Tokyo, Japan</b></td><td>('11682769', 'Daiki Ikami', 'daiki ikami')<br/>('2759239', 'Toshihiko Yamasaki', 'toshihiko yamasaki')<br/>('1712839', 'Kiyoharu Aizawa', 'kiyoharu aizawa')</td><td>{ikami, yamasaki, aizawa}@hal.t.u-tokyo.ac.jp +</td></tr><tr><td>029b53f32079063047097fa59cfc788b2b550c4b</td><td></td><td></td><td></td></tr><tr><td>02bd665196bd50c4ecf05d6852a4b9ba027cd9d0</td><td></td><td></td><td></td></tr><tr><td>026b5b8062e5a8d86c541cfa976f8eee97b30ab8</td><td>MDLFace: Memorability Augmented Deep Learning for Video Face Recognition +<br/>IIIT-Delhi, India +</td><td>('1931069', 'Gaurav Goswami', 'gaurav goswami')<br/>('1875774', 'Romil Bhardwaj', 'romil bhardwaj')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')</td><td>{gauravgs,romil11092,rsingh,mayank}@iiitd.ac.in +</td></tr><tr><td>0235b2d2ae306b7755483ac4f564044f46387648</td><td>Recognition of Facial Attributes +<br/>using Adaptive Sparse Representations +<br/>of Random Patches +<br/>1 Department of Computer Science +<br/>Pontificia Universidad Cat´olica de Chile +<br/>http://dmery.ing.puc.cl +<br/>2 Department of Computer Science & Engineering +<br/><b>University of Notre Dame</b><br/>http://www.nd.edu/~kwb +</td><td>('1797475', 'Domingo Mery', 'domingo mery')</td><td></td></tr><tr><td>02467703b6e087799e04e321bea3a4c354c5487d</td><td>To appear in the CVPR Workshop on Biometrics, June 2016 +<br/>Grouper: Optimizing Crowdsourced Face Annotations∗ +<br/>Noblis +<br/>Noblis +<br/>Noblis +<br/>Noblis +<br/><b>Michigan State University</b></td><td>('9453012', 'Jocelyn C. Adams', 'jocelyn c. adams')<br/>('7996649', 'Kristen C. Allen', 'kristen c. allen')<br/>('15282121', 'Tim Miller', 'tim miller')<br/>('1718102', 'Nathan D. Kalka', 'nathan d. kalka')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>jocelyn.adams@noblis.org +<br/>kristen.allen@noblis.org +<br/>timothy.miller@noblis.org +<br/>nathan.kalka@noblis.org +<br/>jain@cse.msu.edu +</td></tr><tr><td>02e39f23e08c2cb24d188bf0ca34141f3cc72d47</td><td>REMOVING ILLUMINATION ARTIFACTS FROM FACE IMAGES USING THE NUISANCE +<br/>ATTRIBUTE PROJECTION +<br/>Vitomir ˇStruc, Boˇstjan Vesnicer, France Miheliˇc, Nikola Paveˇsi´c +<br/><b>Faculty of Electrical Engineering, University of Ljubljana, Tr za ska 25, SI-1000 Ljubljana, Slovenia</b></td><td></td><td></td></tr><tr><td>023be757b1769ecb0db810c95c010310d7daf00b</td><td>YANG, MOU, ZHANG ET AL.: FACE ALIGNMENT ASSISTED BY HEAD POSE ESTIMATION1 +<br/>Face Alignment Assisted by Head Pose +<br/>Estimation +<br/>1 Computer Laboratory +<br/><b>University of Cambridge</b><br/>Cambridge, UK +<br/>2 School of EECS +<br/><b>Queen Mary University of London</b><br/>London, UK +<br/>3 Faculty of Arts & Sciences +<br/><b>Harvard University</b><br/>Cambridge, MA, US +</td><td>('2966679', 'Heng Yang', 'heng yang')<br/>('2734386', 'Wenxuan Mou', 'wenxuan mou')<br/>('40491398', 'Yichi Zhang', 'yichi zhang')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')<br/>('1781916', 'Hatice Gunes', 'hatice gunes')<br/>('39626495', 'Peter Robinson', 'peter robinson')</td><td>heng.yang@cl.cam.ac.uk +<br/>w.mou@qmul.ac.uk +<br/>yichizhang@fas.harvard.edu +<br/>i.patras@qmul.ac.uk +<br/>h.gunes@qmul.ac.uk +<br/>peter.robinson@cl.cam.ac.uk +</td></tr><tr><td>0278acdc8632f463232e961563e177aa8c6d6833</td><td>Selective Transfer Machine for Personalized +<br/>Facial Expression Analysis +<br/>1 INTRODUCTION +<br/>Index Terms—Facial expression analysis, personalization, domain adaptation, transfer learning, support vector machine (SVM) +<br/>A UTOMATIC facial AU detection confronts a number of +</td><td>('39336289', 'Wen-Sheng Chu', 'wen-sheng chu')<br/>('3141839', 'Fernando De la Torre', 'fernando de la torre')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td></td></tr><tr><td>0209389b8369aaa2a08830ac3b2036d4901ba1f1</td><td>DenseReg: Fully Convolutional Dense Shape Regression In-the-Wild +<br/>Rıza Alp G¨uler 1 +<br/>1INRIA-CentraleSup´elec, France +<br/><b>Imperial College London, UK</b><br/><b>University College London, UK</b></td><td>('2814229', 'George Trigeorgis', 'george trigeorgis')<br/>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('2796644', 'Patrick Snape', 'patrick snape')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('2010660', 'Iasonas Kokkinos', 'iasonas kokkinos')</td><td>1riza.guler@inria.fr +<br/>2{g.trigeorgis, e.antonakos, p.snape,s.zafeiriou}@imperial.ac.uk +<br/>3i.kokkinos@cs.ucl.ac.uk +</td></tr><tr><td>02c993d361dddba9737d79e7251feca026288c9c</td><td></td><td></td><td></td></tr><tr><td>02239ae5e922075a354169f75f684cad8fdfd5ab</td><td>Commonly Uncommon: +<br/>Semantic Sparsity in Situation Recognition +<br/><b>Computer Science and Engineering, University of Washington, Seattle, WA</b><br/><b>Allen Institute for Arti cial Intelligence (AI2), Seattle, WA</b><br/><b>University of Virginia, Charlottesville, VA</b></td><td>('2064210', 'Mark Yatskar', 'mark yatskar')<br/>('2004053', 'Vicente Ordonez', 'vicente ordonez')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')</td><td>[my89, lsz, ali]@cs.washington.edu, vicente@cs.virginia.edu +</td></tr><tr><td>02d650d8a3a9daaba523433fbe93705df0a7f4b1</td><td>How Does Aging Affect Facial Components? +<br/><b>Michigan State University</b></td><td>('40653304', 'Charles Otto', 'charles otto')<br/>('34393045', 'Hu Han', 'hu han')</td><td>{ottochar,hhan,jain}@cse.msu.edu +</td></tr><tr><td>0294f992f8dfd8748703f953925f9aee14e1b2a2</td><td>Blur-Robust Face Recognition via +<br/>Transformation Learning +<br/><b>Beijing University of Posts and Telecommunications, Beijing, China</b></td><td>('40448827', 'Jun Li', 'jun li')<br/>('1690083', 'Chi Zhang', 'chi zhang')<br/>('23224233', 'Jiani Hu', 'jiani hu')<br/>('1774956', 'Weihong Deng', 'weihong deng')</td><td></td></tr><tr><td>02820c1491b10a1ff486fed32c269e4077c36551</td><td>Active User Authentication for Smartphones: A Challenge +<br/>Data Set and Benchmark Results +<br/>1Department of Electrical and Computer Engineering and the Center for Automation Research, +<br/><b>UMIACS, University of Maryland, College Park, MD</b><br/><b>Rutgers, The State University of New Jersey, 508 CoRE, 94 Brett Rd, Piscataway, NJ</b></td><td>('3152615', 'Upal Mahbub', 'upal mahbub')<br/>('40599829', 'Sayantan Sarkar', 'sayantan sarkar')</td><td>{umahbub, ssarkar2, rama}@umiacs.umd.edu +<br/>vishal.m.patel@rutgers.edu∗ +</td></tr><tr><td>a40edf6eb979d1ddfe5894fac7f2cf199519669f</td><td>Improving Facial Attribute Prediction using Semantic Segmentation +<br/>Center for Research in Computer Vision +<br/><b>University of Central Florida</b></td><td>('3222250', 'Mahdi M. Kalayeh', 'mahdi m. kalayeh')<br/>('40206014', 'Boqing Gong', 'boqing gong')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>Mahdi@eecs.ucf.edu +<br/>bgong@crcv.ucf.edu +<br/>shah@crcv.ucf.edu +</td></tr><tr><td>a46283e90bcdc0ee35c680411942c90df130f448</td><td></td><td></td><td></td></tr><tr><td>a4a5ad6f1cc489427ac1021da7d7b70fa9a770f2</td><td>Yudistira and Kurita EURASIP Journal on Image and Video +<br/>Processing (2017) 2017:85 +<br/>DOI 10.1186/s13640-017-0235-9 +<br/>EURASIP Journal on Image +<br/>and Video Processing +<br/>RESEARCH +<br/>Open Access +<br/>Gated spatio and temporal convolutional +<br/>neural network for activity recognition: +<br/>towards gated multimodal deep learning +</td><td>('2035597', 'Novanto Yudistira', 'novanto yudistira')<br/>('1742728', 'Takio Kurita', 'takio kurita')</td><td></td></tr><tr><td>a4876b7493d8110d4be720942a0f98c2d116d2a0</td><td>Multi-velocity neural networks for gesture recognition in videos +<br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, MA +</td><td>('37381309', 'Otkrist Gupta', 'otkrist gupta')<br/>('2283049', 'Dan Raviv', 'dan raviv')<br/>('1717566', 'Ramesh Raskar', 'ramesh raskar')</td><td>otkrist@mit.edu +<br/>raviv@mit.edu +<br/>raskar@media.mit.edu +</td></tr><tr><td>a40f8881a36bc01f3ae356b3e57eac84e989eef0</td><td>End-to-end semantic face segmentation with conditional +<br/>random fields as convolutional, recurrent and adversarial +<br/>networks +</td><td>('3038211', 'Umut Güçlü', 'umut güçlü')<br/>('1920611', 'Meysam Madadi', 'meysam madadi')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('1857280', 'Xavier Baró', 'xavier baró')<br/>('38485168', 'Rob van Lier', 'rob van lier')<br/>('2052286', 'Marcel van Gerven', 'marcel van gerven')</td><td></td></tr><tr><td>a4a0b5f08198f6d7ea2d1e81bd97fea21afe3fc3</td><td>E +<br/>Feature Transfer +<br/>MSc Thesis +<br/>written by +<br/>degree of +<br/>Master of Science +<br/><b>at the Delft University of Technology</b><br/>Date of the public defense: Members of the Thesis Committee: +<br/>August 31, 2017 +<br/>Prof. Marcel Reinders +<br/>Dr. Julian Urbano Merino +<br/>Dr. Gonzalez Adrlana (Bosch) +</td><td>('1694101', 'Yue Liu', 'yue liu')<br/>('37806314', 'Silvia-Laura Pintea', 'silvia-laura pintea')<br/>('30445013', 'Jan van Gemert', 'jan van gemert')<br/>('2372050', 'Ildiko Suveg', 'ildiko suveg')<br/>('30445013', 'Jan van Gemert', 'jan van gemert')<br/>('37806314', 'Silvia-Laura Pintea', 'silvia-laura pintea')<br/>('2372050', 'Ildiko Suveg', 'ildiko suveg')</td><td></td></tr><tr><td>a46086e210c98dcb6cb9a211286ef906c580f4e8</td><td>Fusing Multi-Stream Deep Networks for Video Classification +<br/><b>Fudan University, Shanghai, China</b><br/>Alibaba Group, Seattle, USA +</td><td>('3099139', 'Zuxuan Wu', 'zuxuan wu')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('31825486', 'Xi Wang', 'xi wang')<br/>('1743864', 'Hao Ye', 'hao ye')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')<br/>('1715001', 'Jun Wang', 'jun wang')</td><td>zxwu, ygj, xwang10, haoye10, xyxue@fudan.edu.cn +<br/>wongjun@gmail.com +</td></tr><tr><td>a44590528b18059b00d24ece4670668e86378a79</td><td>Learning the Hierarchical Parts of Objects by Deep +<br/>Non-Smooth Nonnegative Matrix Factorization +</td><td>('19275690', 'Jinshi Yu', 'jinshi yu')<br/>('1764724', 'Guoxu Zhou', 'guoxu zhou')<br/>('1747156', 'Andrzej Cichocki', 'andrzej cichocki')<br/>('1795838', 'Shengli Xie', 'shengli xie')</td><td></td></tr><tr><td>a472d59cff9d822f15f326a874e666be09b70cfd</td><td>VISUAL LEARNING WITH WEAKLY LABELED VIDEO +<br/>A DISSERTATION +<br/>SUBMITTED TO THE DEPARTMENT OF COMPUTER SCIENCE +<br/>AND THE COMMITTEE ON GRADUATE STUDIES +<br/><b>OF STANFORD UNIVERSITY</b><br/>IN PARTIAL FULFILLMENT OF THE REQUIREMENTS +<br/>FOR THE DEGREE OF +<br/>DOCTOR OF PHILOSOPHY +<br/>May 2015 +</td><td>('3355264', 'Kevin Tang', 'kevin tang')</td><td></td></tr><tr><td>a4c430b7d849a8f23713dc283794d8c1782198b2</td><td>Video Concept Embedding +<br/>1. Introduction +<br/>In the area of natural language processing, there has been +<br/>much success in learning distributed representations for +<br/>words as vectors. Doing so has an advantage over using +<br/>simple labels, or a one-hot coding scheme for representing +<br/>individual words. In learning distributed vector representa- +<br/>tions for words, we manage to capture semantic relatedness +<br/>of words in vector distance. For example, the word vector +<br/>for ”car” and ”road” should end up being closer together in +<br/>the vector space representation than ”car” and ”penguin”. +<br/>This has been very useful in NLP areas of machine transla- +<br/>tion and semantic understanding. +<br/>In the computer vision domain, video understanding is a +<br/>very important topic. +<br/>It is made hard due to the large +<br/>amount of high dimensional data in videos. One strategy +<br/>to address this is to summarize a video into concepts (eg. +<br/>running, climbing, cooking). This allows us to represent a +<br/>video in a very natural way to humans, such as a sequence +<br/>of semantic events. However this has the same shortcom- +<br/>ings that one-hot coding of words have. +<br/>The goal of this project is to find a meaningful way to em- +<br/>bed video concepts into a vector space. The hope would +<br/>be to capture semantic relatedness of concepts in a vector +<br/>representation, essentially doing for videos what word2vec +<br/>did for text. Having a vector representation for video con- +<br/>cepts would help in areas such as semantic video retrieval +<br/>and video classification, as it would provide a statistically +<br/>meaningful and robust way of representing videos as lower +<br/>dimensional vectors. An interesting thing would be to ob- +<br/>serve if such a vector representation would result in ana- +<br/>logical reasoning using simple vector arithmetic. +<br/>Figure 1 shows an example of concepts detected at differ- +<br/>ent snapshots in the same video. For example, consider +<br/>the scenario where the concepts Kicking a ball, Soccer and +<br/>Running are detected in the three snapshots respectively +<br/>(from left to right). Since, these snapshots belong in the +<br/>same video, we expect that these concepts are semantically +<br/>similar and that they should lie close in the resulting em- +<br/>bedding space. The aim of this project is to find a vector +<br/>space embedding for the space of concepts such that vector +<br/>representations for semantically similar concepts (in this +<br/>Figure 1. Example snapshots from the same video +<br/>case, Running, Kicking and Soccer) lie in the vicinity of +<br/>each other. +<br/>2. Related Work +<br/>(Mikolov et al., 2013a) introduces the popular skip-gram +<br/>model to learn distributed representations of words from +<br/>very large linguistic datasets. Specifically, it uses each +<br/>word as an input to a log-linear classifier and predict words +<br/>within a certain range before and after the current word in +<br/>the dataset. +<br/>(Mikolov et al., 2013b) extends this model +<br/>to learn representations for phrases, in addition to words, +<br/>and also improve the quality of vectors and training speed. +<br/>These works also show that the skip-gram model exhibits +<br/>a linear structure that enables it to perform reasoning using +<br/>basic vector arithmetic. The skip-gram model from these +<br/>works is the basis of our model in learning representations +<br/>for concepts. +<br/>(Le & Mikolov, 2014) extends the concept of word vectors +<br/>to sentences and paragraphs. Their approach is more in- +<br/>volved than a simple bag of words approach, in that it tries +<br/>to capture the nature of the words in the paragraph. They +<br/>construct the paragraph vector in such a way that it can be +<br/>used to predict the word vectors that are contained inside +<br/>the paragraph. They do this by first learning word vectors, +<br/>such that the probability of a word vector given its context +<br/>is maximized. To learn paragraph vectors, the paragraph +<br/>is essentially treated as a word, and the words it contains +<br/>become the context. This provides a key insight in how +<br/>a set of concept vectors can be used together to provide a +<br/>more meaningful vector representation for videos, which +<br/>can then be used for retrieval. +<br/>(Hu et al.) utilizes structured knowledge in the data to learn +<br/>distributed representations that improve semantic related- +</td><td>('2387189', 'Anirudh Vemula', 'anirudh vemula')<br/>('32203964', 'Rahul Nallamothu', 'rahul nallamothu')<br/>('9619757', 'Syed Zahir Bokhari', 'syed zahir bokhari')</td><td>AVEMULA1@ANDREW.CMU.EDU +<br/>RNALLAMO@ANDREW.CMU.EDU +<br/>SBOKHARI@ANDREW.CMU.EDU +</td></tr><tr><td>a4cc626da29ac48f9b4ed6ceb63081f6a4b304a2</td><td></td><td></td><td></td></tr><tr><td>a4f37cfdde3af723336205b361aefc9eca688f5c</td><td>Recent Advances +<br/>in Face Recognition +</td><td></td><td></td></tr><tr><td>a481e394f58f2d6e998aa320dad35c0d0e15d43c</td><td>Selectively Guiding Visual Concept Discovery +<br/><b>Colorado State University</b><br/>Fort Collins, Colorado +</td><td>('2857477', 'Maggie Wigness', 'maggie wigness')<br/>('1694404', 'Bruce A. Draper', 'bruce a. draper')<br/>('1757322', 'J. Ross Beveridge', 'j. ross beveridge')</td><td>mwigness,draper,ross@cs.colostate.edu +</td></tr><tr><td>a30869c5d4052ed1da8675128651e17f97b87918</td><td>Fine-Grained Comparisons with Attributes +</td><td>('2206630', 'Aron Yu', 'aron yu')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>a3ebacd8bcbc7ddbd5753935496e22a0f74dcf7b</td><td>First International Workshop on Adaptive Shot Learning +<br/>for Gesture Understanding and Production +<br/>ASL4GUP 2017 +<br/>Held in conjunction with IEEE FG 2017, in May 30, 2017, +<br/>Washington DC, USA +</td><td></td><td></td></tr><tr><td>a3d8b5622c4b9af1f753aade57e4774730787a00</td><td>Pose-Aware Person Recognition +<br/>Anoop Namboodiri (cid:63) +<br/>(cid:63) CVIT, IIIT Hyderabad, India +<br/>† Facebook AI Research +</td><td>('37956314', 'Vijay Kumar', 'vijay kumar')<br/>('2210374', 'Manohar Paluri', 'manohar paluri')<br/>('1694502', 'C. V. Jawahar', 'c. v. jawahar')</td><td></td></tr><tr><td>a322479a6851f57a3d74d017a9cb6d71395ed806</td><td>Towards Pose Invariant Face Recognition in the Wild +<br/><b>National University of Singapore</b><br/><b>National University of Defense Technology</b><br/><b>Nanyang Technological University</b><br/>4Panasonic R&D Center Singapore +<br/><b>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences</b><br/><b>Qihoo 360 AI Institute</b></td><td>('2668358', 'Sugiri Pranata', 'sugiri pranata')<br/>('3493398', 'Shengmei Shen', 'shengmei shen')<br/>('1757173', 'Junliang Xing', 'junliang xing')<br/>('46509407', 'Jian Zhao', 'jian zhao')<br/>('5524736', 'Yu Cheng', 'yu cheng')<br/>('33419682', 'Lin Xiong', 'lin xiong')<br/>('2757639', 'Jianshu Li', 'jianshu li')<br/>('40345914', 'Fang Zhao', 'fang zhao')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')</td><td></td></tr><tr><td>a3017bb14a507abcf8446b56243cfddd6cdb542b</td><td>Face Localization and Recognition in Varied +<br/>Expressions and Illumination +<br/>Hui-Yu Huang, Shih-Hang Hsu +<br/> +</td><td></td><td></td></tr><tr><td>a3c8c7da177cd08978b2ad613c1d5cb89e0de741</td><td>A Spatio-temporal Approach for Multiple +<br/>Object Detection in Videos Using Graphs +<br/>and Probability Maps +<br/><b>University of S ao Paulo, S ao Paulo, Brazil</b><br/>2 Institut Mines T´el´ecom, T´el´ecom ParisTech, CNRS LTCI, Paris, France +</td><td>('1863046', 'Henrique Morimitsu', 'henrique morimitsu')<br/>('1695917', 'Isabelle Bloch', 'isabelle bloch')</td><td>henriquem87@gmail.com +</td></tr><tr><td>a378fc39128107815a9a68b0b07cffaa1ed32d1f</td><td>Determining a Suitable Metric When using Non-negative Matrix Factorization∗ +<br/>Computer Vision Center, Dept. Inform`atica +<br/>Universitat Aut`onoma de Barcelona +<br/>08193 Bellaterra, Barcelona, Spain +</td><td>('1761407', 'David Guillamet', 'david guillamet')</td><td>{davidg,jordi}@cvc.uab.es +</td></tr><tr><td>a34d75da87525d1192bda240b7675349ee85c123</td><td>Naive-Deep Face Recognition: Touching the Limit of LFW Benchmark or Not? +<br/>Face++, Megvii Inc. +<br/>Face++, Megvii Inc. +<br/>Face++, Megvii Inc. +</td><td>('1848243', 'Erjin Zhou', 'erjin zhou')<br/>('2695115', 'Zhimin Cao', 'zhimin cao')<br/>('2274228', 'Qi Yin', 'qi yin')</td><td>zej@megvii.com +<br/>czm@megvii.com +<br/>yq@megvii.com +</td></tr><tr><td>a301ddc419cbd900b301a95b1d9e4bb770afc6a3</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>DECK: Discovering Event Composition Knowledge from +<br/>Web Images for Zero-Shot Event Detection and Recounting in Videos +<br/><b>University of Southern California</b><br/><b>IIIS, Tsinghua University</b><br/>‡ Google Research +</td><td>('2551285', 'Chuang Gan', 'chuang gan')<br/>('1726241', 'Chen Sun', 'chen sun')</td><td></td></tr><tr><td>a3dc109b1dff3846f5a2cc1fe2448230a76ad83f</td><td>J.Savitha et al, International Journal of Computer Science and Mobile Computing, Vol.4 Issue.4, April- 2015, pg. 722-731 +<br/>Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/>A Monthly Journal of Computer Science and Information Technology +<br/>ISSN 2320–088X +<br/> IJCSMC, Vol. 4, Issue. 4, April 2015, pg.722 – 731 +<br/> RESEARCH ARTICLE +<br/>ACTIVE APPEARANCE MODEL AND PCA +<br/>BASED FACE RECOGNITION SYSTEM +<br/>Mrs. J.Savitha M.Sc., M.Phil. +<br/><b>Ph.D Research Scholar, Karpagam University, Coimbatore, Tamil Nadu, India</b><br/>Dr. A.V.Senthil Kumar +<br/><b>Director, Hindustan College of Arts and Science, Coimbatore, Tamil Nadu, India</b></td><td></td><td>Email: savitha.sanjay1@gmail.com +<br/>Email: avsenthilkumar@gmail.com +</td></tr><tr><td>a3f69a073dcfb6da8038607a9f14eb28b5dab2db</td><td>Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18) +<br/>1184 +</td><td></td><td></td></tr><tr><td>a38045ed82d6800cbc7a4feb498e694740568258</td><td>UNLV Theses, Dissertations, Professional Papers, and Capstones +<br/>5-2010 +<br/>African American and Caucasian males' evaluation +<br/>of racialized female facial averages +<br/>Rhea M. Watson +<br/><b>University of Nevada Las Vegas</b><br/>Follow this and additional works at: http://digitalscholarship.unlv.edu/thesesdissertations +<br/>Part of the Cognition and Perception Commons, Race and Ethnicity Commons, and the Social +<br/>Psychology Commons +<br/>Repository Citation +<br/>Watson, Rhea M., "African American and Caucasian males' evaluation of racialized female facial averages" (2010). UNLV Theses, +<br/>Dissertations, Professional Papers, and Capstones. 366. +<br/>http://digitalscholarship.unlv.edu/thesesdissertations/366 +</td><td></td><td>This Thesis is brought to you for free and open access by Digital Scholarship@UNLV. It has been accepted for inclusion in UNLV Theses, Dissertations, +<br/>Professional Papers, and Capstones by an authorized administrator of Digital Scholarship@UNLV. For more information, please contact +<br/>digitalscholarship@unlv.edu. +</td></tr><tr><td>a3f684930c5c45fcb56a2b407d26b63879120cbf</td><td>LPM for Fast Action Recognition with Large Number of Classes +<br/>School of Electrical Engineering and Computer Scinece +<br/><b>University of Ottawa, Ottawa, On, Canada</b><br/>Department of Electronics and Information Engineering +<br/><b>Hua Zhong University of Science and Technology, Wuhan, China</b><br/>1. Introduction +<br/>In this paper, we provide an overview of the Local Part +<br/>Model system for the THUMOS 2013: Action Recognition +<br/>with a Large Number of Classes1 evaluations. Our system +<br/>uses a combination of fast random sampling feature extrac- +<br/>tion and local part model feature representation. +<br/>Over the last decade, the advances in the area of com- +<br/>puter vision and pattern recognition have fuelled a large +<br/>amount of research with great progress in human action +<br/>recognition. Much of the early progress [1, 5, 14] has been +<br/>reported on atomic actions with several categories based +<br/>on staged videos captured under controlled settings, such +<br/>as KTH [14] and Weizmann [1]. More recently, there are +<br/>emerging interests for sophisticated algorithms in recogniz- +<br/>ing actions from realistic video. Such interests involve two +<br/>prospects: 1) In comparison to image classification evalu- +<br/>ating millions of images with over one thousand categories, +<br/>action recognition is still at its initial stage. It is important +<br/>to develop reliable, automatic methods which scale to large +<br/>numbers of action categories captured in realistic settings. +<br/>2) With over 100 hours of videos are uploaded to YouTube +<br/>every minute2, and millions of surveillance cameras all over +<br/>the world, the need for efficient recognition of the visual +<br/>events in the video is crucial for real world applications. +<br/>Recent studies [5, 10, 11, 21] have shown that lo- +<br/>cal spatio-temporal features can achieve remarkable per- +<br/>formance when represented by popular bag-of-features +<br/>method. A recent trend is the use of dense sampled points +<br/>[16, 21] and trajectories [7, 19] to improve the perfor- +<br/>mance. Local Part Model [15] achieved state-of-the-art per- +<br/>formance on real-life datasets with high efficiency when +<br/>combined with random sampling over high density sam- +<br/>1http://crcv.ucf.edu/ICCV13-Action-Workshop/index.html +<br/>2http://www.youtube.com/yt/press/statistics.html +<br/>pling grids. +<br/>In this paper, we focus on recognize human +<br/>action “in the wild” with large number of classes. More +<br/>specifically, we aim to improve the state-of-the-art Local +<br/>Part Model method on large scale real-life action datasets. +<br/>The paper is organized as follows: The next section re- +<br/>views the LPM algorithm. Section 3 introduces four differ- +<br/>ent descriptors we will use. In section 4, we present some +<br/>experimental results and analysis. The paper is completed +<br/>with a brief conclusion. The code for computing random +<br/>sampling with Local Part Model is available on-line3. +<br/>2. LPM algorithm +<br/>Inspired by the multiscale, deformable part model [6] +<br/>for object classification, we proposed a 3D multiscale part +<br/>model in [16]. However, instead of adopting deformable +<br/>“parts”, we used “parts” with fixed size and location on the +<br/>purpose of maintaining both structural information and lo- +<br/>cal events ordering for action recognition. As shown in Fig- +<br/>ure 1, the local part model includes both a coarse primi- +<br/>tive level root feature covering event-content statistics and +<br/>higher resolution overlapping part filters incorporating lo- +<br/>cal structural and temporal relations. +<br/>More recently, we [15] applied random sampling method +<br/>with local part model over a very dense sampling grid +<br/>and achieved state-of-the-art performance on realistic large +<br/>scale datasets with potential for real-time recognition. Un- +<br/>der the local part model, a feature consists of a coarse global +<br/>root filter and several fine overlapped part filters. The root +<br/>filter is extracted on the video at half the resolution. This +<br/>way, a high density grid can be defined with far less sam- +<br/>ples. For every coarse root filter, a group of fine part filters +<br/>are computed at full video resolution and at locations rela- +<br/>tive to their root filter reference position. These part filters +<br/>3https://github.com/fshi/actionMBH +</td><td>('36925389', 'Feng Shi', 'feng shi')<br/>('1745632', 'Emil Petriu', 'emil petriu')</td><td>fshi98@gmail.com, {laganier, petriu}@site.uottawa.ca +<br/>zhenhaiyu@mail.hust.edu.cn +</td></tr><tr><td>a3f78cc944ac189632f25925ba807a0e0678c4d5</td><td>Action Recognition in Realistic Sports Videos +</td><td>('1799979', 'Khurram Soomro', 'khurram soomro')<br/>('40029556', 'Amir Roshan Zamir', 'amir roshan zamir')</td><td></td></tr><tr><td>a33f20773b46283ea72412f9b4473a8f8ad751ae</td><td></td><td></td><td></td></tr><tr><td>a3a6a6a2eb1d32b4dead9e702824375ee76e3ce7</td><td>Multiple Local Curvature Gabor Binary +<br/>Patterns for Facial Action Recognition +<br/>Signal Processing Laboratory (LTS5), +<br/>´Ecole Polytechnique F´ed´erale de Lausanne, Switzerland +</td><td>('2383305', 'Nuri Murat Arar', 'nuri murat arar')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')</td><td>{anil.yuce,murat.arar,jean-philippe.thiran}@epfl.ch +</td></tr><tr><td>a32c5138c6a0b3d3aff69bcab1015d8b043c91fb</td><td>Downloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Electronic-Imaging on 9/19/2018 +<br/>Terms of Use: https://www.spiedigitallibrary.org/terms-of-use +<br/>Videoredaction:asurveyandcomparisonofenablingtechnologiesShaganSahAmeyaShringiRaymondPtuchaAaronBurryRobertLoceShaganSah,AmeyaShringi,RaymondPtucha,AaronBurry,RobertLoce,“Videoredaction:asurveyandcomparisonofenablingtechnologies,”J.Electron.Imaging26(5),051406(2017),doi:10.1117/1.JEI.26.5.051406. </td><td></td><td></td></tr><tr><td>a32d4195f7752a715469ad99cb1e6ebc1a099de6</td><td>Hindawi Publishing Corporation +<br/>e Scientific World Journal +<br/>Volume 2014, Article ID 749096, 10 pages +<br/>http://dx.doi.org/10.1155/2014/749096 +<br/>Research Article +<br/>The Potential of Using Brain Images for Authentication +<br/><b>College of Mechatronic Engineering and Automation, National University of Defense Technology</b><br/>Changsha, Hunan 410073, China +<br/>Received 6 May 2014; Accepted 19 June 2014; Published 10 July 2014 +<br/>Academic Editor: Wangmeng Zuo +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Biometric recognition (also known as biometrics) refers to the automated recognition of individuals based on their biological or +<br/>behavioral traits. Examples of biometric traits include fingerprint, palmprint, iris, and face. The brain is the most important and +<br/>complex organ in the human body. Can it be used as a biometric trait? In this study, we analyze the uniqueness of the brain and +<br/>try to use the brain for identity authentication. The proposed brain-based verification system operates in two stages: gray matter +<br/>extraction and gray matter matching. A modified brain segmentation algorithm is implemented for extracting gray matter from +<br/>an input brain image. Then, an alignment-based matching algorithm is developed for brain matching. Experimental results on two +<br/>data sets show that the proposed brain recognition system meets the high accuracy requirement of identity authentication. Though +<br/>currently the acquisition of the brain is still time consuming and expensive, brain images are highly unique and have the potential +<br/>possibility for authentication in view of pattern recognition. +<br/>1. Introduction +<br/>Identity authentication is an important task for different +<br/>applications including access control, ATM card verification, +<br/>and forensic affairs. Compared with conventional methods +<br/>(e.g., key, ID card, and password), biometric recognition +<br/>is more resistant to social engineering attacks (e.g., theft). +<br/>Biometric recognition is also intrinsically superior that makes +<br/>it unforgettable. During the past few decades, biometric tech- +<br/>nologies have shown more and more importance in various +<br/>applications [1, 2]. Among them, recognition technologies +<br/>based on fingerprint [3, 4], palmprint [5, 6], iris [7, 8], and +<br/>face [9, 10] are the most popular. +<br/>The brain is the center of the nervous system and the most +<br/>important and complex organ in the human body. Though +<br/>different brains may be alike in the way they act and have +<br/>similar traits, scientists have confirmed that no two brains are +<br/>or will ever be the same [11]. Both genes (what we inherit) +<br/>and experience (what we learn) could allow individual brains +<br/>to develop in distinctly different ways. Recent studies show +<br/>that the so-called jumping genes, which ensure that identical +<br/>twins are different, may also influence the brains [12]. All +<br/>these studies show that the human brain is a work of genius in +<br/>its design and capabilities, and it is unique. Though brain gray +<br/>matter will change with age or disease, it shows steadiness in +<br/>adulthood [13, 14]. The question we are interested in this study +<br/>is as follows: can we use the brain for identity authentication? +<br/>This paper analyzes the uniqueness of human brain +<br/>and proposes to use the brain for personal identification +<br/>(authentication). Compared with other biometric techniques, +<br/>brain recognition is more resistant to forgery (e.g., fake +<br/>fingerprints [15]) and spoofing (e.g., face disguise [16]). Brain +<br/>recognition is also more reliable to identify the escapee +<br/>since one’s brain can hardly be modified, whereas other +<br/>biologic traits may be altered, such as altered fingerprints [17]. +<br/>Palaniappan and Mandic [18] established a Visual Evoked +<br/>Potential- (VEP-) based biometrics, and simulations have +<br/>indicated the significant potential of brain electrical activity +<br/>as a biometric tool. However, VEP is not robust to the +<br/>activity of brain. Aloui et al. [19] extracted characteristics of +<br/>brain images and used them in an application as a biometric +<br/>tool to identify individuals. Their method just uses a single +<br/>slice of the brain and thus suffers from the influence of +<br/>noise. Another drawback of this method is that it only uses +</td><td>('40326124', 'Fanglin Chen', 'fanglin chen')<br/>('8526311', 'Zongtan Zhou', 'zongtan zhou')<br/>('1730001', 'Hui Shen', 'hui shen')<br/>('2517668', 'Dewen Hu', 'dewen hu')<br/>('40326124', 'Fanglin Chen', 'fanglin chen')</td><td>Correspondence should be addressed to Dewen Hu; dwhu@nudt.edu.cn +</td></tr><tr><td>a3d78bc94d99fdec9f44a7aa40c175d5a106f0b9</td><td>Recognizing Violence in Movies +<br/>CIS400/401 Project Final Report +<br/>Univ. of Pennsylvania +<br/>Philadelphia, PA +<br/>Univ. of Pennsylvania +<br/>Philadelphia, PA +<br/>Ben Sapp +<br/>Univ. of Pennsylvania +<br/>Philadelphia, PA +<br/>Univ. of Pennsylvania +<br/>Philadelphia, PA +</td><td>('1908780', 'Lei Kang', 'lei kang')<br/>('1685978', 'Ben Taskar', 'ben taskar')</td><td>kanglei@seas.upenn.edu +<br/>mjiawei@seas.upenn.edu +<br/>bensapp@cis.upenn.edu +<br/>taskar@cis.upenn.edu +</td></tr><tr><td>a3eab933e1b3db1a7377a119573ff38e780ea6a3</td><td>978-1-4244-4296-6/10/$25.00 ©2010 IEEE +<br/>838 +<br/>ICASSP 2010 +</td><td></td><td></td></tr><tr><td>a308077e98a611a977e1e85b5a6073f1a9bae6f0</td><td>Hindawi Publishing Corporation +<br/>e Scientific World Journal +<br/>Volume 2014, Article ID 810368, 15 pages +<br/>http://dx.doi.org/10.1155/2014/810368 +<br/>Review Article +<br/>Intelligent Screening Systems for Cervical Cancer +<br/><b>Faculty of Engineering Building, University of Malaya, 50603 Kuala Lumpur, Malaysia</b><br/>Received 24 December 2013; Accepted 11 February 2014; Published 11 May 2014 +<br/>Academic Editors: S. Balochian, V. Bhatnagar, and Y. Zhang +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Advent of medical image digitalization leads to image processing and computer-aided diagnosis systems in numerous clinical +<br/>applications. These technologies could be used to automatically diagnose patient or serve as second opinion to pathologists. This +<br/>paper briefly reviews cervical screening techniques, advantages, and disadvantages. The digital data of the screening techniques +<br/>are used as data for the computer screening system as replaced in the expert analysis. Four stages of the computer system are +<br/>enhancement, features extraction, feature selection, and classification reviewed in detail. The computer system based on cytology +<br/>data and electromagnetic spectra data achieved better accuracy than other data. +<br/>1. Introduction +<br/>Cervical cancer is a leading cause of mortality and morbidity, +<br/>which comprises approximately 12% of all cancers in women +<br/>worldwide according to World Health Organization (WHO). +<br/>In fact, the annual global statistics of WHO estimated 470 +<br/>600 new cases and 233 400 deaths from cervical cancer +<br/>around the year 2000. As reported in National Cervical +<br/>Cancer Coalition (NCCC) in 2010, cervical cancer is a cancer +<br/>of the cervix which is commonly caused by a virus named +<br/>Human Papillomavirus (HPV) [1]. The virus can damage +<br/>cells in the cervix, namely, squamous cells and glandular +<br/>cells that may develop into squamous cell carcinoma (cancer +<br/>of the squamous cells) and adenocarcinoma (cancer of the +<br/>glandular cells), respectively. Squamous cell carcinoma can +<br/>be thought of as similar to skin cancer because it begins on +<br/>the surface of the ectocervix. Adenocarcinoma begins further +<br/>inside the uterus, in the mucus-producing gland cells of the +<br/>endocervix [2]. +<br/>Cervical cancer develops from normal to precancerous +<br/>cells (dysplasia) over a period of two to three decades [3]. +<br/>Even though the dysplasia cells look like cancer cells, they +<br/>are not malignant cells. These cells are known as cervical +<br/>intraepithelial neoplasia (CIN) which is usually of low grade, +<br/>and they only affect the surface of the cervical tissue. The +<br/>majority will regress back to normal spontaneously. Over +<br/>time, a small proportion will continue to develop into cancer. +<br/>Based on WHO system, the level of CIN growth can be +<br/>divided into grades 1, 2, and 3. It should be noted that at least +<br/>two-thirds of the CIN 1 lesions, half of the CIN 2 lesions, and +<br/>one-third of the CIN 3 lesions will regress back to normal [3]. +<br/>The median ages of patients with these different precursor +<br/>grades are 25, 29, and 34 years, respectively. Ultimately, a +<br/>small proportion will develop into infiltrating cancer, usually +<br/>from the age of 45 years onwards. +<br/>In 1994, the Bethesda system was introduced to simplify +<br/>the WHO system. This system divided all cervical epithelial +<br/>precursor lesions into two groups: the Low-grade Squamous +<br/>Intraepithelial Lesion (LSIL) and High-grade Squamous +<br/>Intraepithelial Lesion (HSIL). The LSIL corresponds to CIN1, +<br/>while the HSIL includes CIN2 and CIN3 [4]. +<br/>Since a period of two to three decades is needed for +<br/>cervical cancer to reach an invasive state, the incidence and +<br/>mortality related to this disease can be significantly reduced +<br/>through early detection and proper treatment. Realizing +<br/>this fact, a variety of screening tests have therefore been +<br/>developed in attempting to be implemented as early cervical +<br/>precancerous screening tools. +<br/>2. Methodology +<br/>This paper reviews 103 journal papers. The papers are +<br/>obtained electronically through 2 major scientific databases: +</td><td>('2905656', 'Yessi Jusman', 'yessi jusman')<br/>('33102280', 'Siew Cheok Ng', 'siew cheok ng')<br/>('2784667', 'Noor Azuan Abu Osman', 'noor azuan abu osman')<br/>('2905656', 'Yessi Jusman', 'yessi jusman')</td><td>Correspondence should be addressed to Siew Cheok Ng; siewcng@um.edu.my and Noor Azuan Abu Osman; azuan@um.edu.my +</td></tr><tr><td>a35dd69d63bac6f3296e0f1d148708cfa4ba80f6</td><td>Audio Visual Emotion Recognition with Temporal Alignment and Perception +<br/>Attention +<br/><b>National Laboratory of Pattern Recognition Institute of Automation, Chinese Academy of Sciences</b><br/><b>Institute of Neuroscience, State Key Laboratory of Neuroscience, CAS Center for Excellence in Brain</b><br/><b>Science and Intelligence Technology, Shanghai Institutes for Biological Sciences, CAS</b></td><td>('1850313', 'Linlin Chao', 'linlin chao')<br/>('37670752', 'Jianhua Tao', 'jianhua tao')<br/>('2740129', 'Minghao Yang', 'minghao yang')<br/>('1704841', 'Ya Li', 'ya li')<br/>('1718662', 'Zhengqi Wen', 'zhengqi wen')</td><td>{linlin.chao, jhtao, mhyang, yli, zqwen}@nlpr.ia.ac.cn +</td></tr><tr><td>a3a34c1b876002e0393038fcf2bcb00821737105</td><td>Face Identification across Different Poses and Illuminations +<br/>with a 3D Morphable Model +<br/>V. Blanz, S. Romdhani, and T. Vetter +<br/><b>University of Freiburg</b><br/>Georges-K¨ohler-Allee 52, 79110 Freiburg, Germany +</td><td></td><td>fvolker, romdhani, vetterg@informatik.uni-freiburg.de +</td></tr><tr><td>a3f1db123ce1818971a57330d82901683d7c2b67</td><td>Poselets and Their Applications in High-Level +<br/>Computer Vision +<br/>Lubomir Bourdev +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2012-52 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2012/EECS-2012-52.html +<br/>May 1, 2012 +</td><td></td><td></td></tr><tr><td>a36c8a4213251d3fd634e8893ad1b932205ad1ca</td><td>Videos from the 2013 Boston Marathon: +<br/>An Event Reconstruction Dataset for +<br/>Synchronization and Localization +<br/>CMU-LTI-018 +<br/><b>Language Technologies Institute</b><br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave., Pittsburgh, PA 15213 +<br/>www.lti.cs.cmu.edu +<br/>© October 1, 2016 +</td><td>('1915796', 'Junwei Liang', 'junwei liang')<br/>('47896638', 'Han Lu', 'han lu')<br/>('2927024', 'Shoou-I Yu', 'shoou-i yu')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td></td></tr><tr><td>a3a97bb5131e7e67316b649bbc2432aaa1a6556e</td><td>Cogn Affect Behav Neurosci +<br/>DOI 10.3758/s13415-013-0170-x +<br/>Role of the hippocampus and orbitofrontal cortex +<br/>during the disambiguation of social cues in working memory +<br/>Chantal E. Stern +<br/><b>Psychonomic Society, Inc</b></td><td>('2973557', 'Karin Schon', 'karin schon')</td><td></td></tr><tr><td>a35d3ba191137224576f312353e1e0267e6699a1</td><td>Increasing security in DRM systems +<br/>through biometric authentication. +<br/>ecuring the exchange +<br/>of intellectual property +<br/>and providing protection +<br/>to multimedia contents in +<br/>distribution systems have enabled the +<br/>advent of digital rights management +<br/>(DRM) systems [5], [14], [21], [47], +<br/>[51], [53]. Rights holders should be able to +<br/>license, monitor, and track the usage of rights +<br/>in a dynamic digital trading environment, espe- +<br/>cially in the near future when universal multimedia +<br/>access (UMA) becomes a reality, and any multimedia +<br/>content will be available anytime, anywhere. In such +<br/>DRM systems, encryption algorithms, access control, +<br/>key management strategies, identification and tracing +<br/>of contents, or copy control will play a prominent role +<br/>to supervise and restrict access to multimedia data, +<br/>avoiding unauthorized or fraudulent operations. +<br/>A key component of any DRM system, also known +<br/>as intellectual property management and protection +<br/>(IPMP) systems in the MPEG-21 framework, is user +<br/>authentication to ensure that +<br/>only those with specific rights are +<br/>able to access the digital informa- +<br/>tion. It is here that biometrics can +<br/>play an essential role, reinforcing securi- +<br/>ty at all stages where customer authentica- +<br/>tion is needed. The ubiquity of users and +<br/>devices, where the same user might want to +<br/>access to multimedia contents from different +<br/>environments (home, car, work, jogging, etc.) and +<br/>also from different devices or media (CD, DVD, +<br/>home computer, laptop, PDA, 2G/3G mobile phones, +<br/>game consoles, etc.) strengthens the need for reliable +<br/>and universal authentication of users. +<br/>Classical user authentication systems have been +<br/>based in something that you have (like a key, an identi- +<br/>fication card, etc.) and/or something that you know +<br/>(like a password, or a PIN). With biometrics, a new +<br/>user authentication paradigm is added: something that +<br/>you are (e.g., fingerprints or face) or something that +<br/>you do or produce (e.g., handwritten signature or +<br/>50 +<br/>IEEE SIGNAL PROCESSING MAGAZINE +<br/>1053-5888/04/$20.00©2004IEEE +<br/>MARCH 2004 +</td><td>('1732220', 'Javier Ortega-Garcia', 'javier ortega-garcia')<br/>('5058247', 'Josef Bigun', 'josef bigun')<br/>('3127386', 'Douglas Reynolds', 'douglas reynolds')<br/>('1775227', 'Joaquin Gonzalez-Rodriguez', 'joaquin gonzalez-rodriguez')</td><td></td></tr><tr><td>a3a2f3803bf403262b56ce88d130af15e984fff0</td><td>Building a Compact Relevant Sample Coverage +<br/>for Relevance Feedback in Content-Based Image +<br/>Retrieval +<br/><b>Tsinghua University, Beijing, China</b><br/>2 Sensing & Control Technology Laboratory, Omron Corporation, Kyoto, Japan +</td><td>('38916673', 'Bangpeng Yao', 'bangpeng yao')<br/>('1679380', 'Haizhou Ai', 'haizhou ai')<br/>('1710195', 'Shihong Lao', 'shihong lao')</td><td></td></tr><tr><td>b56f3a7c50bfcd113d0ba84e6aa41189e262d7ae</td><td>Harvesting Motion Patterns in Still Images from the Internet +<br/><b>ITCS, Institute for Interdisciplinary Information Sciences, Tsinghua University, Beijing</b><br/><b>University of California, San Diego, La Jolla</b></td><td></td><td>Jiajun Wu (jiajunwu.cs@gmail.com) +<br/>Yining Wang (ynwang.yining@gmail.com) +<br/>Zhulin Li (li-zl12@mails.tsinghua.edu.cn) +<br/>Zhuowen Tu (ztu@ucsd.edu) +</td></tr><tr><td>b5968e7bb23f5f03213178c22fd2e47af3afa04c</td><td>Multi-Human Parsing in the Wild +<br/><b>National University of Singapore</b><br/><b>Beijing Jiaotong University</b><br/>March 16, 2018 +</td><td>('2757639', 'Jianshu Li', 'jianshu li')<br/>('2263674', 'Yidong Li', 'yidong li')<br/>('46509407', 'Jian Zhao', 'jian zhao')<br/>('1715286', 'Terence Sim', 'terence sim')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>b5cd9e5d81d14868f1a86ca4f3fab079f63a366d</td><td>Tag-based Video Retrieval by Embedding Semantic Content in a Continuous +<br/>Word Space +<br/><b>University of Southern California</b><br/>Ram Nevatia +<br/>Cees G.M. Snoek +<br/><b>University of Amsterdam</b></td><td>('3407713', 'Arnav Agharwal', 'arnav agharwal')<br/>('3407447', 'Rama Kovvuri', 'rama kovvuri')</td><td>{agharwal,nkovvuri,nevatia}@usc.edu +<br/>cgmsnoek@uva.nl +</td></tr><tr><td>b558be7e182809f5404ea0fcf8a1d1d9498dc01a</td><td>Bottom-up and top-down reasoning with convolutional latent-variable models +<br/>UC Irvine +<br/>UC Irvine +</td><td>('2894848', 'Peiyun Hu', 'peiyun hu')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td>peiyunh@ics.uci.edu +<br/>dramanan@ics.uci.edu +</td></tr><tr><td>b5cd8151f9354ee38b73be1d1457d28e39d3c2c6</td><td>Finding Celebrities in Video +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2006-77 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2006/EECS-2006-77.html +<br/>May 23, 2006 +</td><td>('3317048', 'Nazli Ikizler', 'nazli ikizler')<br/>('1865836', 'Jai Vasanth', 'jai vasanth')<br/>('1744452', 'David Forsyth', 'david forsyth')</td><td></td></tr><tr><td>b5fc4f9ad751c3784eaf740880a1db14843a85ba</td><td>SIViP (2007) 1:225–237 +<br/>DOI 10.1007/s11760-007-0016-5 +<br/>ORIGINAL PAPER +<br/>Significance of image representation for face verification +<br/>Received: 29 August 2006 / Revised: 28 March 2007 / Accepted: 28 March 2007 / Published online: 1 May 2007 +<br/>© Springer-Verlag London Limited 2007 +</td><td>('2627097', 'Anil Kumar Sao', 'anil kumar sao')<br/>('1783087', 'B. V. K. Vijaya Kumar', 'b. v. k. vijaya kumar')</td><td></td></tr><tr><td>b562def2624f59f7d3824e43ecffc990ad780898</td><td></td><td></td><td></td></tr><tr><td>b506aa23949b6d1f0c868ad03aaaeb5e5f7f6b57</td><td><b>UNIVERSITY OF CALIFORNIA</b><br/>RIVERSIDE +<br/>Modeling Social and Temporal Context for Video Analysis +<br/>A Dissertation submitted in partial satisfaction +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>in +<br/>Computer Science +<br/>by +<br/>June 2015 +<br/>Dissertation Committee: +<br/>Dr. Christian R. Shelton, Chairperson +<br/>Dr. Tao Jiang +<br/>Dr. Stefano Lonardi +<br/>Dr. Amit Roy-Chowdhury +</td><td>('12561781', 'Zhen Qin', 'zhen qin')</td><td></td></tr><tr><td>b599f323ee17f12bf251aba928b19a09bfbb13bb</td><td>AUTONOMOUS QUADCOPTER VIDEOGRAPHER +<br/>by +<br/>REY R. COAGUILA +<br/>B.S. Universidad Peruana de Ciencias Aplicadas, 2009 +<br/>A thesis submitted in partial fulfillment of the requirements +<br/>for the degree of Master of Science in Computer Science +<br/>in the Department of Electrical Engineering and Computer Science +<br/><b>in the College of Engineering and Computer Science</b><br/><b>at the University of Central Florida</b><br/>Orlando, Florida +<br/>Spring Term +<br/>2015 +<br/>Major Professor: Gita R. Sukthankar +</td><td></td><td></td></tr><tr><td>b5f2846a506fc417e7da43f6a7679146d99c5e96</td><td>UCF101: A Dataset of 101 Human Actions +<br/>Classes From Videos in The Wild +<br/>CRCV-TR-12-01 +<br/>November 2012 +<br/>Keywords: Action Dataset, UCF101, UCF50, Action Recognition +<br/>Center for Research in Computer Vision +<br/><b>University of Central Florida</b><br/>4000 Central Florida Blvd. +<br/>Orlando, FL 32816-2365 USA +</td><td>('1799979', 'Khurram Soomro', 'khurram soomro')<br/>('40029556', 'Amir Roshan Zamir', 'amir roshan zamir')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td></td></tr><tr><td>b5da4943c348a6b4c934c2ea7330afaf1d655e79</td><td>Facial Landmarks Detection by Self-Iterative Regression based +<br/>Landmarks-Attention Network +<br/><b>University of Chinese Academy of Sciences, Beijing, China</b><br/>2 Microsoft Research Asia, Beijing, China +</td><td>('33325349', 'Tao Hu', 'tao hu')<br/>('3245785', 'Honggang Qi', 'honggang qi')<br/>('1697982', 'Jizheng Xu', 'jizheng xu')<br/>('1689702', 'Qingming Huang', 'qingming huang')</td><td>hutao16@mails.ucas.ac.cn, hgqi@ucas.ac.cn +</td></tr><tr><td>b5402c03a02b059b76be829330d38db8e921e4b5</td><td>Mei, et al, Hybridized KNN and SVM for gene expression data classification +<br/>Hybridized KNN and SVM for gene expression data classification +<br/><b>Zhengzhou University, Zhengzhou, Henan 450052, China</b><br/>Received October 22, 2008 +</td><td>('39156927', 'Zhen Mei', 'zhen mei')<br/>('2380760', 'Qi Shen', 'qi shen')<br/>('35476967', 'Baoxian Ye', 'baoxian ye')</td><td></td></tr><tr><td>b5160e95192340c848370f5092602cad8a4050cd</td><td>IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, TO APPEAR +<br/>Video Classification With CNNs: Using The Codec +<br/>As A Spatio-Temporal Activity Sensor +</td><td>('33998511', 'Aaron Chadha', 'aaron chadha')<br/>('2822935', 'Alhabib Abbas', 'alhabib abbas')<br/>('2747620', 'Yiannis Andreopoulos', 'yiannis andreopoulos')</td><td></td></tr><tr><td>b52c0faba5e1dc578a3c32a7f5cfb6fb87be06ad</td><td>Journal of Applied Research and +<br/>Technology +<br/>ISSN: 1665-6423 +<br/>Centro de Ciencias Aplicadas y +<br/>Desarrollo Tecnológico +<br/>México +<br/> +<br/>Hussain Shah, Jamal; Sharif, Muhammad; Raza, Mudassar; Murtaza, Marryam; Ur-Rehman, Saeed +<br/>Robust Face Recognition Technique under Varying Illumination +<br/>Journal of Applied Research and Technology, vol. 13, núm. 1, febrero, 2015, pp. 97-105 +<br/>Centro de Ciencias Aplicadas y Desarrollo Tecnológico +<br/>Distrito Federal, México +<br/>Available in: http://www.redalyc.org/articulo.oa?id=47436895009 +<br/> How to cite +<br/> Complete issue +<br/> More information about this article +<br/> Journal's homepage in redalyc.org +<br/>Scientific Information System +<br/>Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal +<br/>Non-profit academic project, developed under the open access initiative +</td><td></td><td>jart@aleph.cinstrum.unam.mx +</td></tr><tr><td>b56530be665b0e65933adec4cc5ed05840c37fc4</td><td>IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 2007 +<br/>©IEEE +<br/>Reducing correspondence ambiguity in loosely labeled training data +<br/><b>University of Arizona</b><br/>Tucson Arizona +</td><td>('1728667', 'Kobus Barnard', 'kobus barnard')</td><td>kobus@cs.arizona.edu +</td></tr><tr><td>b5f4e617ac3fc4700ec8129fcd0dcf5f71722923</td><td>Hierarchical Wavelet Networks for Facial Feature Localization +<br/>Rog·erio S. Feris +<br/>Microsoft Research +<br/>Redmond, WA 98052 +<br/>U.S.A. +<br/>Volker Kr¤uger +<br/><b>University of Maryland, CFAR</b><br/><b>College Park, MD</b><br/>U.S.A. +</td><td>('1936061', 'Jim Gemmell', 'jim gemmell')</td><td></td></tr><tr><td>b52886610eda6265a2c1aaf04ce209c047432b6d</td><td>Microexpression Identification and Categorization +<br/>using a Facial Dynamics Map +</td><td>('1684875', 'Feng Xu', 'feng xu')<br/>('2247926', 'Junping Zhang', 'junping zhang')</td><td></td></tr><tr><td>b51b4ef97238940aaa4f43b20a861eaf66f67253</td><td>Hindawi Publishing Corporation +<br/>EURASIP Journal on Image and Video Processing +<br/>Volume 2008, Article ID 184618, 16 pages +<br/>doi:10.1155/2008/184618 +<br/>Research Article +<br/>Unsupervised Modeling of Objects and Their Hierarchical +<br/>Contextual Interactions +<br/><b>Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA</b><br/>Received 11 June 2008; Accepted 2 September 2008 +<br/>Recommended by Simon Lucey +<br/>A successful representation of objects in literature is as a collection of patches, or parts, with a certain appearance and position. +<br/>The relative locations of the different parts of an object are constrained by the geometry of the object. Going beyond a single +<br/>object, consider a collection of images of a particular scene category containing multiple (recurring) objects. The parts belonging +<br/>to different objects are not constrained by such a geometry. However, the objects themselves, arguably due to their semantic +<br/>relationships, demonstrate a pattern in their relative locations. Hence, analyzing the interactions among the parts across the +<br/>collection of images can allow for extraction of the foreground objects, and analyzing the interactions among these objects +<br/>can allow for a semantically meaningful grouping of these objects, which characterizes the entire scene. These groupings are +<br/>typically hierarchical. We introduce hierarchical semantics of objects (hSO) that captures this hierarchical grouping. We propose +<br/>an approach for the unsupervised learning of the hSO from a collection of images of a particular scene. We also demonstrate the +<br/>use of the hSO in providing context for enhanced object localization in the presence of significant occlusions, and show its superior +<br/>performance over a fully connected graphical model for the same task. +<br/>Copyright © 2008 D. Parikh and T. Chen. This is an open access article distributed under the Creative Commons Attribution +<br/>License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly +<br/>cited. +<br/>1. +<br/>INTRODUCTION +<br/>Objects that tend to cooccur in scenes are often semantically +<br/>related. Hence, they demonstrate a characteristic grouping +<br/>behavior according to their relative positions in the scene. +<br/>Some groupings are tighter than others, and thus a hierarchy +<br/>of these groupings among these objects can be observed in a +<br/>collection of images of similar scenes. It is this hierarchy that +<br/>we refer to as the hierarchical semantics of objects (hSO). +<br/>This can be better understood with an example. +<br/>Consider an office scene. Most offices, as seen in Figure 1, +<br/>are likely to have, for instance, a chair, a phone, a monitor, +<br/>and a keyboard. If we analyze a collection of images taken +<br/>from such office settings, we would observe that across +<br/>images, the monitor and keyboard are more or less in the +<br/>same position with respect to each other, and hence can be +<br/>considered to be part of the same super object at a lower level +<br/>in the hSO structure, say a computer. Similarly, the computer +<br/>may usually be somewhere in the vicinity of the phone, and +<br/>so the computer and the phone belong to the same super +<br/>object at a higher level, say the desk area. But the chair and +<br/>the desk area may be placed relatively arbitrarily in the scene +<br/>with respect to each other, more so than any of the other +<br/>objects, and hence belong to a common super object only +<br/>at the highest level in the hierarchy, that is, the scene itself. +<br/>A possible hSO that would describe such an office scene is +<br/>shown in Figure 1. Along with the structure, the hSO may +<br/>also store other information such as the relative position of +<br/>the objects and their cooccurrence counts as parameters. +<br/>The hSO is motivated from an interesting thought +<br/>exercise: at what scale is an object defined? Are the individual +<br/>keys on a keyboard objects, or the entire keyboard, or is +<br/>the entire computer an object? The definition of an object +<br/>is blurry, and the hSO exploits this to allow incorporation +<br/>of semantic information of the scene layout. The leaves of +<br/>the hSO are a collection of parts and represent the objects, +<br/>while the various levels in the hSO represent the super objects +</td><td>('1713589', 'Devi Parikh', 'devi parikh')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td>Correspondence should be addressed to Devi Parikh, dparikh@andrew.cmu.edu +</td></tr><tr><td>b5d7c5aba7b1ededdf61700ca9d8591c65e84e88</td><td>INTERSPEECH 2010 +<br/>Data Pruning for Template-based Automatic Speech Recognition +<br/><b>ESAT, Katholieke Universiteit Leuven, Leuven, Belgium</b></td><td>('1717646', 'Dino Seppi', 'dino seppi')</td><td>dino.seppi@esat.kuleuven.be, dirk.vancompernolle@esat.kuleuven.be +</td></tr><tr><td>b5c749f98710c19b6c41062c60fb605e1ef4312a</td><td>Evaluating Two-Stream CNN for Video Classification +<br/>School of Computer Science, Shanghai Key Lab of Intelligent Information Processing, +<br/><b>Fudan University, Shanghai, China</b></td><td>('1743864', 'Hao Ye', 'hao ye')<br/>('3099139', 'Zuxuan Wu', 'zuxuan wu')<br/>('3066866', 'Rui-Wei Zhao', 'rui-wei zhao')<br/>('31825486', 'Xi Wang', 'xi wang')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')</td><td>{haoye10, zxwu,rwzhao14, xwang10, ygj, xyxue}@fudan.edu.cn +</td></tr><tr><td>b5857b5bd6cb72508a166304f909ddc94afe53e3</td><td>SSIG and IRISA at Multimodal Person Discovery +<br/>1Department of Computer Science, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil +<br/>2IRISA & Inria Rennes , CNRS, Rennes, France +</td><td>('2823797', 'Cassio E. dos Santos', 'cassio e. dos santos')<br/>('1708671', 'Guillaume Gravier', 'guillaume gravier')<br/>('1679142', 'William Robson Schwartz', 'william robson schwartz')</td><td>cass@dcc.ufmg.br, guig@irisa.fr, william@dcc.ufmg.br +</td></tr><tr><td>b59f441234d2d8f1765a20715e227376c7251cd7</td><td></td><td></td><td></td></tr><tr><td>b51e3d59d1bcbc023f39cec233f38510819a2cf9</td><td>CBMM Memo No. 003 +<br/>March 27, 2014 +<br/>Can a biologically-plausible hierarchy effectively +<br/>replace face detection, alignment, and +<br/>recognition pipelines? +<br/>by +</td><td>('1694846', 'Qianli Liao', 'qianli liao')<br/>('2211263', 'Youssef Mroueh', 'youssef mroueh')</td><td></td></tr><tr><td>b54c477885d53a27039c81f028e710ca54c83f11</td><td>1201 +<br/>Semi-Supervised Kernel Mean Shift Clustering +</td><td>('34817359', 'Saket Anand', 'saket anand')<br/>('3323332', 'Sushil Mittal', 'sushil mittal')<br/>('2577513', 'Oncel Tuzel', 'oncel tuzel')<br/>('1729185', 'Peter Meer', 'peter meer')</td><td></td></tr><tr><td>b503f481120e69b62e076dcccf334ee50559451e</td><td>Recognition of Facial Action Units with Action +<br/>Unit Classifiers and An Association Network +<br/>1Department of Electronic and Information Engineering, The Hong Kong Polytechnic +<br/><b>University, Hong Kong</b><br/><b>Chu Hai College of Higher Education, Hong Kong</b></td><td>('2366262', 'JunKai Chen', 'junkai chen')<br/>('1715231', 'Zenghai Chen', 'zenghai chen')<br/>('8590720', 'Zheru Chi', 'zheru chi')<br/>('1965426', 'Hong Fu', 'hong fu')</td><td>Junkai.Chen@connect.polyu.hk, Zenghai.Chen@connect.polyu.hk +<br/>chi.zheru@polyu.edu.hk, hongfu@chuhai.edu.hk +</td></tr><tr><td>b55d0c9a022874fb78653a0004998a66f8242cad</td><td>Hybrid Facial Representations +<br/>for Emotion Recognition +<br/>Automatic facial expression recognition is a widely +<br/>studied problem in computer vision and human-robot +<br/>interaction. There has been a range of studies for +<br/>representing facial descriptors for facial expression +<br/>recognition. Some prominent descriptors were presented +<br/>in the first facial expression recognition and analysis +<br/>challenge (FERA2011). In that competition, the Local +<br/>Gabor Binary Pattern Histogram Sequence descriptor +<br/>showed the most powerful description capability. In this +<br/>paper, we introduce hybrid facial representations for facial +<br/>expression recognition, which have more powerful +<br/>description capability with lower dimensionality. Our +<br/>descriptors consist of a block-based descriptor and a pixel- +<br/>based descriptor. The block-based descriptor represents +<br/>the micro-orientation and micro-geometric structure +<br/>information. The pixel-based descriptor represents texture +<br/>information. We validate our descriptors on two public +<br/>databases, and the results show that our descriptors +<br/>perform well with a relatively low dimensionality. +<br/>Keywords: Facial expression recognition, Histograms of +<br/>Oriented Gradients, HOG, Local Binary Pattern, LBP, +<br/>Rotated Local Binary Pattern, RLBP, Gabor filter, GF. +<br/> +<br/>Manuscript received Mar. 31, 2013; revised Aug. 29, 2013; accepted Sept. 23, 2013. +<br/>This work was supported by the R&D program of the Korea Ministry of Knowledge and +<br/><b>Economy (MKE) and the Korea Evaluation Institute of Industrial Technology (KEIT</b><br/>[10041826, Development of emotional features sensing, diagnostics and distribution s/w +<br/>platform for measurement of multiple intelligence from young children]. +<br/>Jaehong Kim +<br/>Daejeon, Rep. of Korea. +<br/>and +<br/>I. Introduction +<br/>Facial expression is a natural and intuitive means for humans +<br/>to express and sense their emotions and intentions. For this +<br/>reason, automatic facial expression recognition has been an +<br/>active research field in computer vision and human-robot +<br/>interaction for a long time [1], [2]. In the case of robots living +<br/>with a family, it is very useful to sense the family members’ +<br/>emotions through facial expressions and respond appropriately. +<br/>There are three stages in the general automatic facial +<br/>expression recognition systems. The first stage is to detect the +<br/>faces and normalize the photographic images of the faces. This +<br/>stage may be based on a holistic facial region or on facial +<br/>components such as the eyes, nose, and mouth. The next stage +<br/>is to extract the facial expression descriptors from the +<br/>normalized faces. Finally, the system classifies the facial +<br/>descriptors into the proper expression categories. +<br/>In this paper, we introduce new facial expression descriptors. +<br/>These descriptors adopt two representations, a block-based +<br/>representation and a pixel-based representation, to reflect the +<br/>micro-orientation, micro-geometric structure, and texture +<br/>information. The descriptors show more powerful description +<br/>capability with low dimensionality than the state-of-the-art +<br/>descriptors. +<br/>II. Previous Work +<br/>Many researchers have shown a range of approaches to +<br/>construct an automatic facial expression recognition system. +<br/>Geometric approaches and texture-based approaches are the +<br/>types. Texture-based approaches have +<br/>most prominent +<br/>generally shown a better performance +<br/>than geometric +<br/>approaches in previous research [3], [4]. In texture-based +<br/>ETRI Journal, Volume 35, Number 6, December 2013 © 2013 +<br/>http://dx.doi.org/10.4218/etrij.13.2013.0054 +<br/>Woo-han Yun et al. 1021 +</td><td>('36034086', 'DoHyung Kim', 'dohyung kim')</td><td>Woo-han Yun (phone: +82 42 860 5804, yochin@etri.re.kr), DoHyung Kim +<br/>(dhkim008@etri.re.kr), Chankyu Park +<br/>(jhkim504@etri.re.kr) are with the IT Convergence Technology Research Laboratory, ETRI, +<br/>(parkck@etri.re.kr), +</td></tr><tr><td>b5930275813a7e7a1510035a58dd7ba7612943bc</td><td>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 26, 1525-1537 (2010) +<br/>Short Paper__________________________________________________ +<br/>Face Recognition Using L-Fisherfaces* +<br/><b>Institute of Information Science</b><br/><b>Beijing Jiaotong University</b><br/>Beijing, 100044 China +<br/><b>College of Information and Electrical Engineering</b><br/><b>Shandong University of Science and Technology</b><br/>Qingdao, 266510 China +<br/>An appearance-based face recognition approach called the L-Fisherfaces is pro- +<br/>posed in this paper, By using Local Fisher Discriminant Embedding (LFDE), the face +<br/>images are mapped into a face subspace for analysis. Different from Linear Discriminant +<br/>Analysis (LDA), which effectively sees only the Euclidean structure of face space, LFDE +<br/>finds an embedding that preserves local information, and obtains a face subspace that +<br/>best detects the essential face manifold structure. Different from Locality Preserving +<br/>Projections (LPP) and Unsupervised Discriminant projections (UDP), which ignore the +<br/>class label information, LFDE searches for the project axes on which the data points of +<br/>different classes are far from each other while requiring data points of the same class to +<br/>be close to each other. We compare the proposed L-Fisherfaces approach with PCA, +<br/>LDA, LPP, and UDP on three different face databases. Experimental results suggest that +<br/>the proposed L-Fisherfaces provides a better representation and achieves higher accuracy +<br/>in face recognition. +<br/>Keywords: face recognition, local Fisher discriminant embedding, manifold learning, lo- +<br/>cality preserving projections, unsupervised discriminant projections +<br/>1. INTRODUCTION +<br/>Face recognition has aroused wide concerns over the past few decades due to its +<br/>potential applications, such as criminal identification, credit card verification, and secu- +<br/>rity system and scene surveillance. In the literature, various algorithms have been proposed +<br/>for this problem [1, 2]. PCA and LDA are two well-known linear subspace-learning tech- +<br/>niques and have become the most popular methods for face recognition [3-5]. Recently, He +<br/>et al. [6, 7] and Yang et al. [8, 9] proposed two manifold learning based methods, +<br/>namely, Locality Preserving Projections (LPP) and unsupervised discriminant projection +<br/>(UDP), for face recognition. LPP is a linear subspace method derived from Laplacian +<br/>Eigenmap [10]. It results in a linear map that optimally preserves local neighborhood +<br/>information and its objective function is to minimize the local scatter of the projected +<br/>data. Unlike LPP, UDP finds a linear map based on the criterion that seeks to maximize +<br/>Received July 29, 2008; revised October 30, 2008; accepted January 8, 2009. +<br/>Communicated by H. Y. Mark Liao. +<br/>* This work was partially supported by the National Natural Science Foundation of China (NSFC, No. 60672062) +<br/>and the Major State Basic Research Development Program of China (973 Program No. 2004CB318005). +<br/>1525 +</td><td>('7924002', 'Cheng-Yuan Zhang', 'cheng-yuan zhang')<br/>('2383779', 'Qiu-Qi Ruan', 'qiu-qi ruan')</td><td></td></tr><tr><td>b59c8b44a568587bc1b61d130f0ca2f7a2ae3b88</td><td>An Enhanced Intelligent Agent with Image Description +<br/>Generation +<br/>Department of Computer Science and Digital Technologies, Facutly of Engineering and +<br/><b>Environment, Northumbria University, Newcastle, NE1 8ST, United Kingdom</b><br/>learning +<br/>for +<br/>techniques +</td><td>('29695322', 'Ben Fielding', 'ben fielding')<br/>('1921534', 'Philip Kinghorn', 'philip kinghorn')<br/>('2801063', 'Kamlesh Mistry', 'kamlesh mistry')<br/>('1712838', 'Li Zhang', 'li zhang')</td><td>{ben.fielding, philip.kinghorn, kamlesh.mistry, li.zhang (corr. author)}@northumbria.ac.uk +</td></tr><tr><td>b59cee1f647737ec3296ccb3daa25c890359c307</td><td>Continuously Reproducing Toolchains in Pattern +<br/>Recognition and Machine Learning Experiments +<br/>A. Anjos +<br/><b>Idiap Research Institute</b><br/>Martigny, Switzerland +<br/>M. G¨unther +<br/>Vision and Security Technology +<br/><b>University of Colorado</b><br/>Colorado Springs, USA +</td><td></td><td>andre.anjos@idiap.ch +<br/>mgunther@vast.uccs.edu +</td></tr><tr><td>b249f10a30907a80f2a73582f696bc35ba4db9e2</td><td>Improved graph-based SFA: Information preservation +<br/>complements the slowness principle +<br/>Institut f¨ur Neuroinformatik +<br/><b>Ruhr-University Bochum, Germany</b></td><td>('2366497', 'Alberto N. Escalante', 'alberto n. escalante')<br/>('1736245', 'Laurenz Wiskott', 'laurenz wiskott')</td><td></td></tr><tr><td>b2a0e5873c1a8f9a53a199eecae4bdf505816ecb</td><td>Hybrid VAE: Improving Deep Generative Models +<br/>using Partial Observations +<br/>Snap Research +<br/>Microsoft Research +</td><td>('1715440', 'Sergey Tulyakov', 'sergey tulyakov')<br/>('2388416', 'Sebastian Nowozin', 'sebastian nowozin')</td><td>stulyakov@snap.com +<br/>{awf,Sebastian.Nowozin}@microsoft.com +</td></tr><tr><td>b2cd92d930ed9b8d3f9dfcfff733f8384aa93de8</td><td>HyperFace: A Deep Multi-task Learning Framework for Face Detection, +<br/>Landmark Localization, Pose Estimation, and Gender Recognition +<br/><b>University of Maryland</b><br/><b>College Park, MD</b></td><td>('26988560', 'Rajeev Ranjan', 'rajeev ranjan')</td><td>rranjan1@umd.edu +</td></tr><tr><td>b216040f110d2549f61e3f5a7261cab128cab361</td><td>2734 +<br/>IEICE TRANS. INF. & SYST., VOL.E100–D, NO.11 NOVEMBER 2017 +<br/>LETTER +<br/>Weighted Voting of Discriminative Regions for Face Recognition∗ +<br/>SUMMARY +<br/>This paper presents a strategy, Weighted Voting of Dis- +<br/>criminative Regions (WVDR), to improve the face recognition perfor- +<br/>mance, especially in Small Sample Size (SSS) and occlusion situations. +<br/>In WVDR, we extract the discriminative regions according to facial key +<br/>points and abandon the rest parts. Considering different regions of face +<br/>make different contributions to recognition, we assign weights to regions +<br/>for weighted voting. We construct a decision dictionary according to the +<br/>recognition results of selected regions in the training phase, and this dic- +<br/>tionary is used in a self-defined loss function to obtain weights. The final +<br/>identity of test sample is the weighted voting of selected regions. In this +<br/>paper, we combine the WVDR strategy with CRC and SRC separately, and +<br/>extensive experiments show that our method outperforms the baseline and +<br/>some representative algorithms. +<br/>key words: discriminative regions, small sample size, occlusion, weighted +<br/>strategy, face recognition +<br/>1. +<br/>Introduction +<br/>Face recognition is one of the most popular and challenging +<br/>problems in computer vision. Many representative methods, +<br/>such as SRC [1] and CRC [2], have achieved good results in +<br/>the controlled condition. However, face recognition with +<br/>occlusion or small training size is still challenging. +<br/>Wright et al. [1] first apply the Sparse Representation +<br/>based Classification (SRC) for face recognition (FR). Zhang +<br/>et al. [2] propose Collaborative Representation based Clas- +<br/>sification (CRC) and claim that it is the CR instead of the +<br/>l1-norm sparsity that truly improves the FR performance. +<br/>However, the performance of classifiers (e.g. SVM [3], SRC +<br/>and CRC) declines dramatically if the training sample size +<br/>is small. Some works have been done to tackle the Small +<br/>Sample Size (SSS) problem. The Extended SRC [4] algo- +<br/>rithm constructs an auxiliary intra-class variant dictionary +<br/>to represent the variations between training and test images, +<br/>while the construction of the dictionary needs extra data. +<br/>Patch-based methods are another effective way to solve the +<br/>SSS problem. +<br/>In [5], Zhu et al. propose the patch-based +<br/>CRC and multi-scale ensemble. Gao et al. [6] propose the +<br/>Regularized Patch-based Representation to solve the SSS +<br/>problem. However, patch-based methods are sensitive to the +<br/>patch size [7], and haven’t noticed the texture distribution of +<br/>a face image. +<br/>Images with disguise or occlusion are hard to clas- +<br/>sify. The recognition rate of many classifiers (e.g. SVM and +<br/>SRC) decreases rapidly when images occluded. Local Con- +<br/>tourlet Combined Patterns (LCCP) [8] reports a good per- +<br/>formance in non-occlusion images but the recognition rate +<br/>decreases in occlusion condition. There are some improve- +<br/>ments [9], [10] for occlusion problem. The recent prob- +<br/>abilistic collaborative representation (ProCRC) [10] jointly +<br/>maximizes the likelihood of test samples with multiple +<br/>classes. +<br/>Instead of splitting the image into patches of same size, +<br/>we extract the face regions according to an alignment algo- +<br/>rithm [11]. Some regions, such as eyes and nose, are dis- +<br/>In addition, different regions +<br/>criminative for recognition. +<br/>have different representation abilities. As Fig. 1 shows, dis- +<br/>criminative ability of regions is affected by type of region +<br/>and training size. So it’s reasonable that the regions are as- +<br/>signed with different weights. +<br/>In this paper, we propose a method termed Weighted +<br/>Voting of Discriminative Regions (WVDR), in which, dis- +<br/>criminative regions are extracted from face images and +<br/>weights are learned from a decision dictionary in training +<br/>Manuscript received June 5, 2017. +<br/>Manuscript revised July 16, 2017. +<br/>Manuscript publicized August 4, 2017. +<br/>The authors are with Shenzhen Key Lab. of Information Sci +<br/>& Tech, Shenzhen Engineering Lab. of IS & DCP Department of +<br/>Electronic Engineering, Graduate School at Shenzhen, Tsinghua +<br/><b>University, China</b><br/>This work was supported by the Natural Science Foun- +<br/>dation of China (No. 61471216, No. 61771276), +<br/>the Na- +<br/>tional Key Research and Development Program of China +<br/>(No. 2016YFB0101001 and 2017YFC0112500) and the Spe- +<br/>cial Foundation for the Development of Strategic Emerging In- +<br/>dustries of Shenzhen (No. JCYJ20170307153940960 and No. +<br/>JCYJ20150831192224146). +<br/>thor) +<br/>DOI: 10.1587/transinf.2017EDL8124 +<br/>Fig. 1 +<br/>Recognition rates (AR database) when using only a single region. +<br/>The s represents the number of training samples per person. The X-axis +<br/>represents the regions extracted from face, and the image means the whole +<br/>face image. +<br/><b>Copyright c(cid:3) 2017 The Institute of Electronics, Information and Communication Engineers</b></td><td>('3196016', 'Wenming Yang', 'wenming yang')<br/>('2183412', 'Riqiang Gao', 'riqiang gao')<br/>('2883861', 'Qingmin Liao', 'qingmin liao')</td><td>a) E-mail: grq15@mails.tsinghua.edu.cn (Corresponding au- +</td></tr><tr><td>b261439b5cde39ec52d932a222450df085eb5a91</td><td>International Journal of Computer Trends and Technology (IJCTT) – volume 24 Number 2 – June 2015 +<br/>Facial Expression Recognition using Analytical Hierarchy +<br/>Process +<br/><b>MTech Student 1, 2, Disha Institute of</b><br/>Management and Technology, Raipur Chhattisgarh, India1, 2 +<br/>to +<br/>its significant contribution +</td><td></td><td></td></tr><tr><td>b234cd7788a7f7fa410653ad2bafef5de7d5ad29</td><td>Unsupervised Temporal Ensemble Alignment +<br/>For Rapid Annotation +<br/>1 CSIRO, Brisbane, QLD, Australia +<br/><b>Queensland University of Technology, Brisbane, QLD, Australia</b><br/><b>Carnegie Mellon University, Pittsburgh, PA, USA</b></td><td>('3231493', 'Ashton Fagg', 'ashton fagg')<br/>('1729760', 'Sridha Sridharan', 'sridha sridharan')<br/>('1820249', 'Simon Lucey', 'simon lucey')</td><td>ashton@fagg.id.au, s.sridharan@qut.edu.au, slucey@cs.cmu.edu +</td></tr><tr><td>b2c60061ad32e28eb1e20aff42e062c9160786be</td><td>Diverse and Controllable Image Captioning with +<br/>Part-of-Speech Guidance +<br/><b>University of Illinois at Urbana-Champaign</b></td><td>('2118997', 'Aditya Deshpande', 'aditya deshpande')<br/>('29956361', 'Jyoti Aneja', 'jyoti aneja')<br/>('46659761', 'Liwei Wang', 'liwei wang')</td><td>{ardeshp2, janeja2, lwang97, aschwing, daf}@illinois.edu +</td></tr><tr><td>b2b535118c5c4dfcc96f547274cdc05dde629976</td><td>JOURNAL OF IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, VOL. XX, NO. X, XXX 2017 +<br/>Automatic Recognition of Facial Displays of +<br/>Unfelt Emotions +<br/>Escalera, Xavier Bar´o, Sylwia Hyniewska, Member, IEEE, J¨uri Allik, +</td><td>('38370357', 'Kaustubh Kulkarni', 'kaustubh kulkarni')<br/>('22197083', 'Ciprian Adrian Corneanu', 'ciprian adrian corneanu')<br/>('22211769', 'Ikechukwu Ofodile', 'ikechukwu ofodile')<br/>('47608164', 'Gholamreza Anbarjafari', 'gholamreza anbarjafari')</td><td></td></tr><tr><td>b235b4ccd01a204b95f7408bed7a10e080623d2e</td><td>Regularizing Flat Latent Variables with Hierarchical Structures +</td><td>('7246002', 'Rongcheng Lin', 'rongcheng lin')<br/>('2703486', 'Huayu Li', 'huayu li')<br/>('38472218', 'Xiaojun Quan', 'xiaojun quan')<br/>('2248826', 'Richang Hong', 'richang hong')<br/>('2737890', 'Zhiang Wu', 'zhiang wu')<br/>('1874059', 'Yong Ge', 'yong ge')</td><td>(cid:117)UNC Charlotte. Email: {rlin4, hli38, yong.ge}@uncc.edu, +<br/>(cid:63) Hefei University of Technology. Email: hongrc@hfut.edu.cn +<br/>† Institute for Infocomm Research. Email: quanx@i2r.a-star.edu.sg +<br/>∓ Nanjing University of Finance and Economics. Email: zawu@seu.edu.cn +</td></tr><tr><td>b29b42f7ab8d25d244bfc1413a8d608cbdc51855</td><td>EFFECTIVE FACE LANDMARK LOCALIZATION VIA SINGLE DEEP NETWORK +<br/>1National Key Laboratory of Fundamental Science on Synthetic Vision +<br/><b>School of Computer Science, Sichuan University, Chengdu, China</b></td><td>('3471145', 'Zongping Deng', 'zongping deng')<br/>('1691465', 'Ke Li', 'ke li')<br/>('7345195', 'Qijun Zhao', 'qijun zhao')<br/>('40600345', 'Yi Zhang', 'yi zhang')<br/>('1715100', 'Hu Chen', 'hu chen')</td><td>3huchen@scu.edu.cn +</td></tr><tr><td>b2e5df82c55295912194ec73f0dca346f7c113f6</td><td>CUHK&SIAT Submission for THUMOS15 Action Recognition Challenge +<br/><b>The Chinese University of Hong Kong, Hong Kong</b><br/><b>Shenzhen key lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('39060754', 'Limin Wang', 'limin wang')<br/>('40184588', 'Zhe Wang', 'zhe wang')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('40612284', 'Yu Qiao', 'yu qiao')</td><td>07wanglimin@gmail.com, buptwangzhe2012@gmail.com, yjxiong@ie.cuhk.edu.hk, yu.qiao@siat.ac.cn +</td></tr><tr><td>b2e6944bebab8e018f71f802607e6e9164ad3537</td><td>Mixed Error Coding for +<br/>Face Recognition with Mixed Occlusions +<br/><b>Zhejiang University of Technology</b><br/>Hangzhou, China +</td><td>('4487395', 'Ronghua Liang', 'ronghua liang')<br/>('34478462', 'Xiao-Xin Li', 'xiao-xin li')</td><td>{rhliang, mordekai}@zjut.edu.cn +</td></tr><tr><td>b2c25af8a8e191c000f6a55d5f85cf60794c2709</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>A Novel Dimensionality Reduction Technique based on +<br/>Kernel Optimization Through Graph Embedding +<br/>N. Vretos, A. Tefas and I. Pitas +<br/>the date of receipt and acceptance should be inserted later +</td><td></td><td></td></tr><tr><td>b239a756f22201c2780e46754d06a82f108c1d03</td><td>Robust Multimodal Recognition via Multitask +<br/>Multivariate Low-Rank Representations +<br/><b>Center for Automation Research, UMIACS, University of Maryland, College Park, MD 20742 USA</b></td><td>('9033105', 'Heng Zhang', 'heng zhang')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{hzhang98, pvishalm, rama}@umiacs.umd.edu +</td></tr><tr><td>b20cfbb2348984b4e25b6b9174f3c7b65b6aed9e</td><td>Learning with Ambiguous Label Distribution for +<br/>Apparent Age Estimation +<br/>Department of Signal Processing +<br/><b>Tampere University of Technology</b><br/>Tampere 33720, Finland +</td><td>('40394658', 'Ke Chen', 'ke chen')</td><td>firstname.lastname@tut.fi +</td></tr><tr><td>d904f945c1506e7b51b19c99c632ef13f340ef4c</td><td>A scalable 3D HOG model for fast object detection and viewpoint estimation +<br/>KU Leuven, ESAT/PSI - iMinds +<br/>Kasteelpark Arenberg 10 B-3001 Leuven, Belgium +</td><td>('3048367', 'Marco Pedersoli', 'marco pedersoli')<br/>('1704728', 'Tinne Tuytelaars', 'tinne tuytelaars')</td><td>firstname.lastname@esat.kuleuven.be +</td></tr><tr><td>d949fadc9b6c5c8b067fa42265ad30945f9caa99</td><td>Rethinking Feature Discrimination and +<br/>Polymerization for Large-scale Recognition +<br/><b>The Chinese University of Hong Kong</b></td><td>('1715752', 'Yu Liu', 'yu liu')<br/>('46382329', 'Hongyang Li', 'hongyang li')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')</td><td>{yuliu, yangli, xgwang}@ee.cuhk.edu.hk +</td></tr><tr><td>d93baa5ecf3e1196b34494a79df0a1933fd2b4ec</td><td>Precise Temporal Action Localization by +<br/>Evolving Temporal Proposals +<br/><b>East China Normal University</b><br/>Shanghai, China +<br/><b>University of Washington</b><br/>Seattle, WA, USA +<br/>Shanghai Advanced Research +<br/><b>Institute, CAS, China</b><br/><b>East China Normal University</b><br/>Shanghai, China +<br/>Shanghai Advanced Research +<br/><b>Institute, CAS, China</b><br/>Liang He +<br/><b>East China Normal University</b><br/>Shanghai, China +</td><td>('31567595', 'Haonan Qiu', 'haonan qiu')<br/>('1803391', 'Yao Lu', 'yao lu')<br/>('3015119', 'Yingbin Zheng', 'yingbin zheng')<br/>('47939010', 'Feng Wang', 'feng wang')<br/>('1743864', 'Hao Ye', 'hao ye')</td><td>hnqiu@ica.stc.sh.cn +<br/>luyao@cs.washington.edu +<br/>zhengyb@sari.ac.cn +<br/>fwang@cs.ecnu.edu.cn +<br/>yeh@sari.ac.cn +<br/>lhe@cs.ecnu.edu.cn +</td></tr><tr><td>d961617db4e95382ba869a7603006edc4d66ac3b</td><td>Experimenting Motion Relativity for Action Recognition +<br/>with a Large Number of Classes +<br/><b>East China Normal University</b><br/>500 Dongchuan Rd., Shanghai, China +</td><td>('39586279', 'Feng Wang', 'feng wang')<br/>('38755510', 'Xiaoyan Li', 'xiaoyan li')</td><td></td></tr><tr><td>d9810786fccee5f5affaef59bc58d2282718af9b</td><td>Adaptive Frame Selection for +<br/>Enhanced Face Recognition in +<br/>Low-Resolution Videos +<br/>by +<br/>Thesis submitted to the +<br/><b>College of Engineering and Mineral Resources</b><br/><b>at West Virginia University</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>Master of Science +<br/>in +<br/>Electrical Engineering +<br/>Arun Ross, PhD., Chair +<br/>Xin Li, PhD. +<br/>Donald Adjeroh, PhD. +<br/>Lane Department of Computer Science and Electrical Engineering +<br/>Morgantown, West Virginia +<br/>2008 +<br/>Keywords: Face Biometrics, Super-Resolution, Optical Flow, Super-Resolution using +<br/>Optical Flow, Adaptive Frame Selection, Inter-Frame Motion Parameter, Image Quality, +<br/>Image-Level Fusion, Score-Level Fusion +</td><td>('2531952', 'Raghavender Reddy Jillela', 'raghavender reddy jillela')<br/>('2531952', 'Raghavender Reddy Jillela', 'raghavender reddy jillela')</td><td></td></tr><tr><td>d94d7ff6f46ad5cab5c20e6ac14c1de333711a0c</td><td>978-1-5090-4117-6/17/$31.00 ©2017 IEEE +<br/>3031 +<br/>ICASSP 2017 +</td><td></td><td></td></tr><tr><td>d930ec59b87004fd172721f6684963e00137745f</td><td>Face Pose Estimation using a +<br/>Tree of Boosted Classifiers +<br/><b>Signal Processing Institute</b><br/>´Ecole Polytechnique F´ed´erale de Lausanne (EPFL) +<br/>September 11, 2006 +</td><td>('1768663', 'Julien Meynet', 'julien meynet')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')</td><td></td></tr><tr><td>d9739d1b4478b0bf379fe755b3ce5abd8c668f89</td><td></td><td></td><td></td></tr><tr><td>d9c4586269a142faee309973e2ce8cde27bda718</td><td>Contextual Visual Similarity +<br/><b>The Robotics Institute</b><br/><b>Carnegie Mellon University</b></td><td>('2461523', 'Xiaofang Wang', 'xiaofang wang')<br/>('37991449', 'Kris M. Kitani', 'kris m. kitani')<br/>('1709305', 'Martial Hebert', 'martial hebert')</td><td>xiaofan2@andrew.cmu.edu {kkitani,hebert}@cs.cmu.edu +</td></tr><tr><td>d912b8d88d63a2f0cb5d58164e7414bfa6b41dfa</td><td>Facial identification problem: A tracking based approach +<br/>Department of Information Technology +<br/><b>University of Milan</b><br/>via Bramante, 65 - 26013, Crema (CR), Italy +<br/>Telephone: +390373898047, Fax: 0373899010 +<br/>AST Group, ST Microelectronics +<br/>via Olivetti, 5 - 20041, +<br/>Agrate Brianza (MI), Italy +<br/>Telephone: +390396037234 +</td><td>('3330245', 'Marco Anisetti', 'marco anisetti')<br/>('2061298', 'Valerio Bellandi', 'valerio bellandi')<br/>('1746044', 'Ernesto Damiani', 'ernesto damiani')<br/>('2666794', 'Fabrizio Beverina', 'fabrizio beverina')</td><td>Email: {anisetti,bellandi,damiani}@dti.unimi.it +<br/>Email: fabrizio.beverina@st.com +</td></tr><tr><td>d9318c7259e394b3060b424eb6feca0f71219179</td><td>406 +<br/>Face Matching and Retrieval Using Soft Biometrics +</td><td>('2222919', 'Unsang Park', 'unsang park')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>d9a1dd762383213741de4c1c1fd9fccf44e6480d</td><td></td><td></td><td></td></tr><tr><td>d963e640d0bf74120f147329228c3c272764932b</td><td>International Journal of Advanced Science and Technology +<br/>Vol.64 (2014), pp.1-10 +<br/>http://dx.doi.org/10.14257/ijast.2014.64.01 +<br/>Image Processing for Face Recognition Rate Enhancement +<br/><b>School of Computer and Information, Hefei University of Technology, Hefei</b><br/><b>University of Technology, Baghdad, Iraq</b><br/>People’s Republic of China +</td><td></td><td>Israa_ameer@yahoo.com +</td></tr><tr><td>d9ef1a80738bbdd35655c320761f95ee609b8f49</td><td> Volume 5, Issue 4, 2015 ISSN: 2277 128X +<br/>International Journal of Advanced Research in +<br/> Computer Science and Software Engineering +<br/> Research Paper +<br/> Available online at: www.ijarcsse.com +<br/>A Research - Face Recognition by Using Near Set Theory +<br/>Department of Computer Science and Engineering +<br/><b>Abha Gaikwad -Patil College of Engineering, Nagpur, Maharashtra, India</b></td><td>('9231464', 'Bhakti Kurhade', 'bhakti kurhade')</td><td></td></tr><tr><td>d9c4b1ca997583047a8721b7dfd9f0ea2efdc42c</td><td>Learning Inference Models for Computer Vision +</td><td></td><td></td></tr><tr><td>d9bad7c3c874169e3e0b66a031c8199ec0bc2c1f</td><td>It All Matters: +<br/>Reporting Accuracy, Inference Time and Power Consumption +<br/>for Face Emotion Recognition on Embedded Systems +<br/><b>Institute of Telecommunications, TU Wien</b><br/>Movidius an Intel Company +<br/>Dexmont Pe˜na +<br/>Movidius an Intel Company +<br/>Movidius an Intel Company +<br/>ALaRI, Faculty of Informatics, USI +</td><td>('48802034', 'Jelena Milosevic', 'jelena milosevic')<br/>('51129064', 'Andrew Forembsky', 'andrew forembsky')<br/>('9151916', 'David Moloney', 'david moloney')<br/>('1697550', 'Miroslaw Malek', 'miroslaw malek')</td><td>jelena.milosevic@tuwien.ac.at +<br/>andrew.forembsky2@mail.dcu.ie +<br/>dexmont.pena@intel.com +<br/>david.moloney@intel.com +<br/>miroslaw.malek@usi.ch +</td></tr><tr><td>d9327b9621a97244d351b5b93e057f159f24a21e</td><td>SCIENCE CHINA +<br/>Information Sciences +<br/>. RESEARCH PAPERS . +<br/>December 2010 Vol. 53 No. 12: 2415–2428 +<br/>doi: 10.1007/s11432-010-4099-1 +<br/>Laplacian smoothing transform for face recognition +<br/>GU SuiCheng, TAN Ying +<br/>& HE XinGui +<br/>Key Laboratory of Machine Perception (MOE); Department of Machine Intelligence, +<br/><b>School of Electronics Engineering and Computer Science; Peking University, Beijing 100871, China</b><br/>Received March 16, 2009; accepted April 1, 2010 +</td><td></td><td></td></tr><tr><td>d915e634aec40d7ee00cbea96d735d3e69602f1a</td><td>Two-Stream convolutional nets for action recognition in untrimmed video +<br/><b>Stanford University</b><br/><b>Stanford University</b></td><td>('3308619', 'Kenneth Jung', 'kenneth jung')<br/>('5590869', 'Song Han', 'song han')</td><td>kjung@stanford.edu +<br/>songhan@stanford.edu +</td></tr><tr><td>aca232de87c4c61537c730ee59a8f7ebf5ecb14f</td><td>EBGM VS SUBSPACE PROJECTION FOR FACE RECOGNITION +<br/>19.5 Km Markopoulou Avenue, P.O. Box 68, Peania, Athens, Greece +<br/>Athens Information Technology +<br/>Keywords: +<br/>Human-Machine Interfaces, Computer Vision, Face Recognition. +</td><td>('40089976', 'Andreas Stergiou', 'andreas stergiou')<br/>('1702943', 'Aristodemos Pnevmatikakis', 'aristodemos pnevmatikakis')<br/>('1725498', 'Lazaros Polymenakos', 'lazaros polymenakos')</td><td></td></tr><tr><td>ac1d97a465b7cc56204af5f2df0d54f819eef8a6</td><td>A Look at Eye Detection for Unconstrained +<br/>Environments +<br/>Key words: Unconstrained Face Recognition, Eye Detection, Machine Learning, +<br/>Correlation Filters, Photo-head Testing Protocol +<br/>1 Introduction +<br/>Eye detection is a necessary processing step for many face recognition algorithms. +<br/>For some of these algorithms, the eye coordinates are required for proper geomet- +<br/>ric normalization before recognition. For others, the eyes serve as reference points +<br/>to locate other significant features on the face, such as the nose and mouth. The +<br/>eyes, containing significant discriminative information, can even be used by them- +<br/>selves as features for recognition. Eye detection is a well studied problem for the +<br/>constrained face recognition problem, where we find controlled distances, lighting, +<br/>and limited pose variation. A far more difficult scenario for eye detection is the un- +<br/>constrained face recognition problem, where we do not have any control over the +<br/>environment or the subject. In this chapter, we will take a look at eye detection for +<br/>the latter, which encompasses problems of flexible authentication, surveillance, and +<br/>intelligence collection. +<br/>A multitude of problems affect the acquisition of face imagery in unconstrained +<br/>environments, with major problems related to lighting, distance, motion and pose. +<br/>Existing work on lighting [14, 7] has focused on algorithmic issues (specifically, +<br/>normalization), and not the direct impact of acquisition. Under difficult acquisition +<br/><b>Vision and Security Technology Lab, University of Colorado at Colorado Springs, Colorado</b><br/>Anderson Rocha +<br/><b>Institute of Computing, University of Campinas (Unicamp), Campinas, Brazil, e-mail: ander</b></td><td>('2613438', 'Walter J. Scheirer', 'walter j. scheirer')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')<br/>('2613438', 'Walter J. Scheirer', 'walter j. scheirer')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')</td><td>USA, e-mail: lastname@uccs.edu +<br/>son.rocha@ic.unicamp.br +</td></tr><tr><td>ac2e44622efbbab525d4301c83cb4d5d7f6f0e55</td><td>A 3D Morphable Model learnt from 10,000 faces +<br/><b>Imperial College London, UK</b><br/>†Great Ormond Street Hospital, UK +<br/><b>Center for Machine Vision and Signal Analysis (CMVS), University of Oulu, Finland</b></td><td>('1848903', 'James Booth', 'james booth')<br/>('2931390', 'Anastasios Roussos', 'anastasios roussos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('5137183', 'Allan Ponniah', 'allan ponniah')<br/>('2421231', 'David Dunaway', 'david dunaway')</td><td>⋆{james.booth,troussos,s.zafeiriou}@imperial.ac.uk, †{allan.ponniah,david.dunaway}@gosh.nhs.uk +</td></tr><tr><td>ac6c3b3e92ff5fbcd8f7967696c7aae134bea209</td><td>Deep Cascaded Bi-Network for +<br/>Face Hallucination(cid:63) +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b><br/><b>University of California, Merced</b></td><td>('2226254', 'Shizhan Zhu', 'shizhan zhu')<br/>('2391885', 'Sifei Liu', 'sifei liu')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>ac855f0de9086e9e170072cb37400637f0c9b735</td><td>Fast Geometrically-Perturbed Adversarial Faces +<br/><b>West Virginia University</b></td><td>('35477977', 'Ali Dabouei', 'ali dabouei')<br/>('30319988', 'Sobhan Soleymani', 'sobhan soleymani')<br/>('8147588', 'Nasser M. Nasrabadi', 'nasser m. nasrabadi')</td><td>{ad0046, ssoleyma}@mix.wvu.edu, {jeremy.dawson, nasser.nasrabadi}@mail.wvu.edu +</td></tr><tr><td>ac21c8aceea6b9495574f8f9d916e571e2fc497f</td><td>Pose-Independent Identity-based Facial Image +<br/>Retrieval using Contextual Similarity +<br/><b>King Abdullah University of Science and Technology 4700, Thuwal, Saudi Arabia</b></td><td>('3036634', 'Islam Almasri', 'islam almasri')</td><td></td></tr><tr><td>ac6a9f80d850b544a2cbfdde7002ad5e25c05ac6</td><td>779 +<br/>Privacy-Protected Facial Biometric Verification +<br/>Using Fuzzy Forest Learning +</td><td>('1690116', 'Ahmed Bouridane', 'ahmed bouridane')<br/>('1691478', 'Danny Crookes', 'danny crookes')<br/>('1739563', 'M. Emre Celebi', 'm. emre celebi')<br/>('39486168', 'Hua-Liang Wei', 'hua-liang wei')</td><td></td></tr><tr><td>aca273a9350b10b6e2ef84f0e3a327255207d0f5</td><td></td><td></td><td></td></tr><tr><td>aca75c032cfb0b2eb4c0ae56f3d060d8875e43f9</td><td>Co-Regularized Ensemble for Feature Selection +<br/><b>School of Computer Science and Technology, Tianjin University, China</b><br/><b>School of Information Technology and Electrical Engineering, The University of Queensland</b><br/>3Tianjin Key Laboratory of Cognitive Computing and Application +</td><td>('2302512', 'Yahong Han', 'yahong han')<br/>('1698559', 'Yi Yang', 'yi yang')<br/>('1720932', 'Xiaofang Zhou', 'xiaofang zhou')</td><td>yahong@tju.edu.cn, yee.i.yang@gmail.com, zxf@itee.uq.edu.au +</td></tr><tr><td>accbd6cd5dd649137a7c57ad6ef99232759f7544</td><td>FACIAL EXPRESSION RECOGNITION WITH LOCAL BINARY PATTERNS +<br/>AND LINEAR PROGRAMMING +<br/>1 Machine Vision Group, Infotech Oulu and Dept. of Electrical and Information Engineering +<br/><b>P. O. Box 4500 Fin-90014 University of Oulu, Finland</b><br/><b>College of Electronics and Information, Northwestern Polytechnic University</b><br/>710072 Xi’an, China +<br/>In this work, we propose a novel approach to recognize facial expressions from static +<br/>images. First, the Local Binary Patterns (LBP) are used to efficiently represent the facial +<br/>images and then the Linear Programming (LP) technique is adopted to classify the seven +<br/>facial expressions anger, disgust, fear, happiness, sadness, surprise and neutral. +<br/>Experimental results demonstrate an average recognition accuracy of 93.8% on the JAFFE +<br/>database, which outperforms the rates of all other reported methods on the same database. +<br/>Introduction +<br/>Facial expression recognition from static +<br/>images is a more challenging problem +<br/>than from image sequences because less +<br/>information for expression actions +<br/>is +<br/>available. However, information in a +<br/>single image is sometimes enough for +<br/>expression recognition, and +<br/>in many +<br/>applications it is also useful to recognize +<br/>single image’s facial expression. +<br/>In the recent years, numerous approaches +<br/>to facial expression analysis from static +<br/>images have been proposed [1] [2]. These +<br/>methods +<br/>face +<br/>representation and similarity measure. +<br/>For instance, Zhang [3] used two types of +<br/>features: the geometric position of 34 +<br/>manually selected fiducial points and a +<br/>set of Gabor wavelet coefficients at these +<br/>points. These two types of features were +<br/>used both independently and jointly with +<br/>a multi-layer perceptron for classification. +<br/>Guo and Dyer [4] also adopted a similar +<br/>face representation, combined with linear +<br/>to carry out +<br/>programming +<br/>selection +<br/>simultaneous +<br/>and +<br/>classifier +<br/>they reported +<br/>technique +<br/>feature +<br/>training, and +<br/>differ +<br/>generally +<br/>in +<br/>a +<br/>simple +<br/>imperative question +<br/>better result. Lyons et al. used a similar face +<br/>representation with +<br/>LDA-based +<br/>classification scheme [5]. All the above methods +<br/>required the manual selection of fiducial points. +<br/>Buciu et al. used ICA and Gabor representation for +<br/>facial expression recognition and reported good result +<br/>on the same database [6]. However, a suitable +<br/>combination of feature extraction and classification is +<br/>still one +<br/>for expression +<br/>recognition. +<br/>In this paper, we propose a novel method for facial +<br/>expression recognition. In the feature extraction step, +<br/>the Local Binary Pattern (LBP) operator is used to +<br/>describe facial expressions. In the classification step, +<br/>seven expressions (anger, disgust, fear, happiness, +<br/>sadness, surprise and neutral) are decomposed into 21 +<br/>expression pairs such as anger-fear, happiness- +<br/>sadness etc. 21 classifiers are produced by the Linear +<br/>Programming (LP) technique, each corresponding to +<br/>one of the 21 expression pairs. A simple binary tree +<br/>tournament scheme with pairwise comparisons is +<br/>Face Representation with Local Binary Patterns +<br/> +<br/>Fig.1 shows the basic LBP operator [7], in which the +<br/>original 3×3 neighbourhood at the left is thresholded +<br/>by the value of the centre pixel, and a binary pattern +</td><td>('4729239', 'Xiaoyi Feng', 'xiaoyi feng')<br/>('1714724', 'Matti Pietikäinen', 'matti pietikäinen')<br/>('1751372', 'Abdenour Hadid', 'abdenour hadid')</td><td>{xiaoyi,mkp,hadid}@ee.oulu.fi +<br/>fengxiao@nwpu.edu.cn +</td></tr><tr><td>ac51d9ddbd462d023ec60818bac6cdae83b66992</td><td>Hindawi Publishing Corporation +<br/>Computational Intelligence and Neuroscience +<br/>Volume 2015, Article ID 709072, 10 pages +<br/>http://dx.doi.org/10.1155/2015/709072 +<br/>Research Article +<br/>An Efficient Robust Eye Localization by Learning +<br/>the Convolution Distribution Using Eye Template +<br/>1Science and Technology on Parallel and Distributed Processing Laboratory, School of Computer, +<br/><b>National University of Defense Technology, Changsha 410073, China</b><br/><b>Informatization Office, National University of Defense Technology, Changsha 410073, China</b><br/>Received 30 January 2015; Accepted 14 April 2015 +<br/>Academic Editor: Ye-Sho Chen +<br/>permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Eye localization is a fundamental process in many facial analyses. In practical use, it is often challenged by illumination, head pose, +<br/>facial expression, occlusion, and other factors. It remains great difficulty to achieve high accuracy with short prediction time and +<br/>low training cost at the same time. This paper presents a novel eye localization approach which explores only one-layer convolution +<br/>map by eye template using a BP network. Results showed that the proposed method is robust to handle many difficult situations. In +<br/>experiments, accuracy of 98% and 96%, respectively, on the BioID and LFPW test sets could be achieved in 10 fps prediction rate +<br/>with only 15-minute training cost. In comparison with other robust models, the proposed method could obtain similar best results +<br/>with greatly reduced training time and high prediction speed. +<br/>1. Introduction +<br/>Eye localization is essential to many face analyses. In analysis +<br/>of the human sentiment, eye focus, and head pose, the loca- +<br/>tion of the eye is indispensable to extract the corresponding +<br/>information there [1]. In face tracing, eye localization is often +<br/>required in real time. In face recognition, many algorithms +<br/>ask for the alignment of the face images based on eye location +<br/>[2]. Inaccurate location may result in the failure of the +<br/>recognition [3, 4]. +<br/>However, real-world eye localization is filled with chal- +<br/>lenges. Face pictures are commonly taken by a projection +<br/>from the 3D space to the 2D plane. Appearance of the face +<br/>image could be influenced by the head pose, facial expression, +<br/>and illumination. Texture around eyes is therefore full of +<br/>change. Moreover, eyes may be occluded by stuffs like glasses +<br/>and hair, as shown in Figure 1. To work in any unexpected +<br/>cases, the algorithm should be robust to those impacts. +<br/>In the design of the eye localization algorithm in practical +<br/>use, prediction accuracy, rate, and the training cost are the +<br/>most concerned factors. A robust algorithm should keep high +<br/>prediction accuracy for varying cases with diverse face poses, +<br/>facial expressions in complex environment with occlusion, +<br/>and illumination changes. For real time applications, high +<br/>prediction rate is required. For some online learning systems +<br/>like the one used for public security, short training time is +<br/>also in demand to quickly adapt the algorithm to different +<br/>working places. Low training cost is also of benefit for the +<br/>tuning of the algorithm. To improve the accuracy in the diffi- +<br/>cult environment, complex model is often applied. However, +<br/>the over complicated model will increase the training cost +<br/>and the prediction time. How to select an approach with +<br/>enough complexity to achieve high prediction accuracy, high +<br/>prediction rate, and low training cost at the same time is still +<br/>a challenge. +<br/>Eye localization approaches could be mainly divided into +<br/>the texture based and the structure based. Texture based +<br/>methods [5–8] learn the features from the image textures. For +<br/>the methods exploring local textures [5, 6], high prediction +<br/>rate could be achieved with simple training. However, they +<br/>are usually not robust to the situation with occlusion and +<br/>distortion due to the limited information from the local area. +<br/>On the other hand, methods like [7, 8] study the global texture +<br/>feature from entire face image by convolution networks. High +</td><td>('1790480', 'Xuan Li', 'xuan li')<br/>('1791001', 'Yong Dou', 'yong dou')<br/>('2223570', 'Xin Niu', 'xin niu')<br/>('2512580', 'Jiaqing Xu', 'jiaqing xu')<br/>('2672701', 'Ruorong Xiao', 'ruorong xiao')<br/>('1790480', 'Xuan Li', 'xuan li')</td><td>Correspondence should be addressed to Xuan Li; lixuan@nudt.edu.cn +</td></tr><tr><td>acc548285f362e6b08c2b876b628efceceeb813e</td><td>Hindawi Publishing Corporation +<br/>Computational and Mathematical Methods in Medicine +<br/>Volume 2014, Article ID 427826, 12 pages +<br/>http://dx.doi.org/10.1155/2014/427826 +<br/>Research Article +<br/>Objectifying Facial Expressivity Assessment of Parkinson’s +<br/>Patients: Preliminary Study +<br/><b>Vrije Universiteit Brussel, 1050 Brussels, Belgium</b><br/><b>Shaanxi Provincial Key Lab on Speech and Image Information Processing, Northwestern Polytechnical University, Xi an, China</b><br/><b>Vrije Universiteit Brussel, 1050 Brussels, Belgium</b><br/><b>Vrije Universiteit Brussel, 1050 Brussels, Belgium</b><br/>Received 9 June 2014; Accepted 22 September 2014; Published 13 November 2014 +<br/>Academic Editor: Justin Dauwels +<br/>permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Patients with Parkinson’s disease (PD) can exhibit a reduction of spontaneous facial expression, designated as “facial masking,” a +<br/>symptom in which facial muscles become rigid. To improve clinical assessment of facial expressivity of PD, this work attempts +<br/>to quantify the dynamic facial expressivity (facial activity) of PD by automatically recognizing facial action units (AUs) and +<br/>estimating their intensity. Spontaneous facial expressivity was assessed by comparing 7 PD patients with 8 control participants. To +<br/>voluntarily produce spontaneous facial expressions that resemble those typically triggered by emotions, six emotions (amusement, +<br/>sadness, anger, disgust, surprise, and fear) were elicited using movie clips. During the movie clips, physiological signals (facial +<br/>electromyography (EMG) and electrocardiogram (ECG)) and frontal face video of the participants were recorded. The participants +<br/>were asked to report on their emotional states throughout the experiment. We first examined the effectiveness of the emotion +<br/>manipulation by evaluating the participant’s self-reports. Disgust-induced emotions were significantly higher than the other +<br/>emotions. Thus we focused on the analysis of the recorded data during watching disgust movie clips. The proposed facial expressivity +<br/>assessment approach captured differences in facial expressivity between PD patients and controls. Also differences between PD +<br/>patients with different progression of Parkinson’s disease have been observed. +<br/>1. Introduction +<br/>One of the manifestations of Parkinson’s disease (PD) is the +<br/>gradual loss of facial mobility and “mask-like” appearance. +<br/>Katsikitis and Pilowsky (1988) [1] stated that PD patients +<br/>were rated as significantly less expressive than an aphasic +<br/>and control group, on a task designed to assess spontaneous +<br/>facial expression. In addition, the spontaneous smiles of PD +<br/>patients are often perceived to be “unfelt,” because of the lack +<br/>of accompanying cheek raises [2]. Jacobs et al. [3] confirmed +<br/>that PD patients show reduced intensity of emotional facial +<br/>expression compared to the controls. In order to assess facial +<br/>expressivity, most research relies on subjective coding of the +<br/>implied researchers, as in aforementioned studies. Tickle- +<br/>Degnen and Lyons [4] found that decreased facial expressivity +<br/>correlated with self-reports of PD patients as well as the +<br/>Unified Parkinson’s Disease Rating Scale (UPDRS) [5]. PD +<br/>patients, who rated their ability to facially express emotions +<br/>as severely affected, did demonstrate less facial expressivity. +<br/>In this paper, we investigate automatic measurements +<br/>of facial expressivity from video recorded PD patients and +<br/>control populations. To the best of our knowledge, in actual +<br/>research, few attempts have been made for designing a +<br/>computer-based quantitative analysis of facial expressivity of +<br/>PD patient. To analyze whether Parkinson’s disease affected +<br/>voluntary expression of facial emotions, Bowers et al. [6] +<br/>videotaped PD patients and healthy control participants +<br/>while they made voluntary facial expression (happy, sad, fear, +<br/>anger, disgust, and surprise). In their approach, the amount of +<br/>facial movements change and timing have been quantified by +</td><td>('40432410', 'Peng Wu', 'peng wu')<br/>('34068333', 'Isabel Gonzalez', 'isabel gonzalez')<br/>('3348420', 'Dongmei Jiang', 'dongmei jiang')<br/>('1970907', 'Hichem Sahli', 'hichem sahli')<br/>('3041213', 'Eric Kerckhofs', 'eric kerckhofs')<br/>('2540163', 'Marie Vandekerckhove', 'marie vandekerckhove')<br/>('40432410', 'Peng Wu', 'peng wu')</td><td>Correspondence should be addressed to Peng Wu; pwu@etro.vub.ac.be +</td></tr><tr><td>acee2201f8a15990551804dd382b86973eb7c0a8</td><td>To Boost or Not to Boost? On the Limits of +<br/>Boosted Trees for Object Detection +<br/><b>Computer Vision and Robotics Research Laboratory</b><br/><b>University of California San Diego</b></td><td>('1802326', 'Eshed Ohn-Bar', 'eshed ohn-bar')</td><td>{eohnbar, mtrivedi}@ucsd.edu +</td></tr><tr><td>ac0d3f6ed5c42b7fc6d7c9e1a9bb80392742ad5e</td><td></td><td></td><td></td></tr><tr><td>ac820d67b313c38b9add05abef8891426edd5afb</td><td></td><td></td><td></td></tr><tr><td>ac9a331327cceda4e23f9873f387c9fd161fad76</td><td>Deep Convolutional Neural Network for Age Estimation based on +<br/>VGG-Face Model +<br/><b>University of Bridgeport</b><br/><b>University of Bridgeport</b><br/>Technology Building, Bridgeport CT 06604 USA +</td><td>('7404315', 'Zakariya Qawaqneh', 'zakariya qawaqneh')<br/>('34792425', 'Arafat Abu Mallouh', 'arafat abu mallouh')<br/>('2791535', 'Buket D. Barkana', 'buket d. barkana')</td><td>Emails: {zqawaqneh; aabumall@my.bridgeport.edu}, bbarkana@bridgeport.edu +</td></tr><tr><td>ac26166857e55fd5c64ae7194a169ff4e473eb8b</td><td>Personalized Age Progression with Bi-level +<br/>Aging Dictionary Learning +</td><td>('2287686', 'Xiangbo Shu', 'xiangbo shu')<br/>('8053308', 'Jinhui Tang', 'jinhui tang')<br/>('3233021', 'Zechao Li', 'zechao li')<br/>('2356867', 'Hanjiang Lai', 'hanjiang lai')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>ac559873b288f3ac28ee8a38c0f3710ea3f986d9</td><td>Team DEEP-HRI Moments in Time Challenge 2018 Technical Report +<br/><b>Hikvision Research Institute</b></td><td>('39816387', 'Chao Li', 'chao li')<br/>('48375401', 'Zhi Hou', 'zhi hou')<br/>('35843399', 'Jiaxu Chen', 'jiaxu chen')<br/>('9162532', 'Jiqiang Zhou', 'jiqiang zhou')<br/>('50322310', 'Di Xie', 'di xie')<br/>('3290437', 'Shiliang Pu', 'shiliang pu')</td><td></td></tr><tr><td>ac8e09128e1e48a2eae5fa90f252ada689f6eae7</td><td>Leolani: a reference machine with a theory of +<br/>mind for social communication +<br/><b>VU University Amsterdam, Computational Lexicology and Terminology Lab, De</b><br/>Boelelaan 1105, 1081HV Amsterdam, The Netherlands +<br/>www.cltl.nl +</td><td>('50998926', 'Bram Kraaijeveld', 'bram kraaijeveld')</td><td>{p.t.j.m.vossen,s.baezsantamaria,l.bajcetic,b.kraaijeveld}@vu.nl +</td></tr><tr><td>ac8441e30833a8e2a96a57c5e6fede5df81794af</td><td>IEEE TRANSACTIONS ON IMAGE PROCESSING +<br/>Hierarchical Representation Learning for Kinship +<br/>Verification +</td><td>('1952698', 'Naman Kohli', 'naman kohli')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2487227', 'Afzel Noore', 'afzel noore')<br/>('2641605', 'Angshul Majumdar', 'angshul majumdar')</td><td></td></tr><tr><td>ac86ccc16d555484a91741e4cb578b75599147b2</td><td>Morphable Face Models - An Open Framework +<br/><b>Gravis Research Group, University of Basel</b></td><td>('3277377', 'Thomas Gerig', 'thomas gerig')<br/>('39550224', 'Clemens Blumer', 'clemens blumer')<br/>('34460642', 'Bernhard Egger', 'bernhard egger')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td></td></tr><tr><td>ac12ba5bf81de83991210b4cd95b4ad048317681</td><td>Combining Deep Facial and Ambient Features +<br/>for First Impression Estimation +<br/><b>Program of Computational Science and Engineering, Bo gazi ci University</b><br/>Bebek, Istanbul, Turkey +<br/><b>Nam k Kemal University</b><br/>C¸ orlu, Tekirda˘g, Turkey +<br/><b>Bo gazi ci University</b><br/>Bebek, Istanbul, Turkey +</td><td>('38007788', 'Heysem Kaya', 'heysem kaya')<br/>('1764521', 'Albert Ali Salah', 'albert ali salah')</td><td>furkan.gurpinar@boun.edu.tr +<br/>hkaya@nku.edu.tr +<br/>salah@boun.edu.tr +</td></tr><tr><td>ac75c662568cbb7308400cc002469a14ff25edfd</td><td>REGULARIZATION STUDIES ON LDA FOR FACE RECOGNITION +<br/>Bell Canada Multimedia Laboratory, The Edward S. Rogers Sr. Department of +<br/><b>Electrical and Computer Engineering, University of Toronto, M5S 3G4, Canada</b></td><td>('1681365', 'Juwei Lu', 'juwei lu')</td><td></td></tr><tr><td>ac9dfbeb58d591b5aea13d13a83b1e23e7ef1fea</td><td>From Gabor Magnitude to Gabor Phase Features: +<br/>Tackling the Problem of Face Recognition under Severe Illumination Changes +<br/>215 +<br/>12 +<br/>X +<br/>From Gabor Magnitude to Gabor Phase +<br/>Features: Tackling the Problem of Face +<br/>Recognition under Severe Illumination Changes +<br/><b>Faculty of Electrical Engineering, University of Ljubljana</b><br/>Slovenia +<br/>1. Introduction +<br/>Among the numerous biometric systems presented in the literature, face recognition +<br/>systems have received a great deal of attention in recent years. The main driving force in the +<br/>development of these systems can be found in the enormous potential face recognition +<br/>technology has in various application domains ranging from access control, human-machine +<br/>interaction and entertainment to homeland security and surveillance (Štruc et al., 2008a). +<br/>While contemporary face recognition techniques have made quite a leap in terms of +<br/>performance over the last two decades, they still struggle with their performance when +<br/>deployed in unconstrained and uncontrolled environments (Gross et al., 2004; Phillips et al., +<br/>2007). In such environments the external conditions present during the image acquisition +<br/>stage heavily influence the appearance of a face in the acquired image and consequently +<br/>affect the performance of the recognition system. It is said that face recognition techniques +<br/>suffer from the so-called PIE problem, which refers to the problem of handling Pose, +<br/>Illumination and Expression variations that are typically encountered in real-life operating +<br/>conditions. In fact, it was emphasized by numerous researchers that the appearance of the +<br/>same face can vary significantly from image to image due to changes of the PIE factors and +<br/>that the variability in the images induced by the these factors can easily surpass the +<br/>variability induced by the subjects’ identity (Gross et al., 2004; Short et al., 2005). To cope +<br/>with image variability induced by the PIE factors, face recognition systems have to utilize +<br/>feature extraction techniques capable of extracting stable and discriminative features from +<br/>facial images regardless of the conditions governing the acquisition procedure. We will +<br/>confine ourselves in this chapter to tackling the problem of illumination changes, as it +<br/>represents the PIE factor which, in our opinion, is the hardest to control when deploying a +<br/>face recognition system, e.g., in access control applications. +<br/>Many feature extraction techniques, among them particularly the appearance based +<br/>methods, have difficulties extracting stable features from images captured under varying +<br/>illumination conditions and, hence, perform poorly when deployed in unconstrained +<br/>environments. Researchers have, therefore, proposed a number of alternatives that should +<br/>compensate for the illumination changes and thus ensure stable face recognition +<br/>performance. +</td><td>('2011218', 'Vitomir Štruc', 'vitomir štruc')<br/>('1753753', 'Nikola Pavešić', 'nikola pavešić')</td><td></td></tr><tr><td>acb83d68345fe9a6eb9840c6e1ff0e41fa373229</td><td>Kernel Methods in Computer Vision: +<br/>Object Localization, Clustering, +<br/>and Taxonomy Discovery +<br/>vorgelegt von +<br/>Matthew Brian Blaschko, M.S. +<br/>aus La Jolla +<br/>Von der Fakult¨at IV - Elektrotechnik und Informatik +<br/>der Technischen Universit¨at Berlin +<br/>zur Erlangung des akademischen Grades +<br/>Doktor der Naturwissenschaften +<br/>Dr. rer. nat. +<br/>genehmigte Dissertation +<br/>Promotionsausschuß: +<br/>Vorsitzender: Prof. Dr. O. Hellwich +<br/>Berichter: Prof. Dr. T. Hofmann +<br/>Berichter: Prof. Dr. K.-R. M¨uller +<br/>Berichter: Prof. Dr. B. Sch¨olkopf +<br/>Tag der wissenschaftlichen Aussprache: 23.03.2009 +<br/>Berlin 2009 +<br/>D83 +</td><td></td><td></td></tr><tr><td>ade1034d5daec9e3eba1d39ae3f33ebbe3e8e9a7</td><td>Multimodal Caricatural Mirror +<br/>(1) : Université catholique de Louvain, Belgium +<br/>(2) Universitat Polytecnica de Barcelona, Spain +<br/>(3) Universidad Polytècnica de Madrid, Spain +<br/><b>Aristotle University of Thessaloniki, Greece</b><br/><b>Bogazici University, Turkey</b><br/>(6) Faculté Polytechnique de Mons, Belgium +</td><td></td><td></td></tr><tr><td>ad8540379884ec03327076b562b63bc47e64a2c7</td><td>Int. J. Bio-Inspired Computation, Vol. 5, No. 3, 2013 +<br/>175 +<br/>Bee royalty offspring algorithm for improvement of +<br/>facial expressions classification model +<br/>Department of Computer Science, +<br/>Mahshahr Branch, +<br/><b>Islamic Azad University</b><br/>Mahshahr, Iran +<br/>*Corresponding author +<br/>Md Jan Nordin +<br/>Centre for Artificial Intelligence Technology, +<br/>Universiti Kebangsaan Malaysia, +<br/>Bangi, Selangor, Malaysia +</td><td>('1880066', 'Amir Jamshidnezhad', 'amir jamshidnezhad')</td><td>E-mail: a.jamshidnejad@yahoo.com +<br/>E-mail: jan@ftsm.ukm.my +</td></tr><tr><td>adce9902dca7f4e8a9b9cf6686ec6a7c0f2a0ba6</td><td>Two Birds, One Stone: Jointly Learning Binary Code for +<br/>Large-scale Face Image Retrieval and Attributes Prediction +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b><br/><b>School of Information Science and Technology, ShanghaiTech University, Shanghai, 200031, China</b></td><td>('38751558', 'Yan Li', 'yan li')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('3035576', 'Haomiao Liu', 'haomiao liu')<br/>('3371529', 'Huajie Jiang', 'huajie jiang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{yan.li, haomiao.liu, huajie.jiang}@vipl.ict.ac.cn, {wangruiping, sgshan, xlchen}@ict.ac.cn +</td></tr><tr><td>adf7ccb81b8515a2d05fd3b4c7ce5adf5377d9be</td><td>Apprentissage de métrique appliqué à la +<br/>détection de changement de page Web et +<br/>aux attributs relatifs +<br/>thieu Cord* +<br/>* Sorbonne Universités, UPMC Univ Paris 06, UMR 7606, LIP6, F-75005, Paris, +<br/>France +<br/>RÉSUMÉ. Nous proposons dans cet article un nouveau schéma d’apprentissage de métrique. +<br/>Basé sur l’exploitation de contraintes qui impliquent des quadruplets d’images, notre approche +<br/>vise à modéliser des relations sémantiques de similarités riches ou complexes. Nous étudions +<br/>comment ce schéma peut être utilisé dans des contextes tels que la détection de régions impor- +<br/>tantes dans des pages Web ou la reconnaissance à partir d’attributs relatifs. +</td><td>('1728523', 'Nicolas Thome', 'nicolas thome')</td><td></td></tr><tr><td>ada73060c0813d957576be471756fa7190d1e72d</td><td>VRPBench: A Vehicle Routing Benchmark Tool +<br/>October 19, 2016 +</td><td>('7660594', 'Guilherme A. Zeni', 'guilherme a. zeni')<br/>('7809605', 'Mauro Menzori', 'mauro menzori')<br/>('1788152', 'Luis A. A. Meira', 'luis a. a. meira')</td><td></td></tr><tr><td>add50a7d882eb38e35fe70d11cb40b1f0059c96f</td><td>High-Fidelity Pose and Expression Normalization for Face Recognition in the Wild +<br/><b>Center for Biometrics and Security Research and National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences</b><br/>Pose and expression normalization is a crucial step to recover the canonical +<br/>view of faces under arbitrary conditions, so as to improve the face recogni- +<br/>tion performance. Most normalization algorithms can be divided in to 2D +<br/>and 3D methods. 2D methods either estimate a flow to simulate the 3D +<br/>geometry transformation or learn appearance transformations between dif- +<br/>ferent poses. 3D methods estimate the depth information with a face model +<br/>and normalize faces through 3D transformations. +<br/>An ideal normalization is desired to preserve the face appearance with +<br/>little artifact and information loss, which we call high-fidelity. However, +<br/>most previous methods fail to satisfy that. In this paper, we present a 3D +<br/>pose and expression normalization method to recover the canonical-view, +<br/>expression-free image with high fidelity. It contains three components: pose +<br/>adaptive 3D Morphable Model (3DMM) fitting, identity preserving normal- +<br/>ization and invisible region filling, which is briefly summarized in Fig. 1. +<br/>Figure 1: Overview of the High-Fidelity Pose and Expression Normalization +<br/>(HPEN) method +<br/>With an input image, the landmarks are detected with the face alignment +<br/>algorithm and we mark the corresponding 3D landmarks on the face model. +<br/>Then the 3DMM can be fitted by minimizing the distance between the 2D +<br/>landmarks and projected 3D landmarks: +<br/>arg +<br/>min +<br/>f ,R,t3d ,αid ,αexp +<br/>(cid:107)s2d − f PR(S + Aidαid + Aexpαexp +t3d)(cid:107) +<br/>(1) +<br/>where αid is the shape parameter, αexp is the expression parameter. f ,R,t3d +<br/>are pose parameters. However, when faces deviate from the frontal pose, the +<br/>correspondence between 2D and 3D landmarks will be broken, which we +<br/>model as “landmark marching”: when pose changes, the contour landmarks +<br/>move along the parallel to the visibility boundary, see Fig. 2(a). To deal with +<br/>the phenomenon we propose an approximation method to adjust contour +<br/>landmarks during 3DMM fitting. The 3D model are firstly projected with +<br/>only yaw and pitch to eliminate in-plane rotation. Then for each parallel, the +<br/>point with extreme x coordinate will be chosen as the marching destimation, +<br/>see Fig. 2(b). +<br/>With the fitted 3DMM, The face can be normalized through 3D trans- +<br/>formations. In this paper we also normalize the external face region which +<br/>contains discriminative information as well. Firstly we mark three groups of +<br/>anchors which are located on the face boundary, face surrounding and image +<br/>contour, see Fig. 3(a). Then their depth are estimated by enlarging the fitted +</td><td>('8362374', 'Xiangyu Zhu', 'xiangyu zhu')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('1716143', 'Dong Yi', 'dong yi')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td></td></tr><tr><td>ad784332cc37720f03df1c576e442c9c828a587a</td><td>Face Recognition Based on Face-Specific Subspace +<br/><b>JDL, Institute of Computing Technology, CAS, P.O. Box 2704, Beijing, China</b><br/><b>Harbin Institute of Technology, Harbin, China</b></td><td>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1698902', 'Wen Gao', 'wen gao')<br/>('1725937', 'Debin Zhao', 'debin zhao')</td><td></td></tr><tr><td>ada42b99f882ba69d70fff68c9ccbaff642d5189</td><td>Semantic Image Segmentation +<br/>and +<br/>Web-Supervised Visual Learning +<br/>D.Phil Thesis +<br/>Robotics Research Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b><br/>Supervisors: +<br/>Professor Andrew Zisserman +<br/>Dr. Antonio Criminisi +<br/>Florian Schroff +<br/><b>St. Anne s College</b><br/>Trinity, 2009 +</td><td></td><td></td></tr><tr><td>ad0d4d5c61b55a3ab29764237cd97be0ebb0ddff</td><td>Weakly Supervised Action Localization by Sparse Temporal Pooling Network +<br/><b>University of California</b><br/>Irvine, CA, USA +<br/>Google +<br/>Venice, CA, USA +<br/><b>Seoul National University</b><br/>Seoul, Korea +</td><td>('1998374', 'Phuc Nguyen', 'phuc nguyen')<br/>('40282288', 'Ting Liu', 'ting liu')<br/>('2775959', 'Gautam Prasad', 'gautam prasad')<br/>('40030651', 'Bohyung Han', 'bohyung han')</td><td>nguyenpx@uci.edu +<br/>{liuti, gautamprasad}@google.com +<br/>bhhan@snu.ac.kr +</td></tr><tr><td>adfaf01773c8af859faa5a9f40fb3aa9770a8aa7</td><td>LARGE SCALE VISUAL RECOGNITION +<br/>A DISSERTATION +<br/>PRESENTED TO THE FACULTY +<br/><b>OF PRINCETON UNIVERSITY</b><br/>IN CANDIDACY FOR THE DEGREE +<br/>OF DOCTOR OF PHILOSOPHY +<br/>RECOMMENDED FOR ACCEPTANCE +<br/>BY THE DEPARTMENT OF +<br/>COMPUTER SCIENCE +<br/>ADVISER: FEI-FEI LI +<br/>JUNE 2012 +</td><td>('8342699', 'JIA DENG', 'jia deng')</td><td></td></tr><tr><td>adf5caca605e07ee40a3b3408f7c7c92a09b0f70</td><td>Line-based PCA and LDA approaches for Face Recognition +<br/><b>Kyung Hee University South of Korea</b></td><td>('1687579', 'Vo Dinh Minh Nhat', 'vo dinh minh nhat')<br/>('1700806', 'Sungyoung Lee', 'sungyoung lee')</td><td>{vdmnhat, sylee}@oslab.khu.ac.kr +</td></tr><tr><td>adaf2b138094981edd615dbfc4b7787693dbc396</td><td>Statistical Methods For Facial +<br/>Shape-from-shading and Recognition +<br/>Submitted for the degree of Doctor of Philosophy +<br/>Department of Computer Science +<br/>20th February 2007 +</td><td>('1687021', 'William A. P. Smith', 'william a. p. smith')</td><td></td></tr><tr><td>ad6745dd793073f81abd1f3246ba4102046da022</td><td></td><td></td><td></td></tr><tr><td>ad9cb522cc257e3c5d7f896fe6a526f6583ce46f</td><td>Real-Time Recognition of Facial Expressions for Affective +<br/>Computing Applications +<br/>by +<br/>A M. Eng. Project submitted in conformity with the requirements +<br/>for the degree of Master of Engineering +<br/>Department of Mechanical and Industrial Engineering +<br/><b>University of Toronto</b></td><td>('26301224', 'Christopher Wang', 'christopher wang')<br/>('26301224', 'Christopher Wang', 'christopher wang')</td><td></td></tr><tr><td>ad08c97a511091e0f59fc6a383615c0cc704f44a</td><td>Towards the improvement of self-service +<br/>systems via emotional virtual agents +<br/>Christopher Martin +<br/>School of Computing & +<br/>Engineering Systems +<br/><b>University of Abertay</b><br/>Bell Street, Dundee +<br/>School of Computing & +<br/>Engineering Systems +<br/><b>University of Abertay</b><br/>Bell Street, Dundee +<br/>School of Computing & +<br/>Engineering Systems +<br/><b>University of Abertay</b><br/>Bell Street, Dundee +<br/>School of Social & Health +<br/>Sciences +<br/><b>University of Abertay</b><br/>Bell Street, Dundee +<br/>Affective computing and emotional agents have been found to have a positive effect on human- +<br/>computer interactions. In order to develop an acceptable emotional agent for use in a self-service +<br/>interaction, two stages of research were identified and carried out; the first to determine which +<br/>facial expressions are present in such an interaction and the second to determine which emotional +<br/>agent behaviours are perceived as appropriate during a problematic self-service shopping task. In +<br/>the first stage, facial expressions associated with negative affect were found to occur during self- +<br/>service shopping interactions, indicating that facial expression detection is suitable for detecting +<br/>negative affective states during self-service interactions. In the second stage, user perceptions of +<br/>the emotional facial expressions displayed by an emotional agent during a problematic self-service +<br/>interaction were gathered. Overall, the expression of disgust was found to be perceived as +<br/>inappropriate while emotionally neutral behaviour was perceived as appropriate, however gender +<br/>differences suggested that females perceived surprise as inappropriate. Results suggest that +<br/>agents should change their behaviour and appearance based on user characteristics such as +<br/>gender. +<br/>Keywords: affective computing, virtual agents, emotions, emotion detection, HCI, computer vision, empathy. +<br/>1. INTRODUCTION +<br/>This paper describes research which contributes +<br/>towards the development of an empathetic system +<br/>which will detect and improve a user’s affective +<br/>state during a problematic self-service interaction +<br/>(SSI) through the use of an affective agent. Self- +<br/>Service Technologies (SSTs) are those which allow +<br/>a person to obtain goods or services from a retailer +<br/>or service provider without the need for another +<br/>person to be involved in the transaction. SSTs are +<br/>used in many situations including high street shops, +<br/>supermarkets and ticket kiosks. The use of SSTs +<br/>may provide benefits such as improved customer +<br/>service (for example allowing 24 hour a day, 7 days +<br/>a week service), +<br/>labour costs and +<br/>improved efficiency (Cho & Fiorito, 2010). Less +<br/>than 5% of causes for dissatisfaction with SST +<br/>interactions were found to be the fault of the +<br/>customer (Meuter et al., 2000; Pujari, 2004), +<br/>indicating that there is a need for businesses and +<br/>SST manufacturers to improve these interactions in +<br/>order to reduce causes for dissatisfaction (Martin et +<br/>al., unpublished). The frustration caused by a +<br/>negative SSI can have a detrimental effect on a +<br/>user’s behavioural intentions towards the retailer, +<br/>impacting the likelihood the user will continue doing +<br/>reduced +<br/>business with them in the future and whether they +<br/>will recommend them to other potential users (Lin & +<br/>Hsieh, 2006; Johnson et al., 2008). By adopting +<br/>affective computing practices in SSI design, such +<br/>as giving computers the ability to detect and react +<br/>intelligently to human emotions and to express their +<br/>own simulated emotions, user experiences may be +<br/>improved (Klein et al., 1999; Jaksic et al., 2006; +<br/>Wang et al., 2009). +<br/>Affective agents have been +<br/>to reduce +<br/>found +<br/>frustration during human-computer +<br/>interactions +<br/>(HCIs) (Klein et al., 1999; Jaksic et al., 2006), +<br/>therefore we are investigating their effectiveness at +<br/>improving negative affective states in a SST user +<br/>during a shopping scenario. We propose a system +<br/>which will detect negative affective states in a user +<br/>and express appropriate empathetic reactions +<br/>using an affective virtual agent. +<br/>Two stages of research were identified. The +<br/>purpose of stage 1 (reported in Martin et al., in +<br/>press) was to investigate whether emotional facial +<br/>expressions are present during SST use, +<br/>to +<br/>determine whether a vision-based emotion detector +<br/>would be suitable for this system. The purpose of +<br/>stage 2 (reported in Martin et al., unpublished) was +<br/>© The Authors. Published by BISL. Proceedings of the BCS HCI 2012 People & Computers XXVI, Birmingham, UK351Work In Progress </td><td>('11111134', 'Leslie Ball', 'leslie ball')<br/>('2529392', 'Jacqueline Archibald', 'jacqueline archibald')<br/>('33069212', 'Lloyd Carson', 'lloyd carson')</td><td>c.martin@abertay.ac.uk +<br/>l.ball@abertay.ac.uk +<br/>j.archibald @abertay.ac.uk +<br/>l.carson@abertay.ac.uk +</td></tr><tr><td>ad2339c48ad4ffdd6100310dcbb1fb78e72fac98</td><td>Video Fill In the Blank using LR/RL LSTMs with Spatial-Temporal Attentions +<br/><b>Center for Research in Computer Vision, University of Central Florida, Orlando, FL</b></td><td>('33209161', 'Amir Mazaheri', 'amir mazaheri')<br/>('46335319', 'Dong Zhang', 'dong zhang')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>amirmazaheri@cs.ucf.edu, dzhang@cs.ucf.edu, shah@crcv.ucf.edu +</td></tr><tr><td>ad247138e751cefa3bb891c2fe69805da9c293d7</td><td>American Journal of Networks and Communications +<br/>2015; 4(4): 90-94 +<br/>Published online July 7, 2015 (http://www.sciencepublishinggroup.com/j/ajnc) +<br/>doi: 10.11648/j.ajnc.20150404.12 +<br/>ISSN: 2326-893X (Print); ISSN: 2326-8964 (Online) +<br/>A Novel Hybrid Method for Face Recognition Based on 2d +<br/>Wavelet and Singular Value Decomposition +<br/><b>Computer Engineering, Faculty of Engineering, Kharazmi University of Tehran, Tehran, Iran</b><br/><b>Islamic Azad University, Shahrood, Iran</b><br/>Email address: +<br/>To cite this article: +<br/>Decomposition. American Journal of Networks and Communications. Vol. 4, No. 4, 2015, pp. 90-94. doi: 10.11648/j.ajnc.20150404.12 +</td><td>('2653670', 'Vahid Haji Hashemi', 'vahid haji hashemi')<br/>('2153844', 'Abdorreza Alavi Gharahbagh', 'abdorreza alavi gharahbagh')<br/>('2653670', 'Vahid Haji Hashemi', 'vahid haji hashemi')<br/>('2153844', 'Abdorreza Alavi Gharahbagh', 'abdorreza alavi gharahbagh')</td><td>hajihashemi.vahid@yahoo.com (V. H. Hashemi), R_alavi@iau-shahrood.ac.ir (A. A. Gharahbagh) +</td></tr><tr><td>adf62dfa00748381ac21634ae97710bb80fc2922</td><td>ViFaI: A trained video face indexing scheme +<br/>1. Introduction +<br/>With the increasing prominence of inexpensive +<br/>video recording devices (e.g., digital camcorders and +<br/>video recording smartphones), +<br/>the average user’s +<br/>video collection today is increasing rapidly. With this +<br/>development, there arises a natural desire to rapidly +<br/>access a subset of one’s collection of videos. The solu- +<br/>tion to this problem requires an effective video index- +<br/>ing scheme. In particular, we must be able to easily +<br/>process a video to extract such indexes. +<br/>Today, there also exist large sets of labeled (tagged) +<br/>face images. One important example is an individual’s +<br/>Facebook profile. Such a set of of tagged images of +<br/>one’s self, family, friends, and colleagues represents +<br/>an extremely valuable potential training set. +<br/>In this work, we explore how to leverage the afore- +<br/>mentioned training set to solve the video indexing +<br/>problem. +<br/>2. Problem Statement +<br/>Use a labeled (tagged) training set of face images +<br/>to extract relevant indexes from a collection of videos, +<br/>and use these indexes to answer boolean queries of the +<br/>form: “videos with ‘Person 1’ OP1 ‘Person 2’ OP2 ... +<br/>OP(N-1) ‘Person N’ ”, where ‘Person N’ corresponds +<br/>to a training label (tag) and OPN is a boolean operand +<br/>such as AND, OR, NOT, XOR, and so on. +<br/>3. Proposed Scheme +<br/>In this section, we outline our proposed scheme to +<br/>address the problem we postulate in the previous sec- +<br/>tion. We provide further details about the system im- +<br/>plementation in Section 4. +<br/>At a high level, we subdivide the problem into two +<br/>key phases: the first ”off-line” executed once, and the +<br/>second ”on-line” phase instantiated upon each query. +<br/>For the purposes of this work, we define an index as +<br/>follows: <video id, tag, frame #>. +<br/>3.1. The training phase +<br/>We first outline Phase 1 (the training or “off-line” +<br/>phase): +<br/>1. Use the labeled training set plus an additional set +<br/>of ‘other’ faces to compute the Fisher Linear Dis- +<br/>criminant (FLD) [1]. +<br/>2. Project the training data onto the space defined by +<br/>the eigenvectors returned by the FLD, and train +<br/>a classifier (first nearest neighbour, then SVM if +<br/>required) using the training features. +<br/>3. Iterate through each frame of each video, detect- +<br/>ing faces [2], classifying detected results, and add +<br/>an index if the detected face corresponds to one of +<br/>the labeled classes from the previous step. +<br/>3.2. The query phase +<br/>Now, we outline Phase 2 (the query or “on-line” +<br/>phase): +<br/>1. Key the indexes on their video id. +<br/>2. For each video, evaluate the boolean query for the +<br/>set of corresponding indexes. +<br/>3. Keep videos for which the boolean query evalu- +<br/>ates true, and discard those for which it evaluates +<br/>false. +<br/>4. Implementation Details +<br/>We are implementing the project in C++, leverag- +<br/>ing the OpenCV v2.2 framework [4]. In this section, +<br/>we will highlight some of the critical implementation +<br/>details of our proposed system. +</td><td>('30006340', 'Nayyar', 'nayyar')<br/>('47384529', 'Audrey Wei', 'audrey wei')</td><td>hnayyar@stanford.edu +<br/>awei1001@stanford.edu +</td></tr><tr><td>bbc4b376ebd296fb9848b857527a72c82828fc52</td><td>Attributes for Improved Attributes +<br/><b>University of Maryland</b><br/><b>College Park, MD</b></td><td>('3351637', 'Emily Hand', 'emily hand')</td><td>emhand@cs.umd.edu +</td></tr><tr><td>bb489e4de6f9b835d70ab46217f11e32887931a2</td><td>Everything you wanted to know about Deep Learning for Computer Vision but were +<br/>afraid to ask +<br/>Moacir A. Ponti, Leonardo S. F. Ribeiro, Tiago S. Nazare +<br/><b>ICMC University of S ao Paulo</b><br/>S˜ao Carlos/SP, 13566-590, Brazil +<br/><b>CVSSP University of Surrey</b><br/>Guildford, GU2 7XH, UK +<br/>tools, +</td><td>('2227956', 'Tu Bui', 'tu bui')<br/>('10710438', 'John Collomosse', 'john collomosse')</td><td>Email: [ponti, leonardo.sampaio.ribeiro, tiagosn]@usp.br +<br/>Email: [t.bui, j.collomosse]@surrey.ac.uk +</td></tr><tr><td>bba281fe9c309afe4e5cc7d61d7cff1413b29558</td><td>Social Cognitive and Affective Neuroscience, 2017, 984–992 +<br/>doi: 10.1093/scan/nsx030 +<br/>Advance Access Publication Date: 11 April 2017 +<br/>Original article +<br/>An unpleasant emotional state reduces working +<br/>memory capacity: electrophysiological evidence +<br/>1Laboratorio de Neurofisiologia do Comportamento, Departamento de Fisiologia e Farmacologia, Instituto +<br/>Biome´dico, Universidade Federal Fluminense, Niteroi, Brazil, 2MograbiLab, Departamento de Psicologia, +<br/>Pontifıcia Universidade Catolica do Rio de Janeiro, Rio de Janeiro, Brazil, and 3Laboratorio de Engenharia +<br/>Pulmonar, Programa de Engenharia Biome´dica, COPPE, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil +</td><td>('18129331', 'Jessica S. B. Figueira', 'jessica s. b. figueira')<br/>('2993713', 'Leticia Oliveira', 'leticia oliveira')<br/>('38252417', 'Mirtes G. Pereira', 'mirtes g. pereira')<br/>('18138365', 'Luiza B. Pacheco', 'luiza b. pacheco')<br/>('6891211', 'Isabela Lobo', 'isabela lobo')<br/>('5663717', 'Gabriel C. Motta-Ribeiro', 'gabriel c. motta-ribeiro')<br/>('1837214', 'Isabel A. David', 'isabel a. david')<br/>('1837214', 'Isabel A. David', 'isabel a. david')</td><td>Fluminense, Rua Hernani Pires de Mello, 101, Niteroi, RJ 24210-130, Brazil. E-mail: isabeldavid@id.uff.br. +</td></tr><tr><td>bb557f4af797cae9205d5c159f1e2fdfe2d8b096</td><td></td><td></td><td></td></tr><tr><td>bb06ef67a49849c169781657be0bb717587990e0</td><td>Impact of Temporal Subsampling on Accuracy and +<br/>Performance in Practical Video Classification +<br/>F. Scheidegger∗†, L. Cavigelli∗, M. Schaffner∗, A. C. I. Malossi†, C. Bekas†, L. Benini∗‡ +<br/>∗ETH Zürich, 8092 Zürich, Switzerland +<br/>†IBM Research - Zürich, 8803 Rüschlikon, Switzerland +<br/>‡Università di Bologna, Italy +</td><td></td><td></td></tr><tr><td>bb22104d2128e323051fb58a6fe1b3d24a9e9a46</td><td>IAJ=JE BH ==OIEI 1 AIIA?A ?= EBH=JE =EO B?KIAI JDA IK>JA +<br/>ABBA?JELAAII B KH =CHEJD +<br/>==OIEI 7IK=O = B=?E= ANFHAIIE ==OIEI IOIJA ?J=EI JDHAA IJ=CAI B=?A =?GKE +<br/>9DAJDAH KIEC *=OAIE= ?=IIEAH " & IKFFHJ LA?JH =?DEA 58 H AKH= +<br/>HACEI E = IECA ?=IIEAH EI = ? IJH=JACO & 0MALAH J = ?= HACEI +</td><td></td><td>)=OEC .=?E= -NFHAIIE >O .KIEC =EB@I +<br/>9A;= +D=C1,2 +DK5C +DA1,3 =@ ;E2EC 0KC1,2,3 +<br/>11IJEJKJA B 1BH=JE 5?EA?A )?=@AE= 5EE?= 6=EM= +<br/>2,AFJ B +FKJAH 5?EA?A =@ 1BH=JE -CEAAHEC =JE= 6=EM= 7ELAHIEJO +<br/>3/H=@K=JA 1IJEJKJA B AJMHEC =@ KJEA@E= =JE= 6=EM= 7ELAHIEJO +<br/>{wychang, song}@iis.sinica.edu.tw; hung@csie.ntu.edu.tw +<br/>)>IJH=?J .A=JKHA HAFHAIAJ=JE =@ ?=IIE?=JE =HA JM =H EIIKAI E B=?E= +<br/>ANFHAIIE ==OIEI 1 JDA F=IJ IJ AJD@I KIA@ AEJDAH DEIJE? H ?= HAFHA +<br/>L=HE=JEI B ANFHAIIEI =@ DEIJE? HAFHAIAJ=JE IJHAIIAI C>= @ELAHIE +<br/>JEAI 6 J=A JDA =@L=J=CAI B >JD = DO>HE@ HAFHAIAJ=JE EI IKCCAIJA@ E JDEI +<br/>F=FAH =@ =EB@ A=HEC EI =FFEA@ J ?D=H=?JAHEA C>= =@ ?= EBH= +<br/>JE @EI?HEE=JELAO 7EA IA AJD@I KIEC KIKFAHLEIA@ =EB@ A=H +<br/>EC =FFH=?DAI A>A@@A@ =EB@I B JDA DO>HE@ HAFHAIAJ=JE =HA A=HA@ >O +<br/>=@FJEC = IKFAHLEIA@ =EB@ A=HEC JA?DEGKA 6 EJACH=JA JDAIA =EB@I +<br/>ABBA?JELAO = BKIE ?=IIEAH EI EJH@K?A@ MDE?D ?= DAF J AFO IKEJ=>A +<br/>?>E=JE MAECDJI B B=?E= ?FAJI J E@AJEBO = ANFHAIIE +FHADA +<br/>IELA ?F=HEII B=?E= ANFHAIIE HA?CEJE =HA E?K@A@ J @AIJH=JA JDA +<br/> 1JH@K?JE +<br/>4A=EEC DK= AJEI F=OI = EFHJ=J HA E DK= ?KE?=JE 6 IJK@O +<br/>DK= >AD=LEH I?EAJE?=O =@ IOIJA=JE?=O AJE ==OIEI EI = EJHECKEC HA +<br/>IA=H?D EIIKA E =O A@I K?D =JJAJE D=I >AA @H=M J JDEI JFE? E ?FKJAH +<br/>LEIE =FFE?=JEI IK?D =I DK=?FKJAH EJAH=?JE H>J ?CEJE =@ >AD=LEH +<br/>IEJE BA=JKHA ANJH=?JE =@ ?=IIE?=JE +<br/>.H BA=JKHA ANJH=?JE = J B AJD@I D=LA >AA FHFIA@ 1 CAAH= IJ AJD +<br/>@I HAFHAIAJ BA=JKHAI E AEJDAH DEIJE? H ?= M=OI 0EIJE? HAFHAIAJ=JE KIAI JDA +<br/>MDA B=?A BH HAFHAIAJ=JE =@ B?KIAI JDA B=?E= L=HE=JEI B C>= =FFA=H=?A +<br/>1 ?JH=IJ ?= HAFHAIAJ=JE =@FJI ?= B=?E= HACEI H BA=JKHAI =@ CELAI =JJA +<br/>JE J JDA IK>JA @ELAHIEJEAI = B=?A 6DKCD IJ HA?AJ IJK@EAI D=LA >AA @EHA?JA@ +<br/>JM=H@I ?= HAFHAIAJ=JE % & C@ HAIA=H?D HAIKJI =HA IJE >J=EA@ >O KIEC +<br/>DEIJE? =FFH=?D 0A?A EJ EI EJAHAIJEC J ANFEJ >JD B JDAEH >AAJI J @A +<br/>LAF = DO>HE@ HAFHAIAJ=JE +<br/>1 =@@EJE J BA=JKHA HAFHAIAJ=JE MA =I EJH@K?A = AJD@ BH ?=IIE?=JE +<br/>AJMHI @EC = IJHC ?=IIEAH EI JDA ?HA E JDA ANEIJEC B=?E= ANFHAIIE ==O +<br/>IEI IJK@EAI 1 JDA =FFH=?DAI JD=J =@FJ ?= B=?E= EBH=JE MAECDJEC JDAIA ?= +</td></tr><tr><td>bbf28f39e5038813afd74cf1bc78d55fcbe630f1</td><td>Style Aggregated Network for Facial Landmark Detection +<br/><b>University of Technology Sydney, 2 The University of Sydney</b></td><td>('9929684', 'Xuanyi Dong', 'xuanyi dong')<br/>('1685212', 'Yan Yan', 'yan yan')<br/>('3001348', 'Wanli Ouyang', 'wanli ouyang')<br/>('1698559', 'Yi Yang', 'yi yang')</td><td>{xuanyi.dong,yan.yan-3}@student.uts.edu.au; +<br/>wanli.ouyang@sydney.edu.au; yi.yang@uts.edu.au +</td></tr><tr><td>bbe1332b4d83986542f5db359aee1fd9b9ba9967</td><td></td><td></td><td></td></tr><tr><td>bbe949c06dc4872c7976950b655788555fe513b8</td><td>Automatic Frequency Band Selection for +<br/>Illumination Robust Face Recognition +<br/><b>Institute of Anthropomatics, Karlsruhe Institute of Technology, Germany</b></td><td>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>{ekenel,rainer.stiefelhagen}@kit.edu +</td></tr><tr><td>bbcb4920b312da201bf4d2359383fb4ee3b17ed9</td><td>RESEARCH ARTICLE +<br/>Robust Face Recognition via Multi-Scale +<br/>Patch-Based Matrix Regression +<br/><b>Institute of Advanced Technology, Nanjing University of Posts and Telecommunications, Nanjing</b><br/><b>China, 2 School of Computer Science and Engineering, Nanjing University of Science and Technology</b><br/><b>Nanjing, 210094, China, 3 School of Automation, Nanjing University of Posts and Telecommunications</b><br/><b>Nanjing, 210023, China, 4 School of Computer Science and Technology, Nanjing University of Posts and</b><br/>Telecommunications, Nanjing, 210023, China +<br/>a11111 +</td><td>('3306402', 'Guangwei Gao', 'guangwei gao')<br/>('2700773', 'Jian Yang', 'jian yang')<br/>('1712078', 'Xiaoyuan Jing', 'xiaoyuan jing')<br/>('35919708', 'Pu Huang', 'pu huang')<br/>('3359690', 'Juliang Hua', 'juliang hua')<br/>('1742990', 'Dong Yue', 'dong yue')</td><td>* csggao@gmail.com +</td></tr><tr><td>bb6bf94bffc37ef2970410e74a6b6dc44a7f4feb</td><td>Situation Recognition with Graph Neural Networks +<br/>Supplementary Material +<br/><b>Uber Advanced Technologies Group, 5Vector Institute</b><br/>We present additional analysis and results of our approach in the supplementary material. First, we analyze the verb +<br/>prediction performance in Sec. 1. In Sec. 2, we present t-SNE [2] plots to visualize the verb and role embeddings. We present +<br/>several examples of the influence of different roles on predicting the verb-frame correctly. This is visualized in Sec. 3 through +<br/>propagation matrices similar to Fig. 7 of the main paper. Finally, in Sec. 4 we include several example predictions that our +<br/>model makes. +<br/>1. Verb Prediction +<br/>We present the verb prediction accuracies for our fully-connected model on the development set in Fig. 1. The random +<br/>performance is close to 0.2% (504 verbs). About 22% of all verbs are classified correctly over 50% of the time. These +<br/>include taxiing, erupting, flossing, microwaving, etc. On the other hand, verbs such as attaching, +<br/>making, placing can have very different image representations, and show prediction accuracies of less than 10%. +<br/>Our model helps improve the role-noun predictions by sharing information across all roles. Nevertheless, if the verb is +<br/>predicted incorrectly, the whole situation is treated as incorrect. Thus, verb prediction performance plays a crucial role. +<br/>Figure 1. Verb prediction accuracy on the development set. Some verbs such as taxiing typically have a similar image (a plane on the +<br/>tarmac), while verbs such as rubbing or twisting can have very different corresponding images. +<br/>taxiinglappingretrievingflickingminingwaxingjugglingcurtsyingcommutingdancingcrushingreadingexaminingdousingdecomposingchoppingdrawingcryingcalmingsniffingmourningsubmergingtwistingcarvingrubbingaskingVerbs0102030405060708090100Accuracy (%) </td><td>('8139953', 'Ruiyu Li', 'ruiyu li')<br/>('2103464', 'Makarand Tapaswi', 'makarand tapaswi')<br/>('2246396', 'Renjie Liao', 'renjie liao')<br/>('1729056', 'Jiaya Jia', 'jiaya jia')<br/>('2422559', 'Raquel Urtasun', 'raquel urtasun')<br/>('37895334', 'Sanja Fidler', 'sanja fidler')<br/>('2043324', 'Hong Kong', 'hong kong')</td><td>ryli@cse.cuhk.edu.hk, {makarand,rjliao,urtasun,fidler}@cs.toronto.edu, leojia9@gmail.com +</td></tr><tr><td>bb7f2c5d84797742f1d819ea34d1f4b4f8d7c197</td><td>TO APPEAR IN TPAMI +<br/>From Images to 3D Shape Attributes +</td><td>('1786435', 'David F. Fouhey', 'david f. fouhey')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>bbf01aa347982592b3e4c9e4f433e05d30e71305</td><td></td><td></td><td></td></tr><tr><td>bbc5f4052674278c96abe7ff9dc2d75071b6e3f3</td><td>Nonlinear Hierarchical Part-based Regression for Unconstrained Face Alignment +<br/>†NEC Laboratories America, Media Analytics +<br/>‡Adobe Research +<br/><b>cid:93)University of North Carolina at Charlotte</b><br/><b>Rutgers, The State University of New Jersey</b></td><td>('39960064', 'Xiang Yu', 'xiang yu')<br/>('1753384', 'Shaoting Zhang', 'shaoting zhang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>xiangyu@nec-labs.com, zlin@adobe.com, szhang16@uncc.edu, dnm@cs.rutgers.edu +</td></tr><tr><td>bbfe0527e277e0213aafe068113d719b2e62b09c</td><td>Dog Breed Classification Using Part Localization +<br/><b>Columbia University</b><br/><b>University of Maryland</b></td><td>('2454675', 'Jiongxin Liu', 'jiongxin liu')<br/>('20615377', 'Angjoo Kanazawa', 'angjoo kanazawa')</td><td></td></tr><tr><td>bbf1396eb826b3826c5a800975047beabde2f0de</td><td></td><td></td><td></td></tr><tr><td>bb451dc2420e1a090c4796c19716f93a9ef867c9</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 104 – No.5, October 2014 +<br/>A Review on: Automatic Movie Character Annotation +<br/>by Robust Face-Name Graph Matching +<br/>Research Scholar +<br/><b>Sinhgad College of</b><br/>Engineering, korti, Pandharpur, +<br/><b>Solapur University, INDIA</b><br/>Gadekar P.R. +<br/>Assistant Professor +<br/><b>Sinhgad College of</b><br/>Engineering, korti, Pandharpur, +<br/><b>Solapur University, INDIA</b><br/>Bandgar Vishal V. +<br/>Assistant Professor +<br/><b>College of Engineering (Poly</b><br/>Pandharpur, Solapur, INDIA +<br/>Bhise Avdhut S. +<br/>HOD, Department of +<br/>Information Technology, +<br/><b>College of Engineering (Poly</b><br/>Pandharpur, Solapur, INDIA +</td><td></td><td></td></tr><tr><td>bbd1eb87c0686fddb838421050007e934b2d74ab</td><td></td><td></td><td></td></tr><tr><td>d73d2c9a6cef79052f9236e825058d5d9cdc1321</td><td>2014-ENST-0040 +<br/>EDITE - ED 130 +<br/>Doctorat ParisTech +<br/>T H È S E +<br/>pour obtenir le grade de docteur délivré par +<br/>TELECOM ParisTech +<br/>Spécialité « Signal et Images » +<br/>présentée et soutenue publiquement par +<br/>le 08 juillet 2014 +<br/>Cutting the Visual World into Bigger Slices for Improved Video +<br/>Concept Detection +<br/>Amélioration de la détection des concepts dans les vidéos par de plus grandes tranches du Monde +<br/>Visuel +<br/>Directeur de thèse : Bernard Mérialdo +<br/>Jury +<br/>M. Philippe-Henri Gosselin, Professeur, INRIA +<br/>M. Georges Quénot, Directeur de recherche CNRS, LIG +<br/>M. Georges Linares, Professeur, LIA +<br/>M. François Brémond, Professeur, INRIA +<br/>M. Bernard Mérialdo, Professeur, EURECOM +<br/>Rapporteur +<br/>Rapporteur +<br/>Examinateur +<br/>Examinateur +<br/>Encadrant +<br/>TELECOM ParisTech +<br/>école de l’Institut Télécom - membre de ParisTech +</td><td>('2135932', 'Usman Farrokh Niaz', 'usman farrokh niaz')</td><td></td></tr><tr><td>d794ffece3533567d838f1bd7f442afee13148fd</td><td>Hand Detection and Tracking in Videos +<br/>for Fine-grained Action Recognition +<br/><b>The University of Electro-Communications, Tokyo</b><br/>1-5-1 Chofugaoka, Chofu, Tokyo, 182-8585 Japan +</td><td>('1681659', 'Keiji Yanai', 'keiji yanai')</td><td></td></tr><tr><td>d78077a7aa8a302d4a6a09fb9737ab489ae169a6</td><td></td><td></td><td></td></tr><tr><td>d7593148e4319df7a288180d920f2822eeecea0b</td><td>LIU, YU, FUNES-MORA, ODOBEZ: DIFFERENTIAL APPROACH FOR GAZE ESTIMATION 1 +<br/>A Differential Approach for Gaze +<br/>Estimation with Calibration +<br/><b>Idiap Research Institute</b><br/>2 Eyeware Tech SA +<br/>Kenneth A. Funes-Mora 2 +</td><td>('1697913', 'Gang Liu', 'gang liu')<br/>('50133842', 'Yu Yu', 'yu yu')<br/>('1719610', 'Jean-Marc Odobez', 'jean-marc odobez')</td><td>gang.liu@idiap.ch +<br/>yu.yu@idiap.ch +<br/>kenneth@eyeware.tech +<br/>odobez@idiap.ch +</td></tr><tr><td>d7312149a6b773d1d97c0c2b847609c07b5255ec</td><td></td><td></td><td></td></tr><tr><td>d7fe2a52d0ad915b78330340a8111e0b5a66513a</td><td>Unpaired Photo-to-Caricature Translation on Faces in +<br/>the Wild +<br/><b>aNo. 238 Songling Road, Ocean University of</b><br/>China, Qingdao, China +</td><td>('4670300', 'Ziqiang Zheng', 'ziqiang zheng')<br/>('50077564', 'Zhibin Yu', 'zhibin yu')<br/>('2336297', 'Haiyong Zheng', 'haiyong zheng')<br/>('49297407', 'Bing Zheng', 'bing zheng')</td><td></td></tr><tr><td>d7cbedbee06293e78661335c7dd9059c70143a28</td><td>MobileFaceNets: Efficient CNNs for Accurate Real- +<br/>Time Face Verification on Mobile Devices +<br/><b>School of Computer and Information Technology, Beijing Jiaotong University, Beijing</b><br/><b>Research Institute, Watchdata Inc., Beijing, China</b><br/>China +</td><td>('39326372', 'Sheng Chen', 'sheng chen')<br/>('1681842', 'Yang Liu', 'yang liu')<br/>('46757550', 'Xiang Gao', 'xiang gao')<br/>('2765914', 'Zhen Han', 'zhen han')</td><td>{sheng.chen, yang.liu.yj, xiang.gao}@watchdata.com, +<br/>zhan@bjtu.edu.cn +</td></tr><tr><td>d7d9c1fa77f3a3b3c2eedbeb02e8e7e49c955a2f</td><td>Automating Image Analysis by Annotating Landmarks with Deep +<br/>Neural Networks +<br/>February 3, 2017 +<br/>Running head: Automatic Annotation of Landmarks +<br/><b>Boston University, Boston, MA</b><br/><b>University of North Carolina at Chapel Hill, Chapel Hill, NC</b><br/>Keywords: automatic landmark localization, annotation, pose estimation, deep neural networks, hawkmoths +<br/>Contents +</td><td>('2025025', 'Mikhail Breslav', 'mikhail breslav')<br/>('1711465', 'Tyson L. Hedrick', 'tyson l. hedrick')<br/>('1749590', 'Stan Sclaroff', 'stan sclaroff')<br/>('1723703', 'Margrit Betke', 'margrit betke')</td><td></td></tr><tr><td>d708ce7103a992634b1b4e87612815f03ba3ab24</td><td>FCVID: Fudan-Columbia Video Dataset +<br/>Available at: http://bigvid.fudan.edu.cn/FCVID/ +<br/>1 OVERVIEW +<br/>Recognizing visual contents in unconstrained videos +<br/>has become a very important problem for many ap- +<br/>plications, such as Web video search and recommen- +<br/>dation, smart content-aware advertising, robotics, etc. +<br/>Existing datasets for video content recognition are +<br/>either small or do not have reliable manual labels. +<br/>In this work, we construct and release a new Inter- +<br/>net video dataset called Fudan-Columbia Video Dataset +<br/>(FCVID), containing 91,223 Web videos (total duration +<br/>4,232 hours) annotated manually according to 239 +<br/>categories. We believe that the release of FCVID can +<br/>stimulate innovative research on this challenging and +<br/>important problem. +<br/>2 COLLECTION AND ANNOTATION +<br/>The categories in FCVID cover a wide range of topics +<br/>like social events (e.g., “tailgate party”), procedural +<br/>events (e.g., “making cake”), objects (e.g., “panda”), +<br/>scenes (e.g., “beach”), etc. These categories were de- +<br/>fined very carefully. Specifically, we conducted user +<br/>surveys and used the organization structures on +<br/>YouTube and Vimeo as references, and browsed nu- +<br/>merous videos to identify categories that satisfy the +<br/>following three criteria: (1) utility — high relevance +<br/>in supporting practical application needs; (2) cover- +<br/>age — a good coverage of the contents that people +<br/>record; and (3) feasibility — likely to be automatically +<br/>recognized in the next several years, and a high +<br/>frequency of occurrence that is sufficient for training +<br/>a recognition algorithm. +<br/>This definition effort led to a set of over 250 candi- +<br/>date categories. For each category, in addition to the +<br/>official name used in the public release, we manually +<br/>defined another alternative name. Videos were then +<br/>downloaded from YouTube searches using the official +<br/>and the alternative names as search terms. The pur- +<br/>pose of using the alternative names was to expand the +<br/>candidate video sets. For each search, we downloaded +<br/>1,000 videos, and after removing duplicate videos and +<br/>some extremely long ones (longer than 30 minutes), +<br/>there were around 1,000–1,500 candidate videos for +<br/>each category. +<br/>All the videos were annotated manually to ensure +<br/>a high precision of the FCVID labels. In order to min- +<br/>imize subjectivity, nearly 20 annotators were involved +<br/>in the task, and a master annotator was assigned to +<br/>monitor the entire process and double-check all the +<br/>found positive videos. Some of the videos are multi- +<br/>labeled, and thus filtering the 1,000–1,500 videos for +<br/>each category with focus on just the single category +<br/>label is not adequate. As checking the existence of all +<br/>the 250+ classes for each video is extremely difficult, +<br/>we use the following strategy to narrow down the “la- +<br/>bel search space” for each video. We first grouped the +<br/>categories according to subjective predictions of label +<br/>co-occurrences, e.g., “wedding reception” & “wed- +<br/>ding ceremony”, “waterfall” & “river”, “hiking” & +<br/>“mountain”, and even “dog” & “birthday”. We then +<br/>annotated the videos not only based on the target cat- +<br/>egory label, but also according to the identified related +<br/>labels. This helped produce a fairly complete label +<br/>set for FCVID, but largely reduced the annotation +<br/>workload. After removing the rare categories with +<br/>less than 100 videos after annotation, the final FCVID +<br/>dataset contains 91,223 videos and 239 categories, +<br/>where 183 are events and 56 are objects, scenes, etc. +<br/>Figure 1 shows the number of videos per category. +<br/>“Dog” has the largest number of positive videos +<br/>(1,136), while “making egg tarts” is the most infre- +<br/>quent category containing only 108 samples. The total +<br/>duration of FCVID is 4,232 hours with an average +<br/>video duration of 167 seconds. Figure 2 further gives +<br/>the average video duration of each category. +<br/>The categories are organized using a hierarchy con- +<br/>taining 11 high-level groups, as visualized in Figure 3. +<br/>3 COMPARISON WITH RELATED DATASETS +<br/>We compare FCVID with the following datasets. Most +<br/>of them have been widely adopted in the existing +<br/>works on video categorization. +<br/>KTH and Weizmann: The KTH [1] and the Weiz- +<br/>mann [2] datasets are well-known benchmarks for +<br/>human action recognition. The former contains 600 +<br/>videos of 6 human actions performed by 25 people +<br/>in four scenarios, and the latter consists of 81 videos +<br/>associated with 9 actions performed by 9 actors. +<br/>Hollywood Human Action: The Hollywood +<br/>dataset [3] contains 8 action classes collected from +<br/>32 Hollywood movies with a total of 430 videos. +</td><td>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('3099139', 'Zuxuan Wu', 'zuxuan wu')<br/>('39811558', 'Jun Wang', 'jun wang')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td></td></tr><tr><td>d78734c54f29e4474b4d47334278cfde6efe963a</td><td>Exploring Disentangled Feature Representation Beyond Face Identification +<br/><b>CUHK-SenseTime Joint Lab, The Chinese University of Hong Kong</b><br/><b>SenseTime Group Limited, 3Peking University</b></td><td>('1715752', 'Yu Liu', 'yu liu')<br/>('22181490', 'Fangyin Wei', 'fangyin wei')<br/>('49895575', 'Jing Shao', 'jing shao')<br/>('37145669', 'Lu Sheng', 'lu sheng')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')</td><td>{yuliu,lsheng,xgwang}@ee.cuhk.edu.hk, weifangyin@pku.edu.cn, +<br/>{shaojing,yanjunjie}@sensetime.com +</td></tr><tr><td>d785fcf71cb22f9c33473cba35f075c1f0f06ffc</td><td>Learning Active Facial Patches for Expression Analysis +<br/><b>Rutgers University, Piscataway, NJ</b><br/><b>Nanjing University of Information Science and Technology, Nanjing, 210044, China</b><br/><b>University of Texas at Arlington, Arlington, TX</b></td><td>('29803023', 'Lin Zhong', 'lin zhong')<br/>('1734954', 'Qingshan Liu', 'qingshan liu')<br/>('39606160', 'Peng Yang', 'peng yang')<br/>('40107085', 'Bo Liu', 'bo liu')<br/>('1768190', 'Junzhou Huang', 'junzhou huang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>{linzhong,qsliu,peyang,lb507,dnm}@cs.rutgers.edu, Jzhuang@uta.edu +</td></tr><tr><td>d79365336115661b0e8dbbcd4b2aa1f504b91af6</td><td>Variational methods for Conditional Multimodal +<br/>Deep Learning +<br/>Department of Computer Science and Automation +<br/><b>Indian Institute of Science</b></td><td>('2686270', 'Gaurav Pandey', 'gaurav pandey')<br/>('2440174', 'Ambedkar Dukkipati', 'ambedkar dukkipati')</td><td>Email{gp88, ad@csa.iisc.ernet.in +</td></tr><tr><td>d7b6bbb94ac20f5e75893f140ef7e207db7cd483</td><td>Griffith Research Online +<br/>https://research-repository.griffith.edu.au +<br/>Face Recognition across Pose: A +<br/>Review +<br/>Author +<br/>Zhang, Paul, Gao, Yongsheng +<br/>Published +<br/>2009 +<br/>Journal Title +<br/>Pattern Recognition +<br/>DOI +<br/>https://doi.org/10.1016/j.patcog.2009.04.017 +<br/>Copyright Statement +<br/>Copyright 2009 Elsevier. This is the author-manuscript version of this paper. Reproduced in accordance +<br/>with the copyright policy of the publisher. Please refer to the journal's website for access to the +<br/>definitive, published version. +<br/>Downloaded from +<br/>http://hdl.handle.net/10072/30193 +</td><td></td><td></td></tr><tr><td>d78373de773c2271a10b89466fe1858c3cab677f</td><td></td><td></td><td></td></tr><tr><td>d78fbd11f12cbc194e8ede761d292dc2c02d38a2</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 8, No. 10, 2017 +<br/>Enhancing Gray Scale Images for Face Detection +<br/>under Unstable Lighting Condition +<br/>Department of Mathematics and Computer Science, +<br/>Faculty of Sciences, PO Box 67 Dschang +<br/><b>University of Dschang, Cameroon</b><br/>DJIMELI TSAMENE Charly +<br/>Department of Mathematics and Computer Science, +<br/>Faculty of Sciences, PO Box 67 Dschang, +<br/><b>University of Dschang, Cameroon</b><br/>techniques compared are: +</td><td></td><td></td></tr><tr><td>d72973a72b5d891a4c2d873daeb1bc274b48cddf</td><td>A New Supervised Dimensionality Reduction Algorithm Using Linear +<br/>Discriminant Analysis and Locality Preserving Projection +<br/>School of Information Engineering +<br/><b>Guangdong Medical College</b><br/>Dongguan, Guangdong, China +<br/>School of Electronics and Information +<br/><b>South China University of Technology</b><br/>Guangzhou, Guangdong, China +</td><td>('2588058', 'DI ZHANG', 'di zhang')<br/>('20374749', 'YUN ZHAO', 'yun zhao')<br/>('31866339', 'MINGHUI DU', 'minghui du')</td><td> haihaiwenqi@163.com, zyun@gdmc.edu.cn +<br/>ecmhdu@scut.edu.cn +</td></tr><tr><td>d700aedcb22a4be374c40d8bee50aef9f85d98ef</td><td>Rethinking Spatiotemporal Feature Learning: +<br/>Speed-Accuracy Trade-offs in Video Classification +<br/>1 Google Research +<br/><b>University of California San Diego</b></td><td>('1817030', 'Saining Xie', 'saining xie')<br/>('40559421', 'Chen Sun', 'chen sun')<br/>('1808244', 'Jonathan Huang', 'jonathan huang')<br/>('1736745', 'Zhuowen Tu', 'zhuowen tu')<br/>('1702318', 'Kevin Murphy', 'kevin murphy')</td><td></td></tr><tr><td>d7d166aee5369b79ea2d71a6edd73b7599597aaa</td><td>Fast Subspace Clustering Based on the +<br/>Kronecker Product +<br/><b>Beihang University 2Gri th University 3University of York, UK</b></td><td>('38840844', 'Lei Zhou', 'lei zhou')<br/>('3042223', 'Xiao Bai', 'xiao bai')<br/>('6820648', 'Xianglong Liu', 'xianglong liu')<br/>('40582215', 'Jun Zhou', 'jun zhou')<br/>('38987678', 'Hancock Edwin', 'hancock edwin')</td><td></td></tr><tr><td>d79f9ada35e4410cd255db39d7cc557017f8111a</td><td>Journal of Eye Movement Research +<br/>7(3):3, 1-8 +<br/>Evaluation of accurate eye corner detection methods for gaze +<br/>estimation +<br/><b>Public University of Navarra, Spain</b><br/>Childrens National Medical Center, USA +<br/><b>Public University of Navarra, Spain</b><br/><b>Public University of Navarra, Spain</b><br/>Accurate detection of iris center and eye corners appears to be a promising +<br/>approach for low cost gaze estimation. +<br/>In this paper we propose novel eye +<br/>inner corner detection methods. Appearance and feature based segmentation +<br/>approaches are suggested. All these methods are exhaustively tested on a realistic +<br/>dataset containing images of subjects gazing at different points on a screen. +<br/>We have demonstrated that a method based on a neural network presents the +<br/>best performance even in light changing scenarios. +<br/>In addition to this method, +<br/>algorithms based on AAM and Harris corner detector present better accuracies +<br/>than recent high performance face points tracking methods such as Intraface. +<br/>Keywords: eye tracking, low cost, eye inner corner +<br/>Introduction +<br/>Research on eye detection and tracking has attracted +<br/>much attention in the last decades. Since it is one of the +<br/>most stable and representative features of the subject, +<br/>eye detection is used in a great variety of applications, +<br/>such as subject identification, human computer inter- +<br/>action as shown in Morimoto and Mimica (2005) and +<br/>gesture recognition as described by Tian, Kanade, and +<br/>Cohn (2000) and Bailenson et al. (2008). +<br/>Human computer interaction based on eye informa- +<br/>tion is one of the most challenging research topics in +<br/>the recent years. According to the literature, the first +<br/>attempts to track the human gaze using cameras be- +<br/>gan in 1974 as shown in the work by Merchant, Mor- +<br/>rissette, and Porterfield (1974). Since then, and espe- +<br/>cially in the last decades, much effort has been devoted +<br/>to improving the performance of eye tracking systems. +<br/>The availability of high performance eye tracking sys- +<br/>tems has provided advances in fields such as usabil- +<br/>ity research as described by Ellis, Candrea, Misner, +<br/>Craig, and Lankford (1998) Poole and Ball (2005) and +<br/>interaction for severely disabled people in works such +<br/>as Bolt (1982), Starker and Bolt (1990) and Vertegaal +<br/>(1999). Gaze tracking systems can be used to deter- +<br/>mine the fixation point of an individual on a computer +<br/>screen, which can in turn be used as a pointer to in- +<br/>teract with the computer. Thus, severely disabled peo- +<br/>ple who cannot communicate with their environment +<br/>using alternative interaction tools can perform several +<br/>tasks by means of their gaze. Performance limitations, +<br/>such as head movement constraints, limit the employ- +<br/>ment of the gaze trackers as interaction tools in other +<br/>areas. Moreover, the limited market for eye tracking +<br/>systems and the specialized hardware they employ, in- +<br/>crease their prices. The eye tracking community has +<br/>identified new application fields, such as video games +<br/>or the automotive industry, as potential markets for the +<br/>technology (Zhang, Bulling, & Gellersen, 2013). How- +<br/>ever, simpler (i.e., lower cost) hardware is needed to +<br/>reach these areas. +<br/>Although web cams offer acceptable resolutions for +<br/>eye tracking purposes, the optics used provide a wider +<br/>field of view in which the whole face appears. By con- +<br/>trast, most of the existing high-performance eye track- +<br/>ing systems employ infrared illumination. +<br/>Infrared +<br/>light-emitting diodes provide a higher image quality +<br/>and produce bright pixels in the image from infrared +<br/>light reflections on the cornea named as glints. Al- +<br/>though some works suggest the combination of light +<br/>sources and web cams to track the eyes as described in +<br/>Sigut and Sidha (2011), the challenge of low-cost sys- +<br/>tems is to avoid the use of light sources to keep the sys- +<br/>tems as simple as possible; hence, the image quality de- +<br/>creases. High-performance eye tracking systems usu- +<br/>ally combine glints and pupil information to compute +<br/>the gaze position on the screen. Accurate pupil detec- +<br/>tion is not feasible in web cam images, and most works +<br/>on this topic focus on iris center. In order to improve +<br/>accuracy, other elements such as eye corners or head +<br/>position are necessary for gaze estimation applications, +<br/>apart from the estimation of both irises. In the work by +<br/>Ince and Yang (2009), they consider that the horizontal +<br/>and vertical deviation of eye movements through eye- +</td><td>('2592332', 'Jose Javier Bengoechea', 'jose javier bengoechea')<br/>('2595143', 'Juan J. Cerrolaza', 'juan j. cerrolaza')<br/>('2175923', 'Arantxa Villanueva', 'arantxa villanueva')<br/>('1752979', 'Rafael Cabeza', 'rafael cabeza')</td><td></td></tr><tr><td>d0e895a272d684a91c1b1b1af29747f92919d823</td><td>Classification of Mouth Action Units using Local Binary Patterns +<br/><b>The American University in Cairo</b><br/>Department of Computer Science, AUC, AUC +<br/>Avenue, P.O. Box 74 New Cairo 11835, Egypt +<br/><b>The American University in Cairo</b><br/>Department of Computer Science, AUC, AUC +<br/>Avenue, P.O. Box 74 New Cairo 11835, Egypt +</td><td>('3298267', 'Sarah Adel Bargal', 'sarah adel bargal')<br/>('3337337', 'Amr Goneid', 'amr goneid')</td><td>s_bargal@aucegypt.edu +<br/>goneid@aucegypt.edu +</td></tr><tr><td>d082f35534932dfa1b034499fc603f299645862d</td><td>TAMING WILD FACES: WEB-SCALE, OPEN-UNIVERSE FACE IDENTIFICATION IN +<br/>STILL AND VIDEO IMAGERY +<br/>by +<br/><b>B.S. University of Central Florida</b><br/><b>M.S. University of Central Florida</b><br/>A dissertation submitted in partial fulfilment of the requirements +<br/>for the degree of Doctor of Philosophy +<br/>in the Department of Electrical Engineering and Computer Science +<br/><b>in the College of Engineering and Computer Science</b><br/><b>at the University of Central Florida</b><br/>Orlando, Florida +<br/>Spring Term +<br/>2014 +<br/>Major Professor: Mubarak Shah +</td><td>('1873759', 'G. ORTIZ', 'g. ortiz')</td><td></td></tr><tr><td>d03265ea9200a993af857b473c6bf12a095ca178</td><td>Multiple deep convolutional neural +<br/>networks averaging for face +<br/>alignment +<br/>Zhouping Yin +<br/>Downloaded From: http://electronicimaging.spiedigitallibrary.org/ on 05/28/2015 Terms of Use: http://spiedl.org/terms </td><td>('7671296', 'Shaohua Zhang', 'shaohua zhang')<br/>('39584289', 'Hua Yang', 'hua yang')</td><td></td></tr><tr><td>d0ac9913a3b1784f94446db2f1fb4cf3afda151f</td><td>Exploiting Multi-modal Curriculum in Noisy Web Data for +<br/>Large-scale Concept Learning +<br/><b>School of Computer Science, Carnegie Mellon University, PA, USA</b><br/><b>School of Mathematics and Statistics, Xi an Jiaotong University, P. R. China</b></td><td>('1915796', 'Junwei Liang', 'junwei liang')<br/>('38782499', 'Lu Jiang', 'lu jiang')<br/>('1803714', 'Deyu Meng', 'deyu meng')</td><td>{junweil, lujiang, alex}@cs.cmu.edu, dymeng@mail.xjtu.edu.cn. +</td></tr><tr><td>d0471d5907d6557cf081edf4c7c2296c3c221a38</td><td>A Constrained Deep Neural Network for Ordinal Regression +<br/><b>Nanyang Technological University</b><br/>Rolls-Royce Advanced Technology Centre +<br/>50 Nanyang Avenue, Singapore, 639798 +<br/>6 Seletar Aerospace Rise, Singapore, 797575 +</td><td>('47908585', 'Yanzhu Liu', 'yanzhu liu')<br/>('1799918', 'Chi Keong Goh', 'chi keong goh')</td><td>liuy0109@e.ntu.edu.sg, adamskong@ntu.edu.sg +<br/>ChiKeong.Goh@Rolls-Royce.com +</td></tr><tr><td>d0eb3fd1b1750242f3bb39ce9ac27fc8cc7c5af0</td><td></td><td></td><td></td></tr><tr><td>d00c335fbb542bc628642c1db36791eae24e02b7</td><td>Article +<br/>Deep Learning-Based Gaze Detection System for +<br/>Automobile Drivers Using a NIR Camera Sensor +<br/><b>Division of Electronics and Electrical Engineering, Dongguk University, 30 Pildong-ro, 1-gil, Jung-gu</b><br/>Received: 5 January 2018; Accepted: 1 February 2018; Published: 3 February 2018 +</td><td>('8683310', 'Rizwan Ali Naqvi', 'rizwan ali naqvi')<br/>('15668895', 'Muhammad Arsalan', 'muhammad arsalan')<br/>('3407484', 'Ganbayar Batchuluun', 'ganbayar batchuluun')<br/>('40376380', 'Hyo Sik Yoon', 'hyo sik yoon')<br/>('4634733', 'Kang Ryoung Park', 'kang ryoung park')</td><td>Seoul 100-715, Korea; rizwanali@dongguk.edu (R.A.N.); arsal@dongguk.edu (M.A.); +<br/>ganabata87@gmail.com (G.B.); yoonhs@dongguk.edu (H.S.Y.) +<br/>* Correspondence: parkgr@dongguk.edu; Tel.: +82-10-3111-7022; Fax: +82-2-2277-8735 +</td></tr><tr><td>d06c8e3c266fbae4026d122ec9bd6c911fcdf51d</td><td>Role for 2D image generated 3D face models in the rehabilitation of facial palsy +<br/><b>Northumbria University, Newcastle Upon-Tyne NE21XE, UK</b><br/>Published in Healthcare Technology Letters; Received on 4th April 2017; Revised on 7th June 2017; Accepted on 7th June 2017 +<br/>The outcome for patients diagnosed with facial palsy has been shown to be linked to rehabilitation. Dense 3D morphable models have been +<br/>shown within the computer vision to create accurate representations of human faces even from single 2D images. This has the potential +<br/>to provide feedback to both the patient and medical expert dealing with the rehabilitation plan. It is proposed that a framework for the +<br/>creation and measuring of patient facial movement consisting of a hybrid 2D facial landmark fitting technique which shows better +<br/>accuracy in testing than current methods and 3D model fitting. +<br/>1. Introduction: Recent medical studies [1–3] have highlighted +<br/>that patients diagnosed and treated with specific types of facial +<br/>paralysis such as Bell’s palsy have outcomes that are directly +<br/>linked to the rehabilitation provided. While various treatment and +<br/>rehabilitation paths exist dependant on the specifics of the facial +<br/>palsy diagnosis, the aim is to restore a degree of facial muscle +<br/>movement +<br/>[4] completed a +<br/>comprehensive study over 5 years of the rehabilitation process +<br/>and outcomes for 303 facial paralysis patients, the key finding +<br/>was the need for specialised therapy plans tailored via feedback +<br/>for the best patient outcomes. While Banks et al [5] have shown +<br/>that quality qualitative feedback to a clinician is required for the +<br/>best development of rehabilitation plans. +<br/>to the patient. Lindsay et al +<br/>Tracking and providing qualitative feedback on the progress +<br/>of rehabilitation for a patient is an area where the application of +<br/>computer vision and machine learning techniques could prove to +<br/>be highly beneficial. Computer vision methods can provide the +<br/>capability of capturing accurate 3D models of the human face +<br/>these in turn can be leveraged to analyse and measure changes in +<br/>face shape and levels of motion [6]. +<br/>Applying 3D face modelling techniques in an automated +<br/>framework for +<br/>tracking facial palsy rehabilitation progression +<br/>has a number of potential benefits. 3D face models generated +<br/>from a 2D face image can provide a detailed topography of an +<br/>individual human face which can be qualitatively measured for +<br/>change over time by a computer system. Potential benefits of +<br/>such an automated system include providing the clinician +<br/>dealing with a patients rehabilitation to gather regular objective +<br/>feedback on the condition and tailor therapy without always +<br/>needing to physically see the patient or providing continuity of +<br/>care if for instance the clinician changes during the rehabilitation +<br/>period. Patients will have a visual evidence in which to see the +<br/>progress that has been made. It has been indicated that patients +<br/>suffering from facial palsy can also be affected by psychol- +<br/>ogical and social problems the capacity to track rehabilitation pri- +<br/>vately within a comfortable setting like their own home may be +<br/>of benefit. +<br/>Some previous studies [7] have looked at the process of aiding +<br/>diagnosis through the application of computer vision techniques +<br/>these have been limited to 2D imaging which measure on a spare +<br/>set of landmarks. The hypothesis is that 3D face modelling consist- +<br/>ing of thousands of landmarks provides a far richer model of the +<br/>face which in turn can present a more accurate measurement +<br/>system for facial motion. +<br/>In this Letter we propose a framework applicable for accurate +<br/>generation of 3D face models of facial palsy patients from 2D +<br/>images applying state-of-the-art methods and a proposed method +<br/>Healthcare Technology Letters, 2017, Vol. 4, Iss. 4, pp. 145–148 +<br/>doi: 10.1049/htl.2017.0023 +<br/>Fig. 1 2D face alignment of 68 landmarks on a facial image which displays +<br/>asymmetric movement, like that of a patient suffering from facial palsy +<br/>of using geometrical features to track rehabilitation and present +<br/>our conclusions. +<br/>2. Proposed system overview: The accuracy of +<br/>the facial +<br/>representation is a key components of any computer-based system +<br/>which aims to measure facial motion. We suggest that the more +<br/>complex a depiction of the individuals patient facial topography +<br/>the greater the potential +<br/>is for the desired level of accuracy. +<br/>Developing such a system requires a framework of methods to +<br/>build and measure such a model. +<br/>As camera systems which perceive depth within an image are not +<br/>currently common place or require specialist and expensive hard- +<br/>ware initially we require a method for face detection and 2D face +<br/>145 +<br/>This is an open access article published by the IET under the +<br/>Creative Commons Attribution License (http://creativecommons. +<br/>org/licenses/by/3.0/) +</td><td>('12667800', 'Gary Storey', 'gary storey')<br/>('40618413', 'Richard Jiang', 'richard jiang')<br/>('1690116', 'Ahmed Bouridane', 'ahmed bouridane')</td><td>✉ E-mail: gary.storey@northumbria.ac.uk +</td></tr><tr><td>d074b33afd95074d90360095b6ecd8bc4e5bb6a2</td><td>December 11, 2007 +<br/>12:8 WSPC/INSTRUCTION FILE +<br/>bauer-2007-ijhr +<br/>International Journal of Humanoid Robotics +<br/>c(cid:13) World Scientific Publishing Company +<br/>Human-Robot Collaboration: A Survey +<br/><b>Institute of Automatic Control Engineering (LSR</b><br/>Technische Universit¨at M¨unchen +<br/>80290 Munich +<br/>Germany +<br/>Received 01.05.2007 +<br/>Revised 29.09.2007 +<br/>Accepted Day Month Year +<br/>As robots are gradually leaving highly structured factory environments and moving into +<br/>human populated environments, they need to possess more complex cognitive abilities. +<br/>They do not only have to operate efficiently and safely in natural, populated environ- +<br/>ments, but also be able to achieve higher levels of cooperation and communication with +<br/>humans. Human-robot collaboration (HRC) is a research field with a wide range of ap- +<br/>plications, future scenarios, and potentially a high economic impact. HRC is an interdis- +<br/>ciplinary research area comprising classical robotics, cognitive sciences, and psychology. +<br/>This article gives a survey of the state of the art of human-robot collaboration. Es- +<br/>tablished methods for intention estimation, action planning, joint action, and machine +<br/>learning are presented together with existing guidelines to hardware design. This article +<br/>is meant to provide the reader with a good overview of technologies and methods for +<br/>HRC. +<br/>Keywords: Human-robot collaboration; intention estimation; action planning; machine +<br/>learning. +<br/>1. Introduction +<br/>Human-robot Collaboration (HRC) is a wide research field with a high economic +<br/>impact. Robots have already started moving out of laboratory and manufacturing +<br/>environments into more complex human working environments such as homes, of- +<br/>fices, hospitals and even outer space. HRC is already used in elderly care1, space +<br/>applications2, and rescue robotics3. The design of robot behaviour, appearance, +<br/>cognitive, and social skills is highly challenging, and requires interdisciplinary co- +<br/>operation between classical robotics, cognitive sciences, and psychology. Humans as +<br/>nondeterministic factors make cognitive sciences and artificial intelligence important +<br/>research fields in HRC. +<br/>This article refers to human-robot collaboration as opposed to human-robot in- +<br/>teraction (HRI) as these two terms hold different meanings4. Interaction is a more +<br/><b>general term, including collaboration. Interaction determines action on someone</b></td><td>('1749896', 'Dirk Wollherr', 'dirk wollherr')<br/>('1732126', 'Martin Buss', 'martin buss')</td><td>ab@tum.de; dw@tum.de; mb@tum.de +</td></tr><tr><td>d04d5692461d208dd5f079b98082eda887b62323</td><td>Subspace learning with frequency regularizer: its application to face recognition +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition, +<br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/>95 Zhongguancun Donglu, Beijing 100190, China. +</td><td>('1704114', 'Xiangsheng Huang', 'xiangsheng huang')<br/>('34679741', 'Stan Z. Li', 'stan z. li')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1716143', 'Dong Yi', 'dong yi')</td><td>{zlei,dyi,szli}@cbsr.ia.ac.cn, xiangsheng.huang@ia.ac.cn +</td></tr><tr><td>d05513c754966801f26e446db174b7f2595805ba</td><td>Everything is in the Face? Represent Faces with +<br/>Object Bank +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>CAS), Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>School of Computer Science, Carnegie Mellon University, PA 15213, USA</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b></td><td>('1731144', 'Xin Liu', 'xin liu')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1688086', 'Shaoxin Li', 'shaoxin li')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td>{xin.liu, shiguang.shan, shaoxin.li}@vipl.ict.ac.cn, alex@cs.cmu.edu; +</td></tr><tr><td>d0509afe9c2c26fe021889f8efae1d85b519452a</td><td>Visual Psychophysics for Making Face +<br/>Recognition Algorithms More Explainable +<br/><b>University of Notre Dame, Notre Dame, IN, 46556, USA</b><br/><b>Perceptive Automata, Inc</b><br/><b>Harvard University, Cambridge, MA 02138, USA</b></td><td>('3849184', 'Brandon RichardWebster', 'brandon richardwebster')<br/>('40901458', 'So Yon Kwon', 'so yon kwon')<br/>('40896426', 'Christopher Clarizio', 'christopher clarizio')<br/>('2503235', 'Samuel E. Anthony', 'samuel e. anthony')<br/>('2613438', 'Walter J. Scheirer', 'walter j. scheirer')</td><td></td></tr><tr><td>d03baf17dff5177d07d94f05f5791779adf3cd5f</td><td></td><td></td><td></td></tr><tr><td>d0144d76b8b926d22411d388e7a26506519372eb</td><td>Improving Regression Performance with Distributional Losses +</td><td>('29905816', 'Ehsan Imani', 'ehsan imani')</td><td></td></tr><tr><td>d02e27e724f9b9592901ac1f45830341d37140fe</td><td>DA-GAN: Instance-level Image Translation by Deep Attention Generative +<br/>Adversarial Networks +<br/>The State Universtiy of New York at Buffalo +<br/>The State Universtiy of New York at Buffalo +<br/>Microsoft Research +<br/>Microsoft Research +</td><td>('2327045', 'Shuang Ma', 'shuang ma')<br/>('1735257', 'Chang Wen Chen', 'chang wen chen')<br/>('3247966', 'Jianlong Fu', 'jianlong fu')<br/>('1724211', 'Tao Mei', 'tao mei')</td><td>shuangma@buffalo.edu +<br/>chencw@buffalo.edu +<br/>jianf@microsoft.com +<br/>tmei@microsoft.com +</td></tr><tr><td>d02b32b012ffba2baeb80dca78e7857aaeececb0</td><td>Human Pose Estimation: Extension and Application +<br/>Thesis submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/>Master of Science (By Research) +<br/>in +<br/>Computer Science and Engineering +<br/>by +<br/>201002052 +<br/>Center for Visual Information Technology +<br/><b>International Institute of Information Technology</b><br/>Hyderabad - 500 032, INDIA +<br/>September 2016 +</td><td>('50226534', 'Digvijay Singh', 'digvijay singh')</td><td>digvijay.singh@research.iiit.ac.in +</td></tr><tr><td>d0a21f94de312a0ff31657fd103d6b29db823caa</td><td>Facial Expression Analysis +</td><td>('1707876', 'Fernando De la Torre', 'fernando de la torre')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td></td></tr><tr><td>d03e4e938bcbc25aa0feb83d8a0830f9cd3eb3ea</td><td>Face Recognition with Patterns of Oriented +<br/>Edge Magnitudes +<br/>1 Vesalis Sarl, Clermont Ferrand, France +<br/>2 Gipsa-lab, Grenoble INP, France +</td><td>('35083213', 'Ngoc-Son Vu', 'ngoc-son vu')<br/>('1788869', 'Alice Caplier', 'alice caplier')</td><td></td></tr><tr><td>d0d7671c816ed7f37b16be86fa792a1b29ddd79b</td><td>Exploring Semantic Inter-Class Relationships (SIR) +<br/>for Zero-Shot Action Recognition +<br/><b>Institute for Interdisciplinary Information Sciences, Tsinghua University, Beijing, China</b><br/><b>Centre for Quantum Computation and Intelligent Systems, University of Technology Sydney, Sydney, Australia</b><br/><b>School of Computer Science, Carnegie Mellon University, Pittsburgh, USA</b><br/><b>College of Computer Science, Zhejiang University, Zhejiang, China</b></td><td>('2551285', 'Chuang Gan', 'chuang gan')<br/>('2735055', 'Ming Lin', 'ming lin')<br/>('39033919', 'Yi Yang', 'yi yang')<br/>('1755711', 'Yueting Zhuang', 'yueting zhuang')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td>ganchuang1990@gmail.com, linming04@gmail.com, +<br/>yiyang@cs.cmu.edu, yzhuang@zju.edu.cn, alex@cs.cmu.edu +</td></tr><tr><td>d01303062b21cd9ff46d5e3ff78897b8499480de</td><td>Multi-task Learning by Maximizing Statistical Dependence +<br/><b>University of Bath</b><br/><b>University of Bath</b><br/><b>University of Bath</b></td><td>('51013428', 'Youssef A. Mejjati', 'youssef a. mejjati')<br/>('1792288', 'Darren Cosker', 'darren cosker')<br/>('1808255', 'Kwang In Kim', 'kwang in kim')</td><td></td></tr><tr><td>d02c54192dbd0798b43231efe1159d6b4375ad36</td><td>3D Reconstruction and Face Recognition Using Kernel-Based +<br/> ICA and Neural Networks +<br/>Dept. of Electrical Dept. of CSIE Dept. of CSIE +<br/><b>Engineering Chaoyang University Nankai Institute of</b><br/><b>National University of Technology Technology</b></td><td>('1734467', 'Cheng-Jian Lin', 'cheng-jian lin')<br/>('1759040', 'Chi-Yung Lee', 'chi-yung lee')</td><td> of Kaohsiung s9527618@cyut.edu.tw cylee@nkc.edu.tw +<br/>cjlin@nuk.edu.tw +</td></tr><tr><td>d00787e215bd74d32d80a6c115c4789214da5edb</td><td>Faster and Lighter Online +<br/>Sparse Dictionary Learning +<br/>Project report +</td><td>('2714145', 'Jeremias Sulam', 'jeremias sulam')</td><td></td></tr><tr><td>d0f54b72e3a3fe7c0e65d7d5a3b30affb275f4c5</td><td>Towards Universal Representation for Unseen Action Recognition +<br/><b>University of California, Merced</b><br/><b>Open Lab, School of Computing, Newcastle University, UK</b><br/><b>Inception Institute of Arti cial Intelligence (IIAI), Abu Dhabi, UAE</b></td><td>('1749901', 'Yi Zhu', 'yi zhu')<br/>('50363618', 'Yang Long', 'yang long')<br/>('1735787', 'Yu Guan', 'yu guan')<br/>('40799321', 'Ling Shao', 'ling shao')</td><td></td></tr><tr><td>be8c517406528edc47c4ec0222e2a603950c2762</td><td>Harrigan / The new handbook of methods in nonverbal behaviour research 02-harrigan-chap02 Page Proof page 7 +<br/>17.6.2005 +<br/>5:45pm +<br/>B A S I C R E S E A RC H +<br/>M E T H O D S A N D +<br/>P RO C E D U R E S +</td><td></td><td></td></tr><tr><td>beb3fd2da7f8f3b0c3ebceaa2150a0e65736d1a2</td><td>RESEARCH PAPER +<br/>International Journal of Recent Trends in Engineering Vol 1, No. 1, May 2009, +<br/>Adaptive Histogram Equalization and Logarithm +<br/>Transform with Rescaled Low Frequency DCT +<br/>Coefficients for Illumination Normalization +<br/>Department of Computer Science and Engineering +<br/>Amity School of Engineering Technology, 580, Bijwasan, New Delhi-110061, India +<br/><b>Affiliated to Guru Gobind Singh Indraprastha University, Delhi, India</b><br/>illumination normalization. The +<br/>lighting conditions. Most of the +</td><td>('2650871', 'Virendra P. Vishwakarma', 'virendra p. vishwakarma')<br/>('2100294', 'Sujata Pandey', 'sujata pandey')</td><td>Email: vpvishwakarma@aset.amity.edu +</td></tr><tr><td>be86d88ecb4192eaf512f29c461e684eb6c35257</td><td>Automatic Attribute Discovery and +<br/>Characterization from Noisy Web Data +<br/><b>Stony Brook University, Stony Brook NY 11794, USA</b><br/><b>Columbia University, New York NY 10027, USA</b><br/><b>University of California, Berkeley, Berkeley CA 94720, USA</b></td><td>('1685538', 'Tamara L. Berg', 'tamara l. berg')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('9676096', 'Jonathan Shih', 'jonathan shih')</td><td>tlberg@cs.sunysb.edu, +<br/>aberg@cs.columbia.edu, +<br/>jmshih@berkeley.edu. +</td></tr><tr><td>be48b5dcd10ab834cd68d5b2a24187180e2b408f</td><td>FOR PERSONAL USE ONLY +<br/>Constrained Low-rank Learning Using Least +<br/>Squares Based Regularization +</td><td>('2420746', 'Ping Li', 'ping li')<br/>('1720236', 'Jun Yu', 'jun yu')<br/>('48958393', 'Meng Wang', 'meng wang')<br/>('1763785', 'Luming Zhang', 'luming zhang')<br/>('1724421', 'Deng Cai', 'deng cai')<br/>('50080046', 'Xuelong Li', 'xuelong li')</td><td></td></tr><tr><td>beb49072f5ba79ed24750108c593e8982715498e</td><td>STUDENT, PROF, COLLABORATOR: BMVC AUTHOR GUIDELINES +<br/>GeneGAN: Learning Object Transfiguration +<br/>and Attribute Subspace from Unpaired Data +<br/>1 Megvii Inc. +<br/>Beijing, China +<br/>2 Department of Information Science, +<br/>School of Mathematical Sciences, +<br/><b>Peking University</b><br/>Beijing, China +</td><td>('35132667', 'Shuchang Zhou', 'shuchang zhou')<br/>('14002400', 'Taihong Xiao', 'taihong xiao')<br/>('1698559', 'Yi Yang', 'yi yang')<br/>('7841666', 'Dieqiao Feng', 'dieqiao feng')<br/>('8159691', 'Qinyao He', 'qinyao he')<br/>('2416953', 'Weiran He', 'weiran he')</td><td>shuchang.zhou@gmail.com +<br/>xiaotaihong@pku.edu.cn +<br/>yangyi@megvii.com +<br/>fdq@megvii.com +<br/>hqy@megvii.com +<br/>hwr@megvii.com +</td></tr><tr><td>be4a20113bc204019ea79c6557a0bece23da1121</td><td>DeepCache: Principled Cache for Mobile Deep Vision +<br/>We present DeepCache, a principled cache design for deep learning +<br/>inference in continuous mobile vision. DeepCache benefits model +<br/>execution efficiency by exploiting temporal locality in input video +<br/>streams. It addresses a key challenge raised by mobile vision: the +<br/>cache must operate under video scene variation, while trading off +<br/>among cacheability, overhead, and loss in model accuracy. At the +<br/>input of a model, DeepCache discovers video temporal locality by ex- +<br/>ploiting the video’s internal structure, for which it borrows proven +<br/>heuristics from video compression; into the model, DeepCache prop- +<br/>agates regions of reusable results by exploiting the model’s internal +<br/>structure. Notably, DeepCache eschews applying video heuristics to +<br/>model internals which are not pixels but high-dimensional, difficult- +<br/>to-interpret data. +<br/>Our implementation of DeepCache works with unmodified deep +<br/>learning models, requires zero developer’s manual effort, and is +<br/>therefore immediately deployable on off-the-shelf mobile devices. +<br/>Our experiments show that DeepCache saves inference execution +<br/>time by 18% on average and up to 47%. DeepCache reduces system +<br/>energy consumption by 20% on average. +<br/>CCS Concepts: • Human-centered computing → Ubiquitous +<br/>and mobile computing; • Computing methodologies → Com- +<br/>puter vision tasks; +<br/>Additional Key Words and Phrases: Deep Learning; Mobile Vision; +<br/>Cache +<br/>INTRODUCTION +<br/>With ubiquitous cameras on mobile and wearable devices, +<br/>continuous mobile vision emerges to enable a variety of com- +<br/><b>pelling applications, including cognitive assistance [29], life</b><br/>style monitoring [61], and street navigation [27]. To support +<br/>continuous mobile vision, Convolutional Neural Network +<br/>2018. XXXX-XXXX/2018/9-ART $15.00 +<br/>https://doi.org/10.1145/3241539.3241563 +<br/>Fig. 1. The overview of DeepCache. +<br/>(CNN) is recognized as the state-of-the-art algorithm: a soft- +<br/>ware runtime, called deep learning engine, ingests a continu- +<br/>ous stream of video images1; for each input frame the engine +<br/>executes a CNN model as a cascade of layers, produces in- +<br/>termediate results called feature maps, and outputs inference +<br/>results. Such CNN executions are known for their high time +<br/>and space complexity, stressing resource-constrained mobile +<br/>devices. Although CNN execution can be offloaded to the +<br/>cloud [2, 34], it becomes increasingly compelling to execute +<br/>CNNs on device [27, 44, 52], which ensures fast inference, pre- +<br/>serves user privacy, and remains unaffected by poor Internet +<br/>connectivity. +<br/>To afford costly CNN on resource-constrained mobile/wear- +<br/>able devices, we set to exploit a mobile video stream’s tempo- +<br/>ral locality, i.e., rich information redundancy among consec- +<br/>utive video frames [27, 51, 52]. Accordingly, a deep learning +<br/>engine can cache results when it executes CNN over a mo- +<br/>bile video, by using input frame contents as cache keys and +<br/>inference results as cache values. Such caching is expected +<br/>to reduce the engine’s resource demand significantly. +<br/>Towards effective caching and result reusing, we face two +<br/>major challenges. 1) Reusable results lookup: Classic caches, +<br/>e.g., the web browser cache, look up cached values (e.g., web +<br/>pages) based on key equivalence (e.g., identical URLs). This +<br/>does not apply to a CNN cache: its keys, i.e., mobile video +<br/>contents, often undergo moderate scene variation over time. +<br/>The variation is caused by environmental changes such as +<br/>1We refer to them as a mobile video stream in the remainder of the paper. +<br/>, Vol. 1, No. 1, Article . Publication date: September 2018. +</td><td>('2529558', 'Mengwei Xu', 'mengwei xu')<br/>('46694806', 'Mengze Zhu', 'mengze zhu')<br/>('3180228', 'Yunxin Liu', 'yunxin liu')<br/>('1774176', 'Felix Xiaozhu Lin', 'felix xiaozhu lin')<br/>('8016688', 'Xuanzhe Liu', 'xuanzhe liu')<br/>('8016688', 'Xuanzhe Liu', 'xuanzhe liu')<br/>('2529558', 'Mengwei Xu', 'mengwei xu')</td><td>xumengwei@pku.edu.cn; Mengze Zhu, Peking University, MoE, Beijing, +<br/>China, zhumz@pku.edu.cn; Yunxin Liu, Microsoft Research, Beijing, China, +<br/>yunxin.liu@microsoft.com; Felix Xiaozhu Lin, Purdue ECE, West Lafayette, +<br/>Indiana, USA, xzl@purdue.edu; Xuanzhe Liu, Peking University, MoE, Bei- +<br/>jing, China, xzl@pku.edu.cn. +</td></tr><tr><td>becd5fd62f6301226b8e150e1a5ec3180f748ff8</td><td>Robust and Practical Face Recognition via +<br/>Structured Sparsity +<br/>1Advanced Digital Sciences Center, Singapore +<br/>2 Microsoft Research Asia, Beijing, China +<br/><b>University of Illinois at Urbana-Champaign</b></td><td>('2370507', 'Kui Jia', 'kui jia')<br/>('1926757', 'Tsung-Han Chan', 'tsung-han chan')<br/>('1700297', 'Yi Ma', 'yi ma')</td><td></td></tr><tr><td>be437b53a376085b01ebd0f4c7c6c9e40a4b1a75</td><td>ISSN (Online) 2321 – 2004 +<br/>ISSN (Print) 2321 – 5526 +<br/> INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN ELECTRICAL, ELECTRONICS, INSTRUMENTATION AND CONTROL ENGINEERING +<br/> Vol. 4, Issue 5, May 2016 +<br/>IJIREEICE +<br/>Face Recognition and Retrieval Using Cross +<br/>Age Reference Coding +<br/> BE, DSCE, Bangalore1 +<br/>Assistant Professor, DSCE, Bangalore2 +</td><td>('4427719', 'Chandrakala', 'chandrakala')</td><td></td></tr><tr><td>bebb8a97b2940a4e5f6e9d3caf6d71af21585eda</td><td>Mapping Emotional Status to Facial Expressions +<br/><b>Tsinghua University</b><br/>Beijing 100084, P. R. China +</td><td>('3165307', 'Yangzhou Du', 'yangzhou du')<br/>('2693354', 'Xueyin Lin', 'xueyin lin')</td><td>dyz99@mails.tsinghua.edu.cn; lxy-dcs@tsinghua.edu.cn +</td></tr><tr><td>be07f2950771d318a78d2b64de340394f7d6b717</td><td>See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/290192867 +<br/>3D HMM-based Facial Expression Recognition +<br/>using Histogram of Oriented Optical Flow +<br/>ARTICLE in SYNTHESIS LECTURES ON ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING · DECEMBER 2015 +<br/>DOI: 10.14738/tmlai.36.1661 +<br/>READS +<br/>12 +<br/>3 AUTHORS, INCLUDING: +<br/>Sheng Kung +<br/><b>Oakland University</b><br/>Djamel Bouchaffra +<br/><b>Institute of Electrical and Electronics Engineers</b><br/>1 PUBLICATION 0 CITATIONS +<br/>57 PUBLICATIONS 402 CITATIONS +<br/>SEE PROFILE +<br/>SEE PROFILE +<br/>All in-text references underlined in blue are linked to publications on ResearchGate, +<br/>letting you access and read them immediately. +<br/>Available from: Djamel Bouchaffra +<br/>Retrieved on: 11 February 2016 +</td><td></td><td></td></tr><tr><td>be4f7679797777f2bc1fd6aad8af67cce5e5ce87</td><td>Interestingness Prediction +<br/>by Robust Learning to Rank(cid:2) +<br/><b>School of EECS, Queen Mary University of London, UK</b><br/><b>School of Mathematical Sciences, Peking University, China</b></td><td>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('1697755', 'Timothy M. Hospedales', 'timothy m. hospedales')<br/>('1700927', 'Tao Xiang', 'tao xiang')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('1746280', 'Yuan Yao', 'yuan yao')</td><td>{y.fu,t.hospedales,t.xiang,s.gong}@qmul.ac.uk, yuany@math.pku.edu.cn +</td></tr><tr><td>beb4546ae95f79235c5f3c0e9cc301b5d6fc9374</td><td>A Modular Approach to Facial Expression Recognition +<br/><b>Cognitive Arti cial Intelligence, Utrecht University, Heidelberglaan 6, 3584 CD, Utrecht</b><br/><b>Intelligent Systems Group, Utrecht University, Padualaan 14, 3508 TB, Utrecht</b></td><td>('31822812', 'Michal Sindlar', 'michal sindlar')<br/>('1727399', 'Marco Wiering', 'marco wiering')</td><td>sindlar@phil.uu.nl +<br/>marco@cs.uu.nl +</td></tr><tr><td>be28ed1be084385f5d389db25fd7f56cd2d7f7bf</td><td>Exploring Computation-Communication Tradeoffs +<br/>in Camera Systems +<br/><b>Paul G. Allen School of Computer Science and Engineering, University of Washington</b><br/><b>University of Washington</b></td><td>('19170117', 'Amrita Mazumdar', 'amrita mazumdar')<br/>('47108160', 'Thierry Moreau', 'thierry moreau')<br/>('37270394', 'Meghan Cowan', 'meghan cowan')<br/>('1698528', 'Armin Alaghi', 'armin alaghi')<br/>('1717411', 'Luis Ceze', 'luis ceze')<br/>('1723213', 'Mark Oskin', 'mark oskin')<br/>('46829693', 'Visvesh Sathe', 'visvesh sathe')</td><td>{amrita,moreau,cowanmeg}@cs.washington.edu, sungk9@uw.edu, {armin,luisceze,oskin}@cs.washington.edu, sathe@uw.edu +</td></tr><tr><td>bebea83479a8e1988a7da32584e37bfc463d32d4</td><td>Discovery of Latent 3D Keypoints via +<br/>End-to-end Geometric Reasoning +<br/>Google AI +</td><td>('37016781', 'Supasorn Suwajanakorn', 'supasorn suwajanakorn')<br/>('2704494', 'Jonathan Tompson', 'jonathan tompson')</td><td>{supasorn, snavely, tompson, mnorouzi}@google.com +</td></tr><tr><td>bed06e7ff0b510b4a1762283640b4233de4c18e0</td><td>Bachelor Project +<br/>Czech +<br/>Technical +<br/><b>University</b><br/>in Prague +<br/>F3 +<br/>Faculty of Electrical Engineering +<br/>Department of Cybernetics +<br/>Face Interpretation Problems on Low +<br/>Quality Images +<br/>Supervisor: Ing. Jan Čech, Ph.D +<br/>May 2018 +</td><td></td><td></td></tr><tr><td>bec31269632c17206deb90cd74367d1e6586f75f</td><td>Large-scale Datasets: Faces with Partial +<br/>Occlusions and Pose Variations in the Wild +<br/><b>Wayne State University</b><br/>Detroit, MI, USA 48120 +</td><td>('2489629', 'Zeyad Hailat', 'zeyad hailat')<br/>('35265528', 'Xuewen Chen', 'xuewen chen')</td><td>Email: ∗tarik alafif@wayne.edu, †zmhailat@wayne.edu, ‡melih.aslan@wayne.edu, §xuewen.chen@wayne.edu +</td></tr><tr><td>be5276e9744c4445fe5b12b785650e8f173f56ff</td><td>Spatio-temporal VLAD Encoding for +<br/>Human Action Recognition in Videos +<br/><b>University of Trento, Italy</b><br/><b>University Politehnica of Bucharest, Romania</b><br/><b>University of Tokyo, Japan</b></td><td>('3429470', 'Ionut C. Duta', 'ionut c. duta')<br/>('1796198', 'Bogdan Ionescu', 'bogdan ionescu')<br/>('1712839', 'Kiyoharu Aizawa', 'kiyoharu aizawa')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')</td><td>{ionutcosmin.duta, niculae.sebe}@unitn.it +<br/>bionescu@imag.pub.ro +<br/>aizawa@hal.t.u-tokyo.ac.jp +</td></tr><tr><td>be57d2aaab615ec8bc1dd2dba8bee41a4d038b85</td><td>Automatic Analysis of Naturalistic Hand-Over-Face Gestures +<br/><b>University of Cambridge</b><br/>One of the main factors that limit the accuracy of facial analysis systems is hand occlusion. As the face +<br/>becomes occluded, facial features are lost, corrupted, or erroneously detected. Hand-over-face occlusions are +<br/>considered not only very common but also very challenging to handle. However, there is empirical evidence +<br/>that some of these hand-over-face gestures serve as cues for recognition of cognitive mental states. In this +<br/>article, we present an analysis of automatic detection and classification of hand-over-face gestures. We detect +<br/>hand-over-face occlusions and classify hand-over-face gesture descriptors in videos of natural expressions +<br/>using multi-modal fusion of different state-of-the-art spatial and spatio-temporal features. We show experi- +<br/>mentally that we can successfully detect face occlusions with an accuracy of 83%. We also demonstrate that +<br/>we can classify gesture descriptors (hand shape, hand action, and facial region occluded) significantly better +<br/>than a na¨ıve baseline. Our detailed quantitative analysis sheds some light on the challenges of automatic +<br/>classification of hand-over-face gestures in natural expressions. +<br/>Categories and Subject Descriptors: I.2.10 [Vision and Scene Understanding]: Video Analysis +<br/>General Terms: Affective Computing, Body Expressions +<br/>Additional Key Words and Phrases: Hand-over-face occlusions, face touches, hand gestures, facial landmarks, +<br/>histograms of oriented gradient, space-time interest points +<br/>ACM Reference Format: +<br/>over-face gestures. ACM Trans. Interact. Intell. Syst. 6, 2, Article 19 (July 2016), 18 pages. +<br/>DOI: http://dx.doi.org/10.1145/2946796 +<br/>1. INTRODUCTION +<br/>Over the past few years, there has been an increasing interest in machine under- +<br/>standing and recognition of people’s affective and cognitive mental states, especially +<br/>based on facial expression analysis. One of the major factors that limits the accuracy +<br/>of facial analysis systems is hand occlusion. People often hold their hands near their +<br/>faces as a gesture in natural conversation. As many facial analysis systems are based +<br/>on geometric or appearance based facial features, such features are lost, corrupted, +<br/>or erroneously detected during occlusion. This results in an incorrect analysis of the +<br/>person’s facial expression. Although face touches are very common, they are under- +<br/>researched, mostly because segmenting of the hand on the face is very challenging, +<br/>as face and hand usually have similar colour and texture. Detection of hand-over-face +<br/>The research leading to these results received partial funding from the European Community’s Seventh +<br/>Framework Programme (FP7/2007-2013) under Grant No. 289021 (ASC-Inclusion). We also thank Yousef +<br/>Jameel and Qualcomm for providing funding as well. +<br/>Authors’ address: The Computer Laboratory, 15 JJ Thomson Avenue, Cambridge CB3 0FD, United Kingdom; +<br/>Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted +<br/>without fee provided that copies are not made or distributed for profit or commercial advantage and that +<br/>copies show this notice on the first page or initial screen of a display along with the full citation. Copyrights for +</td><td>('2022940', 'Marwa Mahmoud', 'marwa mahmoud')<br/>('39626495', 'Peter Robinson', 'peter robinson')<br/>('2022940', 'Marwa Mahmoud', 'marwa mahmoud')<br/>('39626495', 'Peter Robinson', 'peter robinson')</td><td>emails: {Marwa.Mahmoud, Tadas.Baltrusaitis, Peter.Robinson}@cl.cam.ac.uk. +</td></tr><tr><td>be4f18e25b06f430e2de0cc8fddcac8585b00beb</td><td>STUDENT, PROF, COLLABORATOR: BMVC AUTHOR GUIDELINES +<br/>A New Face Recognition Algorithm based on +<br/>Dictionary Learning for a Single Training +<br/>Sample per Person +<br/>Ian Wassell +<br/>Computer Laboratory, +<br/><b>University of Cambridge</b></td><td>('1681842', 'Yang Liu', 'yang liu')</td><td>yl504@cam.ac.uk +<br/>ijw24@cam.ac.uk +</td></tr><tr><td>bef503cdfe38e7940141f70524ee8df4afd4f954</td><td></td><td></td><td></td></tr><tr><td>beab10d1bdb0c95b2f880a81a747f6dd17caa9c2</td><td>DeepDeblur: Fast one-step blurry face images restoration +<br/>Tsinghua Unversity +</td><td>('2766905', 'Lingxiao Wang', 'lingxiao wang')<br/>('2112160', 'Yali Li', 'yali li')<br/>('1678689', 'Shengjin Wang', 'shengjin wang')</td><td>wlx16@mails.tsinghua.edu.cn, liyali@ocrserv.ee.tsinghua.edu.cn, wgsgj@tsinghua.edu.cn +</td></tr><tr><td>b331ca23aed90394c05f06701f90afd550131fe3</td><td>Zhou et al. EURASIP Journal on Image and Video Processing (2018) 2018:49 +<br/>https://doi.org/10.1186/s13640-018-0287-5 +<br/>EURASIP Journal on Image +<br/>and Video Processing +<br/>R ES EAR CH +<br/>Double regularized matrix factorization for +<br/>image classification and clustering +<br/>Open Access +</td><td>('39147685', 'Wei Zhou', 'wei zhou')<br/>('7513726', 'Chengdong Wu', 'chengdong wu')<br/>('46583983', 'Jianzhong Wang', 'jianzhong wang')<br/>('9305845', 'Xiaosheng Yu', 'xiaosheng yu')<br/>('50130800', 'Yugen Yi', 'yugen yi')</td><td></td></tr><tr><td>b3b532e8ea6304446b1623e83b0b9a96968f926c</td><td>Joint Network based Attention for Action Recognition +<br/>1 National Engineering Laboratory for Video Technology, School of EE&CS, +<br/><b>Peking University, Beijing, China</b><br/>2 Cooperative Medianet Innovation Center, China +<br/>3 School of Information and Electronics, +<br/><b>Beijing Institute of Technology, Beijing, China</b></td><td>('38179026', 'Yemin Shi', 'yemin shi')<br/>('1705972', 'Yonghong Tian', 'yonghong tian')<br/>('5765799', 'Yaowei Wang', 'yaowei wang')<br/>('34097174', 'Tiejun Huang', 'tiejun huang')</td><td></td></tr><tr><td>b37f57edab685dba5c23de00e4fa032a3a6e8841</td><td>Towards Social Interaction Detection in Egocentric Photo-streams +<br/><b>University of Barcelona and Computer Vision Centre, Barcelona, Spain</b><br/>Recent advances in wearable camera technology have +<br/>led to novel applications in the field of Preventive Medicine. +<br/>For some of them, such as cognitive training of elderly peo- +<br/>ple by digital memories and detection of unhealthy social +<br/>trends associated to neuropsychological disorders, social in- +<br/>teraction are of special interest. Our purpose is to address +<br/>this problem in the domain of egocentric photo-streams cap- +<br/>tured by a low temporal resolution wearable camera (2fpm). +<br/>These cameras are suited for collecting visual information +<br/>for long period of time, as required by the aforementioned +<br/>applications. The major difficulties to be handled in this +<br/>context are the sparsity of observations as well as the unpre- +<br/>dictability of camera motion and attention orientation due +<br/>to the fact that the camera is worn as part of clothing (see +<br/>Fig. 1). Inspired by the theory of F-formation which is a +<br/>pattern that people tend to follow when interacting [5], our +<br/>proposed approach consists of three steps: multi-faces as- +<br/>signment, social signals extraction and interaction detection +<br/>of the individuals with the camera wearer (see Fig. 2). +<br/>1. Multi-face Assignment +<br/>While person detection and tracking in classical videos +<br/>have been active research areas for a long time, the problem +<br/>of people assignment in low temporal resolution egocen- +<br/>tric photo-streams is still unexplored. To address such an +<br/>issue, we proposed a novel method for multi-face assign- +<br/>ment in egocentric photo-streams, we called extended-Bag- +<br/>of-Tracklets (eBoT) [2]. This approach basically consists +<br/>of 4 major sequential modules: seed and tracklet gener- +<br/>ation, grouping tracklets into eBoT, prototypes extraction +<br/>and occlusion treatment. Prior to any computation, first, a +<br/>temporal segmentation algorithm [6] is applied to extract +<br/>segments characterized by similar visual properties. Later +<br/>on, a face detector is applied on all the frames of a seg- +<br/>ment to detect visible faces on them [8]. Based on the ratio +<br/>between the number of frames with detected faces and the +<br/>total number of frames of the segment, we extract segments +<br/>containing trackable persons. The next steps are applied on +<br/>these extracted segments, hereafter referred to as sequences. +<br/>Figure 1. Example of social interaction (first row) and non-social +<br/>interaction (second row) in egocentric photo-streams. +<br/>• Seed and tracklet generation: The set of collected +<br/>bounding boxes that surround the face of each per- +<br/>son throughout the sequence, are called seeds. For +<br/>each seed, a set of correspondences to it is generated +<br/>along the sequence by propagating the seed forward +<br/>and backward employing the deep-matching technique +<br/>[7] that lead to form a tracklet. To propagate a seed +<br/>found in a frame, in all the frames of the sequence, the +<br/>region of the frames most similar to the seed is found +<br/>as the one having the highest deep-matching score. +<br/>• Grouping tracklets into Bag-of-tracklets (eBoT): +<br/>Assuming that tracklets generated by seeds belong- +<br/>ing to the same person in a sequence, are likely to +<br/>be similar to each other, we group them into a set of +<br/>non-overlapping eBoTs. Since seeds corresponding to +<br/>false positive detections generate unreliable tracklets +<br/>and unreliable eBoTs, we defined a measure based on +<br/>the density of the eBoTs to exclude unreliable eBoTs. +<br/>• Prototypes extraction: A prototype extracted from an +<br/>eBoT, should best represent all tracklets in the eBoT, +<br/>and therefore, it should best localize a person’s face in +<br/>each frame. As the prototype frame, the frame whose +<br/>bounding box has the biggest intersection with the rest +<br/>of the tracklets in that frame is chosen. +<br/>• Occlusion treatment: Estimation of occluded frames +<br/>is a very helpful feature since it allows us to exclude +<br/>occluded frames which do not convey many informa- +<br/>tion from final prototypes. To this goal, we define a +<br/>frame confidence measure to assign a confidence value +</td><td>('2084534', 'Maedeh Aghaei', 'maedeh aghaei')<br/>('2837527', 'Mariella Dimiccoli', 'mariella dimiccoli')<br/>('1724155', 'Petia Radeva', 'petia radeva')</td><td>aghaei.maya@gmail.com +</td></tr><tr><td>b3154d981eca98416074538e091778cbc031ca29</td><td>Pedestrian Attribute Analysis +<br/>Using a Top-View Camera in a Public Space +<br/><b>The University of Tokyo</b><br/>7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan +<br/><b>School of Electrical and Computer Engineering, Cornell University</b><br/>116 Ward Hall, Ithaca, NY 14853, USA +<br/>3 JSPS Postdoctoral Fellow for Research Abroad +</td><td>('2759239', 'Toshihiko Yamasaki', 'toshihiko yamasaki')<br/>('21152852', 'Tomoaki Matsunami', 'tomoaki matsunami')</td><td>{yamasaki,matsunami}@hal.t.u-tokyo.ac.jp +</td></tr><tr><td>b3cb91a08be4117d6efe57251061b62417867de9</td><td>T. Swearingen and A. Ross. "A label propagation approach for predicting missing biographic labels in +<br/>A Label Propagation Approach for +<br/>Predicting Missing Biographic Labels +<br/>in Face-Based Biometric Records +</td><td>('3153117', 'Thomas Swearingen', 'thomas swearingen')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td></td></tr><tr><td>b340f275518aa5dd2c3663eed951045a5b8b0ab1</td><td>Visual Inference of Human Emotion and Behaviour +<br/>Dept of Computer Science +<br/><b>Queen Mary College, London</b><br/>Dept of Computer Science +<br/><b>Queen Mary College, London</b><br/>Dept of Computer Science +<br/><b>Queen Mary College, London</b><br/>England, UK +<br/>England, UK +<br/>England, UK +</td><td>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('1700927', 'Tao Xiang', 'tao xiang')</td><td>sgg@dcs.qmul.ac.uk +<br/>cfshan@dcs.qmul.ac.uk +<br/>txiang@dcs.qmul.ac.uk +</td></tr><tr><td>b3200539538eca54a85223bf0ec4f3ed132d0493</td><td>Action Anticipation with RBF Kernelized +<br/>Feature Mapping RNN +<br/>Hartley[0000−0002−5005−0191] +<br/><b>The Australian National University, Australia</b></td><td>('11519650', 'Yuge Shi', 'yuge shi')</td><td></td></tr><tr><td>b3b467961ba66264bb73ffe00b1830d7874ae8ce</td><td>Finding Tiny Faces +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Figure 1: We describe a detector that can find around 800 faces out of the reportedly 1000 present, by making use of novel +<br/>characterizations of scale, resolution, and context to find small objects. Detector confidence is given by the colorbar on the +<br/>right: can you confidently identify errors? +</td><td>('2894848', 'Peiyun Hu', 'peiyun hu')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td>{peiyunh,deva}@cs.cmu.edu +</td></tr><tr><td>b3ba7ab6de023a0d58c741d6abfa3eae67227caf</td><td>Zero-Shot Activity Recognition with Verb Attribute Induction +<br/>Paul G. Allen School of Computer Science & Engineering +<br/><b>University of Washington</b><br/>Seattle, WA 98195, USA +</td><td>('2545335', 'Rowan Zellers', 'rowan zellers')<br/>('1699545', 'Yejin Choi', 'yejin choi')</td><td>{rowanz,yejin}@cs.washington.edu +</td></tr><tr><td>b375db63742f8a67c2a7d663f23774aedccc84e5</td><td>Brain-inspired Classroom Occupancy +<br/>Monitoring on a Low-Power Mobile Platform +<br/><b>Electronic and Information Engineering, University of Bologna, Italy</b><br/>†Integrated Systems Laboratory, ETH Zurich, Switzerland +</td><td>('1721381', 'Francesco Conti', 'francesco conti')<br/>('1785226', 'Antonio Pullini', 'antonio pullini')<br/>('1710649', 'Luca Benini', 'luca benini')</td><td>f.conti@unibo.it,{pullinia,lbenini}@iis.ee.ethz.ch +</td></tr><tr><td>b3330adb131fb4b6ebbfacce56f1aec2a61e0869</td><td>Emotion recognition using facial images +<br/>School of Electrical and Electronics Engineering +<br/>Department of Electronics and Communication Engineering +<br/><b>SASTRA University, Thanjavur, Tamil Nadu, India</b></td><td>('9365696', 'Siva sankari', 'siva sankari')</td><td> ramya.ece.sk@gmail.com, siva.ece.ds@gmail.com, knr@ece.sastra.edu +</td></tr><tr><td>b3c60b642a1c64699ed069e3740a0edeabf1922c</td><td>Max-Margin Object Detection +</td><td>('29250541', 'Davis E. King', 'davis e. king')</td><td>davis@dlib.net +</td></tr><tr><td>b3f3d6be11ace907c804c2d916830c85643e468d</td><td><b>University of Toulouse</b><br/><b>University of Toulouse II Le Mirail</b><br/>PhD in computer sciences / artificial intelligence +<br/>A Logical Framework for +<br/>Trust-Related Emotions: +<br/>Formal and Behavioral Results +<br/>by +<br/>Co-supervisors: +<br/>Toulouse, September 2010 +</td><td>('1759342', 'Manh Hung NGUYEN', 'manh hung nguyen')<br/>('3107309', 'Jean-François BONNEFON', 'jean-françois bonnefon')<br/>('1733042', 'Dominique LONGIN', 'dominique longin')</td><td></td></tr><tr><td>b3f7c772acc8bc42291e09f7a2b081024a172564</td><td> www.ijmer.com Vol. 3, Issue. 5, Sep - Oct. 2013 pp-3225-3230 ISSN: 2249-6645 +<br/>International Journal of Modern Engineering Research (IJMER) +<br/>A novel approach for performance parameter estimation of face +<br/>recognition based on clustering, shape and corner detection +<br/><b></b><br/> +</td><td>('1904292', 'Prashant Jain', 'prashant jain')</td><td></td></tr><tr><td>b3c398da38d529b907b0bac7ec586c81b851708f</td><td>Face Recognition under Varying Lighting Conditions Using Self Quotient +<br/>Image +<br/><b>Institute of Automation, Chinese Academy of</b><br/>Sciences, Beijing, 100080, China, +</td><td>('29948255', 'Haitao Wang', 'haitao wang')<br/>('1744302', 'Yangsheng Wang', 'yangsheng wang')</td><td>Email: {htwang,wys}@nlpr.ia.ac.cn +</td></tr><tr><td>b32cf547a764a4efa475e9c99a72a5db36eeced6</td><td>UvA-DARE (Digital Academic Repository) +<br/>Mimicry of ingroup and outgroup emotional expressions +<br/>Sachisthal, M.S.M.; Sauter, D.A.; Fischer, A.H. +<br/>Published in: +<br/>Comprehensive Results in Social Psychology +<br/>DOI: +<br/>10.1080/23743603.2017.1298355 +<br/>Link to publication +<br/>Citation for published version (APA): +<br/>Sachisthal, M. S. M., Sauter, D. A., & Fischer, A. H. (2016). Mimicry of ingroup and outgroup emotional +<br/>expressions. Comprehensive Results in Social Psychology, 1(1-3), 86-105. DOI: +<br/>10.1080/23743603.2017.1298355 +<br/>General rights +<br/>It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), +<br/>other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). +<br/>Disclaimer/Complaints regulations +<br/>If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating +<br/>your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask +<br/><b>the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam</b><br/>The Netherlands. You will be contacted as soon as possible. +<br/>Download date: 08 Aug 2018 +<br/><b>UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl</b></td><td></td><td></td></tr><tr><td>b3658514a0729694d86a8b89c875a66cde20480c</td><td>Improving the Robustness of Subspace Learning +<br/>Techniques for Facial Expression Recognition +<br/><b>Aristotle University of Thessaloniki</b><br/>Box 451, 54124 Thessaloniki, Greece +</td><td>('2342345', 'Dimitris Bolis', 'dimitris bolis')<br/>('2447585', 'Anastasios Maronidis', 'anastasios maronidis')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>email: {mpolis, amaronidis, tefas, pitas}@aiia.csd.auth.gr (cid:63) +</td></tr><tr><td>b3b4a7e29b9186e00d2948a1d706ee1605fe5811</td><td>Paper +<br/>Image Preprocessing +<br/>for Illumination Invariant Face +<br/>Verification +<br/><b>Institute of Radioelectronics, Warsaw University of Technology, Warsaw, Poland</b></td><td>('3031283', 'Mariusz Leszczyński', 'mariusz leszczyński')</td><td></td></tr><tr><td>b32631f456397462b3530757f3a73a2ccc362342</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>3069 +</td><td></td><td></td></tr><tr><td>b33e8db8ccabdfc49211e46d78d09b14557d4cba</td><td>Face Expression Recognition and Analysis: +<br/>1 +<br/>The State of the Art +<br/><b>College of Computing, Georgia Institute of Technology</b></td><td>('3115428', 'Vinay Bettadapura', 'vinay bettadapura')</td><td>vinay@gatech.edu +</td></tr><tr><td>b3afa234996f44852317af382b98f5f557cab25a</td><td></td><td></td><td></td></tr><tr><td>df90850f1c153bfab691b985bfe536a5544e438b</td><td>FACE TRACKING ALGORITHM ROBUST TO POSE, +<br/>ILLUMINATION AND FACE EXPRESSION CHANGES: A 3D +<br/>PARAMETRIC MODEL APPROACH +<br/><b></b><br/>via Bramante 65 - 26013, Crema (CR), Italy +<br/>Luigi Arnone, Fabrizio Beverina +<br/>STMicroelectronics - Advanced System Technology Group +<br/>via Olivetti 5 - 20041, Agrate Brianza, Italy +<br/>Keywords: +<br/>Face tracking, expression changes, FACS, illumination changes. +</td><td>('3330245', 'Marco Anisetti', 'marco anisetti')<br/>('2061298', 'Valerio Bellandi', 'valerio bellandi')</td><td></td></tr><tr><td>df8da144a695269e159fb0120bf5355a558f4b02</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>International Conference on Recent Trends in engineering & Technology - 2013(ICRTET'2013) +<br/>Face Recognition using PCA and Eigen Face +<br/>Approach +<br/>ME EXTC [VLSI & Embedded System] +<br/>Sinhgad Academy of Engineering +<br/>EXTC Department +<br/>Pune, India +</td><td></td><td></td></tr><tr><td>dfd934ae448a1b8947d404b01303951b79b13801</td><td>Christopher A. Longmore +<br/><b>University of Plymouth, UK</b><br/><b>Bournemouth University, UK</b><br/>Andrew W. Young +<br/><b>University of York, UK</b><br/>The importance of internal facial features in learning new +<br/>faces +<br/>Running head: FACIAL FEATURES IN LEARNING NEW FACES +<br/>Address of correspondence: +<br/>Chris Longmore +<br/>School of Psychology +<br/>Faculty of Health and Human Sciences +<br/><b>Plymouth University</b><br/>Drake Circus +<br/>Plymouth +<br/>PL4 8AA +<br/>Tel: +44 (0)1752 584890 +<br/>Fax: +44 (0)1752 584808 +</td><td>('39557512', 'Chang Hong Liu', 'chang hong liu')</td><td>Email: chris.longmore@plymouth.ac.uk +</td></tr><tr><td>df577a89830be69c1bfb196e925df3055cafc0ed</td><td>Shift: A Zero FLOP, Zero Parameter Alternative to Spatial Convolutions +<br/>UC Berkeley +</td><td>('3130257', 'Bichen Wu', 'bichen wu')<br/>('40417702', 'Alvin Wan', 'alvin wan')<br/>('27577617', 'Xiangyu Yue', 'xiangyu yue')<br/>('1755487', 'Sicheng Zhao', 'sicheng zhao')<br/>('30096597', 'Noah Golmant', 'noah golmant')<br/>('3647010', 'Amir Gholaminejad', 'amir gholaminejad')<br/>('30503077', 'Joseph Gonzalez', 'joseph gonzalez')<br/>('1732330', 'Kurt Keutzer', 'kurt keutzer')</td><td>{bichen,alvinwan,xyyue,phj,schzhao,noah.golmant,amirgh,jegonzal,keutzer}@berkeley.edu +</td></tr><tr><td>df0e280cae018cebd5b16ad701ad101265c369fa</td><td>Deep Attributes from Context-Aware Regional Neural Codes +<br/><b>Image Processing Center, Beihang University</b><br/>2 Intel Labs China +<br/><b>Columbia University</b></td><td>('2780589', 'Jianwei Luo', 'jianwei luo')<br/>('35423937', 'Jianguo Li', 'jianguo li')<br/>('1715001', 'Jun Wang', 'jun wang')<br/>('1791565', 'Zhiguo Jiang', 'zhiguo jiang')<br/>('6060281', 'Yurong Chen', 'yurong chen')</td><td></td></tr><tr><td>dfabe7ef245ca68185f4fcc96a08602ee1afb3f7</td><td></td><td></td><td></td></tr><tr><td>df51dfe55912d30fc2f792561e9e0c2b43179089</td><td>Face Hallucination using Linear Models of Coupled +<br/>Sparse Support +<br/>grid and fuse them to suppress the aliasing caused by under- +<br/>sampling [5], [6]. On the other hand, learning based meth- +<br/>ods use coupled dictionaries to learn the mapping relations +<br/>between low- and high- resolution image pairs to synthesize +<br/>high-resolution images from low-resolution images [4], [7]. +<br/>The research community has lately focused on the latter +<br/>category of super-resolution methods, since they can provide +<br/>higher quality images and larger magnification factors. +</td><td>('1805605', 'Reuben A. Farrugia', 'reuben a. farrugia')<br/>('1780587', 'Christine Guillemot', 'christine guillemot')</td><td></td></tr><tr><td>df2c685aa9c234783ab51c1aa1bf1cb5d71a3dbb</td><td>SREFI: Synthesis of Realistic Example Face Images +<br/><b>University of Notre Dame, USA</b><br/><b>FaceTec, Inc</b></td><td>('40061203', 'Sandipan Banerjee', 'sandipan banerjee')<br/>('3365839', 'John S. Bernhard', 'john s. bernhard')<br/>('2613438', 'Walter J. Scheirer', 'walter j. scheirer')<br/>('1799014', 'Kevin W. Bowyer', 'kevin w. bowyer')<br/>('1704876', 'Patrick J. Flynn', 'patrick j. flynn')</td><td>{sbanerj1, wscheire, kwb, flynn}@nd.edu +<br/>jsbernhardjr@gmail.com +</td></tr><tr><td>df054fa8ee6bb7d2a50909939d90ef417c73604c</td><td>Image Quality-Aware Deep Networks Ensemble for Efficient +<br/>Gender Recognition in the Wild +<br/><b>Augmented Vision Lab, Technical University Kaiserslautern, Kaiserslautern, Germany</b><br/><b>German Research Center for Arti cial Intelligence (DFKI), Kaiserslautern, Germany</b><br/>Keywords: +<br/>Gender, Face, Deep Neural Networks, Quality, In the Wild +</td><td>('2585383', 'Mohamed Selim', 'mohamed selim')<br/>('40810260', 'Suraj Sundararajan', 'suraj sundararajan')<br/>('1771057', 'Alain Pagani', 'alain pagani')<br/>('1807169', 'Didier Stricker', 'didier stricker')</td><td>{mohamed.selim, alain.pagani, didier.stricker}@dfki.uni-kl.de, s lakshmin13@informatik.uni-kl.de +</td></tr><tr><td>df80fed59ffdf751a20af317f265848fe6bfb9c9</td><td>1666 +<br/>Learning Deep Sharable and Structural +<br/>Detectors for Face Alignment +</td><td>('40387982', 'Hao Liu', 'hao liu')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('2632601', 'Jianjiang Feng', 'jianjiang feng')<br/>('25060740', 'Jie Zhou', 'jie zhou')</td><td></td></tr><tr><td>dfd8602820c0e94b624d02f2e10ce6c798193a25</td><td>STRUCTURED ANALYSIS DICTIONARY LEARNING FOR IMAGE CLASSIFICATION +<br/>Department of Electrical and Computer Engineering +<br/><b>North Carolina State University, Raleigh, NC, USA</b><br/>†Army Research Office, RTP, Raleigh, NC, USA +</td><td>('49501811', 'Wen Tang', 'wen tang')<br/>('1733181', 'Ashkan Panahi', 'ashkan panahi')<br/>('1769928', 'Hamid Krim', 'hamid krim')<br/>('2622498', 'Liyi Dai', 'liyi dai')</td><td>{wtang6, apanahi, ahk}@ncsu.edu, liyi.dai@us.army.mil +</td></tr><tr><td>dff838ba0567ef0a6c8fbfff9837ea484314efc6</td><td>Progress Report, MSc. Dissertation: On-line +<br/>Random Forest for Face Detection +<br/>School of Computer Science +<br/><b>The University of Manchester</b><br/>May 9, 2014 +<br/>Contents +<br/>1 Introduction +<br/>2 Background +<br/>3 Research Methods +<br/>3.1 What the project involves . . . . . . . . . . . . . . . . . . . . . . +<br/>3.2 The project plan and evaluation of the plan . . . . . . . . . . . . +<br/>4 Progress +<br/>4.1 Quality attributes +<br/>4.2 Prototypes +<br/>. . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>4.2.1 PGM Image . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>4.2.2 Working with Haar-like features and Integral Image +<br/>. . . +<br/>4.2.3 Accesing the Webcam Driver . . . . . . . . . . . . . . . . +<br/>4.2.4 The On-line Random Forest . . . . . . . . . . . . . . . . . +<br/>4.2.5 The First version of the User Interface . . . . . . . . . . . +<br/>4.3 Open discussion about the On-line Random Forest . . . . . . . . +<br/>5 Next Steps and Conclusions +<br/>6 References +<br/>10 +<br/>10 +<br/>11 +<br/>11 +<br/>12 +<br/>13 +<br/>15 +<br/>15 +<br/>16 +<br/>17 +<br/>18 +</td><td></td><td></td></tr><tr><td>dfa80e52b0489bc2585339ad3351626dee1a8395</td><td>Human Action Forecasting by Learning Task Grammars +</td><td>('22237490', 'Tengda Han', 'tengda han')<br/>('36541522', 'Jue Wang', 'jue wang')<br/>('2691929', 'Anoop Cherian', 'anoop cherian')<br/>('2377076', 'Stephen Gould', 'stephen gould')</td><td></td></tr><tr><td>df71a00071d5a949f9c31371c2e5ee8b478e7dc8</td><td>Using Opportunistic Face Logging +<br/>from Smartphone to Infer Mental +<br/>Health: Challenges and Future +<br/>Directions +<br/><b>Dartmouth College</b><br/><b>Dartmouth College</b><br/><b>Dartmouth College</b><br/>Permission to make digital or hard copies of all or part of this work for personal +<br/>or classroom use is granted without fee provided that copies are not made or +<br/>distributed for profit or commercial advantage and that copies bear this notice +<br/>and the full citation on the first page. Copyrights for components of this work +</td><td>('1698066', 'Rui Wang', 'rui wang')<br/>('1690035', 'Andrew T. Campbell', 'andrew t. campbell')<br/>('2253140', 'Xia Zhou', 'xia zhou')</td><td>rui.wang@cs.dartmouth.edu +<br/>campbell@cs.dartmouth.edu +<br/>xia@cs.dartmouth.edu +</td></tr><tr><td>df9269657505fcdc1e10cf45bbb8e325678a40f5</td><td>INTERSPEECH 2016 +<br/>September 8–12, 2016, San Francisco, USA +<br/>Open-Domain Audio-Visual Speech Recognition: A Deep Learning Approach +<br/><b>Carnegie Mellon University</b></td><td>('37467623', 'Yajie Miao', 'yajie miao')<br/>('1740721', 'Florian Metze', 'florian metze')</td><td>{ymiao,fmetze}@cs.cmu.edu +</td></tr><tr><td>dfb6aa168177d4685420fcb184def0aa7db7cddb</td><td>The Effect of Lighting Direction/Condition on the Performance +<br/>of Face Recognition Algorithms +<br/><b>West Virginia University, Morgantown, WV</b><br/><b>University of Miami, Coral Gables, FL</b></td><td>('1722978', 'Gamal Fahmy', 'gamal fahmy')<br/>('4562956', 'Ahmed El-Sherbeeny', 'ahmed el-sherbeeny')<br/>('9449390', 'Mohamed Abdel-Mottaleb', 'mohamed abdel-mottaleb')<br/>('16279046', 'Hany Ammar', 'hany ammar')</td><td></td></tr><tr><td>df2841a1d2a21a0fc6f14fe53b6124519f3812f9</td><td>Learning Image Attributes +<br/>using the Indian Buffet Process +<br/>Department of Computer Science +<br/><b>Brown University</b><br/>Providence, RI 02912 +<br/>Department of Computer Science +<br/><b>Brown University</b><br/>Providence, RI 02912 +</td><td>('2059199', 'Soravit Changpinyo', 'soravit changpinyo')<br/>('1799035', 'Erik B. Sudderth', 'erik b. sudderth')</td><td>schangpi@cs.brown.edu +<br/>sudderth@cs.brown.edu +</td></tr><tr><td>dfecaedeaf618041a5498cd3f0942c15302e75c3</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>A Recursive Framework for Expression Recognition: From +<br/>Web Images to Deep Models to Game Dataset +<br/>Received: date / Accepted: date +</td><td>('48625314', 'Wei Li', 'wei li')</td><td></td></tr><tr><td>df5fe0c195eea34ddc8d80efedb25f1b9034d07d</td><td>Robust Modified Active Shape Model for Automatic Facial Landmark +<br/>Annotation of Frontal Faces +</td><td>('2363348', 'Keshav Seshadri', 'keshav seshadri')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td></td></tr><tr><td>df2494da8efa44d70c27abf23f73387318cf1ca8</td><td>RESEARCH ARTICLE +<br/>Supervised Filter Learning for Representation +<br/>Based Face Recognition +<br/><b>College of Computer Science and Information Technology, Northeast Normal University, Changchun</b><br/><b>China, 2 Changchun Institute of Optics, Fine Mechanics and Physics, CAS, Changchun, China, 3 School of</b><br/><b>Software, Jiangxi Normal University, Nanchang, China, 4 School of Statistics, Capital University of</b><br/>Economics and Business, Beijing, China +<br/>a11111 +</td><td>('2498586', 'Chao Bi', 'chao bi')<br/>('1684635', 'Lei Zhang', 'lei zhang')<br/>('7009658', 'Miao Qi', 'miao qi')<br/>('5858971', 'Caixia Zheng', 'caixia zheng')<br/>('3042163', 'Yugen Yi', 'yugen yi')<br/>('1831935', 'Jianzhong Wang', 'jianzhong wang')<br/>('1751108', 'Baoxue Zhang', 'baoxue zhang')</td><td>* wangjz019@nenu.edu.cn (JW); zhangbaoxue@cueb.edu.cn (BZ) +</td></tr><tr><td>df674dc0fc813c2a6d539e892bfc74f9a761fbc8</td><td>IOSR Journal of Computer Engineering (IOSR-JCE) +<br/>e-ISSN: 2278-0661, p- ISSN: 2278-8727Volume 10, Issue 6 (May. - Jun. 2013), PP 21-29 +<br/>www.iosrjournals.org +<br/>An Image Mining System for Gender Classification & Age +<br/>Prediction Based on Facial Features +<br/> 1.Ms.Dhanashri Shirkey , 2Prof.Dr.S.R.Gupta, +<br/>M.E(Scholar),Department Computer Science & Engineering, PRMIT & R, Badnera +<br/>Asstt.Prof. Department Computer Science & Engineering, PRMIT & R, Badnera +</td><td></td><td></td></tr><tr><td>dad7b8be074d7ea6c3f970bd18884d496cbb0f91</td><td>Super-Sparse Regression for Fast Age +<br/>Estimation From Faces at Test Time +<br/><b>University of Cagliari</b><br/>Piazza d’Armi, 09123 Cagliari, Italy +<br/>WWW home page: http://prag.diee.unica.it +</td><td>('2272441', 'Ambra Demontis', 'ambra demontis')<br/>('1684175', 'Battista Biggio', 'battista biggio')<br/>('1716261', 'Giorgio Fumera', 'giorgio fumera')<br/>('1710171', 'Fabio Roli', 'fabio roli')</td><td>{ambra.demontis,battista.biggio,fumera,roli}@diee.unica.it +</td></tr><tr><td>daf05febbe8406a480306683e46eb5676843c424</td><td>Robust Subspace Segmentation with Block-diagonal Prior +<br/><b>National University of Singapore, Singapore</b><br/><b>Key Lab. of Machine Perception, School of EECS, Peking University, China</b><br/><b>National University of Singapore, Singapore</b></td><td>('33221685', 'Jiashi Feng', 'jiashi feng')<br/>('33383055', 'Zhouchen Lin', 'zhouchen lin')<br/>('1678675', 'Huan Xu', 'huan xu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>1{a0066331,eleyans}@nus.edu.sg, 2zlin@pku.edu.cn, 3mpexuh@nus.edu.sg +</td></tr><tr><td>da4170c862d8ae39861aa193667bfdbdf0ecb363</td><td>Multi-task CNN Model for Attribute Prediction +</td><td>('3282196', 'Abrar H. Abdulnabi', 'abrar h. abdulnabi')<br/>('22804340', 'Gang Wang', 'gang wang')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('2370507', 'Kui Jia', 'kui jia')</td><td></td></tr><tr><td>da15344a4c10b91d6ee2e9356a48cb3a0eac6a97</td><td></td><td></td><td></td></tr><tr><td>da5bfddcfe703ca60c930e79d6df302920ab9465</td><td></td><td></td><td></td></tr><tr><td>dac2103843adc40191e48ee7f35b6d86a02ef019</td><td>854 +<br/>Unsupervised Celebrity Face Naming in Web Videos +</td><td>('2172810', 'Lei Pang', 'lei pang')<br/>('1751681', 'Chong-Wah Ngo', 'chong-wah ngo')</td><td></td></tr><tr><td>dae420b776957e6b8cf5fbbacd7bc0ec226b3e2e</td><td>RECOGNIZING EMOTIONS IN SPONTANEOUS FACIAL EXPRESSIONS +<br/>Institut f¨ur Nachrichtentechnik +<br/>Universit¨at Karlsruhe (TH), Germany +</td><td>('2500636', 'Michael Grimm', 'michael grimm')<br/>('1787004', 'Kristian Kroschel', 'kristian kroschel')</td><td>grimm@int.uni-karlsruhe.de +</td></tr><tr><td>daa02cf195818cbf651ef81941a233727f71591f</td><td>Face recognition system on Raspberry Pi +<br/><b>Institute of Electronics and Computer Science</b><br/>14 Dzerbenes Street, Riga, LV 1006, Latvia +</td><td>('2059963', 'Olegs Nikisins', 'olegs nikisins')<br/>('2337567', 'Rihards Fuksis', 'rihards fuksis')<br/>('3199162', 'Arturs Kadikis', 'arturs kadikis')<br/>('3310787', 'Modris Greitans', 'modris greitans')</td><td></td></tr><tr><td>daa52dd09b61ee94945655f0dde216cce0ebd505</td><td>Recognizing Micro-Actions and Reactions from Paired Egocentric Videos +<br/><b>The University of Tokyo</b><br/><b>Carnegie Mellon University</b><br/><b>The University of Tokyo</b><br/>Tokyo, Japan +<br/>Pittsburgh, PA, USA +<br/>Tokyo, Japan +</td><td>('1899753', 'Ryo Yonetani', 'ryo yonetani')<br/>('37991449', 'Kris M. Kitani', 'kris m. kitani')<br/>('9467266', 'Yoichi Sato', 'yoichi sato')</td><td>yonetani@iis.u-tokyo.ac.jp +<br/>kkitani@cs.cmu.edu +<br/>ysato@iis.u-tokyo.ac.jp +</td></tr><tr><td>daba8f0717f3f47c272f018d0a466a205eba6395</td><td></td><td></td><td></td></tr><tr><td>daefac0610fdeff415c2a3f49b47968d84692e87</td><td>New Orleans, Louisiana, June 1 - 6, 2018. c(cid:13)2018 Association for Computational Linguistics +<br/>Proceedings of NAACL-HLT 2018, pages 1481–1491 +<br/>1481 +</td><td></td><td></td></tr><tr><td>b49affdff167f5d170da18de3efa6fd6a50262a2</td><td>Author manuscript, published in "Workshop on Faces in 'Real-Life' Images: Detection, Alignment, and Recognition, Marseille : France +<br/>(2008)" +</td><td></td><td></td></tr><tr><td>b4d694961d3cde43ccef7d8fcf1061fe0d8f97f3</td><td>Rapid Face Recognition Using Hashing +<br/><b>Australian National University, and NICTA</b><br/><b>Australian National University, and NICTA</b><br/>Canberra, Australia +<br/>Canberra, Australia +<br/><b>NICTA, and Australian National University</b><br/>Canberra, Australia +</td><td>('3177281', 'Qinfeng Shi', 'qinfeng shi')<br/>('1711119', 'Hanxi Li', 'hanxi li')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')</td><td></td></tr><tr><td>b4ee1b468bf7397caa7396cfee2ab5f5ed6f2807</td><td>A short review and primer on electromyography +<br/>in human computer interaction applications +<br/><b>Helsinki Collegium for Advanced Studies, University of Helsinki, Finland</b><br/><b>Helsinki Institute for Information Technology, Aalto University, Finland</b><br/><b>School of Business, Aalto University, Finland</b><br/><b>Quantitative Employee unit, Finnish Institute of Occupational Health</b><br/>POBox 40, Helsinki, 00250, Finland +<br/><b>Cognitive Brain Research Unit, Institute of Behavioural Sciences, University of</b><br/>Helsinki, Finland +</td><td>('1751008', 'Niklas Ravaja', 'niklas ravaja')<br/>('1713422', 'Jari Torniainen', 'jari torniainen')</td><td>benjamin.cowley@ttl.fi, +</td></tr><tr><td>b446bcd7fb78adfe346cf7a01a38e4f43760f363</td><td>To appear in ICB 2018 +<br/>Longitudinal Study of Child Face Recognition +<br/><b>Michigan State University</b><br/>East Lansing, MI, USA +<br/><b>Malaviya National Institute of Technology</b><br/>Jaipur, India +<br/><b>Michigan State University</b><br/>East Lansing, MI, USA +</td><td>('32623642', 'Debayan Deb', 'debayan deb')<br/>('2117075', 'Neeta Nain', 'neeta nain')<br/>('1739705', 'Anil K. Jain', 'anil k. jain')</td><td>debdebay@msu.edu +<br/>nnain.cse@mnit.ac.in +<br/>jain@cse.msu.edu +</td></tr><tr><td>b417b90fa0c288bbaab1aceb8ebc7ec1d3f33172</td><td>Face Aging with Contextual Generative Adversarial Nets +<br/>SKLOIS, IIE, CAS +<br/>SKLOIS, IIE, CAS +<br/>School of Cyber Security, UCAS +<br/>SKLOIS, IIE, CAS +<br/><b>University of Trento, Italy</b><br/><b>Qihoo 360 AI Institute, Beijing, China</b><br/><b>National University of singapore</b><br/>SKLOIS, IIE, CAS +<br/>School of Cyber Security, UCAS +<br/><b>Nanjing University of Science and</b><br/>Technology +</td><td>('38110120', 'Si Liu', 'si liu')<br/>('7760591', 'Renda Bao', 'renda bao')<br/>('39711014', 'Yao Sun', 'yao sun')<br/>('1699978', 'Wei Wang', 'wei wang')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('4661961', 'Defa Zhu', 'defa zhu')<br/>('2287686', 'Xiangbo Shu', 'xiangbo shu')</td><td>liusi@iie.ac.cn +<br/>roger bao@163.com +<br/>sunyao@iie.ac.cn +<br/>wangwei1990@gmail.com +<br/>eleyans@nus.edu.sg +<br/>18502408950@163.com +<br/>shuxb@njust.edu.cn +</td></tr><tr><td>b41374f4f31906cf1a73c7adda6c50a78b4eb498</td><td>This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. +<br/>Iterative Gaussianization: From ICA to +<br/>Random Rotations +</td><td>('2732577', 'Valero Laparra', 'valero laparra')<br/>('1684246', 'Gustavo Camps-Valls', 'gustavo camps-valls')<br/>('2186866', 'Jesús Malo', 'jesús malo')</td><td></td></tr><tr><td>b42a97fb47bcd6bfa72e130c08960a77ee96f9ab</td><td>FACIAL EXPRESSION RECOGNITION BASED ON GRAPH-PRESERVING SPARSE +<br/>NON-NEGATIVE MATRIX FACTORIZATION +<br/><b>Institute of Information Science</b><br/><b>Beijing Jiaotong University</b><br/>Beijing 100044, P.R. China +<br/>Qiuqi Ruan +<br/>ACCESS Linnaeus Center +<br/><b>KTH Royal Institute of Technology, Stockholm</b><br/>School of Electrical Engineering +</td><td>('3247912', 'Ruicong Zhi', 'ruicong zhi')<br/>('1749334', 'Markus Flierl', 'markus flierl')</td><td>{05120370, qqruan}@bjtu.edu.cn +<br/>{ruicong, mflierl, bastiaan}@kth.se +</td></tr><tr><td>b4d209845e1c67870ef50a7c37abaf3770563f3e</td><td>GHODRATI, GAVVES, SNOEK: VIDEO TIME +<br/>Video Time: Properties, Encoders and +<br/>Evaluation +<br/>Cees G. M. Snoek +<br/>QUVA Lab +<br/><b>University of Amsterdam</b><br/>Netherlands +</td><td>('3060081', 'Amir Ghodrati', 'amir ghodrati')<br/>('2304222', 'Efstratios Gavves', 'efstratios gavves')</td><td>{a.ghodrati,egavves,cgmsnoek}@uva.nl +</td></tr><tr><td>b4d7ca26deb83cec1922a6964c1193e8dd7270e7</td><td></td><td></td><td></td></tr><tr><td>b4ee64022cc3ccd14c7f9d4935c59b16456067d3</td><td>Unsupervised Cross-Domain Image Generation +</td><td>('40084473', 'Davis Rempe', 'davis rempe')<br/>('9184695', 'Haotian Zhang', 'haotian zhang')</td><td></td></tr><tr><td>b40290a694075868e0daef77303f2c4ca1c43269</td><td>第 40 卷 第 4 期 +<br/>2014 年 4 月 +<br/>自 动 化 学 报 +<br/>ACTA AUTOMATICA SINICA +<br/>Vol. 40, No. 4 +<br/>April, 2014 +<br/>融合局部与全局信息的头发形状模型 +<br/>王 楠 1 艾海舟 1 +<br/>摘 要 头发在人体表观中具有重要作用, 然而, 因为缺少有效的形状模型, 头发分割仍然是一个非常具有挑战性的问题. 本 +<br/>文提出了一种基于部件的模型, 它对头发形状以及环境变化更加鲁棒. 该模型将局部与全局信息相结合以描述头发的形状. 局 +<br/>部模型通过一系列算法构建, 包括全局形状词表生成, 词表分类器学习以及参数优化; 而全局模型刻画不同的发型, 采用支持 +<br/>向量机 (Support vector machine, SVM) 来学习, 它为所有潜在的发型配置部件并确定势函数. 在消费者图片上的实验证明 +<br/>了本文算法在头发形状多变和复杂环境等条件下的准确性与有效性. +<br/>关键词 头发形状建模, 部件模型, 部件配置算法, 支持向量机 +<br/>引用格式 王楠, 艾海舟. 融合局部与全局信息的头发形状模型. 自动化学报, 2014, 40(4): 615−623 +<br/>DOI 10.3724/SP.J.1004.2014.00615 +<br/>Combining Local and Global Information for Hair Shape Modeling +<br/>AI Hai-Zhou1 +</td><td>('3666771', 'WANG Nan', 'wang nan')</td><td></td></tr><tr><td>b4362cd87ad219790800127ddd366cc465606a78</td><td>Sensors 2015, 15, 26756-26768; doi:10.3390/s151026756 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>A Smartphone-Based Automatic Diagnosis System for Facial +<br/>Nerve Palsy +<br/><b>Interdisciplinary Program of Bioengineering, Seoul National University, Seoul 03080, Korea</b><br/><b>Head and Neck Surgery, Seoul National University</b><br/><b>College of Medicine, Seoul National University</b><br/>Seoul 03080, Korea +<br/>Fax: +82-2-870-3863 (Y.H.K.); +82-2-3676-1175 (K.S.P.). +<br/>Academic Editor: Ki H. Chon +<br/>Received: 31 July 2015 / Accepted: 19 October 2015 / Published: 21 October 2015 +</td><td>('31812715', 'Hyun Seok Kim', 'hyun seok kim')<br/>('2189639', 'So Young Kim', 'so young kim')<br/>('40219387', 'Young Ho Kim', 'young ho kim')<br/>('1972762', 'Kwang Suk Park', 'kwang suk park')</td><td>E-Mail: khs0330kr@bmsil.snu.ac.kr +<br/>Boramae Medical Center, Seoul 07061, Korea; E-Mail: sossi81@hanmail.net +<br/>* Authors to whom correspondence should be addressed; E-Mails: yhkiment@gmail.com (Y.H.K.); +<br/>pks@bmsil.snu.ac.kr (K.S.P.); Tel.: +82-2-870-2442 (Y.H.K.); +82-2-2072-3135 (K.S.P.); +</td></tr><tr><td>b4f4b0d39fd10baec34d3412d53515f1a4605222</td><td>Every Picture Tells a Story: +<br/>Generating Sentences from Images +<br/>1 Computer Science Department +<br/><b>University of Illinois at Urbana-Champaign</b><br/>2 Computer Vision Group, School of Mathematics +<br/><b>Institute for studies in theoretical Physics and Mathematics(IPM</b></td><td>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('1888731', 'Mohsen Hejrati', 'mohsen hejrati')<br/>('21160985', 'Mohammad Amin Sadeghi', 'mohammad amin sadeghi')<br/>('35527128', 'Peter Young', 'peter young')<br/>('3125805', 'Cyrus Rashtchian', 'cyrus rashtchian')<br/>('3118681', 'Julia Hockenmaier', 'julia hockenmaier')</td><td>{afarhad2,pyoung2,crashtc2,juliahmr,daf}@illinois.edu +<br/>{m.a.sadeghi,mhejrati}@gmail.com +</td></tr><tr><td>b4b0bf0cbe1a2c114adde9fac64900b2f8f6fee4</td><td>Autonomous Learning Framework Based on Online Hybrid +<br/>Classifier for Multi-view Object Detection in Video +<br/><b>aSchool of Electronic Information and Mechanics, China University of Geosciences, Wuhan, Hubei 430074, China</b><br/><b>bSchool of Automation, China University of Geosciences, Wuhan, Hubei 430074, China</b><br/><b>cHuizhou School Affiliated to Beijing Normal University, Huizhou 516002, China</b><br/>dNational Key Laboratory of Science and Technology on Multispectral Information Processing, School of Automation, Huazhong +<br/><b>University of Science and Technology, Wuhan, 430074, China</b></td><td>('2588731', 'Dapeng Luo', 'dapeng luo')</td><td></td></tr><tr><td>b43b6551ecc556557b63edb8b0dc39901ed0343b</td><td>ICA AND GABOR REPRESENTATION FOR FACIAL EXPRESSION RECOGNITION +<br/>I. Buciu C. Kotropoulos +<br/>and I. Pitas +<br/><b>Aristotle University of Thessaloniki</b></td><td></td><td>GR-54124, Thessaloniki, Box 451, Greece, {nelu,costas,pitas}@zeus.csd.auth.gr +</td></tr><tr><td>a255a54b8758050ea1632bf5a88a201cd72656e1</td><td>Nonparametric Facial Feature Localization +<br/>J. K. Aggarwal +<br/><b>Computer and Vision Research Center</b><br/><b>The University of Texas at Austin</b></td><td>('2622649', 'Birgi Tamersoy', 'birgi tamersoy')<br/>('1713065', 'Changbo Hu', 'changbo hu')</td><td>birgi@utexas.edu +<br/>changbo.hu@gmail.com +<br/>aggarwaljk@mail.utexas.edu +</td></tr><tr><td>a2b9cee7a3866eb2db53a7d81afda72051fe9732</td><td>Reconstructing a Fragmented Face from an Attacked +<br/>Secure Identification Protocol +<br/>Department of Computer Science +<br/><b>University of Texas at Austin</b><br/>May 6, 2011 +</td><td>('39573884', 'Andy Luong', 'andy luong')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>aluong@cs.utexas.edu +</td></tr><tr><td>a285b6edd47f9b8966935878ad4539d270b406d1</td><td>Sensors 2011, 11, 9573-9588; doi:10.3390/s111009573 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>Facial Expression Recognition Based on Local Binary Patterns +<br/>and Kernel Discriminant Isomap +<br/><b>Taizhou University, Taizhou 317000, China</b><br/><b>School of Physics and Electronic Engineering, Taizhou University, Taizhou 318000, China</b><br/>Tel.: +86-576-8513-7178; Fax: ++86-576-8513-7178. +<br/>Received: 31 August 2011; in revised form: 27 September 2011 / Accepted: 9 October 2011 / +<br/>Published: 11 October 2011 +</td><td>('48551029', 'Xiaoming Zhao', 'xiaoming zhao')<br/>('1695589', 'Shiqing Zhang', 'shiqing zhang')</td><td>E-Mail: tzczsq@163.com +<br/>* Author to whom correspondence should be addressed; E-Mail: tzxyzxm@163.com; +</td></tr><tr><td>a2bd81be79edfa8dcfde79173b0a895682d62329</td><td>Multi-Objective Vehicle Routing Problem Applied to +<br/>Large Scale Post Office Deliveries +<br/>Zenia +<br/><b>aSchool of Technology, University of Campinas</b><br/>Paschoal Marmo, 1888, Limeira, SP, Brazil +</td><td>('1788152', 'Luis A. A. Meira', 'luis a. a. meira')<br/>('37279198', 'Paulo S. Martins', 'paulo s. martins')<br/>('7809605', 'Mauro Menzori', 'mauro menzori')</td><td></td></tr><tr><td>a2359c0f81a7eb032cff1fe45e3b80007facaa2a</td><td>Towards Structured Analysis of Broadcast Badminton Videos +<br/>C.V.Jawahar +<br/>CVIT, KCIS, IIIT Hyderabad +</td><td>('2964097', 'Anurag Ghosh', 'anurag ghosh')<br/>('48039353', 'Suriya Singh', 'suriya singh')</td><td>{anurag.ghosh, suriya.singh}@research.iiit.ac.in, jawahar@iiit.ac.in +</td></tr><tr><td>a2eb90e334575d9b435c01de4f4bf42d2464effc</td><td>A NEW SPARSE IMAGE REPRESENTATION +<br/>ALGORITHM APPLIED TO FACIAL +<br/>EXPRESSION RECOGNITION +<br/>Ioan Buciu and Ioannis Pitas +<br/>Department of Informatics +<br/><b>Aristotle University of Thessaloniki, GR-541 24 Thessaloniki, Greece</b><br/>Phone: +30-231-099-6361 +<br/>Fax: +30-231-099-8453 +<br/>Web: http://poseidon.csd.auth.gr +</td><td></td><td>E-mail: nelu,pitas@zeus.csd.auth.gr +</td></tr><tr><td>a25106a76af723ba9b09308a7dcf4f76d9283589</td><td> Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/> A Monthly Journal of Computer Science and Information Technology +<br/>ISSN 2320–088X +<br/> IJCSMC, Vol. 3, Issue. 4, April 2014, pg.139 – 146 +<br/> RESEARCH ARTICLE +<br/>Local Octal Pattern: A Proficient Feature +<br/>Extraction for Face Recognition +<br/><b>Computer Science and Engineering, Easwari Engineering College, India</b><br/><b>Computer Science and Engineering, Anna University, India</b></td><td>('3263740', 'S Chitrakala', 's chitrakala')</td><td>1 nithya.jagan90@gamil.com +<br/>2 suchitra.s@srmeaswari.ac.in +<br/>3 ckgops@gmail.com +</td></tr><tr><td>a2d9c9ed29bbc2619d5e03320e48b45c15155195</td><td></td><td></td><td></td></tr><tr><td>a29a22878e1881d6cbf6acff2d0b209c8d3f778b</td><td>Benchmarking Still-to-Video Face Recognition +<br/>via Partial and Local Linear Discriminant +<br/>Analysis on COX-S2V Dataset +<br/><b>Key Lab of Intelligent Information Processing, Institute of Computing Technology</b><br/>Chinese Academy of Sciences, Beijing 100190, China +<br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b><br/>3OMRON Social Solutions Co. Ltd, Kyoto, Japan +<br/><b>College of Information Science and Engineering, Xinjiang University</b></td><td>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1705483', 'Haihong Zhang', 'haihong zhang')<br/>('1710195', 'Shihong Lao', 'shihong lao')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{zhiwu.huang, shiguang.shan}@vipl.ict.ac.cn, +<br/>angelazhang@ssb.kusatsu.omron.co.jp, lao@ari.ncl.omron.co.jp, +<br/>ghalipk@xju.edu.cn, xilin.chen@vipl.ict.ac.cn +</td></tr><tr><td>a2429cc2ccbabda891cc5ae340b24ad06fcdbed5</td><td>Discovering the Signatures of Joint Attention in Child-Caregiver Interaction +<br/>Department of Computer Science +<br/>Department of Psychology +<br/><b>Stanford University</b><br/>Department of Psychology +<br/><b>Stanford University</b><br/>Department of Computer Science +<br/><b>Stanford University</b><br/>Department of Psychology +<br/><b>Stanford University</b></td><td>('2536223', 'Michael C. Frank', 'michael c. frank')<br/>('7211962', 'Laura Soriano', 'laura soriano')<br/>('3147852', 'Guido Pusiol', 'guido pusiol')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td>guido@cs.stanford.edu +<br/>lsoriano@stanford.edu +<br/>feifeili@stanford.edu +<br/>mcfrank@stanford.edu +</td></tr><tr><td>a2b54f4d73bdb80854aa78f0c5aca3d8b56b571d</td><td></td><td></td><td></td></tr><tr><td>a27735e4cbb108db4a52ef9033e3a19f4dc0e5fa</td><td>Intention from Motion +</td><td>('40063519', 'Andrea Zunino', 'andrea zunino')<br/>('3393678', 'Jacopo Cavazza', 'jacopo cavazza')<br/>('34465973', 'Atesh Koul', 'atesh koul')<br/>('37783905', 'Andrea Cavallo', 'andrea cavallo')<br/>('1834966', 'Cristina Becchio', 'cristina becchio')<br/>('1727204', 'Vittorio Murino', 'vittorio murino')</td><td></td></tr><tr><td>a2bcfba155c990f64ffb44c0a1bb53f994b68a15</td><td>The Photoface Database +<br/><b>Imperial College London</b><br/>180 Queen’s Gate, London SW7 2AZ UK. +<br/><b>Machine Vision Lab, Faculty of Environment and Technology, University of the West of England</b><br/><b>cid:63) Faculty of Computing, Information Systems and Mathematics, Kingston University London</b><br/>Frenchay Campus, Bristol BS16 1QY UK. +<br/>Exhibition Road, South Kensington Campus, London SW7 2AZ UK. +<br/>River House, 53-57 High Street, Kingston upon Thames, Surrey KT1 1LQ UK. +<br/><b>Imperial College London</b><br/><b>Informatics and Telematics Institute, Centre of Research and Technology - Hellas</b><br/>6th km Xarilaou - Thermi, Thessaloniki 57001 Greece +</td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1689047', 'Vasileios Argyriou', 'vasileios argyriou')<br/>('2871609', 'Maria Petrou', 'maria petrou')</td><td>{s.zafeiriou,maria.petrou}@imperial.ac.uk, vasileios.argyriou@kinston.ac.uk +<br/>{mark.hansen,gary.atkinson,melvyn.smith,lyndon.smith}@uwe.ac.uk. ∗ +</td></tr><tr><td>a2fbaa0b849ecc74f34ebb36d1442d63212b29d2</td><td> Volume 5, Issue 6, June 2015 ISSN: 2277 128X +<br/>International Journal of Advanced Research in +<br/> Computer Science and Software Engineering +<br/> Research Paper +<br/> Available online at: www.ijarcsse.com +<br/>An Efficient Approach to Face Recognition of Surgically +<br/>Altered Images +<br/>Department of computer science and engineering +<br/><b>SUS college of Engineering and Technology</b><br/>Tangori, District, Mohali, Punjab, India +</td><td></td><td></td></tr><tr><td>a50b4d404576695be7cd4194a064f0602806f3c4</td><td>In Proceedings of BMVC, Edimburgh, UK, September 2006 +<br/>Efficiently estimating facial expression and +<br/>illumination in appearance-based tracking +<br/>†ESCET, U. Rey Juan Carlos +<br/>C/ Tulip´an, s/n +<br/>28933 M´ostoles, Spain +<br/>‡Facultad Inform´atica, UPM +<br/>Campus de Montegancedo s/n +<br/>28660 Boadilla del Monte, Spain +<br/>http://www.dia.fi.upm.es/~pcr +</td><td>('1778998', 'Luis Baumela', 'luis baumela')</td><td></td></tr><tr><td>a59cdc49185689f3f9efdf7ee261c78f9c180789</td><td>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING XX, XXX-XXX (2015) +<br/>A New Approach for Learning Discriminative Dictionary +<br/>for Pattern Classification +<br/>THUY THI NGUYEN1, BINH THANH HUYNH2 AND SANG VIET DINH2 +<br/>1Faculty of Information Technology +<br/><b>Vietnam National University of Agriculture</b><br/>Trau Quy town, Gialam, Hanoi, Vietnam +<br/>2School of Information and Communication Technology +<br/><b>Hanoi University of Science and Technology</b><br/>No 1, Dai Co Viet Street, Hanoi, Vietnam +<br/>Dictionary learning (DL) for sparse coding based classification has been widely re- +<br/>searched in pattern recognition in recent years. Most of the DL approaches focused on +<br/>the reconstruction performance and the discriminative capability of the learned dictionary. +<br/>This paper proposes a new method for learning discriminative dictionary for sparse rep- +<br/>resentation based classification, called Incoherent Fisher Discrimination Dictionary +<br/>Learning (IFDDL). IFDDL combines the Fisher Discrimination Dictionary Learning +<br/>(FDDL) method, which learns a structured dictionary where the class labels and the dis- +<br/>crimination criterion are exploited, and the Incoherent Dictionary Learning (IDL) method, +<br/>which learns a dictionary where the mutual incoherence between pairs of atoms is ex- +<br/>ploited. In the combination, instead of considering the incoherence between atoms in a +<br/>single shared dictionary as in IDL, we propose to incorporate the incoherence between +<br/>pairs of atoms within each sub-dictionary, which represent a specific object class. This +<br/>aims to increase discrimination capacity of between basic atoms in sub-dictionaries. The +<br/>combination allows one to exploit the advantages of both methods and the discrimination +<br/>capacity of the entire dictionary. Extensive experiments have been conducted on bench- +<br/>mark image data sets for Face recognition (ORL database, Extended Yale B database, AR +<br/>database) and Digit recognition (the USPS database). The experimental results show that +<br/>our proposed method outperforms most of state-of-the-art methods for sparse coding and +<br/>DL based classification, meanwhile maintaining similar complexity. +<br/>Keywords: dictionary learning, sparse coding, fisher criterion, pattern recognition, object +<br/>classification +<br/>1. INTRODUCTION +<br/>Sparse representation (or sparse coding) has been widely used in many problems of +<br/>image processing and computer vision [1, 2], audio processing [3, 4], as well as classifi- +<br/>cation [5-9] and archived very impressive results. In this model, an input signal is de- +<br/>composed by a sparse linear combination of a few atoms from an over-complete diction- +<br/>ary. In general, the goal of sparse representation is to represent input signals by a linear +<br/>combination of atoms (or words). This is done by minimizing the reconstruction error +<br/>under a sparsity constraint: +<br/>min +<br/>D X +<br/>|| +<br/>A DX +<br/>|| +<br/>X +<br/>|| +<br/>|| +<br/>Received February 15, 2015; revised June 18, 2015; accepted July 9, 2015. +<br/>Communicated by Hsin-Min Wang. +<br/>xxx +<br/>(1) +</td><td></td><td>E-mail: myngthuy@gmail.com +<br/>E-mail: {binhht; sangdv}@soict.hust.edu.vn +</td></tr><tr><td>a5e5094a1e052fa44f539b0d62b54ef03c78bf6a</td><td>Detection without Recognition for Redaction +<br/><b>Rochester Institute of Technology - 83 Lomb Memorial Drive, Rochester, NY USA</b><br/>2Conduent, Conduent Labs - US, 800 Phillips Rd, MS128, Webster, NY USA, 14580 +</td><td>('3424086', 'Shagan Sah', 'shagan sah')<br/>('40492623', 'Ram Longman', 'ram longman')<br/>('29980978', 'Ameya Shringi', 'ameya shringi')<br/>('1736673', 'Robert Loce', 'robert loce')<br/>('39834006', 'Majid Rabbani', 'majid rabbani')<br/>('32847225', 'Raymond Ptucha', 'raymond ptucha')</td><td>Email: sxs4337@rit.edu +</td></tr><tr><td>a5c8fc1ca4f06a344b53dc81ebc6d87f54896722</td><td>Learning to see people like people +<br/><b>University of California, San Diego</b><br/>9500 Gilman Dr, La Jolla, CA 92093 +<br/><b>University of California, San Diego</b><br/>9500 Gilman Dr, La Jolla, CA 92093 +<br/><b>Purdue University</b><br/>610 Purdue Mall, West Lafayette, IN 47907 +<br/>Garrison Cottrell +<br/><b>University of California, San Diego</b><br/>9500 Gilman Dr, La Jolla, CA 92093 +</td><td>('9409376', 'Amanda Song', 'amanda song')<br/>('13212680', 'Chad Atalla', 'chad atalla')<br/>('11157727', 'Linjie Li', 'linjie li')</td><td>feijuejuanling@gmail.com +<br/>li2477@purdue.edu +<br/>catalla@ucsd.edu +<br/>gary@ucsd.edu +</td></tr><tr><td>a5ade88747fa5769c9c92ffde9b7196ff085a9eb</td><td>Why is Facial Expression Analysis in the Wild +<br/>Challenging? +<br/><b>Institute for Anthropomatics</b><br/><b>Karlsruhe Institute of Technology, Germany</b><br/>Hazım Kemal Ekenel +<br/>Faculty of Computer and Informatics +<br/><b>Istanbul Technical University, Turkey</b><br/><b>Institute for Anthropomatics</b><br/><b>Karlsruhe Institute of Technology, Germany</b></td><td>('40303076', 'Tobias Gehrig', 'tobias gehrig')</td><td>tobias.gehrig@kit.edu +<br/>ekenel@itu.edu.tr +</td></tr><tr><td>a56c1331750bf3ac33ee07004e083310a1e63ddc</td><td>Vol. xx, pp. x +<br/>c(cid:13) xxxx Society for Industrial and Applied Mathematics +<br/>x–x +<br/>Efficient Point-to-Subspace Query in (cid:96)1 with Application to Robust Object +<br/>Instance Recognition +</td><td>('1699024', 'Ju Sun', 'ju sun')<br/>('2580421', 'Yuqian Zhang', 'yuqian zhang')<br/>('1738310', 'John Wright', 'john wright')</td><td></td></tr><tr><td>a54e0f2983e0b5af6eaafd4d3467b655a3de52f4</td><td>Face Recognition Using Convolution Filters and +<br/>Neural Networks +<br/>Head, Dept. of E&E,PEC +<br/>Sec-12, Chandigarh – 160012 +<br/>Department of CSE & IT, PEC +<br/>Sec-12, Chandigarh – 160012 +<br/>C.P. Singh +<br/>Physics Department, CFSL, +<br/>Sec-36, Chandigarh - 160036 +<br/>a +<br/>of +<br/>to: (a) +<br/>potential method +</td><td>('1734714', 'V. Rihani', 'v. rihani')<br/>('2927010', 'Amit Bhandari', 'amit bhandari')</td><td>vrihani@yahoo.com +<br/>amit.bhandari@yahoo.com +<br/>cpureisingh@yahoo.com +</td></tr><tr><td>a5625cfe16d72bd00e987857d68eb4d8fc3ce4fb</td><td>VFSC: A Very Fast Sparse Clustering to Cluster Faces +<br/>from Videos +<br/><b>University of Science, VNU-HCMC, Ho Chi Minh city, Vietnam</b></td><td>('2187730', 'Dinh-Luan Nguyen', 'dinh-luan nguyen')<br/>('1780348', 'Minh-Triet Tran', 'minh-triet tran')</td><td>1212223@student.hcmus.edu.vn +<br/>tmtriet@fit.hcmus.edu.vn +</td></tr><tr><td>a5f11c132eaab258a7cea2d681875af09cddba65</td><td>A spatiotemporal model with visual attention for +<br/>video classification +<br/>Department of Electrical and Computer Engineering +<br/><b>University of California San Diego, La Jolla, California, USA</b><br/>paper proposes a spatiotemporal model in which CNN and +<br/>RNN are concatenated, as shown in Fig. 1. +</td><td>('2493180', 'Mo Shan', 'mo shan')<br/>('50365495', 'Nikolay Atanasov', 'nikolay atanasov')</td><td>Email: {moshan, natanasov}@eng.ucsd.edu +</td></tr><tr><td>a546fd229f99d7fe3cf634234e04bae920a2ec33</td><td>RESEARCH ARTICLE +<br/>Fast Fight Detection +<br/>1 Department of Systems Engineering and Automation, E.T.S.I. Industriales, Ciudad Real, Castilla-La +<br/><b>Mancha, Spain, Imperial College, London, UK</b></td><td>('5463808', 'Ismael Serrano Gracia', 'ismael serrano gracia')<br/>('8952654', 'Oscar Deniz Suarez', 'oscar deniz suarez')<br/>('8219927', 'Gloria Bueno Garcia', 'gloria bueno garcia')<br/>('1700968', 'Tae-Kyun Kim', 'tae-kyun kim')</td><td>* ismael.serrano@uclm.es (ISG); oscar.deniz@uclm.es (ODS); gloria.bueno@uclm.es (GBG) +</td></tr><tr><td>a538b05ebb01a40323997629e171c91aa28b8e2f</td><td>Rectified Linear Units Improve Restricted Boltzmann Machines +<br/>Geoffrey E. Hinton +<br/><b>University of Toronto, Toronto, ON M5S 2G4, Canada</b></td><td>('4989209', 'Vinod Nair', 'vinod nair')</td><td>vnair@cs.toronto.edu +<br/>hinton@cs.toronto.edu +</td></tr><tr><td>a57ee5a8fb7618004dd1def8e14ef97aadaaeef5</td><td>Fringe Projection Techniques: Whither we are? +<br/><b>Applied computing and mechanics laboratory, Swiss Federal Institute of Technology, 1015 Lausanne, Switzerland</b><br/>During recent years, the use of fringe projection techniques +<br/>for generating three-dimensional (3D) surface information has +<br/>become one of the most active research areas in optical metrol- +<br/>ogy. +<br/>Its applications range from measuring the 3D shape of +<br/>MEMS components to the measurement of flatness of large +<br/>panels (2.5 m × .45 m). The technique has found various ap- +<br/>plications in diverse fields: biomedical applications such as +<br/>3D intra-oral dental measurements [1], non-invasive 3D imag- +<br/>ing and monitoring of vascular wall deformations [2], human +<br/>body shape measurement for shape guided radiotherapy treat- +<br/>ment [3, 4], lower back deformation measurement [5], detection +<br/>and monitoring of scoliosis [6], inspection of wounds [7, 8] +<br/>and skin topography measurement for use in cosmetology [9, +<br/>10, 11]; +<br/>industrial and scientific applications such as char- +<br/>acterization of MEMS components [12, 13], vibration analy- +<br/>sis [14, 15], refractometry [16], global measurement of free +<br/>surface deformations [17, 18], local wall thickness measure- +<br/>ment of forced sheet metals [19], corrosion analysis [20, 21], +<br/>measurement of surface roughness [22, 23], reverse engineer- +<br/>ing [24, 25, 26], quality control of printed circuit board man- +<br/>ufacturing [27, 28, 29] and heat-flow visualization [30]; kine- +<br/>matics applications such as measuring the shape and position +<br/>of a moving object/creature [31, 32] and the study of kinemat- +<br/>ical parameters of dragonfly in free flight [33, 34]; biometric +<br/>identification applications such as 3D face reconstruction for +<br/>the development of robust face recognition systems [35, 36]; +<br/>cultural heritage and preservation [37, 38, 39] etc. +<br/>One of the outstanding features of some of the fringe pro- +<br/>jection techniques is their ability to provide high-resolution, +<br/>whole-field 3D reconstruction of objects in a non-contact man- +<br/>ner at video frame rates. This feature has backed the technique +<br/>to pervade new areas of applications such as security systems, +<br/>gaming and virtual reality. To gain insights into the series of +<br/>contributions that have helped in unfolding the technique to ac- +<br/>quire this feature, the reader is referred to the review articles in +<br/>this special issue by Song Zhang, and Xianyu Su et al. +<br/>A typical fringe projection profilometry system is shown in +<br/>Fig 1. +<br/>It consists of a projection unit, an image acquisition +<br/>unit and a processing/analysis unit. Measurement of shape +<br/>through fringe projection techniques involves (1) projecting a +<br/>structured pattern (usually a sinusoidal fringe pattern) onto the +<br/>object surface, (2) recording the image of the fringe pattern +<br/>that is phase modulated by the object height distribution, (3) +<br/>calculating the phase modulation by analyzing the image with +<br/>one of the fringe analysis techniques (such as Fourier transform +<br/>Figure 1: Fringe projection profilometry system +<br/>method, phase stepping and spatial phase detection methods- +<br/>most of them generate wrapped phase distribution) (4) using a +<br/>suitable phase unwrapping algorithm to get continuous phase +<br/>distribution which is proportional to the object height varia- +<br/>tions, and finally (5) calibrating the system for mapping the +<br/>unwrapped phase distribution to real world 3-D co-ordinates. +<br/>Fig. 2 shows the flowchart that depicts different steps involved +<br/>in the measurement of height distribution of an object using the +<br/>fringe projection technique and the role of each step. A pic- +<br/>torial representation of the same with more details is shown in +<br/>Fig. 3. +<br/>During the last three decades, fringe projection techniques +<br/>have developed tremendously due to the contribution of large +<br/>number of researchers and the developments can be broadly +<br/>categorized as follows: design or structure of the pattern +<br/>used for projection [40, 41, 42, 43, 44, 45, 46, 47, 48, 49], +<br/>method of generating and projecting the patterns [50, 51, 52, +<br/>53, 54, 55, 56, 57, 58, 59, 60, 61, 62], study of errors +<br/>caused by the equipment used and proposing possible correc- +<br/>tions [63, 64, 65, 66], developing new fringe analysis meth- +<br/>ods to extract underlying phase distribution [67, 68, 69, 70, +<br/>71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83], improv- +<br/>ing existing fringe analysis methods [84, 85, 86, 87, 88, 89, +<br/>90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100], phase unwrapping +<br/>algorithms [101, 102, 103, 104, 105, 106, 107, 108, 109], cal- +<br/>ibration techniques [110, 111, 112, 113, 114, 115, 116, 117, +<br/>118, 119, 120, 121, 122, 123], scale of measurement (mi- +<br/>Preprint submitted to Optics and Lasers in Engineering +<br/>September 1, 2009 +</td><td>('1694155', 'Sai Siva Gorthi', 'sai siva gorthi')<br/>('32741407', 'Pramod Rastogi', 'pramod rastogi')</td><td></td></tr><tr><td>a5ae7fe2bb268adf0c1cd8e3377f478fca5e4529</td><td>Exemplar Hidden Markov Models for Classification of Facial Expressions in +<br/>Videos +<br/>Univ. of California San Diego +<br/>Univ. of Canberra, Australian +<br/>Univ. of California San Diego +<br/>Marian Bartlett +<br/>California, USA +<br/><b>National University</b><br/>Australia +<br/>California, USA +</td><td>('1735697', 'Abhinav Dhall', 'abhinav dhall')<br/>('39707211', 'Karan Sikka', 'karan sikka')</td><td>ksikka@ucsd.edu +<br/>mbartlett@ucsd.edu +<br/>abhinav.dhall@anu.edu +</td></tr><tr><td>a55efc4a6f273c5895b5e4c5009eabf8e5ed0d6a</td><td>818 +<br/>Continuous Head Movement Estimator for +<br/>Driver Assistance: Issues, Algorithms, +<br/>and On-Road Evaluations +<br/>Mohan Manubhai Trivedi, Fellow, IEEE +</td><td>('1947383', 'Ashish Tawari', 'ashish tawari')<br/>('1841835', 'Sujitha Martin', 'sujitha martin')</td><td></td></tr><tr><td>a51d5c2f8db48a42446cc4f1718c75ac9303cb7a</td><td>Cross-validating Image Description Datasets and Evaluation Metrics +<br/>Department of Computer Science +<br/><b>University of Shef eld, UK</b></td><td>('2635321', 'Josiah Wang', 'josiah wang')</td><td>{j.k.wang, r.gaizauskas}@sheffield.ac.uk +</td></tr><tr><td>a52d9e9daf2cb26b31bf2902f78774bd31c0dd88</td><td>Understanding and Designing Convolutional Networks +<br/>for Local Recognition Problems +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2016-97 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-97.html +<br/>May 13, 2016 +</td><td>('34703740', 'Jonathan Long', 'jonathan long')</td><td></td></tr><tr><td>a51882cfd0706512bf50e12c0a7dd0775285030d</td><td>Cross-Modal Face Matching: Beyond Viewed +<br/>Sketches +<br/><b>Beijing University of Posts and Telecommunications, Beijing, China. 2School of</b><br/><b>Electronic Engineering and Computer Science Queen Mary University of London</b><br/>London E1 4NS, United Kingdom +</td><td>('2961830', 'Shuxin Ouyang', 'shuxin ouyang')<br/>('1705408', 'Yi-Zhe Song', 'yi-zhe song')<br/>('7823169', 'Xueming Li', 'xueming li')</td><td></td></tr><tr><td>a5c04f2ad6a1f7c50b6aa5b1b71c36af76af06be</td><td></td><td></td><td></td></tr><tr><td>a503eb91c0bce3a83bf6f524545888524b29b166</td><td></td><td></td><td></td></tr><tr><td>a5a44a32a91474f00a3cda671a802e87c899fbb4</td><td>Moments in Time Dataset: one million +<br/>videos for event understanding +</td><td>('2526653', 'Mathew Monfort', 'mathew monfort')<br/>('1804424', 'Bolei Zhou', 'bolei zhou')<br/>('3298267', 'Sarah Adel Bargal', 'sarah adel bargal')<br/>('50112310', 'Alex Andonian', 'alex andonian')<br/>('12082007', 'Tom Yan', 'tom yan')<br/>('40544169', 'Kandan Ramakrishnan', 'kandan ramakrishnan')<br/>('33421444', 'Quanfu Fan', 'quanfu fan')<br/>('1856025', 'Carl Vondrick', 'carl vondrick')<br/>('31735139', 'Aude Oliva', 'aude oliva')</td><td></td></tr><tr><td>a52581a7b48138d7124afc7ccfcf8ec3b48359d0</td><td>http://www.jos.org.cn +<br/>Tel/Fax: +86-10-62562563 +<br/>ISSN 1000-9825, CODEN RUXUEW +<br/>Journal of Software, Vol.17, No.3, March 2006, pp.525−534 +<br/>DOI: 10.1360/jos170525 +<br/>© 2006 by Journal of Software. All rights reserved. +<br/>基于 3D 人脸重建的光照、姿态不变人脸识别 +<br/>柴秀娟 1+, 山世光 2, 卿来云 2, 陈熙霖 2, 高 文 1,2 +<br/>1(哈尔滨工业大学 计算机学院,黑龙江 哈尔滨 150001) +<br/>2(中国科学院 计算技术研究所 ICT-ISVISION 面像识别联合实验室,北京 100080) +<br/>Pose and Illumination Invariant Face Recognition Based on 3D Face Reconstruction +<br/><b>Harbin Institute of Technology, Harbin 150001, China</b><br/><b>ICT-ISVISION Joint RandD Laboratory for Face Recognition, Institute of Computer Technology, The Chinese Academy of Sciences</b><br/>Beijing 100080, China) +<br/>Chai XJ, Shan SG, Qing LY, Chen XL, Gao W. Pose and illumination invariant face recognition based on 3D +<br/>face reconstruction. Journal of Software, 2006,17(3):525−534. http://www.jos.org.cn/1000-9825/17/525.htm +</td><td>('2100752', 'GAO Wen', 'gao wen')</td><td>E-mail: jos@iscas.ac.cn +<br/>+ Corresponding author: Phn: +86-10-58858300 ext 314, Fax: +86-10-58858301, E-mail: xjchai@jdl.ac.cn, http://www.jdl.ac.cn/ +</td></tr><tr><td>bd0265ba7f391dc3df9059da3f487f7ef17144df</td><td>Data-Driven Sparse Sensor Placement +<br/><b>University of Washington, Seattle, WA 98195, United States</b><br/><b>University of Washington, Seattle, WA 98195, United States</b><br/><b>University of Washington, Seattle, WA 98195, United States</b></td><td>('37119658', 'Krithika Manohar', 'krithika manohar')<br/>('1824880', 'Bingni W. Brunton', 'bingni w. brunton')<br/>('1937069', 'J. Nathan Kutz', 'j. nathan kutz')<br/>('3083169', 'Steven L. Brunton', 'steven l. brunton')</td><td></td></tr><tr><td>bd572e9cbec095bcf5700cb7cd73d1cdc2fe02f4</td><td>Hindawi +<br/>Computational Intelligence and Neuroscience +<br/>Volume 2018, Article ID 7068349, 13 pages +<br/>https://doi.org/10.1155/2018/7068349 +<br/>Review Article +<br/>Deep Learning for Computer Vision: A Brief Review +<br/><b>Technological Educational Institute of Athens, 12210 Athens, Greece</b><br/><b>National Technical University of Athens, 15780 Athens, Greece</b><br/>Received 17 June 2017; Accepted 27 November 2017; Published 1 February 2018 +<br/>Academic Editor: Diego Andina +<br/>License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly +<br/>cited. +<br/>Over the last years deep learning methods have been shown to outperform previous state-of-the-art machine learning techniques +<br/>in several fields, with computer vision being one of the most prominent cases. This review paper provides a brief overview of some +<br/>of the most significant deep learning schemes used in computer vision problems, that is, Convolutional Neural Networks, Deep +<br/>Boltzmann Machines and Deep Belief Networks, and Stacked Denoising Autoencoders. A brief account of their history, structure, +<br/>advantages, and limitations is given, followed by a description of their applications in various computer vision tasks, such as object +<br/>detection, face recognition, action and activity recognition, and human pose estimation. Finally, a brief overview is given of future +<br/>directions in designing deep learning schemes for computer vision problems and the challenges involved therein. +<br/>1. Introduction +<br/>Deep learning allows computational models of multiple +<br/>processing layers to learn and represent data with multiple +</td><td>('3393001', 'Nikolaos Doulamis', 'nikolaos doulamis')<br/>('2594647', 'Athanasios Voulodimos', 'athanasios voulodimos')<br/>('3393144', 'Anastasios Doulamis', 'anastasios doulamis')<br/>('1806369', 'Eftychios Protopapadakis', 'eftychios protopapadakis')<br/>('2594647', 'Athanasios Voulodimos', 'athanasios voulodimos')</td><td>Correspondence should be addressed to Athanasios Voulodimos; thanosv@mail.ntua.gr +</td></tr><tr><td>bd6099429bb7bf248b1fd6a1739e744512660d55</td><td>Submitted 11/09; Revised 5/10; Published 8/10 +<br/>Regularized Discriminant Analysis, Ridge Regression and Beyond +<br/><b>College of Computer Science and Technology</b><br/><b>Zhejiang University</b><br/>Hangzhou, Zhejiang 310027, China +<br/>Computer Science Division and Department of Statistics +<br/><b>University of California</b><br/>Berkeley, CA 94720-1776, USA +<br/>Editor: Inderjit Dhillon +</td><td>('1739312', 'Zhihua Zhang', 'zhihua zhang')<br/>('1779165', 'Guang Dai', 'guang dai')<br/>('1682914', 'Congfu Xu', 'congfu xu')<br/>('1694621', 'Michael I. Jordan', 'michael i. jordan')</td><td>ZHZHANG@ZJU.EDU.CN +<br/>GUANG.GDAI@GMAIL.COM +<br/>XUCONGFU@ZJU.EDU.CN +<br/>JORDAN@CS.BERKELEY.EDU +</td></tr><tr><td>bd0e100a91ff179ee5c1d3383c75c85eddc81723</td><td>Okutama-Action: An Aerial View Video Dataset for Concurrent Human Action +<br/>Detection∗ +<br/><b>Technical University of Munich, Munich, 2KTH Royal Institute of Technology, Stockholm</b><br/><b>Polytechnic University of Catalonia, Barcelona, 4National Taiwan University, Taipei, 5University of</b><br/><b>Tokyo, Tokyo, 6National Institute of Informatics, Tokyo</b></td><td>('39393520', 'Mohammadamin Barekatain', 'mohammadamin barekatain')<br/>('19185012', 'Hsueh-Fu Shih', 'hsueh-fu shih')<br/>('47427148', 'Samuel Murray', 'samuel murray')<br/>('1943224', 'Kotaro Nakayama', 'kotaro nakayama')<br/>('47972365', 'Yutaka Matsuo', 'yutaka matsuo')<br/>('2356111', 'Helmut Prendinger', 'helmut prendinger')</td><td>m.barekatain@tum.de, miquelmr@kth.se, r03945026@ntu.edu.tw, samuelmu@kth.se, +<br/>nakayama@weblab.t.u-tokyo.ac.jp, matsuo@weblab.t.u-tokyo.ac.jp, helmut@nii.ac.jp +</td></tr><tr><td>bd8f3fef958ebed5576792078f84c43999b1b207</td><td>BUAA-iCC at ImageCLEF 2015 Scalable +<br/>Concept Image Annotation Challenge +<br/><b>Intelligent Recognition and Image Processing Lab, Beihang University, Beijing</b><br/>100191, P.R.China +<br/>http://irip.buaa.edu.cn/ +<br/><b>School of Information Technology and Management, University of International</b><br/>Business and Economics, Beijing 100029, P.R.China +</td><td>('40013375', 'Yunhong Wang', 'yunhong wang')<br/>('2097309', 'Jiaxin Chen', 'jiaxin chen')<br/>('34288046', 'Ningning Liu', 'ningning liu')<br/>('1712838', 'Li Zhang', 'li zhang')</td><td>yhwang@buaa.edu.cn; chenjiaxinX@gmail.com. +<br/>ningning.liu@uibe.edu.cn +</td></tr><tr><td>bd9eb65d9f0df3379ef96e5491533326e9dde315</td><td></td><td></td><td></td></tr><tr><td>bd07d1f68486052b7e4429dccecdb8deab1924db</td><td></td><td></td><td></td></tr><tr><td>bd0201b32e7eca7818468f2b5cb1fb4374de75b9</td><td> International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056 +<br/> Volume: 02 Issue: 02 | May-2015 www.irjet.net p-ISSN: 2395-0072 +<br/>FACIAL EMOTION EXPRESSIONS RECOGNITION WITH BRAIN ACTIVITES +<br/>USING KINECT SENSOR V2 +<br/>Ph.D student Hesham A. ALABBASI, Doctoral School of Automatic Control and Computers, +<br/><b>University POLITEHNICA of Bucharest, Bucharest, Romania</b><br/>Bucharest, Bucharest, Romania. +<br/><b>Alin Moldoveanu, Faculty of Automatic Control and Computers, University POLITEHNICA of Bucharest</b><br/>Bucharest, Romania. +<br/><b>Ph.D student Zaid Shhedi, Doctoral School of Automatic Control and Computers, University</b><br/>POLITEHNICA of Bucharest, Bucharest, Romania. +<br/>is emotional +<br/>sensor, Face tracking SDK, Neural network, Brain +<br/>activities. +<br/>Key Words: Facial expressions, Facial features, Kinect +<br/>visual Studio 2013 (C++) and Matlab 2015 to recognize +<br/>eight expressions. +<br/>---------------------------------------------------------------------***--------------------------------------------------------------------- +</td><td>('3124644', 'Florica Moldoveanu', 'florica moldoveanu')</td><td></td></tr><tr><td>bd8e2d27987be9e13af2aef378754f89ab20ce10</td><td></td><td></td><td></td></tr><tr><td>bd236913cfe07896e171ece9bda62c18b8c8197e</td><td>Deep Learning with Energy-efficient Binary Gradient Cameras +<br/>∗NVIDIA, +<br/><b>Carnegie Mellon University</b></td><td>('39131476', 'Suren Jayasuriya', 'suren jayasuriya')<br/>('39775678', 'Orazio Gallo', 'orazio gallo')<br/>('2931118', 'Jinwei Gu', 'jinwei gu')<br/>('1690538', 'Jan Kautz', 'jan kautz')</td><td></td></tr><tr><td>bd379f8e08f88729a9214260e05967f4ca66cd65</td><td>Learning Compositional Visual Concepts with Mutual Consistency +<br/><b>School of Electrical and Computer Engineering, Cornell University, Ithaca NY</b><br/><b>Nancy E. and Peter C. Meinig School of Biomedical Engineering, Cornell University, Ithaca NY</b><br/>3Siemens Corporate Technology, Princeton NJ +<br/>Figure 1: We propose ConceptGAN, a framework that can jointly learn, transfer and compose concepts to generate semantically meaningful +<br/>images, even in subdomains with no training data (highlighted) while the state-of-the-art methods such as CycleGAN [49] fail to do so. +</td><td>('3303727', 'Yunye Gong', 'yunye gong')<br/>('1976152', 'Srikrishna Karanam', 'srikrishna karanam')<br/>('3311781', 'Ziyan Wu', 'ziyan wu')<br/>('2692770', 'Kuan-Chuan Peng', 'kuan-chuan peng')<br/>('39497207', 'Jan Ernst', 'jan ernst')<br/>('1767099', 'Peter C. Doerschuk', 'peter c. doerschuk')</td><td>{yg326,pd83}@cornell.edu,{first.last}@siemens.com +</td></tr><tr><td>bd13f50b8997d0733169ceba39b6eb1bda3eb1aa</td><td>Occlusion Coherence: Detecting and Localizing Occluded Faces +<br/><b>University of California at Irvine, Irvine, CA</b></td><td>('1898210', 'Golnaz Ghiasi', 'golnaz ghiasi')<br/>('3157443', 'Charless C. Fowlkes', 'charless c. fowlkes')</td><td></td></tr><tr><td>bd21109e40c26af83c353a3271d0cd0b5c4b4ade</td><td>Attentive Sequence to Sequence Translation for Localizing Clips of Interest +<br/>by Natural Language Descriptions +<br/><b>Zhejiang University</b><br/><b>University of Technology Sydney</b><br/><b>Zhejiang University</b><br/><b>University of Technology Sydney</b><br/><b>Hikvision Research Institute</b></td><td>('1819984', 'Ke Ning', 'ke ning')<br/>('2948393', 'Linchao Zhu', 'linchao zhu')<br/>('50140409', 'Ming Cai', 'ming cai')<br/>('1698559', 'Yi Yang', 'yi yang')<br/>('2603725', 'Di Xie', 'di xie')</td><td>ningke@zju.edu.cn +<br/>zhulinchao7@gmail.com +<br/>Yi.Yang@uts.edu.au +<br/>xiedi@hikvision.com +</td></tr><tr><td>bd8b7599acf53e3053aa27cfd522764e28474e57</td><td>Learning Long Term Face Aging Patterns +<br/>from Partially Dense Aging Databases +<br/>Jinli Suo1,2,3 +<br/><b>Graduate University of Chinese Academy of Sciences(CAS), 100190, China</b><br/>2Key Lab of Intelligent Information Processing of CAS, +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>Lotus Hill Institute for Computer Vision and Information Science, 436000, China</b><br/><b>School of Electronic Engineering and Computer Science, Peking University, 100871, China</b></td><td>('1698902', 'Wen Gao', 'wen gao')<br/>('1710220', 'Xilin Chen', 'xilin chen')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')</td><td>wgao@pku.edu.cn +<br/>jlsuo@jdl.ac.cn +<br/>{xlchen,sgshan}@ict.ac.cn +</td></tr><tr><td>bd8f77b7d3b9d272f7a68defc1412f73e5ac3135</td><td>SphereFace: Deep Hypersphere Embedding for Face Recognition +<br/><b>Georgia Institute of Technology</b><br/><b>Carnegie Mellon University</b><br/><b>Sun Yat-Sen University</b></td><td>('36326884', 'Weiyang Liu', 'weiyang liu')<br/>('1751019', 'Zhiding Yu', 'zhiding yu')<br/>('1779453', 'Le Song', 'le song')</td><td>wyliu@gatech.edu, {yandongw,yzhiding}@andrew.cmu.edu, lsong@cc.gatech.edu +</td></tr><tr><td>bd26dabab576adb6af30484183c9c9c8379bf2e0</td><td>SCUT-FBP: A Benchmark Dataset for +<br/>Facial Beauty Perception +<br/>School of Electronic and Information Engineering +<br/><b>South China University of Technology, Guangzhou 510640, China</b></td><td>('2361818', 'Duorui Xie', 'duorui xie')<br/>('2521432', 'Lingyu Liang', 'lingyu liang')<br/>('1703322', 'Lianwen Jin', 'lianwen jin')<br/>('1720015', 'Jie Xu', 'jie xu')<br/>('4997446', 'Mengru Li', 'mengru li')</td><td>*Email: lianwen.jin@gmail.com +</td></tr><tr><td>bd78a853df61d03b7133aea58e45cd27d464c3cf</td><td>A Sparse Representation Approach to Facial +<br/>Expression Recognition Based on LBP plus LFDA +<br/>Computer science and Engineering Department, +<br/><b>Government College of Engineering, Aurangabad [Autonomous</b><br/>Station Road, Aurangabad, Maharashtra, India. +</td><td></td><td></td></tr><tr><td>bd9c9729475ba7e3b255e24e7478a5acb393c8e9</td><td>Interpretable Partitioned Embedding for Customized Fashion Outfit +<br/>Composition +<br/><b>Zhejiang University, Hangzhou, China</b><br/><b>Arizona State University, Phoenix, Arizona</b><br/>♭Alibaba Group, Hangzhou, China +</td><td>('7357719', 'Zunlei Feng', 'zunlei feng')<br/>('46218293', 'Zhenyun Yu', 'zhenyun yu')<br/>('7607499', 'Yezhou Yang', 'yezhou yang')<br/>('9633703', 'Yongcheng Jing', 'yongcheng jing')<br/>('46179768', 'Junxiao Jiang', 'junxiao jiang')<br/>('1727111', 'Mingli Song', 'mingli song')</td><td></td></tr><tr><td>bd2d7c7f0145028e85c102fe52655c2b6c26aeb5</td><td>Attribute-based People Search: Lessons Learnt from a +<br/>Practical Surveillance System +<br/>Rogerio Feris +<br/>IBM Watson +<br/>http://rogerioferis.com +<br/>Russel Bobbitt +<br/>IBM Watson +<br/>Lisa Brown +<br/>IBM Watson +<br/>IBM Watson +</td><td>('1767897', 'Sharath Pankanti', 'sharath pankanti')</td><td>bobbitt@us.ibm.com +<br/>lisabr@us.ibm.com +<br/>sharat@us.ibm.com +</td></tr><tr><td>bd9157331104a0708aa4f8ae79b7651a5be797c6</td><td>SLAC: A Sparsely Labeled Dataset for Action Classification and Localization +<br/><b>Massachusetts Institute of Technology, 2Facebook Applied Machine Learning, 3Dartmouth College</b></td><td>('1683002', 'Hang Zhao', 'hang zhao')<br/>('3305169', 'Zhicheng Yan', 'zhicheng yan')<br/>('1804138', 'Heng Wang', 'heng wang')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')<br/>('1690178', 'Antonio Torralba', 'antonio torralba')</td><td>{hangzhao, torralba}@mit.edu, {zyan3, hengwang, torresani}@fb.com +</td></tr><tr><td>bdbba95e5abc543981fb557f21e3e6551a563b45</td><td>International Journal of Computational Intelligence and Applications +<br/>Vol. 17, No. 2 (2018) 1850008 (15 pages) +<br/>#.c The Author(s) +<br/>DOI: 10.1142/S1469026818500086 +<br/>Speeding up the Hyperparameter Optimization of Deep +<br/>Convolutional Neural Networks +<br/>Knowledge Technology, Department of Informatics +<br/>Universit€at Hamburg +<br/>Vogt-K€olln-Str. 30, Hamburg 22527, Germany +<br/>Received 15 August 2017 +<br/>Accepted 23 March 2018 +<br/>Published 18 June 2018 +<br/>Most learning algorithms require the practitioner to manually set the values of many hyper- +<br/>parameters before the learning process can begin. However, with modern algorithms, the +<br/>evaluation of a given hyperparameter setting can take a considerable amount of time and the +<br/>search space is often very high-dimensional. We suggest using a lower-dimensional represen- +<br/>tation of the original data to quickly identify promising areas in the hyperparameter space. This +<br/>information can then be used to initialize the optimization algorithm for the original, higher- +<br/>dimensional data. We compare this approach with the standard procedure of optimizing the +<br/>hyperparameters only on the original input. +<br/>We perform experiments with various state-of-the-art hyperparameter optimization algo- +<br/>rithms such as random search, the tree of parzen estimators (TPEs), sequential model-based +<br/>algorithm con¯guration (SMAC), and a genetic algorithm (GA). Our experiments indicate that +<br/>it is possible to speed up the optimization process by using lower-dimensional data repre- +<br/>sentations at the beginning, while increasing the dimensionality of the input later in the opti- +<br/>mization process. This is independent of the underlying optimization procedure, making the +<br/>approach promising for many existing hyperparameter optimization algorithms. +<br/>Keywords: Hyperparameter optimization; hyperparameter importance; convolutional neural +<br/>networks; genetic algorithm; Bayesian optimization. +<br/>1. Introduction +<br/>The performance of many contemporary machine learning algorithms depends cru- +<br/>cially on the speci¯c initialization of hyperparameters such as the general architec- +<br/>ture, the learning rate, regularization parameters, and many others.1,2 Indeed, +<br/>This is an Open Access article published by World Scienti¯c Publishing Company. It is distributed under +<br/>the terms of the Creative Commons Attribution 4.0 (CC-BY) License. Further distribution of this work is +<br/>permitted, provided the original work is properly cited. +<br/>1850008-1 +<br/>Int. J. Comp. Intel. Appl. 2018.17. Downloaded from www.worldscientific.comby WSPC on 07/18/18. Re-use and distribution is strictly not permitted, except for Open Access articles. </td><td>('11634287', 'Tobias Hinz', 'tobias hinz')<br/>('2632932', 'Sven Magg', 'sven magg')<br/>('1736513', 'Stefan Wermter', 'stefan wermter')</td><td>*hinz@informatik.uni-hamburg.de +<br/>†navarro@informatik.uni-hamburg.de +<br/>‡magg@informatik.uni-hamburg.de +<br/>wermter@informatik.uni-hamburg.de +</td></tr><tr><td>bd70f832e133fb87bae82dfaa0ae9d1599e52e4b</td><td>Combining Classifier for Face Identification +<br/><b>HCI Lab., Samsung Advanced Institute of Technology, Yongin, Korea</b><br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, Guildford, UK</b></td><td>('1700968', 'Tae-Kyun Kim', 'tae-kyun kim')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td>taekyun@sait.samsung.co.kr +<br/>J.Kittler@surrey.ac.uk +</td></tr><tr><td>d1dfdc107fa5f2c4820570e369cda10ab1661b87</td><td>Super SloMo: High Quality Estimation of Multiple Intermediate Frames +<br/>for Video Interpolation +<br/>Erik Learned-Miller1 +<br/>1UMass Amherst +<br/>2NVIDIA 3UC Merced +</td><td>('40175280', 'Huaizu Jiang', 'huaizu jiang')<br/>('3232265', 'Deqing Sun', 'deqing sun')<br/>('2745026', 'Varun Jampani', 'varun jampani')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')<br/>('1690538', 'Jan Kautz', 'jan kautz')</td><td>{hzjiang,elm}@cs.umass.edu,{deqings,vjampani,jkautz}@nvidia.com, mhyang@ucmerced.edu +</td></tr><tr><td>d185f4f05c587e23c0119f2cdfac8ea335197ac0</td><td> 33 +<br/>Chapter III +<br/>Facial Expression Analysis, +<br/>Modeling and Synthesis: +<br/>Overcoming the Limitations of +<br/>Artificial Intelligence with the Art +<br/>of the Soluble +<br/><b>Eindhoven University of Technology, The Netherlands</b><br/><b>Ritsumeikan University, Japan</b></td><td>('1728894', 'Christoph Bartneck', 'christoph bartneck')<br/>('1709339', 'Michael J. Lyons', 'michael j. lyons')</td><td></td></tr><tr><td>d140c5add2cddd4a572f07358d666fe00e8f4fe1</td><td>Statistically Learned Deformable Eye Models +<br/><b>Imperial College London</b></td><td>('2575567', 'Joan Alabort-i-Medina', 'joan alabort-i-medina')<br/>('37539937', 'Bingqing Qu', 'bingqing qu')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td></td></tr><tr><td>d1dae2993bdbb2667d1439ff538ac928c0a593dc</td><td>International Journal of Computational Intelligence and Informatics, Vol. 3: No. 1, April - June 2013 +<br/>Gamma Correction Technique Based Feature Extraction +<br/>for Face Recognition System +<br/>P Kumar +<br/>Electronics and Communication Engineering +<br/><b>K S Rangasamy College of Technology</b><br/>Electronics and Communication Engineering +<br/><b>K S Rangasamy College of Technology</b><br/>Tamilnadu, India +<br/>Tamilnadu, India +</td><td>('9316812', 'B Vinothkumar', 'b vinothkumar')</td><td>Vinoeee58@gmail.com +<br/>kumar@ksrct.ac.in +</td></tr><tr><td>d1f58798db460996501f224fff6cceada08f59f9</td><td>Transferrable Representations for Visual Recognition +<br/>Jeffrey Donahue +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2017-106 +<br/>http://www2.eecs.berkeley.edu/Pubs/TechRpts/2017/EECS-2017-106.html +<br/>May 14, 2017 +</td><td></td><td></td></tr><tr><td>d115c4a66d765fef596b0b171febca334cea15b5</td><td>Combining Stacked Denoising Autoencoders and +<br/>Random Forests for Face Detection +<br/><b>Swansea University</b><br/>Singleton Park, Swansea SA2 8PP, United Kingdom +<br/>http://csvision.swan.ac.uk +</td><td>('6248353', 'Jingjing Deng', 'jingjing deng')<br/>('2168049', 'Xianghua Xie', 'xianghua xie')<br/>('13154093', 'Michael Edwards', 'michael edwards')</td><td>*x.xie@swansea.ac.uk +</td></tr><tr><td>d1a43737ca8be02d65684cf64ab2331f66947207</td><td>IJB–S: IARPA Janus Surveillance Video Benchmark (cid:3) +<br/>Kevin O’Connor z +</td><td>('1718102', 'Nathan D. Kalka', 'nathan d. kalka')<br/>('48889427', 'Stephen Elliott', 'stephen elliott')<br/>('8033275', 'Brianna Maze', 'brianna maze')<br/>('40205896', 'James A. Duncan', 'james a. duncan')<br/>('40577714', 'Julia Bryan', 'julia bryan')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>d122d66c51606a8157a461b9d7eb8b6af3d819b0</td><td>Vol-3 Issue-4 2017 +<br/>IJARIIE-ISSN(O)-2395-4396 +<br/>AUTOMATED RECOGNITION OF FACIAL +<br/>EXPRESSIONS +<br/><b>METs Institute of Engineering</b><br/>Adgoan,Nashik,Maharashtra. +<br/>Adgoan, Nashik, Maharashtra. +<br/> +</td><td></td><td></td></tr><tr><td>d142e74c6a7457e77237cf2a3ded4e20f8894e1a</td><td>HUMAN EMOTION ESTIMATION FROM +<br/>EEG AND FACE USING STATISTICAL +<br/>FEATURES AND SVM +<br/>1,3Department of Information Technologies, +<br/><b>University of telecommunications and post, Sofia, Bulgaria</b><br/> 2,4Department of Telecommunications, +<br/><b>University of telecommunications and post, Sofia, Bulgaria</b></td><td>('40110188', 'Strahil Sokolov', 'strahil sokolov')<br/>('3050423', 'Yuliyan Velchev', 'yuliyan velchev')<br/>('2283935', 'Svetla Radeva', 'svetla radeva')<br/>('2512835', 'Dimitar Radev', 'dimitar radev')</td><td></td></tr><tr><td>d1082eff91e8009bf2ce933ac87649c686205195</td><td>(will be inserted by the editor) +<br/>Pruning of Error Correcting Output Codes by +<br/>Optimization of Accuracy-Diversity Trade off +<br/>S¨ureyya ¨Oz¨o˘g¨ur Aky¨uz · Terry +<br/>Windeatt · Raymond Smith +<br/>Received: date / Accepted: date +</td><td></td><td></td></tr><tr><td>d1959ba4637739dcc6cc6995e10fd41fd6604713</td><td><b>Rochester Institute of Technology</b><br/>RIT Scholar Works +<br/>Theses +<br/>5-2017 +<br/>Thesis/Dissertation Collections +<br/>Deep Learning for Semantic Video Understanding +<br/>Follow this and additional works at: http://scholarworks.rit.edu/theses +<br/>Recommended Citation +<br/><b>Kulhare, Sourabh, "Deep Learning for Semantic Video Understanding" (2017). Thesis. Rochester Institute of Technology. Accessed</b><br/>from +<br/>This Thesis is brought to you for free and open access by the Thesis/Dissertation Collections at RIT Scholar Works. It has been accepted for inclusion +</td><td>('10376365', 'Sourabh Kulhare', 'sourabh kulhare')</td><td>sk1846@rit.edu +<br/>in Theses by an authorized administrator of RIT Scholar Works. For more information, please contact ritscholarworks@rit.edu. +</td></tr><tr><td>d1881993c446ea693bbf7f7d6e750798bf958900</td><td>Large-Scale YouTube-8M Video Understanding with Deep Neural Networks +<br/><b>Institute for System Programming</b><br/><b>Institute for System Programming</b><br/>ispras.ru +</td><td>('34125461', 'Manuk Akopyan', 'manuk akopyan')<br/>('19228325', 'Eshsou Khashba', 'eshsou khashba')</td><td>manuk@ispras.ru +</td></tr><tr><td>d1d6f1d64a04af9c2e1bdd74e72bd3ffac329576</td><td>Neural Face Editing with Intrinsic Image Disentangling +<br/><b>Stony Brook University 2Adobe Research 3 CentraleSup elec, Universit e Paris-Saclay</b></td><td>('2496409', 'Zhixin Shu', 'zhixin shu')</td><td>1{zhshu,samaras}@cs.stonybrook.edu +<br/>2{yumer,hadap,sunkaval,elishe}@adobe.com +</td></tr><tr><td>d69df51cff3d6b9b0625acdcbea27cd2bbf4b9c0</td><td></td><td></td><td></td></tr><tr><td>d61578468d267c2d50672077918c1cda9b91429b</td><td>Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/> A Monthly Journal of Computer Science and Information Technology +<br/>ISSN 2320–088X +<br/>IJCSMC, Vol. 3, Issue. 9, September 2014, pg.314 – 323 +<br/> RESEARCH ARTICLE +<br/>Face Image Retrieval Using Pose Specific +<br/>Set Sparse Feature Representation +<br/><b>Viswajyothi College of Engineering and Technology Kerala, India</b><br/><b>Viswajyothi College of Engineering and Technology Kerala, India</b></td><td>('3163376', 'Sebastian George', 'sebastian george')</td><td>afeefengg@gmail.com +</td></tr><tr><td>d687fa99586a9ad229284229f20a157ba2d41aea</td><td>Journal of Intelligent Learning Systems and Applications, 2013, 5, 115-122 +<br/>http://dx.doi.org/10.4236/jilsa.2013.52013 Published Online May 2013 (http://www.scirp.org/journal/jilsa) +<br/>115 +<br/>Face Recognition Based on Wavelet Packet Coefficients +<br/>and Radial Basis Function Neural Networks +<br/><b>Virudhunagar Hindu Nadars Senthikumara Nadar College, Virudhunagar</b><br/><b>Computer Applications, Ayya Nadar Janaki Ammal College, Sivakasi, India</b><br/>Received December 12th, 2012; revised April 19th, 2013; accepted April 26th, 2013 +<br/>tributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any me- +<br/>dium, provided the original work is properly cited. +</td><td></td><td>Email: *kathirvalavakumar@yahoo.com, jebaarul07@yahoo.com +</td></tr><tr><td>d69719b42ee53b666e56ed476629a883c59ddf66</td><td>Learning Facial Action Units from Web Images with +<br/>Scalable Weakly Supervised Clustering +<br/>Aleix M. Martinez3 +<br/><b>School of Comm. and Info. Engineering, Beijing University of Posts and Telecom</b><br/><b>Robotics Institute, Carnegie Mellon University</b><br/><b>The Ohio State University</b></td><td>('2393320', 'Kaili Zhao', 'kaili zhao')</td><td></td></tr><tr><td>d647099e571f9af3a1762f895fd8c99760a3916e</td><td>Exploring Facial Expressions with Compositional Features +<br/><b>Rutgers University</b><br/>110 Frelinghuysen Road, Piscataway, NJ 08854, USA +</td><td>('39606160', 'Peng Yang', 'peng yang')<br/>('1734954', 'Qingshan Liu', 'qingshan liu')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>peyang@cs.rutgers.edu, qsliu@cs.rutgers.edu, dnm@cs.rutgers.edu +</td></tr><tr><td>d69271c7b77bc3a06882884c21aa1b609b3f76cc</td><td>FaceBoxes: A CPU Real-time Face Detector with High Accuracy +<br/><b>CBSR and NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, China</b></td><td>('3220556', 'Shifeng Zhang', 'shifeng zhang')</td><td>{shifeng.zhang,xiangyu.zhu,zlei,hailin.shi,xiaobo.wang,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>d6a9ea9b40a7377c91c705f4c7f206a669a9eea2</td><td>Visual Representations for Fine-grained +<br/>Categorization +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2015-244 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2015/EECS-2015-244.html +<br/>December 17, 2015 +</td><td>('40565777', 'Ning Zhang', 'ning zhang')</td><td></td></tr><tr><td>d6ca3dc01de060871839d5536e8112b551a7f9ff</td><td>Sleep-deprived Fatigue Pattern Analysis using Large-Scale Selfies from Social Media +<br/>Computer Science Department +<br/>Computer Science Department +<br/><b>University of Rochester</b><br/><b>University of Rochester</b><br/>Rochester, USA +<br/>Rochester, USA +<br/>Department of Psychiatry +<br/><b>University of Rochester</b><br/>Rochester, USA +<br/>Computer Science Department +<br/><b>University of Rochester</b><br/>Rochester, USA +</td><td>('1901094', 'Xuefeng Peng', 'xuefeng peng')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')<br/>('39226140', 'Catherine Glenn', 'catherine glenn')<br/>('35678395', 'Li-Kai Chi', 'li-kai chi')<br/>('13171221', 'Jingyao Zhan', 'jingyao zhan')</td><td>xpeng4@u.rochester.edu +<br/>jiebo.luo@rochester.edu +<br/>catherine.glenn@rochester.edu +<br/>{lchi3, jzhan}@u.rochester.edu +</td></tr><tr><td>d671a210990f67eba9b2d3dda8c2cb91575b4a7a</td><td>Journal of Machine Learning Research () +<br/>Submitted ; Published +<br/>Social Environment Description from Data Collected with a +<br/>Wearable Device +<br/>Computer Vision Center +<br/><b>Autonomous University of Barcelona</b><br/>Barcelona, Spain +<br/>Editor: Radeva Petia, Pujol Oriol +</td><td>('7629833', 'Pierluigi Casale', 'pierluigi casale')</td><td>pierluigi@cvc.uab.cat +</td></tr><tr><td>d61e794ec22a4d4882181da17316438b5b24890f</td><td>Detecting Sensor Level Spoof Attacks Using Joint +<br/>Encoding of Temporal and Spatial Features +<br/><b>The Hong Kong Polytechnic University, Hong Kong</b></td><td>('1690410', 'Jun Liu', 'jun liu')<br/>('1684016', 'Ajay Kumar', 'ajay kumar')</td><td></td></tr><tr><td>d65b82b862cf1dbba3dee6541358f69849004f30</td><td>Contents lists available at ScienceDirect +<br/>j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c v i u +<br/>2.5D Elastic graph matching +<br/><b>Imperial College, London, UK</b><br/>a r t i c l e +<br/>i n f o +<br/>a b s t r a c t +<br/>Article history: +<br/>Received 29 November 2009 +<br/>Accepted 1 December 2010 +<br/>Available online 17 March 2011 +<br/>Keywords: +<br/>Elastic graph matching +<br/>3D face recognition +<br/>Multiscale mathematical morphology +<br/>Geodesic distances +<br/>In this paper, we propose novel elastic graph matching (EGM) algorithms for face recognition assisted by +<br/>the availability of 3D facial geometry. More specifically, we conceptually extend the EGM algorithm in +<br/>order to exploit the 3D nature of human facial geometry for face recognition/verification. In order to +<br/>achieve that, first we extend the matching module of the EGM algorithm in order to capitalize on the +<br/>2.5D facial data. Furthermore, we incorporate the 3D geometry into the multiscale analysis used and +<br/>build a novel geodesic multiscale morphological pyramid of dilations/erosions in order to fill the graph +<br/>jets. We show that the proposed advances significantly enhance the performance of EGM algorithms. +<br/>We demonstrate the efficiency of the proposed advances in the face recognition/verification problem +<br/>using photometric stereo. +<br/>Ó 2011 Elsevier Inc. All rights reserved. +<br/>1. Introduction +</td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('2871609', 'Maria Petrou', 'maria petrou')</td><td></td></tr><tr><td>d6102a7ddb19a185019fd2112d2f29d9258f6dec</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>3721 +</td><td></td><td></td></tr><tr><td>d6bfa9026a563ca109d088bdb0252ccf33b76bc6</td><td>Unsupervised Temporal Segmentation of Facial Behaviour +<br/>Department of Computer Science and Engineering, IIT Kanpur +</td><td>('2094658', 'Abhishek Kar', 'abhishek kar')<br/>('2676758', 'Prithwijit Guha', 'prithwijit guha')</td><td>{akar,amit}@iitk.ac.in, prithwijit.guha@tcs.com +</td></tr><tr><td>d67dcaf6e44afd30c5602172c4eec1e484fc7fb7</td><td>Illumination Normalization for Robust Face Recognition +<br/>Using Discrete Wavelet Transform +<br/><b>Mahanakorn University of Technology</b><br/>51 Cheum-Sampan Rd., Nong Chok, Bangkok, THAILAND 10530 +</td><td>('2337544', 'Amnart Petpon', 'amnart petpon')<br/>('1805935', 'Sanun Srisuk', 'sanun srisuk')</td><td>ta tee473@hotmail.com, sanun@mut.ac.th +</td></tr><tr><td>d6c7092111a8619ed7a6b01b00c5f75949f137bf</td><td>A Novel Feature Extraction Technique for Facial Expression +<br/>Recognition +<br/> 1 Department of Computer Science, School of Applied Statistics, +<br/><b>National Institute of Development Administration</b><br/>Bangkok, 10240, Thailand +<br/> +<br/>2 Department of Computer Science, School of Applied Statistics, +<br/><b>National Institute of Development Administration</b><br/>Bangkok, 10240, Thailand +<br/> +</td><td>('7484236', 'Mohammad Shahidul Islam', 'mohammad shahidul islam')<br/>('2291161', 'Surapong Auwatanamongkol', 'surapong auwatanamongkol')</td><td></td></tr><tr><td>d68dbb71b34dfe98dee0680198a23d3b53056394</td><td>VIVA Face-off Challenge: Dataset Creation and Balancing Privacy +<br/><b>University of California, San Diego</b><br/>9500 Gilman Drive, La Jolla, CA 92093 +<br/>1. Introduction +<br/>Vision for intelligent vehicles is a growing area of re- +<br/>search [5] for many practical reasons including the rela- +<br/>tively inexpensive nature of camera sensing units and even +<br/>more the non-contact and non-intrusive manner of obser- +<br/>vation. The latter is of critical importance when observing +<br/>the driver inside the vehicle cockpit because no sensing unit +<br/>should impede the driver’s primary task of driving. One +<br/>of the key tasks in observing the driver is to estimate the +<br/>driver’s gaze direction. From a vision sensing perspective, +<br/>for driver gaze estimation, two of the fundamental building +<br/>blocks are face detection and head pose estimation. +<br/>Figure 1. A sample of challenging instances due to varying illumi- +<br/>nation, occlusions and camera perspectives. +<br/>In literature, vision based systems for face detection and +<br/>head pose estimation have progressed significantly in the +<br/>last decade. However, the limits of the state-of-the-art sys- +<br/>tems have not been tested thoroughly on a common pool +<br/>of challenging dataset as the one we propose in this work. +<br/>Using our database, we want to benchmark existing algo- +<br/>rithms to highlight problems and deficiencies in current +<br/>approaches and, simultaneously, progress the development +<br/>of future algorithms to tackle this problem. Furthermore, +<br/>while introducing a new benchmarking database, we also +<br/>raise awareness of privacy protection systems [4] necessary +<br/>to protect the identity of driver’s in such databases. +<br/>2. In-the Wild Dataset +<br/>In recent years, literature has introduced a few in-the- +<br/>wild datasets (e.g. Helen [2] and COFW [1]) but nothing +<br/>like the challenges from real-world driving scenario are pre- +<br/>sented in such databases. Therefore, we introduce a never +<br/>before seen challenging database of driver’s faces under +<br/>varying illumination (e.g. sunny and cloudy), in the pres- +<br/>ence of typical partially occluding objects (e.g. eyewear and +<br/>hats) or actions (e.g. hand movements),in blur from head +<br/>motions, under different camera configurations and from +<br/>different drivers. A small sample of these challenging in- +<br/>stances are depicted in Figure 1. +<br/>Three major efforts have been put forth in creating this +<br/>challenging database. One is in the data collection itself +<br/>which was done by instrumenting vehicles at UCSD-LISA +<br/>and having multiple drivers drive the instrumented vehicle +<br/>year around. Second is in extracting challenging instances +<br/>from more than a hundred hours of video data. The final +<br/>effort has been in ground truth annotations (e.g. face posi- +<br/>tion and head pose). Preliminary evaluation of the state-of- +<br/>the art head pose algorithms on a small validation part of +<br/>this dataset is shown in Table 1. Here detection rate is the +<br/>number of sample images where an algorithm produced an +<br/>output over the total number of sample images. It is evident +<br/>that no one algorithm is yet to reach high detection rate and +<br/>low error values in head pose. +<br/>3. Balancing Privacy +<br/>In current literature, there is a lack of publicly available +<br/>naturalistic driving data largely due to concerns over indi- +<br/>vidual privacy. Camera sensors looking at a driver, which +</td><td>('1841835', 'Sujitha Martin', 'sujitha martin')<br/>('1713989', 'Mohan M. Trivedi', 'mohan m. trivedi')</td><td>scmartin@ucsd.edu, mtrivedi@ucsd.edu +</td></tr><tr><td>d666ce9d783a2d31550a8aa47da45128a67304a7</td><td>On Relating Visual Elements to City Statistics +<br/><b>University of California, Berkeley</b><br/>Maneesh Agrawala† +<br/><b>University of California, Berkeley</b><br/><b>University of California, Berkeley</b><br/>(c) Visual Elements for Thefts in San Francisco +<br/>(a) Predicted High Theft Location in Oakland +<br/>(b) Predicted Low Theft Location in Oakland +<br/>(d) Predicted Theft Rate in Oakland +<br/>Figure 1: Our system automatically computes a predictor from a set of Google StreetView images of areas where a statistic was observed. In this example +<br/>we use a predictor generated from reports of theft in San Francisco to predict the probability of thefts occurring in Oakland. Our system can predict high +<br/>theft rate areas (a) and low theft rates area (b) based solely on street-level images from the areas. Visually, the high theft area exhibits a marked quality of +<br/>disrepair (bars on the windows, unkempt facades, etc), a visual cue that the probability of theft is likely higher. Our method automatically computes machine +<br/>learning models that detect visual elements similar to these cues (c) from San Francisco. To compute predictions, we use the models to detect the presence of +<br/>these visual elements in an image and combine all of the detections according to an automatically learned set of weights. Our resulting predictions are 63% +<br/>accurate in this case and can be computed everywhere in Oakland (d) as they only rely on images as input. +</td><td>('2288243', 'Sean M. Arietta', 'sean m. arietta')<br/>('1752236', 'Ravi Ramamoorthi', 'ravi ramamoorthi')</td><td></td></tr><tr><td>d6fb606e538763282e3942a5fb45c696ba38aee6</td><td></td><td></td><td></td></tr><tr><td>bcee40c25e8819955263b89a433c735f82755a03</td><td>Biologically inspired vision for human-robot +<br/>interaction +<br/>M. Saleiro, M. Farrajota, K. Terzi´c, S. Krishna, J.M.F. Rodrigues, and J.M.H. +<br/>du Buf +<br/><b>Vision Laboratory, LARSyS, University of the Algarve, 8005-139 Faro, Portugal</b></td><td></td><td>{masaleiro, mafarrajota, kterzic, jrodrig, dubuf}@ualg.pt, +<br/>saikrishnap2003@gmail.com, +</td></tr><tr><td>bc6de183cd8b2baeebafeefcf40be88468b04b74</td><td>Age Group Recognition using Human Facial Images +<br/>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 126 – No.13, September 2015 +<br/>Dept. of Electronics and Telecommunication +<br/><b>Government College of Engineering</b><br/>Aurangabad, Maharashtra, India +</td><td>('31765215', 'Shailesh S. Kulkarni', 'shailesh s. kulkarni')</td><td></td></tr><tr><td>bcf19b964e7d1134d00332cf1acf1ee6184aff00</td><td>1922 +<br/>IEICE TRANS. INF. & SYST., VOL.E100–D, NO.8 AUGUST 2017 +<br/>LETTER +<br/>Trajectory-Set Feature for Action Recognition +<br/>SUMMARY We propose a feature for action recognition called +<br/>Trajectory-Set (TS), on top of the improved Dense Trajectory (iDT). +<br/>The TS feature encodes only trajectories around densely sampled inter- +<br/>est points, without any appearance features. Experimental results on the +<br/>UCF50 action dataset demonstrates that TS is comparable to state-of-the- +<br/>arts, and outperforms iDT; the accuracy of 95.0%, compared to 91.7% by +<br/>iDT. +<br/>key words: action recognition, trajectory, improved Dense Trajectory +<br/>the two-stream CNN [2] that uses a single frame and a opti- +<br/>cal flow stack. In their paper stacking trajectories was also +<br/>reported but did not perform well, probably the sparseness +<br/>of trajectories does not fit to CNN architectures. In contrast, +<br/>we take a hand-crafted approach that can be fused later with +<br/>CNN outputs. +<br/>1. +<br/>Introduction +<br/>Action recognition has been well studied in the computer +<br/>vision literature [1] because it is an important and challeng- +<br/>ing task. Deep learning approaches have been proposed +<br/>recently [2]–[4], however still a hand-crafted feature, im- +<br/>proved Dense Trajectory (iDT) [5], [6], is comparable in +<br/>performance. Moreover, top performances of deep learn- +<br/>ing approaches are obtained by combining the iDT fea- +<br/>ture [3], [7], [8]. +<br/>In this paper, we propose a novel hand-crafted feature +<br/>for action recognition, called Trajectory-Set (TS), that en- +<br/>codes trajectories in a local region of a video. The con- +<br/>tribution of this paper is summarized as follows. We pro- +<br/>pose another hand-crafted feature that can be combined with +<br/>deep learning approaches. Hand-crafted features are com- +<br/>plement to deep learning approaches, however a little effort +<br/>has been done in this direction after iDT. Second, the pro- +<br/>posed TS feature focuses on the better handling of motions +<br/>in the scene. The iDT feature uses trajectories of densely +<br/>samples interest points in a simple way, while we explore +<br/>here the way to extract a rich information from trajectories. +<br/>The proposed TS feature is complement to appearance in- +<br/>formation such as HOG and objects in the scene, which can +<br/>be computed separately and combined afterward in a late +<br/>fusion fashion. +<br/>There are two relate works relevant to our work. One +<br/>is trajectons [9] that uses a global dictionary of trajectories +<br/>in a video to cluster representative trajectories as snippets. +<br/>Our TS feature is computed locally, not globally, inspired +<br/>by the success of local image descriptors [10]. The other is +<br/>Manuscript received March 2, 2017. +<br/>Manuscript revised April 27, 2017. +<br/>Manuscript publicized May 10, 2017. +<br/><b>The authors are with Hiroshima University, Higashihiroshima</b><br/>shi, 739–8527 Japan. +<br/>DOI: 10.1587/transinf.2017EDL8049 +<br/>2. Dense Trajectory +<br/>Here we briefly summarize the improved dense trajectory +<br/>(iDT) [6] on which we base for the proposed method. First, +<br/>the image pyramid for a particular frame at time t in a video +<br/>is constructed, and interest points are densely sampled at +<br/>each level of the pyramid. Next, interest points are tracked +<br/>in the following L frames (L = 15 by default). Then, the +<br/>iDT is computed by using local features such as HOG (His- +<br/>togram of Oriented Gradient) [10], HOF (Histogram of Op- +<br/>tical Flow), and MBH (Motion Boundary Histograms) [11] +<br/>along the trajectory tube; a stack of patches centered at the +<br/>trajectory in the frames. +<br/>, pt1 +<br/>In fact, Tt0,tL +<br/>For example, between two points in time t0 and tL, a +<br/>, . . . , ptL in frames {t0, t1, +<br/>trajectory Tt0,tL has points pt0 +<br/>. . . , tL}. +<br/>is a vector of displacement be- +<br/>tween frames rather than point coordinates, that is, Tt0,tL +<br/>(v0, v1, . . . , vL−1) where vi = pi−1 − pi. Local features such as +<br/>HOGti are computed with a patch centered at pti in frame at +<br/>time ti. +<br/>To improve the performance, the global motion is re- +<br/>moved by computing homography, and background trajec- +<br/>tories are removed by using a people detector. The Fisher +<br/>vector encoding [12] is used to compute an iDT feature of a +<br/>video. +<br/>3. Proposed Trajectory-Set Feature +<br/>We think that extracted trajectories might have rich informa- +<br/>tion discriminative enough for classifying different actions, +<br/>even although trajectories have no appearance information. +<br/>As shown in Fig. 1, different actions are expected to have +<br/>different trajectories, regardless of appearance, texture, or +<br/>shape of the video frame contents. However a single trajec- +<br/>tory Tt0,tL may be severely affected by inaccurate tracking +<br/>results and an irregular motion in the frame. +<br/>We instead propose to aggregate nearby trajectories to +<br/>form a Trajectory-Set (TS) feature. First, a frame is divided +<br/>into non-overlapping cells of M × M pixels as shown in +<br/><b>Copyright c(cid:2) 2017 The Institute of Electronics, Information and Communication Engineers</b></td><td>('47916686', 'Kenji Matsui', 'kenji matsui')<br/>('1744862', 'Toru Tamaki', 'toru tamaki')<br/>('1688940', 'Bisser Raytchev', 'bisser raytchev')<br/>('1686272', 'Kazufumi Kaneda', 'kazufumi kaneda')</td><td>a) E-mail: tamaki@hiroshima-u.ac.jp +</td></tr><tr><td>bc9003ad368cb79d8a8ac2ad025718da5ea36bc4</td><td>Technische Universit¨at M¨unchen +<br/>Bildverstehen und Intelligente Autonome Systeme +<br/>Facial Expression Recognition With A +<br/>Three-Dimensional Face Model +<br/>Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Informatik der Technischen Uni- +<br/>versit¨at M¨unchen zur Erlangung des akademischen Grades eines +<br/>Doktors der Naturwissenschaften +<br/>genehmigten Dissertation. +<br/>Vorsitzender: +<br/>Univ.-Prof. Dr. Johann Schlichter +<br/>Pr¨ufer der Dissertation: 1. Univ.-Prof. Dr. Bernd Radig (i.R.) +<br/>2. Univ.-Prof. Gudrun J. Klinker, Ph.D. +<br/>Die Dissertation wurde am 04.07.2011 bei der Technischen Universit¨at M¨unchen +<br/>eingereicht und durch die Fakult¨at f¨ur Informatik am 02.12.2011 angenommen. +</td><td>('50565622', 'Christoph Mayer', 'christoph mayer')</td><td></td></tr><tr><td>bc15a2fd09df7046e7e8c7c5b054d7f06c3cefe9</td><td>Using Deep Autoencoders for Facial Expression +<br/>Recognition +<br/><b>COMSATS Institute of Information Technology, Islamabad</b><br/><b>Information Technology University (ITU), Punjab, Lahore, Pakistan</b><br/><b>National University of Sciences and Technology (NUST), Islamabad, Pakistan</b></td><td>('24040678', 'Siddique Latif', 'siddique latif')<br/>('1734917', 'Junaid Qadir', 'junaid qadir')</td><td>engr.ussman@gmail.com, slatif.msee15seecs@seecs.edu.pk, junaid.qadir@itu.edu.pk +</td></tr><tr><td>bcc346f4a287d96d124e1163e4447bfc47073cd8</td><td></td><td></td><td></td></tr><tr><td>bc27434e376db89fe0e6ef2d2fabc100d2575ec6</td><td>Faceless Person Recognition; +<br/>Privacy Implications in Social Media +<br/><b>Max-Planck Institute for Informatics</b><br/>Person A training samples. +<br/>Is this person A ? +<br/>Fig. 1: An illustration of one of the scenarios considered: can a vision system +<br/>recognise that the person in the right image is the same as the tagged person in +<br/>the left images, even when the head is obfuscated? +</td><td>('2390510', 'Seong Joon Oh', 'seong joon oh')<br/>('1798000', 'Rodrigo Benenson', 'rodrigo benenson')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td>{joon, benenson, mfritz, schiele}@mpi-inf.mpg.de +</td></tr><tr><td>bcc172a1051be261afacdd5313619881cbe0f676</td><td>978-1-5090-4117-6/17/$31.00 ©2017 IEEE +<br/>2197 +<br/>ICASSP 2017 +</td><td></td><td></td></tr><tr><td>bcfeac1e5c31d83f1ed92a0783501244dde5a471</td><td></td><td></td><td></td></tr><tr><td>bc12715a1ddf1a540dab06bf3ac4f3a32a26b135</td><td>An Analysis of the State of the Art in Multiple Object Tracking +<br/>Tracking the Trackers: +<br/><b>Technical University Munich, Germany</b><br/><b>University of Adelaide, Australia</b><br/>3Photogrammetry and Remote Sensing, ETH Z¨urich, Switzerland +<br/>4TU Darmstadt, Germany +</td><td>('34761498', 'Anton Milan', 'anton milan')<br/>('1803034', 'Konrad Schindler', 'konrad schindler')<br/>('34493380', 'Stefan Roth', 'stefan roth')</td><td></td></tr><tr><td>bc910ca355277359130da841a589a36446616262</td><td>Conditional High-order Boltzmann Machine: +<br/>A Supervised Learning Model for Relation Learning +<br/>1Center for Research on Intelligent Perception and Computing +<br/>National Laboratory of Pattern Recognition +<br/>2Center for Excellence in Brain Science and Intelligence Technology +<br/><b>Institute of Automation, Chinese Academy of Sciences</b></td><td>('39937384', 'Yan Huang', 'yan huang')<br/>('40119691', 'Wei Wang', 'wei wang')<br/>('22985667', 'Liang Wang', 'liang wang')</td><td>{yhuang, wangwei, wangliang}@nlpr.ia.ac.cn +</td></tr><tr><td>bc2852fa0a002e683aad3fb0db5523d1190d0ca5</td><td></td><td></td><td></td></tr><tr><td>bc866c2ced533252f29cf2111dd71a6d1724bd49</td><td>Sensors 2014, 14, 19561-19581; doi:10.3390/s141019561 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>A Multi-Modal Face Recognition Method Using Complete Local +<br/>Derivative Patterns and Depth Maps +<br/><b>Institute of Microelectronics, Tsinghua University, Beijing 100084, China</b><br/>Tel.: +86-10-6279-4398. +<br/>External Editor: Vittorio M.N. Passaro +<br/>Received: 8 August 2014; in revised form: 3 October 2014 / Accepted: 13 October 2014 / +<br/>Published: 20 October 2014 +</td><td>('3817476', 'Shouyi Yin', 'shouyi yin')<br/>('34585208', 'Xu Dai', 'xu dai')<br/>('12263637', 'Peng Ouyang', 'peng ouyang')<br/>('1743798', 'Leibo Liu', 'leibo liu')<br/>('1803672', 'Shaojun Wei', 'shaojun wei')</td><td>E-Mails: daixu@gmail.com (X.D.); oyangpeng12@163.com (P.O.); liulb@tsinghua.edu.cn (L.L.); +<br/>wsj@tsinghua.edu.cn (S.W.) +<br/>* Author to whom correspondence should be addressed; E-Mail: yinsy@tsinghua.edu.cn; +</td></tr><tr><td>bc8e11b8cdf0cfbedde798a53a0318e8d6f67e17</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Deep Learning for Fixed Model Reuse∗ +<br/><b>National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, 210023, China</b><br/>Collaborative Innovation Center of Novel Software Technology and Industrialization, Nanjing, 210023, China +</td><td>('1708973', 'Yang Yang', 'yang yang')<br/>('1721819', 'De-Chuan Zhan', 'de-chuan zhan')<br/>('3750883', 'Ying Fan', 'ying fan')<br/>('2192443', 'Yuan Jiang', 'yuan jiang')<br/>('1692625', 'Zhi-Hua Zhou', 'zhi-hua zhou')</td><td>{yangy, zhandc, fany, jiangy, zhouzh}@lamda.nju.edu.cn +</td></tr><tr><td>bcb99d5150d792001a7d33031a3bd1b77bea706b</td><td></td><td></td><td></td></tr><tr><td>bc811a66855aae130ca78cd0016fd820db1603ec</td><td>Towards three-dimensional face recognition in the real +<br/>To cite this version: +<br/>HAL Id: tel-00998798 +<br/>https://tel.archives-ouvertes.fr/tel-00998798 +<br/>Submitted on 2 Jun 2014 +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>teaching and research institutions in France or +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +</td><td>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')<br/>('47144044', 'Li', 'li')</td><td></td></tr><tr><td>bc98027b331c090448492eb9e0b9721e812fac84</td><td>Journal of Intelligent Learning Systems and Applications, 2012, 4, 266-273 +<br/>http://dx.doi.org/10.4236/jilsa.2012.44027 Published Online November 2012 (http://www.SciRP.org/journal/jilsa) +<br/>Face Representation Using Combined Method of Gabor +<br/>Filters, Wavelet Transformation and DCV and Recognition +<br/>Using RBF +<br/><b>VHNSN College, Virudhunagar, ANJA College</b><br/>Sivakasi, India. +<br/>Received April 27th, 2012; revised July 19th, 2012; accepted July 26th, 2012 +</td><td>('39000426', 'Kathirvalavakumar Thangairulappan', 'kathirvalavakumar thangairulappan')<br/>('15392239', 'Jebakumari Beulah Vasanthi Jeyasingh', 'jebakumari beulah vasanthi jeyasingh')</td><td>Email: *kathirvalavakumar@yahoo.com, jebaarul07@yahoo.com +</td></tr><tr><td>bc9af4c2c22a82d2c84ef7c7fcc69073c19b30ab</td><td>MoCoGAN: Decomposing Motion and Content for Video Generation +<br/>Snap Research +<br/>NVIDIA +</td><td>('1715440', 'Sergey Tulyakov', 'sergey tulyakov')<br/>('9536217', 'Ming-Yu Liu', 'ming-yu liu')<br/>('50030951', 'Xiaodong Yang', 'xiaodong yang')<br/>('1690538', 'Jan Kautz', 'jan kautz')</td><td>stulyakov@snap.com +<br/>{mingyul,xiaodongy,jkautz}@nvidia.com +</td></tr><tr><td>bcac3a870501c5510df80c2a5631f371f2f6f74a</td><td>CVPR +<br/>#1387 +<br/>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>045 +<br/>046 +<br/>047 +<br/>048 +<br/>049 +<br/>050 +<br/>051 +<br/>052 +<br/>053 +<br/>CVPR 2013 Submission #1387. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +<br/>CVPR +<br/>#1387 +<br/>Structured Face Hallucination +<br/>Anonymous CVPR submission +<br/>Paper ID 1387 +</td><td></td><td></td></tr><tr><td>ae8d5be3caea59a21221f02ef04d49a86cb80191</td><td>Published as a conference paper at ICLR 2018 +<br/>SKIP RNN: LEARNING TO SKIP STATE UPDATES IN +<br/>RECURRENT NEURAL NETWORKS +<br/>†Barcelona Supercomputing Center, ‡Google Inc, +<br/><b>Universitat Polit`ecnica de Catalunya, Columbia University</b></td><td>('2447185', 'Brendan Jou', 'brendan jou')<br/>('1711068', 'Jordi Torres', 'jordi torres')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>{victor.campos, jordi.torres}@bsc.es, bjou@google.com, +<br/>xavier.giro@upc.edu, shih.fu.chang@columbia.edu +</td></tr><tr><td>aed321909bb87c81121c841b21d31509d6c78f69</td><td></td><td></td><td></td></tr><tr><td>ae936628e78db4edb8e66853f59433b8cc83594f</td><td></td><td></td><td></td></tr><tr><td>ae0765ebdffffd6e6cc33c7705df33b7e8478627</td><td>Self-Reinforced Cascaded Regression for Face Alignment +<br/><b>DUT-RU International School of Information Science and Engineering, Dalian University of Technology, Dalian, China</b><br/>2Key Laboratory for Ubiquitous Network and Service Software of Liaoning Province, Dalian, China +<br/><b>School of Mathematical Science, Dalian University of Technology, Dalian, China</b></td><td>('1710408', 'Xin Fan', 'xin fan')<br/>('34469457', 'Risheng Liu', 'risheng liu')<br/>('3453975', 'Kang Huyan', 'kang huyan')<br/>('3013708', 'Yuyao Feng', 'yuyao feng')<br/>('7864960', 'Zhongxuan Luo', 'zhongxuan luo')</td><td>{xin.fan, rsliu, zxluo}@dlut.edu.cn, huyankang@hotmail.com yyaofeng@gmail.com +</td></tr><tr><td>aefc7c708269b874182a5c877fb6dae06da210d4</td><td>Deep Learning of Invariant Features via Simulated +<br/>Fixations in Video +<br/><b>Stanford University, CA</b><br/><b>Stanford University, CA</b><br/><b>NEC Laboratories America, Inc., Cupertino, CA</b></td><td>('2860351', 'Will Y. Zou', 'will y. zou')<br/>('1682028', 'Shenghuo Zhu', 'shenghuo zhu')<br/>('1701538', 'Andrew Y. Ng', 'andrew y. ng')<br/>('38701713', 'Kai Yu', 'kai yu')</td><td>{wzou, ang}@cs.stanford.edu +<br/>{zsh, kyu}@sv.nec-labs.com +</td></tr><tr><td>ae2cf545565c157813798910401e1da5dc8a6199</td><td>Mahkonen et al. EURASIP Journal on Image and Video +<br/>Processing (2018) 2018:61 +<br/>https://doi.org/10.1186/s13640-018-0303-9 +<br/>EURASIP Journal on Image +<br/>and Video Processing +<br/>RESEARCH +<br/>Open Access +<br/>Cascade of Boolean detector +<br/>combinations +</td><td>('3292563', 'Katariina Mahkonen', 'katariina mahkonen')</td><td></td></tr><tr><td>aebb9649bc38e878baef082b518fa68f5cda23a5</td><td> +</td><td></td><td></td></tr><tr><td>aeaf5dbb3608922246c7cd8a619541ea9e4a7028</td><td>Weakly Supervised Facial Action Unit Recognition through Adversarial Training +<br/><b>University of Science and Technology of China, Hefei, Anhui, China</b></td><td>('46217896', 'Guozhu Peng', 'guozhu peng')<br/>('1791319', 'Shangfei Wang', 'shangfei wang')</td><td>gzpeng@mail.ustc.edu.cn, sfwang@ustc.edu.cn +</td></tr><tr><td>ae836e2be4bb784760e43de88a68c97f4f9e44a1</td><td>Semi-Supervised Dimensionality Reduction∗ +<br/>1National Laboratory for Novel Software Technology +<br/><b>Nanjing University, Nanjing 210093, China</b><br/>2Department of Computer Science and Engineering +<br/><b>Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China</b></td><td>('51326748', 'Daoqiang Zhang', 'daoqiang zhang')<br/>('46228434', 'Zhi-Hua Zhou', 'zhi-hua zhou')<br/>('1680768', 'Songcan Chen', 'songcan chen')</td><td>dqzhang@nuaa.edu.cn +<br/>zhouzh@nju.edu.cn +<br/>s.chen@nuaa.edu.cn +</td></tr><tr><td>ae5bb02599244d6d88c4fe466a7fdd80aeb91af4</td><td>Analysis of Recognition Algorithms using Linear, Generalized Linear, and +<br/>Generalized Linear Mixed Models +<br/>Dept. of Computer Science +<br/><b>Colorado State University</b><br/>Fort Colllins, CO 80523 +<br/>Dept. of Statistics +<br/><b>Colorado State University</b><br/>Fort Collins, CO 80523 +</td><td>('1757322', 'J. Ross Beveridge', 'j. ross beveridge')<br/>('1750370', 'Geof H. Givens', 'geof h. givens')</td><td></td></tr><tr><td>ae18ccb35a1a5d7b22f2a5760f706b1c11bf39a9</td><td>Sensing Highly Non-Rigid Objects with RGBD +<br/>Sensors for Robotic Systems +<br/>A Dissertation +<br/>Presented to +<br/>the Graduate School of +<br/><b>Clemson University</b><br/>In Partial Fulfillment +<br/>of the Requirements for the Degree +<br/>Doctor of Philosophy +<br/>Computer Engineering +<br/>by +<br/>May 2013 +<br/>Accepted by: +<br/>Dr. Stanley T. Birchfield, Committee Chair +</td><td>('2181472', 'Bryan Willimon', 'bryan willimon')<br/>('26607413', 'Ian D. Walker', 'ian d. walker')<br/>('1724942', 'Adam W. Hoover', 'adam w. hoover')<br/>('2171076', 'Damon L. Woodard', 'damon l. woodard')</td><td></td></tr><tr><td>aeeea6eec2f063c006c13be865cec0c350244e5b</td><td>Induced Disgust, Happiness and Surprise: an Addition to the MMI Facial +<br/>Expression Database +<br/><b>Imperial College London / Twente University</b><br/>Department of Computing / EEMCS +<br/>180 Queen’s Gate / Drienerlolaan 5 +<br/>London / Twente +</td><td>('1795528', 'Michel F. Valstar', 'michel f. valstar')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>Michel.Valstar@imperial.ac.uk, M.Pantic@imperial.ac.uk +</td></tr><tr><td>ae9257f3be9f815db8d72819332372ac59c1316b</td><td>P SY CH O L O GIC AL SC I E NC E +<br/>Research Article +<br/>Deciphering the Enigmatic Face +<br/>The Importance of Facial Dynamics in Interpreting Subtle +<br/>Facial Expressions +<br/><b>University of Pittsburgh and 2University of British Columbia, Vancouver, British Columbia, Canada</b></td><td>('2059653', 'Zara Ambadar', 'zara ambadar')</td><td></td></tr><tr><td>ae89b7748d25878c4dc17bdaa39dd63e9d442a0d</td><td>On evaluating face tracks in movies +<br/>To cite this version: +<br/>in movies. IEEE International Conference on Image Processing (ICIP 2013), Sep 2013, Melbourne, +<br/>Australia. 2013. <hal-00870059> +<br/>HAL Id: hal-00870059 +<br/>https://hal.inria.fr/hal-00870059 +<br/>Submitted on 4 Oct 2013 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('2889451', 'Alexey Ozerov', 'alexey ozerov')<br/>('2712091', 'Jean-Ronan Vigouroux', 'jean-ronan vigouroux')<br/>('39255836', 'Louis Chevallier', 'louis chevallier')<br/>('1799777', 'Patrick Pérez', 'patrick pérez')<br/>('2889451', 'Alexey Ozerov', 'alexey ozerov')<br/>('2712091', 'Jean-Ronan Vigouroux', 'jean-ronan vigouroux')<br/>('39255836', 'Louis Chevallier', 'louis chevallier')<br/>('1799777', 'Patrick Pérez', 'patrick pérez')</td><td></td></tr><tr><td>ae1de0359f4ed53918824271c888b7b36b8a5d41</td><td>Low-cost Automatic Inpainting for Artifact Suppression in Facial Images +<br/>Thomaz4 +<br/><b>Scienti c Visualization and Computer Graphics, University of Groningen, Nijenborgh 9, Groningen, The Netherlands</b><br/>2Department of Computing, National Laboratory of Scientific Computation, Petr´opolis, Brazil +<br/><b>Paran a Federal University, Curitiba, Brazil</b><br/><b>University Center of FEI, S ao Bernardo do Campo, Brazil</b><br/>Keywords: +<br/>Image inpainting, Face reconstruction, Statistical Decision, Image Quality Index +</td><td>('1686665', 'Alexandru Telea', 'alexandru telea')</td><td>{a.sobiecki, a.c.telea}@rug.nl, gilson@lncc.br, neves@ufpr.br, cet@fei.edu.br +</td></tr><tr><td>ae4390873485c9432899977499c3bf17886fa149</td><td>FACIAL EXPRESSION RECOGNITION USING +<br/>DIGITALISED FACIAL FEATURES BASED ON +<br/>ACTIVE SHAPE MODEL +<br/><b>Institute for Arts, Science and Technology</b><br/><b>Glyndwr University</b><br/>Wrexham, United Kingdom +</td><td>('39048426', 'Nan Sun', 'nan sun')<br/>('11832393', 'Zheng Chen', 'zheng chen')<br/>('1818364', 'Richard Day', 'richard day')</td><td>bruce.n.sun@gmail.com1 +<br/>z.chen@glyndwr.ac.uk2 +<br/>r.day@glyndwr.ac.uk3 +</td></tr><tr><td>aeff403079022683b233decda556a6aee3225065</td><td>DeepFace: Face Generation using Deep Learning +</td><td>('31560532', 'Hardie Cate', 'hardie cate')<br/>('6415321', 'Fahim Dalvi', 'fahim dalvi')<br/>('8815003', 'Zeshan Hussain', 'zeshan hussain')</td><td>ccate@stanford.edu +<br/>fdalvi@cs.stanford.edu +<br/>zeshanmh@stanford.edu +</td></tr><tr><td>ae753fd46a744725424690d22d0d00fb05e53350</td><td>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>Describing Clothing by Semantic Attributes +<br/>Anonymous ECCV submission +<br/>Paper ID 727 +</td><td></td><td></td></tr><tr><td>aea4128ba18689ff1af27b90c111bbd34013f8d5</td><td>Efficient k-Support Matrix Pursuit +<br/><b>National University of Singapore</b><br/><b>School of Software, Sun Yat-sen University, China</b><br/><b>School of Information Science and Technology, Sun Yat-sen University, China</b><br/><b>School of Computer Science, South China Normal University, China</b></td><td>('2356867', 'Hanjiang Lai', 'hanjiang lai')<br/>('2493641', 'Yan Pan', 'yan pan')<br/>('33224509', 'Canyi Lu', 'canyi lu')<br/>('1704995', 'Yong Tang', 'yong tang')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>{laihanj,canyilu}@gmail.com, panyan5@mail.sysu.edu.cn, +<br/>ytang@scnu.edu.cn, eleyans@nus.edu.sg +</td></tr><tr><td>ae2c71080b0e17dee4e5a019d87585f2987f0508</td><td>Research Paper: Emotional Face Recognition in Children +<br/>With Attention Deficit/Hyperactivity Disorder: Evidence +<br/>From Event Related Gamma Oscillation +<br/>CrossMark +<br/><b>School of Advanced Technologies in Medicine, Tehran University of Medical Sciences, Tehran, Iran</b><br/><b>School of Medicine, Tehran University of Medical Sciences, Tehran, Iran</b><br/><b>Research Center for Cognitive and Behavioral Sciences, Tehran University of Medical Sciences, Tehran, Iran</b><br/><b>Amirkabir University of Technology, Tehran, Iran</b><br/>Use your device to scan +<br/>and read the article online +<br/>Citation: Sarraf Razavi, M., Tehranidoost, M., Ghassemi, F., Purabassi, P., & Taymourtash, A. (2017). Emotional Face Rec- +<br/>ognition in Children With Attention Deficit/Hyperactivity Disorder: Evidence From Event Related Gamma Oscillation. Basic +<br/>and Clinical Neuroscience, 8(5):419-426. https://doi.org/10.18869/NIRP.BCN.8.5.419 +<br/> : : https://doi.org/10.18869/NIRP.BCN.8.5.419 +<br/>Article info: +<br/>Received: 03 Feb. 2017 +<br/>First Revision: 29 Feb. 2017 +<br/>Accepted: 11 Jul. 2017 +<br/>Key Words: +<br/>Emotional face +<br/>recognition, Event- +<br/>Related Oscillation +<br/>(ERO), Gamma band +<br/>activity, Attention Deficit +<br/>Hyperactivity Disorder +<br/>(ADHD) +<br/>A B S T R A C T +<br/>Introduction: Children with attention-deficit/hyperactivity disorder (ADHD) have some +<br/>impairment in emotional relationship which can be due to problems in emotional processing. +<br/>The present study investigated neural correlates of early stages of emotional face processing in +<br/>this group compared with typically developing children using the Gamma Band Activity (GBA). +<br/>Methods: A total of 19 children diagnosed with ADHD (Combined type) based on DSM-IV +<br/>classification were compared with 19 typically developing children matched on age, gender, and +<br/>IQ. The participants performed an emotional face recognition while their brain activities were +<br/>recorded using an event-related oscillation procedure. +<br/>Results: The results indicated that ADHD children compared to normal group showed a significant +<br/>reduction in the gamma band activity, which is thought to reflect early perceptual emotion +<br/>discrimination for happy and angry emotions (P<0.05). +<br/>Conclusion: The present study supports the notion that individuals with ADHD have some +<br/>impairments in early stage of emotion processing which can cause their misinterpretation of +<br/>emotional faces. +<br/>1. Introduction +<br/>DHD is a common neurodevelopmental +<br/>disorder characterized by inattentiveness +<br/>and hyperactivity/impulsivity (American +<br/>Psychiatric Association, 2013). Individu- +<br/>als with ADHD also show problems in social and emo- +<br/><b>tional functions, including the effective assessment of</b><br/>the emotional state of others. It is important to set the +<br/>adaptive behavior of human facial expressions in social +<br/>interactions (Cadesky, Mota, & Schachar, 2000; Corbett +<br/>& Glidden, 2000). Based on the evidence, frontotem- +<br/>poral-posterior and fronto striatal cerebellar systems +<br/>are involved in emotional functions. These regions may +<br/>contribute to impairments of emotional recognition in +<br/>ADHD (Corbett & Glidden, 2000; Dickstein, Bannon, +<br/>Xavier Castellanos, & Milham, 2006; Durston, Van +<br/>Belle, & De Zeeuw, 2011). +<br/>* Corresponding Author: +<br/><b>Amirkabir University of Technology, Tehran, Iran</b><br/>Tel:+98 (912) 3260661 +<br/>419 +<br/>Basic and ClinicalSeptember, October 2017, Volume 8, Number 5 </td><td>('29928144', 'Mahdiyeh Sarraf Razavi', 'mahdiyeh sarraf razavi')<br/>('7171067', 'Mehdi Tehranidoost', 'mehdi tehranidoost')<br/>('34494047', 'Farnaz Ghassemi', 'farnaz ghassemi')<br/>('29839761', 'Parivash Purabassi', 'parivash purabassi')<br/>('29933673', 'Athena Taymourtash', 'athena taymourtash')<br/>('34494047', 'Farnaz Ghassemi', 'farnaz ghassemi')</td><td>E-mail: ghassemi@aut.ac.ir +</td></tr><tr><td>ae4e2c81c8a8354c93c4b21442c26773352935dd</td><td></td><td></td><td></td></tr><tr><td>ae85c822c6aec8b0f67762c625a73a5d08f5060d</td><td>This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. +<br/>The final version of record is available at http://dx.doi.org/10.1109/TPAMI.2014.2353624 +<br/>IEEE TRANSACTION ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. M, NO. N, MONTH YEAR +<br/>Retrieving Similar Styles to Parse Clothing +</td><td>('1721910', 'Kota Yamaguchi', 'kota yamaguchi')<br/>('1772294', 'M. Hadi Kiapour', 'm. hadi kiapour')<br/>('35258350', 'Luis E. Ortiz', 'luis e. ortiz')<br/>('1685538', 'Tamara L. Berg', 'tamara l. berg')</td><td></td></tr><tr><td>ae5f32e489c4d52e7311b66060c7381d932f4193</td><td>Appearance-and-Relation Networks for Video Classification +<br/><b>State Key Laboratory for Novel Software Technology, Nanjing University, China</b><br/>2Computer Vision Laboratory, ETH Zurich, Switzerland +<br/>3Google Research +</td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('47113208', 'Wei Li', 'wei li')<br/>('50135099', 'Wen Li', 'wen li')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>ae71f69f1db840e0aa17f8c814316f0bd0f6fbbf</td><td>Contents lists available at ScienceDirect +<br/>j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c o m p h u m b e h +<br/>Full length article +<br/>That personal profile image might jeopardize your rental opportunity! +<br/>On the relative impact of the seller's facial expressions upon buying +<br/>behavior on Airbnb™* +<br/>a Faculty of Technology, Westerdals Oslo School of Arts, Communication and Technology, Oslo, Norway +<br/><b>b School of Business, Reykjavik University, Reykjavik, Iceland</b><br/><b>c Cardiff Business School, Cardiff University, Cardiff, United Kingdom</b><br/>a r t i c l e i n f o +<br/>a b s t r a c t +<br/>Article history: +<br/>Received 29 November 2016 +<br/>Received in revised form +<br/>2 February 2017 +<br/>Accepted 9 February 2017 +<br/>Available online 10 February 2017 +<br/>Keywords: +<br/>Sharing economy +<br/>Peer-to-peer +<br/>Facial expressions +<br/>Evolutionary psychology +<br/>Approach and avoidance +<br/>Conjoint study +<br/>Airbnb is an online marketplace for peer-to-peer accommodation rental services. In contrast to tradi- +<br/>tional rental services, personal profile images, i.e. the sellers' facial images, are present along with the +<br/>housing on offer. This study aims to investigate the impact of a seller's facial image and their expression +<br/>upon buyers' behavior in this context. The impact of facial expressions was investigated together with +<br/>other relevant variables (price and customer ratings). Findings from a conjoint study (n ¼ 139) show that +<br/>the impact of a seller's facial expression on buying behavior in an online peer-to-peer context is sig- +<br/>nificant. A negative facial expression and absence of facial image (head silhouette) abates approach and +<br/>evokes avoidance tendencies to explore a specific web page on Airbnb, and, simultaneously decrease the +<br/>likelihood to rent. The reverse effect was true for neutral and positive facial expressions. We found that a +<br/>negative and positive facial expression had more impact on likelihood to rent, for women than for men. +<br/>Further analysis shows that the absence of facial image and an angry facial expression cannot be +<br/>compensated for by a low price and top customer ratings related to likelihood to rent. Practitioners +<br/>should keep in mind that the presence/absence of facial images and their inherent expressions have a +<br/>significant impact in the peer-to-peer accommodation rental services. +<br/>© 2017 Elsevier Ltd. All rights reserved. +<br/>1. Introduction +<br/>The sharing economy, characterized by peer-to-peer trans- +<br/>actions, has seen immense growth recently. These marketplaces are +<br/>defined by direct transactions between individuals (buyers and +<br/>sellers), while the marketplace itself is provided by a third party +<br/>(Botsman & Rogers, 2011). According to a recent survey by Penn +<br/>Schoen Berland (2016), 22% of American adults have already +<br/>offered something to this market, and 42% had used the service to +<br/>buy a product or a service. PricewaterhouseCoopers (PwC) (2014), +<br/>has predicted that these sharing economy sectors will be worth +<br/>* The authors express their thanks to Dr. R. G. Vishnu Menon for assistance with +<br/>the conjoint analysis. +<br/>* Corresponding author. Westerdals Oslo School of Arts, Communication and +<br/>Technology, Faculty of Technology, Christian Kroghs Gate 32, 0186, Oslo, Norway. +<br/>http://dx.doi.org/10.1016/j.chb.2017.02.029 +<br/>0747-5632/© 2017 Elsevier Ltd. All rights reserved. +<br/>around $335 billion by 2025. Their research further indicates that +<br/>the most important growth sectors are lending and crowd funding, +<br/>online staffing, and peer-to-peer accommodation. Participants in +<br/>the peer-to-peer market tend to be motivated by new economic, +<br/>environmental, and social factors (Bucher, Fieseler, & Lutz, 2016; +<br/>B€ocker & Meelen, 2016; Schor, 2014) as this marketplace has +<br/>some additional attributes compared to more traditional forms of +<br/>commerce. The behavior of buyers on the peer-to-peer marketplace +<br/>is, however, not well understood. +<br/>Airbnb is a peer-to-peer platform that facilitates accommoda- +<br/>tion rental services. This marketplace offers intangible experienced +<br/>goods (Levitt, 1981, pp. 94e102), which are typically produced and +<br/>consumed simultaneously (Gr€onroos, 1978). The sellers are co- +<br/>producers of the service experience. Thus, the quality of renting +<br/>an apartment on Airbnb cannot be verified before the buyer has +<br/>started using the service. The Sellers on Airbnb are, therefore, an +<br/>integrated part of the service that is delivered, and are expected to +<br/>fulfill the buyer's needs throughout their stay. Consequently, +</td><td>('2372119', 'Asle Fagerstrøm', 'asle fagerstrøm')<br/>('10665177', 'Sanchit Pawar', 'sanchit pawar')<br/>('3617093', 'Valdimar Sigurdsson', 'valdimar sigurdsson')<br/>('3232722', 'Mirella Yani-De-Soriano', 'mirella yani-de-soriano')</td><td>E-mail address: asle.fagerstrom@westerdals.no (A. Fagerstrøm). +</td></tr><tr><td>d893f75206b122973cdbf2532f506912ccd6fbe0</td><td>Facial Expressions with Some Mixed +<br/>Expressions Recognition Using Neural +<br/>Networks +<br/>Dr.R.Parthasarathi, V.Lokeswar Reddy, K.Vishnuthej, G.Vishnu Vandan +<br/>Department of Information Technology +<br/><b>Pondicherry Engineering College</b><br/>Puducherry-605014, India +</td><td></td><td></td></tr><tr><td>d861c658db2fd03558f44c265c328b53e492383a</td><td>Automated Face Extraction and Normalization of 3D Mesh Data +</td><td>('10423763', 'Jia Wu', 'jia wu')<br/>('1905646', 'Raymond Tse', 'raymond tse')<br/>('1809809', 'Linda G. Shapiro', 'linda g. shapiro')</td><td></td></tr><tr><td>d84a48f7d242d73b32a9286f9b148f5575acf227</td><td>Global and Local Consistent Age Generative +<br/>Adversarial Networks +<br/>Center for Research on Intelligent Perception and Computing, CASIA, Beijing, China +<br/>National Laboratory of Pattern Recognition, CASIA, Beijing, China +<br/><b>University of Chinese Academy of Sciences, Beijing, China</b></td><td>('2112221', 'Peipei Li', 'peipei li')<br/>('33079499', 'Yibo Hu', 'yibo hu')<br/>('39763795', 'Qi Li', 'qi li')<br/>('1705643', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>Email: peipei.li, yibo.hu@cripac.ia.ac.cn, qli,rhe,znsun@nlpr.ia.ac.cn +</td></tr><tr><td>d8f0bda19a345fac81a1d560d7db73f2b4868836</td><td><b>UNIVERSITY OF CALIFORNIA</b><br/>RIVERSIDE +<br/>Online Activity Understanding and Labeling in Natural Videos +<br/>A Dissertation submitted in partial satisfaction +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>in +<br/>Computer Science +<br/>by +<br/>August 2016 +<br/>Dissertation Committee: +<br/>Dr. Amit K. Roy-Chowdhury, Chairperson +<br/>Dr. Eamonn Keogh +<br/>Dr. Evangelos Christidis +<br/>Dr. Christian Shelton +</td><td>('38514801', 'Mahmudul Hasan', 'mahmudul hasan')</td><td></td></tr><tr><td>d82b93f848d5442f82154a6011d26df8a9cd00e7</td><td>NEURAL NETWORK BASED AGE CLASSIFICATION USING +<br/>LINEAR WAVELET TRANSFORMS +<br/>1Department of Computer Science & Engineering, +<br/><b>Sathyabama University Old Mamallapuram Road, Chennai, India</b><br/><b>Electronics Engineering, National Institute of Technical Teachers</b><br/>Training & Research, Taramani, Chennai, India +</td><td></td><td>E-mail : 1nithyaranjith2002@yahoo.co.in, 2gkvel@rediffmail.com +</td></tr><tr><td>d8722ffbca906a685abe57f3b7b9c1b542adfa0c</td><td><b>University of Twente</b><br/>Faculty: Electrical Engineering, Mathematics and Computer Science +<br/>Department: Computer Science +<br/>Group: Human Media Interaction +<br/>Facial Expression Analysis for Human +<br/>Computer Interaction +<br/>Recognizing emotions in an intelligent tutoring system by facial +<br/>expression analysis from a video stream +<br/>M. Ghijsen +<br/>November 2004 +<br/>Examination committee: +<br/>Dr. D.K.J. Heylen +<br/>Prof.dr.ir. A Nijholt +<br/>Dr.ir. H.J.A. op den Akker +<br/>Dr. M. Poel +<br/>Ir. R.J. Rienks +</td><td></td><td></td></tr><tr><td>d8896861126b7fd5d2ceb6fed8505a6dff83414f</td><td>In-Plane Rotational Alignment of Faces by Eye and Eye-Pair Detection +<br/>M.F. Karaaba1, O. Surinta1, L.R.B. Schomaker1 and M.A. Wiering1 +<br/><b>Institute of Arti cial Intelligence and Cognitive Engineering (ALICE), University of Groningen</b><br/>Nijenborgh 9, Groningen 9747AG, The Netherlands +<br/>Keywords: +<br/>Eye-pair Detection, Eye Detection, Face Alignment, Face Recognition, Support Vector Machine +</td><td></td><td>{m.f.karaaba, o.surinta, l.r.b.schomaker, m.a.wiering}@rug.nl +</td></tr><tr><td>d83d2fb5403c823287f5889b44c1971f049a1c93</td><td>Motiv Emot +<br/>DOI 10.1007/s11031-013-9353-6 +<br/>O R I G I N A L P A P E R +<br/>Introducing the sick face +<br/>Ó Springer Science+Business Media New York 2013 +</td><td>('3947094', 'Sherri C. Widen', 'sherri c. widen')</td><td></td></tr><tr><td>d8b568392970b68794a55c090c4dd2d7f90909d2</td><td>PDA Face Recognition System +<br/>Using Advanced Correlation +<br/>Filters +<br/>Chee Kiat Ng +<br/>2005 +<br/>Advisor: Prof. Khosla/Reviere +</td><td></td><td></td></tr><tr><td>d83ae5926b05894fcda0bc89bdc621e4f21272da</td><td>version of the following thesis: +<br/>Frugal Forests: Learning a Dynamic and Cost Sensitive +<br/>Feature Extraction Policy for Anytime Activity Classification +</td><td>('1794409', 'Kristen Grauman', 'kristen grauman')<br/>('1728389', 'Peter Stone', 'peter stone')</td><td></td></tr><tr><td>d86fabd4498c8feaed80ec342d254fb877fb92f5</td><td>Y. GOUTSU: REGION-OBJECT RELEVANCE-GUIDED VRD +<br/>Region-Object Relevance-Guided +<br/>Visual Relationship Detection +<br/><b>National Institute of Informatics</b><br/>Tokyo, Japan +</td><td>('2897806', 'Yusuke Goutsu', 'yusuke goutsu')</td><td>goutsu@nii.ac.jp +</td></tr><tr><td>d8bf148899f09a0aad18a196ce729384a4464e2b</td><td>FACIAL EXPRESSION RECOGNITION AND EXPRESSION +<br/>INTENSITY ESTIMATION +<br/>A dissertation submitted to the +<br/>Graduate School—New Brunswick +<br/><b>Rutgers, The State University of New Jersey</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>Doctor of Philosophy +<br/>Graduate Program in Computer Science +<br/>Written under the direction of +<br/>and approved by +<br/>New Brunswick, New Jersey +<br/>May, 2011 +</td><td>('1683829', 'PENG YANG', 'peng yang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td></td></tr><tr><td>d80a3d1f3a438e02a6685e66ee908446766fefa9</td><td>ZHANG ET AL.: QUANTIFYING FACIAL AGE BY POSTERIOR OF AGE COMPARISONS +<br/>Quantifying Facial Age by Posterior of +<br/>Age Comparisons +<br/>1 SenseTime Group Limited +<br/>2 Department of Information Engineering, +<br/><b>The Chinese University of Hong Kong</b></td><td>('6693591', 'Yunxuan Zhang', 'yunxuan zhang')<br/>('46457827', 'Li Liu', 'li liu')<br/>('46651787', 'Cheng Li', 'cheng li')<br/>('1717179', 'Chen Change Loy', 'chen change loy')</td><td>zhangyunxuan@sensetime.com +<br/>liuli@sensetime.com +<br/>chengli@sensetime.com +<br/>ccloy@ie.cuhk.edu.hk +</td></tr><tr><td>d850aff9d10a01ad5f1d8a1b489fbb3998d0d80e</td><td><b>UNIVERSITY OF CALIFORNIA</b><br/>IRVINE +<br/>Recognizing and Segmenting Objects in the Presence of Occlusion and Clutter +<br/>DISSERTATION +<br/>submitted in partial satisfaction of the requirements +<br/>for the degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>in Computer Science +<br/>by +<br/>Dissertation Committee: +<br/>Professor Charless Fowlkes, Chair +<br/>Professor Deva Ramanan +<br/>Professor Alexander Ihler +<br/>2016 +</td><td>('1898210', 'Golnaz Ghiasi', 'golnaz ghiasi')</td><td></td></tr><tr><td>d89cfed36ce8ffdb2097c2ba2dac3e2b2501100d</td><td>Robust Face Recognition via Multimodal Deep +<br/>Face Representation +</td><td>('37990555', 'Changxing Ding', 'changxing ding')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td></td></tr><tr><td>ab8f9a6bd8f582501c6b41c0e7179546e21c5e91</td><td>Nonparametric Face Verification Using a Novel +<br/>Face Representation +</td><td>('3326805', 'Hae Jong Seo', 'hae jong seo')<br/>('1718280', 'Peyman Milanfar', 'peyman milanfar')</td><td></td></tr><tr><td>ab58a7db32683aea9281c188c756ddf969b4cdbd</td><td>Efficient Solvers for Sparse Subspace Clustering +</td><td>('50333204', 'Stephen Becker', 'stephen becker')</td><td></td></tr><tr><td>ab734bac3994b00bf97ce22b9abc881ee8c12918</td><td>Log-Euclidean Metric Learning on Symmetric Positive Definite Manifold +<br/>with Application to Image Set Classification +<br/>†Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b><br/>§Cooperative Medianet Innovation Center, China +</td><td>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('3046528', 'Xianqiu Li', 'xianqiu li')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>ZHIWU.HUANG@VIPL.ICT.AC.CN +<br/>WANGRUIPING@ICT.AC.CN +<br/>SGSHAN@ICT.AC.CN +<br/>XIANQIU.LI@VIPL.ICT.AC.CN +<br/>XLCHEN@ICT.AC.CN +</td></tr><tr><td>aba770a7c45e82b2f9de6ea2a12738722566a149</td><td>Face Recognition in the Scrambled Domain via Salience-Aware +<br/>Ensembles of Many Kernels +<br/>Jiang, R., Al-Maadeed, S., Bouridane, A., Crookes, D., & Celebi, M. E. (2016). Face Recognition in the +<br/>Scrambled Domain via Salience-Aware Ensembles of Many Kernels. IEEE Transactions on Information +<br/>Forensics and Security, 11(8), 1807-1817. DOI: 10.1109/TIFS.2016.2555792 +<br/>Published in: +<br/>Document Version: +<br/>Peer reviewed version +<br/><b>Queen's University Belfast - Research Portal</b><br/><b>Link to publication record in Queen's University Belfast Research Portal</b><br/>Publisher rights +<br/><b>c) 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting</b><br/>republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, +<br/>or reuse of any copyrighted components of this work in other works. +<br/>General rights +<br/><b>Copyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or other</b><br/>copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated +<br/>with these rights. +<br/>Take down policy +<br/>The Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made to +<br/>ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the +<br/>Download date:05. Nov. 2018 +</td><td></td><td>Research Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk. +</td></tr><tr><td>ab0f9bc35b777eaefff735cb0dd0663f0c34ad31</td><td>Semi-Supervised Learning of Geospatial Objects +<br/>Through Multi-Modal Data Integration +<br/>Electrical Engineering and Computer Science +<br/><b>University of California, Merced, CA</b></td><td>('1698559', 'Yi Yang', 'yi yang')</td><td>Email: snewsam@ucmerced.edu +</td></tr><tr><td>abb396490ba8b112f10fbb20a0a8ce69737cd492</td><td>Robust Face Recognition Using Color +<br/>Information +<br/><b>New Jersey Institute of Technology</b></td><td>('2047820', 'Zhiming Liu', 'zhiming liu')<br/>('39664966', 'Chengjun Liu', 'chengjun liu')</td><td>Newark, New Jersey 07102, USA. femail:zl9@njit.edug +</td></tr><tr><td>ab989225a55a2ddcd3b60a99672e78e4373c0df1</td><td>Sample, Computation vs Storage Tradeoffs for +<br/>Classification Using Tensor Subspace Models +</td><td>('9039699', 'Mohammadhossein Chaghazardi', 'mohammadhossein chaghazardi')<br/>('1980683', 'Shuchin Aeron', 'shuchin aeron')</td><td></td></tr><tr><td>abac0fa75281c9a0690bf67586280ed145682422</td><td>Describable Visual Attributes for Face Images +<br/>Submitted in partial fulfillment of the +<br/>requirements for the degree +<br/>of Doctor of Philosophy +<br/>in the Graduate School of Arts and Sciences +<br/><b>COLUMBIA UNIVERSITY</b><br/>2011 +</td><td>('40192613', 'Neeraj Kumar', 'neeraj kumar')</td><td></td></tr><tr><td>ab6776f500ed1ab23b7789599f3a6153cdac84f7</td><td>International Journal of Scientific & Engineering Research, Volume 6, Issue 4, April-2015 1212 +<br/>ISSN 2229-5518 +<br/>A Survey on Various Facial Expression +<br/>Techniques +</td><td>('2122870', 'Joy Bhattacharya', 'joy bhattacharya')</td><td></td></tr><tr><td>ab1719f573a6c121d7d7da5053fe5f12de0182e7</td><td>Combining Visual Recognition +<br/>and Computational Linguistics +<br/>Linguistic Knowledge for Visual Recognition +<br/>and Natural Language Descriptions +<br/>of Visual Content +<br/>Thesis for obtaining the title of +<br/>Doctor of Engineering Science +<br/>(Dr.-Ing.) +<br/>of the Faculty of Natural Science and Technology I +<br/><b>of Saarland University</b><br/>by +<br/>Saarbrücken +<br/>March 2014 +</td><td>('34849128', 'Marcus Rohrbach', 'marcus rohrbach')</td><td></td></tr><tr><td>ab2b09b65fdc91a711e424524e666fc75aae7a51</td><td>Multi-modal Biomarkers to Discriminate Cognitive State* +<br/>1MIT Lincoln Laboratory, Lexington, Massachusetts, USA +<br/>2USARIEM, 3NSRDEC +<br/>1. Introduction +<br/>Multimodal biomarkers based on behavorial, neurophysiolgical, and cognitive measurements have +<br/>recently obtained increasing popularity in the detection of cognitive stress- and neurological-based +<br/>disorders. Such conditions are significantly and adversely affecting human performance and quality +<br/>of life for a large fraction of the world’s population. Example modalities used in detection of these +<br/>conditions include voice, facial expression, physiology, eye tracking, gait, and EEG analysis. +<br/>Toward the goal of finding simple, noninvasive means to detect, predict and monitor cognitive +<br/>stress and neurological conditions, MIT Lincoln Laboratory is developing biomarkers that satisfy +<br/>three criteria. First, we seek biomarkers that reflect core components of cognitive status such as +<br/>working memory capacity, processing speed, attention, and arousal. Second, and as importantly, we +<br/>seek biomarkers that reflect timing and coordination relations both within components of each +<br/>modality and across different modalities. This is based on the hypothesis that neural coordination +<br/>across different parts of the brain is essential in cognition (Figure 1). An example of timing and +<br/>coordination within a modality is the set of finely timed and synchronized physiological +<br/>components of speech production, while an example of coordination across modalities is the timing +<br/>and synchrony that occurs across speech and facial expression while speaking. Third, we seek +<br/>multimodal biomarkers that contribute in a complementary fashion under various channel and +<br/>background conditions. In this chapter, as an illustration of this biomarker approach we focus on +<br/>cognitive stress and the particular case of detecting different cognitive load levels. We also briefly +<br/>show how similar feature-extraction principles can be applied to a neurological condition through +<br/>the example of major depression disorder (MDD). MDD is one of several neurological disorders +<br/>where multi-modal biomarkers based on principles of timing and coordination are important for +<br/>detection [11]-[22]. In our cognitive load experiments, we use two easily obtained noninvasive +<br/>modalities, voice and face, and show how these two modalities can be fused to produce results on +<br/>par with more invasive, “gold-standard” EEG measurements. Vocal and facial biomarkers will also +<br/>be used in our MDD case study. In both application areas we focus on timing and coordination +<br/>relations within the components of each modality. +<br/>* Distribution A: public release.This work is sponsored by the Assistant Secretary of Defense for Research & Engineering under Air Force contract +<br/>#FA8721-05-C-0002. Opinions,interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the United States +<br/>Government. +</td><td>('1718470', 'Thomas F. Quatieri', 'thomas f. quatieri')<br/>('48628822', 'James R. Williamson', 'james r. williamson')<br/>('2794344', 'Christopher J. Smalt', 'christopher j. smalt')<br/>('38799981', 'Tejash Patel', 'tejash patel')<br/>('2894484', 'Brian S. Helfer', 'brian s. helfer')<br/>('3051832', 'Daryush D. Mehta', 'daryush d. mehta')<br/>('35718569', 'Kristin Heaton', 'kristin heaton')<br/>('47534051', 'Marianna Eddy', 'marianna eddy')<br/>('49739272', 'Joseph Moran', 'joseph moran')</td><td>[quatieri,jrw]@ll.mit.edu +</td></tr><tr><td>ab87dfccb1818bdf0b41d732da1f9335b43b74ae</td><td>SUBMITTED TO IEEE TRANSACTIONS ON SIGNAL PROCESSING +<br/>Structured Dictionary Learning for Classification +</td><td>('36657778', 'Yuanming Suo', 'yuanming suo')<br/>('31507586', 'Minh Dao', 'minh dao')<br/>('35210356', 'Umamahesh Srinivas', 'umamahesh srinivas')<br/>('3346079', 'Vishal Monga', 'vishal monga')<br/>('1709073', 'Trac D. Tran', 'trac d. tran')</td><td></td></tr><tr><td>abc1ef570bb2d7ea92cbe69e101eefa9a53e1d72</td><td>Raisonnement abductif en logique de +<br/>description exploitant les domaines concrets +<br/>spatiaux pour l’interprétation d’images +<br/>1. LTCI, Télécom ParisTech, Université Paris-Saclay, Paris, France +<br/><b>Universit Paris-Dauphine, PSL Research University, CNRS, UMR</b><br/>LAMSADE, 75016 Paris, France +<br/>RÉSUMÉ. L’interprétation d’images a pour objectif non seulement de détecter et reconnaître des +<br/>objets dans une scène mais aussi de fournir une description sémantique tenant compte des in- +<br/>formations contextuelles dans toute la scène. Le problème de l’interprétation d’images peut être +<br/>formalisé comme un problème de raisonnement abductif, c’est-à-dire comme la recherche de la +<br/>meilleure explication en utilisant une base de connaissances. Dans ce travail, nous présentons +<br/>une nouvelle approche utilisant une méthode par tableau pour la génération et la sélection +<br/>d’explications possibles d’une image donnée lorsque les connaissances, exprimées dans une +<br/>logique de description, comportent des concepts décrivant les objets mais aussi les relations +<br/>spatiales entre ces objets. La meilleure explication est sélectionnée en exploitant les domaines +<br/>concrets pour évaluer le degré de satisfaction des relations spatiales entre les objets. +</td><td>('4156317', 'Yifan Yang', 'yifan yang')<br/>('1773774', 'Jamal Atif', 'jamal atif')<br/>('1695917', 'Isabelle Bloch', 'isabelle bloch')</td><td>{yifan.yang,isabelle.bloch}@telecom-paristech.fr +<br/>jamal.atif@dauphine.fr +</td></tr><tr><td>abba1bf1348a6f1b70a26aac237338ee66764458</td><td>Facial Action Unit Detection Using Attention and Relation Learning +<br/><b>Shanghai Jiao Tong University, China</b><br/><b>School of Computer Science and Technology, Tianjin University, China</b><br/><b>School of Computer Science and Engineering, Nanyang Technological University, Singapore</b><br/>4 Tencent YouTu, China +<br/><b>School of Computer Science and Software Engineering, East China Normal University, China</b></td><td>('3403352', 'Zhiwen Shao', 'zhiwen shao')<br/>('1771215', 'Zhilei Liu', 'zhilei liu')<br/>('1688642', 'Jianfei Cai', 'jianfei cai')<br/>('10609538', 'Yunsheng Wu', 'yunsheng wu')<br/>('8452947', 'Lizhuang Ma', 'lizhuang ma')</td><td>shaozhiwen@sjtu.edu.cn, zhileiliu@tju.edu.cn, asjfcai@ntu.edu.sg +<br/>simonwu@tencent.com, ma-lz@cs.sjtu.edu.cn +</td></tr><tr><td>abdd17e411a7bfe043f280abd4e560a04ab6e992</td><td>Pose-Robust Face Recognition via Deep Residual Equivariant Mapping +<br/><b>The Chinese University of Hong Kong</b><br/>2SenseTime Research +</td><td>('9963152', 'Kaidi Cao', 'kaidi cao')<br/>('46651787', 'Cheng Li', 'cheng li')</td><td>{ry017, ccloy, xtang}@ie.cuhk.edu.hk +<br/>{caokaidi, chengli}@sensetime.com +</td></tr><tr><td>ab1dfcd96654af0bf6e805ffa2de0f55a73c025d</td><td></td><td></td><td></td></tr><tr><td>abeda55a7be0bbe25a25139fb9a3d823215d7536</td><td>UNIVERSITATPOLITÈCNICADECATALUNYAProgramadeDoctorat:AUTOMÀTICA,ROBÒTICAIVISIÓTesiDoctoralUnderstandingHuman-CentricImages:FromGeometrytoFashionEdgarSimoSerraDirectors:FrancescMorenoNoguerCarmeTorrasMay2015 </td><td></td><td></td></tr><tr><td>ab427f0c7d4b0eb22c045392107509451165b2ba</td><td>LEARNING SCALE RANGES FOR THE EXTRACTION OF REGIONS OF +<br/>INTEREST +<br/><b>Western Kentucky University</b><br/>Department of Mathematics and Computer Science +<br/><b>College Heights Blvd, Bowling Green, KY</b></td><td>('1682467', 'Qi Li', 'qi li')<br/>('2446364', 'Zachary Bessinger', 'zachary bessinger')</td><td></td></tr><tr><td>ab1900b5d7cf3317d17193e9327d57b97e24d2fc</td><td></td><td></td><td></td></tr><tr><td>ab8fb278db4405f7db08fa59404d9dd22d38bc83</td><td>UNIVERSITÉ DE GENÈVE +<br/>Département d'Informatique +<br/>FACULTÉ DES SCIENCES +<br/>Implicit and Automated Emotional +<br/>Tagging of Videos +<br/>THÈSE +<br/>présenté à la Faculté des sciences de l'Université de Genève +<br/>pour obtenir le grade de Docteur ès sciences, mention informatique +<br/>par +<br/>de +<br/>Téhéran (IRAN) +<br/>Thèse No 4368 +<br/>GENÈVE +<br/>Repro-Mail - Université de Genève +<br/>2011 +</td><td>('1809085', 'Thierry Pun', 'thierry pun')<br/>('2463695', 'Mohammad SOLEYMANI', 'mohammad soleymani')</td><td></td></tr><tr><td>e5e5f31b81ed6526c26d277056b6ab4909a56c6c</td><td>Revisit Multinomial Logistic Regression in Deep Learning: +<br/>Data Dependent Model Initialization for Image Recognition +<br/><b>University of Illinois at Urbana-Champaign</b><br/>2Ping An Property&Casualty Insurance Company of China, +<br/>3Microsoft +</td><td>('50563570', 'Bowen Cheng', 'bowen cheng')<br/>('1972288', 'Rong Xiao', 'rong xiao')<br/>('3133575', 'Yandong Guo', 'yandong guo')<br/>('1689532', 'Yuxiao Hu', 'yuxiao hu')<br/>('38504661', 'Jianfeng Wang', 'jianfeng wang')<br/>('48571185', 'Lei Zhang', 'lei zhang')</td><td>1bcheng9@illinois.edu +<br/>2xiaorong283@pingan.com.cn +<br/>3yandong.guo@live.com, yuxiaohu@msn.com, {jianfw, leizhang}@microsoft.com +</td></tr><tr><td>e5737ffc4e74374b0c799b65afdbf0304ff344cb</td><td></td><td></td><td></td></tr><tr><td>e506cdb250eba5e70c5147eb477fbd069714765b</td><td>Heterogeneous Face Recognition +<br/>By +<br/>Brendan F. Klare +<br/>A Dissertation +<br/>Submitted to +<br/><b>Michigan State University</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>Doctor of Philosophy +<br/>Computer Science and Engineering +<br/>2012 +</td><td></td><td></td></tr><tr><td>e572c42d8ef2e0fadedbaae77c8dfe05c4933fbf</td><td>A Visual Historical Record of American High School Yearbooks +<br/>A Century of Portraits: +<br/><b>University of California Berkeley</b><br/><b>Brown University</b><br/><b>University of California Berkeley</b></td><td>('2361255', 'Shiry Ginosar', 'shiry ginosar')<br/>('2660664', 'Kate Rakelly', 'kate rakelly')<br/>('33385802', 'Sarah Sachs', 'sarah sachs')<br/>('2130100', 'Brian Yin', 'brian yin')<br/>('1763086', 'Alexei A. Efros', 'alexei a. efros')</td><td></td></tr><tr><td>e5823a9d3e5e33e119576a34cb8aed497af20eea</td><td>DocFace+: ID Document to Selfie* Matching +</td><td>('9644181', 'Yichun Shi', 'yichun shi')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>e5dfd17dbfc9647ccc7323a5d62f65721b318ba9</td><td></td><td></td><td></td></tr><tr><td>e510f2412999399149d8635a83eca89c338a99a1</td><td>Journal of Advanced Computer Science and Technology, 1 (4) (2012) 266-283 +<br/>c(cid:13)Science Publishing Corporation +<br/>www.sciencepubco.com/index.php/JACST +<br/>Face Recognition using Block-Based +<br/>DCT Feature Extraction +<br/>1Department of Electronics and Communication Engineering, +<br/><b>M S Ramaiah Institute of Technology, Bangalore, Karnataka, India</b><br/>2Department of Electronics and Communication Engineering, +<br/><b>S J B Institute of Technology, Bangalore, Karnataka, India</b></td><td>('2472608', 'K Manikantan', 'k manikantan')<br/>('3389602', 'Vaishnavi Govindarajan', 'vaishnavi govindarajan')<br/>('35084871', 'V V S Sasi Kiran', 'v v s sasi kiran')<br/>('1687245', 'S Ramachandran', 's ramachandran')</td><td>E-mail: kmanikantan@msrit.edu +<br/>E-mail: vaish.india@gmail.com +<br/>E-mail: sasikiran.f4@gmail.com +<br/>E-mail: ramachandr@gmail.com +</td></tr><tr><td>e56c4c41bfa5ec2d86c7c9dd631a9a69cdc05e69</td><td>Human Activity Recognition Based on Wearable +<br/>Sensor Data: A Standardization of the +<br/>State-of-the-Art +<br/>Smart Surveillance Interest Group, Computer Science Department +<br/>Universidade Federal de Minas Gerais, Brazil +</td><td>('2954974', 'Antonio C. Nazare', 'antonio c. nazare')<br/>('1679142', 'William Robson Schwartz', 'william robson schwartz')</td><td>Email: {arturjordao, antonio.nazare, jessicasena, william}@dcc.ufmg.br +</td></tr><tr><td>e59813940c5c83b1ce63f3f451d03d34d2f68082</td><td>Faculty of Informatics - Papers (Archive) +<br/>Faculty of Engineering and Information Sciences +<br/><b>University of Wollongong</b><br/>Research Online +<br/>2008 +<br/>A real-time facial expression recognition system for +<br/>online games +<br/>Publication Details +<br/>Zhan, C., Li, W., Ogunbona, P. & Safaei, F. (2008). A real-time facial expression recognition system for online games. International +<br/>Journal of Computer Games Technology, 2008 (Article No. 10), 1-7. +<br/>Research Online is the open access institutional repository for the +<br/><b>University of Wollongong. For further information contact the UOW</b></td><td>('3283367', 'Ce Zhan', 'ce zhan')<br/>('1685696', 'Wanqing Li', 'wanqing li')<br/>('1719314', 'Philip Ogunbona', 'philip ogunbona')<br/>('1803733', 'Farzad Safaei', 'farzad safaei')</td><td>University of Wollongong, czhan@uow.edu.au +<br/>University of Wollongong, wanqing@uow.edu.au +<br/>University of Wollongong, philipo@uow.edu.au +<br/>University of Wollongong, farzad@uow.edu.au +<br/>Library: research-pubs@uow.edu.au +</td></tr><tr><td>e5b301ee349ba8e96ea6c71782295c4f06be6c31</td><td>The Case for Onloading Continuous High-Datarate Perception to the Phone +<br/><b>University of Washington</b><br/>Microsoft Research +</td><td>('1871038', 'Seungyeop Han', 'seungyeop han')<br/>('3041721', 'Matthai Philipose', 'matthai philipose')</td><td></td></tr><tr><td>e569f4bd41895028c4c009e5b46b935056188e91</td><td>SIMONYAN et al.: FISHER VECTOR FACES IN THE WILD +<br/>Fisher Vector Faces in the Wild +<br/>Visual Geometry Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b><br/>Omkar M. Parkhi +<br/>Andrea Vedaldi +<br/>Andrew Zisserman +</td><td>('34838386', 'Karen Simonyan', 'karen simonyan')</td><td>karen@robots.ox.ac.uk +<br/>omkar@robots.ox.ac.uk +<br/>vedaldi@robots.ox.ac.uk +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>e5fbffd3449a2bfe0acb4ec339a19f5b88fff783</td><td>WILES, KOEPKE, ZISSERMAN: SELF-SUP. FACIAL ATTRIBUTE FROM VIDEO +<br/>Self-supervised learning of a facial attribute +<br/>embedding from video +<br/>Visual Geometry Group +<br/><b>University of Oxford</b><br/>Oxford, UK +</td><td>('8792285', 'Olivia Wiles', 'olivia wiles')<br/>('47104886', 'A. Sophia Koepke', 'a. sophia koepke')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>ow@robots.ox.ac.uk +<br/>koepke@robots.ox.ac.uk +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>e5342233141a1d3858ed99ccd8ca0fead519f58b</td><td>ISSN: 2277 – 9043 +<br/>International Journal of Advanced Research in Computer Science and Electronics Engineering (IJARCSEE) +<br/>Volume 2, Issue 2, February 2013 +<br/>Finger print and Palm print based Multibiometric +<br/>Authentication System with GUI Interface +<br/><b>PG Scholar, Dr.Pauls Engineering College, Villupuram District, Tamilnadu, India</b><br/><b>Dr.Pauls Engineering College, Villupuram District, Tamilnadu, India</b></td><td></td><td></td></tr><tr><td>e52be9a083e621d9ed29c8e9914451a6a327ff59</td><td>UvA-DARE (Digital Academic Repository) +<br/>Communication and Automatic Interpretation of Affect from Facial Expressions +<br/>Salah, A.A.; Sebe, N.; Gevers, T. +<br/>Published in: +<br/>Affective computing and interaction: psychological, cognitive, and neuroscientific perspectives +<br/>Link to publication +<br/>Citation for published version (APA): +<br/>Salah, A. A., Sebe, N., & Gevers, T. (2010). Communication and Automatic Interpretation of Affect from Facial +<br/>Expressions. In D. Gökçay, & G. Yildirim (Eds.), Affective computing and interaction: psychological, cognitive, +<br/>and neuroscientific perspectives (pp. 157-183). Hershey, PA: Information Science Reference. +<br/>General rights +<br/>It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), +<br/>other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). +<br/>Disclaimer/Complaints regulations +<br/>If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating +<br/>your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask +<br/><b>the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam</b><br/>The Netherlands. You will be contacted as soon as possible. +<br/>Download date: 12 Sep 2017 +<br/><b>UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl</b></td><td></td><td></td></tr><tr><td>e5d53a335515107452a30b330352cad216f88fc3</td><td>Generalized Loss-Sensitive Adversarial Learning +<br/>with Manifold Margins +<br/>Laboratory for MAchine Perception and LEarning (MAPLE) +<br/>http://maple.cs.ucf.edu/ +<br/><b>University of Central Florida, Orlando FL 32816, USA</b></td><td>('46232436', 'Marzieh Edraki', 'marzieh edraki')<br/>('2272096', 'Guo-Jun Qi', 'guo-jun qi')</td><td>m.edraki@knights.ucf.edu, guojun.qi@ucf.edu +</td></tr><tr><td>e5799fd239531644ad9270f49a3961d7540ce358</td><td>KINSHIP CLASSIFICATION BY MODELING FACIAL FEATURE HEREDITY +<br/><b>Cornell University 2Eastman Kodak Company</b></td><td>('2666471', 'Ruogu Fang', 'ruogu fang')<br/>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td></td></tr><tr><td>e5eb7fa8c9a812d402facfe8e4672670541ed108</td><td>Performance of PCA Based Semi-supervised +<br/>Learning in Face Recognition Using MPEG-7 +<br/>Edge Histogram Descriptor +<br/>Department of Computer Science and Engineering +<br/><b>Bangladesh University of Engineering and Technology(BUET</b><br/>Dhaka-1000, Bangladesh +</td><td>('3034202', 'Sheikh Motahar Naim', 'sheikh motahar naim')<br/>('9248625', 'Abdullah Al Farooq', 'abdullah al farooq')<br/>('1990532', 'Md. Monirul Islam', 'md. monirul islam')</td><td>Email: {shafin buet, naim sbh2007, saurav00001}@yahoo.com, mmislam@cse.buet.ac.bd +</td></tr><tr><td>e22adcd2a6a7544f017ec875ce8f89d5c59e09c8</td><td>Published in Proc. of IEEE 9th International Conference on Biometrics: Theory, Applications and Systems (BTAS), (Los +<br/>Angeles, CA), October 2018. +<br/>Gender Privacy: An Ensemble of Semi Adversarial Networks for Confounding +<br/>Arbitrary Gender Classifiers +<br/><b>Computer Science and Engineering, Michigan State University, East Lansing, USA</b><br/><b>University of Wisconsin Madison, USA</b></td><td>('5456235', 'Vahid Mirjalili', 'vahid mirjalili')<br/>('2562040', 'Sebastian Raschka', 'sebastian raschka')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td>mirjalil@cse.msu.edu +<br/>mail@sebastianraschka.com +<br/>rossarun@cse.msu.edu +</td></tr><tr><td>e27c92255d7ccd1860b5fb71c5b1277c1648ed1e</td><td></td><td></td><td></td></tr><tr><td>e200c3f2849d56e08056484f3b6183aa43c0f13a</td><td></td><td></td><td></td></tr><tr><td>e2d265f606cd25f1fd72e5ee8b8f4c5127b764df</td><td>Real-Time End-to-End Action Detection +<br/>with Two-Stream Networks +<br/><b>School of Engineering, University of Guelph</b><br/><b>Vector Institute for Arti cial Intelligence</b><br/><b>Canadian Institute for Advanced Research</b></td><td>('35933395', 'Alaaeldin El-Nouby', 'alaaeldin el-nouby')<br/>('3861110', 'Graham W. Taylor', 'graham w. taylor')</td><td>{aelnouby,gwtaylor}@uoguelph.ca +</td></tr><tr><td>e293a31260cf20996d12d14b8f29a9d4d99c4642</td><td>Published as a conference paper at ICLR 2017 +<br/>LR-GAN: LAYERED RECURSIVE GENERATIVE AD- +<br/>VERSARIAL NETWORKS FOR IMAGE GENERATION +<br/>Virginia Tech +<br/>Blacksburg, VA +<br/>Facebook AI Research +<br/>Menlo Park, CA +<br/><b>Georgia Institute of Technology</b><br/>Atlanta, GA +</td><td>('2404941', 'Jianwei Yang', 'jianwei yang')<br/>('39248118', 'Anitha Kannan', 'anitha kannan')<br/>('1746610', 'Dhruv Batra', 'dhruv batra')</td><td>jw2yang@vt.edu +<br/>akannan@fb.com +<br/>{dbatra, parikh}@gatech.edu +</td></tr><tr><td>e20e2db743e8db1ff61279f4fda32bf8cf381f8e</td><td>Deep Cross Polarimetric Thermal-to-visible Face Recognition +<br/><b>West Virginia University</b></td><td>('6779960', 'Seyed Mehdi Iranmanesh', 'seyed mehdi iranmanesh')<br/>('35477977', 'Ali Dabouei', 'ali dabouei')<br/>('2700951', 'Hadi Kazemi', 'hadi kazemi')<br/>('8147588', 'Nasser M. Nasrabadi', 'nasser m. nasrabadi')</td><td>{seiranmanesh, ad0046, hakazemi}@mix.wvu.edu, {nasser.nasrabadi}@mail.wvu.edu +</td></tr><tr><td>f437b3884a9e5fab66740ca2a6f1f3a5724385ea</td><td>Human Identification Technical Challenges +<br/>DARPA +<br/>3701 N. Fairfax Dr +<br/>Arlington, VA 22203 +</td><td>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')</td><td>jphillips@darpa.mil +</td></tr><tr><td>f412d9d7bc7534e7daafa43f8f5eab811e7e4148</td><td>Durham Research Online +<br/>Deposited in DRO: +<br/>16 December 2014 +<br/>Version of attached le: +<br/>Accepted Version +<br/>Peer-review status of attached le: +<br/>Peer-reviewed +<br/>Citation for published item: +<br/>Kirk, H. E. and Hocking, D. R. and Riby, D. M. and Cornish, K. M. (2013) 'Linking social behaviour and +<br/>anxiety to attention to emotional faces in Williams syndrome.', Research in developmental disabilities., 34 +<br/>(12). pp. 4608-4616. +<br/>Further information on publisher's website: +<br/>http://dx.doi.org/10.1016/j.ridd.2013.09.042 +<br/>Publisher's copyright statement: +<br/>NOTICE: this is the author's version of a work that was accepted for publication in Research in Developmental +<br/>Disabilities. Changes resulting from the publishing process, such as peer review, editing, corrections, structural +<br/>formatting, and other quality control mechanisms may not be re ected in this document. Changes may have been made +<br/>to this work since it was submitted for publication. A de nitive version was subsequently published in Research in +<br/>Developmental Disabilities, 34, 12, December 2013, 10.1016/j.ridd.2013.09.042. +<br/>Additional information: +<br/>Use policy +<br/>The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for +<br/>personal research or study, educational, or not-for-pro t purposes provided that: +<br/>• a full bibliographic reference is made to the original source +<br/>• a link is made to the metadata record in DRO +<br/>• the full-text is not changed in any way +<br/>The full-text must not be sold in any format or medium without the formal permission of the copyright holders. +<br/>Please consult the full DRO policy for further details. +<br/><b>Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom</b><br/>Tel : +44 (0)191 334 3042 | Fax : +44 (0)191 334 2971 +<br/>http://dro.dur.ac.uk +</td><td></td><td></td></tr><tr><td>f43eeb578e0ca48abfd43397bbd15825f94302e4</td><td>Optical Computer Recognition of Facial Expressions +<br/>Associated with Stress Induced by Performance +<br/>Demands +<br/>DINGES DF, RIDER RL, DORRIAN J, MCGLINCHEY EL, ROGERS NL, +<br/>CIZMAN Z, GOLDENSTEIN SK, VOGLER C, VENKATARAMAN S, METAXAS +<br/>DN. Optical computer recognition of facial expressions associated +<br/>with stress induced by performance demands. Aviat Space Environ +<br/>Med 2005; 76(6, Suppl.):B172– 82. +<br/>Application of computer vision to track changes in human facial +<br/>expressions during long-duration spaceflight may be a useful way to +<br/>unobtrusively detect the presence of stress during critical operations. To +<br/>develop such an approach, we applied optical computer recognition +<br/>(OCR) algorithms for detecting facial changes during performance while +<br/>people experienced both low- and high-stressor performance demands. +<br/>Workload and social feedback were used to vary performance stress in +<br/>60 healthy adults (29 men, 31 women; mean age 30 yr). High-stressor +<br/>scenarios involved more difficult performance tasks, negative social +<br/>feedback, and greater time pressure relative to low workload scenarios. +<br/>Stress reactions were tracked using self-report ratings, salivary cortisol, +<br/>and heart rate. Subjects also completed personality, mood, and alexi- +<br/>thymia questionnaires. To bootstrap development of the OCR algorithm, +<br/>we had a human observer, blind to stressor condition, identify the +<br/>expressive elements of the face of people undergoing high- vs. low- +<br/>stressor performance. Different sets of videos of subjects’ faces during +<br/>performance conditions were used for OCR algorithm training. Subjec- +<br/>tive ratings of stress, task difficulty, effort required, frustration, and +<br/>negative mood were significantly increased during high-stressor perfor- +<br/>mance bouts relative to low-stressor bouts (all p ⬍ 0.01). The OCR +<br/>algorithm was refined to provide robust 3-d tracking of facial expres- +<br/>sions during head movement. Movements of eyebrows and asymmetries +<br/>in the mouth were extracted. These parameters are being used in a +<br/>Hidden Markov model to identify high- and low-stressor conditions. +<br/>Preliminary results suggest that an OCR algorithm using mouth and +<br/>eyebrow regions has the potential +<br/>to discriminate high- from low- +<br/>stressor performance bouts in 75– 88% of subjects. The validity of the +<br/>workload paradigm to induce differential levels of stress in facial ex- +<br/>pressions was established. The paradigm also provided the basic stress- +<br/>related facial expressions required to establish a prototypical OCR al- +<br/>gorithm to detect such changes. Efforts are underway to further improve +<br/>the OCR algorithm by adding facial touching and automating applica- +<br/>tion of the deformable masks and OCR algorithms to video footage of the +<br/>moving faces as a prelude to blind validation of the automated ap- +<br/>proach. +<br/>Keywords: optical computer recognition, computer vision, workload, +<br/>performance, stress, human face, cortisol, heart rate, astronauts, Markov +<br/>models. +<br/>ASTRONAUTS ARE required to perform mission- +<br/>critical tasks at a high level of functional capability +<br/>throughout spaceflight. While they can be trained to +<br/>cope with, and/or adapt to some stressors of space- +<br/>flight, stressful reactions can and have occurred during +<br/>long-duration missions, especially when operational +<br/>performance demands become elevated when unex- +<br/>pected and/or underestimated operational require- +<br/>ments occurred while crews were already experiencing +<br/>work-related stressors (13,28,42,43,52,57,66). In some of +<br/>these instances, stressed flight crews have withdrawn +<br/>from voice communications with ground controllers +<br/>(7,66), or when pressed to continue performing, made +<br/>errors that could have jeopardized the mission (13,28). +<br/>Consequently, there is a need to identify when during +<br/>operational demands astronauts are experiencing be- +<br/>havioral stress associated with performance demands. +<br/>This is especially important as mission durations in- +<br/>crease in length and ultimately involve flight to other +<br/>locations in the solar system. +<br/>Facial Expressions of Stress +<br/>Measurement of human emotional expressions via +<br/><b>the face, including negative affect and distress, dates</b><br/>back to Darwin (14), but in recent years has been un- +<br/>dergoing extensive scientific study (46). Although cul- +<br/>tural differences can intensify facial expression of emo- +<br/>tions (53), there is considerable scientific evidence that +<br/>select emotions are communicated in distinct facial dis- +<br/>plays across cultures, age, and gender (45). Because +<br/>many techniques for monitoring stress reactions are +<br/>impractical, unreliable, or obtrusive in spaceflight, we +<br/>seek to develop a novel, objective, unobtrusive com- +<br/>puter vision system to continuously track facial expres- +<br/>sions during performance demands, to detect when +<br/>From the Unit for Experimental Psychiatry, Department of Psychi- +<br/><b>atry, University of Pennsylvania School of Medicine, Philadelphia, PA</b><br/>(D. F. Dinges, R. L. Rider, J. Dorrian, E. L. McGlinchey, N. L. Rogers, +<br/>Z. Cizman); and the Center for Computational Biomedicine, Imaging +<br/><b>and Modeling, Rutgers University</b><br/>New Brunswick, NJ (S. K. Goldstein, C. Vogler, S. Venkataraman, +<br/>D. N. Metaxas). +<br/>Address reprint requests to: David F. Dinges, Ph.D., Professor and +<br/>Director, Unit for Experimental Psychiatry, Department of Psychiatry, +<br/><b>University of Pennsylvania School of Medicine, 1013 Blockley Hall</b><br/>med.upenn.edu. +<br/>Reprint & Copyright © by Aerospace Medical Association, Alexan- +<br/>dria, VA. +<br/>B172 +</td><td>('5515440', 'Jillian Dorrian', 'jillian dorrian')<br/>('4940404', 'Ziga Cizman', 'ziga cizman')<br/>('2467082', 'Christian Vogler', 'christian vogler')<br/>('2898034', 'Sundara Venkataraman', 'sundara venkataraman')</td><td>423 Guardian Drive, Philadelphia, PA 19104-6021; dinges@mail. +</td></tr><tr><td>f442a2f2749f921849e22f37e0480ac04a3c3fec</td><td></td><td></td><td> Critical Features for Face Recognition in Humans and Machines Naphtali Abudarham1, Lior Shkiller1, Galit Yovel1,2 1School of Psychological Sciences, 2Sagol School of Neuroscience Tel Aviv University, Tel Aviv, Israel Correspondence regarding this manuscript should be addressed to: Galit Yovel School of Psychological Sciences & Sagol School of Neuroscience Tel Aviv University Tel Aviv, 69978, Israel Email: gality@post.tau.ac.il, </td></tr><tr><td>f4f9697f2519f1fe725ee7e3788119ed217dca34</td><td>Selfie-Presentation in Everyday Life: A Large-scale +<br/>Characterization of Selfie Contexts on Instagram +<br/><b>Georgia Institute of Technology</b><br/>North Ave NW +<br/>Atlanta, GA 30332 +</td><td>('10799246', 'Julia Deeb-Swihart', 'julia deeb-swihart')<br/>('39723397', 'Christopher Polack', 'christopher polack')<br/>('1809407', 'Eric Gilbert', 'eric gilbert')</td><td>{jdeeb3, cfpolack,gilbert,irfan}@gatech.edu +</td></tr><tr><td>f4f6fc473effb063b7a29aa221c65f64a791d7f4</td><td>Downloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Electronic-Imaging on 4/20/2018 Terms of Use: https://www.spiedigitallibrary.org/terms-of-use +<br/>FacialexpressionrecognitioninthewildbasedonmultimodaltexturefeaturesBoSunLiandongLiGuoyanZhouJunHeBoSun,LiandongLi,GuoyanZhou,JunHe,“Facialexpressionrecognitioninthewildbasedonmultimodaltexturefeatures,”J.Electron.Imaging25(6),061407(2016),doi:10.1117/1.JEI.25.6.061407. </td><td></td><td></td></tr><tr><td>f4c01fc79c7ead67899f6fe7b79dd1ad249f71b0</td><td></td><td></td><td></td></tr><tr><td>f4373f5631329f77d85182ec2df6730cbd4686a9</td><td>Soft Computing manuscript No. +<br/>(will be inserted by the editor) +<br/>Recognizing Gender from Human Facial Regions using +<br/>Genetic Algorithm +<br/>Received: date / Accepted: date +</td><td>('24069279', 'Avirup Bhattacharyya', 'avirup bhattacharyya')<br/>('40813600', 'Partha Pratim Roy', 'partha pratim roy')<br/>('32614479', 'Samarjit Kar', 'samarjit kar')</td><td></td></tr><tr><td>f4210309f29d4bbfea9642ecadfb6cf9581ccec7</td><td>An Agreement and Sparseness-based Learning Instance Selection +<br/>and its Application to Subjective Speech Phenomena +<br/>1 Machine Intelligence & Signal Processing Group, MMK, Technische Universit¨at M¨unchen, Germany +<br/><b>Imperial College London, United Kingdom</b></td><td>('30512170', 'Zixing Zhang', 'zixing zhang')<br/>('1751126', 'Florian Eyben', 'florian eyben')<br/>('39629517', 'Jun Deng', 'jun deng')</td><td>zixing.zhang@tum.de +</td></tr><tr><td>f47404424270f6a20ba1ba8c2211adfba032f405</td><td>International Journal of Emerging Technology and Advanced Engineering +<br/>Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 5, May 2012) +<br/>Identification of Face Age range Group using Neural +<br/>Network +</td><td>('7530203', 'Sneha Thakur', 'sneha thakur')</td><td> 1sne_thakur@yahoo.co.in +<br/> 2ligendra@rediffmail.com +</td></tr><tr><td>f4d30896c5f808a622824a2d740b3130be50258e</td><td>DS++: A Flexible, Scalable and Provably Tight Relaxation for Matching Problems +<br/><b>Weizmann Institute of Science</b></td><td>('3046344', 'Nadav Dym', 'nadav dym')<br/>('3416939', 'Haggai Maron', 'haggai maron')<br/>('3232072', 'Yaron Lipman', 'yaron lipman')</td><td></td></tr><tr><td>f4ebbeb77249d1136c355f5bae30f02961b9a359</td><td>Human Computation for Attribute and Attribute Value Acquisition +<br/>School of Computer Science +<br/><b>Carnegie Melon University</b></td><td>('2987829', 'Edith Law', 'edith law')<br/>('1717452', 'Burr Settles', 'burr settles')<br/>('2681926', 'Aaron Snook', 'aaron snook')<br/>('2762792', 'Harshit Surana', 'harshit surana')<br/>('3328108', 'Luis von Ahn', 'luis von ahn')<br/>('39182987', 'Tom Mitchell', 'tom mitchell')</td><td>edith@cmu.edu +</td></tr><tr><td>f4aed1314b2d38fd8f1b9d2bc154295bbd45f523</td><td>Subspace Clustering using Ensembles of +<br/>K-Subspaces +<br/>Department of Electrical and Computer Engineering +<br/><b>University of Michigan, Ann Arbor</b></td><td>('1782134', 'John Lipor', 'john lipor')<br/>('5250186', 'David Hong', 'david hong')<br/>('2358258', 'Dejiao Zhang', 'dejiao zhang')<br/>('1682385', 'Laura Balzano', 'laura balzano')</td><td>{lipor,dahong,dejiao,girasole}@umich.edu +</td></tr><tr><td>f42dca4a4426e5873a981712102aa961be34539a</td><td>Next-Flow: Hybrid Multi-Tasking with Next-Frame Prediction to Boost +<br/>Optical-Flow Estimation in the Wild +<br/><b>University of Freiburg</b><br/>Germany +</td><td>('31656404', 'Nima Sedaghat', 'nima sedaghat')</td><td>nima@cs.uni-freiburg.de +</td></tr><tr><td>f3ca2c43e8773b7062a8606286529c5bc9b3ce25</td><td>Deep Clustering via Joint Convolutional Autoencoder Embedding and Relative +<br/>Entropy Minimization +<br/><b>Electrical and Computer Engineering, University of Pittsburgh, USA</b><br/><b>Computer Science and Engineering, University of Texas at Arlington, USA</b><br/><b>cid:93)School of Electronic Engineering, Xidian University, China</b><br/><b>cid:92)School of Information Technologies, University of Sydney, Australia</b></td><td>('2331771', 'Kamran Ghasedi Dizaji', 'kamran ghasedi dizaji')<br/>('10797930', 'Amirhossein Herandi', 'amirhossein herandi')<br/>('1748032', 'Heng Huang', 'heng huang')</td><td>kamran.ghasedi@gmail.com, amirhossein.herandi@uta.edu, chdeng@mail.xidian.edu.cn +<br/>tom.cai@sydney.edu.au, heng.huang@pitt.edu +</td></tr><tr><td>f3fcaae2ea3e998395a1443c87544f203890ae15</td><td></td><td></td><td></td></tr><tr><td>f3015be0f9dbc1a55b6f3dc388d97bb566ff94fe</td><td>A Study on the Effective Approach +<br/>to Illumination-Invariant Face Recognition +<br/>Based on a Single Image +<br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, 518055, China</b><br/>2 Shenzhen Key Laboratory for Visual Computing and Analytics, Shenzhen, 518055, China +</td><td>('31361063', 'Jiapei Zhang', 'jiapei zhang')<br/>('2002129', 'Xiaohua Xie', 'xiaohua xie')</td><td>{jp.zhang,xiaohua.xie}@siat.ac.cn, +<br/>sysuxiexh@gmail.com +</td></tr><tr><td>f3d9e347eadcf0d21cb0e92710bc906b22f2b3e7</td><td>NosePose: a competitive, landmark-free +<br/>methodology for head pose estimation in the wild +<br/>IMAGO Research Group - Universidade Federal do Paran´a +</td><td>('37435823', 'Antonio C. P. Nascimento', 'antonio c. p. nascimento')<br/>('1800955', 'Olga R. P. Bellon', 'olga r. p. bellon')</td><td>{flavio,antonio.paes,olga,luciano}@ufpr.br +</td></tr><tr><td>f3a59d85b7458394e3c043d8277aa1ffe3cdac91</td><td>Query-Free Attacks on Industry-Grade Face Recognition Systems under Resource +<br/>Constraints +<br/><b>Chinese University of Hong Kong</b><br/><b>Indiana University</b><br/><b>Chinese University of Hong Kong</b></td><td>('1807925', 'Di Tang', 'di tang')<br/>('47119002', 'XiaoFeng Wang', 'xiaofeng wang')<br/>('3297454', 'Kehuan Zhang', 'kehuan zhang')</td><td>td016@ie.cuhk.edu.hk +<br/>xw7@indiana.edu +<br/>khzhang@ie.cuhk.edu.hk +</td></tr><tr><td>f3f77b803b375f0c63971b59d0906cb700ea24ed</td><td>Advances in Electrical and Computer Engineering Volume 9, Number 3, 2009 +<br/>Feature Extraction for Facial Expression +<br/>Recognition based on Hybrid Face Regions +<br/>Seyed M. LAJEVARDI, Zahir M. HUSSAIN +<br/><b>RMIT University, Australia</b></td><td></td><td>seyed.lajevardi @ rmit.edu.au +</td></tr><tr><td>f355e54ca94a2d8bbc598e06e414a876eb62ef99</td><td></td><td></td><td></td></tr><tr><td>f3df296de36b7c114451865778e211350d153727</td><td>Spatio-Temporal Facial Expression Recognition Using Convolutional +<br/>Neural Networks and Conditional Random Fields +<br/><b>University of Denver, Denver, CO</b></td><td>('3093835', 'Mohammad H. Mahoor', 'mohammad h. mahoor')</td><td>behzad.hasani@du.edu, and mmahoor@du.edu +</td></tr><tr><td>f3ea181507db292b762aa798da30bc307be95344</td><td>Covariance Pooling for Facial Expression Recognition +<br/>†Computer Vision Lab, ETH Zurich, Switzerland +<br/>‡VISICS, KU Leuven, Belgium +</td><td>('32610154', 'Dinesh Acharya', 'dinesh acharya')<br/>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('35268081', 'Danda Pani Paudel', 'danda pani paudel')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{acharyad, zhiwu.huang, paudel, vangool}@vision.ee.ethz.ch +</td></tr><tr><td>f3fed71cc4fc49b02067b71c2df80e83084b2a82</td><td>Published as a conference paper at ICLR 2018 +<br/>LEARNING SPARSE LATENT REPRESENTATIONS WITH +<br/>THE DEEP COPULA INFORMATION BOTTLENECK +<br/><b>University of Basel, Switzerland</b></td><td>('30069186', 'Aleksander Wieczorek', 'aleksander wieczorek')<br/>('30537851', 'Mario Wieser', 'mario wieser')<br/>('2620254', 'Damian Murezzan', 'damian murezzan')<br/>('39891341', 'Volker Roth', 'volker roth')</td><td>{firstname.lastname}@unibas.ch +</td></tr><tr><td>f3cf10c84c4665a0b28734f5233d423a65ef1f23</td><td>Title +<br/>Temporal Exemplar-based Bayesian Networks for facial +<br/>expression recognition +<br/>Author(s) +<br/>Shang, L; Chan, KP +<br/>Citation +<br/>Proceedings - 7Th International Conference On Machine +<br/>Learning And Applications, Icmla 2008, 2008, p. 16-22 +<br/>Issued Date +<br/>2008 +<br/>URL +<br/>http://hdl.handle.net/10722/61208 +<br/>Rights +<br/>This work is licensed under a Creative Commons Attribution- +<br/>NonCommercial-NoDerivatives 4.0 International License.; +<br/>International Conference on Machine Learning and Applications +<br/>Proceedings. Copyright © IEEE.; ©2008 IEEE. Personal use of +<br/>this material is permitted. However, permission to +<br/>reprint/republish this material for advertising or promotional +<br/>purposes or for creating new collective works for resale or +<br/>redistribution to servers or lists, or to reuse any copyrighted +<br/>component of this work in other works must be obtained from +<br/>the IEEE. +</td><td></td><td></td></tr><tr><td>f35a493afa78a671b9d2392c69642dcc3dd2cdc2</td><td>Automatic Attribute Discovery with Neural +<br/>Activations +<br/><b>University of North Carolina at Chapel Hill, USA</b><br/>2 NTT Media Intelligence Laboratories, Japan +<br/><b>Tohoku University, Japan</b></td><td>('3302783', 'Sirion Vittayakorn', 'sirion vittayakorn')<br/>('1706592', 'Takayuki Umeda', 'takayuki umeda')<br/>('2023568', 'Kazuhiko Murasaki', 'kazuhiko murasaki')<br/>('1745497', 'Kyoko Sudo', 'kyoko sudo')<br/>('1718872', 'Takayuki Okatani', 'takayuki okatani')<br/>('1721910', 'Kota Yamaguchi', 'kota yamaguchi')</td><td></td></tr><tr><td>f3b7938de5f178e25a3cf477107c76286c0ad691</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, MARCH 2017 +<br/>Object Detection with Deep Learning: A Review +</td><td>('33698309', 'Zhong-Qiu Zhao', 'zhong-qiu zhao')<br/>('36659418', 'Peng Zheng', 'peng zheng')<br/>('51132438', 'Shou-tao Xu', 'shou-tao xu')<br/>('1748808', 'Xindong Wu', 'xindong wu')</td><td></td></tr><tr><td>ebedc841a2c1b3a9ab7357de833101648281ff0e</td><td></td><td></td><td></td></tr><tr><td>eb526174fa071345ff7b1fad1fad240cd943a6d7</td><td>Deeply Vulnerable – A Study of the Robustness of Face Recognition to +<br/>Presentation Attacks +</td><td>('1990628', 'Amir Mohammadi', 'amir mohammadi')<br/>('1952348', 'Sushil Bhattacharjee', 'sushil bhattacharjee')</td><td></td></tr><tr><td>eb100638ed73b82e1cce8475bb8e180cb22a09a2</td><td>Temporal Action Detection with Structured Segment Networks +<br/><b>The Chinese University of Hong Kong</b><br/>2Computer Vision Laboratory, ETH Zurich, Switzerland +</td><td>('47827548', 'Yue Zhao', 'yue zhao')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('2765994', 'Zhirong Wu', 'zhirong wu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('1807606', 'Dahua Lin', 'dahua lin')</td><td></td></tr><tr><td>eb6ee56e085ebf473da990d032a4249437a3e462</td><td>Age/Gender Classification with Whole-Component +<br/>Convolutional Neural Networks (WC-CNN) +<br/><b>University of Southern California, Los Angeles, CA 90089, USA</b></td><td>('39004239', 'Chun-Ting Huang', 'chun-ting huang')<br/>('7022231', 'Yueru Chen', 'yueru chen')<br/>('35521292', 'Ruiyuan Lin', 'ruiyuan lin')<br/>('9363144', 'C.-C. Jay Kuo', 'c.-c. jay kuo')</td><td>E-mail: {chuntinh, yueruche, ruiyuanl}@usc.edu, cckuo@sipi.usc.edu +</td></tr><tr><td>eb8519cec0d7a781923f68fdca0891713cb81163</td><td>Temporal Non-Volume Preserving Approach to Facial Age-Progression and +<br/>Age-Invariant Face Recognition +<br/><b>Computer Science and Software Engineering, Concordia University, Montr eal, Qu ebec, Canada</b><br/>2 CyLab Biometrics Center and the Department of Electrical and Computer Engineering, +<br/><b>Carnegie Mellon University, Pittsburgh, PA, USA</b></td><td>('1876581', 'Chi Nhan Duong', 'chi nhan duong')<br/>('2687827', 'Kha Gia Quach', 'kha gia quach')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('6131978', 'T. Hoang Ngan Le', 't. hoang ngan le')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td>{chinhand, kquach, kluu, thihoanl}@andrew.cmu.edu, msavvid@ri.cmu.edu +</td></tr><tr><td>ebb1c29145d31c4afa3c9be7f023155832776cd3</td><td>CASME II: An Improved Spontaneous Micro-Expression +<br/>Database and the Baseline Evaluation +<br/><b>State Key Laboratory of Brain and Cognitive Science, Institute of Psychology, Chinese Academy of Sciences, Beijing, China, 2 University of Chinese Academy of Sciences</b><br/><b>Beijing, China, 3 Center for Machine Vision Research, Department of Computer Science and Engineering, University of Oulu, Oulu, Finland, 4 TNList, Department of</b><br/><b>Computer Science and Technology, Tsinghua University, Beijing, China</b></td><td>('9185305', 'Wen-Jing Yan', 'wen-jing yan')<br/>('39522870', 'Xiaobai Li', 'xiaobai li')<br/>('2819642', 'Su-Jing Wang', 'su-jing wang')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')<br/>('1715826', 'Yong-Jin Liu', 'yong-jin liu')<br/>('1838009', 'Yu-Hsin Chen', 'yu-hsin chen')<br/>('1684007', 'Xiaolan Fu', 'xiaolan fu')</td><td></td></tr><tr><td>eb566490cd1aa9338831de8161c6659984e923fd</td><td>From Lifestyle Vlogs to Everyday Interactions +<br/>EECS Department, UC Berkeley +</td><td>('1786435', 'David F. Fouhey', 'david f. fouhey')<br/>('1763086', 'Alexei A. Efros', 'alexei a. efros')<br/>('1689212', 'Jitendra Malik', 'jitendra malik')</td><td></td></tr><tr><td>eb9312458f84a366e98bd0a2265747aaed40b1a6</td><td>1-4244-1437-7/07/$20.00 ©2007 IEEE +<br/>IV - 473 +<br/>ICIP 2007 +</td><td></td><td></td></tr><tr><td>eb716dd3dbd0f04e6d89f1703b9975cad62ffb09</td><td>Copyright +<br/>by +<br/>2012 +</td><td>('1883898', 'Yong Jae Lee', 'yong jae lee')</td><td></td></tr><tr><td>eb4d2ec77fae67141f6cf74b3ed773997c2c0cf6</td><td>Int. J. Information Technology and Management, Vol. 11, Nos. 1/2, 2012 +<br/>35 +<br/>A new soft biometric approach for keystroke +<br/>dynamics based on gender recognition +<br/><b>GREYC Research Lab</b><br/>ENSICAEN – Université de Caen Basse Normandie – CNRS, +<br/>14000 Caen, France +<br/>Fax: +33-231538110 +<br/>*Corresponding author +</td><td>('2615638', 'Romain Giot', 'romain giot')<br/>('1793765', 'Christophe Rosenberger', 'christophe rosenberger')</td><td>E-mail: romain.giot@ensicaen.fr +<br/>E-mail: christophe.rosenberger@ensicaen.fr +</td></tr><tr><td>ebb7cc67df6d90f1c88817b20e7a3baad5dc29b9</td><td>Journal of Computational Mathematics +<br/>Vol.xx, No.x, 200x, 1–25. +<br/>http://www.global-sci.org/jcm +<br/>doi:?? +<br/>Fast algorithms for Higher-order Singular Value Decomposition +<br/>from incomplete data* +<br/><b>University of Alabama, Tuscaloosa, AL</b></td><td>('40507939', 'Yangyang Xu', 'yangyang xu')</td><td>Email: yangyang.xu@ua.edu +</td></tr><tr><td>ebabd1f7bc0274fec88a3dabaf115d3e226f198f</td><td>Driver drowsiness detection system based on feature +<br/>representation learning using various deep networks +<br/>School of Electrical Engineering, KAIST, +<br/>Guseong-dong, Yuseong-gu, Dajeon, Rep. of Korea +</td><td>('1989730', 'Sanghyuk Park', 'sanghyuk park')<br/>('1773194', 'Fei Pan', 'fei pan')<br/>('3315036', 'Sunghun Kang', 'sunghun kang')</td><td>{shine0624, feipan, sunghun.kang, cd yoo}@kaist.ac.kr +</td></tr><tr><td>eb70c38a350d13ea6b54dc9ebae0b64171d813c9</td><td>On Graph-Structured Discrete +<br/>Labelling Problems in Computer +<br/>Vision: Learning, Inference and +<br/>Applications +<br/>Submitted in partial fulfillment of the requirements for +<br/>the degree of +<br/>Doctor of Philosophy +<br/>in +<br/>Electrical and Computer Engineering +<br/><b>M.S., Electrical and Computer Engineering, Carnegie Mellon University</b><br/><b>B.Tech., Electronics Engineering, Institute of Technology, Banaras Hindu University</b><br/><b>Carnegie Mellon University</b><br/>August, 2010 +</td><td>('1746610', 'Dhruv Batra', 'dhruv batra')</td><td></td></tr><tr><td>ebb9d53668205c5797045ba130df18842e3eadef</td><td></td><td></td><td></td></tr><tr><td>eb027969f9310e0ae941e2adee2d42cdf07d938c</td><td>VGGFace2: A dataset for recognising faces across pose and age +<br/><b>Visual Geometry Group, University of Oxford</b></td><td>('46632720', 'Qiong Cao', 'qiong cao')<br/>('46980108', 'Li Shen', 'li shen')<br/>('10096695', 'Weidi Xie', 'weidi xie')<br/>('3188342', 'Omkar M. Parkhi', 'omkar m. parkhi')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>{qiong,lishen,weidi,omkar,az}@robots.ox.ac.uk +</td></tr><tr><td>eb48a58b873295d719827e746d51b110f5716d6c</td><td>Face Alignment Using K-cluster Regression Forests +<br/>With Weighted Splitting +</td><td>('2393538', 'Marek Kowalski', 'marek kowalski')<br/>('1930272', 'Jacek Naruniec', 'jacek naruniec')</td><td></td></tr><tr><td>eb7b387a3a006609b89ca5ed0e6b3a1d5ecb5e5a</td><td>Facial Expression Recognition using Neural +<br/>Network +<br/><b>National Cheng Kung University</b><br/>Tainan, Taiwan, R.O.C. +<br/> +</td><td>('1751725', 'Shen-Chuan Tai', 'shen-chuan tai')<br/>('2142418', 'Yu-Yi Liao', 'yu-yi liao')<br/>('1925097', 'Chien-Shiang Hong', 'chien-shiang hong')</td><td>sctai@mail.ncku.edu.tw hhf93d@lily.ee.ncku.edu.tw zgz@lily.ee.ncku.edu.tw +<br/>lyy94d@lily.ee.ncku.edu.tw hcs95d@dcmc.ee.ncku.edu.tw +</td></tr><tr><td>ebd5df2b4105ba04cef4ca334fcb9bfd6ea0430c</td><td>Fast Localization of Facial Landmark Points +<br/><b>University of Zagreb, Faculty of Electrical Engineering and Computing, Unska 3, 10000 Zagreb, Croatia</b><br/><b>Link oping University, SE-581 83 Link oping, Sweden</b><br/>March 28, 2014 +</td><td>('3013350', 'Miroslav Frljak', 'miroslav frljak')<br/>('1767736', 'Robert Forchheimer', 'robert forchheimer')</td><td></td></tr><tr><td>ebf204e0a3e137b6c24e271b0d55fa49a6c52b41</td><td>Master of Science Thesis in Electrical Engineering +<br/><b>Link ping University</b><br/>Visual Tracking Using +<br/>Deep Motion Features +</td><td>('8161428', 'Susanna Gladh', 'susanna gladh')</td><td></td></tr><tr><td>c71f36c9376d444075de15b1102b4974481be84d</td><td>3D Morphable Models: Data +<br/>Pre-Processing, Statistical Analysis and +<br/>Fitting +<br/>Submitted for the degree of Doctor of Philosophy +<br/>Department of Computer Science +<br/><b>The University of York</b><br/>June, 2011 +</td><td>('37519514', 'Ankur Patel', 'ankur patel')</td><td></td></tr><tr><td>c7c53d75f6e963b403057d8ba5952e4974a779ad</td><td><b>Purdue University</b><br/>Purdue e-Pubs +<br/>Open Access Theses +<br/>8-2016 +<br/>Theses and Dissertations +<br/>Aging effects in automated face recognition +<br/><b>Purdue University</b><br/>Follow this and additional works at: http://docs.lib.purdue.edu/open_access_theses +<br/>Recommended Citation +<br/>Agamez, Miguel Cedeno, "Aging effects in automated face recognition" (2016). Open Access Theses. 930. +<br/>http://docs.lib.purdue.edu/open_access_theses/930 +<br/>additional information. +</td><td></td><td>This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact epubs@purdue.edu for +</td></tr><tr><td>c79cf7f61441195404472102114bcf079a72138a</td><td>Pose-Invariant 2D Face Recognition by Matching +<br/>Using Graphical Models +<br/>Submitted for the Degree of +<br/>Doctor of Philosophy +<br/>from the +<br/><b>University of Surrey</b><br/>Center for Vision, Speech and Signal Processing +<br/>Faculty of Engineering and Physical Sciences +<br/><b>University of Surrey</b><br/>Guildford, Surrey GU2 7XH, U.K. +<br/>September 2010 +</td><td>('1690611', 'Shervin Rahimzadeh Arashloo', 'shervin rahimzadeh arashloo')<br/>('1690611', 'Shervin Rahimzadeh Arashloo', 'shervin rahimzadeh arashloo')</td><td></td></tr><tr><td>c73dd452c20460f40becb1fd8146239c88347d87</td><td>Manifold Constrained Low-Rank Decomposition +<br/>1State Key Laboratory of Satellite Navigation System and Equipment Technology, Shijiazhuang, China +<br/><b>Center for Research in Computer Vision (CRCV), University of Central Florida (UCF</b><br/><b>School of Automation Science and Electrical Engineering, Beihang University, Beijing, China</b><br/>4 Istituto Italiano di Tecnologia, Genova, Italy +</td><td>('9497155', 'Chen Chen', 'chen chen')<br/>('1740430', 'Baochang Zhang', 'baochang zhang')<br/>('1714730', 'Alessio Del Bue', 'alessio del bue')<br/>('1727204', 'Vittorio Murino', 'vittorio murino')</td><td>chenchen870713@gmail.com, alessio.delbue@iit.it, bczhang@buaa.edu.cn, vittorio.murino@iit.it ∗ +</td></tr><tr><td>c7e4c7be0d37013de07b6d829a3bf73e1b95ad4e</td><td>The International Journal of Multimedia & Its Applications (IJMA) Vol.5, No.5, October 2013 +<br/>DYNEMO: A VIDEO DATABASE OF NATURAL FACIAL +<br/>EXPRESSIONS OF EMOTIONS +<br/>1LIP, Univ. Grenoble Alpes, BP 47 - 38040 Grenoble Cedex 9, France +<br/>2LIG, Univ. Grenoble Alpes, BP 53 - 38041 Grenoble Cedex 9, France +</td><td>('3209946', 'Anna Tcherkassof', 'anna tcherkassof')<br/>('20944713', 'Damien Dupré', 'damien dupré')<br/>('2357225', 'Brigitte Meillon', 'brigitte meillon')<br/>('2872246', 'Nadine Mandran', 'nadine mandran')<br/>('1870899', 'Michel Dubois', 'michel dubois')<br/>('1828394', 'Jean-Michel Adam', 'jean-michel adam')</td><td></td></tr><tr><td>c72e6992f44ce75a40f44be4365dc4f264735cfb</td><td>Story Understanding in Video +<br/>Advertisements +<br/>Department of Computer Science +<br/><b>University of Pittsburgh</b><br/>Pennsylvania, United States +</td><td>('9085797', 'Keren Ye', 'keren ye')<br/>('51150048', 'Kyle Buettner', 'kyle buettner')<br/>('1770205', 'Adriana Kovashka', 'adriana kovashka')<br/>('9085797', 'Keren Ye', 'keren ye')<br/>('51150048', 'Kyle Buettner', 'kyle buettner')<br/>('1770205', 'Adriana Kovashka', 'adriana kovashka')</td><td>yekeren@cs.pitt.edu +<br/>buettnerk@pitt.edu +<br/>kovashka@cs.pitt.edu +</td></tr><tr><td>c74aba9a096379b3dbe1ff95e7af5db45c0fd680</td><td>Neuro-Fuzzy Analysis of Facial Action Units +<br/>and Expressions +<br/>Digital Signal Processing Lab, Department of Computer Engineering +<br/><b>Sharif University of Technology</b><br/>Tehran, Iran, Tel: +98 21 6616 4632 +</td><td>('1736464', 'Mahmoud Khademi', 'mahmoud khademi')<br/>('2936650', 'Mohammad Taghi Manzuri', 'mohammad taghi manzuri')<br/>('1702826', 'Mohammad Hadi Kiapour', 'mohammad hadi kiapour')</td><td>khademi@ce.sharif.edu, manzuri@sharif.edu, kiapour@ee.sharif.edu +</td></tr><tr><td>c7de0c85432ad17a284b5b97c4f36c23f506d9d1</td><td>INTERSPEECH 2011 +<br/>RANSAC-based Training Data Selection for Speaker State Recognition +<br/><b>Multimedia, Vision and Graphics Laboratory, Koc University, Istanbul, Turkey</b><br/><b>Bahc es ehir University, Istanbul, Turkey</b><br/><b>Ozye gin University, Istanbul, Turkey</b></td><td>('1777185', 'Elif Bozkurt', 'elif bozkurt')<br/>('1749677', 'Engin Erzin', 'engin erzin')</td><td>ebozkurt, eerzin@ku.edu.tr, cigdem.eroglu@bahcesehir.edu.tr, tanju.erdem@ozyegin.edu.tr +</td></tr><tr><td>c7c5f0fe1fcaf3787c7f78f7dc62f3497dcfdf3c</td><td>THE IMPACT OF PRODUCT PHOTO ON ONLINE CONSUMER +<br/>PURCHASE INTENTION: AN IMAGE-PROCESSING ENABLED +<br/>EMPIRICAL STUDY +</td><td>('39306563', 'Xin Li', 'xin li')<br/>('2762720', 'Mengyue Wang', 'mengyue wang')<br/>('39016300', 'Yubo Chen', 'yubo chen')</td><td>Xin.Li.PhD@gmail.com +<br/>Kong, menwang-c@my.cityu.edu.hk +<br/>chenyubo@sem.tsinghua.edu.cn +</td></tr><tr><td>c7f752eea91bf5495a4f6e6a67f14800ec246d08</td><td>EXPLORING THE TRANSFER +<br/>LEARNING ASPECT OF DEEP +<br/>NEURAL NETWORKS IN FACIAL +<br/>INFORMATION PROCESSING +<br/><b>A DISSERTATION SUBMITTED TO THE UNIVERSITY OF MANCHESTER</b><br/>FOR THE DEGREE OF MASTER OF SCIENCE +<br/>IN THE FACULTY OF ENGINEERING AND PHYSICAL SCIENCES +<br/>2015 +<br/>By +<br/>Crefeda Faviola Rodrigues +<br/>School of Computer Science +</td><td></td><td></td></tr><tr><td>c71217b2b111a51a31cf1107c71d250348d1ff68</td><td>One Network to Solve Them All — Solving Linear Inverse Problems +<br/>using Deep Projection Models +<br/><b>Carnegie Mellon University, Pittsburgh, PA</b></td><td>('2088535', 'Chun-Liang Li', 'chun-liang li')<br/>('1783087', 'B. V. K. Vijaya Kumar', 'b. v. k. vijaya kumar')<br/>('1745861', 'Aswin C. Sankaranarayanan', 'aswin c. sankaranarayanan')</td><td></td></tr><tr><td>c758b9c82b603904ba8806e6193c5fefa57e9613</td><td>Heterogeneous Face Recognition with CNNs +<br/>INRIA Grenoble, Laboratoire Jean Kuntzmann +</td><td>('2143851', 'Shreyas Saxena', 'shreyas saxena')<br/>('34602236', 'Jakob Verbeek', 'jakob verbeek')</td><td>{firstname.lastname}@inria.fr +</td></tr><tr><td>c7c03324833ba262eeaada0349afa1b5990c1ea7</td><td>A Wearable Face Recognition System on Google +<br/>Glass for Assisting Social Interactions +<br/><b>Institute for Infocomm Research, Singapore</b></td><td>('1709001', 'Bappaditya Mandal', 'bappaditya mandal')<br/>('35718875', 'Liyuan Li', 'liyuan li')<br/>('1694051', 'Cheston Tan', 'cheston tan')</td><td>Email address: bmandal@i2r.a-star.edu.sg (∗Contact author: Bappaditya Mandal); +<br/>{scchia, lyli, vijay, cheston-tan, joohwee}@i2r.a-star.edu.sg +</td></tr><tr><td>c76f64e87f88475069f7707616ad9df1719a6099</td><td>T-RECS: Training for Rate-Invariant +<br/>Embeddings by Controlling Speed for Action +<br/>Recognition +<br/><b>University of Michigan</b></td><td>('31646172', 'Madan Ravi Ganesh', 'madan ravi ganesh')<br/>('24337238', 'Eric Hofesmann', 'eric hofesmann')<br/>('40893359', 'Byungsu Min', 'byungsu min')<br/>('40893002', 'Nadha Gafoor', 'nadha gafoor')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')</td><td></td></tr><tr><td>c7f0c0636d27a1d45b8fcef37e545b902195d937</td><td>Towards Around-Device Interaction using Corneal Imaging +<br/><b>Coburg University</b><br/><b>Coburg University</b></td><td>('49770541', 'Daniel Schneider', 'daniel schneider')<br/>('2708269', 'Jens Grubert', 'jens grubert')</td><td>daniel.schneider@hs-coburg.de +<br/>jg@jensgrubert.de +</td></tr><tr><td>c7c8d150ece08b12e3abdb6224000c07a6ce7d47</td><td>DeMeshNet: Blind Face Inpainting for Deep MeshFace Verification +<br/>National Laboratory of Pattern Recognition, CASIA +<br/>Center for Research on Intelligent Perception and Computing, CASIA +</td><td>('50202300', 'Shu Zhang', 'shu zhang')</td><td>{shu.zhang,rhe,tnt}@nlpr.ia.ac.cn +</td></tr><tr><td>c78fdd080df01fff400a32fb4cc932621926021f</td><td>Robust Automatic Facial Expression Detection +<br/>Method +<br/><b>Institute for Pattern Recognition and Artificial Intelligence/ Huazhong University of Science and Technology, Wuhan</b><br/><b>Institute for Pattern Recognition and Artificial Intelligence/ Huazhong University of Science and Technology, Wuhan</b><br/>China +<br/>China +</td><td>('33024921', 'Yan Ouyang', 'yan ouyang')<br/>('1707161', 'Nong Sang', 'nong sang')</td><td>Email:oyy_01@163.com +<br/>Email: nsang@hust.edu.cn +</td></tr><tr><td>c74b1643a108939c6ba42ae4de55cb05b2191be5</td><td>NON-NEGATIVE MATRIX FACTORIZATION FOR FACE +<br/>ILLUMINATION ANALYSIS +<br/><b>CVSSP, University of Surrey</b><br/><b>CVSSP, University of Surrey</b><br/><b>CVSSP, University of Surrey</b><br/>Guildford, Surrey +<br/>UK GU2 7XH +<br/>Guildford, Surrey +<br/>UK GU2 7XH +<br/>Guildford, Surrey +<br/>UK GU2 7XH +</td><td>('38746097', 'Xuan Zou', 'xuan zou')<br/>('39685698', 'Wenwu Wang', 'wenwu wang')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td>xuan.zou@surrey.ac.uk +<br/>w.wang@surrey.ac.uk +<br/>j.kittler@surrey.ac.uk +</td></tr><tr><td>c75e6ce54caf17b2780b4b53f8d29086b391e839</td><td>ExpNet: Landmark-Free, Deep, 3D Facial Expressions +<br/><b>Institute for Robotics and Intelligent Systems, USC, CA, USA</b><br/><b>Information Sciences Institute, USC, CA, USA</b><br/><b>The Open University of Israel, Israel</b></td><td>('1752756', 'Feng-Ju Chang', 'feng-ju chang')<br/>('46634688', 'Anh Tuan Tran', 'anh tuan tran')<br/>('1756099', 'Tal Hassner', 'tal hassner')<br/>('11269472', 'Iacopo Masi', 'iacopo masi')</td><td>{fengjuch,anhttran,iacopoma,nevatia,medioni}@usc.edu, hassner@openu.ac.il +</td></tr><tr><td>c0723e0e154a33faa6ff959d084aebf07770ffaf</td><td>Interpolation Between Eigenspaces Using +<br/>Rotation in Multiple Dimensions +<br/><b>Graduate School of Information Science, Nagoya University, Japan</b><br/>2 No Japan Society for the Promotion of Science +<br/><b>Japan</b></td><td>('1685524', 'Tomokazu Takahashi', 'tomokazu takahashi')<br/>('2833316', 'Lina', 'lina')<br/>('1679187', 'Ichiro Ide', 'ichiro ide')<br/>('1680642', 'Yoshito Mekada', 'yoshito mekada')<br/>('1725612', 'Hiroshi Murase', 'hiroshi murase')</td><td>ttakahashi@murase.m.is.nagoya-u.ac.jp +</td></tr><tr><td>c03f48e211ac81c3867c0e787bea3192fcfe323e</td><td>INTERSPEECH 2016 +<br/>September 8–12, 2016, San Francisco, USA +<br/>Mahalanobis Metric Scoring Learned from Weighted Pairwise Constraints in +<br/>I-vector Speaker Recognition System +<br/><b>School of Computer Information Engineering, Jiangxi Normal University, Nanchang, China</b></td><td>('3308432', 'Zhenchun Lei', 'zhenchun lei')<br/>('2947033', 'Yanhong Wan', 'yanhong wan')<br/>('1853437', 'Jian Luo', 'jian luo')<br/>('2956877', 'Yingen Yang', 'yingen yang')</td><td>zhenchun.lei@hotmail.com, wyanhhappy@126.com, +<br/>luo.jian@hotmail.com, ygyang@jxnu.edu.cn +</td></tr><tr><td>c038beaa228aeec174e5bd52460f0de75e9cccbe</td><td>Temporal Segment Networks for Action +<br/>Recognition in Videos +</td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('48708388', 'Zhe Wang', 'zhe wang')<br/>('40612284', 'Yu Qiao', 'yu qiao')<br/>('1807606', 'Dahua Lin', 'dahua lin')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>c043f8924717a3023a869777d4c9bee33e607fb5</td><td>Emotion Separation Is Completed Early and It Depends +<br/>on Visual Field Presentation +<br/><b>Lab for Human Brain Dynamics, RIKEN Brain Science Institute, Wakoshi, Saitama, Japan, 2 Lab for Human Brain Dynamics, AAI Scientific Cultural Services Ltd., Nicosia</b><br/>Cyprus +</td><td>('2259342', 'Lichan Liu', 'lichan liu')<br/>('2348276', 'Andreas A. Ioannides', 'andreas a. ioannides')</td><td></td></tr><tr><td>c05a7c72e679745deab9c9d7d481f7b5b9b36bdd</td><td>NPS-CS-11-005 +<br/> +<br/> +<br/>NAVAL +<br/>POSTGRADUATE +<br/>SCHOOL +<br/>MONTEREY, CALIFORNIA +<br/>by +<br/>BIOMETRIC CHALLENGES FOR FUTURE DEPLOYMENTS: +<br/>A STUDY OF THE IMPACT OF GEOGRAPHY, CLIMATE, CULTURE, +<br/> AND SOCIAL CONDITIONS ON THE EFFECTIVE +<br/>COLLECTION OF BIOMETRICS +<br/>April 2011 +<br/>Approved for public release; distribution is unlimited +</td><td>('3337733', 'Paul C. Clark', 'paul c. clark')</td><td></td></tr><tr><td>c03e01717b2d93f04cce9b5fd2dcfd1143bcc180</td><td>Locality-constrained Active Appearance Model +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>CAS), Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b></td><td>('1874505', 'Xiaowei Zhao', 'xiaowei zhao')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1695600', 'Xiujuan Chai', 'xiujuan chai')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>mathzxw2002@gmail.com,{sgshan,chaixiujuan,xlchen}@ict.ac.cn +</td></tr><tr><td>c0ff7dc0d575658bf402719c12b676a34271dfcd</td><td>A New Incremental Optimal Feature Extraction +<br/>Method for On-line Applications +<br/><b>K. N. Toosi University of</b><br/>Technology, Tehran, Iran +<br/>21−Σ +</td><td>('2784763', 'Youness Aliyari Ghassabeh', 'youness aliyari ghassabeh')<br/>('2060085', 'Hamid Abrishami Moghaddam', 'hamid abrishami moghaddam')</td><td>y_aliyari@ee.kntu.ac.ir, moghadam@saba.kntu.ac.ir +</td></tr><tr><td>c02847a04a99a5a6e784ab580907278ee3c12653</td><td>Fine Grained Video Classification for +<br/>Endangered Bird Species Protection +<br/>Non-Thesis MS Final Report +<br/>1. Introduction +<br/>1.1 Background +<br/>This project is about detecting eagles in videos. Eagles are endangered species at the brim of +<br/>extinction since 1980s. With the bans of harmful pesticides, the number of eagles keep increasing. +<br/>However, recent studies on golden eagles’ activities in the vicinity of wind turbines have shown +<br/>significant number of turbine blade collisions with eagles as the major cause of eagles’ mortality. [1] +<br/>This project is a part of a larger research project to build an eagle detection and deterrent system +<br/>on wind turbine toward reducing eagles’ mortality. [2] The critical component of this study is a +<br/>computer vision system for eagle detection in videos. The key requirement are that the system should +<br/>work in real time and detect eagles at a far distance from the camera (i.e. in low resolution). +<br/>There are three different bird species in my dataset - falcon, eagle and seagull. The reason for +<br/>involving only these three species is based on the real world situation. Wind turbines are always +<br/>installed near coast and mountain hill where falcons and seagulls will be the majority. So my model +<br/>will classify the minority eagles out of other bird species during the immigration season and protecting +<br/>them by using the deterrent system. +<br/>1.2 Brief Approach +<br/>Our approach represents a unified deep-learning architecture for eagle detection. Given videos, +<br/>our goal is to detect eagle species at far distance from the camera, using both appearance and bird +<br/>motion cues, so as to meet the recall-precision rates set by the user. Detecting eagle is a challenging +<br/>task because of the following reasons. Frist, an eagle flies fast and high in the sky which means that +<br/>we need a lens with wide angle such that captures their movement. However, a camera with wide +<br/>angle produces a low resolution and low quality video and the detailed appearance of bird is +<br/>compromised. Second, current neural network typically take as input low resolution images. This is +<br/>because a higher resolution image will require larger filters and deeper networks which is turn hard to +<br/>train [3]. So it is not clear whether the low resolution will cause challenge for fine-grained +<br/>classification task. Last but not the least, there is not a large training database like PASCAL, MNIST +</td><td>('2355840', 'Chenyu Wang', 'chenyu wang')</td><td></td></tr><tr><td>c0c8d720658374cc1ffd6116554a615e846c74b5</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>Modeling Multimodal Clues in a Hybrid Deep +<br/>Learning Framework for Video Classification +</td><td>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('3099139', 'Zuxuan Wu', 'zuxuan wu')<br/>('8053308', 'Jinhui Tang', 'jinhui tang')<br/>('3233021', 'Zechao Li', 'zechao li')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td></td></tr><tr><td>c035c193eed5d72c7f187f0bc880a17d217dada0</td><td>Local Gradient Gabor Pattern (LGGP) with Applications in +<br/>Face Recognition, Cross-spectral Matching and Soft +<br/>Biometrics +<br/><b>West Virginia University</b><br/><b>Michigan State University</b><br/>Morgantown, WV, USA +<br/>East Lansing, MI, USA +</td><td>('1751335', 'Cunjian Chen', 'cunjian chen')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td></td></tr><tr><td>c0cdaeccff78f49f4604a6d263dc6eb1bb8707d5</td><td>Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'16 | +<br/>263 +<br/>MLP Neural Network Based Approach for +<br/>Facial Expression Analysis +<br/><b>Kent State University, Kent, Ohio, USA</b><br/>2 Department of Robotic Engineering, AU-TNB, Tehran, Iran +<br/>the efficiency of +</td><td></td><td></td></tr><tr><td>c00f402b9cfc3f8dd2c74d6b3552acbd1f358301</td><td>LEARNING DEEP REPRESENTATION FROM COARSE TO FINE FOR FACE ALIGNMENT +<br/><b>Shanghai Jiao Tong University, China</b></td><td>('3403352', 'Zhiwen Shao', 'zhiwen shao')<br/>('7406856', 'Shouhong Ding', 'shouhong ding')<br/>('3450479', 'Yiru Zhao', 'yiru zhao')<br/>('3451401', 'Qinchuan Zhang', 'qinchuan zhang')<br/>('8452947', 'Lizhuang Ma', 'lizhuang ma')</td><td>{shaozhiwen, feiben, yiru.zhao, qinchuan.zhang}@sjtu.edu.cn, ma-lz@cs.sjtu.edu.cn +</td></tr><tr><td>c089c7d8d1413b54f59fc410d88e215902e51638</td><td>TVParser: An Automatic TV Video Parsing Method +<br/><b>National Lab of Pattern Recognition, Institute of Automation</b><br/>Chinese Academy of Sciences, Beijing, China, 100190 +<br/><b>China-Singapore Institute of Digital Media, Singapore</b></td><td>('1690954', 'Chao Liang', 'chao liang')<br/>('1688633', 'Changsheng Xu', 'changsheng xu')<br/>('1709439', 'Jian Cheng', 'jian cheng')<br/>('1694235', 'Hanqing Lu', 'hanqing lu')</td><td>fcliang,csxu,jcheng,luhqg@nlpr.ia.ac.cn +</td></tr><tr><td>c0ee89dc2dad76147780f96294de9e421348c1f4</td><td>Efficiently detecting outlying behavior in +<br/>video-game players +<br/><b>Interdisciplinary Program in Visual Information Processing, Korea University, Seoul, Korea</b><br/><b>School of Games, Hongik University, Seoul, Korea</b><br/><b>Korea University</b><br/>Seoul, Korea +<br/>4 AI Lab, NCSOFT, Seongnam, Korea +</td><td>('7652095', 'Young Bin Kim', 'young bin kim')<br/>('40267433', 'Shin Jin Kang', 'shin jin kang')<br/>('4972813', 'Sang Hyeok Lee', 'sang hyeok lee')<br/>('5702793', 'Jang Young Jung', 'jang young jung')<br/>('3000093', 'Hyeong Ryeol Kam', 'hyeong ryeol kam')<br/>('2013790', 'Jung Lee', 'jung lee')<br/>('2467280', 'Young Sun Kim', 'young sun kim')<br/>('3103240', 'Joonsoo Lee', 'joonsoo lee')<br/>('22232963', 'Chang Hun Kim', 'chang hun kim')</td><td></td></tr><tr><td>c0ca6b992cbe46ea3003f4e9b48f4ef57e5fb774</td><td>A Two-Layer Representation For Large-Scale Action Recognition +<br/><b>Institute of Image Communication and Network Engineering, Shanghai Jiao Tong University</b><br/>2Shanghai Key Lab of Digital Media Processing and Transmission, 3Microsoft Research Asia +<br/><b>University of California, San Diego</b></td><td>('1701941', 'Jun Zhu', 'jun zhu')<br/>('2450889', 'Baoyuan Wang', 'baoyuan wang')<br/>('1795291', 'Xiaokang Yang', 'xiaokang yang')<br/>('38790729', 'Wenjun Zhang', 'wenjun zhang')<br/>('1736745', 'Zhuowen Tu', 'zhuowen tu')</td><td>{zhujun.sjtu,zhuowen.tu}@gmail.com, baoyuanw@microsoft.com, {xkyang,zhangwenjun}@sjtu.edu.cn +</td></tr><tr><td>c00df53bd46f78ae925c5768d46080159d4ef87d</td><td>Learning Bag-of-Features Pooling for Deep Convolutional Neural Networks +<br/><b>Aristotle University of Thessaloniki</b><br/>Thessaloniki, Greece +</td><td>('3200630', 'Nikolaos Passalis', 'nikolaos passalis')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')</td><td>passalis@csd.auth.gr, tefas@aiia.csd.auth.gr +</td></tr><tr><td>c0d5c3aab87d6e8dd3241db1d931470c15b9e39d</td><td></td><td></td><td></td></tr><tr><td>c05441dd1bc418fb912a6fafa84c0659a6850bf0</td><td>Received on 16th July 2014 +<br/>Revised on 11th September 2014 +<br/>Accepted on 23rd September 2014 +<br/>doi: 10.1049/iet-cvi.2014.0200 +<br/>www.ietdl.org +<br/>ISSN 1751-9632 +<br/>Face recognition under varying illumination based on +<br/>adaptive homomorphic eight local directional patterns +<br/><b>Utah State University, Logan, UT 84322-4205, USA</b></td><td>('2147212', 'Mohammad Reza Faraji', 'mohammad reza faraji')<br/>('1725739', 'Xiaojun Qi', 'xiaojun qi')</td><td>E-mail: Mohammadreza.Faraji@aggiemail.usu.edu +</td></tr><tr><td>eee8a37a12506ff5df72c402ccc3d59216321346</td><td>Uredniki: +<br/>dr. Tomaž Erjavec +<br/>Odsek za tehnologije znanja +<br/>Institut »Jožef Stefan«, Ljubljana +<br/>dr. Jerneja Žganec Gros +<br/>Alpineon d.o.o, Ljubljana +<br/>Založnik: Institut »Jožef Stefan«, Ljubljana +<br/>Tisk: Birografika BORI d.o.o. +<br/>Priprava zbornika: Mitja Lasič +<br/>Oblikovanje naslovnice: dr. Damjan Demšar +<br/>Tiskano iz predloga avtorjev +<br/>Naklada: 50 +<br/>Ljubljana, oktober 2008 +<br/>Konferenco IS 2008 sofinancirata +<br/>Ministrstvo za visoko šolstvo, znanost in tehnologijo +<br/>Institut »Jožef Stefan« +<br/>ISSN 1581-9973 +<br/>CIP - Kataložni zapis o publikaciji +<br/>Narodna in univerzitetna knjižnica, Ljubljana +<br/>004.934(082) +<br/>81'25:004.6(082) +<br/>004.8(063) +<br/>oktober 2008, Ljubljana, Slovenia : zbornik 11. mednarodne +<br/>Proceedings of the Sixth Language Technologies Conference, October +<br/>16th-17th, 2008 : proceedings of the 11th International +<br/>Multiconference Information Society - IS 2008, volume C / uredila, +<br/>edited by Tomaž Erjavec, Jerneja Žganec Gros. - Ljubljana : +<br/>1581-9973) +<br/>ISBN 978-961-264-006-4 +<br/>družba 4. Information society 5. Erjavec, Tomaž, 1960- 6. +<br/>Ljubljana) +<br/>241520896 +</td><td></td><td></td></tr><tr><td>ee6b503ab512a293e3088fdd7a1c893a77902acb</td><td>Automatic Name-Face Alignment to Enable Cross-Media News Retrieval +<br/>*School of Computer Science, Shanghai Key Laboratory of Intelligent Information Processing, +<br/><b>The University of North Carolina at Charlotte, USA</b><br/><b>Fudan University, Shanghai, China</b></td><td>('7550713', 'Yuejie Zhang', 'yuejie zhang')<br/>('1721131', 'Wei Wu', 'wei wu')<br/>('1678662', 'Yang Li', 'yang li')<br/>('1751513', 'Cheng Jin', 'cheng jin')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')<br/>('2344620', 'Jianping Fan', 'jianping fan')</td><td>*{yjzhang, 10210240122, 11210240052, jc, xyxue}@fudan.edu.cn, +jfan@uncc.edu +</td></tr><tr><td>ee18e29a2b998eddb7f6663bb07891bfc7262248</td><td>1119 +<br/>Local Linear Discriminant Analysis Framework +<br/>Using Sample Neighbors +</td><td>('38162192', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>eeb6d084f9906c53ec8da8c34583105ab5ab8284</td><td>12 +<br/>Generation of Facial Expression Map using +<br/>Supervised and Unsupervised Learning +<br/><b>Akita Prefectural University</b><br/><b>Akita University</b><br/>Japan +<br/>1. Introduction +<br/>Recently, studies of human face recognition have been conducted vigorously (Fasel & +<br/>Luettin, 2003; Yang et al., 2002; Pantic & Rothkrantz, 2000a; Zhao et al., 2000; Hasegawa et +<br/>al., 1997; Akamatsu, 1997). Such studies are aimed at the implementation of an intelligent +<br/>man-machine interface. Especially, studies of facial expression recognition for human- +<br/>machine emotional communication are attracting attention (Fasel & Luettin, 2003; Pantic & +<br/>Rothkrantz, 2000a; Tian et al., 2001; Pantic & Rothkrantz, 2000b; Lyons et al., 1999; Lyons et +<br/>al., 1998; Zhang et al., 1998). +<br/>The shape (static diversity) and motion (dynamic diversity) of facial components such as the +<br/>eyebrows, eyes, nose, and mouth manifest expressions. Considering facial expressions from +<br/>the perspective of static diversity because facial configurations differ among people, it is +<br/>presumed that a facial expression pattern appearing on a face when facial expression is +<br/>manifested includes person-specific features. In addition, from the viewpoint of dynamic +<br/>diversity, because the dynamic change of facial expression originates in a person-specific +<br/>facial expression pattern, it is presumed that the displacement vector of facial components +<br/>has person-specific features. The properties of the human face described above reveal the +<br/>following tasks. +<br/>The first task is to generalize a facial expression recognition model. Numerous conventional +<br/>approaches have attempted generalization of a facial expression recognition model. They +<br/>use the distance of motion of feature points set on a face and the motion vectors of facial +<br/>muscle movements in its arbitrary regions as feature values. Typically, such methods assign +<br/>that information to so-called Action Units (AUs) of a Facial Action Coding System (FACS) +<br/>(Ekman & Friesen, 1978). In fact, AUs are described qualitatively. Therefore, no objective +<br/>criteria pertain to the setting positions of feature points and regions. They all depend on a +<br/>particular researcher’s experience. However, features representing facial expressions are +<br/>presumed to differ among subjects. Accordingly, a huge effort is necessary to link +<br/>quantitative features with qualitative AUs for each subject and to derive universal features +<br/>therefrom. It is also suspected that a generalized facial expression recognition model that is +<br/>applicable to all subjects would disregard person-specific features of facial expressions that are +<br/>borne originally by each subject. For all the reasons described above, it is an important task to +<br/>establish a method to extract person-specific features using a common approach to every +<br/>subject, and to build a facial expression recognition model that incorporates these features. +<br/>Source: Machine Learning, Book edited by: Abdelhamid Mellouk and Abdennacer Chebira, +<br/> ISBN 978-3-902613-56-1, pp. 450, February 2009, I-Tech, Vienna, Austria +<br/>www.intechopen.com +</td><td>('1932760', 'Masaki Ishii', 'masaki ishii')<br/>('2052920', 'Kazuhito Sato', 'kazuhito sato')<br/>('1738333', 'Hirokazu Madokoro', 'hirokazu madokoro')<br/>('21063785', 'Makoto Nishida', 'makoto nishida')</td><td></td></tr><tr><td>ee815f60dc4a090fa9fcfba0135f4707af21420d</td><td>EAC-Net: A Region-based Deep Enhancing and Cropping Approach for +<br/>Facial Action Unit Detection +<br/><b>Grove School of Engineering, CUNY City College, NY, USA</b><br/>2 Department of Computer Science, CUNY Graduate Center, NY, USA +<br/><b>Engineering and Applied Science, SUNY Binghamton University, NY, USA</b></td><td>('48625314', 'Wei Li', 'wei li')</td><td></td></tr><tr><td>eed7920682789a9afd0de4efd726cd9a706940c8</td><td>Computers to Help with Conversations: +<br/>Affective Framework to Enhance Human Nonverbal Skills +<br/>by +<br/>Mohammed Ehsan Hoque +<br/><b>B.S., Pennsylvania State University</b><br/><b>M.S., University of Memphis</b><br/>Submitted to the Program in Media Arts and Sciences, +<br/>School of Architecture and Planning, +<br/>In partial fulfilment of the requirements for the degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>at the +<br/><b>MASSACHUSETTS INSTITUTE OF TECHNOLOGY</b><br/>September 2013 +<br/><b>Massachusetts Institute of Technology 2013. All rights reserved</b><br/>Author +<br/>Certified by +<br/>Accepted by +<br/> Program in Media Arts and Sciences +<br/>August 15, 2013 +<br/> Rosalind W. Picard +<br/> Professor of Media Arts and Sciences +<br/> Program in Media Arts and Sciences, MIT +<br/> Thesis supervisor +<br/>Pattie Maes +<br/>Associate Academic Head +<br/>Program in Media Arts and Sciences, MIT +</td><td></td><td></td></tr><tr><td>ee7093e91466b81d13f4d6933bcee48e4ee63a16</td><td>Discovering Person Identity via +<br/>Large-Scale Observations +<br/><b>Interactive and Digital Media Institute, National University of Singapore, SG</b><br/><b>School of Computing, National University of Singapore, SG</b></td><td>('3026404', 'Yongkang Wong', 'yongkang wong')<br/>('1986874', 'Lekha Chaisorn', 'lekha chaisorn')<br/>('1744045', 'Mohan S. Kankanhalli', 'mohan s. kankanhalli')</td><td></td></tr><tr><td>ee461d060da58d6053d2f4988b54eff8655ecede</td><td></td><td></td><td></td></tr><tr><td>eefb8768f60c17d76fe156b55b8a00555eb40f4d</td><td>Subspace Scores for Feature Selection in Computer Vision +</td><td>('2032038', 'Cameron Musco', 'cameron musco')<br/>('2767340', 'Christopher Musco', 'christopher musco')</td><td>cnmusco@mit.edu +<br/>cpmusco@mit.edu +</td></tr><tr><td>ee463f1f72a7e007bae274d2d42cd2e5d817e751</td><td>Automatically Extracting Qualia Relations for the Rich Event Ontology +<br/><b>University of Colorado Boulder, 2U.S. Army Research Lab</b></td><td>('51203051', 'Ghazaleh Kazeminejad', 'ghazaleh kazeminejad')<br/>('3202888', 'Claire Bonial', 'claire bonial')<br/>('1783500', 'Susan Windisch Brown', 'susan windisch brown')<br/>('1728285', 'Martha Palmer', 'martha palmer')</td><td>{ghazaleh.kazeminejad, susan.brown, martha.palmer}@colorado.edu +<br/>claire.n.bonial.civ@mail.mil +</td></tr><tr><td>eed1dd2a5959647896e73d129272cb7c3a2e145c</td><td></td><td></td><td></td></tr><tr><td>ee92d36d72075048a7c8b2af5cc1720c7bace6dd</td><td>FACE RECOGNITION USING MIXTURES OF PRINCIPAL COMPONENTS +<br/>Video and Display Processing +<br/>Philips Research USA +<br/>Briarcliff Manor, NY 10510 +</td><td>('1727257', 'Deepak S. Turaga', 'deepak s. turaga')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td>deepak.turaga@philips.com +</td></tr><tr><td>ee418372b0038bd3b8ae82bd1518d5c01a33a7ec</td><td>CSE 255 Winter 2015 Assignment 1: Eye Detection using Histogram +<br/>of Oriented Gradients and Adaboost Classifier +<br/>Electrical and Computer Engineering Department +<br/><b>University of California, San Diego</b></td><td>('2812409', 'Kevan Yuen', 'kevan yuen')</td><td>kcyuen@eng.ucsd.edu +</td></tr><tr><td>eee06d68497be8bf3a8aba4fde42a13aa090b301</td><td>CR-GAN: Learning Complete Representations for Multi-view Generation +<br/><b>Rutgers University</b><br/><b>University of North Carolina at Charlotte</b></td><td>('6812347', 'Yu Tian', 'yu tian')<br/>('4340744', 'Xi Peng', 'xi peng')<br/>('33860220', 'Long Zhao', 'long zhao')<br/>('1753384', 'Shaoting Zhang', 'shaoting zhang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>{yt219, px13, lz311, dnm}@cs.rutgers.edu, szhang16@uncc.edu +</td></tr><tr><td>eee2d2ac461f46734c8e674ae14ed87bbc8d45c6</td><td>Generalized Rank Pooling for Activity Recognition +<br/>1Australian Centre for Robotic Vision, 2Data61/CSIRO +<br/><b>The Australian National University, Canberra, Australia</b></td><td>('2691929', 'Anoop Cherian', 'anoop cherian')<br/>('1688071', 'Basura Fernando', 'basura fernando')<br/>('23911916', 'Mehrtash Harandi', 'mehrtash harandi')<br/>('49384847', 'Stephen Gould', 'stephen gould')</td><td>firstname.lastname@{anu.edu.au, data61.csiro.au} +</td></tr><tr><td>eed93d2e16b55142b3260d268c9e72099c53d5bc</td><td>ICFVR 2017: 3rd International Competition on Finger Vein Recognition +<br/><b>Chittagong University of Engineering and Technology</b><br/>∗ These authors contributed equally to this work +<br/><b>Peking University</b><br/>2Shenzhen Maidi Technology Co., LTD. +<br/>3TigerIT +</td><td>('46867002', 'Yi Zhang', 'yi zhang')<br/>('2560109', 'Houjun Huang', 'houjun huang')<br/>('38728899', 'Haifeng Zhang', 'haifeng zhang')<br/>('3142600', 'Liao Ni', 'liao ni')<br/>('47210488', 'Wei Xu', 'wei xu')<br/>('1694788', 'Nasir Uddin Ahmed', 'nasir uddin ahmed')<br/>('9336364', 'Md. Shakil Ahmed', 'md. shakil ahmed')<br/>('9372198', 'Yilun Jin', 'yilun jin')<br/>('23100665', 'Yingjie Chen', 'yingjie chen')<br/>('35273470', 'Jingxuan Wen', 'jingxuan wen')<br/>('39201759', 'Wenxin Li', 'wenxin li')</td><td></td></tr><tr><td>eedfb384a5e42511013b33104f4cd3149432bd9e</td><td>Multimodal Probabilistic Person +<br/>Tracking and Identification +<br/>in Smart Spaces +<br/>zur Erlangung des akademischen Grades eines +<br/>Doktors der Ingenieurwissenschaften +<br/>der Fakultät für Informatik +<br/>der Universität Fridericiana zu Karlsruhe (TH) +<br/>genehmigte +<br/>Dissertation +<br/>von +<br/>aus Karlsruhe +<br/>Tag der mündlichen Prüfung: 20.11.2009 +<br/>Erster Gutachter: +<br/>Zweiter Gutachter: +<br/>Prof. Dr. A. Waibel +<br/>Prof. Dr. R. Stiefelhagen +</td><td>('1701229', 'Keni Bernardin', 'keni bernardin')</td><td></td></tr><tr><td>c94b3a05f6f41d015d524169972ae8fd52871b67</td><td>The Fastest Deformable Part Model for Object Detection +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences, China</b></td><td>('1721677', 'Junjie Yan', 'junjie yan')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('39774417', 'Longyin Wen', 'longyin wen')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jjyan,zlei,lywen,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>c9424d64b12a4abe0af201e7b641409e182babab</td><td>Article +<br/>Which, When, and How: Hierarchical Clustering with +<br/>Human–Machine Cooperation +<br/>Academic Editor: Tom Burr +<br/>Received: 3 November 2016; Accepted: 14 December 2016; Published: 21 December 2016 +</td><td>('1751849', 'Huanyang Zheng', 'huanyang zheng')<br/>('1703691', 'Jie Wu', 'jie wu')</td><td>Computer and Information Sciences, Temple University, PA 19121, USA; jiewu@temple.edu +<br/>* Correspondence: huanyang.zheng@temple.edu; Tel.: +1-215-204-8450 +</td></tr><tr><td>c91103e6612fa7e664ccbc3ed1b0b5deac865b02</td><td>Automatic facial expression recognition using +<br/>statistical-like moments +<br/><b>Integrated Research Center, Universit`a Campus Bio-Medico di Roma</b><br/>Via Alvaro del Portillo, 00128 Roma, Italy +</td><td>('1679260', 'Giulio Iannello', 'giulio iannello')<br/>('1720099', 'Paolo Soda', 'paolo soda')</td><td>{r.dambrosio, g.iannello, p.soda}@unicampus.it +</td></tr><tr><td>c903af0d69edacf8d1bff3bfd85b9470f6c4c243</td><td></td><td></td><td></td></tr><tr><td>c97a5f2241cc6cd99ef0c4527ea507a50841f60b</td><td>Person Search in Videos with One Portrait +<br/>Through Visual and Temporal Links +<br/><b>CUHK-SenseTime Joint Lab, The Chinese University of Hong Kong</b><br/><b>Tsinghua University</b><br/>3 SenseTime Research +</td><td>('39360892', 'Qingqiu Huang', 'qingqiu huang')<br/>('40584026', 'Wentao Liu', 'wentao liu')<br/>('1807606', 'Dahua Lin', 'dahua lin')</td><td>{hq016,dhlin}@ie.cuhk.edu.hk +<br/>liuwtwinter@gmail.com +</td></tr><tr><td>c95cd36779fcbe45e3831ffcd3314e19c85defc5</td><td>FACE RECOGNITION USING MULTI-MODAL LOW-RANK DICTIONARY LEARNING +<br/><b>University of Alberta, Edmonton, Canada</b></td><td>('1807674', 'Homa Foroughi', 'homa foroughi')<br/>('2627414', 'Moein Shakeri', 'moein shakeri')<br/>('1772846', 'Nilanjan Ray', 'nilanjan ray')<br/>('1734058', 'Hong Zhang', 'hong zhang')</td><td></td></tr><tr><td>c9e955cb9709f16faeb0c840f4dae92eb875450a</td><td>Proposal of Novel Histogram Features +<br/>for Face Detection +<br/><b>Harbin Institute of Technology, School of Computer Science and Technology</b><br/>P.O.Box 1071, Harbin, Heilongjiang 150001, China +<br/><b>Heilongjiang University, College of Computer Science and Technology, China</b></td><td>('2607285', 'Haijing Wang', 'haijing wang')<br/>('40426020', 'Peihua Li', 'peihua li')<br/>('1821107', 'Tianwen Zhang', 'tianwen zhang')</td><td>ninhaijing@yahoo.com +<br/>peihualj@hotmail.com +</td></tr><tr><td>c92bb26238f6e30196b0c4a737d8847e61cfb7d4</td><td>BEYOND CONTEXT: EXPLORING SEMANTIC SIMILARITY FOR TINY FACE DETECTION +<br/><b>School of Computer Science, Northwestern Polytechnical University, P.R.China</b><br/><b>Global Big Data Technologies Centre (GBDTC), University of Technology Sydney, Australia</b><br/><b>School of Data and Computer Science, Sun Yat-sen University, P.R.China</b></td><td>('24336288', 'Yue Xi', 'yue xi')<br/>('3104013', 'Jiangbin Zheng', 'jiangbin zheng')<br/>('1714410', 'Wenjing Jia', 'wenjing jia')<br/>('3031842', 'Hanhui Li', 'hanhui li')</td><td></td></tr><tr><td>c9bbd7828437e70cc3e6863b278aa56a7d545150</td><td>Unconstrained Fashion Landmark Detection via +<br/>Hierarchical Recurrent Transformer Networks +<br/><b>The Chinese University of Hong Kong</b><br/>2SenseTime Group Limited +</td><td>('1979911', 'Sijie Yan', 'sijie yan')<br/>('3243969', 'Ziwei Liu', 'ziwei liu')<br/>('47571885', 'Ping Luo', 'ping luo')<br/>('1725421', 'Shi Qiu', 'shi qiu')</td><td>{ys016,lz013,pluo,xtang}@ie.cuhk.edu.hk,sqiu@sensetime.com,xgwang@ee.cuhk.edu.hk +</td></tr><tr><td>c9f588d295437009994ddaabb64fd4e4c499b294</td><td>Predicting Professions through +<br/>Probabilistic Model under Social Context +<br/><b>Northeastern University</b><br/>Boston, MA, 02115 +</td><td>('2025056', 'Ming Shao', 'ming shao')<br/>('2897748', 'Liangyue Li', 'liangyue li')<br/>('1708679', 'Yun Fu', 'yun fu')</td><td>mingshao@ccs.neu.edu, {liangyue, yunfu}@ece.neu.edu +</td></tr><tr><td>c92da368a6a886211dc759fe7b1b777a64d8b682</td><td>International Journal of Science and Advanced Technology (ISSN 2221-8386) Volume 1 No 2 April 2011 +<br/>http://www.ijsat.com +<br/>Face Recognition System based on +<br/>Face Pose Estimation and +<br/>Frontal Face Pose Synthesis +<br/>Department of Electrical Engineering +<br/><b>National Chiao-Tung University</b><br/>Hsinchu, Taiwan, R.O.C +<br/>Department of Electrical Engineering +<br/><b>National Chiao-Tung University</b><br/>Hsinchu, Taiwan, R.O.C +</td><td>('4525043', 'Kuo-Yu Chiu', 'kuo-yu chiu')<br/>('1707677', 'Sheng-Fuu Lin', 'sheng-fuu lin')</td><td>Alvin_cgr@hotmail.com +</td></tr><tr><td>c98983592777952d1751103b4d397d3ace00852d</td><td>Face Synthesis from Facial Identity Features +<br/>Google Research +<br/>Google Research +<br/><b>University of Massachusetts Amherst</b><br/>Google Research +<br/>Google Research +<br/>CSAIL, MIT and Google Research +</td><td>('39578349', 'Forrester Cole', 'forrester cole')<br/>('8707513', 'Aaron Sarna', 'aaron sarna')<br/>('2636941', 'David Belanger', 'david belanger')<br/>('1707347', 'Dilip Krishnan', 'dilip krishnan')<br/>('2138834', 'Inbar Mosseri', 'inbar mosseri')<br/>('1768236', 'William T. Freeman', 'william t. freeman')</td><td>fcole@google.com +<br/>sarna@google.com +<br/>belanger@cs.umass.edu +<br/>dilipkay@google.com +<br/>inbarm@google.com +<br/>wfreeman@google.com +</td></tr><tr><td>c9367ed83156d4d682cefc59301b67f5460013e0</td><td>Geometry-Contrastive GAN for Facial Expression +<br/>Transfer +<br/><b>Institute of Software, Chinese Academy of Sciences</b></td><td>('35790820', 'Fengchun Qiao', 'fengchun qiao')<br/>('35996065', 'Zirui Jiao', 'zirui jiao')<br/>('3238696', 'Zhihao Li', 'zhihao li')<br/>('1804472', 'Hui Chen', 'hui chen')<br/>('7643981', 'Hongan Wang', 'hongan wang')</td><td></td></tr><tr><td>fc1e37fb16006b62848def92a51434fc74a2431a</td><td>DRAFT +<br/>A Comprehensive Analysis of Deep Regression +</td><td>('2793152', 'Pablo Mesejo', 'pablo mesejo')<br/>('1780201', 'Xavier Alameda-Pineda', 'xavier alameda-pineda')<br/>('1794229', 'Radu Horaud', 'radu horaud')</td><td></td></tr><tr><td>fc5bdb98ff97581d7c1e5eb2d24d3f10714aa192</td><td>Initialization Strategies of Spatio-Temporal +<br/>Convolutional Neural Networks +<br/><b>University of Toronto</b></td><td>('2711409', 'Elman Mansimov', 'elman mansimov')<br/>('2897313', 'Nitish Srivastava', 'nitish srivastava')<br/>('1776908', 'Ruslan Salakhutdinov', 'ruslan salakhutdinov')</td><td></td></tr><tr><td>fc20149dfdff5fdf020647b57e8a09c06e11434b</td><td>Submitted 8/06; Revised 1/07; Published 5/07 +<br/>Local Discriminant Wavelet Packet Coordinates for Face Recognition +<br/>Center for Computer Vision and Department of Mathematics +<br/><b>Sun Yat-Sen (Zhongshan) University</b><br/>Guangzhou, 510275 China +<br/>Department of Electric Engineering +<br/><b>City University of Hong Kong</b><br/>83 Tat Chee Avenue +<br/>Kowloon, Hong Kong, China +<br/>Editor: Donald Geman +</td><td>('5692650', 'Chao-Chun Liu', 'chao-chun liu')<br/>('1726138', 'Dao-Qing Dai', 'dao-qing dai')<br/>('1718530', 'Hong Yan', 'hong yan')</td><td>STSDDQ@MAIL.SYSU.EDU.CN +<br/>H.YAN@CITYU.EDU.HK +</td></tr><tr><td>fc516a492cf09aaf1d319c8ff112c77cfb55a0e5</td><td></td><td></td><td></td></tr><tr><td>fc0f5859a111fb17e6dcf6ba63dd7b751721ca61</td><td>Design of an Automatic +<br/>Facial Expression Detector +<br/>An essay presented for the degree +<br/>of +<br/>M.Math +<br/>Applied Mathematics +<br/><b>University of Waterloo</b><br/>2018/01/26 +</td><td>('2662893', 'Jian Liang', 'jian liang')</td><td></td></tr><tr><td>fcbec158e6a4ace3d4311b26195482b8388f0ee9</td><td>Face Recognition from Still Images and Videos +<br/>Center for Automation Research (CfAR) and +<br/>Department of Electrical and Computer Engineering +<br/><b>University of Maryland, College Park, MD</b><br/>I. INTRODUCTION +<br/>In most situations, identifying humans using faces is an effortless task for humans. Is this true for computers? +<br/>This very question defines the field of automatic face recognition [7], [31], [62], one of the most active research +<br/>areas in computer vision, pattern recognition, and image understanding. +<br/>Over the past decade, the problem of face recognition has attracted substantial attention from various disciplines +<br/>and has witnessed a skyrocketing growth of the literature. Below, we mainly emphasize some key perspectives of +<br/>the face recognition problem. +<br/>A. Biometric perspective +<br/>Face is a biometric. As a consequence, face recognition finds wide applications in authentication, security, and +<br/>so on. One recent application is the US-VISIT system by the Department of Homeland Security (DHS), collecting +<br/>foreign passengers’ fingerprints and face images. +<br/>Biometric signatures of a person characterize the physiological or behavioral characteristics. Physiological bio- +<br/>metrics are innate or naturally occuring, while behavioral biometrics arise from mannerisms or traits that are learned +<br/>or acquired. Table I lists commonly used biometrics. Biometric technologies provide the foundation for an extensive +<br/>array of highly secure identification and personal verification solutions. Compared to conventional identification and +<br/>verification methods based on personal identification numbers (PINs) or passwords, biometric technologies offer +<br/>many advantages. First, biometrics are individualized traits while passwords may be used or stolen by someone +<br/>other than the authorized user. Also, biometric is very convenient since there is nothing to carry or remember. In +<br/>addition, biometric technologies are becoming more accurate and less expensive. +<br/>Among all biometrics listed in Table I, the face is a very unique one because it is the only biometric belonging +<br/>to both physiological and behavioral categories. While the physiological part of the face has been widely exploited +<br/>Partially supported by NSF ITR Grant 03-25119. Zhou is now with Integrated Data Systems Department, Siemens Corporate Research, +<br/>November 5, 2004 +<br/>DRAFT +</td><td>('1682187', 'Shaohua Kevin Zhou', 'shaohua kevin zhou')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>Email: {shaohua, rama}@cfar.umd.edu +<br/>Princeton, NJ 08540. His current email address is kzhou@scr.siemens.com. +</td></tr><tr><td>fcd3d69b418d56ae6800a421c8b89ef363418665</td><td>Effects of Aging over Facial Feature Analysis and Face +<br/>Recognition +<br/>Bogaziçi Un. Electronics Eng. Dept. March 2010 +</td><td>('3398552', 'Bilgin Esme', 'bilgin esme')</td><td></td></tr><tr><td>fcd77f3ca6b40aad6edbd1dab9681d201f85f365</td><td>c(cid:13)Copyright 2014 +</td><td>('3299424', 'Miro Enev', 'miro enev')</td><td></td></tr><tr><td>fcf91995dc4d9b0cee84bda5b5b0ce5b757740ac</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Asymmetric Discrete Graph Hashing +<br/><b>University of Florida, Gainesville, FL, 32611, USA</b></td><td>('2766473', 'Xiaoshuang Shi', 'xiaoshuang shi')<br/>('2082604', 'Fuyong Xing', 'fuyong xing')<br/>('46321210', 'Kaidi Xu', 'kaidi xu')<br/>('2599018', 'Manish Sapkota', 'manish sapkota')<br/>('49576071', 'Lin Yang', 'lin yang')</td><td>xsshi2015@ufl.edu +</td></tr><tr><td>fc798314994bf94d1cde8d615ba4d5e61b6268b6</td><td>Face Recognition: face in video, age invariance, +<br/>and facial marks +<br/>By +<br/>A DISSERTATION +<br/>Submitted to +<br/><b>Michigan State University</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>Computer Science +<br/>2009 +</td><td>('2222919', 'Unsang Park', 'unsang park')</td><td></td></tr><tr><td>fc23a386c2189f221b25dbd0bb34fcd26ccf60fa</td><td>A Discriminative Latent Model of Object +<br/>Classes and Attributes +<br/><b>School of Computing Science, Simon Fraser University, Canada</b></td><td>('40457160', 'Yang Wang', 'yang wang')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td>{ywang12,mori}@cs.sfu.ca +</td></tr><tr><td>fc68c5a3ab80d2d31e6fd4865a7ff2b4ab66ca9f</td><td>This is a preprint of the paper presented at the 11th International Conference Beyond Databases, Architectures and +<br/>Structures (BDAS 2015), May 26-29 2015 in Ustroń, Poland and published in the Communications in Computer and +<br/>Information Science Volume 521, 2015, pp 585-597. DOI: 10.1007/978-3-319-18422-7_52 +<br/>Evaluation Criteria for Affect-Annotated Databases +<br/><b>Faculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology, Poland</b></td><td>('2414357', 'Agnieszka Landowska', 'agnieszka landowska')<br/>('3271448', 'Mariusz Szwoch', 'mariusz szwoch')<br/>('3175073', 'Wioleta Szwoch', 'wioleta szwoch')</td><td>szwoch@eti.pg.gda.pl +</td></tr><tr><td>fc2bad3544c7c8dc7cd182f54888baf99ed75e53</td><td>Efficient Retrieval for Large Scale Metric +<br/>Learning +<br/><b>Institute for Computer Graphics and Vision</b><br/><b>Graz University of Technology, Austria</b></td><td>('1791182', 'Peter M. Roth', 'peter m. roth')<br/>('3628150', 'Horst Bischof', 'horst bischof')</td><td>{koestinger,pmroth,bischof}@icg.tugraz.at +</td></tr><tr><td>fcf8bb1bf2b7e3f71fb337ca3fcf3d9cf18daa46</td><td>MANUSCRIPT SUBMITTED TO IEEE TRANS. PATTERN ANAL. MACH. INTELL., JULY 2010 +<br/>Feature Selection via Sparse Approximation for +<br/>Face Recognition +</td><td>('1944073', 'Yixiong Liang', 'yixiong liang')<br/>('31685288', 'Lei Wang', 'lei wang')<br/>('2090968', 'Yao Xiang', 'yao xiang')<br/>('6609276', 'Beiji Zou', 'beiji zou')</td><td></td></tr><tr><td>fcbf808bdf140442cddf0710defb2766c2d25c30</td><td>IJCV manuscript No. +<br/>(will be inserted by the editor) +<br/>Unsupervised Semantic Action Discovery from Video +<br/>Collections +<br/>Received: date / Accepted: date +</td><td>('3114252', 'Ozan Sener', 'ozan sener')<br/>('1681995', 'Ashutosh Saxena', 'ashutosh saxena')</td><td></td></tr><tr><td>fdff2da5bdca66e0ab5874ef58ac2205fb088ed7</td><td>Continuous Supervised Descent Method for +<br/>Facial Landmark Localisation +<br/>1Universitat Oberta de Catalunya, 156 Rambla del Poblenou, Barcelona, Spain +<br/>2Universitat de Barcelona, 585 Gran Via de les Corts Catalanes, Barcelona, Spain +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, USA</b><br/>4Computer Vision Center, O Building, UAB Campus, Bellaterra, Spain +<br/><b>University of Pittsburgh, Pittsburgh, PA, USA</b></td><td>('3305641', 'Marc Oliu', 'marc oliu')<br/>('1733113', 'Takeo Kanade', 'takeo kanade')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')</td><td></td></tr><tr><td>fdfd57d4721174eba288e501c0c120ad076cdca8</td><td>An Analysis of Action Recognition Datasets for +<br/>Language and Vision Tasks +<br/><b>Institute for Language, Cognition and Computation</b><br/><b>School of Informatics, University of Edinburgh</b><br/>10 Crichton Street, Edinburgh EH8 9AB +</td><td>('2921001', 'Spandana Gella', 'spandana gella')<br/>('48716849', 'Frank Keller', 'frank keller')</td><td>S.Gella@sms.ed.ac.uk, keller@inf.ed.ac.uk +</td></tr><tr><td>fd4ac1da699885f71970588f84316589b7d8317b</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 6, NO. 1, JANUARY 2007 +<br/>Supervised Descent Method +<br/>for Solving Nonlinear Least Squares +<br/>Problems in Computer Vision +</td><td>('3182065', 'Xuehan Xiong', 'xuehan xiong')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td></td></tr><tr><td>fd33df02f970055d74fbe69b05d1a7a1b9b2219b</td><td>Single Shot Temporal Action Detection +<br/><b>Shanghai Jiao Tong University, China. 2Columbia University, USA</b><br/><b>Cooperative Medianet Innovation Center (CMIC), Shanghai Jiao Tong University, China</b></td><td>('6873935', 'Tianwei Lin', 'tianwei lin')<br/>('1758267', 'Xu Zhao', 'xu zhao')<br/>('2195345', 'Zheng Shou', 'zheng shou')</td><td>{wzmsltw,zhaoxu}@sjtu.edu.cn,zs2262@columbia.edu +</td></tr><tr><td>fdf533eeb1306ba418b09210387833bdf27bb756</td><td>951 +</td><td></td><td></td></tr><tr><td>fdda5852f2cffc871fd40b0cb1aa14cea54cd7e3</td><td>Im2Flow: Motion Hallucination from Static Images for Action Recognition +<br/>UT Austin +<br/>UT Austin +<br/>UT Austin +</td><td>('3387849', 'Ruohan Gao', 'ruohan gao')<br/>('50398746', 'Bo Xiong', 'bo xiong')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>rhgao@cs.utexas.edu +<br/>bxiong@cs.utexas.edu +<br/>grauman@cs.utexas.edu +</td></tr><tr><td>fdfaf46910012c7cdf72bba12e802a318b5bef5a</td><td>Computerized Face Recognition in Renaissance +<br/>Portrait Art +</td><td>('18640672', 'Ramya Srinivasan', 'ramya srinivasan')<br/>('3007257', 'Conrad Rudolph', 'conrad rudolph')<br/>('1688416', 'Amit Roy-Chowdhury', 'amit roy-chowdhury')</td><td></td></tr><tr><td>fd15e397629e0241642329fc8ee0b8cd6c6ac807</td><td>Semi-Supervised Clustering with Neural Networks +<br/>IIIT-Delhi, India +</td><td>('2200208', 'Ankita Shukla', 'ankita shukla')<br/>('39866663', 'Gullal Singh Cheema', 'gullal singh cheema')<br/>('34817359', 'Saket Anand', 'saket anand')</td><td>{ankitas, gullal1408, anands}@iiitd.ac.in +</td></tr><tr><td>fde41dc4ec6ac6474194b99e05b43dd6a6c4f06f</td><td>Multi-Expert Gender Classification on Age Group by Integrating Deep Neural +<br/>Networks +<br/><b>Yonsei University</b><br/>50 Yonsei-ro, Seodaemun-gu, Seoul 03722, Republic of Korea. +</td><td>('51430701', 'Jun Beom Kho', 'jun beom kho')</td><td>kojb87@hanmail.net +</td></tr><tr><td>fd9feb21b3d1fab470ff82e3f03efce6a0e67a1f</td><td><b>University of Twente</b><br/>Department of Services, Cybersecurity and Safety +<br/>Master Thesis +<br/>Deep Verification Learning +<br/>Author: +<br/>F.H.J. Hillerstr¨om +<br/>Committee: +<br/>Prof. Dr. Ir. R.N.J. Veldhuis +<br/>Dr. Ir. L.J. Spreeuwers +<br/>Dr. Ir. D. Hiemstra +<br/>December 5, 2016 +</td><td></td><td></td></tr><tr><td>fdca08416bdadda91ae977db7d503e8610dd744f</td><td> +<br/>ICT-2009.7.1 +<br/>KSERA Project +<br/>2010-248085 +<br/>Deliverable D3.1 +<br/>Deliverable D3.1 +<br/>Human Robot Interaction +<br/>Human Robot Interaction +<br/>18 October 2010 +<br/>Public Document +<br/>The KSERA project (http://www.ksera +<br/>KSERA project (http://www.ksera-project.eu) has received funding from the European Commission +<br/>project.eu) has received funding from the European Commission +<br/>under the 7th Framework Programme (FP7) for Research and Technological Development under grant +<br/>under the 7th Framework Programme (FP7) for Research and Technological Development under grant +<br/>under the 7th Framework Programme (FP7) for Research and Technological Development under grant +<br/>agreement n°2010-248085. +</td><td></td><td></td></tr><tr><td>fd53be2e0a9f33080a9db4b5a5e416e24ae8e198</td><td>Apparent Age Estimation Using Ensemble of Deep Learning Models +<br/>Refik Can Mallı∗ +<br/>Mehmet Ayg¨un∗ +<br/>Hazım Kemal Ekenel +<br/><b>Istanbul Technical University</b><br/>Istanbul, Turkey +</td><td></td><td>{mallir,aygunme,ekenel}@itu.edu.tr +</td></tr><tr><td>fd71ae9599e8a51d8a61e31e6faaaf4a23a17d81</td><td>Action Detection from a Robot-Car Perspective +<br/>Universit´a degli Studi Federico II +<br/>Naples, Italy +<br/><b>Oxford Brookes University</b><br/>Oxford, UK +</td><td>('39078800', 'Valentina Fontana', 'valentina fontana')<br/>('51149466', 'Manuele Di Maio', 'manuele di maio')<br/>('51152717', 'Stephen Akrigg', 'stephen akrigg')<br/>('1931660', 'Gurkirt Singh', 'gurkirt singh')<br/>('49348905', 'Suman Saha', 'suman saha')<br/>('1754181', 'Fabio Cuzzolin', 'fabio cuzzolin')</td><td>vale.fontana@studenti.unina.it, man.dimaio@gmail.com +<br/>15057204@brookes.ac.uk, gurkirt.singh-2015@brookes.ac.uk, +<br/>suman.saha-2014@brookes.ac.uk, fabio.cuzzolin@brookes.ac.uk +</td></tr><tr><td>fd96432675911a702b8a4ce857b7c8619498bf9f</td><td>Improved Face Detection and Alignment using Cascade +<br/>Deep Convolutional Network +<br/>†Beijing Key Laboratory of Intelligent Information Technology, School of +<br/><b>Computer Science, Beijing Institute of Technology, Beijing 100081, P.R.China</b><br/><b>China Mobile Research Institute, Xuanwu Men West Street, Beijing</b></td><td>('22244104', 'Weilin Cong', 'weilin cong')<br/>('2901725', 'Sanyuan Zhao', 'sanyuan zhao')<br/>('1698061', 'Hui Tian', 'hui tian')<br/>('34926055', 'Jianbing Shen', 'jianbing shen')</td><td></td></tr><tr><td>fd10b0c771a2620c0db294cfb82b80d65f73900d</td><td>Identifying The Most Informative Features Using A Structurally Interacting Elastic Net +<br/><b>Central University of Finance and Economics, Beijing, China</b><br/><b>Xiamen University, Xiamen, Fujian, China</b><br/><b>University of York, York, UK</b></td><td>('2290930', 'Lixin Cui', 'lixin cui')<br/>('1749518', 'Lu Bai', 'lu bai')<br/>('47295137', 'Zhihong Zhang', 'zhihong zhang')<br/>('49416727', 'Yue Wang', 'yue wang')<br/>('1679753', 'Edwin R. Hancock', 'edwin r. hancock')</td><td></td></tr><tr><td>fd7b6c77b46420c27725757553fcd1fb24ea29a8</td><td>MEXSVMs: Mid-level Features for Scalable Action Recognition +<br/><b>Dartmouth College</b><br/>6211 Sudikoff Lab, Hanover, NH 03755 +<br/>Dartmouth Computer Science Technical Report TR2013-726 +</td><td>('1687325', 'Du Tran', 'du tran')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')</td><td>{dutran,lorenzo}@cs.dartmouth.edu +</td></tr><tr><td>fdb33141005ca1b208a725796732ab10a9c37d75</td><td>Int.J.Appl. Math. Comput.Sci.,2016,Vol. 26,No. 2,451–465 +<br/>DOI: 10.1515/amcs-2016-0032 +<br/>A CONNECTIONIST COMPUTATIONAL METHOD FOR FACE RECOGNITION +<br/>, JOS ´E A. GIRONA-SELVA a +<br/>aDepartment of Computer Technology +<br/><b>University of Alicante, 03690, San Vicente del Raspeig, Alicante, Spain</b><br/>In this work, a modified version of the elastic bunch graph matching (EBGM) algorithm for face recognition is introduced. +<br/>First, faces are detected by using a fuzzy skin detector based on the RGB color space. Then, the fiducial points for the facial +<br/>graph are extracted automatically by adjusting a grid of points to the result of an edge detector. After that, the position of +<br/>the nodes, their relation with their neighbors and their Gabor jets are calculated in order to obtain the feature vector defining +<br/>each face. A self-organizing map (SOM) framework is shown afterwards. Thus, the calculation of the winning neuron and +<br/>the recognition process are performed by using a similarity function that takes into account both the geometric and texture +<br/>information of the facial graph. The set of experiments carried out for our SOM-EBGM method shows the accuracy of our +<br/>proposal when compared with other state-of the-art methods. +<br/>Keywords: pattern recognition, face recognition, neural networks, self-organizing maps. +<br/>1. +<br/>Introduction +<br/>libraries, +<br/>In recent years, there has been intensive research carried +<br/>to develop complex security systems involving +<br/>out +<br/>biometric features. +<br/>Automated biometric systems +<br/>are being widely used in many applications such +<br/>as surveillance, digital +<br/>law +<br/>enforcement, human computer intelligent interaction, and +<br/>banking, among others. For applications requiring high +<br/>levels of security, biometrics can be integrated with other +<br/>authentication means such as smart cards and passwords. +<br/>In relation to this, face recognition is an emerging research +<br/>area and, in the next few years, it is supposed to be +<br/>extensively used for automatic human recognition systems +<br/>in many of the applications mentioned before. +<br/>forensic work, +<br/>One of the most popular methods for face recognition +<br/>is elastic graph bunch matching (EBGM), proposed by +<br/>Wiskott et al. (1997). This method is an evolution of the +<br/>so-called dynamic link architecture (DLA) (Kotropoulos +<br/>and Pitas, 1997). The main idea in elastic graph matching +<br/>is to represent a face starting from a set of reference or +<br/>fiducial points known as landmarks. These fiducial points +<br/>have a spatial coherence, as they are connected using a +<br/>graph structure. Therefore, EBGM represents faces as +<br/>facial graphs with nodes at those facial landmarks (such +<br/>Corresponding author +<br/>as eyes, the tip of the nose, etc.). Considering these nodes, +<br/>geometric information can be extracted, and both distance +<br/>and angle metrics can be defined accordingly. +<br/>This algorithm takes into account that facial images +<br/>have many nonlinear features (variations in lighting, +<br/>pose and expression) that are not generally considered +<br/>in linear analysis methods, such as linear discriminant +<br/>analysis (LDA) or principal component analysis (PCA) +<br/>(Shin and Park, 2011). Moreover, it is particularly robust +<br/>when out-of-plane rotations appear. However, the main +<br/>drawback of this method is that it requires an accurate +<br/>location of the fiducial points. +<br/>Artificial neural networks (ANNs) are one of the +<br/>most often used paradigms to address problems in +<br/>artificial intelligence (Ba´nka et al., 2014; Kayarvizhy +<br/>et al., 2014; Tran et al., 2014; Kumar and Kumar, +<br/>2015). Among the different approaches of ANNs, the self +<br/>organizing map (SOM) has special features for association +<br/>and pattern classification (Kohonen, 2001), and it is one of +<br/>the most popular neural network models. This technique +<br/>is suitable in situations where there is an inaccuracy or a +<br/>lack of formalization of the problem to be solved. In these +<br/>cases, there is no precise mathematical formulation of +<br/>the relationship between the input patterns (Azor´ın-L´opez +<br/>et al., 2014). +<br/>The SOM makes use of an unsupervised learning +</td><td>('2274078', 'Francisco A. Pujol', 'francisco a. pujol')</td><td>e-mail: {fpujol,hmora}@dtic.ua.es,jags20@alu.ua.es +</td></tr><tr><td>fdbacf2ff0fc21e021c830cdcff7d347f2fddd8e</td><td>ORIGINAL RESEARCH +<br/>published: 17 August 2018 +<br/>doi: 10.3389/fnhum.2018.00327 +<br/>Recognizing Frustration of Drivers +<br/>From Face Video Recordings and +<br/>Brain Activation Measurements With +<br/>Functional Near-Infrared +<br/>Spectroscopy +<br/><b>Institute of Transportation Systems, German Aerospace Center (DLR), Braunschweig</b><br/><b>Germany, University of Oldenburg, Oldenburg, Germany</b><br/>Experiencing frustration while driving can harm cognitive processing, result in aggressive +<br/>behavior and hence negatively influence driving performance and traffic safety. Being +<br/>able to automatically detect frustration would allow adaptive driver assistance and +<br/>automation systems to adequately react to a driver’s frustration and mitigate potential +<br/>negative consequences. To identify reliable and valid indicators of driver’s frustration, +<br/>we conducted two driving simulator experiments. In the first experiment, we aimed to +<br/>reveal facial expressions that indicate frustration in continuous video recordings of the +<br/>driver’s face taken while driving highly realistic simulator scenarios in which frustrated +<br/>or non-frustrated emotional states were experienced. An automated analysis of facial +<br/>expressions combined with multivariate logistic regression classification revealed that +<br/>frustrated time intervals can be discriminated from non-frustrated ones with accuracy +<br/>of 62.0% (mean over 30 participants). A further analysis of the facial expressions +<br/>revealed that frustrated drivers tend to activate muscles in the mouth region (chin +<br/>raiser, lip pucker, lip pressor). In the second experiment, we measured cortical activation +<br/>with almost whole-head functional near-infrared spectroscopy (fNIRS) while participants +<br/>experienced frustrating and non-frustrating driving simulator scenarios. Multivariate +<br/>logistic regression applied to the fNIRS measurements allowed us to discriminate +<br/>between frustrated and non-frustrated driving intervals with higher accuracy of 78.1% +<br/>(mean over 12 participants). Frustrated driving intervals were indicated by increased +<br/>activation in the inferior frontal, putative premotor and occipito-temporal cortices. +<br/>Our results show that facial and cortical markers of +<br/>frustration can be informative +<br/>for time resolved driver state identification in complex realistic driving situations. The +<br/>markers derived here can potentially be used as an input for future adaptive driver +<br/>assistance and automation systems that detect driver frustration and adaptively react +<br/>to mitigate it. +<br/>Keywords: frustration, driver state recognition, facial expressions, functional near-infrared spectroscopy, adaptive +<br/>automation +<br/>Edited by: +<br/>Guido P. H. Band, +<br/><b>Leiden University, Netherlands</b><br/>Reviewed by: +<br/>Paola Pinti, +<br/><b>University College London</b><br/>United Kingdom +<br/>Edmund Wascher, +<br/>Leibniz-Institut für Arbeitsforschung +<br/>an der TU Dortmund (IfADo), +<br/>Germany +<br/>*Correspondence: +<br/>Received: 17 April 2018 +<br/>Accepted: 25 July 2018 +<br/>Published: 17 August 2018 +<br/>Citation: +<br/>Ihme K, Unni A, Zhang M, Rieger JW +<br/>and Jipp M (2018) Recognizing +<br/>Frustration of Drivers From Face +<br/>Video Recordings and Brain +<br/>Activation Measurements With +<br/>Functional Near-Infrared +<br/>Spectroscopy. +<br/>Front. Hum. Neurosci. 12:327. +<br/>doi: 10.3389/fnhum.2018.00327 +<br/>Frontiers in Human Neuroscience | www.frontiersin.org +<br/>August 2018 | Volume 12 | Article 327 +</td><td>('2873465', 'Klas Ihme', 'klas ihme')<br/>('34722642', 'Anirudh Unni', 'anirudh unni')<br/>('48984951', 'Meng Zhang', 'meng zhang')<br/>('2743311', 'Jochem W. Rieger', 'jochem w. rieger')<br/>('50093361', 'Meike Jipp', 'meike jipp')<br/>('2873465', 'Klas Ihme', 'klas ihme')</td><td>klas.ihme@dlr.de +</td></tr><tr><td>fd892e912149e3f5ddd82499e16f9ea0f0063fa3</td><td>GazeDirector: Fully Articulated Eye Gaze Redirection in Video +<br/><b>University of Cambridge, UK 2Carnegie Mellon University, USA</b><br/><b>Max Planck Institute for Informatics, Germany</b><br/>4Microsoft +</td><td>('34399452', 'Erroll Wood', 'erroll wood')<br/>('49933077', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td></td></tr><tr><td>fde0180735699ea31f6c001c71eae507848b190f</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 76– No.3, August 2013 +<br/>Face Detection and Sex Identification from Color Images +<br/>using AdaBoost with SVM based Component Classifier +<br/>Lecturer, Department of EEE +<br/><b>University of Information</b><br/>Technology and Sciences +<br/>(UITS) +<br/>Dhaka, Bangladesh +<br/>B.Sc. in EEE +<br/><b>International University of</b><br/>Business Agriculture and +<br/>Technology (IUBAT) +<br/>Dhaka-1230, Bangladesh +<br/>Lecturer, Department of EEE +<br/><b>International University of</b><br/>Business Agriculture and +<br/>Technology (IUBAT) +<br/>Dhaka-1230, Bangladesh +</td><td>('1804849', 'Tonmoy Das', 'tonmoy das')<br/>('2832495', 'Md. Hafizur Rahman', 'md. hafizur rahman')</td><td></td></tr><tr><td>fdf8e293a7618f560e76bd83e3c40a0788104547</td><td>Interspecies Knowledge Transfer for Facial Keypoint Detection +<br/><b>University of California, Davis</b><br/><b>Zhejiang University</b><br/><b>University of California, Davis</b></td><td>('35157022', 'Maheen Rashid', 'maheen rashid')<br/>('10734287', 'Xiuye Gu', 'xiuye gu')<br/>('1883898', 'Yong Jae Lee', 'yong jae lee')</td><td>mhnrashid@ucdavis.edu +<br/>gxy0922@zju.edu.cn +<br/>yongjaelee@ucdavis.edu +</td></tr><tr><td>fd615118fb290a8e3883e1f75390de8a6c68bfde</td><td>Joint Face Alignment with Non-Parametric +<br/>Shape Models +<br/><b>University of Wisconsin Madison</b><br/>http://www.cs.wisc.edu/~lizhang/projects/joint-align/ +</td><td>('1893050', 'Brandon M. Smith', 'brandon m. smith')<br/>('40396555', 'Li Zhang', 'li zhang')</td><td></td></tr><tr><td>fdaf65b314faee97220162980e76dbc8f32db9d6</td><td>Accepted Manuscript +<br/>Face recognition using both visible light image and near-infrared image and a deep +<br/>network +<br/>PII: +<br/>DOI: +<br/>Reference: +<br/>S2468-2322(17)30014-8 +<br/>10.1016/j.trit.2017.03.001 +<br/>TRIT 41 +<br/>To appear in: +<br/>CAAI Transactions on Intelligence Technology +<br/>Received Date: 30 January 2017 +<br/>Accepted Date: 28 March 2017 +<br/>Please cite this article as: K. Guo, S. Wu, Y. Xu, Face recognition using both visible light image and +<br/>near-infrared image and a deep network, CAAI Transactions on Intelligence Technology (2017), doi: +<br/>10.1016/j.trit.2017.03.001. +<br/>This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to +<br/>our customers we are providing this early version of the manuscript. The manuscript will undergo +<br/>copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please +<br/>note that during the production process errors may be discovered which could affect the content, and all +<br/>legal disclaimers that apply to the journal pertain. +</td><td>('48477652', 'Kai Guo', 'kai guo')<br/>('40200363', 'Shuai Wu', 'shuai wu')</td><td></td></tr><tr><td>f22d6d59e413ee255e5e0f2104f1e03be1a6722e</td><td>Lattice Long Short-Term Memory for Human Action Recognition +<br/><b>The Hong Kong University of Science and Technology</b><br/><b>Stanford University</b><br/><b>South China University of Technology</b></td><td>('41191188', 'Lin Sun', 'lin sun')<br/>('2370507', 'Kui Jia', 'kui jia')<br/>('1794604', 'Kevin Chen', 'kevin chen')<br/>('2131088', 'Bertram E. Shi', 'bertram e. shi')<br/>('1702137', 'Silvio Savarese', 'silvio savarese')</td><td></td></tr><tr><td>f24e379e942e134d41c4acec444ecf02b9d0d3a9</td><td>International Scholarly Research Network +<br/>ISRN Machine Vision +<br/>Volume 2012, Article ID 505974, 7 pages +<br/>doi:10.5402/2012/505974 +<br/>Research Article +<br/>Analysis of Facial Images across Age Progression by Humans +<br/><b>Temple University, Philadelphia, PA 19122, USA</b><br/><b>Temple University, Philadelphia, PA 19122, USA</b><br/><b>West Virginia University, Morgantown, WV 26506, USA</b><br/>Received 25 July 2011; Accepted 25 August 2011 +<br/>Academic Editors: O. Ghita and R.-H. Park +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>The appearance of human faces can undergo large variations over aging progress. Analysis of facial image taken over age +<br/>progression recently attracts increasing attentions in computer-vision community. Human abilities for such analysis are, however, +<br/>less studied. In this paper, we conduct a thorough study of human ability on two tasks, face verification and age estimation, for +<br/>facial images taken at different ages. Detailed and rigorous experimental analysis is provided, which helps understanding roles of +<br/>different factors including age group, age gap, race, and gender. In addition, our study also leads to an interesting observation: for +<br/>age estimation, photos from adults are more challenging than that from young people. We expect the study to provide a reference +<br/>for machine-based solutions. +<br/>1. Introduction +<br/>Human faces are important in revealing the personal char- +<br/>acteristic and understanding visual data. The facial research +<br/>has been studied over several decades in computer vision +<br/>community [1, 2]. Analysis facial images across age pro- +<br/>gression recently attracts increasing research attention [3] +<br/>because of its important real-life applications. For example, +<br/>facial appearance predictor of missing people and ID photo +<br/>automatic update system are playing important roles in +<br/>simulating face aging of human beings. Age estimation can +<br/>also be applied to age-restricted vending machine [4]. Most +<br/>recent studies (see Section 2) of age-related facial image +<br/>analysis mainly focus on three tasks: face verification, age +<br/>estimation, and age effect simulation. In comparison, it +<br/>remains unclear how humans perform on these tasks. +<br/>In this paper, we study human ability on face verification +<br/>and age estimation for face photos taken at across age +<br/>progression. Such studies are important in that it not only +<br/>provides a reference for future machine-based solutions, +<br/>but also provides insight on how different factors (e.g., age +<br/>gaps, gender, etc.) affect facial analysis algorithms. There are +<br/>previous works on human performance for face recognition +<br/>and age estimation; however, most of them are either +<br/>focusing on nonage related issues such as lighting [5] or +<br/>limited by the scale of image datasets (e.g., [6]). Taking +<br/>advantage of the recent available MORPH dataset [7], which +<br/>to the best of our knowledge is the largest publicly available +<br/>face aging dataset, we are able to conduct thorough human +<br/>studies on facial analysis tasks. +<br/>For face verification, the task is to let a human subject +<br/>decide whether two photos come from the same person (at +<br/>different ages). In addition to report the general performance +<br/>on our human subjects’ performance, we also analyze the +<br/><b>e ects of di erence factors, including age group, age gap</b><br/>race, and gender. In addition, we also compare human +<br/>performance with previous reported baseline algorithm. For +<br/>age estimation, similarly, we report and analyze human +<br/>performance for general cases as well as for different factors. +<br/>Compared to a previous study on the FGNet database [8], +<br/>our study implies that age estimation are harder for photos +<br/>from adults than those from young people. +<br/>The rest of the paper is organized as follows. Section 2 +<br/>shows the related works on different databases. Section 3 +<br/>describes the details of human experiments of face-recog- +<br/>nition and age-estimation problems. Then, in Section 4, +</td><td>('38129124', 'Jingting Zeng', 'jingting zeng')<br/>('1805398', 'Haibin Ling', 'haibin ling')<br/>('1686678', 'Longin Jan Latecki', 'longin jan latecki')<br/>('1822413', 'Guodong Guo', 'guodong guo')<br/>('38129124', 'Jingting Zeng', 'jingting zeng')</td><td>Correspondence should be addressed to Haibin Ling, hbling@temple.edu +</td></tr><tr><td>f2b13946d42a50fa36a2c6d20d28de2234aba3b4</td><td>Adaptive Facial Expression Recognition Using Inter-modal +<br/>Top-down Context +<br/>Ravi Kiran +<br/>Sarvadevabhatla +<br/><b>Honda Research Institute USA</b><br/>425 National Ave, Suite 100 +<br/>Mountain View 94043, USA +<br/>Neural Prosthetics Lab +<br/>Department of Electrical and +<br/>Computer Engineering +<br/><b>McGill University</b><br/>Montreal H3A 2A7, Canada +<br/>Neural Prosthetics Lab +<br/>Department of Electrical and +<br/>Computer Engineering +<br/><b>McGill University</b><br/>Montreal H3A 2A7, Canada +<br/><b>Honda Research Institute USA</b><br/>425 National Ave, Suite 100 +<br/>Mountain View 94043, USA +</td><td>('1708927', 'Mitchel Benovoy', 'mitchel benovoy')<br/>('2003327', 'Sam Musallam', 'sam musallam')<br/>('1692465', 'Victor Ng-Thow-Hing', 'victor ng-thow-hing')</td><td>RSarvadevabhatla@hra.com +<br/>benovoym@mcgill.ca +<br/>sam.musallam@mcgill.ca +<br/>vngthowhing@hra.com +</td></tr><tr><td>f2c30594d917ea915028668bc2a481371a72a14d</td><td>Scene Understanding Using Internet Photo Collections +<br/>A dissertation submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/><b>University of Washington</b><br/>2010 +<br/>Program Authorized to Offer Degree: Computer Science and Engineering +</td><td>('35577716', 'Ian Simon', 'ian simon')</td><td></td></tr><tr><td>f2ad9b43bac8c2bae9dea694f6a4e44c760e63da</td><td>A Study on Illumination Invariant Face Recognition Methods +<br/>Based on Multiple Eigenspaces +<br/>1National Laboratory for Novel Software Technology +<br/><b>Nanjing University, Nanjing 210093, P.R.China</b><br/>2Department of Computer Science +<br/><b>North Dakota State University, Fargo, ND58105, USA</b></td><td>('7878359', 'Wu-Jun Li', 'wu-jun li')<br/>('2697799', 'Chong-Jun Wang', 'chong-jun wang')<br/>('1737124', 'Bin Luo', 'bin luo')</td><td>Email: {liwujun, chjwang}@ai.nju.edu.cn +<br/>Email: Dianxiang.xu@ndsu.nodak.edu +</td></tr><tr><td>f2e9494d0dca9fb6b274107032781d435a508de6</td><td></td><td></td><td></td></tr><tr><td>f2c568fe945e5743635c13fe5535af157b1903d1</td><td></td><td></td><td></td></tr><tr><td>f2a7f9bd040aa8ea87672d38606a84c31163e171</td><td>Human Action Recognition without Human +<br/><b>National Institute of Advanced Industrial Science and Technology (AIST</b><br/>Tsukuba, Ibaraki, Japan +</td><td>('1713046', 'Yun He', 'yun he')<br/>('3393640', 'Soma Shirakabe', 'soma shirakabe')<br/>('1732705', 'Yutaka Satoh', 'yutaka satoh')<br/>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')</td><td>{yun.he, shirakabe-s, yu.satou, hirokatsu.kataoka}@aist.go.jp +</td></tr><tr><td>f257300b2b4141aab73f93c146bf94846aef5fa1</td><td>Eigen Evolution Pooling for Human Action Recognition +<br/><b>Stony Brook University, Stony Brook, NY 11794, USA</b></td><td>('2295608', 'Yang Wang', 'yang wang')<br/>('49701507', 'Vinh Tran', 'vinh tran')<br/>('2356016', 'Minh Hoai', 'minh hoai')</td><td>{wang33, tquangvinh, minhhoai}@cs.stonybrook.edu +</td></tr><tr><td>f20e0eefd007bc310d2a753ba526d33a8aba812c</td><td>Lee et al.: RGB-D FACE RECOGNITION WITH A DEEP LEARNING APPROACH +<br/>Accurate and robust face recognition from +<br/>RGB-D images with a deep learning +<br/>approach +<br/>Yuancheng Lee +<br/>http://cv.cs.nthu.edu.tw/php/people/profile.php?uid=150 +<br/>http://cv.cs.nthu.edu.tw/php/people/profile.php?uid=153 +<br/>Ching-Wei Tseng +<br/>http://cv.cs.nthu.edu.tw/php/people/profile.php?uid=156 +<br/>Computer Vision Lab, +<br/>Department of +<br/>Computer Science, +<br/>National Tsing Hua +<br/><b>University</b><br/>Hsinchu, Taiwan +<br/>http://www.cs.nthu.edu.tw/~lai/ +</td><td>('7557765', 'Jiancong Chen', 'jiancong chen')<br/>('1696527', 'Shang-Hong Lai', 'shang-hong lai')</td><td></td></tr><tr><td>f26097a1a479fb6f32b27a93f8f32609cfe30fdc</td><td></td><td></td><td></td></tr><tr><td>f231046d5f5d87e2ca5fae88f41e8d74964e8f4f</td><td>We are IntechOpen, +<br/>the first native scientific +<br/>publisher of Open Access books +<br/>3,350 +<br/>108,000 +<br/>1.7 M +<br/>Open access books available +<br/>International authors and editors +<br/>Downloads +<br/>Our authors are among the +<br/>151 +<br/>Countries delivered to +<br/>TOP 1% +<br/>12.2% +<br/>most cited scientists +<br/>Contributors from top 500 universities +<br/>Selection of our books indexed in the Book Citation Index +<br/>in Web of Science™ Core Collection (BKCI) +<br/>Interested in publishing with us? +<br/>Numbers displayed above are based on latest data collected. +<br/>For more information visit www.intechopen.com +</td><td></td><td>Contact book.department@intechopen.com +</td></tr><tr><td>f28b7d62208fdaaa658716403106a2b0b527e763</td><td>Clustering-driven Deep Embedding with Pairwise Constraints +<br/><b>JACOB GOLDBERGER, Bar-Ilan University</b><br/>Fig. 1. Employing deep embeddings for clustering 3D shapes. Above, we use PCA to visualize the output embedding of point clouds of chairs. We also highlight +<br/>(in unique colors) a few random clusters and display a few representative chairs from these clusters. +<br/>Recently, there has been increasing interest to leverage the competence +<br/>of neural networks to analyze data. In particular, new clustering meth- +<br/>ods that employ deep embeddings have been presented. In this paper, we +<br/>depart from centroid-based models and suggest a new framework, called +<br/>Clustering-driven deep embedding with PAirwise Constraints (CPAC), for +<br/>non-parametric clustering using a neural network. We present a clustering- +<br/>driven embedding based on a Siamese network that encourages pairs of data +<br/>points to output similar representations in the latent space. Our pair-based +<br/>model allows augmenting the information with labeled pairs to constitute a +<br/>semi-supervised framework. Our approach is based on analyzing the losses +<br/>associated with each pair to refine the set of constraints. We show that clus- +<br/>tering performance increases when using this scheme, even with a limited +<br/>amount of user queries. We demonstrate how our architecture is adapted +<br/>for various types of data and present the first deep framework to cluster 3D +<br/>shapes. +<br/>INTRODUCTION +<br/>Autoencoders provide means to analyze data without supervision. +<br/>Autoencoders based on deep neural networks include non-linear +<br/>neurons which significantly strengthen the power of the analysis. +<br/>The key idea is that the encoders project the data into an embedding +<br/>latent space, where the L2 proximity among the projected elements +<br/>better expresses their similarity. To further enhance the data prox- +<br/>imity in the embedding space, the encoder can be encouraged to +<br/>form tight clusters in the embedding space. Xie et al. [2016] have +<br/>presented an unsupervised embedding driven by a centroid-based +<br/>clustering. They have shown that their deep embedding leads to +<br/>better clustering of the data. More advanced clustering-driven em- +<br/>bedding techniques have been recently presented [Dizaji et al. 2017; +<br/>Yang et al. 2016]. These techniques are all centroid-based and para- +<br/>metric, in the sense that the number of clusters is known a-priori. +<br/>In this paper, we present a clustering-driven embedding technique +<br/>that allows semi-supervision. The idea is to depart from centroid- +<br/>based methods and use pairwise constraints to drive the clustering. +<br/>Most, or all the constraints, can be learned with no supervision, +<br/>while possibly a small portion of the data is supervised. More specifi- +<br/>cally, we adopt robust continuous clustering (RCC) [Shah and Koltun +<br/>2017] as a driving mechanism to encourage a tight clustering of the +<br/>embedded data. +<br/>The idea is to extract pairwise constraints using a mutual k- +<br/>nearest neighbors analysis, and use these pairs as must-link con- +<br/>straints. With no supervision, the set of constraints is imperfect +<br/>and contains false positive pairs on one hand. Our technique allows +<br/>removing false positive pairs and strengthening true positive pairs +<br/>actively by a user. We present an approach that analyzes the losses +<br/>associated with the pairs to form a set of false positive candidates. +<br/>See Figure 2(b)-(c) for a visualization of the distribution of the data +</td><td>('40901326', 'Sharon Fogel', 'sharon fogel')<br/>('1793313', 'Hadar Averbuch-Elor', 'hadar averbuch-elor')<br/>('1701009', 'Daniel Cohen-Or', 'daniel cohen-or')</td><td></td></tr><tr><td>f214bcc6ecc3309e2efefdc21062441328ff6081</td><td></td><td></td><td></td></tr><tr><td>f5149fb6b455a73734f1252a96a9ce5caa95ae02</td><td>Low-Rank-Sparse Subspace Representation for Robust Regression +<br/><b>Harbin Institute of Technology</b><br/><b>Harbin Institute of Technology;Shenzhen University</b><br/>Harbin, China +<br/>Harbin, China;Shenzhen, China +<br/><b>The University of Sydney</b><br/><b>Harbin Institute of Technology</b><br/>Sydney, Australia +<br/>Harbin, China +</td><td>('1747644', 'Yongqiang Zhang', 'yongqiang zhang')<br/>('1887263', 'Daming Shi', 'daming shi')<br/>('1750488', 'Junbin Gao', 'junbin gao')<br/>('2862899', 'Dansong Cheng', 'dansong cheng')</td><td>seekever@foxmail.com +<br/>d.m.shi@hotmail.com +<br/>junbin.gao@sydney.edu.au +<br/>cdsinhit@hit.edu.cn +</td></tr><tr><td>f58d584c4ac93b4e7620ef6e5a8f20c6f6da295e</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Feature Selection Guided Auto-Encoder +<br/>1Department of Electrical & Computer Engineering, +<br/><b>College of Computer and Information Science</b><br/><b>Northeastern University, Boston, MA, USA</b></td><td>('47673521', 'Shuyang Wang', 'shuyang wang')<br/>('2788685', 'Zhengming Ding', 'zhengming ding')<br/>('1708679', 'Yun Fu', 'yun fu')</td><td>{shuyangwang, allanding, yunfu}@ece.neu.edu +</td></tr><tr><td>f5eb0cf9c57716618fab8e24e841f9536057a28a</td><td>Rethinking Feature Distribution for Loss Functions in Image Classification +<br/><b>Tsinghua University, Beijing, China</b><br/><b>University of at Urbana-Champaign, Illinois, USA</b></td><td>('47718901', 'Weitao Wan', 'weitao wan')<br/>('1752427', 'Jiansheng Chen', 'jiansheng chen')<br/>('8802368', 'Yuanyi Zhong', 'yuanyi zhong')<br/>('2641581', 'Tianpeng Li', 'tianpeng li')</td><td>wwt16@mails.tsinghua.edu.cn +<br/>yuanyiz2@illinois.edu +<br/>ltp16@mails.tsinghua.edu.cn +<br/>jschenthu@mail.tsinghua.edu.cn +</td></tr><tr><td>f571fe3f753765cf695b75b1bd8bed37524a52d2</td><td>Submodular Attribute Selection for Action +<br/>Recognition in Video +<br/>Jinging Zheng +<br/><b>UMIACS, University of Maryland</b><br/><b>College Park, MD, USA</b><br/>Noah’s Ark Lab +<br/>Huawei Technologies +<br/><b>UMIACS, University of Maryland</b><br/><b>National Institute of Standards and Technology</b><br/><b>College Park, MD, USA</b><br/>Gaithersburg, MD, USA +</td><td>('34145947', 'Zhuolin Jiang', 'zhuolin jiang')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')</td><td>zjngjng@umiacs.umd.edu +<br/>zhuolin.jiang@huawei.com +<br/>rama@umiacs.umd.edu +<br/>jonathon.phillips@nist.gov +</td></tr><tr><td>f5fae7810a33ed67852ad6a3e0144cb278b24b41</td><td>Multilingual Gender Classification with Multi-view +<br/>Deep Learning +<br/>Notebook for PAN at CLEF 2018 +<br/><b>Jo ef Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia</b><br/>2 Jožef Stefan International Postgraduate School, Jamova 39, 1000 Ljubljana, Slovenia +<br/><b>USHER Institute, University of Edinburgh, United Kingdom</b></td><td>('22684661', 'Matej Martinc', 'matej martinc')<br/>('40235216', 'Senja Pollak', 'senja pollak')</td><td>{matej.martinc,blaz.skrlj,senja.pollak}@ijs.si +</td></tr><tr><td>f5af4e9086b0c3aee942cb93ece5820bdc9c9748</td><td>ENHANCING PERSON ANNOTATION +<br/>FOR PERSONAL PHOTO MANAGEMENT +<br/>USING CONTENT AND CONTEXT +<br/>BASED TECHNOLOGIES +<br/>By +<br/>THESIS DIRECTED BY: PROF. NOEL E. O’CONNOR +<br/>A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE +<br/>DEGREE OF DOCTOR OF PHILOSOPHY +<br/>September 2008 +<br/>SCHOOL OF ELECTRONIC ENGINEERING +<br/><b>DUBLIN CITY UNIVERSITY</b></td><td>('2668569', 'Saman H. Cooray', 'saman h. cooray')</td><td></td></tr><tr><td>f5770dd225501ff3764f9023f19a76fad28127d4</td><td>Real Time Online Facial Expression Transfer +<br/>with Single Video Camera +</td><td></td><td></td></tr><tr><td>f5aee1529b98136194ef80961ba1a6de646645fe</td><td>Large-Scale Learning of +<br/>Discriminative Image Representations +<br/>D.Phil Thesis +<br/>Robotics Research Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b><br/>Supervisors: +<br/>Professor Andrew Zisserman +<br/>Doctor Antonio Criminisi +<br/><b>Mans eld College</b><br/>Trinity Term, 2013 +</td><td>('34838386', 'Karen Simonyan', 'karen simonyan')</td><td></td></tr><tr><td>f52efc206432a0cb860155c6d92c7bab962757de</td><td>MUGSHOT DATABASE ACQUISITION IN VIDEO SURVEILLANCE NETWORKS USING +<br/>INCREMENTAL AUTO-CLUSTERING QUALITY MEASURES +<br/>Computer Science Department +<br/><b>University of Kentucky</b><br/>Lexington, KY, 40508 +</td><td>('3237043', 'Quanren Xiong', 'quanren xiong')</td><td></td></tr><tr><td>f519723238701849f1160d5a9cedebd31017da89</td><td>Impact of multi-focused images on recognition of soft biometric traits +<br/>aEURECOM, Campus SophiaTech, 450 Route des Chappes, CS 50193 - 06904 Biot Sophia +<br/> +<br/>Antipolis cedex, FRANCE +</td><td>('24362694', 'V. Chiesa', 'v. chiesa')</td><td></td></tr><tr><td>f5eb411217f729ad7ae84bfd4aeb3dedb850206a</td><td>Tackling Low Resolution for Better Scene Understanding +<br/>Thesis submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/>MS in Computer Science and Engineering +<br/>By Research +<br/>by +<br/>201202172 +<br/><b>International Institute of Information Technology</b><br/>Hyderabad - 500 032, INDIA +<br/>July 2018 +</td><td>('41033644', 'Harish Krishna', 'harish krishna')</td><td>harishkrishna.v@research.iiit.ac.in +</td></tr><tr><td>f558af209dd4c48e4b2f551b01065a6435c3ef33</td><td>International Journal of Emerging Technology in Computer Science & Electronics (IJETCSE) +<br/>ISSN: 0976-1353 Volume 23 Issue 1 –JUNE 2016. +<br/>AN ENHANCED ATTRIBUTE +<br/>RERANKING DESIGN FOR WEB IMAGE +<br/>SEARCH +<br/>#Student,Cse, CIET, Lam,Guntur, India +<br/>* Assistant Professort,Cse, CIET, Lam,Guntur , India +</td><td>('4384318', 'G K Kishore Babu', 'g k kishore babu')</td><td></td></tr><tr><td>e378ce25579f3676ca50c8f6454e92a886b9e4d7</td><td>Robust Video Super-Resolution with Learned Temporal Dynamics +<br/><b>University of Illinois at Urbana-Champaign 2Adobe Research</b><br/><b>Facebook 4Texas AandM University 5IBM Research</b></td><td>('1771885', 'Ding Liu', 'ding liu')<br/>('2969311', 'Zhangyang Wang', 'zhangyang wang')</td><td></td></tr><tr><td>e393a038d520a073b9835df7a3ff104ad610c552</td><td>Automatic temporal segment +<br/>detection via bilateral long short- +<br/>term memory recurrent neural +<br/>networks +<br/>detection via bilateral long short-term memory recurrent neural networks,” J. +<br/>Electron. Imaging 26(2), 020501 (2017), doi: 10.1117/1.JEI.26.2.020501. +<br/>Downloaded From: http://electronicimaging.spiedigitallibrary.org/ on 03/03/2017 Terms of Use: http://spiedigitallibrary.org/ss/termsofuse.aspx </td><td>('49447269', 'Bo Sun', 'bo sun')<br/>('7886608', 'Siming Cao', 'siming cao')<br/>('49264106', 'Jun He', 'jun he')<br/>('8834504', 'Lejun Yu', 'lejun yu')<br/>('2089565', 'Liandong Li', 'liandong li')<br/>('49447269', 'Bo Sun', 'bo sun')<br/>('7886608', 'Siming Cao', 'siming cao')<br/>('49264106', 'Jun He', 'jun he')<br/>('8834504', 'Lejun Yu', 'lejun yu')<br/>('2089565', 'Liandong Li', 'liandong li')</td><td></td></tr><tr><td>e35b09879a7df814b2be14d9102c4508e4db458b</td><td>Optimal Sensor Placement and +<br/>Enhanced Sparsity for Classification +<br/><b>University of Washington, Seattle, WA 98195, United States</b><br/><b>University of Washington, Seattle, WA 98195, United States</b><br/><b>Institute for Disease Modeling, Intellectual Ventures Laboratory, Bellevue, WA 98004, United States</b></td><td>('1824880', 'Bingni W. Brunton', 'bingni w. brunton')<br/>('3083169', 'Steven L. Brunton', 'steven l. brunton')<br/>('2424683', 'Joshua L. Proctor', 'joshua l. proctor')<br/>('1937069', 'J. Nathan Kutz', 'j. nathan kutz')</td><td></td></tr><tr><td>e3b324101157daede3b4d16bdc9c2388e849c7d4</td><td>Robust Real-Time 3D Face Tracking from RGBD Videos under Extreme Pose, +<br/>Depth, and Expression Variations +<br/>Hai X. Pham +<br/><b>Rutgers University, USA</b></td><td>('1736042', 'Vladimir Pavlovic', 'vladimir pavlovic')</td><td>{hxp1,vladimir}@cs.rutgers.edu +</td></tr><tr><td>e3657ab4129a7570230ff25ae7fbaccb4ba9950c</td><td></td><td></td><td></td></tr><tr><td>e315959d6e806c8fbfc91f072c322fb26ce0862b</td><td>An Efficient Face Recognition System Based on Sub-Window +<br/>International Journal of Soft Computing and Engineering (IJSCE) +<br/>ISSN: 2231-2307, Volume-1, Issue-6, January 2012 +<br/>Extraction Algorithm +</td><td>('1696227', 'Manish Gupta', 'manish gupta')<br/>('36776003', 'Govind sharma', 'govind sharma')</td><td></td></tr><tr><td>e3c011d08d04c934197b2a4804c90be55e21d572</td><td>How to Train Triplet Networks with 100K Identities? +<br/>Orion Star +<br/>Beijing, China +<br/>Orion Star +<br/>Beijing, China +<br/>Orion Star +<br/>Beijing, China +</td><td>('1747751', 'Chong Wang', 'chong wang')<br/>('46447079', 'Xue Zhang', 'xue zhang')<br/>('26403761', 'Xipeng Lan', 'xipeng lan')</td><td>chongwang.nlpr@gmail.com +<br/>yuannixue@126.com +<br/>xipeng.lan@gmail.com +</td></tr><tr><td>e39a0834122e08ba28e7b411db896d0fdbbad9ba</td><td>1368 +<br/>Maximum Likelihood Estimation of Depth Maps +<br/>Using Photometric Stereo +</td><td>('2964822', 'Adam P. Harrison', 'adam p. harrison')<br/>('39367958', 'Dileepan Joseph', 'dileepan joseph')</td><td></td></tr><tr><td>e3bb83684817c7815f5005561a85c23942b1f46b</td><td>Face Verification using Correlation Filters +<br/>Electrical and Computer Eng. Dept, +<br/>Electrical and Computer Eng. Dept, +<br/>Electrical and Computer Eng. Dept, +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213, U.S.A. +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213, U.S.A. +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213, U.S.A. +</td><td>('1794486', 'Marios Savvides', 'marios savvides')<br/>('36754879', 'Vijaya Kumar', 'vijaya kumar')<br/>('34607721', 'Pradeep Khosla', 'pradeep khosla')</td><td>msavvid@ri.cmu.edu +<br/>kumar@ece.cmu.edu +<br/>pkk@ece.cmu.edu +</td></tr><tr><td>e30dc2abac4ecc48aa51863858f6f60c7afdf82a</td><td>Facial Signs and Psycho-physical Status Estimation for Well-being +<br/>Assessment +<br/>F. Chiarugi, G. Iatraki, E. Christinaki, D. Manousos, G. Giannakakis, M. Pediaditis, +<br/>A. Pampouchidou, K. Marias and M. Tsiknakis +<br/><b>Computational Medicine Laboratory, Institute of Computer Science, Foundation for Research and Technology - Hellas</b><br/>70013 Vasilika Vouton, Heraklion, Crete, Greece +<br/>Keywords: +<br/>Facial Expression, Stress, Anxiety, Feature Selection, Well-being Evaluation, FACS, FAPS, Classification. +</td><td></td><td>{chiarugi, giatraki, echrist, mandim, ggian, mped, pampouch, kmarias, tsiknaki}@ics.forth.gr +</td></tr><tr><td>e3e2c106ccbd668fb9fca851498c662add257036</td><td>Appearance, Context and Co-occurrence Ensembles for +<br/>Identity Recognition in Personal Photo Collections +<br/><b>University of Colorado at Colorado Springs</b><br/>T.E.Boult1 +<br/>2AT&T Labs-Research, Middletown, NJ +</td><td>('27469806', 'Archana Sapkota', 'archana sapkota')<br/>('33692583', 'Raghuraman Gopalan', 'raghuraman gopalan')<br/>('2900213', 'Eric Zavesky', 'eric zavesky')</td><td>1 {asapkota,tboult}@vast.uccs.edu +<br/>2{raghuram,ezavesky}@research.att.com +</td></tr><tr><td>e379e73e11868abb1728c3acdc77e2c51673eb0d</td><td>In S.Li and A.Jain, (ed). Handbook of Face Recognition. Springer-Verlag, 2005 +<br/>Face Databases +<br/><b>The Robotics Inistitute, Carnegie Mellon University</b><br/>5000 Forbes Avenue, Pittsburgh, PA 15213 +<br/>Because of its nonrigidity and complex three-dimensional (3D) structure, the appearance of a face is affected by a large +<br/>number of factors including identity, face pose, illumination, facial expression, age, occlusion, and facial hair. The develop- +<br/>ment of algorithms robust to these variations requires databases of sufficient size that include carefully controlled variations +<br/>of these factors. Furthermore, common databases are necessary to comparatively evaluate algorithms. Collecting a high +<br/>quality database is a resource-intensive task: but the availability of public face databases is important for the advancement of +<br/>the field. In this chapter we review 27 publicly available databases for face recognition, face detection, and facial expression +<br/>analysis. +<br/>1 Databases for Face Recognition +<br/>Face recognition continues to be one of the most popular research areas of computer vision and machine learning. Along +<br/>with the development of face recognition algorithms, a comparatively large number of face databases have been collected. +<br/>However, many of these databases are tailored to the specific needs of the algorithm under development. In this section +<br/>we review publicly available databases that are of demonstrated use to others in the community. At the beginning of each +<br/><b>subsection a table summarizing the key features of the database is provided, including (where available) the number of</b><br/>subjects, recording conditions, image resolution, and total number of images. Table 1 gives an overview of the recording +<br/>conditions for all databases discussed in this section. Owing to space constraints not all databases are discussed at the same +<br/>level of detail. Abbreviated descriptions of a number of mostly older databases are included in Section 1.13. The scope of +<br/>this section is limited to databases containing full face imagery. Note, however, that there are databases of subface images +<br/>available, such as the recently released CASIA Iris database [23]. +<br/>1.1 AR Database +<br/>No. of subjects +<br/>116 +<br/>Conditions +<br/>Facial expressions +<br/>Illumination +<br/>Occlusion +<br/>Time +<br/>Image Resolution +<br/>No. of Images +<br/>768 × 576 +<br/>3288 +<br/>http://rvl1.ecn.purdue.edu/˜aleix/aleix face DB.html +<br/>The AR database was collected at the Computer Vision Center in Barcelona, Spain in 1998 [25]. It contains images of +<br/>116 individuals (63 men and 53 women). The imaging and recording conditions (camera parameters, illumination setting, +<br/>camera distance) were carefully controlled and constantly recalibrated to ensure that settings are identical across subjects. +<br/>The resulting RGB color images are 768 × 576 pixels in size. The subjects were recorded twice at a 2–week interval. During +<br/>each session 13 conditions with varying facial expressions, illumination and occlusion were captured. Figure 1 shows an +<br/>example for each condition. So far, more than 200 research groups have accessed the database. +</td><td>('33731953', 'Ralph Gross', 'ralph gross')</td><td>Email: {rgross}@cs.cmu.edu +</td></tr><tr><td>e39a66a6d1c5e753f8e6c33cd5d335f9bc9c07fa</td><td><b>University of Massachusetts - Amherst</b><br/>Dissertations +<br/>5-1-2012 +<br/>Dissertations and Theses +<br/>Weakly Supervised Learning for Unconstrained +<br/>Face Processing +<br/>Follow this and additional works at: http://scholarworks.umass.edu/open_access_dissertations +<br/>Recommended Citation +<br/>Huang, Gary B., "Weakly Supervised Learning for Unconstrained Face Processing" (2012). Dissertations. Paper 559. +</td><td>('3219900', 'Gary B. Huang', 'gary b. huang')</td><td>ScholarWorks@UMass Amherst +<br/>University of Massachusetts - Amherst, garybhuang@gmail.com +<br/>This Open Access Dissertation is brought to you for free and open access by the Dissertations and Theses at ScholarWorks@UMass Amherst. It has +<br/>been accepted for inclusion in Dissertations by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact +<br/>scholarworks@library.umass.edu. +</td></tr><tr><td>e3a6e9ddbbfc4c5160082338d46808cea839848a</td><td>Vision-Based Classification of Developmental Disorders +<br/>Using Eye-Movements +<br/><b>Stanford University, USA</b><br/><b>Stanford University, USA</b><br/><b>Stanford University, USA</b><br/><b>Stanford University, USA</b><br/><b>Stanford University, USA</b></td><td>('3147852', 'Guido Pusiol', 'guido pusiol')<br/>('1811529', 'Andre Esteva', 'andre esteva')<br/>('3472674', 'Arnold Milstein', 'arnold milstein')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td></td></tr><tr><td>e3c8e49ffa7beceffca3f7f276c27ae6d29b35db</td><td>Families in the Wild (FIW): Large-Scale Kinship Image +<br/>Database and Benchmarks +<br/><b>Northeastern University, Boston, USA</b><br/><b>College of Computer and Information Science, Northeastern University, Boston, USA</b></td><td>('4056993', 'Joseph P. Robinson', 'joseph p. robinson')<br/>('49248003', 'Ming Shao', 'ming shao')<br/>('47096713', 'Yue Wu', 'yue wu')<br/>('1708679', 'Yun Fu', 'yun fu')</td><td>{jrobins1, mingshao, yuewu, yunfu}@ece.neu.edu +</td></tr><tr><td>e38371b69be4f341baa95bc854584e99b67c6d3a</td><td>DYAN: A Dynamical Atoms-Based Network +<br/>For Video Prediction(cid:63) +<br/><b>Electrical and Computer Engineering, Northeastern University, Boston, MA</b><br/>http://robustsystems.coe.neu.edu +</td><td>('40366599', 'WenQian Liu', 'wenqian liu')<br/>('1785252', 'Abhishek Sharma', 'abhishek sharma')<br/>('30929906', 'Octavia Camps', 'octavia camps')<br/>('1687866', 'Mario Sznaier', 'mario sznaier')</td><td>liu.wenqi,sharma.abhis@husky.neu.edu, camps,msznaier@northeastern.edu +</td></tr><tr><td>e3917d6935586b90baae18d938295e5b089b5c62</td><td>152 +<br/>Face Localization and Authentication +<br/>Using Color and Depth Images +</td><td>('1807962', 'Filareti Tsalakanidou', 'filareti tsalakanidou')<br/>('1744180', 'Sotiris Malassiotis', 'sotiris malassiotis')<br/>('1721460', 'Michael G. Strintzis', 'michael g. strintzis')</td><td></td></tr><tr><td>e328d19027297ac796aae2470e438fe0bd334449</td><td>Automatic Micro-expression Recognition from +<br/>Long Video using a Single Spotted Apex +<br/>1 Faculty of Computer Science & Information Technology, +<br/><b>University of Malaya, Kuala Lumpur, Malaysia</b><br/>2 Faculty of Computing & Informatics, +<br/><b>Multimedia University, Cyberjaya, Malaysia</b><br/>3 Faculty of Engineering, +<br/><b>Multimedia University, Cyberjaya, Malaysia</b></td><td>('39888137', 'Sze-Teng Liong', 'sze-teng liong')<br/>('2339975', 'John See', 'john see')<br/>('1713159', 'KokSheik Wong', 'koksheik wong')</td><td>szeteng1206@hotmail.com,koksheik@um.edu.my +<br/>johnsee@mmu.edu.my +<br/>raphael@mmu.edu.my +</td></tr><tr><td>e3144f39f473e238374dd4005c8b83e19764ae9e</td><td>Next-Flow: Hybrid Multi-Tasking with Next-Frame Prediction to Boost +<br/>Optical-Flow Estimation in the Wild +<br/><b>University of Freiburg</b><br/>Germany +</td><td>('31656404', 'Nima Sedaghat', 'nima sedaghat')</td><td>nima@cs.uni-freiburg.de +</td></tr><tr><td>e3a6e5a573619a97bd6662b652ea7d088ec0b352</td><td>Compare and Contrast: Learning Prominent Visual Differences +<br/><b>The University of Texas at Austin</b></td><td>('50357985', 'Steven Chen', 'steven chen')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>cfeb26245b57dd10de8f187506d4ed5ce1e2b7dd</td><td>CapsNet comparative performance evaluation for image +<br/>classification +<br/><b>University of Waterloo, ON, Canada</b></td><td>('30421594', 'Rinat Mukhometzianov', 'rinat mukhometzianov')<br/>('36957611', 'Juan Carrillo', 'juan carrillo')</td><td></td></tr><tr><td>cffebdf88e406c27b892857d1520cb2d7ccda573</td><td>LEARNING FROM LARGE-SCALE VISUAL DATA +<br/>FOR ROBOTS +<br/>A Dissertation +<br/>Presented to the Faculty of the Graduate School +<br/><b>of Cornell University</b><br/>in Partial Fulfillment of the Requirements for the Degree of +<br/>Doctor of Philosophy +<br/>by +<br/>Ozan S¸ener +<br/>August 2016 +</td><td></td><td></td></tr><tr><td>cfa572cd6ba8dfc2ee8ac3cc7be19b3abff1a8a2</td><td></td><td></td><td></td></tr><tr><td>cfffae38fe34e29d47e6deccfd259788176dc213</td><td>TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. X, NO. X, DECEMBER 2012 +<br/>Matrix Completion for Weakly-supervised +<br/>Multi-label Image Classification +</td><td>('1707876', 'Fernando De la Torre', 'fernando de la torre')<br/>('2884203', 'Alexandre Bernardino', 'alexandre bernardino')</td><td></td></tr><tr><td>cfd4004054399f3a5f536df71f9b9987f060f434</td><td>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. ??, NO. ??, ?? 20?? +<br/>Person Recognition in Personal Photo Collections +</td><td>('2390510', 'Seong Joon Oh', 'seong joon oh')<br/>('1798000', 'Rodrigo Benenson', 'rodrigo benenson')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td></td></tr><tr><td>cfd933f71f4a69625390819b7645598867900eab</td><td>INTERNATIONAL JOURNAL OF TECHNOLOGY ENHANCEMENTS AND EMERGING ENGINEERING RESEARCH, VOL 3, ISSUE 03 55 +<br/>ISSN 2347-4289 +<br/>Person Authentication Using Face And Palm Vein: +<br/>A Survey Of Recognition And Fusion Techniques +<br/><b>College of Engineering, Pune, India</b><br/>Image Processing & Machine Vision Section, Electronics & Instrumentation Services Division, BARC +</td><td>('38561481', 'Dhanashree Vaidya', 'dhanashree vaidya')<br/>('2623250', 'Madhuri A. Joshi', 'madhuri a. joshi')</td><td>Email: preethimedu@gmail.com, dvaidya33@gmail.com, hod.extc@coep.ac.in, maj.extc@coep.ac.in, skar@barc.gov.in +</td></tr><tr><td>cfb8bc66502fb5f941ecdb22aec1fdbfdb73adce</td><td></td><td></td><td></td></tr><tr><td>cf875336d5a196ce0981e2e2ae9602580f3f6243</td><td>7 What 1 +<br/>Rosalind W. Picard +<br/>It Mean for a Computer to "Have" Emotions? +<br/>There is a lot of talk about giving machines emotions, some of +<br/>it fluff. Recently at a large technical meeting, a researcher stood up +<br/>and talked of how a Bamey stuffed animal [the purple dinosaur for +<br/>kids) "has emotions." He did not define what he meant by this, but +<br/>after repeating it several times, it became apparent that children +<br/>attributed emotions to Barney, and that Barney had deliberately +<br/>expressive behaviors that would encourage the kids to think. Bar- +<br/>ney had emotions. But kids have attributed emotions to dolls and +<br/>stuffed animals for as long a s we know; and most of my technical +<br/>colleagues would agree that such toys have never had and still do +<br/>not have emotions. What is different now that prompts a researcher +<br/>to make such a claim? Is the computational plush an example of a +<br/>computer that really does have emotions? +<br/>If not Barney, then what would be an example of a computa- +<br/>tional system that has emotions? I am not a philosopher, and this +<br/>paper will not be a discussion of the meaning of this question in +<br/>any philosophical sense. However, as an engineer I am interested +<br/>in what capabilities I would require a machine to have before I +<br/>would say that it "has emotions," if that is even possible. +<br/>Theorists still grappl~ with the problem of defining emotion, +<br/>after many decades of discussion, and no clean definition looks +<br/>likely to emerge. Even without a precise definition, one can still +<br/>begin to say concrete things about certain components of emotion, +<br/>at least based on what is known about human and animal emo- +<br/>tions. Of course, much is still u d a o w n about human emotions, so +<br/>we are nowhere near being able to model them, much less dupli- +<br/>cate all their functions in machines.'~lso, all scientific findings are +<br/>subject to revision-history has certainly taught us humility, that +<br/>what scientists believed to be true at one point has often been +<br/>changed at a later date. +<br/>I wish to begin by mentioning four motivations for giving +<br/>machines certain emotional abilities (and there are more). One goal +<br/>is to build robots and synthetic characters that can emulate living +<br/>humans and animals-for example, to build a humanoid robot. A +<br/>I +</td><td></td><td></td></tr><tr><td>cfd8c66e71e98410f564babeb1c5fd6f77182c55</td><td>Comparative Study of Coarse Head Pose Estimation +<br/><b>IBM T.J. Watson Research Center</b><br/>Hawthorne, NY 10532 +</td><td>('34609371', 'Lisa M. Brown', 'lisa m. brown')<br/>('40383812', 'Ying-Li Tian', 'ying-li tian')</td><td>{lisabr,yltian}@us.ibm.com +</td></tr><tr><td>cf54a133c89f730adc5ea12c3ac646971120781c</td><td></td><td></td><td></td></tr><tr><td>cfbb2d32586b58f5681e459afd236380acd86e28</td><td>Improving Alignment of Faces for Recognition +<br/>Christopher J. Pal +<br/>D´epartement de g´enie informatique et g´enie logiciel +<br/>´Ecole Polytechnique de Montr´eal, +<br/>D´epartement de g´enie informatique et g´enie logiciel +<br/>´Ecole Polytechnique de Montr´eal, +<br/>Qu´ebec, Canada +<br/>Qu´ebec, Canada +</td><td>('2811524', 'Md. Kamrul Hasan', 'md. kamrul hasan')</td><td>md-kamrul.hasan@polymtl.ca +<br/>christopher.pal@polymtl.ca +</td></tr><tr><td>cfa92e17809e8d20ebc73b4e531a1b106d02b38c</td><td>Advances in Data Analysis and Classification manuscript No. +<br/>(will be inserted by the editor) +<br/>Parametric Classification with Soft Labels using the +<br/>Evidential EM Algorithm +<br/>Linear Discriminant Analysis vs. Logistic Regression +<br/>Received: date / Accepted: date +</td><td>('1772306', 'Benjamin Quost', 'benjamin quost')<br/>('2259794', 'Shoumei Li', 'shoumei li')</td><td></td></tr><tr><td>cf5c9b521c958b84bb63bea9d5cbb522845e4ba7</td><td>Towards Arbitrary-View Face Alignment by Recommendation Trees∗ +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b><br/>2SenseTime Group +</td><td>('2226254', 'Shizhan Zhu', 'shizhan zhu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>zs014@ie.cuhk.edu.hk, chengli@sensetime.com, ccloy@ie.cuhk.edu.hk, xtang@ie.cuhk.edu.hk +</td></tr><tr><td>cf5a0115d3f4dcf95bea4d549ec2b6bdd7c69150</td><td>Detection of emotions from video in non-controlled +<br/>environment +<br/>To cite this version: +<br/>Processing. Universit´e Claude Bernard - Lyon I, 2013. English. <NNT : 2013LYO10227>. +<br/><tel-01166539v2> +<br/>HAL Id: tel-01166539 +<br/>https://tel.archives-ouvertes.fr/tel-01166539v2 +<br/>Submitted on 23 Jun 2015 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('1943666', 'Rizwan Ahmed Khan', 'rizwan ahmed khan')<br/>('1943666', 'Rizwan Ahmed Khan', 'rizwan ahmed khan')</td><td></td></tr><tr><td>cfdc632adcb799dba14af6a8339ca761725abf0a</td><td>Probabilistic Formulations of Regression with Mixed +<br/>Guidance +</td><td>('38688704', 'Aubrey Gress', 'aubrey gress')<br/>('38673135', 'Ian Davidson', 'ian davidson')</td><td>adgress@ucdavis.edu, davidson@cs.ucdavis.edu +</td></tr><tr><td>cfa931e6728a825caada65624ea22b840077f023</td><td>Deformable Generator Network: Unsupervised Disentanglement of +<br/>Appearance and Geometry +<br/><b>College of Automation, Harbin Engineering University, Heilongjiang, China</b><br/><b>University of California, Los Angeles, California, USA</b></td><td>('7306249', 'Xianglei Xing', 'xianglei xing')<br/>('9659905', 'Ruiqi Gao', 'ruiqi gao')<br/>('50495880', 'Tian Han', 'tian han')<br/>('3133970', 'Song-Chun Zhu', 'song-chun zhu')<br/>('39092098', 'Ying Nian Wu', 'ying nian wu')</td><td></td></tr><tr><td>cfc30ce53bfc204b8764ebb764a029a8d0ad01f4</td><td>Regularizing Deep Neural Networks by Noise: +<br/>Its Interpretation and Optimization +<br/>Dept. of Computer Science and Engineering, POSTECH, Korea +</td><td>('2018393', 'Hyeonwoo Noh', 'hyeonwoo noh')<br/>('2205770', 'Tackgeun You', 'tackgeun you')<br/>('8511875', 'Jonghwan Mun', 'jonghwan mun')<br/>('40030651', 'Bohyung Han', 'bohyung han')</td><td>{shgusdngogo,tackgeun.you,choco1916,bhhan}@postech.ac.kr +</td></tr><tr><td>cff911786b5ac884bb71788c5bc6acf6bf569eff</td><td>Multi-task Learning of Cascaded CNN for +<br/>Facial Attribute Classification +<br/><b>School of Information Science and Engineering, Xiamen University, Xiamen 361005, China</b><br/><b>School of Computer and Information Engineering, Xiamen University of Technology, Xiamen 361024, China</b></td><td>('41034942', 'Ni Zhuang', 'ni zhuang')<br/>('40461734', 'Yan Yan', 'yan yan')<br/>('47336404', 'Si Chen', 'si chen')<br/>('37414077', 'Hanzi Wang', 'hanzi wang')</td><td>Email: ni.zhuang@foxmail.com, {yanyan, hanzi.wang}@xmu.edu.cn, chensi@xmut.edu.cn +</td></tr><tr><td>cf09e2cb82961128302b99a34bff91ec7d198c7c</td><td>OFFICE ENTRANCE CONTROL WITH FACE RECOGNITION +<br/> Dept. of Computer Science and Information Engineering, +<br/><b>National Taiwan University, Taiwan</b><br/> Dept. of Computer Science and Information Engineering, +<br/><b>National Taiwan University, Taiwan</b></td><td>('1721106', 'Yun-Che Tsai', 'yun-che tsai')<br/>('1703041', 'Chiou-Shann Fuh', 'chiou-shann fuh')</td><td>E-mail: jpm9ie8c@gmail.com +<br/>E-mail: fuh@csie.ntu.edu.tw +</td></tr><tr><td>cfc4aa456d9da1a6fabd7c6ca199332f03e35b29</td><td><b>University of Amsterdam and Renmin University at TRECVID</b><br/>Searching Video, Detecting Events and Describing Video +<br/><b>University of Amsterdam</b><br/><b>Zhejiang University</b><br/>Amsterdam, The Netherlands +<br/>Hangzhou, China +<br/><b>Renmin University of China</b><br/>Beijing, China +</td><td>('46741353', 'Cees G. M. Snoek', 'cees g. m. snoek')<br/>('40240283', 'Jianfeng Dong', 'jianfeng dong')<br/>('9931285', 'Xirong Li', 'xirong li')<br/>('48631563', 'Xiaoxu Wang', 'xiaoxu wang')<br/>('24332496', 'Qijie Wei', 'qijie wei')<br/>('2896042', 'Weiyu Lan', 'weiyu lan')<br/>('2304222', 'Efstratios Gavves', 'efstratios gavves')<br/>('13142264', 'Noureldien Hussein', 'noureldien hussein')<br/>('1769315', 'Dennis C. Koelma', 'dennis c. koelma')<br/>('1705182', 'Arnold W. M. Smeulders', 'arnold w. m. smeulders')</td><td></td></tr><tr><td>cf805d478aeb53520c0ab4fcdc9307d093c21e52</td><td>Finding Tiny Faces in the Wild with Generative Adversarial Network +<br/>Mingli Ding2 +<br/><b>Visual Computing Center, King Abdullah University of Science and Technology (KAUST</b><br/><b>School of Electrical Engineering and Automation, Harbin Institute of Technology (HIT</b><br/><b>Institute of Software, Chinese Academy of Sciences (CAS</b><br/>Figure1. The detection results of tiny faces in the wild. (a) is the original low-resolution blurry face, (b) is the result of +<br/>re-sizing directly by a bi-linear kernel, (c) is the generated image by the super-resolution method, and our result (d) is learned +<br/>by the super-resolution (×4 upscaling) and refinement network simultaneously. Best viewed in color and zoomed in. +</td><td>('2860057', 'Yancheng Bai', 'yancheng bai')<br/>('48378890', 'Yongqiang Zhang', 'yongqiang zhang')<br/>('2931652', 'Bernard Ghanem', 'bernard ghanem')</td><td>baiyancheng20@gmail.com +<br/>{zhangyongqiang, dingml}@hit.edu.cn +<br/>bernard.ghanem@kaust.edu.sa +</td></tr><tr><td>cfdc4d0f8e1b4b9ced35317d12b4229f2e3311ab</td><td>Quaero at TRECVID 2010: Semantic Indexing +<br/>1UJF-Grenoble 1 / UPMF-Grenoble 2 / Grenoble INP / CNRS, LIG UMR 5217, Grenoble, F-38041, France +<br/><b>Karlsruhe Institute of Technology, P.O. Box 3640, 76021 Karlsruhe, Germany</b></td><td>('2357942', 'Bahjat Safadi', 'bahjat safadi')<br/>('1921500', 'Yubing Tong', 'yubing tong')<br/>('1981024', 'Franck Thollard', 'franck thollard')<br/>('40303076', 'Tobias Gehrig', 'tobias gehrig')<br/>('3025777', 'Hazim Kemal Ekenel', 'hazim kemal ekenel')</td><td></td></tr><tr><td>cf86616b5a35d5ee777585196736dfafbb9853b5</td><td>This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. +<br/>Learning Multiscale Active Facial Patches for +<br/>Expression Analysis +</td><td>('29803023', 'Lin Zhong', 'lin zhong')<br/>('1734954', 'Qingshan Liu', 'qingshan liu')<br/>('39606160', 'Peng Yang', 'peng yang')<br/>('1768190', 'Junzhou Huang', 'junzhou huang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td></td></tr><tr><td>cacd51221c592012bf2d9e4894178c1c1fa307ca</td><td> +<br/>ISSN: 2277-3754 +<br/>ISO 9001:2008 Certified +<br/>International Journal of Engineering and Innovative Technology (IJEIT) +<br/>Volume 4, Issue 11, May 2015 +<br/>Face and Expression Recognition Techniques: A +<br/>Review +<br/> +<br/>Advanced Communication & Signal Processing Laboratory, Department of Electronics & Communication +<br/><b>engineering, Government College of Engineering Kannur, Kerala, India</b></td><td>('35135054', 'A. Ranjith Ram', 'a. ranjith ram')</td><td></td></tr><tr><td>ca0363d29e790f80f924cedaf93cb42308365b3d</td><td>Facial Expression Recognition in Image Sequences +<br/>using Geometric Deformation Features and Support +<br/>Vector Machines +<br/><b>yAristotle University of Thessaloniki</b><br/>Department of Informatics +<br/>Box 451 +<br/>54124 Thessaloniki, Greece +</td><td>('1754270', 'Irene Kotsia', 'irene kotsia')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>email: fekotsia,pitasg@aiia.csd.auth.gr +</td></tr><tr><td>cad52d74c1a21043f851ae14c924ac689e197d1f</td><td>From Ego to Nos-vision: +<br/>Detecting Social Relationships in First-Person Views +<br/>Universit`a degli Studi di Modena e Reggio Emilia +<br/>Via Vignolese 905, 41125 Modena - Italy +</td><td>('2452552', 'Stefano Alletto', 'stefano alletto')<br/>('2275344', 'Giuseppe Serra', 'giuseppe serra')<br/>('2175529', 'Simone Calderara', 'simone calderara')<br/>('2059900', 'Francesco Solera', 'francesco solera')<br/>('1741922', 'Rita Cucchiara', 'rita cucchiara')</td><td>{name.surname}@unimore.it +</td></tr><tr><td>cac8bb0e393474b9fb3b810c61efdbc2e2c25c29</td><td></td><td></td><td></td></tr><tr><td>ca54d0a128b96b150baef392bf7e498793a6371f</td><td>Improve Pedestrian Attribute Classification by +<br/>Weighted Interactions from Other Attributes +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences</b></td><td>('1739258', 'Jianqing Zhu', 'jianqing zhu')<br/>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>jianqingzhu@foxmail.com, {scliao, zlei, szli}@cbsr.ia.ac.cn +</td></tr><tr><td>cad24ba99c7b6834faf6f5be820dd65f1a755b29</td><td>Understanding hand-object +<br/>manipulation by modeling the +<br/>contextual relationship between actions, +<br/>grasp types and object attributes +<br/>Journal Title +<br/>XX(X):1–14 +<br/>c(cid:13)The Author(s) 2016 +<br/>Reprints and permission: +<br/>sagepub.co.uk/journalsPermissions.nav +<br/>DOI: 10.1177/ToBeAssigned +<br/>www.sagepub.com/ +</td><td>('3172280', 'Minjie Cai', 'minjie cai')<br/>('37991449', 'Kris M. Kitani', 'kris m. kitani')<br/>('9467266', 'Yoichi Sato', 'yoichi sato')</td><td></td></tr><tr><td>cadba72aa3e95d6dcf0acac828401ddda7ed8924</td><td>THÈSE PRÉSENTÉE À LA FACULTÉ DES SCIENCES +<br/>POUR L’OBTENTION DU GRADE DE DOCTEUR ÈS SCIENCES +<br/>Algorithms and VLSI Architectures +<br/>for Low-Power Mobile Face Verification +<br/>par +<br/>Acceptée sur proposition du jury: +<br/>Prof. F. Pellandini, directeur de thèse +<br/>PD Dr. M. Ansorge, co-directeur de thèse +<br/>Prof. P.-A. Farine, rapporteur +<br/>Dr. C. Piguet, rapporteur +<br/>Soutenue le 2 juin 2005 +<br/>INSTITUT DE MICROTECHNIQUE +<br/>UNIVERSITÉ DE NEUCHÂTEL +<br/>2006 +</td><td>('1844418', 'Jean-Luc Nagel', 'jean-luc nagel')</td><td></td></tr><tr><td>ca37eda56b9ee53610c66951ee7ca66a35d0a846</td><td>Semantic Concept Discovery for Large-Scale Zero-Shot Event Detection +<br/><b>Centre for Quantum Computation and Intelligent Systems, University of Technology Sydney</b><br/><b>Language Technologies Institute, Carnegie Mellon University</b><br/><b>Carnegie Mellon University</b></td><td>('1729163', 'Xiaojun Chang', 'xiaojun chang')<br/>('39033919', 'Yi Yang', 'yi yang')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')<br/>('1752601', 'Eric P. Xing', 'eric p. xing')</td><td>{cxj273, yee.i.yang}@gmail.com, {alex, epxing, yaoliang}@cs.cmu.edu +</td></tr><tr><td>ca606186715e84d270fc9052af8500fe23befbda</td><td>Using Subclass Discriminant Analysis, Fuzzy Integral and Symlet Decomposition for +<br/>Face Recognition +<br/>Department of Electrical Engineering, +<br/>Iran Univ. of Science and Technology, +<br/>Narmak, Tehran, Iran +<br/>Department of Electrical Engineering, +<br/>Iran Univ. of Science and Technology, +<br/>Department of Electrical Engineering, +<br/>Iran Univ. of Science and Technology, +<br/>Narmak, Tehran, Iran +<br/>Narmak, Tehran, Iran +</td><td>('9267982', 'Seyed Mohammad Seyedzade', 'seyed mohammad seyedzade')<br/>('2532375', 'Sattar Mirzakuchaki', 'sattar mirzakuchaki')<br/>('2535533', 'Amir Tahmasbi', 'amir tahmasbi')</td><td>Email: sm.seyedzade@ieee.org +<br/>Email: m_kuchaki@iust.ac.ir +<br/>Email: a.tahmasbi@ieee.org +</td></tr><tr><td>e48fb3ee27eef1e503d7ba07df8eb1524c47f4a6</td><td>Illumination invariant face recognition and impostor rejection +<br/>using different MINACE filter algorithms +<br/><b>Carnegie Mellon University, Pittsburgh, PA</b></td><td>('8142777', 'Rohit Patnaik', 'rohit patnaik')<br/>('34925745', 'David Casasent', 'david casasent')</td><td></td></tr><tr><td>e4bf70e818e507b54f7d94856fecc42cc9e0f73d</td><td>IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 +<br/>FACE RECOGNITION UNDER VARYING BLUR IN AN +<br/>UNCONSTRAINED ENVIRONMENT +<br/><b>M.Tech, Information Technology, Madras Institute of Technology, TamilNadu, India</b><br/><b>Information Technology, Madras Institute of Technology, TamilNadu, India, email</b></td><td></td><td>anubhapearl@gmail.com +<br/>hemalatha.ch@gmail.com +</td></tr><tr><td>e4bc529ced68fae154e125c72af5381b1185f34e</td><td>PERCEPTUAL GOAL SPECIFICATIONS FOR REINFORCEMENT LEARNING +<br/>A Thesis Proposal +<br/>Presented to +<br/>The Academic Faculty +<br/>by +<br/>In Partial Fulfillment +<br/>of the Requirements for the Degree +<br/>Doctor of Philosophy in the +<br/>School of Interactive Computing +<br/><b>Georgia Institute of Technology</b><br/>November 2017 +</td><td>('12313871', 'Ashley D. Edwards', 'ashley d. edwards')</td><td></td></tr><tr><td>e465f596d73f3d2523dbf8334d29eb93a35f6da0</td><td></td><td></td><td></td></tr><tr><td>e4aeaf1af68a40907fda752559e45dc7afc2de67</td><td></td><td></td><td></td></tr><tr><td>e4c3d5d43cb62ac5b57d74d55925bdf76205e306</td><td></td><td></td><td></td></tr><tr><td>e42998bbebddeeb4b2bedf5da23fa5c4efc976fa</td><td>Generic Active Appearance Models Revisited +<br/><b>Imperial College London, United Kingdom</b><br/><b>School of Computer Science, University of Lincoln, United Kingdom</b><br/><b>Faculty of Electrical Engineering, Mathematics and Computer Science, University</b><br/>of Twente, The Netherlands +</td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('2575567', 'Joan Alabort-i-Medina', 'joan alabort-i-medina')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{gt204, ja310, s.zafeiriou, m.pantic}@imperial.ac.uk +</td></tr><tr><td>e4a1b46b5c639d433d21b34b788df8d81b518729</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>Side Information for Face Completion: a Robust +<br/>PCA Approach +</td><td>('4091869', 'Niannan Xue', 'niannan xue')<br/>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('1902288', 'Shiyang Cheng', 'shiyang cheng')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td></td></tr><tr><td>e4c81c56966a763e021938be392718686ba9135e</td><td></td><td></td><td>3,100+OPEN ACCESS BOOKS103,000+INTERNATIONALAUTHORS AND EDITORS106+ MILLIONDOWNLOADSBOOKSDELIVERED TO151 COUNTRIESAUTHORS AMONGTOP 1%MOST CITED SCIENTIST12.2%AUTHORS AND EDITORSFROM TOP 500 UNIVERSITIESSelection of our books indexed in theBook Citation Index in Web of Science™Core Collection (BKCI)Chapter from the book Visual Cortex - Current Status and PerspectivesDownloaded from: http://www.intechopen.com/books/visual-cortex-current-status-and-perspectivesPUBLISHED BYWorld's largest Science,Technology & Medicine Open Access book publisherInterested in publishing with InTechOpen?Contact us at book.department@intechopen.com </td></tr><tr><td>e4e95b8bca585a15f13ef1ab4f48a884cd6ecfcc</td><td>Face Recognition with Independent Component Based +<br/>Super-resolution +<br/>aFaculty of Engineering and Natural Sciences, Sabanci Univ., Istanbul, Turkiye, 34956 +<br/>bSchool of Elec. and Comp. Eng. , Georgia Inst. of Tech., Atlanta, GA, USA, 30332-0250 +</td><td>('1844879', 'Osman Gokhan Sezer', 'osman gokhan sezer')<br/>('3975060', 'Yucel Altunbasak', 'yucel altunbasak')<br/>('31849282', 'Aytul Ercil', 'aytul ercil')</td><td></td></tr><tr><td>e4df83b7424842ff5864c10fa55d38eae1c45fac</td><td>Hindawi Publishing Corporation +<br/>Discrete Dynamics in Nature and Society +<br/>Volume 2009, Article ID 916382, 8 pages +<br/>doi:10.1155/2009/916382 +<br/>Research Article +<br/>Locally Linear Discriminate Embedding for +<br/>Face Recognition +<br/><b>Faculty of Information Science and Technology, Multimedia University, 75450 Melaka, Malaysia</b><br/>Received 21 January 2009; Accepted 12 October 2009 +<br/>Recommended by B. Sagar +<br/>A novel method based on the local nonlinear mapping is presented in this research. The method +<br/>is called Locally Linear Discriminate Embedding (cid:2)LLDE(cid:3). LLDE preserves a local linear structure +<br/>of a high-dimensional space and obtains a compact data representation as accurately as possible +<br/>in embedding space (cid:2)low dimensional(cid:3) before recognition. For computational simplicity and fast +<br/>processing, Radial Basis Function (cid:2)RBF(cid:3) classifier is integrated with the LLDE. RBF classifier +<br/>is carried out onto low-dimensional embedding with reference to the variance of the data. To +<br/>validate the proposed method, CMU-PIE database has been used and experiments conducted in +<br/>this research revealed the efficiency of the proposed methods in face recognition, as compared to +<br/>the linear and non-linear approaches. +<br/>the Creative Commons Attribution License, which permits unrestricted use, distribution, and +<br/>reproduction in any medium, provided the original work is properly cited. +<br/>1. Introduction +<br/>Linear subspace analysis has been extensively applied to face recognition. A successful face +<br/>recognition methodology is largely dependent on the particular choice of features used by +<br/>the classifier. Linear methods are easy to understand and are very simple to implement, but +<br/>the linearity assumption does not hold in many real-world scenarios. Face appearance lies in +<br/>a high-dimensional nonlinear manifold. A disadvantage of the linear techniques is that they +<br/>fail to capture the characteristics of the nonlinear appearance manifold. This is due to the +<br/>fact that the linear methods extract features only from the input space without considering +<br/>the nonlinear information between the components of the input data. However, a globally +<br/>nonlinear mapping can often be approximated using a linear mapping in a local region. This +<br/>has motivated the design of the nonlinear mapping methods in this study. +<br/>The history of the nonlinear mapping is long; it can be traced back to Sammon’s +<br/>mapping in 1969 (cid:5)1(cid:6). Over time, different techniques have been proposed such as the +<br/>projection pursuit (cid:5)2(cid:6), the projection pursuit regression (cid:5)3(cid:6), self-organizing maps or SOM +</td><td>('2008201', 'Eimad E. Abusham', 'eimad e. abusham')<br/>('32191265', 'E. K. Wong', 'e. k. wong')<br/>('32191265', 'E. K. Wong', 'e. k. wong')</td><td>Correspondence should be addressed to Eimad E. Abusham, eimad.eldin@mmu.edu.my +</td></tr><tr><td>e4e3faa47bb567491eaeaebb2213bf0e1db989e1</td><td>Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-16) +<br/>Empirical Risk Minimization for Metric +<br/>Learning Using Privileged Information +<br/><b>School of Computer and Information, Hefei University of Technology, China</b><br/><b>Centre for Quantum Computation and Intelligent Systems, FEIT, University of Technology Sydney, Australia</b></td><td>('2028727', 'Xun Yang', 'xun yang')<br/>('15970836', 'Meng Wang', 'meng wang')<br/>('1763785', 'Luming Zhang', 'luming zhang')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td>{hfutyangxun, eric.mengwang, zglumg}@gmail.com; +<br/>dacheng.tao@uts.edu.au; +</td></tr><tr><td>e43ea078749d1f9b8254e0c3df4c51ba2f4eebd5</td><td>Facial Expression Recognition Based on Constrained +<br/>Local Models and Support Vector Machines +</td><td>('1901962', 'Nikolay Neshov', 'nikolay neshov')<br/>('34945173', 'Ivo Draganov', 'ivo draganov')<br/>('1750280', 'Agata Manolova', 'agata manolova')</td><td></td></tr><tr><td>e476cbcb7c1de73a7bcaeab5d0d59b8b3c4c1cbf</td><td></td><td></td><td></td></tr><tr><td>e4c2f8e4aace8cb851cb74478a63d9111ca550ae</td><td>DISTRIBUTED ONE-CLASS LEARNING +<br/><b>cid:63)Queen Mary University of London, Imperial College London</b></td><td>('9920557', 'Ali Shahin Shamsabadi', 'ali shahin shamsabadi')<br/>('1763096', 'Hamed Haddadi', 'hamed haddadi')<br/>('1713138', 'Andrea Cavallaro', 'andrea cavallaro')</td><td></td></tr><tr><td>e475e857b2f5574eb626e7e01be47b416deff268</td><td>Facial Emotion Recognition Using Nonparametric +<br/>Weighted Feature Extraction and Fuzzy Classifier +</td><td>('2121174', 'Maryam Imani', 'maryam imani')<br/>('1801348', 'Gholam Ali Montazer', 'gholam ali montazer')</td><td></td></tr><tr><td>e4391993f5270bdbc621b8d01702f626fba36fc2</td><td>Author manuscript, published in "18th Scandinavian Conference on Image Analysis (2013)" +<br/> DOI : 10.1007/978-3-642-38886-6_31 +</td><td></td><td></td></tr><tr><td>e43045a061421bd79713020bc36d2cf4653c044d</td><td>A New Representation of Skeleton Sequences for 3D Action Recognition +<br/><b>The University of Western Australia</b><br/><b>Murdoch University</b></td><td>('2796959', 'Qiuhong Ke', 'qiuhong ke')<br/>('1698675', 'Mohammed Bennamoun', 'mohammed bennamoun')<br/>('1782428', 'Senjian An', 'senjian an')</td><td>qiuhong.ke@research.uwa.edu.au +<br/>{mohammed.bennamoun,senjian.an,farid.boussaid}@uwa.edu.au +<br/>f.sohel@murdoch.edu.au +</td></tr><tr><td>e4d8ba577cabcb67b4e9e1260573aea708574886</td><td>UM SISTEMA DE RECOMENDAC¸ ˜AO INTELIGENTE BASEADO EM V´IDIO +<br/>AULAS PARA EDUCAC¸ ˜AO A DIST ˆANCIA +<br/>Gaspare Giuliano Elias Bruno +<br/>Tese de Doutorado apresentada ao Programa +<br/>de P´os-gradua¸c˜ao em Engenharia de Sistemas e +<br/>Computa¸c˜ao, COPPE, da Universidade Federal +<br/>do Rio de Janeiro, como parte dos requisitos +<br/>necess´arios `a obten¸c˜ao do t´ıtulo de Doutor em +<br/>Engenharia de Sistemas e Computa¸c˜ao. +<br/>Orientadores: Edmundo Albuquerque de +<br/>Souza e Silva +<br/>Rosa Maria Meri Le˜ao +<br/>Rio de Janeiro +<br/>Janeiro de 2016 +</td><td></td><td></td></tr><tr><td>e475deadd1e284428b5e6efd8fe0e6a5b83b9dcd</td><td>Accepted in Pattern Recognition Letters +<br/>Pattern Recognition Letters +<br/>journal homepage: www.elsevier.com +<br/>Are you eligible? Predicting adulthood from face images via class specific mean +<br/>autoencoder +<br/>IIIT-Delhi, New Delhi, 110020, India +<br/>Article history: +<br/>Received 15 March 2017 +</td><td>('2220719', 'Maneet Singh', 'maneet singh')<br/>('1925017', 'Shruti Nagpal', 'shruti nagpal')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('39129417', 'Richa Singh', 'richa singh')</td><td></td></tr><tr><td>e4abc40f79f86dbc06f5af1df314c67681dedc51</td><td>Head Detection with Depth Images in the Wild +<br/>Department of Engineering ”Enzo Ferrari” +<br/><b>University of Modena and Reggio Emilia, Italy</b><br/>Keywords: +<br/>Head Detection, Head Localization, Depth Maps, Convolutional Neural Network +</td><td>('6125279', 'Diego Ballotta', 'diego ballotta')<br/>('12010968', 'Guido Borghi', 'guido borghi')<br/>('1723285', 'Roberto Vezzani', 'roberto vezzani')<br/>('1741922', 'Rita Cucchiara', 'rita cucchiara')</td><td>{name.surname}@unimore.it +</td></tr><tr><td>e4d0e87d0bd6ead4ccd39fc5b6c62287560bac5b</td><td>Implicit Video Multi-Emotion Tagging by Exploiting Multi-Expression +<br/>Relations +</td><td>('1771215', 'Zhilei Liu', 'zhilei liu')<br/>('1791319', 'Shangfei Wang', 'shangfei wang')<br/>('3558606', 'Zhaoyu Wang', 'zhaoyu wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td></td></tr><tr><td>e48e94959c4ce799fc61f3f4aa8a209c00be8d7f</td><td>Hindawi Publishing Corporation +<br/>The Scientific World Journal +<br/>Volume 2013, Article ID 135614, 6 pages +<br/>http://dx.doi.org/10.1155/2013/135614 +<br/>Research Article +<br/>Design of an Efficient Real-Time Algorithm Using Reduced +<br/>Feature Dimension for Recognition of Speed Limit Signs +<br/><b>Sogang University, Seoul 121-742, Republic of Korea</b><br/>2 Samsung Techwin R&D Center, Security Solution Division, 701 Sampyeong-dong, Bundang-gu, Seongnam-si, +<br/>Gyeonggi 463-400, Republic of Korea +<br/>Received 28 August 2013; Accepted 1 October 2013 +<br/>Academic Editors: P. Daponte, M. Nappi, and N. Nishchal +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>We propose a real-time algorithm for recognition of speed limit signs from a moving vehicle. Linear Discriminant Analysis (LDA) +<br/>required for classification is performed by using Discrete Cosine Transform (DCT) coefficients. To reduce feature dimension in +<br/>LDA, DCT coefficients are selected by a devised discriminant function derived from information obtained by training. Binarization +<br/>and thinning are performed on a Region of Interest (ROI) obtained by preprocessing a detected ROI prior to DCT for further +<br/>reduction of computation time in DCT. This process is performed on a sequence of image frames to increase the hit rate of +<br/>recognition. Experimental results show that arithmetic operations are reduced by about 60%, while hit rates reach about 100% +<br/>compared to previous works. +<br/>1. Introduction +<br/>Driver safety is the main concern of the advanced vehicle +<br/>system which became implementable due to the develop- +<br/>ment of the autonomous driving, automatic control, and +<br/>imaging technology. An advanced vehicle system gives driver +<br/>information related to safety by sensing the surroundings +<br/>automatically [1]. Speed limit signs recognition is regarded +<br/>to be helpful in safety for drivers using advanced vehicle +<br/>system. The system needs to recognize the speed limit sign +<br/>in the distance quickly and accurately in order to give +<br/>the driver precaution in time since vehicle is moving fast. +<br/>But existing algorithms perform recognition by using many +<br/>features extracted from captured image, requiring a large +<br/>amount of arithmetic operations for classification [2]. +<br/>Several classification algorithms have been proposed, +<br/>which include Neural Networks [2, 3], Support Vector +<br/>Machine (SVM) [2], and Linear Discriminant Analysis +<br/>(LDA) [2, 4]. Among these, SVM has relatively higher recog- +<br/>nition rate, and LDA is used in many classification applica- +<br/>tions due to its low computational complexity. However, its +<br/>computational complexity needs to be further reduced to be +<br/>used in real-time application. It can be achieved by reducing +<br/>the number of inputs of LDA. +<br/>This paper proposes an efficient real-time algorithm for +<br/>recognition of speed limit signs by using reduced feature +<br/>dimension. In this research study, DCT is employed and parts +<br/>of Discrete Cosine Transform (DCT) coefficients are used as +<br/>inputs to LDA instead of features extracted from image. DCT +<br/>coefficients are selected by a devised discriminant function. +<br/>To further reduce DCT computation time, binarization and +<br/>thinning are applied to the detected Region of Interest (ROI). +<br/>Image of speed limit sign in the distance obtained from cam- +<br/>era has a low resolution and it gives poor rate of recognition. +<br/>To resolve this problem, this paper proposes a recognition +<br/>system using classification results on a sequence of frames. +<br/>It can enhance hit rate of recognition by accumulating the +<br/>probability of single frame recognition. +<br/>2. Background +<br/>In this section, LDA is briefly described, which is popularly +<br/>employed for classification. LDA is a classical statistical +</td><td>('2012225', 'Hanmin Cho', 'hanmin cho')<br/>('5984008', 'Seungwha Han', 'seungwha han')<br/>('6348959', 'Sun-Young Hwang', 'sun-young hwang')<br/>('2012225', 'Hanmin Cho', 'hanmin cho')</td><td>Correspondence should be addressed to Sun-Young Hwang; hwang@sogang.ac.kr +</td></tr><tr><td>e496d6be415038de1636bbe8202cac9c1cea9dbe</td><td>Facial Expression Recognition in Older Adults using +<br/>Deep Machine Learning +<br/><b>National Research Council of Italy, Institute for Microelectronics and Microsystems, Lecce</b><br/>Italy +</td><td>('2886068', 'Andrea Caroppo', 'andrea caroppo')<br/>('1796761', 'Alessandro Leone', 'alessandro leone')<br/>('1737181', 'Pietro Siciliano', 'pietro siciliano')</td><td>{andrea.caroppo,alessandro.leone,pietro.siciliano}@le.imm.cnr.it +</td></tr><tr><td>e43cc682453cf3874785584fca813665878adaa7</td><td>www.ijecs.in +<br/>International Journal Of Engineering And Computer Science ISSN:2319-7242 +<br/>Volume 3 Issue 10 October, 2014 Page No.8830-8834 +<br/>Face Recognition using Local Derivative Pattern Face +<br/>Descriptor +<br/>Department of Electronics and Telecommunication +<br/><b>Datta Meghe College of Engineering</b><br/>Airoli, Navi Mumbai, India 1,2 +<br/>Mob: 99206746061 +<br/>Mob: 99870353142 +</td><td></td><td>pranitachavan42@gmail.com 1 +<br/>djpethe@gmail.com 2 +</td></tr><tr><td>fec6648b4154fc7e0892c74f98898f0b51036dfe</td><td>A Generic Face Processing +<br/>Framework: Technologies, +<br/>Analyses and Applications +<br/>A Thesis Submitted in Partial Ful(cid:12)lment +<br/>of the Requirements for the Degree of +<br/>Master of Philosophy +<br/>in +<br/>Computer Science and Engineering +<br/>Supervised by +<br/><b>c(cid:13)The Chinese University of Hong Kong</b><br/>July 2003 +<br/><b>The Chinese University of Hong Kong holds the copyright of this thesis. Any</b><br/>person(s) intending to use a part or whole of the materials in the thesis in +<br/>a proposed publication must seek copyright release from the Dean of the +<br/>Graduate School. +</td><td>('1681775', 'Michael R. Lyu', 'michael r. lyu')</td><td></td></tr><tr><td>fea0a5ed1bc83dd1b545a5d75db2e37a69489ac9</td><td>Enhancing Recommender Systems for TV by Face Recognition +<br/><b>iMinds - Ghent University, Technologiepark 15, B-9052 Ghent, Belgium</b><br/>Keywords: +<br/>Recommender System, Face Recognition, Face Detection, TV, Emotion Detection. +</td><td>('1738833', 'Toon De Pessemier', 'toon de pessemier')<br/>('3441798', 'Damien Verlee', 'damien verlee')<br/>('1698239', 'Luc Martens', 'luc martens')</td><td>{toon.depessemier, luc.martens}@intec.ugent.be +</td></tr><tr><td>fe9c460d5ca625402aa4d6dd308d15a40e1010fa</td><td>Neural Architecture for Temporal Emotion +<br/>Classification +<br/>Universit¨at Ulm, Neuroinformatik, Germany +</td><td>('1681327', 'Roland Schweiger', 'roland schweiger')<br/>('2331203', 'Pierre Bayerl', 'pierre bayerl')<br/>('1706025', 'Heiko Neumann', 'heiko neumann')</td><td>froland.schweiger,pierre.bayerl,heiko.neumanng@informatik.uni-ulm.de +</td></tr><tr><td>fe7e3cc1f3412bbbf37d277eeb3b17b8b21d71d5</td><td>IOSR Journal of VLSI and Signal Processing (IOSR-JVSP) +<br/>Volume 6, Issue 2, Ver. I (Mar. -Apr. 2016), PP 47-53 +<br/>e-ISSN: 2319 – 4200, p-ISSN No. : 2319 – 4197 +<br/>www.iosrjournals.org +<br/>Performance Evaluation of Gabor Wavelet Features for Face +<br/>Representation and Recognition +<br/><b>Bapuji Institute of Engineering and Technology Davanagere, Karnataka, India</b><br/><b>University B.D.T.College of Engineering, Visvesvaraya</b><br/><b>Technological University, Davanagere, Karnataka, India</b></td><td>('2038371', 'M. E. Ashalatha', 'm. e. ashalatha')<br/>('3283067', 'Mallikarjun S. Holi', 'mallikarjun s. holi')</td><td></td></tr><tr><td>fe464b2b54154d231671750053861f5fd14454f5</td><td>Multi Joint Action in CoTeSys +<br/>- Setup and Challenges - +<br/>Technical report CoTeSys-TR-10-01 +<br/>D. Brˇsˇci´c, F. Rohrm¨uller, O. Kourakos, S. Sosnowski, D. Althoff, M. Lawitzky, +<br/>{drazen, rohrm, omirosk, sosnowski, dalthoff, lawitzky, moertl, rambow, vicky, +<br/>M. Eggers, C. Mayer, T. Kruse, A. Kirsch, M. Beetz and B. Radig 2 +<br/>T. Lorenz and A. Schub¨o 4 +<br/>P. Basili and S. Glasauer 5 +<br/>W. Maier and E. Steinbach 7 +<br/><b>Institute of Automatic Control</b><br/>4 Experimental Psychology Unit +<br/>Engineering +<br/>Department of Psychology +<br/>Department of Electrical Engineering +<br/>Ludwig-Maximilians-Universit¨at +<br/>and Information Technology +<br/>Technische Universit¨at M¨unchen +<br/>Arcisstraße 21, 80333 M¨unchen +<br/>2Intelligent Autonomous Systems +<br/>Department of Informatics +<br/>M¨unchen +<br/>Leopoldstraße 13, 80802 M¨unchen +<br/>5Center for Sensorimotor Research +<br/>Clinical Neurosciences and +<br/>Department of Neurology +<br/>Technische Universit¨at M¨unchen +<br/>Ludwig-Maximilians-Universit¨at +<br/>Boltzmannstraße 3, 85748 Garching +<br/>M¨unchen +<br/>bei M¨unchen +<br/>Marchionistraße 23, 81377 M¨unchen +<br/><b>Institute for Human-Machine</b><br/>6Robotics and Embedded Systems +<br/>Communication +<br/>Department of Informatics +<br/>Department of Electrical Engineering +<br/>Technische Universit¨at M¨unchen +<br/>and Information Technology +<br/>Boltzmannstraße 3, 85748 Garching +<br/>Technische Universit¨at M¨unchen +<br/>Arcisstraße 21, 80333 M¨unchen +<br/>bei M¨unchen +<br/><b>Institute for Media Technology</b><br/>Department of Electrical Engineering +<br/>and Information Technology +<br/>Technische Universit¨at M¨unchen +<br/>Arcisstraße 21, 80333 M¨unchen +</td><td>('46953125', 'X. Zang', 'x. zang')<br/>('47824592', 'W. Wang', 'w. wang')<br/>('48172476', 'A. Bannat', 'a. bannat')<br/>('30849638', 'G. Panin', 'g. panin')</td><td>medina, xueliang zang, wangwei, dirk, kuehnlen, hirche, buss}@lsr.ei.tum.de +<br/>{eggers, mayerc, kruset, kirsch, beetz, radig}@in.tum.de +<br/>{blume, bannat, rehrl, wallhoff}@tum.de +<br/>{lorenz, schuboe}@psy.lmu.de +<br/>{p.basili,s.glasauer}@lrz.uni-muenchen.de +<br/>{lenz,roeder,panin,knoll}@in.tum.de +<br/>{werner.maier, eckehard.steinbach}@tum.de +</td></tr><tr><td>fe7c0bafbd9a28087e0169259816fca46db1a837</td><td></td><td></td><td></td></tr><tr><td>fe5df5fe0e4745d224636a9ae196649176028990</td><td><b>University of Massachusetts - Amherst</b><br/>Dissertations +<br/>9-1-2010 +<br/>Dissertations and Theses +<br/>Using Context to Enhance the Understanding of +<br/>Face Images +<br/>Follow this and additional works at: http://scholarworks.umass.edu/open_access_dissertations +<br/>Recommended Citation +<br/>Jain, Vidit, "Using Context to Enhance the Understanding of Face Images" (2010). Dissertations. Paper 287. +</td><td>('2246870', 'Vidit Jain', 'vidit jain')</td><td>ScholarWorks@UMass Amherst +<br/>University of Massachusetts - Amherst, vidit.jain@gmail.com +<br/>This Open Access Dissertation is brought to you for free and open access by the Dissertations and Theses at ScholarWorks@UMass Amherst. It has +<br/>been accepted for inclusion in Dissertations by an authorized administrator of ScholarWorks@UMass Amherst. For more information, please contact +<br/>scholarworks@library.umass.edu. +</td></tr><tr><td>fe961cbe4be0a35becd2d722f9f364ec3c26bd34</td><td>Computer-based Tracking, Analysis, and Visualization of Linguistically +<br/>Significant Nonmanual Events in American Sign Language (ASL) +<br/><b>Boston University / **Rutgers University / ***Gallaudet University</b><br/><b>Boston University, Linguistics Program, 621 Commonwealth Avenue, Boston, MA</b><br/><b>Rutgers University, Computer and Information Sciences, 110 Frelinghuysen Road, Piscataway, NJ</b><br/><b>Gallaudet University, Technology Access Program, 800 Florida Ave NE, Washington, DC</b></td><td>('1732359', 'Carol Neidle', 'carol neidle')<br/>('38079056', 'Jingjing Liu', 'jingjing liu')<br/>('39132952', 'Bo Liu', 'bo liu')<br/>('4340744', 'Xi Peng', 'xi peng')<br/>('2467082', 'Christian Vogler', 'christian vogler')<br/>('1711560', 'Dimitris Metaxas', 'dimitris metaxas')</td><td>E-mail: carol@bu.edu, jl1322@cs.rutgers.edu, lb507@cs.rutgers.edu, px13@cs.rutgers.edu, +<br/>christian.vogler@gallaudet.edu, dnm@ cs.rutgers.edu +</td></tr><tr><td>feb6e267923868bff6e2108603d00fdfd65251ca</td><td>February 1, 2013 15:16 WSPC/INSTRUCTION FILE +<br/>S0218213012500297 +<br/>International Journal on Artificial Intelligence Tools +<br/>Vol. 22, No. 1 (2013) 1250029 (30 pages) +<br/>c(cid:13) World Scientific Publishing Company +<br/>DOI: 10.1142/S0218213012500297 +<br/>UNSUPERVISED DISCOVERY OF VISUAL FACE CATEGORIES +<br/><b>Institute of Systems Engineering, Southeast University, Nanjing, China</b><br/><b>University of Nevada, Reno, USA</b><br/><b>College of Computer and Information Sciences</b><br/><b>King Saud University, Riyadh 11543, Saudi Arabia</b><br/><b>College of Computer and Information Sciences</b><br/><b>King Saud University, P.O. Box 51178, Riyadh 11543, Saudi Arabia</b><br/>GHULAM MUHAMMAD +<br/><b>College of Computer and Information Sciences</b><br/><b>King Saud University, Riyadh 11543, Saudi Arabia</b><br/>Received 30 January 2012 +<br/>Accepted 10 May 2012 +<br/>Published +<br/>Human faces can be arranged into different face categories using information from common visual +<br/>cues such as gender, ethnicity, and age. It has been demonstrated that using face categorization as a +<br/>precursor step to face recognition improves recognition rates and leads to more graceful errors.1 +<br/>Although face categorization using common visual cues yields meaningful face categories, +<br/>developing accurate and robust gender, ethnicity, and age categorizers is a challenging issue. +<br/>Moreover, it limits the overall number of possible face categories and, in practice, yields unbalanced +<br/>face categories which can compromise recognition performance. This paper investigates ways to +<br/>automatically discover a categorization of human faces from a collection of unlabeled face images +<br/>without relying on predefined visual cues. Specifically, given a set of face images from a group of +<br/>known individuals (i.e., gallery set), our goal is finding ways to robustly partition the gallery set +<br/>(i.e., face categories). The objective is being able to assign novel images of the same individuals +<br/>(i.e., query set) to the correct face category with high accuracy and robustness. To address the issue +<br/>of face category discovery, we represent faces using local features and apply unsupervised learning +<br/>(i.e., clustering). To categorize faces in novel images, we employ nearest-neighbor algorithms +<br/>1250029-1 +</td><td>('2884262', 'Shicai Yang', 'shicai yang')<br/>('1808451', 'George Bebis', 'george bebis')<br/>('2363759', 'Muhammad Hussain', 'muhammad hussain')<br/>('39344692', 'Anwar M. Mirza', 'anwar m. mirza')</td><td>shicai.yang@gmail.com +<br/>bebis@cse.unr.edu +<br/>mhussain@ksu.edu.sa +<br/>ghulam@ksu.edu.sa +<br/>anwar.m.mirza@gmail.com +</td></tr><tr><td>fe48f0e43dbdeeaf4a03b3837e27f6705783e576</td><td></td><td></td><td></td></tr><tr><td>fea83550a21f4b41057b031ac338170bacda8805</td><td>Learning a Metric Embedding +<br/>for Face Recognition +<br/>using the Multibatch Method +<br/>Orcam Ltd., Jerusalem, Israel +</td><td>('46273386', 'Oren Tadmor', 'oren tadmor')<br/>('1743988', 'Yonatan Wexler', 'yonatan wexler')<br/>('31601132', 'Tal Rosenwein', 'tal rosenwein')<br/>('2554670', 'Shai Shalev-Shwartz', 'shai shalev-shwartz')<br/>('3140335', 'Amnon Shashua', 'amnon shashua')</td><td>firstname.lastname@orcam.com +</td></tr><tr><td>feeb0fd0e254f38b38fe5c1022e84aa43d63f7cc</td><td>EURECOM +<br/>Multimedia Communications Department +<br/>and +<br/>Mobile Communications Department +<br/>2229, route des Crˆetes +<br/>B.P. 193 +<br/>06904 Sophia-Antipolis +<br/>FRANCE +<br/>Research Report RR-11-255 +<br/>Search Pruning with Soft Biometric Systems: +<br/>Efficiency-Reliability Tradeoff +<br/>June 1st, 2011 +<br/>Last update June 1st, 2011 +<br/>1EURECOM’s research is partially supported by its industrial members: BMW Group, Cisco, +<br/>Monaco Telecom, Orange, SAP, SFR, Sharp, STEricsson, Swisscom, Symantec, Thales. +</td><td>('3299530', 'Antitza Dantcheva', 'antitza dantcheva')<br/>('15758502', 'Arun Singh', 'arun singh')<br/>('1688531', 'Petros Elia', 'petros elia')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td></td></tr><tr><td>fe108803ee97badfa2a4abb80f27fa86afd9aad9</td><td></td><td></td><td></td></tr><tr><td>fe0c51fd41cb2d5afa1bc1900bbbadb38a0de139</td><td>Rahman et al. EURASIP Journal on Image and Video Processing (2015) 2015:35 +<br/>DOI 10.1186/s13640-015-0090-5 +<br/>RESEARCH +<br/>Open Access +<br/>Bayesian face recognition using 2D +<br/>Gaussian-Hermite moments +</td><td>('47081388', 'S. M. Mahbubur Rahman', 's. m. mahbubur rahman')<br/>('2021126', 'Tamanna Howlader', 'tamanna howlader')</td><td></td></tr><tr><td>c8db8764f9d8f5d44e739bbcb663fbfc0a40fb3d</td><td>Modeling for part-based visual object +<br/>detection based on local features +<br/>Von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +<br/>der Rheinisch-Westf¨alischen Technischen Hochschule Aachen +<br/>zur Erlangung des akademischen Grades eines Doktors +<br/>der Ingenieurwissenschaften genehmigte Dissertation +<br/>vorgelegt von +<br/>Diplom-Ingenieur +<br/>aus Neuss +<br/>Berichter: +<br/>Univ.-Prof. Dr.-Ing. Jens-Rainer Ohm +<br/>Univ.-Prof. Dr.-Ing. Til Aach +<br/>Tag der m¨undlichen Pr¨ufung: 28. September 2011 +<br/>Diese Dissertation ist auf den Internetseiten der +<br/>Hochschulbibliothek online verf¨ugbar. +</td><td>('2447988', 'Mark Asbach', 'mark asbach')</td><td></td></tr><tr><td>c86e6ed734d3aa967deae00df003557b6e937d3d</td><td>Generative Adversarial Networks with +<br/>Decoder-Encoder Output Noise +<br/>conditional distribution of their neighbors. In [32], Portilla and +<br/>Simoncelli proposed a parametric texture model based on joint +<br/>statistics, which uses a decomposition method that is called +<br/>steerable pyramid decomposition to decompose the texture +<br/>of images. An example-based super-resolution algorithm [11] +<br/>was proposed in 2002, which uses a Markov network to model +<br/>the spatial relationship between the pixels of an image. A +<br/>scene completion algorithm [16] was proposed in 2007, which +<br/>applied a semantic scene match technique. These traditional +<br/>algorithms can be applied to particular image generation tasks, +<br/>such as texture synthesis and super-resolution. Their common +<br/>characteristic is that they predict the images pixel by pixel +<br/>rather than generate an image as a whole, and the basic idea +<br/>of them is to make an interpolation according to the existing +<br/>part of the images. Here, the problem is, given a set of images, +<br/>can we generate totally new images with the same distribution +<br/>of the given ones? +</td><td>('2421012', 'Guoqiang Zhong', 'guoqiang zhong')<br/>('46874300', 'Wei Gao', 'wei gao')<br/>('3142351', 'Yongbin Liu', 'yongbin liu')<br/>('47796538', 'Youzhao Yang', 'youzhao yang')</td><td></td></tr><tr><td>c8a4b4fe5ff2ace9ab9171a9a24064b5a91207a3</td><td>LOCATING FACIAL LANDMARKS WITH BINARY MAP CROSS-CORRELATIONS +<br/>J´er´emie Nicolle +<br/>K´evin Bailly +<br/>Univ. Pierre & Marie Curie, ISIR - CNRS UMR 7222, F-75005, Paris - France +</td><td>('3074790', 'Vincent Rapp', 'vincent rapp')<br/>('1680828', 'Mohamed Chetouani', 'mohamed chetouani')</td><td>{nicolle, bailly, rapp, chetouani}@isir.upmc.fr +</td></tr><tr><td>c87f7ee391d6000aef2eadb49f03fc237f4d1170</td><td>1 +<br/>A real-time and unsupervised face Re-Identification system for Human-Robot +<br/>Interaction +<br/><b>Intelligent Behaviour Understanding Group, Imperial College London, London, UK</b><br/>A B S T R A C T +<br/>In the context of Human-Robot Interaction (HRI), face Re-Identification (face Re-ID) aims to verify if certain detected faces have already been +<br/>observed by robots. The ability of distinguishing between different users is crucial in social robots as it will enable the robot to tailor the interaction +<br/>strategy toward the users’ individual preferences. So far face recognition research has achieved great success, however little attention has been paid +<br/>to the realistic applications of Face Re-ID in social robots. In this paper, we present an effective and unsupervised face Re-ID system which +<br/>simultaneously re-identifies multiple faces for HRI. This Re-ID system employs Deep Convolutional Neural Networks to extract features, and an +<br/>online clustering algorithm to determine the face’s ID. Its performance is evaluated on two datasets: the TERESA video dataset collected by the +<br/>TERESA robot, and the YouTube Face Dataset (YTF Dataset). We demonstrate that the optimised combination of techniques achieves an overall +<br/>93.55% accuracy on TERESA dataset and an overall 90.41% accuracy on YTF dataset. We have implemented the proposed method into a software +<br/>module in the HCI^2 Framework [1] for it to be further integrated into the TERESA robot [2], and has achieved real-time performance at 10~26 +<br/>Frames per second. +<br/>Keywords: Real-Time Face Re-Identification, Open Set Re-ID, Multiple Re-ID, Human-Robot Interaction, CNN Descriptors, Online Clustering +<br/>1. Introduction +<br/>Face recognition problem is one of the oldest topics in +<br/>Computer Vision [3]. Recently, the interest in this problem has +<br/>been revamped, mostly due to the observation that standard face +<br/>recognition approaches do not perform well in real-time +<br/>scenarios where faces can be rotated, occluded, and under +<br/>unconstrained illumination. Face recognition tasks are generally +<br/>classified into two categories: +<br/>1. Face Verification. Given two face images, the task of face +<br/>verification is to determine if these two faces belong to the same +<br/>person. +<br/>2. Face Identification. This refers to the process of finding the +<br/>identity of an unknown face image given a database of known +<br/>faces. +<br/>However, there are certain situations where a third type of +<br/>face recognition is needed: face re-identification (face Re-ID). In +<br/>the context of Human-Robot Interaction (HRI), the goal of face +<br/>Re-ID is to determine if certain faces have been seen by the robot +<br/>before, and if so, to determine their identity. +<br/>Generally, a real-time and unsupervised face re-identification +<br/>system is required to achieve effective interactions between +<br/>humans and robots. In the realistic scenarios of HRI, the face re- +<br/>identification task is confronted with the following challenges: +<br/>a. The system needs to be able to build and update the run- +<br/>time user gallery on the fly as there is usually no prior +<br/>knowledge about the interaction targets in advance. +<br/>b. The system should achieve high processing speed in +<br/>order for the robot to maintain real-time interaction with +<br/>the users. +<br/>c. The method should be robust against high intra-class +<br/>illumination changes, partial +<br/>variance caused by +<br/> +<br/>occlusion, pose variation, and/or the display of facial +<br/>expressions. +<br/>d. The system should achieve high recognition accuracy on +<br/>low-quality images resulted from motion blur (when the +<br/>robot and / or the user is moving), out-of-focus blur, +<br/>and/or over /under-exposure. +<br/>Recently, deep-learning approaches, especially Convolutional +<br/>Neural Networks (CNNs), have achieved great success in solving +<br/>face recognition problems [4]–[8]. Comparing +<br/>to classic +<br/>approaches, deep-learning-based methods are characterised by +<br/>their powerful feature extraction abilities. However, as existing +<br/>works mostly focused on traditional face identification problems, +<br/>the potential applications of deep-learning-based methods in +<br/>solving face Re-ID problems is yet to be explored. +<br/>that can work effectively +<br/>In this paper, we present a real-time unsupervised face re- +<br/>identification system +<br/>in an +<br/>unconstrained environment. Firstly, we employ a pre-trained +<br/>CNN [7] as the feature extractor and try to improve its +<br/>performance and processing speed in HRI context by utilising a +<br/>variety of pre-processing techniques. In the Re-Identification step, +<br/>we then use an online clustering algorithm to build and update a +<br/>run-time face gallery and to output the probe faces’ ID. +<br/>Experiments show that our system can achieve a Re-ID accuracy +<br/>of 93.55% and 90.41% on the TERESA video dataset and the +<br/>YTF Dataset respectively and is able to achieve a real-time +<br/>processing speed of 10~26 FPS. +<br/>2. Related Works +<br/>Various methods [9]–[15] have been developed to solve the +<br/>person Re-ID problem in surveillance context. However, most of +<br/>them [9]–[13] are unsuitable to HRI applications as these +<br/>approaches often rely on soft biometrics (i.e. clothing’s colours +<br/>and textures) that are unavailable to the robot (which usually only +<br/>sees the user’s face). Due to the unavailability of such soft +<br/>biometrics, it is difficult to apply person re-identification +</td><td>('2563750', 'Yujiang Wang', 'yujiang wang')<br/>('49927631', 'Jie Shen', 'jie shen')<br/>('2403354', 'Stavros Petridis', 'stavros petridis')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>c866a2afc871910e3282fd9498dce4ab20f6a332</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Surveillance Face Recognition Challenge +<br/>Received: date / Accepted: date +</td><td>('5314735', 'Zhiyi Cheng', 'zhiyi cheng')</td><td></td></tr><tr><td>c8ca6a2dc41516c16ea0747e9b3b7b1db788dbdd</td><td>1 Department of Computer Science +<br/><b>Rutgers University</b><br/>New Jersey, USA +<br/>2 Department of Computer Science +<br/><b>The University of Texas at Arlington</b><br/>Texas, USA +<br/>PENG, XI: TRACK FACIAL POINTS IN UNCONSTRAINED VIDEOS +<br/>Track Facial Points in Unconstrained Videos +</td><td>('4340744', 'Xi Peng', 'xi peng')<br/>('40420376', 'Qiong Hu', 'qiong hu')<br/>('1768190', 'Junzhou Huang', 'junzhou huang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>xipeng.cs@rutgers.edu +<br/>qionghu.cs@rutgers.edu +<br/>jzhuang@uta.edu +<br/>dnm@cs.rutgers.edu +</td></tr><tr><td>c8292aa152a962763185e12fd7391a1d6df60d07</td><td>Camera Distance from Face Images +<br/><b>University of California, San Diego</b><br/>9500 Gilman Drive, La Jolla, CA, USA +</td><td>('25234832', 'Arturo Flores', 'arturo flores')</td><td>{aflores,echristiansen,kriegman,sjb}@cs.ucsd.edu +</td></tr><tr><td>c82c147c4f13e79ad49ef7456473d86881428b89</td><td></td><td></td><td></td></tr><tr><td>c84233f854bbed17c22ba0df6048cbb1dd4d3248</td><td>Exploring Locally Rigid Discriminative +<br/>Patches for Learning Relative Attributes +<br/>http://researchweb.iiit.ac.in/~yashaswi.verma/ +<br/>http://www.iiit.ac.in/~jawahar/ +<br/>CVIT +<br/>IIIT-Hyderabad, India +<br/>http://cvit.iiit.ac.in +</td><td>('1694502', 'C. V. Jawahar', 'c. v. jawahar')<br/>('2169614', 'Yashaswi Verma', 'yashaswi verma')<br/>('1694502', 'C. V. Jawahar', 'c. v. jawahar')</td><td></td></tr><tr><td>c829be73584966e3162f7ccae72d9284a2ebf358</td><td>shuttleNet: A biologically-inspired RNN with loop connection and parameter +<br/>sharing +<br/>1 National Engineering Laboratory for Video Technology, School of EE&CS, +<br/><b>Peking University, Beijing, China</b><br/>2 Cooperative Medianet Innovation Center, China +<br/>3 School of Information and Electronics, +<br/><b>Beijing Institute of Technology, Beijing, China</b></td><td>('38179026', 'Yemin Shi', 'yemin shi')<br/>('1705972', 'Yonghong Tian', 'yonghong tian')<br/>('5765799', 'Yaowei Wang', 'yaowei wang')<br/>('34097174', 'Tiejun Huang', 'tiejun huang')</td><td></td></tr><tr><td>c87d5036d3a374c66ec4f5870df47df7176ce8b9</td><td>ORIGINAL RESEARCH +<br/>published: 12 July 2018 +<br/>doi: 10.3389/fpsyg.2018.01190 +<br/>Temporal Dynamics of Natural Static +<br/>Emotional Facial Expressions +<br/>Decoding: A Study Using Event- and +<br/>Eye Fixation-Related Potentials +<br/><b>GIPSA-lab, Institute of Engineering, Universit Grenoble Alpes, Centre National de la Recherche Scienti que, Grenoble INP</b><br/>Grenoble, France, 2 Department of Conception and Control of Aeronautical and Spatial Vehicles, Institut Supérieur de +<br/>l’Aéronautique et de l’Espace, Université Fédérale de Toulouse, Toulouse, France, 3 Laboratoire InterUniversitaire de +<br/>Psychologie – Personnalité, Cognition, Changement Social, Université Grenoble Alpes, Université Savoie Mont Blanc, +<br/>Grenoble, France, 4 Exploration Fonctionnelle du Système Nerveux, Pôle Psychiatrie, Neurologie et Rééducation +<br/>Neurologique, CHU Grenoble Alpes, Grenoble, France, 5 Université Grenoble Alpes, Inserm, CHU Grenoble Alpes, Grenoble +<br/>Institut des Neurosciences, Grenoble, France +<br/>This study aims at examining the precise temporal dynamics of the emotional facial +<br/>decoding as it unfolds in the brain, according to the emotions displayed. To characterize +<br/>this processing as it occurs in ecological settings, we focused on unconstrained visual +<br/>explorations of natural emotional faces (i.e., free eye movements). The General Linear +<br/>Model (GLM; Smith and Kutas, 2015a,b; Kristensen et al., 2017a) enables such a +<br/>depiction. It allows deconvolving adjacent overlapping responses of the eye fixation- +<br/>related potentials (EFRPs) elicited by the subsequent fixations and the event-related +<br/>potentials (ERPs) elicited at the stimuli onset. Nineteen participants were displayed +<br/>with spontaneous static facial expressions of emotions (Neutral, Disgust, Surprise, and +<br/>Happiness) from the DynEmo database (Tcherkassof et al., 2013). Behavioral results +<br/>on participants’ eye movements show that the usual diagnostic features in emotional +<br/>decoding (eyes for negative facial displays and mouth for positive ones) are consistent +<br/>with the literature. The impact of emotional category on both the ERPs and the EFRPs +<br/>elicited by the free exploration of the emotional faces is observed upon the temporal +<br/>dynamics of the emotional facial expression processing. Regarding the ERP at stimulus +<br/>onset, there is a significant emotion-dependent modulation of the P2–P3 complex +<br/>and LPP components’ amplitude at the left frontal site for the ERPs computed by +<br/>averaging. Yet, the GLM reveals the impact of subsequent fixations on the ERPs time- +<br/>locked on stimulus onset. Results are also in line with the valence hypothesis. The +<br/>observed differences between the two estimation methods (Average vs. GLM) suggest +<br/>the predominance of the right hemisphere at the stimulus onset and the implication +<br/>of the left hemisphere in the processing of the information encoded by subsequent +<br/>fixations. Concerning the first EFRP, the Lambda response and the P2 component are +<br/>modulated by the emotion of surprise compared to the neutral emotion, suggesting +<br/>Edited by: +<br/>Eva G. Krumhuber, +<br/><b>University College London</b><br/>United Kingdom +<br/>Reviewed by: +<br/>Marie Arsalidou, +<br/><b>National Research University Higher</b><br/>School of Economics, Russia +<br/>Jaana Simola, +<br/><b>University of Helsinki, Finland</b><br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Emotion Science, +<br/>a section of the journal +<br/>Frontiers in Psychology +<br/>Received: 07 March 2018 +<br/>Accepted: 20 June 2018 +<br/>Published: 12 July 2018 +<br/>Citation: +<br/>Guérin-Dugué A, Roy RN, +<br/>Kristensen E, Rivet B, Vercueil L and +<br/>Tcherkassof A (2018) Temporal +<br/>Dynamics of Natural Static Emotional +<br/>Facial Expressions Decoding: A Study +<br/>Using Event- and Eye Fixation-Related +<br/>Potentials. Front. Psychol. 9:1190. +<br/>doi: 10.3389/fpsyg.2018.01190 +<br/>Frontiers in Psychology | www.frontiersin.org +<br/>July 2018 | Volume 9 | Article 1190 +</td><td>('7200702', 'Anne Guérin-Dugué', 'anne guérin-dugué')<br/>('20903548', 'Raphaëlle N. Roy', 'raphaëlle n. roy')<br/>('33987947', 'Emmanuelle Kristensen', 'emmanuelle kristensen')<br/>('48223466', 'Bertrand Rivet', 'bertrand rivet')<br/>('2544058', 'Laurent Vercueil', 'laurent vercueil')<br/>('3209946', 'Anna Tcherkassof', 'anna tcherkassof')<br/>('7200702', 'Anne Guérin-Dugué', 'anne guérin-dugué')</td><td>anne.guerin@gipsa-lab.grenoble-inp.fr +</td></tr><tr><td>c8e84cdff569dd09f8d31e9f9ba3218dee65e961</td><td>Dictionaries for Image and Video-based Face Recognition +<br/><b>Center for Automation Research, UMIACS, University of Maryland, College Park, MD 20742, USA</b><br/><b>National Institute of Standards and Technology, Gaithersburg, MD 20899, USA</b><br/>In recent years, sparse representation and dictionary learning-based methods have emerged as +<br/>powerful tools for efficiently processing data in non-traditional ways. A particular area of promise +<br/>for these theories is face recognition. +<br/>In this paper, we review the role of sparse representation +<br/>and dictionary learning for efficient face identification and verification. Recent face recognition +<br/>algorithms from still images, videos, and ambiguously label imagery are reviewed. In particular, +<br/>discriminative dictionary learning algorithms as well as methods based on weakly supervised learning +<br/>and domain adaptation are summarized. Some of the compelling challenges and issues that confront +<br/>research in face recognition using sparse representations and dictionary learning are outlined. +<br/>OCIS codes: (150.0150) Machine vision; (100.5010) Pattern recognition; (150.1135) Algorithms; +<br/>(100.0100) Image processing. +<br/>I. +<br/>INTRODUCTION +<br/>Face recognition is a challenging problem that has been +<br/>actively researched for over two decades [59]. Current +<br/>systems work very well when the test image is captured +<br/>under controlled conditions [35]. However, their perfor- +<br/>mance degrades significantly when the test image con- +<br/>tains variations that are not present in the training im- +<br/>ages. Some of these variations include illumination, pose, +<br/>expression, cosmetics, and aging. +<br/>It has been observed that since human faces have sim- +<br/>ilar overall configuration, face images can be described +<br/>by a relatively low dimensional subspace. As a result, +<br/>holistic dimensionality reduction subspace methods such +<br/>as Principle Component Analysis (PCA) [51], Linear +<br/>Discriminant Analysis (LDA) [3], [17] and Independent +<br/>Component Analysis (ICA) [2] have been proposed for +<br/>the task of face recognition. These approaches can be +<br/>classified into either generative or discriminative meth- +<br/>ods. An advantage of using generative approaches is their +<br/>reduced sensitivity to noise [59], [55]. +<br/>In recent years, generative and discriminative ap- +<br/>proaches based on sparse representations have been gain- +<br/>ing a lot of traction in biometrics recognition [32]. +<br/>In +<br/>sparse representation, given a signal and a redundant dic- +<br/>tionary, the goal is to represent this signal as a sparse lin- +<br/>ear combination of elements (also known as atoms) from +<br/>this dictionary. Finding a sparse representation entails +<br/>solving a convex optimization problem. Using sparse rep- +<br/>resentation, one can extract semantic information from +<br/>the signal. For instance, one can sparsely represent a test +<br/>sample in an overcomplete dictionary whose elements are +<br/>the training samples themselves, provided that sufficient +<br/>training samples are available from each class [55]. An in- +<br/>teresting property of sparse representations is that they +<br/>are robust to noise and occlusion. For instance, good +<br/>performance under partial occlusion, missing data and +<br/>variations in background has been demonstrated in many +<br/>sparsity-based methods [55], [38]. The ability of sparse +<br/>representations to extract meaningful information is due +<br/>in part to the fact that face images belonging to the same +<br/>person lie on a low-dimensional manifold. +<br/>In order to successfully apply sparse representation to +<br/>face recognition problems, one needs to correctly choose +<br/>an appropriate dictionary. Rather than using a pre- +<br/>determined dictionary, e.g. wavelets, one can train an +<br/>overcomplete data-driven dictionary. An appropriately +<br/>trained data-driven dictionary can simultaneously span +<br/>the subspace of all faces and support optimal discrimi- +<br/>nation of the classes. These dictionaries tend to provide +<br/>better classification accuracy than a predetermined dic- +<br/>tionary [31]. +<br/>Data-driven dictionaries can produce state-of-the-art +<br/>results in various face recognition tasks. However, when +<br/>the target data has a different distribution than the +<br/>source data, the learned sparse representation may not +<br/>be optimal. As a result, one needs to adapt these learned +<br/>representations from one domain to the other. The prob- +<br/>lem of transferring a representation or classifier from one +<br/>domain to the other is known as domain adaptation or +<br/>domain transfer learning [22], [42]. +<br/>In this paper, we summarize some of the recent ad- +<br/>vances in still- and video-based face recognition using +<br/>sparse representation and dictionary learning. Discrimi- +<br/>native dictionary learning algorithms as well as methods +<br/>based on weakly supervised learning and domain adapta- +<br/>tion are summarized. These examples show that sparsity +<br/>and dictionary learning are powerful tools for face recog- +<br/>nition. Understanding how well these algorithms work +<br/>can greatly improve our insights into some of the most +<br/>compelling challenges in still- and video-based face recog- +<br/>nition. +<br/>A. Organization of the paper +<br/>This paper is organized as follows. In Section II, we +<br/>briefly review the idea behind sparse representation and +<br/>dictionary learning. Section III presents some recent +</td><td>('1751078', 'Yi-Chen Chen', 'yi-chen chen')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>∗ Corresponding author: pvishalm@umiacs.umd.edu +</td></tr><tr><td>c8829013bbfb19ccb731bd54c1a885c245b6c7d7</td><td>Flexible Template and Model Matching Using Image Intensity +<br/><b>University College London</b><br/>Department of Computer Science +<br/>Gower Street, London, United Kingdom +</td><td>('31557997', 'Bernard F. Buxton', 'bernard f. buxton')<br/>('1797883', 'Vasileios Zografos', 'vasileios zografos')</td><td>{B.Buxton, V.Zografos}@cs.ucl.ac.uk +</td></tr><tr><td>c81ee278d27423fd16c1a114dcae486687ee27ff</td><td>Search Based Face Annotation Using Weakly +<br/>Labeled Facial Images +<br/><b>Savitribai Phule Pune University</b><br/><b>D.Y.Patil Institute of Engineering and Technology, Pimpri, Pune</b><br/>Mahatma Phulenagar, 120/2 Mahaganpati soc, Chinchwad, Pune-19, MH, India +<br/><b>D.Y.Patil Institute of Engineering and Technology, Pimpri, Pune-18, Savitribai Phule Pune University</b><br/>DYPIET, Pimpri, Pune-18, MH, India +</td><td>('15731441', 'Shital Shinde', 'shital shinde')<br/>('3392505', 'Archana Chaugule', 'archana chaugule')</td><td></td></tr><tr><td>c83a05de1b4b20f7cd7cd872863ba2e66ada4d3f</td><td>BREUER, KIMMEL: A DEEP LEARNING PERSPECTIVE ON FACIAL EXPRESSIONS +<br/>A Deep Learning Perspective on the Origin +<br/>of Facial Expressions +<br/>Department of Computer Science +<br/><b>Technion - Israel Institute of Technology</b><br/>Technion City, Haifa, Israel +<br/>Figure 1: Demonstration of the filter visualization process. +</td><td>('50484701', 'Ran Breuer', 'ran breuer')<br/>('1692832', 'Ron Kimmel', 'ron kimmel')</td><td>rbreuer@cs.technion.ac.il +<br/>ron@cs.technion.ac.il +</td></tr><tr><td>c88ce5ef33d5e544224ab50162d9883ff6429aa3</td><td>Face Match for Family Reunification: +<br/>Real-world Face Image Retrieval +<br/>U.S. National Library of Medicine, 8600 Rockville Pike, Bethesda, MD 20894, USA +<br/><b>Central Washington University, 400 E. University Way, Ellensburg, WA 98926, USA</b></td><td>('1744255', 'Eugene Borovikov', 'eugene borovikov')<br/>('34928283', 'Michael Gill', 'michael gill')<br/>('35029039', 'Szilárd Vajda', 'szilárd vajda')</td><td>(FaceMatch@NIH.gov) +<br/>(Szilard.Vajda@cwu.edu) +</td></tr><tr><td>c822bd0a005efe4ec1fea74de534900a9aa6fb93</td><td>Face Recognition Committee Machines: +<br/>Dynamic Vs. Static Structures +<br/>Department of Computer Science and Engineering +<br/><b>The Chinese University of Hong Kong</b><br/>Shatin, Hong Kong +</td><td>('2899702', 'Ho-Man Tang', 'ho-man tang')<br/>('1681775', 'Michael R. Lyu', 'michael r. lyu')<br/>('1706259', 'Irwin King', 'irwin king')</td><td>fhmtang, lyu, kingg@cse.cuhk.edu.hk +</td></tr><tr><td>c88c21eb9a8e08b66c981db35f6556f4974d27a8</td><td>Attribute Learning +<br/>Using Joint Human and Machine Computation +<br/>Edith Law +<br/>April 2011 +<br/>Machine Learning Department +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>Thesis Committee: +<br/>Luis von Ahn (co-Chair) +<br/>Tom Mitchell (co-Chair) +<br/>Jaime Carbonell +<br/>Eric Horvitz, Microsoft Research +<br/>Rob Miller, MIT +<br/>Submitted in partial fulfillment of the requirements +<br/>for the degree of Doctor of Philosophy. +<br/>Copyright c(cid:13) 2011 Edith Law +</td><td></td><td></td></tr><tr><td>c8adbe00b5661ab9b3726d01c6842c0d72c8d997</td><td>Deep Architectures for Face Attributes +<br/>Computer Vision and Machine Learning Group, Flickr, Yahoo, +</td><td>('3469274', 'Tobi Baumgartner', 'tobi baumgartner')<br/>('31922487', 'Jack Culpepper', 'jack culpepper')</td><td>{tobi, jackcul}@yahoo-inc.com +</td></tr><tr><td>fb4545782d9df65d484009558e1824538030bbb1</td><td></td><td></td><td></td></tr><tr><td>fbf196d83a41d57dfe577b3a54b1b7fa06666e3b</td><td>Extreme Learning Machine for Large-Scale +<br/>Action Recognition +<br/><b>Bo gazi ci University, Turkey</b></td><td>('1764521', 'Albert Ali Salah', 'albert ali salah')</td><td></td></tr><tr><td>fb2cc3501fc89f92f5ee130d66e69854f8a9ddd1</td><td>Learning Discriminative Features via Label Consistent Neural Network +<br/>†Raytheon BBN Technologies, Cambridge, MA, 02138 +<br/><b>University of Maryland, College Park, MD</b></td><td>('34145947', 'Zhuolin Jiang', 'zhuolin jiang')<br/>('1691470', 'Yaming Wang', 'yaming wang')<br/>('2502892', 'Viktor Rozgic', 'viktor rozgic')</td><td>{zjiang,wandrews,vrozgic}@bbn.com, {wym,lsd}@umiacs.umd.edu +</td></tr><tr><td>fbb6ee4f736519f7231830a8e337b263e91f06fe</td><td>Illumination Robust Facial Feature Detection via +<br/>Decoupled Illumination and Texture Features +<br/><b>University of Waterloo, Waterloo ON N2L3G1, Canada</b><br/>WWW home page: http://vip.uwaterloo.ca/ (cid:63) +</td><td>('2797326', 'Brendan Chwyl', 'brendan chwyl')<br/>('1685952', 'Alexander Wong', 'alexander wong')<br/>('1720258', 'David A. Clausi', 'david a. clausi')</td><td>{bchwyl,a28wong,dclausi}@uwaterloo.ca, +</td></tr><tr><td>fb87045600da73b07f0757f345a937b1c8097463</td><td>JIA, YANG, ZHU, KUANG, NIU, CHAN: RCCR FOR LARGE POSE +<br/>Reflective Regression of 2D-3D Face Shape +<br/>Across Large Pose +<br/><b>The University of Hong Kong</b><br/><b>National University of Defense</b><br/>Technology +<br/>3 Tencent Inc. +<br/>4 Sensetime Inc. +</td><td>('34760532', 'Xuhui Jia', 'xuhui jia')<br/>('2966679', 'Heng Yang', 'heng yang')<br/>('35130187', 'Xiaolong Zhu', 'xiaolong zhu')<br/>('1874900', 'Zhanghui Kuang', 'zhanghui kuang')<br/>('1939702', 'Yifeng Niu', 'yifeng niu')<br/>('40392393', 'Kwok-Ping Chan', 'kwok-ping chan')</td><td>xhjia@cs.hku.hk +<br/>yanghengnudt@gmail.com +<br/>lucienzhu@gmail.com +<br/>kuangzhanghui@sensetime.com +<br/>niuyifeng@nudt.edu.cn +<br/>kpchan@cs.hku.hk +</td></tr><tr><td>fb85867c989b9ee6b7899134136f81d6372526a9</td><td>Learning to Align Images using Weak Geometric Supervision +<br/><b>Georgia Institute of Technology</b><br/>2 Microsoft Research +</td><td>('1703391', 'Jing Dong', 'jing dong')<br/>('3288815', 'Byron Boots', 'byron boots')<br/>('2038264', 'Frank Dellaert', 'frank dellaert')<br/>('1757937', 'Sudipta N. Sinha', 'sudipta n. sinha')</td><td></td></tr><tr><td>fb5280b80edcf088f9dd1da769463d48e7b08390</td><td></td><td></td><td></td></tr><tr><td>fb54d3c37dc82891ff9dc7dd8caf31de00c40d6a</td><td>Beauty and the Burst: +<br/>Remote Identification of Encrypted Video Streams +<br/><b>Tel Aviv University, Cornell Tech</b><br/>Cornell Tech +<br/><b>Tel Aviv University, Columbia University</b></td><td>('39347554', 'Roei Schuster', 'roei schuster')<br/>('1723945', 'Vitaly Shmatikov', 'vitaly shmatikov')<br/>('2337345', 'Eran Tromer', 'eran tromer')</td><td>rs864@cornell.edu +<br/>shmat@cs.cornell.edu +<br/>tromer@cs.tau.ac.il +</td></tr><tr><td>fba464cb8e3eff455fe80e8fb6d3547768efba2f</td><td> +<br/>International Journal of Engineering and Applied Sciences (IJEAS) +<br/> ISSN: 2394-3661, Volume-3, Issue-2, February 2016 +<br/>Survey Paper on Emotion Recognition +<br/> +</td><td>('40502287', 'Prachi Shukla', 'prachi shukla')<br/>('2229305', 'Sandeep Patil', 'sandeep patil')</td><td></td></tr><tr><td>fbb2f81fc00ee0f257d4aa79bbef8cad5000ac59</td><td>Reading Hidden Emotions: Spontaneous +<br/>Micro-expression Spotting and Recognition +</td><td>('50079101', 'Xiaobai Li', 'xiaobai li')<br/>('1836646', 'Xiaopeng Hong', 'xiaopeng hong')<br/>('39056318', 'Antti Moilanen', 'antti moilanen')<br/>('47932625', 'Xiaohua Huang', 'xiaohua huang')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')</td><td></td></tr><tr><td>fb084b1fe52017b3898c871514cffcc2bdb40b73</td><td>RESEARCH ARTICLE +<br/>Illumination Normalization of Face Image +<br/>Based on Illuminant Direction Estimation and +<br/>Improved Retinex +<br/><b>School of Electronic and Information Engineering, Beihang University, Beijing, 100191, China</b><br/><b>Polytechnic University of Milan, Milan, 20156, Italy, 3 Applied Electronics</b><br/><b>University POLITEHNICA Timisoara, Timisoara, 300223, Romania</b></td><td>('1699804', 'Jizheng Yi', 'jizheng yi')<br/>('1724834', 'Xia Mao', 'xia mao')<br/>('35153304', 'Lijiang Chen', 'lijiang chen')<br/>('3399189', 'Yuli Xue', 'yuli xue')<br/>('1734732', 'Alberto Rovetta', 'alberto rovetta')<br/>('1860887', 'Catalin-Daniel Caleanu', 'catalin-daniel caleanu')</td><td>* clj@ee.buaa.edu.cn +</td></tr><tr><td>fb9ad920809669c1b1455cc26dbd900d8e719e61</td><td>3D Gaze Estimation from Remote RGB-D Sensors +<br/>THÈSE NO 6680 (2015) +<br/>PRÉSENTÉE LE 9 OCTOBRE 2015 +<br/>À LA FACULTÉ DES SCIENCES ET TECHNIQUES DE L'INGÉNIEUR +<br/>LABORATOIRE DE L'IDIAP +<br/>PROGRAMME DOCTORAL EN GÉNIE ÉLECTRIQUE +<br/>ÉCOLE POLYTECHNIQUE FÉDÉRALE DE LAUSANNE +<br/>POUR L'OBTENTION DU GRADE DE DOCTEUR ÈS SCIENCES +<br/>PAR +<br/>acceptée sur proposition du jury: +<br/>Prof. K. Aminian, président du jury +<br/>Dr J.-M. Odobez, directeur de thèse +<br/>Prof. L.-Ph. Morency, rapporteur +<br/>Prof. D. Witzner Hansen, rapporteur +<br/>Dr R. Boulic, rapporteur +<br/>Suisse +<br/>2015 +</td><td>('9206411', 'Kenneth Alberto Funes Mora', 'kenneth alberto funes mora')</td><td></td></tr><tr><td>ed28e8367fcb7df7e51963add9e2d85b46e2d5d6</td><td>International J. of Engg. Research & Indu. Appls. (IJERIA). +<br/>ISSN 0974-1518, Vol.9, No. III (December 2016), pp.23-42 +<br/>A NOVEL APPROACH OF FACE RECOGNITION USING +<br/>CONVOLUTIONAL NEURAL NETWORKS WITH AUTO +<br/>ENCODER +<br/>1 Research Scholar, Dept. of Electronics & Communication Engineering, +<br/><b>Rayalaseema University Kurnool, Andhra Pradesh</b><br/> 2 Research Supervisor, Professor, Dept. of Electronics & Communication Engineering, +<br/><b>Madanapalle Institute of Technology and Science, Madanapalle, Andhra Pradesh</b></td><td>('7006226', 'S. A. K JILANI', 's. a. k jilani')</td><td></td></tr><tr><td>ed0cf5f577f5030ac68ab62fee1cf065349484cc</td><td>Revisiting Data Normalization for +<br/>Appearance-Based Gaze Estimation +<br/><b>Max Planck Institute for Informatics</b><br/>Saarland Informatics Campus, +<br/>Graduate School of Information +<br/>Science and Technology, Osaka +<br/><b>Max Planck Institute for Informatics</b><br/>Saarland Informatics Campus, +<br/>Germany +<br/><b>University, Japan</b><br/>Germany +</td><td>('2520795', 'Xucong Zhang', 'xucong zhang')<br/>('1751242', 'Yusuke Sugano', 'yusuke sugano')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')</td><td>xczhang@mpi-inf.mpg.de +<br/>sugano@ist.osaka-u.ac.jp +<br/>bulling@mpi-inf.mpg.de +</td></tr><tr><td>edde81b2bdd61bd757b71a7b3839b6fef81f4be4</td><td>SHIH, MALLYA, SINGH, HOIEM: MULTI-PROPOSAL PART LOCALIZATION +<br/>Part Localization using Multi-Proposal +<br/>Consensus for Fine-Grained Categorization +<br/><b>University of Illinois</b><br/>Urbana-Champaign +<br/>IL, US +</td><td>('2525469', 'Kevin J. Shih', 'kevin j. shih')<br/>('36508529', 'Arun Mallya', 'arun mallya')<br/>('37415643', 'Saurabh Singh', 'saurabh singh')<br/>('2433269', 'Derek Hoiem', 'derek hoiem')</td><td>kjshih2@illinois.edu +<br/>amallya2@illinois.edu +<br/>ss1@illinois.edu +<br/>dhoiem@illinois.edu +</td></tr><tr><td>edf98a925bb24e39a6e6094b0db839e780a77b08</td><td>Simplex Representation for Subspace Clustering +<br/><b>The Hong Kong Polytechnic University, Hong Kong SAR, China</b><br/><b>School of Mathematics and Statistics, Xi an Jiaotong University, Xi an, China</b><br/>Spectral clustering based methods have achieved leading performance on subspace clustering problem. State-of-the-art subspace +<br/>clustering methods follow a three-stage framework: compute a coefficient matrix from the data by solving an optimization problem; +<br/>construct an affinity matrix from the coefficient matrix; and obtain the final segmentation by applying spectral clustering to the +<br/>affinity matrix. To construct a feasible affinity matrix, these methods mostly employ the operations of exponentiation, absolutely +<br/>symmetrization, or squaring, etc. However, all these operations will force the negative entries (which cannot be explicitly avoided) +<br/>the data. In this paper, we introduce the simplex representation (SR) to remedy this problem of representation based subspace +<br/>clustering. We propose an SR based least square regression (SRLSR) model to construct a physically more meaningful affinity matrix +<br/>by integrating the nonnegative property of graph into the representation coefficient computation while maintaining the discrimination +<br/>of original data. The SRLSR model is reformulated as a linear equality-constrained problem, which is solved efficiently under the +<br/>alternating direction method of multipliers framework. Experiments on benchmark datasets demonstrate that the proposed SRLSR +<br/>algorithm is very efficient and outperforms state-of-the-art subspace clustering methods on accuracy. +<br/>Index Terms—Subspace clustering, simplex representation, spectral clustering. +<br/>I. INTRODUCTION +<br/>H IGH-dimensional data are commonly observed in var- +<br/>ious computer vision and image processing prob- +<br/>lems. Contrary to their high-dimensional appearance, +<br/>the +<br/>latent structure of those data usually lie in a union of +<br/>low-dimensional subspaces [1]. Recovering the latent low- +<br/>dimensional subspaces from the high-dimensional observation +<br/>can not only reduce the computational cost and memory +<br/>requirements of subsequent algorithms, but also reduce the +<br/>learning and computer vision tasks, we need to find the clusters +<br/>of high-dimensional data such that each cluster can be fitted +<br/>by a subspace, which is referred to as the subspace clustering +<br/>(SC) problem [1]. +<br/>SC has been extensively studied in the past decades [2]– +<br/>[33]. Most of existing SC methods can be categorized into +<br/>four categories: iterative based methods [2], [3], algebraic +<br/>based methods [4]–[6], statistical based methods [7]–[10], and +<br/>spectral clustering based methods [14]–[33]. Among these four +<br/>categories, spectral clustering based methods have become the +<br/>mainstream due to their theoretical guarantees and promising +<br/>performance on real-world applications such as motion seg- +<br/>mentation [16] and face clustering [18]. The spectral clustering +<br/>based methods usually follow a three-step framework: Step +<br/>1) obtain a coefficient matrix of the data points by solving +<br/>an optimization problem, which usually incorporates sparse +<br/>or low rank regularizations due to their good mathematical +<br/>properties; Step 2) construct an affinity matrix from the +<br/>coefficient matrix by employing exponentiation [14], abso- +<br/>lutely symmetrization [15], [16], [20], [23]–[31], and squaring +<br/>operations [17]–[19], [32], [33], etc.; Step 3) apply spectral +<br/>analysis techniques [34] to the affinity matrix and obtain the +<br/>final clusters of the data points. +<br/>Most spectral clustering based methods [14]–[33] obtain +<br/>the expected coefficient matrix under the self-expressiveness +<br/>property [15], [16], which states that each data point in a union +<br/>of multiple subspaces can be linearly represented by the other +<br/>data points in the same subspace. However, in some real-world +<br/>applications, the data points lie in a union of multiple affine +<br/>subspaces rather than linear subspaces [16]. A trivial solution +<br/>is to ignore the affine structure of the data points and directly +<br/>perform clustering as in the subspaces of linear structures. +<br/>A non-negligible drawback of this solution is the increasing +<br/>dimension of the intersection of two subspaces, which can +<br/>make the subspaces indistinguishable from each other [16]. To +<br/>cluster data points lying in affine subspaces instead of linear +<br/>subspaces, the affine constraint is introduced [15], [16], in +<br/>which each data point can be written as an affine combination +<br/>of other points with the sum of coefficients being one. +<br/>Despite their high clustering accuracy, most of spectral +<br/>clustering based methods [14]–[33] suffer from three major +<br/>drawbacks. First, under the affine constraint, the coefficient +<br/>vector is not flexible enough to handle real-world applications +<br/>Second, negative coefficients cannot be fully avoided since +<br/>the existing methods do not explicitly consider non-negative +<br/>constraint +<br/>in real-world applications, +<br/>it is physically problematic to reconstruct a data point by +<br/>allowing the others to “cancel each other out” with complex +<br/>additions and subtractions [35]. Thus, most of these methods +<br/>are limited by being stranded at this physical bottleneck. Third, +<br/>the exponentiation, absolutely symmetrization, and squaring +<br/>operations in Step 2 will force the negative coefficients to +<br/>among the data points. +<br/>in Step 1. However, +<br/>To solve the three drawbacks mentioned above, we intro- +<br/>duce the Simplex Representation (SR) for spectral clustering +<br/>based SC. Specifically, the SR is introduced from two in- +<br/>terdependent aspects. First, to broaden its adaptivity to real +<br/>scenarios, we extend the affine constraint to the scaled affine +<br/>constraint, in which the coefficient vector in the optimization +</td><td>('47882783', 'Jun Xu', 'jun xu')<br/>('1803714', 'Deyu Meng', 'deyu meng')<br/>('48571185', 'Lei Zhang', 'lei zhang')</td><td></td></tr><tr><td>ed08ac6da6f8ead590b390b1d14e8a9b97370794</td><td> +<br/> +<br/> +<br/> +<br/> +<br/> ISSN(Online): 2320-9801 +<br/> +<br/> ISSN (Print): 2320-9798 +<br/>International Journal of Innovative Research in Computer +<br/>and Communication Engineering +<br/>(An ISO 3297: 2007 Certified Organization) +<br/>Vol. 3, Issue 9, September 2015 +<br/>An Efficient Approach for 3D Face +<br/>Recognition Using ANN Based Classifiers +<br/><b>Shri Shivaji College, Parbhani, M.S, India</b><br/><b>Arts, Commerce and Science College, Gangakhed, M.S, India</b><br/><b>Dnyanopasak College Parbhani, M.S, India</b></td><td>('34443070', 'Vaibhav M. Pathak', 'vaibhav m. pathak')</td><td></td></tr><tr><td>ed9d11e995baeec17c5d2847ec1a8d5449254525</td><td>Efficient Gender Classification Using a Deep LDA-Pruned Net +<br/><b>McGill University</b><br/>845 Sherbrooke Street W, Montreal, QC H3A 0G4, Canada +</td><td>('48087399', 'Qing Tian', 'qing tian')<br/>('1699104', 'Tal Arbel', 'tal arbel')<br/>('1713608', 'James J. Clark', 'james j. clark')</td><td>{qtian,arbel,clark}@cim.mcgill.ca +</td></tr><tr><td>edef98d2b021464576d8d28690d29f5431fd5828</td><td>Pixel-Level Alignment of Facial Images +<br/>for High Accuracy Recognition +<br/>Using Ensemble of Patches +</td><td>('1782221', 'Hoda Mohammadzade', 'hoda mohammadzade')<br/>('35809715', 'Amirhossein Sayyafan', 'amirhossein sayyafan')<br/>('24033665', 'Benyamin Ghojogh', 'benyamin ghojogh')</td><td></td></tr><tr><td>ed04e161c953d345bcf5b910991d7566f7c486f7</td><td>Combining facial expression analysis and synthesis on a +<br/>Mirror my emotions! +<br/>robot +</td><td>('2185308', 'Stefan Sosnowski', 'stefan sosnowski')<br/>('39124596', 'Christoph Mayer', 'christoph mayer')<br/>('1699132', 'Bernd Radig', 'bernd radig')</td><td></td></tr><tr><td>ed07856461da6c7afa4f1782b5b607b45eebe9f6</td><td>3D Morphable Models as Spatial Transformer Networks +<br/><b>University of York, UK</b><br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, UK</b></td><td>('39180407', 'Anil Bas', 'anil bas')<br/>('39976184', 'Patrik Huber', 'patrik huber')<br/>('1687021', 'William A. P. Smith', 'william a. p. smith')<br/>('46649582', 'Muhammad Awais', 'muhammad awais')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td>{ab1792,william.smith}@york.ac.uk, {p.huber,m.a.rana,j.kittler}@surrey.ac.uk +</td></tr><tr><td>ed1886e233c8ecef7f414811a61a83e44c8bbf50</td><td>Deep Alignment Network: A convolutional neural network for robust face +<br/>alignment +<br/><b>Warsaw University of Technology</b></td><td>('2393538', 'Marek Kowalski', 'marek kowalski')<br/>('1930272', 'Jacek Naruniec', 'jacek naruniec')<br/>('1760267', 'Tomasz Trzcinski', 'tomasz trzcinski')</td><td>m.kowalski@ire.pw.edu.pl, j.naruniec@ire.pw.edu.pl, t.trzcinski@ii.pw.edu.pl +</td></tr><tr><td>edd7504be47ebc28b0d608502ca78c0aea6a65a2</td><td>Recurrent Residual Learning for Action +<br/>Recognition +<br/><b>University of Bonn, Germany</b></td><td>('3434584', 'Ahsan Iqbal', 'ahsan iqbal')<br/>('32774629', 'Alexander Richard', 'alexander richard')<br/>('2946643', 'Juergen Gall', 'juergen gall')</td><td>{iqbalm,richard,kuehne,gall}@iai.uni-bonn.de +</td></tr><tr><td>ed388878151a3b841f95a62c42382e634d4ab82e</td><td>DenseImage Network: Video Spatial-Temporal Evolution +<br/>Encoding and Understanding +<br/><b>Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, China</b></td><td>('3162023', 'Xiaokai Chen', 'xiaokai chen')<br/>('2027479', 'Ke Gao', 'ke gao')</td><td>{chenxiaokai,kegao}@ict.ac.cn +</td></tr><tr><td>edbb8cce0b813d3291cae4088914ad3199736aa0</td><td>Proceedings of the Twenty-Fifth AAAI Conference on Artificial Intelligence +<br/>Efficient Subspace Segmentation via Quadratic Programming +<br/><b>College of Computer Science and Technology, Zhejiang University, China</b><br/><b>National University of Singapore, Singapore</b><br/><b>School of Information Systems, Singapore Management University, Singapore</b></td><td>('35019367', 'Shusen Wang', 'shusen wang')<br/>('2026127', 'Tiansheng Yao', 'tiansheng yao')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('38203359', 'Jialie Shen', 'jialie shen')</td><td>wssatzju@gmail.com, eleyuanx@nus.edu.sg, tsyaoo@gmail.com, eleyans@nus.edu.sg, jlshen@smu.edu.sg +</td></tr><tr><td>edff76149ec44f6849d73f019ef9bded534a38c2</td><td>Privacy-Preserving Visual Learning Using +<br/>Doubly Permuted Homomorphic Encryption +<br/><b>The University of Tokyo</b><br/>Tokyo, Japan +<br/><b>Michigan State University</b><br/>East Lansing, MI, USA +<br/><b>The University of Tokyo</b><br/>Tokyo, Japan +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA, USA +</td><td>('1899753', 'Ryo Yonetani', 'ryo yonetani')<br/>('2232940', 'Vishnu Naresh Boddeti', 'vishnu naresh boddeti')<br/>('9467266', 'Yoichi Sato', 'yoichi sato')<br/>('37991449', 'Kris M. Kitani', 'kris m. kitani')</td><td>yonetani@iis.u-tokyo.ac.jp +<br/>vishnu@msu.edu +<br/>kkitani@cs.cmu.edu +<br/>ysato@iis.u-tokyo.ac.jp +</td></tr><tr><td>ed96f2eb1771f384df2349879970065a87975ca7</td><td>Adversarial Attacks on Face Detectors using Neural +<br/>Net based Constrained Optimization +<br/>Department of Electrical and +<br/>Computer Engineering +<br/><b>University of Toronto</b><br/>Department of Electrical and +<br/>Computer Engineering +<br/><b>University of Toronto</b></td><td>('26418299', 'Avishek Joey Bose', 'avishek joey bose')<br/>('3241876', 'Parham Aarabi', 'parham aarabi')</td><td>Email: joey.bose@mail.utoronto.ca +<br/>Email: parham@ecf.utoronto.ca +</td></tr><tr><td>c178a86f4c120eca3850a4915134fff44cbccb48</td><td></td><td></td><td></td></tr><tr><td>c1d2d12ade031d57f8d6a0333cbe8a772d752e01</td><td>Journal of Math-for-Industry, Vol.2(2010B-5), pp.147–156 +<br/>Convex optimization techniques for the efficient recovery of a sparsely +<br/>corrupted low-rank matrix +<br/>D 案 +<br/>Received on August 10, 2010 / Revised on August 31, 2010 +<br/>E 案 +</td><td>('2372029', 'Silvia Gandy', 'silvia gandy')<br/>('1685085', 'Isao Yamada', 'isao yamada')</td><td></td></tr><tr><td>c180f22a9af4a2f47a917fd8f15121412f2d0901</td><td>Facial Expression Recognition by ICA with +<br/>Selective Prior +<br/>Department of Information Processing, School of Information Science, +<br/><b>Japan Advanced Institute of Science and Technology, Ishikawa-ken 923-1211, Japan</b></td><td>('1753878', 'Fan Chen', 'fan chen')<br/>('1791753', 'Kazunori Kotani', 'kazunori kotani')</td><td>{chen-fan, ikko}@jaist.ac.jp +</td></tr><tr><td>c146aa6d56233ce700032f1cb179700778557601</td><td>3D Morphable Models as Spatial Transformer Networks +<br/><b>University of York, UK</b><br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, UK</b></td><td>('39180407', 'Anil Bas', 'anil bas')<br/>('39976184', 'Patrik Huber', 'patrik huber')<br/>('1687021', 'William A. P. Smith', 'william a. p. smith')<br/>('9170545', 'Muhammad Awais', 'muhammad awais')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td>{ab1792,william.smith}@york.ac.uk, {p.huber,m.a.rana,j.kittler}@surrey.ac.uk +</td></tr><tr><td>c1f07ec629be1c6fe562af0e34b04c54e238dcd1</td><td>A Novel Facial Feature Localization Method Using Probabilistic-like Output* +<br/>Microsoft Research Asia +<br/> +<br/>Other methods utilize the face structure information and +<br/>heuristically search the facial features within the facial +<br/>regions [12]. Though the method is fast in localizing feature +<br/>points, it might be sensitive to some noises, such as eye +<br/>glasses, and thus fail in localization. +<br/>To address these problems, we proposed a learning-based +<br/>facial feature localization method under probabilistic-like +<br/>framework. We modified an object detection method [12] so +<br/>that it could generate a unified probabilistic-like output for +<br/>each point. We therefore proposed an algorithm to locate +<br/>the facial features using this probabilistic-like output. +<br/>Because this method is learning-based, it is robust to pose, +<br/>illumination, expression and appearance variations. The +<br/>localization speed of the proposed method is extremely fast. +<br/>It takes only about 10ms on the computer with a P4 1.3G +<br/>CPU to locate five feature points and the accuracy is +<br/>comparable with hand labeled results. +<br/>This paper is organized as follows. Section 2 first +<br/>describes the algorithm to calculate probabilistic-like output, +<br/>and then presents the proposed localization approach based +<br/>on the probabilistic-like output. Experiments will be given +<br/>at Section 3. Section 4 gives the conclusion remarks and +<br/>discusses future works. +<br/>2. FACIAL FEATURE POINT LOCALIZATION +<br/>The framework of the proposed method is illustrated in +<br/>Figure 1. +<br/>Figure 1.Feature Point Localization Framework +<br/><b>ECE dept, University of Miami</b><br/>1251 Memorial Drive, EB406 +<br/>Coral Gables, Florida, 33124, U.S. +<br/> +</td><td>('1684635', 'Lei Zhang', 'lei zhang')<br/>('9310930', 'Long', 'long')<br/>('8392859', 'Mingjing Li', 'mingjing li')<br/>('38188346', 'Hongjiang Zhang', 'hongjiang zhang')<br/>('1679242', 'Longbin Chen', 'longbin chen')</td><td>{leizhang, mjli,hjzhang}@microsoft.com +<br/>longzhu@msrchina.research.microsoft.com +<br/>l.chen6@umiami.edu +</td></tr><tr><td>c1cc2a2a1ab66f6c9c6fabe28be45d1440a57c3d</td><td>Dual-Agent GANs for Photorealistic and Identity +<br/>Preserving Profile Face Synthesis +<br/><b>National University of Singapore</b><br/>3 Panasonic R&D Center Singapore +<br/><b>National University of Defense Technology</b><br/><b>Franklin. W. Olin College of Engineering</b><br/><b>Qihoo 360 AI Institute</b></td><td>('46509484', 'Jian Zhao', 'jian zhao')<br/>('33419682', 'Lin Xiong', 'lin xiong')<br/>('2757639', 'Jianshu Li', 'jianshu li')<br/>('40345914', 'Fang Zhao', 'fang zhao')<br/>('2513111', 'Zhecan Wang', 'zhecan wang')<br/>('2668358', 'Sugiri Pranata', 'sugiri pranata')<br/>('3493398', 'Shengmei Shen', 'shengmei shen')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')</td><td>{zhaojian90, jianshu}@u.nus.edu +<br/>{lin.xiong, karlekar.jayashree, sugiri.pranata, shengmei.shen}@sg.panasonic.com +<br/>zhecan.wang@students.olin.edu +<br/>{elezhf, eleyans, elefjia}@u.nus.edu +</td></tr><tr><td>c10a15e52c85654db9c9343ae1dd892a2ac4a279</td><td>Int J Comput Vis (2012) 100:134–153 +<br/>DOI 10.1007/s11263-011-0494-3 +<br/>Learning the Relative Importance of Objects from Tagged Images +<br/>for Retrieval and Cross-Modal Search +<br/>Received: 16 December 2010 / Accepted: 23 August 2011 / Published online: 18 October 2011 +<br/>© Springer Science+Business Media, LLC 2011 +</td><td>('35788904', 'Sung Ju Hwang', 'sung ju hwang')</td><td></td></tr><tr><td>c1fc70e0952f6a7587b84bf3366d2e57fc572fd7</td><td></td><td></td><td></td></tr><tr><td>c1dfabe36a4db26bf378417985a6aacb0f769735</td><td>Journal of Computer Vision and Image Processing, NWPJ-201109-50 +<br/>1 +<br/>Describing Visual Scene through EigenMaps +<br/> +</td><td>('2630005', 'Shizhi Chen', 'shizhi chen')<br/>('35484757', 'YingLi Tian', 'yingli tian')</td><td></td></tr><tr><td>c1482491f553726a8349337351692627a04d5dbe</td><td></td><td></td><td></td></tr><tr><td>c1ff88493721af1940df0d00bcfeefaa14f1711f</td><td>CVPR +<br/>#1369 +<br/>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>045 +<br/>046 +<br/>047 +<br/>048 +<br/>049 +<br/>050 +<br/>051 +<br/>052 +<br/>053 +<br/>CVPR 2010 Submission #1369. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +<br/>CVPR +<br/>#1369 +<br/>Subspace Regression: Predicting a Subspace from one Sample +<br/>Anonymous CVPR submission +<br/>Paper ID 1369 +</td><td></td><td></td></tr><tr><td>c11eb653746afa8148dc9153780a4584ea529d28</td><td>Global and Local Consistent Wavelet-domain Age +<br/>Synthesis +</td><td>('2112221', 'Peipei Li', 'peipei li')<br/>('49995036', 'Yibo Hu', 'yibo hu')<br/>('1705643', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td></td></tr><tr><td>c1ebbdb47cb6a0ed49c4d1cf39d7565060e6a7ee</td><td>Robust Facial Landmark Localization Based on +</td><td>('19254504', 'Yiyun Pan', 'yiyun pan')<br/>('7934466', 'Junwei Zhou', 'junwei zhou')<br/>('46636537', 'Yongsheng Gao', 'yongsheng gao')<br/>('2065968', 'Shengwu Xiong', 'shengwu xiong')</td><td></td></tr><tr><td>c17a332e59f03b77921942d487b4b102b1ee73b6</td><td>Learning an appearance-based gaze estimator +<br/>from one million synthesised images +<br/>Tadas Baltruˇsaitis2 +</td><td>('34399452', 'Erroll Wood', 'erroll wood')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')<br/>('39626495', 'Peter Robinson', 'peter robinson')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')</td><td>1University of Cambridge, United Kingdom {erroll.wood,peter.robinson}@cam.ac.uk +<br/>2Carnegie Mellon University, United States {tbaltrus,morency}@cs.cmu.edu +<br/>3Max Planck Institute for Informatics, Germany bulling@mpi-inf.mpg.de +</td></tr><tr><td>c1e76c6b643b287f621135ee0c27a9c481a99054</td><td></td><td></td><td></td></tr><tr><td>c10b0a6ba98aa95d740a0d60e150ffd77c7895ad</td><td>HANSELMANN, YAN, NEY: DEEP FISHER FACES +<br/>Deep Fisher Faces +<br/>Human Language Technology and +<br/>Pattern Recognition Group +<br/><b>RWTH Aachen University</b><br/>Aachen, Germany +</td><td>('1804963', 'Harald Hanselmann', 'harald hanselmann')<br/>('35362682', 'Shen Yan', 'shen yan')<br/>('1685956', 'Hermann Ney', 'hermann ney')</td><td>hanselmann@cs.rwth-aachen.de +<br/>shen.yan@rwth-aachen.de +<br/>ney@cs.rwth-aachen.de +</td></tr><tr><td>c1298120e9ab0d3764512cbd38b47cd3ff69327b</td><td>Disguised Faces in the Wild +<br/>IIIT-Delhi, India +<br/><b>IBM TJ Watson Research Center, USA</b><br/>Rama Chellappa +<br/><b>University of Maryland, College Park, USA</b></td><td>('2573268', 'Vineet Kushwaha', 'vineet kushwaha')<br/>('2220719', 'Maneet Singh', 'maneet singh')<br/>('50631607', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('47733712', 'Nalini Ratha', 'nalini ratha')</td><td>{maneets, rsingh, mayank}@iiitd.ac.in +<br/>ratha@us.ibm.com +<br/>rama@umiacs.umd.ed +</td></tr><tr><td>c1dd69df9dfbd7b526cc89a5749f7f7fabc1e290</td><td>Unconstrained face identification with multi-scale block-based +<br/>correlation +<br/>Gaston, J., MIng, J., & Crookes, D. (2016). Unconstrained face identification with multi-scale block-based +<br/>correlation. In Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal +<br/><b>Processing (pp. 1477-1481). [978-1-5090-4117-6/17] Institute of Electrical and Electronics Engineers (IEEE</b><br/>Published in: +<br/>Proceedings of the 2017 IEEE International Conference on Acoustics, Speech and Signal Processing +<br/>Document Version: +<br/>Peer reviewed version +<br/><b>Queen's University Belfast - Research Portal</b><br/><b>Link to publication record in Queen's University Belfast Research Portal</b><br/>Publisher rights +<br/>© 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future +<br/><b>media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or</b><br/>redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. +<br/>General rights +<br/><b>Copyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or other</b><br/>copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated +<br/>with these rights. +<br/>Take down policy +<br/>The Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made to +<br/>ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the +<br/>Download date:29. Nov. 2017 +</td><td></td><td>Research Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk. +</td></tr><tr><td>c68ec931585847b37cde9f910f40b2091a662e83</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 9, No. 6, 2018 +<br/>A Comparative Evaluation of Dotted Raster- +<br/>Stereography and Feature-Based Techniques for +<br/>Automated Face Recognition +<br/>S. Talha Ahsan +<br/>Department of Computer Science +<br/>Department of Electrical Engineering +<br/><b>Usman Institute of Technology</b><br/><b>Usman Institute of Technology</b><br/>Karachi, Pakistan +<br/>Karachi, Pakistan +<br/>Department of Computer Science +<br/><b>Usman Institute of Technology</b><br/>Karachi, Pakistan +<br/>and +<br/>feature-based +<br/>system. The +<br/>techniques +<br/>two candidate +</td><td>('49508503', 'Muhammad Wasim', 'muhammad wasim')<br/>('3251091', 'Lubaid Ahmed', 'lubaid ahmed')<br/>('33238128', 'Syed Faisal Ali', 'syed faisal ali')</td><td></td></tr><tr><td>c696c9bbe27434cb6279223a79b17535cd6e88c8</td><td>International Journal of Information Technology Vol.11 No.9 2005 +<br/>* +<br/>Discriminant Analysis +<br/>Facial Expression Recognition with Pyramid Gabor +<br/>Features and Complete Kernel Fisher Linear +<br/>1 School of Electronic and Information Engineering, South China +<br/><b>University of Technology, Guangzhou, 510640, P.R.China</b><br/><b>Motorola China Research Center, Shanghai, 210000, P.R.China</b></td><td>('30193721', 'Duan-Duan Yang', 'duan-duan yang')<br/>('2949795', 'Lian-Wen Jin', 'lian-wen jin')<br/>('9215052', 'Jun-Xun Yin', 'jun-xun yin')<br/>('1751744', 'Li-Xin Zhen', 'li-xin zhen')<br/>('34824270', 'Jian-Cheng Huang', 'jian-cheng huang')</td><td>{ddyang, eelwjin,eejxyin}@scut.edu.cn +<br/>{Li-Xin.Zhen, Jian-Cheng.Huang}@motorola.com +</td></tr><tr><td>c65e4ffa2c07a37b0bb7781ca4ec2ed7542f18e3</td><td>Recurrent Neural Networks for Facial Action Unit +<br/>Recognition from Image Sequences +<br/>School of Computer Science +<br/><b>University of Witwatersrand</b><br/>Private Bag 3, Wits 2050, South Africa +<br/>Department of Computer Science +<br/><b>University of the Western Cape</b><br/>Bellville, South Africa +<br/><b>Middle East Technical University</b><br/>Northern Cyprus Campus +<br/>Güzelyurt, Mersin10, Turkey +</td><td>('1903882', 'H Nyongesa', 'h nyongesa')</td><td>Hima.vadapalli@wits.ac.za +<br/>hnyongesa@uwc.ac.za +<br/>Omlin@metu.edu.tr +</td></tr><tr><td>c614450c9b1d89d5fda23a54dbf6a27a4b821ac0</td><td>Vol.60: e17160480, January-December 2017 +<br/>http://dx.doi.org/10.1590/1678-4324-2017160480 +<br/>ISSN 1678-4324 Online Edition +<br/>1 +<br/>Engineering,Technology and Techniques +<br/>BRAZILIAN ARCHIVES OF +<br/>BIOLOGY AND TECHNOLOGY +<br/>A N I N T E R N A T I O N A L J O U R N A L +<br/>Face Image Retrieval of Efficient Sparse Code words and +<br/>Multiple Attribute in Binning Image +<br/><b>Srm Easwari Engineering College, Ramapuram, Bharathi Salai, Chennai, Tamil Nadu, India</b></td><td></td><td></td></tr><tr><td>c6096986b4d6c374ab2d20031e026b581e7bf7e9</td><td>A Framework for Using Context to +<br/>Understand Images of People +<br/>Submitted in partial fulfillment of the +<br/>requirements for the +<br/>degree of Doctor of Philosophy +<br/>Department of Electrical and Computer Engineering +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>May 2009 +<br/>Thesis Committee: +<br/>Tsuhan Chen, Chair +</td><td>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')<br/>('1763086', 'Alexei A. Efros', 'alexei a. efros')<br/>('1709305', 'Martial Hebert', 'martial hebert')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')<br/>('1794486', 'Marios Savvides', 'marios savvides')<br/>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')</td><td></td></tr><tr><td>c6608fdd919f2bc4f8d7412bab287527dcbcf505</td><td>Unsupervised Alignment of Natural +<br/>Language with Video +<br/>by +<br/>Submitted in Partial Fulfillment +<br/>of the +<br/>Requirements for the Degree +<br/>Doctor of Philosophy +<br/>Supervised by +<br/>Professor Daniel Gildea +<br/>Department of Computer Science +<br/>Arts, Sciences and Engineering +<br/>Edmund A. Hajim School of Engineering and Applied Sciences +<br/><b>University of Rochester</b><br/>Rochester, New York +<br/>2015 +</td><td>('2296971', 'Iftekhar Naim', 'iftekhar naim')</td><td></td></tr><tr><td>c6f3399edb73cfba1248aec964630c8d54a9c534</td><td>A Comparison of CNN-based Face and Head Detectors for +<br/>Real-Time Video Surveillance Applications +<br/>1 ´Ecole de technologie sup´erieure, Universit´e du Qu´ebec, Montreal, Canada +<br/>2 Genetec Inc., Montreal, Canada +</td><td>('38993564', 'Le Thanh Nguyen-Meidine', 'le thanh nguyen-meidine')<br/>('1697195', 'Eric Granger', 'eric granger')<br/>('40185782', 'Madhu Kiran', 'madhu kiran')<br/>('38755219', 'Louis-Antoine Blais-Morin', 'louis-antoine blais-morin')</td><td>lethanh@livia.etsmtl.ca, eric.granger@etsmtl.ca, mkiran@livia.etsmtl.ca +<br/>lablaismorin@genetec.com +</td></tr><tr><td>c62c910264658709e9bf0e769e011e7944c45c90</td><td>Recent Progress of Face Image Synthesis +<br/>National Laboratory of Pattern Recognition, CASIA +<br/>Center for Research on Intelligent Perception and Computing, CASIA +<br/>Center for Excellence in Brain Science and Intelligence Technology, CAS +<br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b></td><td>('9702077', 'Zhihe Lu', 'zhihe lu')<br/>('7719475', 'Zhihang Li', 'zhihang li')<br/>('1680853', 'Jie Cao', 'jie cao')<br/>('1705643', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>{luzhihe2016, lizhihang2016, caojie2016}@ia.ac.cn, {rhe, znsun}@nlpr.ia.ac.cn +</td></tr><tr><td>c678920facffd35853c9d185904f4aebcd2d8b49</td><td>Learning to Anonymize Faces for +<br/>Privacy Preserving Action Detection +<br/>1 EgoVid Inc., South Korea +<br/><b>University of California, Davis</b></td><td>('10805888', 'Zhongzheng Ren', 'zhongzheng ren')<br/>('1883898', 'Yong Jae Lee', 'yong jae lee')<br/>('1766489', 'Michael S. Ryoo', 'michael s. ryoo')</td><td>{zzren,yongjaelee}@ucdavis.edu, mryoo@egovid.com +</td></tr><tr><td>c660500b49f097e3af67bb14667de30d67db88e3</td><td>www.elsevier.com/locate/cviu +<br/>Facial asymmetry quantification for +<br/>expression invariant human identification +<br/>and Sinjini Mitrac +<br/><b>a The Robotics Institute, Carnegie Mellon University, 5000 Forbes Avenue, Pittsburgh, PA 15213, USA</b><br/><b>University of Pittsburgh, Pittsburgh, PA 15260, USA</b><br/><b>Carnegie Mellon University, Pittsburgh, PA 15213, USA</b><br/>Received 15 February 2002; accepted 24 March 2003 +</td><td>('1689241', 'Yanxi Liu', 'yanxi liu')<br/>('2185899', 'Karen L. Schmidt', 'karen l. schmidt')</td><td></td></tr><tr><td>c6241e6fc94192df2380d178c4c96cf071e7a3ac</td><td>Action Recognition with Trajectory-Pooled Deep-Convolutional Descriptors +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen key lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('33427555', 'Yu Qiao', 'yu qiao')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>07wanglimin@gmail.com, yu.qiao@siat.ac.cn, xtang@ie.cuhk.edu.hk +</td></tr><tr><td>c6ffa09c4a6cacbbd3c41c8ae7a728b0de6e10b6</td><td>This article appeared in a journal published by Elsevier. The attached +<br/>copy is furnished to the author for internal non-commercial research +<br/><b>and education use, including for instruction at the authors institution</b><br/>and sharing with colleagues. +<br/><b>Other uses, including reproduction and distribution, or selling or</b><br/>licensing copies, or posting to personal, institutional or third party +<br/>websites are prohibited. +<br/>In most cases authors are permitted to post their version of the +<br/>article (e.g. in Word or Tex form) to their personal website or +<br/>institutional repository. Authors requiring further information +<br/>regarding Elsevier’s archiving and manuscript policies are +<br/>encouraged to visit: +<br/>http://www.elsevier.com/copyright +</td><td></td><td></td></tr><tr><td>c6526dd3060d63a6c90e8b7ff340383c4e0e0dd8</td><td>OPEN +<br/>Received: 22 December 2015 +<br/>Accepted: 04 April 2016 +<br/>Published: 21 April 2016 +<br/>Anxiety promotes memory for +<br/>mood-congruent faces but does not +<br/>alter loss aversion +<br/><b>Pathological anxiety is associated with disrupted cognitive processing, including working memory and</b><br/>decision-making. In healthy individuals, experimentally-induced state anxiety or high trait anxiety +<br/>often results in the deployment of adaptive harm-avoidant behaviours. However, how these processes +<br/>affect cognition is largely unknown. To investigate this question, we implemented a translational +<br/>within-subjects anxiety induction, threat of shock, in healthy participants reporting a wide range of +<br/>trait anxiety scores. Participants completed a gambling task, embedded within an emotional working +<br/>memory task, with some blocks under unpredictable threat and others safe from shock. Relative to the +<br/>safe condition, threat of shock improved recall of threat-congruent (fearful) face location, especially in +<br/>highly trait anxious participants. This suggests that threat boosts working memory for mood-congruent +<br/>stimuli in vulnerable individuals, mirroring memory biases in clinical anxiety. By contrast, Bayesian +<br/>analysis indicated that gambling decisions were better explained by models that did not include threat +<br/>or treat anxiety, suggesting that: (i) higher-level executive functions are robust to these anxiety +<br/>manipulations; and (ii) decreased risk-taking may be specific to pathological anxiety. These findings +<br/>provide insight into the complex interactions between trait anxiety, acute state anxiety and cognition, +<br/>and may help understand the cognitive mechanisms underlying adaptive anxiety. +<br/>Anxiety disorders constitute a major global health burden1, and are characterized by negative emotional process- +<br/>ing biases, as well as disrupted working memory and decision-making2,3. On the other hand, anxiety can also be +<br/>an adaptive response to stress, stimulating individuals to engage in harm-avoidant behaviours. Influential the- +<br/>ories of pathological anxiety propose that clinical anxiety emerges through dysregulation of adaptive anxiety4,5. +<br/>Therefore, in order to understand how this dysregulation emerges in pathological anxiety, it is crucial to first +<br/>understand the cognitive features associated with adaptive or ‘non-pathological’ anxiety, in other words anxiety +<br/>levels that can vary within and between individuals but do not result in the development of clinical symptoms +<br/>associated with anxiety disorders. +<br/><b>Several methods exists to induce anxiety in healthy individuals, including threat of shock (ToS), the Trier</b><br/>social stressor test (TSST), and the cold pressor test (CPT). During the ToS paradigm, subjects typically perform +<br/>a cognitive task while either at risk of or safe from rare, but unpleasant, electric shocks. Compared to the other +<br/>methodologies, ToS has the advantage of allowing for within-subjects, within-sessions, designs (for a review +<br/>on its effects on cognition, see Robinson et al.2), and ensures the task is performed while being anxious, rather +<br/>than after being relieved from the stressor. In addition, ToS paradigms have good translational analogues6, are +<br/>well-validated7, and are thus considered a reliable model for examining adaptive anxiety in healthy individuals. +<br/>Because the engagement of adaptive anxiety processes may vary with individuals’ vulnerability to developing +<br/>pathological anxiety8–10, we were also interested in examining how the effects of state anxiety induced by threat +<br/>of shock interact with dispositional or trait anxiety, as reflected in self-report questionnaire scores such as the +<br/>State-Trait Anxiety Inventory11 (STAI). High levels of self-reported trait anxiety are indeed considered a strong +<br/>vulnerability factor in the development of pathological anxiety4,12. +<br/>The extent to which induced state anxiety (elicited by the laboratory procedures discussed above) and +<br/>trait anxiety interact to alter cognition has rarely been studied10. In particular, does induced anxiety have a +<br/><b>Institute of Cognitive Neuroscience, University College London, London WC1N 3AR, UK. 2Affective Brain</b><br/><b>Lab, University College London, London WC1H 0AP, UK. 3Clinical</b><br/><b>Psychopharmacology Unit, Educational and Health Psychology, University College</b><br/>London, WC1E 7HB. *These authors contributed equally to this work. †These authors jointly supervised this work. +</td><td>('4177273', 'Chandni Hindocha', 'chandni hindocha')</td><td>Correspondence and requests for materials should be addressed to C.J.C. (email: caroline.charpentier.11@ucl.ac.uk) +</td></tr><tr><td>c62c07de196e95eaaf614fb150a4fa4ce49588b4</td><td>Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18) +<br/>1078 +</td><td></td><td></td></tr><tr><td>c65a394118d34beda5dd01ae0df163c3db88fceb</td><td>In press : Proceedings of the 30th European Conference On Information Retrieval +<br/>Glasgow, March-April 2008 +<br/>Finding the Best Picture: +<br/>Cross-Media Retrieval of Content +<br/><b>Katholieke Universiteit Leuven</b><br/>Celestijnenlaan 200A, B-3001 Heverlee, Belgium +<br/>http://www.cs.kuleuven.be/~liir/ +</td><td>('1797588', 'Koen Deschacht', 'koen deschacht')<br/>('1802161', 'Marie-Francine Moens', 'marie-francine moens')</td><td>{Koen.Deschacht,Marie-Francine.Moens}@cs.kuleuven.be +</td></tr><tr><td>ec90d333588421764dff55658a73bbd3ea3016d2</td><td>Research Article +<br/>Protocol for Systematic Literature Review of Face +<br/>Recognition in Uncontrolled Environment +<br/><b>Bacha Khan University, Charsadda, KPK, Pakistan</b></td><td>('12144785', 'Faizan Ullah', 'faizan ullah')<br/>('46463663', 'Sabir Shah', 'sabir shah')<br/>('49669073', 'Dilawar Shah', 'dilawar shah')<br/>('12579194', 'Shujaat Ali', 'shujaat ali')</td><td>faizanullah@bkuc.edu.pk +</td></tr><tr><td>ec8ec2dfd73cf3667f33595fef84c95c42125945</td><td>Pose-Invariant Face Alignment with a Single CNN +<br/><b>Michigan State University</b><br/>2Visualization Group, Bosch Research and Technology Center North America +</td><td>('2357264', 'Amin Jourabloo', 'amin jourabloo')<br/>('3876303', 'Mao Ye', 'mao ye')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('3334600', 'Liu Ren', 'liu ren')</td><td>1,2 {jourablo, liuxm}@msu.edu, {mao.ye2, liu.ren}@us.bosch.com +</td></tr><tr><td>ec1e03ec72186224b93b2611ff873656ed4d2f74</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>3D Reconstruction of “In-the-Wild” Faces in +<br/>Images and Videos +</td><td>('47456731', 'James Booth', 'james booth')<br/>('2931390', 'Anastasios Roussos', 'anastasios roussos')<br/>('31243357', 'Evangelos Ververas', 'evangelos ververas')<br/>('2015036', 'Stylianos Ploumpis', 'stylianos ploumpis')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')</td><td></td></tr><tr><td>ec12f805a48004a90e0057c7b844d8119cb21b4a</td><td>Distance-Based Descriptors and Their +<br/>Application in the Task of Object Detection +<br/><b>Technical University of Ostrava, FEECS</b><br/>17. Listopadu 15, 708 33 Ostrava-Poruba, Czech Republic +</td><td>('2467747', 'Radovan Fusek', 'radovan fusek')<br/>('2557877', 'Eduard Sojka', 'eduard sojka')</td><td>{radovan.fusek,eduard.sojka}@vsb.cz +</td></tr><tr><td>ec22eaa00f41a7f8e45ed833812d1ac44ee1174e</td><td></td><td></td><td></td></tr><tr><td>ec54000c6c0e660dd99051bdbd7aed2988e27ab8</td><td>TWO IN ONE: JOINT POSE ESTIMATION AND FACE RECOGNITION WITH P2CA1 +<br/>*Dept. Teoria del Senyal i Comunicacions - Universitat Politècnica de Catalunya, Barcelona, Spain +<br/>+Dipartimento di Elettronica e Informazione - Politecnico di Milano, Meiland, Italy +</td><td>('2771575', 'Francesc Tarres', 'francesc tarres')<br/>('31936578', 'Antonio Rama', 'antonio rama')<br/>('2158932', 'Davide Onofrio', 'davide onofrio')<br/>('1729506', 'Stefano Tubaro', 'stefano tubaro')</td><td>{tarres, alrama}@gps.tsc.upc.edu +<br/>{d.onofrio, tubaro}@elet.polimi.it +</td></tr><tr><td>ec0104286c96707f57df26b4f0a4f49b774c486b</td><td>758 +<br/>An Ensemble CNN2ELM for Age Estimation +</td><td>('40402919', 'Mingxing Duan', 'mingxing duan')<br/>('39893222', 'Kenli Li', 'kenli li')<br/>('34373985', 'Keqin Li', 'keqin li')</td><td></td></tr><tr><td>ec05078be14a11157ac0e1c6b430ac886124589b</td><td>Longitudinal Face Aging in the Wild - Recent Deep Learning Approaches +<br/><b>Concordia University</b><br/>Montreal, Quebec, Canada +<br/><b>Concordia University</b><br/>Montreal, Quebec, Canada +<br/>CyLab Biometrics Center +<br/>Dept. of Electrical and Computer Engineering +<br/><b>Carnegie Mellon University Pittsburgh, PA, USA</b><br/><b>Concordia University</b><br/>Montreal, Quebec, Canada +</td><td>('1876581', 'Chi Nhan Duong', 'chi nhan duong')<br/>('2687827', 'Kha Gia Quach', 'kha gia quach')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('1699922', 'Tien D. Bui', 'tien d. bui')</td><td>Email: c duon@encs.concordia.ca +<br/>Email: k q@encs.concordia.ca +<br/>Email: kluu@andrew.cmu.edu +<br/>Email: bui@encs.concordia.ca +</td></tr><tr><td>4e7ed13e541b8ed868480375785005d33530e06d</td><td>Face Recognition Using Deep Multi-Pose Representations +<br/>Ram Nevatiab Gerard Medionib +<br/>Prem Natarajana +<br/><b>aInformation Sciences Institute</b><br/><b>University of Southern California</b><br/>Marina Del Rey, CA +<br/><b>b Institute for Robotics and Intelligent Systems</b><br/><b>University of Southern California</b><br/>Los Angeles, California +<br/><b>cThe Open University</b><br/>Raanana, Israel +</td><td>('1746738', 'Yue Wu', 'yue wu')<br/>('38696444', 'Stephen Rawls', 'stephen rawls')<br/>('35840854', 'Shai Harel', 'shai harel')<br/>('11269472', 'Iacopo Masi', 'iacopo masi')<br/>('1689391', 'Jongmoo Choi', 'jongmoo choi')<br/>('2955822', 'Jatuporn Toy Leksut', 'jatuporn toy leksut')<br/>('5911467', 'Jungyeon Kim', 'jungyeon kim')<br/>('1756099', 'Tal Hassner', 'tal hassner')</td><td></td></tr><tr><td>4e30107ee6a2e087f14a7725e7fc5535ec2f5a5f</td><td>Представление новостных сюжетов с помощью +<br/>событийных фотографий +<br/>© М.М. Постников +<br/> © Б.В. Добров +<br/>Московский государственный университет имени М.В. Ломоносова +<br/>факультет вычислительной математики и кибернетики, +<br/>Москва, Россия +<br/>Аннотация. Рассмотрена задача аннотирования новостного сюжета изображениями, +<br/>ассоциированными с конкретными текстами сюжета. Введено понятие «событийной фотографии», +<br/>содержащей конкретную информацию, дополняющую текст сюжета. Для решения задачи применены +<br/>нейронные сети с использованием переноса обучения (Inception v3) для специальной размеченной +<br/>коллекции из 4114 изображений. Средняя точность полученных результатов составила более 94,7%. +<br/>Ключевые слова: событийная фотография, новостные иллюстрации, перенос обучения. +<br/>News Stories Representation Using Event Photos +<br/>© M.M. Postnikov +<br/> © B.V. Dobrov +<br/><b>Lomonosov Moscow State University, Faculty of Computational Mathematics and Cybernetics</b><br/>Moscow, Russia +</td><td></td><td>mihanlg@yandex.ru +<br/> dobrov_bv@mail.ru +<br/>mihanlg@yandex.ru +<br/> dobrov_bv@mail.ru +</td></tr><tr><td>4e5dc3b397484326a4348ccceb88acf309960e86</td><td>Hindawi Publishing Corporation +<br/>e Scientific World Journal +<br/>Volume 2014, Article ID 219732, 12 pages +<br/>http://dx.doi.org/10.1155/2014/219732 +<br/>Research Article +<br/>Secure Access Control and Large Scale Robust Representation +<br/>for Online Multimedia Event Detection +<br/><b>School of Computer Science and Engineering, South China University of Technology, Guangzhou 510006, China</b><br/><b>School of Computer Science, Carnegie Mellon University, Pittsburgh, PA 15213, USA</b><br/><b>School of Computer Science, Wuyi University, Jiangmen 529020, China</b><br/><b>State Key Laboratory of Pulp and Paper Engineering, South China University of Technology, Guangzhou 510640, China</b><br/>Received 2 April 2014; Accepted 30 June 2014; Published 22 July 2014 +<br/>Academic Editor: Vincenzo Eramo +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>We developed an online multimedia event detection (MED) system. However, there are a secure access control issue and a large +<br/>scale robust representation issue when we want to integrate traditional event detection algorithms into the online environment. For +<br/>the first issue, we proposed a tree proxy-based and service-oriented access control (TPSAC) model based on the traditional role +<br/>based access control model. Verification experiments were conducted on the CloudSim simulation platform, and the results showed +<br/>that the TPSAC model is suitable for the access control of dynamic online environments. For the second issue, inspired by the +<br/>object-bank scene descriptor, we proposed a 1000-object-bank (1000OBK) event descriptor. Feature vectors of the 1000OBK were +<br/>extracted from response pyramids of 1000 generic object detectors which were trained on standard annotated image datasets, such +<br/>as the ImageNet dataset. A spatial bag of words tiling approach was then adopted to encode these feature vectors for bridging the gap +<br/>between the objects and events. Furthermore, we performed experiments in the context of event classification on the challenging +<br/>TRECVID MED 2012 dataset, and the results showed that the robust 1000OBK event descriptor outperforms the state-of-the-art +<br/>approaches. +<br/>1. Introduction +<br/>As one of the most interesting aspects of multimedia content +<br/>analysis, the multimedia event detection (MED) is becoming +<br/>an important research area for computer vision in recent +<br/><b>years. According to the definition by the National Institute</b><br/>of Standards and Technology (NIST) [1], an event (1) is +<br/>a complex activity occurring at a specific place and time, +<br/>(2) involves people interacting with other people and/or +<br/>objects, (3) consists of a number of human actions, processes, +<br/>and activities that are loosely or tightly organized and that +<br/>have significant temporal and semantic relationships to the +<br/>overarching activity, and (4) is directly observable. A MED +<br/>task is to indicate whether an event is occurred in a specified +<br/>test clip based on a standard event kit [1], which includes an +<br/>event name, a textual definition, a textual explication with an +<br/>attribute list, an evidential description, and a set of illustrative +<br/>video examples. Although there are many other definitions +<br/>available, such as the MED definitions from the NIST, the +<br/>research on the MED is still far from reaching its maturity. +<br/>Most of the current researches are focused on specific areas, +<br/>such as sports video [2], news video [3], and surveillance +<br/>video [4]. These approaches do not perform well when used +<br/>for the online or web based event detection due to two types +<br/>of issues, which are the secure access control issue and the +<br/>large scale robust representation issue. Thus, we developed an +<br/>online multimedia event detection system, trying to provide +<br/>general MED services. +<br/>The first issue is about how we can obtain a secure access +<br/>control for the online multimedia event detection system. +<br/>Compared with that of traditional distributed systems, it is a +<br/>kind of service relationships between access control subjects +<br/>and objects in the online multimedia event detection system. +<br/>The service could establish, recombine, destruct, and even +<br/>inherit efficiently to requested parameters which cannot be +<br/>satisfied well by traditional access control models, such as +</td><td>('1706701', 'Changyu Liu', 'changyu liu')<br/>('40371462', 'Bin Lu', 'bin lu')<br/>('1780591', 'Huiling Li', 'huiling li')<br/>('1706701', 'Changyu Liu', 'changyu liu')</td><td>Correspondence should be addressed to Bin Lu; lbscut@gmail.com +</td></tr><tr><td>4e6c17966efae956133bf8f22edeffc24a0470c1</td><td>Face Classification: A Specialized Benchmark +<br/>Study +<br/>1School of Electronic, Electrical and Communication Engineering +<br/>2Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>University of Chinese Academy of Sciences</b><br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/><b>Macau University of Science and Technology</b></td><td>('37614515', 'Jiali Duan', 'jiali duan')<br/>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('2950852', 'Shuai Zhou', 'shuai zhou')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jli.duan,shuaizhou.palm}@gmail.com, {scliao,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>4e1836914bbcf94dc00e604b24b1b0d6d7b61e66</td><td>Dynamic Facial Expression Recognition Using Boosted +<br/>Component-based Spatiotemporal Features and +<br/>Multi-Classifier Fusion +<br/>1. Machine Vision Group, Department of Electrical and Information Engineering, +<br/><b>University of Oulu, Finland</b><br/><b>Research Center for Learning Science, Southeast University, China</b><br/>http://www.ee.oulu.fi/mvg +</td><td>('18780812', 'Xiaohua Huang', 'xiaohua huang')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')<br/>('40608983', 'Wenming Zheng', 'wenming zheng')</td><td>{huang.xiaohua,gyzhao,mkp}@ee.oulu.fi +<br/>wenming_zheng@seu.edu.cn +</td></tr><tr><td>4e4fa167d772f34dfffc374e021ab3044566afc3</td><td>Learning Low-Rank Representations with Classwise +<br/>Block-Diagonal Structure for Robust Face Recognition +<br/><b>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences</b><br/><b>School of Computer Science, Nanjing University of Science and Technology</b><br/><b>University of Maryland, College Park</b></td><td>('1689181', 'Yong Li', 'yong li')<br/>('38188270', 'Jing Liu', 'jing liu')<br/>('3233021', 'Zechao Li', 'zechao li')<br/>('34868330', 'Yangmuzi Zhang', 'yangmuzi zhang')<br/>('1694235', 'Hanqing Lu', 'hanqing lu')<br/>('38450168', 'Songde Ma', 'songde ma')</td><td>{yong.li,jliu,luhq}@nlpr.ia.ac.cn, zechao.li@gmail.com, ymzhang@umiacs.umd.edu, masd@most.cn +</td></tr><tr><td>4e32fbb58154e878dd2fd4b06398f85636fd0cf4</td><td>A Hierarchical Matcher using Local Classifier Chains +<br/>L. Zhang and I.A. Kakadiaris +<br/>Computational Biomedicine Lab, 4849 Calhoun Rd, Rm 373, Houston, TX 77204 +</td><td></td><td></td></tr><tr><td>4ed54d5093d240cc3644e4212f162a11ae7d1e3b</td><td>Learning Visual Compound Models from Parallel +<br/>Image-Text Datasets +<br/><b>Bielefeld University</b><br/><b>University of Toronto</b></td><td>('2872318', 'Jan Moringen', 'jan moringen')<br/>('1724954', 'Sven Wachsmuth', 'sven wachsmuth')<br/>('1792908', 'Suzanne Stevenson', 'suzanne stevenson')</td><td>{jmoringe,swachsmu}@techfak.uni-bielefeld.de +<br/>{sven,suzanne}@cs.toronto.edu +</td></tr><tr><td>4e8c608fc4b8198f13f8a68b9c1a0780f6f50105</td><td>How Related Exemplars Help Complex Event Detection in Web Videos? +<br/><b>ITEE, The University of Queensland, Australia</b><br/><b>ECE, National University of Singapore, Singapore</b><br/>§† +<br/><b>School of Computer Science, Carnegie Mellon University, USA</b></td><td>('39033919', 'Yi Yang', 'yi yang')<br/>('1727419', 'Zhigang Ma', 'zhigang ma')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')<br/>('2351434', 'Zhongwen Xu', 'zhongwen xu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>{yiyang,kevinma,alex}@cs.cmu.edu z.xu3@uq.edu.au +</td></tr><tr><td>4ea53e76246afae94758c1528002808374b75cfa</td><td>Lasbela, U. J.Sci. Techl., vol.IV , pp. 57-70, 2015 +<br/>Review ARTICLE +<br/>A Review of Scholastic Examination and Models for Face Recognition +<br/> ISSN 2306-8256 +<br/>and Retrieval in Video +<br/> +<br/><b>SBK Women s University, Quetta, Balochistan</b><br/><b>University of Balochistan, Quetta</b><br/><b>University of Balochistan, Quetta</b><br/><b>Institute of Biochemistry, University of Balochistan, Quetta</b></td><td>('35415301', 'Varsha Sachdeva', 'varsha sachdeva')<br/>('2139801', 'Junaid Baber', 'junaid baber')<br/>('3343681', 'Maheen Bakhtyar', 'maheen bakhtyar')<br/>('1903979', 'Muzamil Bokhari', 'muzamil bokhari')<br/>('1702753', 'Imran Ali', 'imran ali')</td><td></td></tr><tr><td>4ed2d7ecb34a13e12474f75d803547ad2ad811b2</td><td>Common Action Discovery and Localization in Unconstrained Videos +<br/>School of Electrical and Electronic Engineering +<br/><b>Nanyang Technological University, Singapore</b></td><td>('1691251', 'Jiong Yang', 'jiong yang')<br/>('34316743', 'Junsong Yuan', 'junsong yuan')</td><td>yang0374@e.ntu.edu.sg, jsyuan@ntu.edu.sg +</td></tr><tr><td>4e97b53926d997f451139f74ec1601bbef125599</td><td>Discriminative Regularization for Generative Models +<br/><b>Montreal Institute for Learning Algorithms, Universit e de Montr eal</b></td><td>('2059369', 'Alex Lamb', 'alex lamb')<br/>('3074927', 'Vincent Dumoulin', 'vincent dumoulin')</td><td>FIRST.LAST@UMONTREAL.CA +</td></tr><tr><td>4e8168fbaa615009d1618a9d6552bfad809309e9</td><td>Deep Convolutional Neural Network Features and the Original Image +<br/><b>School of Behavioral and Brain Sciences, The University of Texas at Dallas, USA</b><br/><b>University of Maryland, College Park, USA</b></td><td>('7493834', 'Connor J. Parde', 'connor j. parde')<br/>('3363752', 'Matthew Q. Hill', 'matthew q. hill')<br/>('15929465', 'Y. Ivette Colon', 'y. ivette colon')<br/>('2716670', 'Swami Sankaranarayanan', 'swami sankaranarayanan')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')</td><td></td></tr><tr><td>4e0636a1b92503469b44e2807f0bb35cc0d97652</td><td>Adversarial Localization Network +<br/><b>Tsinghua University</b><br/><b>Stanford University</b><br/><b>Stanford University</b></td><td>('2548303', 'Lijie Fan', 'lijie fan')<br/>('3303970', 'Shengjia Zhao', 'shengjia zhao')<br/>('2490652', 'Stefano Ermon', 'stefano ermon')</td><td>flj14@mails.tsinghua.edu.cn +<br/>sjzhao@stanford.edu +<br/>ermon@stanford.edu +</td></tr><tr><td>4e27fec1703408d524d6b7ed805cdb6cba6ca132</td><td>SSD-Sface: Single shot multibox detector for small faces +<br/>C. Thuis +</td><td></td><td></td></tr><tr><td>4e6c9be0b646d60390fe3f72ce5aeb0136222a10</td><td>Long-term Temporal Convolutions +<br/>for Action Recognition +</td><td>('1785596', 'Ivan Laptev', 'ivan laptev')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>4ea4116f57c5d5033569690871ba294dc3649ea5</td><td>Multi-View Face Alignment Using 3D Shape Model for +<br/>View Estimation +<br/><b>Tsinghua University</b><br/>2Core Technology Center, Omron Corporation +</td><td>('1739678', 'Yanchao Su', 'yanchao su')<br/>('1679380', 'Haizhou Ai', 'haizhou ai')<br/>('1710195', 'Shihong Lao', 'shihong lao')</td><td>ahz@mail.tsinghua.edu.cn +</td></tr><tr><td>4e444db884b5272f3a41e4b68dc0d453d4ec1f4c</td><td></td><td></td><td></td></tr><tr><td>4ef0a6817a7736c5641dc52cbc62737e2e063420</td><td>International Journal of Advanced Computer Research (ISSN (Print): 2249-7277 ISSN (Online): 2277-7970) +<br/>Volume-4 Number-4 Issue-17 December-2014 +<br/>Study of Face Recognition Techniques +<br/>Received: 10-November-2014; Revised: 18-December-2014; Accepted: 23-December-2014 +<br/>©2014 ACCENTS +</td><td>('7874804', 'Sangeeta Kaushik', 'sangeeta kaushik')<br/>('33551600', 'R. B. Dubey', 'r. b. dubey')<br/>('1680807', 'Abhimanyu Madan', 'abhimanyu madan')</td><td></td></tr><tr><td>4e4d034caa72dce6fca115e77c74ace826884c66</td><td>RESEARCH ARTICLE +<br/>Sex differences in facial emotion recognition +<br/>across varying expression intensity levels +<br/>from videos +<br/><b>University of Bath, Bath, Somerset, United Kingdom</b><br/>☯ These authors contributed equally to this work. +<br/>¤ Current address: Social and Affective Neuroscience Laboratory, Centre for Health and Biological Sciences, +<br/><b>Mackenzie Presbyterian University, S o Paulo, S o Paulo, Brazil</b></td><td>('2708124', 'Chris Ashwin', 'chris ashwin')<br/>('39455300', 'Mark Brosnan', 'mark brosnan')</td><td>* tanja.wingenbach@bath.edu +</td></tr><tr><td>4e7ebf3c4c0c4ecc48348a769dd6ae1ebac3bf1b</td><td></td><td></td><td></td></tr><tr><td>4e0e49c280acbff8ae394b2443fcff1afb9bdce6</td><td>Automatic learning of gait signatures for people identification +<br/>F.M. Castro +<br/>Univ. of Malaga +<br/>fcastro<at>uma.es +<br/>M.J. Mar´ın-Jim´enez +<br/>Univ. of Cordoba +<br/>mjmarin<at>uco.es +<br/>N. Guil +<br/>Univ. of Malaga +<br/>nguil<at>uma.es +<br/>N. P´erez de la Blanca +<br/>Univ. of Granada +<br/>nicolas<at>ugr.es +</td><td></td><td></td></tr><tr><td>4e4e8fc9bbee816e5c751d13f0d9218380d74b8f</td><td></td><td></td><td></td></tr><tr><td>20a88cc454a03d62c3368aa1f5bdffa73523827b</td><td></td><td></td><td></td></tr><tr><td>20a432a065a06f088d96965f43d0055675f0a6c1</td><td>In: Proc. of the 25th Int. Conference on Artificial Neural Networks (ICANN) +<br/>Part II, LNCS 9887, pp. 80-87, Barcelona, Spain, September 2016 +<br/>The final publication is available at Springer via +<br/>http://dx.doi.org//10.1007/978-3-319-44781-0_10 +<br/>The Effects of Regularization on Learning Facial +<br/>Expressions with Convolutional Neural Networks +<br/><b></b><br/>Vogt-Koelln-Strasse 30, 22527 Hamburg, Germany +<br/>http://www.informatik.uni-hamburg.de/WTM +</td><td>('11634287', 'Tobias Hinz', 'tobias hinz')<br/>('1736513', 'Stefan Wermter', 'stefan wermter')</td><td>{4hinz,barros,wermter}@informatik.uni-hamburg.de +</td></tr><tr><td>20a3ce81e7ddc1a121f4b13e439c4cbfb01adfba</td><td>Sparse-MVRVMs Tree for Fast and Accurate +<br/>Head Pose Estimation in the Wild +<br/>Augmented Vision Research Group, +<br/><b>German Research Center for Arti cial Intelligence (DFKI</b><br/>Tripstaddterstr. 122, 67663 Kaiserslautern, Germany +<br/><b>Technical University of Kaiserslautern</b><br/>http://www.av.dfki.de +</td><td>('2585383', 'Mohamed Selim', 'mohamed selim')<br/>('1771057', 'Alain Pagani', 'alain pagani')<br/>('1807169', 'Didier Stricker', 'didier stricker')</td><td>{mohamed.selim,alain.pagani,didier.stricker}@dfki.uni-kl.de +</td></tr><tr><td>20b994a78cd1db6ba86ea5aab7211574df5940b3</td><td>Enriched Long-term Recurrent Convolutional Network +<br/>for Facial Micro-Expression Recognition +<br/><b>Faculty of Computing and Informatics, Multimedia University, Malaysia</b><br/><b>Faculty of Engineering, Multimedia University, Malaysia</b><br/><b>Shanghai Jiao Tong University, China</b></td><td>('30470673', 'Huai-Qian Khor', 'huai-qian khor')<br/>('2339975', 'John See', 'john see')<br/>('8131625', 'Weiyao Lin', 'weiyao lin')</td><td>Emails: 1hqkhor95@gmail.com, 2johnsee@mmu.edu.my, 3raphael@mmu.edu.my, 4wylin@sjtu.edu.cn +</td></tr><tr><td>2004afb2276a169cdb1f33b2610c5218a1e47332</td><td>Hindawi +<br/>Computational Intelligence and Neuroscience +<br/>Volume 2018, Article ID 3803627, 11 pages +<br/>https://doi.org/10.1155/2018/3803627 +<br/>Research Article +<br/>Deep Convolutional Neural Network Used in Single Sample per +<br/>Person Face Recognition +<br/><b>School of Information Engineering, Wuyi University, Jiangmen 529020, China</b><br/>Received 27 November 2017; Revised 23 May 2018; Accepted 26 July 2018; Published 23 August 2018 +<br/>Academic Editor: Jos´e Alfredo Hern´andez-P´erez +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Face recognition (FR) with single sample per person (SSPP) is a challenge in computer vision. Since there is only one sample to be +<br/>trained, it makes facial variation such as pose, illumination, and disguise difficult to be predicted. To overcome this problem, this paper +<br/>proposes a scheme combined traditional and deep learning (TDL) method to process the task. First, it proposes an expanding sample +<br/>method based on traditional approach. Compared with other expanding sample methods, the method can be used easily and +<br/>conveniently. Besides, it can generate samples such as disguise, expression, and mixed variation. Second, it uses transfer learning and +<br/>introduces a well-trained deep convolutional neural network (DCNN) model and then selects some expanding samples to fine-tune the +<br/>DCNN model. 0ird, the fine-tuned model is used to implement experiment. Experimental results on AR face database, Extend Yale B +<br/>face database, FERET face database, and LFW database demonstrate that TDL achieves the state-of-the-art performance in SSPP FR. +<br/>1. Introduction +<br/>As artificial +<br/>intelligence (AI) becomes more and more +<br/>popular, computer vision (CV) also has been proved to be +<br/>a very hot topic in academic such as face recognition [1], +<br/>facial expression recognition [2], and object recognition [3]. +<br/>It is well known that the basic and important foundation in +<br/>CV is that there are an amount of training samples. But in +<br/>actual scenarios such as immigration management, fugitive +<br/>tracing, and video surveillance, there may be only one +<br/>sample, which leads to single sample per person (SSPP) +<br/>problem such as gait recognition [4], face recognition (FR) +<br/>[5, 6], and low-resolution face recognition [7] in CV. +<br/>However, as the widely use of second-generation ID card +<br/>which is convenient to be collected, SSPP FR becomes one of +<br/>the most popular topics no matter in academic or in +<br/>industry. +<br/>Beymer and Poggio [8] proposed one example view +<br/>problem in 1996. In [8], it was researched that how to +<br/>perform face recognition (FR) using one example view. +<br/>Firstly, it exploited prior knowledge to generate multiple +<br/>virtual views. 0en, the example view and these multiple +<br/>virtual views were used as example views in a view-based, +<br/>pose-invariant +<br/>face recognizer. Later, SSPP FR became +<br/>a popular research topic at the beginning of the 21st century. +<br/>Recently, many methods have been proposed. Generally +<br/>speaking, these methods can be summarized in five basic +<br/>methods: direct method, generic learning method, patch- +<br/>based method, expanding sample method, and deep learning +<br/>(DL) method. Direct method does experiment based on the +<br/>SSPP directly by using an algorithm. Generic learning +<br/>method is the way that using an auxiliary dataset to build +<br/>a generic dataset from which some variation information +<br/>can be learned by single sample. Patch-based method par- +<br/>titions single sample into several patches first, then extracts +<br/>features on these patches, respectively, and does classifica- +<br/>tion finally. 0e expanding sample method is with some +<br/>special means such as perturbation-based method [9, 10], +<br/>photometric transforms, and geometric distortion [11] to +<br/>increase sample so that abundant training samples can be +<br/>used to process this task. 0e DL method uses the DL model +<br/>to perform the research. +<br/>Attracted by the good performance of DCNN, inspired +<br/>by [12] and driven by AI, in this paper, a scheme combined +</td><td>('9363278', 'Junying Zeng', 'junying zeng')<br/>('12054657', 'Xiaoxiao Zhao', 'xiaoxiao zhao')<br/>('2926767', 'Junying Gan', 'junying gan')<br/>('40552250', 'Chaoyun Mai', 'chaoyun mai')<br/>('1716453', 'Fan Wang', 'fan wang')<br/>('3003242', 'Yikui Zhai', 'yikui zhai')<br/>('9363278', 'Junying Zeng', 'junying zeng')</td><td>Correspondence should be addressed to Xiaoxiao Zhao; xiaoxiao-zhao@foxmail.com +</td></tr><tr><td>20e504782951e0c2979d9aec88c76334f7505393</td><td>Robust LSTM-Autoencoders for Face De-Occlusion +<br/>in the Wild +</td><td>('37182704', 'Fang Zhao', 'fang zhao')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')<br/>('39913117', 'Jian Zhao', 'jian zhao')<br/>('1898172', 'Wenhan Yang', 'wenhan yang')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>209324c152fa8fab9f3553ccb62b693b5b10fb4d</td><td>CROWDSOURCED VISUAL KNOWLEDGE REPRESENTATIONS +<br/>VISUAL GENOME +<br/>A THESIS +<br/>SUBMITTED TO THE DEPARTMENT OF COMPUTER SCIENCE +<br/>AND THE COMMITTEE ON GRADUATE STUDIES +<br/><b>OF STANFORD UNIVERSITY</b><br/>IN PARTIAL FULFILLMENT OF THE REQUIREMENTS +<br/>FOR THE DEGREE OF +<br/>MASTERS OF SCIENCE +<br/>March 2016 +</td><td>('2580593', 'Ranjay Krishna', 'ranjay krishna')</td><td></td></tr><tr><td>2050847bc7a1a0453891f03aeeb4643e360fde7d</td><td>Accio: A Data Set for Face Track Retrieval +<br/>in Movies Across Age +<br/><b>Istanbul Technical University, Istanbul, Turkey</b><br/><b>Karlsruhe Institute of Technology, Karlsruhe, Germany</b></td><td>('2398366', 'Esam Ghaleb', 'esam ghaleb')<br/>('2103464', 'Makarand Tapaswi', 'makarand tapaswi')<br/>('2256981', 'Ziad Al-Halah', 'ziad al-halah')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>{ghalebe, ekenel}@itu.edu.tr, {tapaswi, ziad.al-halah, rainer.stiefelhagen}@kit.edu +</td></tr><tr><td>20ade100a320cc761c23971d2734388bfe79f7c5</td><td>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +<br/>Subspace Clustering via Good Neighbors +</td><td>('1755872', 'Jufeng Yang', 'jufeng yang')<br/>('1780418', 'Jie Liang', 'jie liang')<br/>('39329211', 'Kai Wang', 'kai wang')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>202d8d93b7b747cdbd6e24e5a919640f8d16298a</td><td>Face Classification via Sparse Approximation +<br/><b>Bilgi University, Dolapdere, Istanbul, TR</b><br/><b>Bo gazici University, Istanbul, TR</b><br/><b>Y ld z Teknik University, Istanbul, TR</b></td><td>('2804969', 'Songul Albayrak', 'songul albayrak')</td><td></td></tr><tr><td>20767ca3b932cbc7b8112db21980d7b9b3ea43a3</td><td></td><td></td><td></td></tr><tr><td>20a16efb03c366fa4180659c2b2a0c5024c679da</td><td>SCREENING RULES FOR OVERLAPPING GROUP LASSO +<br/><b>Carnegie Mellon University</b><br/>Recently, to solve large-scale lasso and group lasso problems, +<br/>screening rules have been developed, the goal of which is to reduce +<br/>the problem size by efficiently discarding zero coefficients using simple +<br/>rules independently of the others. However, screening for overlapping +<br/>group lasso remains an open challenge because the overlaps between +<br/>groups make it infeasible to test each group independently. In this +<br/>paper, we develop screening rules for overlapping group lasso. To ad- +<br/>dress the challenge arising from groups with overlaps, we take into +<br/>account overlapping groups only if they are inclusive of the group +<br/>being tested, and then we derive screening rules, adopting the dual +<br/>polytope projection approach. This strategy allows us to screen each +<br/>group independently of each other. In our experiments, we demon- +<br/>strate the efficiency of our screening rules on various datasets. +<br/>1. Introduction. We propose efficient screening rules for regression +<br/>with the overlapping group lasso penalty. Our goal is to develop simple +<br/>rules to discard groups with zero coefficients in the optimization problem +<br/>with the following form: +<br/>(cid:13)(cid:13)βg +<br/>(cid:13)(cid:13)2 , +<br/>ng +<br/>(1.1) +<br/>min +<br/>(cid:107)y − Xβ(cid:107)2 +<br/>2 + λ +<br/>(cid:88) +<br/>g∈G +<br/>where X ∈ RN×J is the input data for J inputs and N samples, y ∈ RN×1 +<br/>is the output vector, β ∈ RJ×1 is the vector of regression coefficients, ng +<br/>is the size of group g, and λ is a regularization parameter that determines +<br/>the sparsity of β. In this setting, G represents a set of groups of coefficients, +<br/>defined a priori, and we allow arbitrary overlap between different groups, +<br/>hence “overlapping” group lasso. Overlapping group lasso is a general model +<br/>that subsumes lasso (Tibshirani, 1996), group lasso (Yuan and Lin, 2006), +<br/>sparse group lasso (Simon et al., 2013), composite absolute penalties (Zhao, +<br/>Rocha and Yu, 2009), and tree lasso (Zhao, Rocha and Yu, 2009; Kim et al., +<br/>2012) with (cid:96)1/(cid:96)2 penalty because they are a specific form of overlapping +<br/>group lasso. +<br/>In this paper, we do not consider the latent group lasso proposed by +<br/>Jacob et al. (Jacob, Obozinski and Vert, 2009), where support is defined +<br/>by the union of groups with nonzero coefficients. Instead, we consider the +</td><td>('1918078', 'Seunghak Lee', 'seunghak lee')<br/>('1752601', 'Eric P. Xing', 'eric p. xing')</td><td></td></tr><tr><td>205b34b6035aa7b23d89f1aed2850b1d3780de35</td><td>504 +<br/>2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +<br/>978-1-4799-2893-4/14/$31.00 ©2014 IEEE +<br/>†Shenzhen Key Lab. of Information Sci&Tech, +<br/><b>Nagaoka University of Technology, Japan</b><br/>RECOGNITION +<br/>1. INTRODUCTION +</td><td></td><td></td></tr><tr><td>20c2a5166206e7ffbb11a23387b9c5edf42b5230</td><td></td><td></td><td></td></tr><tr><td>20e505cef6d40f896e9508e623bfc01aa1ec3120</td><td>Fast Online Incremental Attribute-based Object +<br/>Classification using Stochastic Gradient Descent and Self- +<br/>Organizing Incremental Neural Network +<br/>Department of Computational Intelligence and Systems Science, +<br/><b>Tokyo Institute of Technology</b><br/>4259 Nagatsuta, Midori-ku, Yokohama, 226-8503 JAPAN +</td><td>('2641676', 'Sirinart Tangruamsub', 'sirinart tangruamsub')<br/>('1711160', 'Aram Kawewong', 'aram kawewong')<br/>('1727786', 'Osamu Hasegawa', 'osamu hasegawa')</td><td>(tangruamsub.s.aa, kawewong.a.aa, hasegawa.o.aa)@m.titech.ac.jp +</td></tr><tr><td>205e4d6e0de81c7dd6c83b737ffdd4519f4f7ffa</td><td>A Model-Based Facial Expression Recognition +<br/>Algorithm using Principal Components Analysis +<br/>N. Vretos, N. Nikolaidis and I.Pitas +<br/><b>Informatics and Telematics Institute</b><br/>Centre for Research and Technology Hellas, Greece +<br/><b>Aristotle University of Thessaloniki</b><br/>Thessaloniki 54124, Greece Tel,Fax: +30-2310996304 +</td><td></td><td>e-mail: vretos,nikolaid,pitas@aiia.csd.auth.gr +</td></tr><tr><td>2098983dd521e78746b3b3fa35a22eb2fa630299</td><td></td><td></td><td></td></tr><tr><td>20b437dc4fc44c17f131713ffcbb4a8bd672ef00</td><td>Head pose tracking from RGBD sensor based on +<br/>direct motion estimation +<br/><b>Warsaw University of Technology, Poland</b></td><td>('1899063', 'Adam Strupczewski', 'adam strupczewski')<br/>('2393538', 'Marek Kowalski', 'marek kowalski')<br/>('1930272', 'Jacek Naruniec', 'jacek naruniec')</td><td></td></tr><tr><td>206e24f7d4b3943b35b069ae2d028143fcbd0704</td><td>Learning Structure and Strength of CNN Filters for Small Sample Size Training +<br/>IIIT-Delhi, India +</td><td>('3390448', 'Rohit Keshari', 'rohit keshari')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('39129417', 'Richa Singh', 'richa singh')</td><td>{rohitk, mayank, rsingh}@iiitd.ac.in +</td></tr><tr><td>208a2c50edb5271a050fa9f29d3870f891daa4dc</td><td>http://www.journalofvision.org/content/11/13/24 +<br/>The resolution of facial expressions of emotion +<br/>Aleix M. Martinez +<br/><b>The Ohio State University, Columbus, OH, USA</b><br/><b>The Ohio State University, Columbus, OH, USA</b><br/><b>Much is known on how facial expressions of emotion are produced, including which individual muscles are most active in</b><br/>each expression. Yet, little is known on how this information is interpreted by the human visual system. This paper presents +<br/>a systematic study of the image dimensionality of facial expressions of emotion. In particular, we investigate how recognition +<br/>degrades when the resolution of the image (i.e., number of pixels when seen as a 5.3 by 8 degree stimulus) is reduced. We +<br/>show that recognition is only impaired in practice when the image resolution goes below 20 30 pixels. A study of the +<br/>confusion tables demonstrates that each expression of emotion is consistently confused by a small set of alternatives and +<br/>that the confusion is not symmetric, i.e., misclassifying emotion a as b does not imply we will mistake b for a. This +<br/>asymmetric pattern is consistent over the different image resolutions and cannot be explained by the similarity of muscle +<br/>activation. Furthermore, although women are generally better at recognizing expressions of emotion at all resolutions, the +<br/>asymmetry patterns are the same. We discuss the implications of these results for current models of face perception. +<br/>Keywords: resolution, facial expressions, emotion +<br/>http://www.journalofvision.org/content/11/13/24, doi:10.1167/11.13.24. +<br/>Introduction +<br/>Emotions are fundamental in studies of cognitive science +<br/>(Damassio, 1995), neuroscience (LeDoux, 2000), social +<br/>psychology (Adolphs, 2003), sociology (Massey, 2002), +<br/>economics (Connolly & Zeelenberg, 2002), human evo- +<br/>lution (Schmidt & Cohn, 2001), and engineering and +<br/>computer science (Pentland, 2000). Emotional states and +<br/>emotional analysis are known to influence or mediate +<br/>behavior and cognitive processing. Many of these emo- +<br/>tional processes may be hidden to an outside observer, +<br/>whereas others are visible through facial expressions of +<br/>emotion. +<br/>Facial expressions of emotion are a consequence of the +<br/>movement of the muscles underneath the skin of our face +<br/>(Duchenne, 1862/1990). The movement of these muscles +<br/>causes the skin of the face to deform in ways that an +<br/>external observer can use to interpret the emotion of that +<br/>person. Each muscle employed to create these facial +<br/>constructs is referred to as an Action Unit (AU). Ekman +<br/>and Friesen (1978) identified those AUs responsible for +<br/>generating the emotions most commonly seen in the +<br/>majority of culturesVanger, sadness, fear, surprise, +<br/>happiness, and disgust. For example, happiness generally +<br/>involves an upper–backward movement of the mouth +<br/>corners; while the mouth is upturned (to produce the +<br/>smile), the cheeks lift and the upper corner of the eyes +<br/>wrinkle. This is known as the Duchenne (1862/1990) +<br/>smile. It requires the activation of two facial muscles: +<br/>the zygomatic major (AU 12) to raise the corners of the +<br/>mouth and the orbicularis oculi (AU 42) to uplift the +<br/>cheeks and form the eye corner wrinkles. The muscles and +<br/>mechanisms used to produce the abovementioned facial +<br/>expressions of emotion are now quite well understood and +<br/>it has been shown that the AUs used in each expression +<br/>are relatively consistent from person to person and among +<br/>distinct cultures (Burrows & Cohn, 2009). +<br/>Yet, as much as we understand the generative process +<br/>of facial expressions of emotion, much still needs to be +<br/>learned about their interpretation by our cognitive system. +<br/>Thus, an important open problem is to define the +<br/>computational (cognitive) space of facial expressions of +<br/>emotion of the human visual system. In the present paper, +<br/>we study the limits of this visual processing of facial +<br/>expressions of emotion and what it tells us about how +<br/>emotions are represented and recognized by our visual +<br/>system. Note that the term “computational space” is used +<br/>here to specify the combination of features (dimensions) +<br/>used by the cognitive system to determine (i.e., analyze +<br/>and classify) +<br/>for each facial +<br/>expression of emotion. +<br/>the appropriate label +<br/>To properly address the problem stated in the preceding +<br/>paragraph, it is worth recalling that some facial expressions +<br/>of emotion may have evolved to enhance or reduce our +<br/>sensory inputs (Susskind et al., 2008). For example, fear is +<br/>associated with a facial expression with open mouth, +<br/>nostrils, and eyes and an inhalation of air, as if to enhance +<br/>the perception of our environment, while the expression of +<br/>disgust closes these channels (Chapman, Kim, Susskind, +<br/>& Anderson, 2009). Other emotions, though, may have +<br/>evolved for communication purposes (Schmidt & Cohn, +<br/>2001). Under this assumption, +<br/>the evolution of this +<br/>capacity to express emotions had to be accompanied by +<br/>doi: 10.1167/11.13.24 +<br/>Received January 25, 2011; published November 30, 2011 +<br/>ISSN 1534-7362 * ARVO +<br/>Downloaded From: http://jov.arvojournals.org/pdfaccess.ashx?url=/data/journals/jov/932792/ on 06/20/2017 </td><td>('2323717', 'Shichuan Du', 'shichuan du')</td><td></td></tr><tr><td>207798603e3089a1c807c93e5f36f7767055ec06</td><td>Modeling the Correlation between +<br/>Modality Semantics and Facial Expressions +<br/>* Key Laboratory of Pervasive Computing, Ministry of Education +<br/>Tsinghua National Laboratory for Information Science and Technology (TNList) +<br/><b>Tsinghua University, Beijing 100084, China</b><br/><b>Tsinghua-CUHK Joint Research Center for Media Sciences, Technologies and Systems</b><br/><b>Graduate School at Shenzhen, Tsinghua University, Shenzhen 518055, China</b><br/>† Human-Computer Communications Laboratory, Department of Systems Engineering and Engineering Management, +<br/><b>The Chinese University of Hong Kong, Hong Kong SAR, China</b></td><td>('25714033', 'Jia Jia', 'jia jia')<br/>('37783013', 'Xiaohui Wang', 'xiaohui wang')<br/>('3860920', 'Zhiyong Wu', 'zhiyong wu')<br/>('7239047', 'Lianhong Cai', 'lianhong cai')</td><td>Contact E-mail: # zywu@sz.tsinghua.edu.cn, * jjia@tsinghua.edu.cn +</td></tr><tr><td>20be15dac7d8a5ba4688bf206ad24cab57d532d6</td><td>Face Shape Recovery and Recognition Using a +<br/>Surface Gradient Based Statistical Model +<br/>1 Centro de Investigaci´on y Estudios Avanzados del I.P.N., Ramos Arizpe 25900, +<br/>Coahuila, Mexico +<br/><b>The University of York, Heslington, York YO10 5DD, United Kingdom</b></td><td>('1679753', 'Edwin R. Hancock', 'edwin r. hancock')</td><td>mario.castelan@cinvestav.edu.mx +<br/>erh@cs.york.ac.uk +</td></tr><tr><td>2059d2fecfa61ddc648be61c0cbc9bc1ad8a9f5b</td><td>TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. 23, NO. 4, APRIL 2015 +<br/>Co-Localization of Audio Sources in Images Using +<br/>Binaural Features and Locally-Linear Regression +<br/>∗ INRIA Grenoble Rhˆone-Alpes, Montbonnot Saint-Martin, France +<br/>† Univ. Grenoble Alpes, GIPSA-Lab, France +<br/>‡ Dept. Electrical Eng., Technion-Israel Inst. of Technology, Haifa, Israel +</td><td>('3307172', 'Antoine Deleforge', 'antoine deleforge')</td><td></td></tr><tr><td>206fbe6ab6a83175a0ef6b44837743f8d5f9b7e8</td><td></td><td></td><td></td></tr><tr><td>2042aed660796b14925db17c0a8b9fbdd7f3ebac</td><td>Saliency in Crowd +<br/>Department of Electrical and Computer Engineering +<br/><b>National University of Singapore, Singapore</b></td><td>('40452812', 'Ming Jiang', 'ming jiang')<br/>('1946538', 'Juan Xu', 'juan xu')<br/>('3243515', 'Qi Zhao', 'qi zhao')</td><td>eleqiz@nus.edu.sg +</td></tr><tr><td>202dc3c6fda654aeb39aee3e26a89340fb06802a</td><td>Spatio-Temporal Instance Learning: +<br/>Action Tubes from Class Supervision +<br/><b>University of Amsterdam</b></td><td>('2606260', 'Pascal Mettes', 'pascal mettes')</td><td></td></tr><tr><td>20111924fbf616a13d37823cd8712a9c6b458cd6</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 130 – No.11, November2015 +<br/>Linear Regression Line based Partial Face Recognition +<br/>Naveena M. +<br/>Department of Studies in +<br/>Computer Science, +<br/>Manasagagothri, +<br/>Mysore. +<br/>Department of Studies in +<br/>Computer Science, +<br/>Manasagagothri, +<br/>Mysore. +<br/>P. Nagabhushan +<br/>Department of Studies in +<br/>Computer Science, +<br/>Manasagagothri, +<br/>Mysore. +<br/>images. In +</td><td>('33377948', 'G. Hemantha Kumar', 'g. hemantha kumar')</td><td></td></tr><tr><td>20ebbcb6157efaacf7a1ceb99f2f3e2fdf1384e6</td><td>Appears in the Second International Conference on Audio- and Video-based Biometric Person Authentication, AVBPA’99, Washington D. C. USA, March 22-24, 1999. +<br/>Comparative Assessment of Independent Component +<br/>Analysis (ICA) for Face Recognition +<br/><b>George Mason University</b><br/><b>University Drive, Fairfax, VA 22030-4444, USA</b><br/> +</td><td>('39664966', 'Chengjun Liu', 'chengjun liu')<br/>('1781577', 'Harry Wechsler', 'harry wechsler')</td><td>@cs.gmu.edu +</td></tr><tr><td>20532b1f80b509f2332b6cfc0126c0f80f438f10</td><td>A deep matrix factorization method for learning +<br/>attribute representations +<br/>Bj¨orn W. Schuller, Senior member, IEEE +</td><td>('2814229', 'George Trigeorgis', 'george trigeorgis')<br/>('2732737', 'Konstantinos Bousmalis', 'konstantinos bousmalis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td></td></tr><tr><td>205af28b4fcd6b569d0241bb6b255edb325965a4</td><td>Intel Serv Robotics (2008) 1:143–157 +<br/>DOI 10.1007/s11370-007-0014-z +<br/>SPECIAL ISSUE +<br/>Facial expression recognition and tracking for intelligent human-robot +<br/>interaction +<br/>Received: 27 June 2007 / Accepted: 6 December 2007 / Published online: 23 January 2008 +<br/>© Springer-Verlag 2008 +</td><td>('1716880', 'Y. Yang', 'y. yang')</td><td></td></tr><tr><td>20cfb4136c1a984a330a2a9664fcdadc2228b0bc</td><td>Sparse Coding Trees with Application to Emotion Classification +<br/><b>Harvard University, Cambridge, MA</b></td><td>('3144257', 'Hsieh-Chung Chen', 'hsieh-chung chen')<br/>('2512314', 'Marcus Z. Comiter', 'marcus z. comiter')<br/>('1731308', 'H. T. Kung', 'h. t. kung')<br/>('1841852', 'Bradley McDanel', 'bradley mcdanel')</td><td></td></tr><tr><td>20c02e98602f6adf1cebaba075d45cef50de089f</td><td>Video Jigsaw: Unsupervised Learning of Spatiotemporal Context for Video +<br/>Action Recognition +<br/><b>Georgia Institute of Technology</b><br/><b>Carnegie Mellon University</b><br/>Irfan Essa +<br/><b>Georgia Institute of Technology</b></td><td>('2308598', 'Unaiza Ahsan', 'unaiza ahsan')<br/>('37714701', 'Rishi Madhok', 'rishi madhok')</td><td>uahsan3@gatech.edu +<br/>rmadhok@andrew.cmu.edu +<br/>irfan@gatech.edu +</td></tr><tr><td>2020e8c0be8fa00d773fd99b6da55029a6a83e3d</td><td>An Evaluation of the Invariance Properties +<br/>of a Biologically-Inspired System +<br/>for Unconstrained Face Recognition +<br/><b>Massachusetts Institute of Technology, Cambridge, MA 02139, USA</b><br/><b>Rowland Institute at Harvard, Cambridge, MA 02142, USA</b></td><td>('30017846', 'Nicolas Pinto', 'nicolas pinto')</td><td>pinto@mit.edu +<br/>cox@rowland.harvard.edu +</td></tr><tr><td>20a0b23741824a17c577376fdd0cf40101af5880</td><td>Learning to track for spatio-temporal action localization +<br/>Zaid Harchaouia,b +<br/>b NYU +<br/>a Inria∗ +</td><td>('2492127', 'Philippe Weinzaepfel', 'philippe weinzaepfel')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td>firstname.lastname@inria.fr +</td></tr><tr><td>18c72175ddbb7d5956d180b65a96005c100f6014</td><td>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 23, NO. 6, +<br/>JUNE 2001 +<br/>643 +<br/>From Few to Many: Illumination Cone +<br/>Models for Face Recognition under +<br/>Variable Lighting and Pose +</td><td>('3230391', 'Athinodoros S. Georghiades', 'athinodoros s. georghiades')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')<br/>('1765887', 'David J. Kriegman', 'david j. kriegman')</td><td></td></tr><tr><td>18636347b8741d321980e8f91a44ee054b051574</td><td>978-1-4244-5654-3/09/$26.00 ©2009 IEEE +<br/>37 +<br/>ICIP 2009 +</td><td></td><td></td></tr><tr><td>18206e1b988389eaab86ef8c852662accf3c3663</td><td></td><td></td><td></td></tr><tr><td>189b1859f77ddc08027e1e0f92275341e5c0fdc6</td><td>Sparse Representations and Distance Learning for +<br/>Attribute based Category Recognition +<br/>1 Center for Imaging Science, 2 Department of Computer Engineering +<br/><b>Rochester Institute of Technology, Rochester, NY</b></td><td>('2272443', 'Grigorios Tsagkatakis', 'grigorios tsagkatakis')</td><td>{gxt6260, andreas.savakis}@rit.edu +</td></tr><tr><td>18a9f3d855bd7728ed4f988675fa9405b5478845</td><td>ISSN: 0976-9102 (ONLINE) +<br/>DOI: 10.21917/ijivp.2013.0103 +<br/> ICTACT JOURNAL ON IMAGE AND VIDEO PROCESSING, NOVEMBER 2013, VOLUME: 04, ISSUE: 02 +<br/>AN ILLUMINATION INVARIANT TEXTURE BASED FACE RECOGNITION +<br/><b>J. P. College of Engineering, India</b><br/><b>Manonmaniam Sundaranar University, India</b><br/><b>St. Xavier s Catholic College of Engineering, India</b></td><td>('2792485', 'K. Meena', 'k. meena')<br/>('3311251', 'A. Suruliandi', 'a. suruliandi')<br/>('1998086', 'Reena Rose', 'reena rose')</td><td>E-mail: meen.nandhu@gmail.com +<br/>E-mail: suruliandi@yahoo.com +<br/>E-mail: mailtoreenarose@yahoo.in +</td></tr><tr><td>181045164df86c72923906aed93d7f2f987bce6c</td><td>RHEINISCH-WESTFÄLISCHE TECHNISCHE +<br/>HOCHSCHULE AACHEN +<br/>KNOWLEDGE-BASED SYSTEMS GROUP +<br/>Detection and Recognition of Human +<br/>Faces using Random Forests for a +<br/>Mobile Robot +<br/>MASTER OF SCIENCE THESIS +<br/>MATRICULATION NUMBER: 26 86 51 +<br/>SUPERVISOR: +<br/>SECOND SUPERVISOR: +<br/>PROF. ENRICO BLANZIERI, PH. D. +<br/>ADVISERS: +</td><td>('1779592', 'GERHARD LAKEMEYER', 'gerhard lakemeyer')<br/>('2181555', 'VAISHAK BELLE', 'vaishak belle')<br/>('1779592', 'GERHARD LAKEMEYER', 'gerhard lakemeyer')<br/>('1686596', 'STEFAN SCHIFFER', 'stefan schiffer')<br/>('1879646', 'THOMAS DESELAERS', 'thomas deselaers')</td><td></td></tr><tr><td>18166432309000d9a5873f989b39c72a682932f5</td><td>LEARNING A WARPED SUBSPACE MODEL OF FACES +<br/>WITH IMAGES OF UNKNOWN POSE AND +<br/>ILLUMINATION +<br/><b>GRASP Laboratory, University of Pennsylvania, 3330 Walnut Street, Philadelphia, PA, USA</b><br/>Keywords: +</td><td>('2720935', 'Jihun Ham', 'jihun ham')<br/>('1732066', 'Daniel D. Lee', 'daniel d. lee')</td><td>jhham@seas.upenn.edu, ddlee@seas.upenn.edu +</td></tr><tr><td>18d5b0d421332c9321920b07e0e8ac4a240e5f1f</td><td>Collaborative Representation Classification +<br/>Ensemble for Face Recognition +</td><td>('2972883', 'Suah Kim', 'suah kim')<br/>('2434811', 'Run Cui', 'run cui')<br/>('1730037', 'Hyoung Joong Kim', 'hyoung joong kim')</td><td></td></tr><tr><td>18d51a366ce2b2068e061721f43cb798177b4bb7</td><td>Cognition and Emotion +<br/>ISSN: 0269-9931 (Print) 1464-0600 (Online) Journal homepage: http://www.tandfonline.com/loi/pcem20 +<br/>Looking into your eyes: observed pupil size +<br/>influences approach-avoidance responses +<br/>eyes: observed pupil size influences approach-avoidance responses, Cognition and Emotion, DOI: +<br/>10.1080/02699931.2018.1472554 +<br/>To link to this article: https://doi.org/10.1080/02699931.2018.1472554 +<br/>View supplementary material +<br/>Published online: 11 May 2018. +<br/>Submit your article to this journal +<br/>View related articles +<br/>View Crossmark data +<br/>Full Terms & Conditions of access and use can be found at +<br/>http://www.tandfonline.com/action/journalInformation?journalCode=pcem20 +</td><td>('47930228', 'Marco Brambilla', 'marco brambilla')<br/>('41074530', 'Marco Biella', 'marco biella')<br/>('47930228', 'Marco Brambilla', 'marco brambilla')<br/>('41074530', 'Marco Biella', 'marco biella')</td><td></td></tr><tr><td>18c6c92c39c8a5a2bb8b5673f339d3c26b8dcaae</td><td>Learning invariant representations and applications +<br/>to face verification +<br/>Center for Brains, Minds and Machines +<br/><b>McGovern Institute for Brain Research</b><br/><b>Massachusetts Institute of Technology</b><br/>Cambridge MA 02139 +</td><td>('1694846', 'Qianli Liao', 'qianli liao')</td><td>lql@mit.edu, jzleibo@mit.edu, tp@ai.mit.edu +</td></tr><tr><td>185263189a30986e31566394680d6d16b0089772</td><td>Efficient Annotation of Objects for Video Analysis +<br/>Thesis submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/>MS in Computer Science and Engineering +<br/>by +<br/>Research +<br/>by +<br/>Sirnam Swetha +<br/>201303014 +<br/><b>International Institute of Information Technology</b><br/>Hyderabad - 500 032, INDIA +<br/>June 2018 +</td><td></td><td>sirnam.swetha@research.iiit.ac.in +</td></tr><tr><td>1885acea0d24e7b953485f78ec57b2f04e946eaf</td><td>Combining Local and Global Features for 3D Face Tracking +<br/>Megvii (face++) Research +</td><td>('40448951', 'Pengfei Xiong', 'pengfei xiong')<br/>('1775836', 'Guoqing Li', 'guoqing li')<br/>('3756559', 'Yuhang Sun', 'yuhang sun')</td><td>{xiongpengfei, liguoqing, sunyuhang}@megvii.com +</td></tr><tr><td>184750382fe9b722e78d22a543e852a6290b3f70</td><td></td><td></td><td></td></tr><tr><td>18b9dc55e5221e704f90eea85a81b41dab51f7da</td><td>Attention-based Temporal Weighted +<br/>Convolutional Neural Network for +<br/>Action Recognition +<br/><b>Xi an Jiaotong University, Xi an, Shannxi 710049, P.R.China</b><br/>2HERE Technologies, Chicago, IL 60606, USA +<br/>3Alibaba Group, Hangzhou, Zhejiang 311121, P.R.China +<br/>4Microsoft Research, Redmond, WA 98052, USA +</td><td>('14800230', 'Jinliang Zang', 'jinliang zang')<br/>('40367806', 'Le Wang', 'le wang')<br/>('46324995', 'Qilin Zhang', 'qilin zhang')<br/>('1786361', 'Zhenxing Niu', 'zhenxing niu')<br/>('1745420', 'Gang Hua', 'gang hua')<br/>('1715389', 'Nanning Zheng', 'nanning zheng')</td><td></td></tr><tr><td>18a849b1f336e3c3b7c0ee311c9ccde582d7214f</td><td>Int J Comput Vis +<br/>DOI 10.1007/s11263-012-0564-1 +<br/>Efficiently Scaling up Crowdsourced Video Annotation +<br/>A Set of Best Practices for High Quality, Economical Video Labeling +<br/>Received: 31 October 2011 / Accepted: 20 August 2012 +<br/>© Springer Science+Business Media, LLC 2012 +</td><td>('1856025', 'Carl Vondrick', 'carl vondrick')</td><td></td></tr><tr><td>18cd79f3c93b74d856bff6da92bfc87be1109f80</td><td>International Journal of Advances in Engineering & Technology, May 2012. +<br/>©IJAET ISSN: 2231-1963 +<br/>AN APPLICATION TO HUMAN FACE PHOTO-SKETCH +<br/>SYNTHESIS AND RECOGNITION +<br/>1Student and 2Professor & Head, +<br/><b>Bharti Vidyapeeth Deemed University, Pune, India</b></td><td>('35541779', 'Amit R. Sharma', 'amit r. sharma')<br/>('2731104', 'Prakash. R. Devale', 'prakash. r. devale')</td><td></td></tr><tr><td>182470fd0c18d0c5979dff75d089f1da176ceeeb</td><td>A Multimodal Annotation Schema for Non-Verbal Affective +<br/>Analysis in the Health-Care Domain +<br/>Federico M. Sukno +<br/>Adrià Ruiz +<br/>Department of Information and Communication Technologies +<br/><b>Pompeu Fabra University, Spain</b><br/>Human-Centered Multimedia +<br/><b>Augsburg University, Germany</b><br/>Louisa Praagst +<br/><b>Institute of Communications Engineering</b><br/><b>Ulm University, Germany</b><br/><b>Information Technologies Institute</b><br/>Centre for Research & Technology Hellas, Greece +</td><td>('33451278', 'Mónica Domínguez', 'mónica domínguez')<br/>('34326647', 'Dominik Schiller', 'dominik schiller')<br/>('2565410', 'Florian Lingenfelser', 'florian lingenfelser')<br/>('8632684', 'Ekeni Kamateri', 'ekeni kamateri')</td><td></td></tr><tr><td>1862cb5728990f189fa91c67028f6d77b5ac94f6</td><td>Speeding Up Tracking by Ignoring Features +<br/>Hamdi Dibeklio˘glu +<br/><b>Pattern Recognition and Bioinformatics Group, Delft University of Technology</b><br/>Mekelweg 4, 2628 CD Delft, The Netherlands +</td><td>('2883723', 'Lu Zhang', 'lu zhang')<br/>('1803520', 'Laurens van der Maaten', 'laurens van der maaten')</td><td>{lu.zhang, h.dibeklioglu, l.j.p.vandermaaten}@tudelft.nl +</td></tr><tr><td>1862bfca2f105fddfc79941c90baea7db45b8b16</td><td>Annotator Rationales for Visual Recognition +<br/><b>University of Texas at Austin</b></td><td>('7408951', 'Jeff Donahue', 'jeff donahue')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>{jdd,grauman}@cs.utexas.edu +</td></tr><tr><td>1886b6d9c303135c5fbdc33e5f401e7fc4da6da4</td><td>Knowledge Guided Disambiguation for Large-Scale +<br/>Scene Classification with Multi-Resolution CNNs +</td><td>('39709927', 'Limin Wang', 'limin wang')<br/>('2072196', 'Sheng Guo', 'sheng guo')<br/>('1739171', 'Weilin Huang', 'weilin huang')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('40285012', 'Yu Qiao', 'yu qiao')</td><td></td></tr><tr><td>1888bf50fd140767352158c0ad5748b501563833</td><td>PA R T 1 +<br/>THE BASICS +</td><td></td><td></td></tr><tr><td>187d4d9ba8e10245a34f72be96dd9d0fb393b1aa</td><td>GAIDON et al.: MINING VISUAL ACTIONS FROM MOVIES +<br/>Mining visual actions from movies +<br/>http://lear.inrialpes.fr/people/gaidon/ +<br/>Marcin Marszałek2 +<br/>http://www.robots.ox.ac.uk/~marcin/ +<br/>http://lear.inrialpes.fr/people/schmid/ +<br/>1 LEAR +<br/>INRIA, LJK +<br/>Grenoble, France +<br/>2 Visual Geometry Group +<br/><b>University of Oxford</b><br/>Oxford, UK +</td><td>('1799820', 'Adrien Gaidon', 'adrien gaidon')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>182f3aa4b02248ff9c0f9816432a56d3c8880706</td><td>Sparse Coding for Classification via Discrimination Ensemble∗ +<br/>1School of Computer Science & Engineering, South China Univ. of Tech., Guangzhou 510006, China +<br/>2School of Automation Science & Engineering, South China Univ. of Tech., Guangzhou 510006, China +<br/><b>National University of Singapore, Singapore</b></td><td>('2217653', 'Yuhui Quan', 'yuhui quan')<br/>('1725160', 'Yong Xu', 'yong xu')<br/>('2111796', 'Yuping Sun', 'yuping sun')<br/>('34881546', 'Yan Huang', 'yan huang')<br/>('39689301', 'Hui Ji', 'hui ji')</td><td>{csyhquan@scut.edu.cn, yxu@scut.edu.cn, ausyp@scut.edu.cn, matjh@nus.edu.sg} +</td></tr><tr><td>18941b52527e6f15abfdf5b86a0086935706e83b</td><td>DeepGUM: Learning Deep Robust Regression with a +<br/>Gaussian-Uniform Mixture Model +<br/>1 Inria Grenoble Rhˆone-Alpes, Montbonnot-Saint-Martin, France, +<br/><b>University of Granada, Granada, Spain</b><br/><b>University of Trento, Trento, Italy</b></td><td>('2793152', 'Pablo Mesejo', 'pablo mesejo')<br/>('1780201', 'Xavier Alameda-Pineda', 'xavier alameda-pineda')<br/>('1794229', 'Radu Horaud', 'radu horaud')</td><td>firstname.name@inria.fr +</td></tr><tr><td>185360fe1d024a3313042805ee201a75eac50131</td><td>299 +<br/>Person De-Identification in Videos +</td><td>('35624289', 'Prachi Agrawal', 'prachi agrawal')<br/>('1729020', 'P. J. Narayanan', 'p. j. narayanan')</td><td></td></tr><tr><td>1824b1ccace464ba275ccc86619feaa89018c0ad</td><td>One Millisecond Face Alignment with an Ensemble of Regression Trees +<br/><b>KTH, Royal Institute of Technology</b><br/>Computer Vision and Active Perception Lab +<br/>Teknikringen 14, Stockholm, Sweden +</td><td>('2626422', 'Vahid Kazemi', 'vahid kazemi')<br/>('1736906', 'Josephine Sullivan', 'josephine sullivan')</td><td>{vahidk,sullivan}@csc.kth.se +</td></tr><tr><td>18dfc2434a95f149a6cbb583cca69a98c9de9887</td><td></td><td></td><td></td></tr><tr><td>27a00f2490284bc0705349352d36e9749dde19ab</td><td>VoxCeleb2: Deep Speaker Recognition +<br/>Visual Geometry Group, Department of Engineering Science, +<br/><b>University of Oxford, UK</b></td><td>('2863890', 'Joon Son Chung', 'joon son chung')<br/>('19263506', 'Arsha Nagrani', 'arsha nagrani')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>{joon,arsha,az}@robots.ox.ac.uk +</td></tr><tr><td>271e2856e332634eccc5e80ba6fa9bbccf61f1be</td><td>3D Spatio-Temporal Face Recognition Using Dynamic Range Model Sequences +<br/>Department of Computer Science +<br/><b>State University of New York at Binghamton, Binghamton, NY</b></td><td>('1681656', 'Yi Sun', 'yi sun')<br/>('8072251', 'Lijun Yin', 'lijun yin')</td><td></td></tr><tr><td>27846b464369095f4909f093d11ed481277c8bba</td><td>Journal of Signal and Information Processing, 2017, 8, 99-112 +<br/>http://www.scirp.org/journal/jsip +<br/>ISSN Online: 2159-4481 +<br/>ISSN Print: 2159-4465 +<br/>Real-Time Face Detection and Recognition in +<br/>Complex Background +<br/><b>Illinois Institute of Technology, Chicago, Illinois, USA</b><br/>How to cite this paper: Zhang, X., Gon- +<br/>not, T. and Saniie, J. (2017) Real-Time +<br/>Face Detection and Recognition in Com- +<br/>plex Background. Journal of Signal and +<br/>Information Processing, 8, 99-112. +<br/>https://doi.org/10.4236/jsip.2017.82007 +<br/>Received: March 25, 2017 +<br/>Accepted: May 16, 2017 +<br/>Published: May 19, 2017 +<br/>Copyright © 2017 by authors and +<br/>Scientific Research Publishing Inc. +<br/>This work is licensed under the Creative +<br/>Commons Attribution International +<br/>License (CC BY 4.0). +<br/>http://creativecommons.org/licenses/by/4.0/ +<br/> +<br/>Open Access +</td><td>('1682913', 'Xin Zhang', 'xin zhang')<br/>('2324553', 'Thomas Gonnot', 'thomas gonnot')<br/>('1691321', 'Jafar Saniie', 'jafar saniie')</td><td></td></tr><tr><td>27eb7a6e1fb6b42516041def6fe64bd028b7614d</td><td>Joint Unsupervised Deformable Spatio-Temporal Alignment of Sequences +<br/><b>Imperial College London, UK</b><br/><b>University of Twente, The Netherlands</b><br/><b>Center for Machine Vision and Signal Analysis, University of Oulu, Finland</b></td><td>('1786302', 'Lazaros Zafeiriou', 'lazaros zafeiriou')<br/>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('1694605', 'Maja Pantic', 'maja pantic')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>⋆{l.zafeiriou12, e.antonakos, s.zafeiriou, m.pantic}@imperial.ac.uk, †PanticM@cs.utwente.nl +</td></tr><tr><td>2717998d89d34f45a1cca8b663b26d8bf10608a9</td><td>Real-time Action Recognition with Enhanced Motion Vector CNNs +<br/><b>Shenzhen key lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b><br/><b>Key Laboratory of Embedded System and Service Computing, Ministry of Education, Tongji University, Shanghai, China</b><br/>3Computer Vision Lab, ETH Zurich, Switzerland +</td><td>('3047890', 'Bowen Zhang', 'bowen zhang')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('1915826', 'Zhe Wang', 'zhe wang')<br/>('33427555', 'Yu Qiao', 'yu qiao')<br/>('2774427', 'Hanli Wang', 'hanli wang')</td><td></td></tr><tr><td>27c66b87e0fbb39f68ddb783d11b5b7e807c76e8</td><td>Fast Simplex-HMM for One-Shot Learning Activity Recognition +<br/><b>Zaragoza University</b><br/>Zaragoza, Spain. +<br/><b>Kingston University</b><br/>London,UK. +</td><td>('1783769', 'Carlos Medrano', 'carlos medrano')<br/>('1687002', 'Dimitrios Makris', 'dimitrios makris')</td><td>[mrodrigo, corrite, ctmedra]@unizar.es +<br/>D.Makris@kingston.ac.uk +</td></tr><tr><td>27a0a7837f9114143717fc63294a6500565294c2</td><td>Face Recognition in Unconstrained Environments: A +<br/>Comparative Study +<br/>To cite this version: +<br/>Environments: A Comparative Study: . Workshop on Faces in ’Real-Life’ Images: Detection, +<br/>Alignment, and Recognition, Oct 2008, Marseille, France. 2008. <inria-00326730> +<br/>HAL Id: inria-00326730 +<br/>https://hal.inria.fr/inria-00326730 +<br/>Submitted on 5 Oct 2008 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('1689681', 'Rodrigo Verschae', 'rodrigo verschae')<br/>('1737300', 'Javier Ruiz-Del-Solar', 'javier ruiz-del-solar')<br/>('34047285', 'Mauricio Correa', 'mauricio correa')<br/>('1689681', 'Rodrigo Verschae', 'rodrigo verschae')<br/>('1737300', 'Javier Ruiz-Del-Solar', 'javier ruiz-del-solar')<br/>('34047285', 'Mauricio Correa', 'mauricio correa')</td><td></td></tr><tr><td>27d709f7b67204e1e5e05fe2cfac629afa21699d</td><td></td><td></td><td></td></tr><tr><td>271df16f789bd2122f0268c3e2fa46bc0cb5f195</td><td>Mining Discriminative Co-occurrence Patterns for Visual Recognition +<br/>School of EEE +<br/><b>Nanyang Technological University</b><br/>Singapore 639798 +<br/>Dept. of Media Analytics +<br/>NEC Laboratories America +<br/>Cupertino, CA, 95014 USA +<br/>EECS Dept. +<br/><b>Northwestern University</b><br/>Evanston, IL, 60208 USA +</td><td>('34316743', 'Junsong Yuan', 'junsong yuan')<br/>('40634508', 'Ming Yang', 'ming yang')<br/>('39955137', 'Ying Wu', 'ying wu')</td><td>jsyuan@ntu.edu.sg +<br/>myang@sv.nec-labs.com +<br/>yingwu@eecs.northwestern.edu +</td></tr><tr><td>275b5091c50509cc8861e792e084ce07aa906549</td><td>Institut für Informatik +<br/>der Technischen +<br/>Universität München +<br/>Dissertation +<br/>Leveraging the User’s Face as a Known Object +<br/>in Handheld Augmented Reality +<br/>Sebastian Bernhard Knorr +</td><td></td><td></td></tr><tr><td>27218ff58c3f0e7d7779fba3bb465d746749ed7c</td><td>Active Learning for Image Ranking +<br/>Over Relative Visual Attributes +<br/>by +<br/>Department of Computer Science +<br/><b>University of Texas at Austin</b></td><td>('2548555', 'Lucy Liang', 'lucy liang')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>276dbb667a66c23545534caa80be483222db7769</td><td>3D Res. 2, 03(2011)4 +<br/>10.1007/3DRes.03(2011)4 +<br/>3DR REVIEW w +<br/>An Introduction to Image-based 3D Surface Reconstruction and a +<br/>Survey of Photometric Stereo Methods +<br/>for +<br/>introduction +<br/>image-based 3D +<br/>techniques. Then we describe +<br/>Received: 21Feburary 2011 / Revised: 20 March 2011 / Accepted: 11 May 2011 +<br/><b>D Research Center, Kwangwoon University and Springer</b></td><td>('1908324', 'Steffen Herbort', 'steffen herbort')</td><td></td></tr><tr><td>270733d986a1eb72efda847b4b55bc6ba9686df4</td><td>We are IntechOpen, +<br/>the first native scientific +<br/>publisher of Open Access books +<br/>3,350 +<br/>108,000 +<br/>1.7 M +<br/>Open access books available +<br/>International authors and editors +<br/>Downloads +<br/>Our authors are among the +<br/>151 +<br/>Countries delivered to +<br/>TOP 1% +<br/>12.2% +<br/>most cited scientists +<br/>Contributors from top 500 universities +<br/>Selection of our books indexed in the Book Citation Index +<br/>in Web of Science™ Core Collection (BKCI) +<br/>Interested in publishing with us? +<br/>Numbers displayed above are based on latest data collected. +<br/>For more information visit www.intechopen.com +</td><td></td><td>Contact book.department@intechopen.com +</td></tr><tr><td>27c6cd568d0623d549439edc98f6b92528d39bfe</td><td>Regressive Tree Structured Model for Facial Landmark Localization +<br/>Artificial Vision Lab., Dept Mechanical Engineering +<br/><b>National Taiwan University of Science and Technology</b></td><td>('2329565', 'Kai-Hsiang Chang', 'kai-hsiang chang')<br/>('2421405', 'Shih-Chieh Huang', 'shih-chieh huang')</td><td>jison@mail.ntust.edu.tw +</td></tr><tr><td>273b0511588ab0a81809a9e75ab3bd93d6a0f1e3</td><td>The final publication is available at Springer via http://dx.doi.org/10.1007/s11042-016-3428-9 +<br/>Recognition of Facial Expressions Based on Salient +<br/>Geometric Features and Support Vector Machines +<br/><b>Korea Electronics Technology Institute, Jeonju-si, Jeollabuk-do 561-844, Rep</b><br/><b>Division of Computer Engineering, Chonbuk National University, Jeonju-si, Jeollabuk-do</b><br/><b>School of Computing Science, Simon Fraser University, Burnaby, B.C., Canada; E-Mail</b><br/>Tel.: +82-63-270-2406; Fax: +82-63-270-2394. +</td><td>('32322842', 'Deepak Ghimire', 'deepak ghimire')<br/>('2034182', 'Joonwhoan Lee', 'joonwhoan lee')<br/>('1689656', 'Ze-Nian Li', 'ze-nian li')<br/>('31984909', 'SungHwan Jeong', 'sunghwan jeong')</td><td>of Korea; E-Mails: (deepak, shjeong)@keti.re.kr +<br/>Rep. of Korea; E-Mail: chlee@jbnu.ac.kr +<br/>li@cs.sfu.ca +<br/>* Author to whom correspondence should be addressed; E-Mail: chlee@jbnu.ac.kr; +</td></tr><tr><td>27169761aeab311a428a9dd964c7e34950a62a6b</td><td>International Journal of the Physical Sciences Vol. 5(13), pp. 2020 -2029, 18 October, 2010 +<br/>Available online at http://www.academicjournals.org/IJPS +<br/>ISSN 1992 - 1950 ©2010 Academic Journals +<br/>Full Length Research Paper +<br/>Face recognition using 3D head scan data based on +<br/>Procrustes distance +<br/><b>Kongju National University, South Korea</b><br/><b>Korean Research Institute of Standards and Science (KRISS), Korea</b><br/>Accepted 6 July, 2010 +<br/>Recently, face recognition has attracted significant attention from the researchers and scientists in +<br/>various fields of research, such as biomedical informatics, pattern recognition, vision, etc due its +<br/>applications in commercially available systems, defense and security purpose. In this paper a practical +<br/>method for face reorganization utilizing head cross section data based on Procrustes analysis is +<br/>proposed. This proposed method relies on shape signatures of the contours extracted from face data. +<br/>The shape signatures are created by calculating the centroid distance of the boundary points, which is +<br/>a translation and rotation invariant signature. The shape signatures for a selected region of interest +<br/>(ROI) are used as feature vectors and authentication is done using them. After extracting feature +<br/>vectors a comparison analysis is performed utilizing Procrustes distance to differentiate their face +<br/>pattern from each other. The proposed scheme attains an equal error rate (EER) of 4.563% for the 400 +<br/>head data for 100 subjects. The performance analysis of face recognition was analyzed based on K +<br/>nearest neighbour classifier. The experimental results presented here verify that the proposed method +<br/>is considerable effective. +<br/>Key words: Face, biometrics, Procrustes distance, equal error rate, k nearest classifier. +<br/>INTRODUCTION +<br/>Perhaps face is the easiest means of identifying a person +<br/>by another person. In general humans can identify +<br/>themselves and others by faces in a scene without hard +<br/>effort, but face recognition systems that implement these +<br/>tasks are very challenging to design. The challenges are +<br/>even extensive when there is a wide range of variation +<br/>due to imaging situations. Both inter- and intra-subject +<br/>variations are related with face images. Physical similarity +<br/>among +<br/>inter-subject +<br/>variation whereas intra-subject variation is dependent on +<br/>the following aspects such as age, head pose facial app- +<br/>roach, presence of light and presence of other obje- +<br/>cts/people etc. However, in face recognition, it has been +<br/>observed that inter-person variations are available due to +<br/>variations in local geometric features. Automatic face +<br/>recognition has been widely studied during the last few +<br/>decades. It is an active research area spanning many di- +<br/>sciplines such as image processing, pattern recognition, +<br/>responsible +<br/>individuals +<br/>for +<br/>is +<br/>computer vision, neural networks, artificial intelligence, +<br/>and biometrics. +<br/>Many researchers from these different disciplines work +<br/>toward the goal of endowing machines or computers with +<br/>the ability to recognize human faces as we human beings +<br/>do, effortlessly, in our everyday life (Brunelli and Poggio, +<br/>1993; Samaria, 1994; Wiskott et al., 1997; Turk and +<br/>Pentland, 1991; Belhumeur et al., 1997; He et al., 2005; +<br/>Wiskott et al., 1997; Lanitis et al., 1995; Cootes et al., +<br/>2001; Brunelli and Poggio, 1993; Turk, 1991; Bellhumer +<br/>et al., 1997). Face recognition has a wide range of +<br/>potential applications +<br/>for commercial, security, and +<br/>forensic purposes. These applications include automated +<br/>crowd +<br/>shot +<br/>identification (e.g., for issuing driver licenses), credit card +<br/>authorization, ATM machine access control, design of +<br/>human computer interfaces, etc. The rapid evaluation in +<br/>face recognition research can be found by the progress +<br/>of systematic evaluation standards that includes the +<br/>FERET, FRVT 2000, FRVT 2002, and XM2VTS +<br/>protocols, and many existing software packages for +<br/>example FaceIt, FaceVACS, FaceSnap Recorder, +<br/>control, mug +<br/>surveillance, +<br/>access +</td><td>('3222448', 'Sikyung Kim', 'sikyung kim')<br/>('2387342', 'Se Jin Park', 'se jin park')</td><td>*Corresponding author. E-mail: mynudding@yahoo.com. +</td></tr><tr><td>27da432cf2b9129dce256e5bf7f2f18953eef5a5</td><td></td><td></td><td></td></tr><tr><td>27961bc8173ac84fdbecacd01e5ed6f7ed92d4bd</td><td>To Appear in The IEEE 6th International Conference on Biometrics: Theory, Applications and +<br/>Systems (BTAS), Sept. 29-Oct. 2, 2013, Washington DC, USA +<br/>Automatic Multi-view Face Recognition via 3D Model Based Pose Regularization +<br/>Department of Computer Science and Engineering +<br/><b>Michigan State University, East Lansing, MI, U.S.A</b></td><td>('1883998', 'Koichiro Niinuma', 'koichiro niinuma')<br/>('34393045', 'Hu Han', 'hu han')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>{niinumak, hhan, jain}@msu.edu +</td></tr><tr><td>27173d0b9bb5ce3a75d05e4dbd8f063375f24bb5</td><td>ISSN : 2248-9622, Vol. 4, Issue 10( Part - 3), October 2014, pp.40-44 +<br/>RESEARCH ARTICLE +<br/> OPEN ACCESS +<br/>Effect of Different Occlusion on Facial Expressions Recognition +<br/><b>RGPV University, Indore</b><br/><b>RGPV University, Indore</b><br/> +</td><td>('2890210', 'Ramchand Hablani', 'ramchand hablani')</td><td></td></tr><tr><td>2784d9212dee2f8a660814f4b85ba564ec333720</td><td>Learning Class-Specific Image Transformations with Higher-Order Boltzmann +<br/>Machines +<br/>Erik Learned-Miller +<br/><b>University of Massachusetts Amherst</b><br/>Amherst, MA +</td><td>('3219900', 'Gary B. Huang', 'gary b. huang')</td><td>{gbhuang,elm}@cs.umass.edu +</td></tr><tr><td>2717b044ae9933f9ab87f16d6c611352f66b2033</td><td>GNAS: A Greedy Neural Architecture Search Method for +<br/>Multi-Attribute Learning +<br/><b>Zhejiang University, 2Southwest Jiaotong University, 3Carnegie Mellon University</b></td><td>('2986516', 'Siyu Huang', 'siyu huang')<br/>('50079147', 'Xi Li', 'xi li')<br/>('1720488', 'Zhongfei Zhang', 'zhongfei zhang')</td><td>{siyuhuang,xilizju,zhongfei}@zju.edu.cn,zhiqicheng@gmail.com,alex@cs.cmu.edu +</td></tr><tr><td>2770b095613d4395045942dc60e6c560e882f887</td><td>GridFace: Face Rectification via Learning Local +<br/>Homography Transformations +<br/>Face++, Megvii Inc. +</td><td>('1848243', 'Erjin Zhou', 'erjin zhou')<br/>('2695115', 'Zhimin Cao', 'zhimin cao')<br/>('40055995', 'Jian Sun', 'jian sun')</td><td>{zej,czm,sunjian}@megvii.com +</td></tr><tr><td>27cccf992f54966feb2ab4831fab628334c742d8</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 64– No.18, February 2013 +<br/>Facial Expression Recognition by Statistical, Spatial +<br/>Features and using Decision Tree +<br/>Assistant Professor +<br/>CSIT Department +<br/>GGV BIlaspur, Chhattisgarh +<br/>India +<br/>Assistant Professor +<br/>Electronics (ECE) Department +<br/>JECRC Jaipur, Rajasthan India +<br/>IshanBhardwaj +<br/>Student of Ph.D. +<br/>Electrical Department +<br/>NIT Raipur, Chhattisgarh India +</td><td>('8836626', 'Nazil Perveen', 'nazil perveen')<br/>('2092589', 'Darshan Kumar', 'darshan kumar')</td><td></td></tr><tr><td>27883967d3dac734c207074eed966e83afccb8c3</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. +<br/>Two-dimensional Maximum Local Variation based on Image Euclidean Distance for Face +<br/>Recognition +<br/><b>State Key Laboratory of Integrated Services Networks, Xidian University, Xi an 710071 China</b><br/><b>State Key Laboratory of CAD and CG, ZHE JIANG University, HangZhou, 310058 China</b><br/><b>The Chinese University of Hong Kong, Hong Kong</b><br/>to +<br/>improve +<br/>in +<br/>images and +<br/>in estimating +</td><td>('38469552', 'Quanxue Gao', 'quanxue gao')</td><td></td></tr><tr><td>270e5266a1f6e76954dedbc2caf6ff61a5fbf8d0</td><td>EmotioNet Challenge: Recognition of facial expressions of emotion in the wild +<br/>Dept. Electrical and Computer Engineering +<br/><b>The Ohio State University</b></td><td>('8038057', 'Ramprakash Srinivasan', 'ramprakash srinivasan')<br/>('9947018', 'Qianli Feng', 'qianli feng')<br/>('1678691', 'Yan Wang', 'yan wang')</td><td></td></tr><tr><td>27f8b01e628f20ebfcb58d14ea40573d351bbaad</td><td>DEPARTMENT OF INFORMATION ENGINEERING AND COMPUTER SCIENCE +<br/>ICT International Doctoral School +<br/>Events based Multimedia Indexing +<br/>and Retrieval +<br/>SUBMITTED TO THE DEPARTMENT OF +<br/>INFORMATION ENGINEERING AND COMPUTER SCIENCE (DISI) +<br/>IN THE PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE +<br/>OF +<br/>DOCTOR OF PHILOSOPHY +<br/>Advisor: +<br/>Examiners: Prof. Marco Carli, Universit`a degli Studi di Roma Tre, Italy +<br/>Prof. Nicola Conci, Universit`a degli Studi di Trento, Italy +<br/>Prof. Pietro Zanuttigh, Universit`a degli Studi di Padova, Italy +<br/>Prof. Giulia Boato, Universit`a degli Studi di Trento, Italy +<br/>December 2017 +</td><td>('36296712', 'Kashif Ahmad', 'kashif ahmad')</td><td></td></tr><tr><td>2742a61d32053761bcc14bd6c32365bfcdbefe35</td><td>Submitted 9/13; Revised 6/14; Published 2/15 +<br/>Learning Transformations for Clustering and Classification +<br/>Department of Electrical and Computer Engineering +<br/><b>Duke University</b><br/>Durham, NC 27708, USA +<br/>Department of Electrical and Computer Engineering +<br/>Department of Computer Science +<br/>Department of Biomedical Engineering +<br/><b>Duke University</b><br/>Durham, NC 27708, USA +<br/>Editor: Ben Recht +</td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('1699339', 'Guillermo Sapiro', 'guillermo sapiro')</td><td>qiang.qiu@duke.edu +<br/>guillermo.sapiro@duke.edu +</td></tr><tr><td>27dafedccd7b049e87efed72cabaa32ec00fdd45</td><td>Unsupervised Visual Alignment with Similarity Graphs +<br/><b>Tampere University of Technology, Finland</b></td><td>('2416841', 'Fatemeh Shokrollahi Yancheshmeh', 'fatemeh shokrollahi yancheshmeh')<br/>('40394658', 'Ke Chen', 'ke chen')</td><td>{fatemeh.shokrollahiyancheshmeh, ke.chen, joni.kamarainen}@tut.fi +</td></tr><tr><td>27a299b834a18e45d73e0bf784bbb5b304c197b3</td><td>Social Role Discovery in Human Events +<br/><b>Stanford University</b><br/>br. maids +<br/>bride +<br/>groom +<br/>gr. man +<br/>Pairwise interaction features +<br/>Social Role Model +<br/>Σ𝛼 +<br/>Σ𝛽 +<br/>Introduction +<br/>• Social Roles describe humans in an event +<br/>•Social roles of humans are dependent on +<br/>- their actions in a social setting +<br/>- their interactions with other roles +<br/>• Obtaining role annotations for training is expensive +<br/>•Goal: Discover role clusters in a social event based on +<br/>role-specific interactions +<br/>1. Input: videos +<br/>with human tracks +<br/>2. Extract unary and +<br/>interaction features +<br/>3. Output: Cluster +<br/>people into social roles +<br/>Our Approach +<br/>- Does not require +<br/>role annotations +<br/>- Clusters people +<br/>into roles based +<br/>on interactions as +<br/>well as person- +<br/>specific features +<br/>Results: Clustering Accuracy +<br/>• New YouTube dataset: ~40 videos with 160-240 people per event +<br/>• Human tracks and ground-truth roles annotated +<br/>Method +<br/>prior +<br/>K-means +<br/>Only unary +<br/>Interaction +<br/>as context +<br/>Birthday Wedding Award +<br/>Function +<br/>62.97% +<br/>31.97% +<br/>69.31% +<br/>77.75% +<br/>20.17% +<br/>29.43% +<br/>39.22% +<br/>38.83% +<br/>29.32% +<br/>33.88% +<br/>38.25% +<br/>41.53% +<br/>Physical +<br/>Training +<br/>65.93% +<br/>57.67% +<br/>76.69% +<br/>77.91% +<br/>No spatial +<br/>43.72% +<br/>No proxemic 43.72% +<br/>44.81% +<br/>Full Model +<br/>36.41% +<br/>39.32% +<br/>42.72% +<br/>79.54% +<br/>79.80% +<br/>83.12% +<br/>82.82% +<br/>77.91% +<br/>82.82% +<br/>• Only unary – No +<br/>interaction feature +<br/>Interaction as +<br/>context – Average +<br/>interaction as unary +<br/>• No spatial – Only +<br/>proxemic interaction +<br/>• No proxemic – Only +<br/>spatial interaction +<br/>Ψ𝑃 +<br/>- Spatio-temporal trajectory features +<br/>- Proxemic[2] interaction features +<br/>Unary features +<br/>- HOG3D and Trajectory to capture action +<br/>- Gender and Color Histogram features +<br/>- Object interaction features +<br/>Ψ𝑢 +<br/>𝒔𝑖 +<br/>𝛼 - Unary feature weight +<br/>𝒔𝑖 +<br/>- Social role assignment +<br/>- Reference role assignment +<br/>Interaction feature weight +<br/>Jointly infer +<br/>by variational +<br/>inference +<br/>Ψ𝑢 +<br/>Ψ𝑝 +<br/>Interaction restricted +<br/>to reference role for +<br/>tractable inference +<br/>• Spatial relations in wedding. Cross-arrow is the position of the reference +<br/>Results: Role Clusters +<br/>role (groom) +<br/>Bride +<br/>Priest +<br/>Brides maid +<br/>Grooms man +<br/>• Color of cross represents ground-truth role for wrong assignments +<br/>bride +<br/>groom +<br/>priest +<br/>grooms men +<br/>brides maid +<br/>b’day person +<br/>parent +<br/>friends +<br/>guest +<br/>presenter +<br/>recipient +<br/>host +<br/>distributor +<br/>[1] V. Ramanathan, B. Yao, L. Fei-Fei. Social Role Discovery in Human Events. In CVPR, 2013. +<br/>[2] Y. Yang, S. Baker, A. Kannan, and D. Ramanan. Recognizing proxemics in personal photos. In CVPR, 2012. +<br/>This work was supported in part by DARPA Minds Eye, NSF, Intel, Microsoft Research, Google Research and the Intelligence Advanced +<br/>Research Projects Activity* (IARPA) via Department of Interior National Business Center contract number D11PC20069. +<br/>* The U.S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright thereon. Disclaimer: The views and conclusions contained herein are +<br/>those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of IARPA, DOI/NBC, or the U.S. Government. +<br/>instructor +<br/>presenter +</td><td>('34066479', 'Vignesh Ramanathan', 'vignesh ramanathan')<br/>('38916673', 'Bangpeng Yao', 'bangpeng yao')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td>{vigneshr, bangpeng, feifeili}@cs.stanford.edu +</td></tr><tr><td>27b1670e1b91ab983b7b1ecfe9eb5e6ba951e0ba</td><td>Comparison between k-nn and svm method +<br/>for speech emotion recognition +<br/><b>Anjuman College of Engineering and Technology, Sadar, Nagpur, India</b></td><td>('27879696', 'Muzaffar Khan', 'muzaffar khan')</td><td></td></tr><tr><td>274f87ad659cd90382ef38f7c6fafc4fc7f0d74d</td><td></td><td></td><td></td></tr><tr><td>27ee8482c376ef282d5eb2e673ab042f5ded99d7</td><td>Scale Normalization for the Distance Maps AAM. +<br/>Avenue de la boulaie, BP 81127, +<br/>35 511 Cesson-S´evign´e, France +<br/>Sup´elec, IETR-SCEE Team +</td><td>('31491147', 'Denis Giri', 'denis giri')<br/>('2861129', 'Maxime Rosenwald', 'maxime rosenwald')<br/>('32420329', 'Benjamin Villeneuve', 'benjamin villeneuve')<br/>('3353560', 'Sylvain Le Gallou', 'sylvain le gallou')</td><td>Email: {denis.giri, maxime.rosenwald, benjamin.villeneuve, sylvain.legallou, renaud.seguier}@supelec.fr +</td></tr><tr><td>4b4106614c1d553365bad75d7866bff0de6056ed</td><td>Unconstrained Facial Images: Database for Face +<br/>Recognition under Real-world Conditions⋆ +<br/>1 Dept. of Computer Science & Engineering +<br/><b>University of West Bohemia</b><br/>Plzeˇn, Czech Republic +<br/>2 NTIS - New Technologies for the Information Society +<br/><b>University of West Bohemia</b><br/>Plzeˇn, Czech Republic +</td><td>('2628715', 'Ladislav Lenc', 'ladislav lenc')</td><td>{llenc,pkral}@kiv.zcu.cz +</td></tr><tr><td>4bb03b27bc625e53d8d444c0ba3ee235d2f17e86</td><td>Reading Between The Lines: Object Localization +<br/>Using Implicit Cues from Image Tags +<br/>Department of Computer Science +<br/><b>University of Texas at Austin</b></td><td>('35788904', 'Sung Ju Hwang', 'sung ju hwang')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>{sjhwang,grauman}@cs.utexas.edu +</td></tr><tr><td>4b89cf7197922ee9418ae93896586c990e0d2867</td><td>LATEX Author Guidelines for CVPR Proceedings +<br/>First Author +<br/>Institution1 +<br/>Institution1 address +</td><td></td><td>firstauthor@i1.org +</td></tr><tr><td>4bc9a767d7e63c5b94614ebdc24a8775603b15c9</td><td><b>University of Trento</b><br/>Doctoral Thesis +<br/>Understanding Visual Information: +<br/>from Unsupervised Discovery to +<br/>Minimal Effort Domain Adaptation +<br/>Author: +<br/>Supervisor: +<br/>Dr. Nicu Sebe +<br/>A thesis submitted in fulfilment of the requirements +<br/>for the degree of Doctor of Philosophy +<br/>in the +<br/>International Doctorate School in Information and Communication Technologies +<br/>Department of Information Engineering and Computer Science +<br/>Multimedia and Human Understanding Group (MHUG) +<br/>April 2015 +</td><td>('2933565', 'Gloria Zen', 'gloria zen')</td><td></td></tr><tr><td>4b519e2e88ccd45718b0fc65bfd82ebe103902f7</td><td>A Discriminative Model for Age Invariant Face +<br/>Recognition +<br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Science, China</b><br/><b>Michigan State University, E. Lansing, MI 48823, USA</b><br/><b>Korea University, Seoul 136-713, Korea</b></td><td>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('2222919', 'Unsang Park', 'unsang park')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>4b3f425274b0c2297d136f8833a31866db2f2aec</td><td>This is a pre-print of the original paper accepted for publication in the CVPR 2017 Biometrics Workshop. +<br/>Toward Open-Set Face Recognition +<br/>Manuel G¨unther +<br/><b>Vision and Security Technology Lab, University of Colorado Colorado Springs</b></td><td>('39616991', 'Steve Cruz', 'steve cruz')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')<br/>('39886114', 'Ethan M. Rudd', 'ethan m. rudd')</td><td>{mgunther,scruz,erudd,tboult}@vast.uccs.edu +</td></tr><tr><td>4b7c110987c1d89109355b04f8597ce427a7cd72</td><td>ORIGINAL RESEARCH ARTICLE +<br/>published: 16 October 2014 +<br/>doi: 10.3389/fnhum.2014.00804 +<br/>Feature- and Face-Exchange illusions: new insights and +<br/>applications for the study of the binding problem +<br/><b>American University, Washington, DC, USA</b><br/><b>University of Nevada, Reno, Reno, NV, USA</b><br/>Edited by: +<br/><b>Baingio Pinna, University of</b><br/>Sassari, Italy +<br/>Reviewed by: +<br/>Stephen Louis Macknik, Barrow +<br/><b>Neurological Institute, USA</b><br/>Susana Martinez-Conde, Barrow +<br/><b>Neurological Institute, USA</b><br/>*Correspondence: +<br/><b>Psychology, American University</b><br/>4400 Massachusetts Avenue NW, +<br/>Washington, DC 20016, USA +<br/>The binding problem is a longstanding issue in vision science: i.e., how are humans able to +<br/>maintain a relatively stable representation of objects and features even though the visual +<br/>system processes many aspects of the world separately and in parallel? We previously +<br/>investigated this issue with a variant of the bounce-pass paradigm, which consists of two +<br/>rectangular bars moving in opposite directions; if the bars are identical and never overlap, +<br/>the motion could equally be interpreted as bouncing or passing. Although bars of different +<br/>colors should be seen as passing each other (since the colors provide more information +<br/>about the bars’ paths), we found “Feature Exchange”: observers reported the paradoxical +<br/>perception that the bars appear to bounce off of each other and exchange colors. Here we +<br/>extend our previous findings with three demonstrations. “Peripheral Feature-Exchange” +<br/>consists of two colored bars that physically bounce (they continually meet in the middle +<br/>of the monitor and return to the sides). When viewed in the periphery, the bars appear +<br/>to stream past each other even though this percept relies on the exchange of features +<br/>and contradicts the information provided by the color of the bars. In “Face-Exchange” +<br/>two different faces physically pass each other. When fixating centrally, observers typically +<br/>report the perception of bouncing faces that swap features, indicating that the Feature +<br/>Exchange effect can occur even with complex objects. In “Face-Go-Round,” one face +<br/>repeatedly moves from left to right on the top of the monitor, and the other from right +<br/>to left at the bottom of the monitor. Observers typically perceive the faces moving in a +<br/>circle—a percept that contradicts information provided by the identity of the faces. We +<br/>suggest that Feature Exchange and the paradigms used to elicit it can be useful for the +<br/>investigation of the binding problem as well as other contemporary issues of interest to +<br/>vision science. +<br/>Keywords: motion perception, object perception, binding problem, visual periphery, animation, bouncing +<br/>streaming illusions, illusion of causality +<br/>INTRODUCTION +<br/>The “binding problem” refers to the observation that the brain +<br/>processes many aspects of the visual world separately and in +<br/>parallel, yet we perceive a unified world, populated by coherent +<br/>objects (James, 1890; Treisman, 1996; Holcombe et al., 2009). The +<br/>implication is that the visual system binds together the output of +<br/>separate processes (which presumably compute features, textures, +<br/>colors, motion gradients, etc.) prior to creating our object-centric +<br/>perceptual world. Two fundamental questions of the binding +<br/>problem can be summarized as follows: (1) How, and under +<br/>what conditions, does the brain combine (or fail to combine) the +<br/>outputs of these separate processes to construct an object rep- +<br/>resentation? (2) How are object representations maintained over +<br/>time and space? +<br/>We recently examined the spatiotemporal conditions and the +<br/>role feature-level processes play in representing and maintaining +<br/>objects (Caplovitz et al., 2011) using a variant of the “bounce- +<br/>pass paradigm” (Metzger, 1934; Michotte, 1946/1963; Kanizsa, +<br/>1969). In a typical version of the bounce pass paradigm, the +<br/>interpretation of motion direction and object correspondence +<br/>direction is intrinsically ambiguous, and the degree to which +<br/>observers report one or the other of the potential percepts has +<br/>been used to study a range of perceptual and cognitive processes. +<br/>For example, versions of this basic paradigm have been used to +<br/>study properties of cross-modal interactions and motion per- +<br/>ception as well as object representations (Bertenthal et al., 1993; +<br/>Watanabe and Shimojo, 1998; Sekuler and Sekuler, 1999; Mitroff +<br/>et al., 2005; Feldman and Tremoulet, 2006). +<br/>The basic paradigm (illustrated in Figure 1A) consists of two +<br/>rectangles; one that moves from right to left while the other moves +<br/>from left to right. The display is ambiguous because the stimulus +<br/>is wholly consistent with each rectangle passing from one side of +<br/>the screen to the other (i.e., the perception of streaming) or as +<br/>bouncing off of the other rectangle and returning to its point of +<br/>origin (i.e., the perception of bouncing). If, at the point of inter- +<br/>section, one rectangle overlaps with the other rectangle observers +<br/>will commonly perceive streaming (Sekuler and Sekuler, 1999). +<br/>In our experiments, this potential cue is removed: at the critical +<br/>point of intersection, the rectangles exactly exchange places and +<br/>thus never have an overlapping edge. When the two rectangles are +<br/>Frontiers in Human Neuroscience +<br/>www.frontiersin.org +<br/>October 2014 | Volume 8 | Article 804 | 1 +<br/>HUMAN NEUROSCIENCE </td><td>('31981243', 'Arthur G. Shapiro', 'arthur g. shapiro')<br/>('8369036', 'Gideon P. Caplovitz', 'gideon p. caplovitz')<br/>('23863232', 'Erica L. Dixon', 'erica l. dixon')<br/>('31981243', 'Arthur G. Shapiro', 'arthur g. shapiro')</td><td>e-mail: arthur.shapiro@american.edu +</td></tr><tr><td>4bd088ba3f42aa1e43ae33b1988264465a643a1f</td><td>Technical Report, IDE0852, May 2008 +<br/>Multiview Face Detection Using +<br/>Gabor Filters and +<br/>Support Vector Machine +<br/>Bachelor’s Thesis in Computer Systems Engineering +<br/>School of Information Science, Computer and Electrical Engineering +<br/> +<br/><b>Halmstad University</b></td><td></td><td></td></tr><tr><td>4bfce41cc72be315770861a15e467aa027d91641</td><td>Active Annotation Translation +<br/>Caltech +<br/>Kristj´an Eldj´arn Hj¨orleifsson +<br/><b>University of Iceland</b><br/>Caltech +</td><td>('3251767', 'Steve Branson', 'steve branson')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>sbranson@caltech.edu +<br/>keh4@hi.is +<br/>perona@caltech.edu +</td></tr><tr><td>4b61d8490bf034a2ee8aa26601d13c83ad7f843a</td><td>A Modulation Module for Multi-task Learning with +<br/>Applications in Image Retrieval +<br/><b>Northwestern University</b><br/>2 AIBee +<br/>3 Bytedance AI Lab +<br/><b>Carnegie Mellon University</b></td><td>('8343585', 'Xiangyun Zhao', 'xiangyun zhao')</td><td></td></tr><tr><td>4bd3de97b256b96556d19a5db71dda519934fd53</td><td>Latent Factor Guided Convolutional Neural Networks for Age-Invariant Face +<br/>Recognition +<br/><b>School of Electronic and Information Engineering, South China University of Technology</b><br/><b>Shenzhen Key Lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('2512949', 'Yandong Wen', 'yandong wen')<br/>('32787758', 'Zhifeng Li', 'zhifeng li')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td>yd.wen@siat.ac.cn, zhifeng.li@siat.ac.cn, yu.qiao@siat.ac.cn +</td></tr><tr><td>4b04247c7f22410681b6aab053d9655cf7f3f888</td><td>Robust Face Recognition by Constrained Part-based +<br/>Alignment +</td><td>('1692992', 'Yuting Zhang', 'yuting zhang')<br/>('2370507', 'Kui Jia', 'kui jia')<br/>('7135663', 'Yueming Wang', 'yueming wang')<br/>('1734380', 'Gang Pan', 'gang pan')<br/>('1926757', 'Tsung-Han Chan', 'tsung-han chan')<br/>('1700297', 'Yi Ma', 'yi ma')</td><td></td></tr><tr><td>4b60e45b6803e2e155f25a2270a28be9f8bec130</td><td>Attribute Based Object Identification +</td><td>('1686318', 'Yuyin Sun', 'yuyin sun')<br/>('1766509', 'Liefeng Bo', 'liefeng bo')<br/>('1731079', 'Dieter Fox', 'dieter fox')</td><td></td></tr><tr><td>4b48e912a17c79ac95d6a60afed8238c9ab9e553</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>Minimum Margin Loss for Deep Face Recognition +</td><td>('49141822', 'Xin Wei', 'xin wei')<br/>('3552546', 'Hui Wang', 'hui wang')<br/>('2986129', 'Huan Wan', 'huan wan')</td><td></td></tr><tr><td>4b5eeea5dd8bd69331bd4bd4c66098b125888dea</td><td>Human Activity Recognition Using Conditional +<br/>Random Fields and Privileged Information +<br/>submitted to +<br/>the designated by the General Assembly Composition of the +<br/>Department of Computer Science & Engineering Inquiry +<br/>Committee +<br/>by +<br/>in partial fulfillment of the Requirements for the Degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>February 2016 +</td><td>('2045915', 'Michalis Vrigkas', 'michalis vrigkas')</td><td></td></tr><tr><td>4bbbee93519a4254736167b31be69ee1e537f942</td><td></td><td></td><td></td></tr><tr><td>4b74f2d56cd0dda6f459319fec29559291c61bff</td><td>CHIACHIA ET AL.: PERSON-SPECIFIC SUBSPACES FOR FAMILIAR FACES +<br/>Person-Specific Subspace Analysis for +<br/>Unconstrained Familiar Face Identification +<br/>David Cox2 +<br/><b>Institute of Computing</b><br/><b>University of Campinas</b><br/>Campinas, Brazil +<br/><b>Rowland Institute</b><br/><b>Harvard University</b><br/>Cambridge, USA +<br/><b>McGovern Institute</b><br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, USA +<br/>4 Department of Computer Science +<br/>Universidade Federal de Minas Gerais +<br/>Belo Horizonte, Brazil +</td><td>('1761151', 'Giovani Chiachia', 'giovani chiachia')<br/>('30017846', 'Nicolas Pinto', 'nicolas pinto')<br/>('1679142', 'William Robson Schwartz', 'william robson schwartz')<br/>('2145405', 'Anderson Rocha', 'anderson rocha')<br/>('1716806', 'Alexandre X. Falcão', 'alexandre x. falcão')</td><td>giovanichiachia@gmail.com +<br/>pinto@mit.edu +<br/>william@dcc.ufmg.br +<br/>anderson.rocha@ic.unicamp.br +<br/>afalcao@ic.unicamp.br +<br/>davidcox@fas.harvard.edu +</td></tr><tr><td>4ba38262fe20fab3e4c80215147b498f83843b93</td><td>MAKIANDCIPOLLA:OBTAININGTHESHAPEOFAMOVINGOBJECT +<br/>Obtaining the Shape of a Moving Object +<br/>with a Specular Surface +<br/>Toshiba Research Europe +<br/><b>Cambridge Research Laboratory</b><br/>Department of Engineering +<br/><b>University of Cambridge</b></td><td>('1801052', 'Atsuto Maki', 'atsuto maki')<br/>('1745672', 'Roberto Cipolla', 'roberto cipolla')</td><td>atsuto.maki@crl.toshiba.co.uk +<br/>cipolla@cam.ac.uk +</td></tr><tr><td>4bbe460ab1b279a55e3c9d9f488ff79884d01608</td><td>GAGAN: Geometry-Aware Generative Adversarial Networks +<br/>Jean Kossaifi∗ +<br/><b>Middlesex University London</b><br/><b>Imperial College London</b></td><td>('47801605', 'Linh Tran', 'linh tran')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{jean.kossaifi;linh.tran;i.panagakis;m.pantic}@imperial.ac.uk +</td></tr><tr><td>4b3eaedac75ac419c2609e131ea9377ba8c3d4b8</td><td>FAST NEWTON ACTIVE APPEARANCE MODELS +<br/>Jean Kossaifi(cid:63) +<br/><b>cid:63) Imperial College London, UK</b><br/><b>University of Lincoln, UK</b><br/><b>University of Twente, The Netherlands</b></td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>4b507a161af8a7dd41e909798b9230f4ac779315</td><td>A Theory of Multiplexed Illumination +<br/>Dept. Electrical Engineering +<br/>Technion - Israel Inst. Technology +<br/>Haifa 32000, ISRAEL +<br/>Dept. Computer Science +<br/><b>Columbia University</b><br/>New York, NY 10027 +</td><td>('2159538', 'Yoav Y. Schechner', 'yoav y. schechner')<br/>('1750470', 'Shree K. Nayar', 'shree k. nayar')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')</td><td>yoav@ee.technion.ac.il +<br/>{nayar,belhumeur}@cs.columbia.edu +</td></tr><tr><td>4b02387c2db968a70b69d98da3c443f139099e91</td><td>Detecting facial landmarks in the video based on a hybrid framework +<br/><b>School of Information Engineering, Guangdong University of Technology, 510006 Guangzhou, China</b><br/><b>School of Electromechanical Engineering, Guangdong University of Technology, 510006 Guangzhou, China</b></td><td>('1850205', 'Nian Cai', 'nian cai')<br/>('3468993', 'Zhineng Lin', 'zhineng lin')<br/>('2686365', 'Fu Zhang', 'fu zhang')<br/>('39038751', 'Guandong Cen', 'guandong cen')<br/>('40465036', 'Han Wang', 'han wang')</td><td></td></tr><tr><td>4b6be933057d939ddfa665501568ec4704fabb39</td><td></td><td></td><td></td></tr><tr><td>4b71d1ff7e589b94e0f97271c052699157e6dc4a</td><td>Hindawi Publishing Corporation +<br/>EURASIP Journal on Advances in Signal Processing +<br/>Volume 2008, Article ID 748483, 18 pages +<br/>doi:10.1155/2008/748483 +<br/>Research Article +<br/>Pose-Encoded Spherical Harmonics for Face Recognition and +<br/>Synthesis Using a Single Image +<br/><b>Center for Automation Research, University of Maryland, College Park, MD 20742, USA</b><br/>2 Vision Technologies Lab, Sarnoff Corporation, Princeton, NJ 08873, USA +<br/>Received 1 May 2007; Accepted 4 September 2007 +<br/>Recommended by Juwei Lu +<br/>Face recognition under varying pose is a challenging problem, especially when illumination variations are also present. In this +<br/>paper, we propose to address one of the most challenging scenarios in face recognition. That is, to identify a subject from a test +<br/>image that is acquired under different pose and illumination condition from only one training sample (also known as a gallery +<br/>image) of this subject in the database. For example, the test image could be semifrontal and illuminated by multiple lighting +<br/>sources while the corresponding training image is frontal under a single lighting source. Under the assumption of Lambertian +<br/>reflectance, the spherical harmonics representation has proved to be effective in modeling illumination variations for a fixed pose. +<br/>In this paper, we extend the spherical harmonics representation to encode pose information. More specifically, we utilize the fact +<br/>that 2D harmonic basis images at different poses are related by close-form linear transformations, and give a more convenient +<br/>transformation matrix to be directly used for basis images. An immediate application is that we can easily synthesize a different +<br/>view of a subject under arbitrary lighting conditions by changing the coefficients of the spherical harmonics representation. A +<br/>more important result is an efficient face recognition method, based on the orthonormality of the linear transformations, for +<br/>solving the above-mentioned challenging scenario. Thus, we directly project a nonfrontal view test image onto the space of frontal +<br/>view harmonic basis images. The impact of some empirical factors due to the projection is embedded in a sparse warping matrix; +<br/>for most cases, we show that the recognition performance does not deteriorate after warping the test image to the frontal view. +<br/>Very good recognition results are obtained using this method for both synthetic and challenging real images. +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>1. +<br/>INTRODUCTION +<br/>Face recognition is one of the most successful applications +<br/>of image analysis and understanding [1]. Given a database of +<br/>training images (sometimes called a gallery set, or gallery im- +<br/>ages), the task of face recognition is to determine the facial ID +<br/>of an incoming test image. Built upon the success of earlier +<br/>efforts, recent research has focused on robust face recogni- +<br/>tion to handle the issue of significant difference between a +<br/>test image and its corresponding training images (i.e., they +<br/>belong to the same subject). Despite significant progress, ro- +<br/>bust face recognition under varying lighting and different +<br/>pose conditions remains to be a challenging problem. The +<br/>problem becomes even more difficult when only one train- +<br/>ing image per subject is available. Recently, methods have +<br/>been proposed to handle the combined pose and illumina- +<br/>tion problem when only one training image is available, for +<br/>example, the method based on morphable models [2] and its +<br/>extension [3] that proposes to handle the complex illumina- +<br/>tion problem by integrating spherical harmonics representa- +<br/>tion [4, 5]. In these methods, either arbitrary illumination +<br/>conditions cannot be handled [2] or the expensive computa- +<br/>tion of harmonic basis images is required for each pose per +<br/>subject [3]. +<br/>Under the assumption of Lambertian reflectance, the +<br/>spherical harmonics representation has proved to be effec- +<br/>tive in modelling illumination variations for a fixed pose. In +<br/>this paper, we extend the harmonic representation to encode +<br/>pose information. We utilize the fact that all the harmonic +<br/>basis images of a subject at various poses are related to each +<br/>other via close-form linear transformations [6, 7], and de- +<br/>rive a more convenient transformation matrix to analytically +<br/>synthesize basis images of a subject at various poses from +<br/>just one set of basis images at a fixed pose, say, the frontal +</td><td>('39265975', 'Zhanfeng Yue', 'zhanfeng yue')<br/>('38480590', 'Wenyi Zhao', 'wenyi zhao')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('39265975', 'Zhanfeng Yue', 'zhanfeng yue')</td><td>Correspondence should be addressed to Zhanfeng Yue, zyue@cfar.umd.edu +</td></tr><tr><td>4b0a2937f64df66cadee459a32ad7ae6e9fd7ed2</td><td>Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset +<br/>Jo˜ao Carreira† +<br/>†DeepMind +<br/><b>University of Oxford</b></td><td>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>joaoluis@google.com +<br/>zisserman@google.com +</td></tr><tr><td>4b4ecc1cb7f048235605975ab37bb694d69f63e5</td><td>Nonlinear Embedding Transform for +<br/>Unsupervised Domain Adaptation +<br/>Center for Cognitive Ubiquitous Computing +<br/><b>Arizona State University, AZ, USA</b></td><td>('3151995', 'Hemanth Venkateswara', 'hemanth venkateswara')<br/>('2471253', 'Shayok Chakraborty', 'shayok chakraborty')<br/>('1743991', 'Sethuraman Panchanathan', 'sethuraman panchanathan')</td><td>{hemanthv,schakr10,panch}@asu.edu +</td></tr><tr><td>4be03fd3a76b07125cd39777a6875ee59d9889bd</td><td>CONTENT-BASED ANALYSIS FOR ACCESSING AUDIOVISUAL ARCHIVES: +<br/>ALTERNATIVES FOR CONCEPT-BASED INDEXING AND SEARCH +<br/>ESAT/PSI - IBBT +<br/>KU Leuven, Belgium +</td><td>('1704728', 'Tinne Tuytelaars', 'tinne tuytelaars')</td><td>Tinne.Tuytelaars@esat.kuleuven.be +</td></tr><tr><td>4be774af78f5bf55f7b7f654f9042b6e288b64bd</td><td>Variational methods for Conditional Multimodal Learning: +<br/>Generating Human Faces from Attributes +<br/><b>Indian Institute of Science</b><br/>Bangalore, India +</td><td>('2686270', 'Gaurav Pandey', 'gaurav pandey')<br/>('2440174', 'Ambedkar Dukkipati', 'ambedkar dukkipati')</td><td>{gp88, ad}@csa.iisc.ernet.in +</td></tr><tr><td>4b321065f6a45e55cb7f9d7b1055e8ac04713b41</td><td>Affective Computing Models for Character +<br/>Animation +<br/>School of Computing and Mathematical Sciences +<br/><b>Liverpool John Moores University</b><br/>Byrom Street, L3 3AF, Liverpool, UK +</td><td>('1794784', 'Abdennour El Rhalibi', 'abdennour el rhalibi')<br/>('36782007', 'Christopher Carter', 'christopher carter')<br/>('1768270', 'Madjid Merabti', 'madjid merabti')</td><td>R.L.Duarte@2010.ljmu.ac.uk;{A.Elrhalibi; C.J.Carter;M.Merabti}@ljmu.ac.uk +</td></tr><tr><td>4b605e6a9362485bfe69950432fa1f896e7d19bf</td><td>To appear in the CVPR Workshop on Biometrics, June 2016 +<br/>A Comparison of Human and Automated Face Verification Accuracy on +<br/>Unconstrained Image Sets∗ +<br/>Noblis +<br/>Noblis +<br/>Noblis +<br/>Noblis +<br/><b>Michigan State University</b></td><td>('1917247', 'Austin Blanton', 'austin blanton')<br/>('7996649', 'Kristen C. Allen', 'kristen c. allen')<br/>('15282121', 'Tim Miller', 'tim miller')<br/>('1718102', 'Nathan D. Kalka', 'nathan d. kalka')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>imaus10@gmail.com +<br/>kristen.allen@noblis.org +<br/>timothy.miller@noblis.org +<br/>nathan.kalka@noblis.org +<br/>jain@cse.msu.edu +</td></tr><tr><td>4b3dd18882ff2738aa867b60febd2b35ab34dffc</td><td>FACIAL FEATURE ANALYSIS OF +<br/>SPONTANEOUS FACIAL EXPRESSION +<br/>Computer Laboratory +<br/><b>University of Cambridge</b><br/>William Gates Building, +<br/>Cambridge CB3 0FD UK +<br/>Department of Computer Science +<br/><b>The American University in Cairo</b><br/>113 Kasr Al Aini Street, +<br/>P.O. Box 2511, Cairo, Egypt +</td><td>('1754451', 'Rana El Kaliouby', 'rana el kaliouby')<br/>('3337337', 'Amr Goneid', 'amr goneid')</td><td>rana.el-kaliouby@cl.cam.ac.uk +<br/>goneid@aucegypt.edu +</td></tr><tr><td>11a2ef92b6238055cf3f6dcac0ff49b7b803aee3</td><td>TOWARDS REDUCTION OF THE TRAINING AND SEARCH RUNNING TIME +<br/>COMPLEXITIES FOR NON-RIGID OBJECT SEGMENTATION +<br/>Instituto de Sistemas e Rob´otica, Instituto Superior T´ecnico, 1049-001 Lisboa, Portugal(a) +<br/><b>Australian Centre for Visual Technologies, The University of Adelaide, Australia (b</b></td><td>('3259175', 'Jacinto C. Nascimento', 'jacinto c. nascimento')<br/>('3265767', 'Gustavo Carneiro', 'gustavo carneiro')</td><td></td></tr><tr><td>11dc744736a30a189f88fa81be589be0b865c9fa</td><td>A Unified Multiplicative Framework for Attribute Learning +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b></td><td>('2582309', 'Kongming Liang', 'kongming liang')<br/>('1783542', 'Hong Chang', 'hong chang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{kongming.liang, hong.chang, shiguang.shan, xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>11a210835b87ccb4989e9ba31e7559bb7a9fd292</td><td>Hub +<br/>ScienceDirect +<br/>Scopus +<br/>SciTopics +<br/>Applications +<br/>Register +<br/>Login +<br/>Go to SciVal Suite +<br/>Search +<br/>Sources +<br/>Analytics +<br/>My alerts +<br/>My list +<br/>My settings +<br/>Quick Search +<br/>View search history | Back to results | < Previous 4 of 11 Next > +<br/>Help +<br/>Download PDF +<br/>Add to 2collab +<br/>Export +<br/>Print +<br/>E-mail +<br/>Create bibliography +<br/>Add to My List +<br/>Cited by since 1996 +<br/>Proceedings of the 2010 10th International Conference on Intelligent Systems Design and +<br/>Applications, ISDA'10 +<br/>2010, Article number 5687029, Pages 1154-1158 +<br/>This article has been cited 0 times in Scopus. +<br/>Inform me when this document is cited in Scopus: +<br/>Set alert +<br/>Set feed +<br/>ISBN: 978-142448135-4 +<br/>DOI: 10.1109/ISDA.2010.5687029 +<br/>Document Type: Conference Paper +<br/>Source Type: Conference Proceeding +<br/><b>Sponsors: Machine Intelligence Research Labs (MIR Labs</b><br/>View references (23) +<br/>My Applications +<br/>Add +<br/>More By These Authors +<br/>The authors of this article have a total of 67 records in +<br/>Scopus: +<br/>(Showing 5 most recent) +<br/>Shekofteh, S.K.,Maryam Baradaran, K.,Toosizadeh, +<br/>S.,Akbarzadeh-T., M.-R.,Hashemi, M. +<br/>Head pose estimation using fuzzy approximator +<br/>augmented by redundant membership functions +<br/>(2010)ICSTE 2010 - 2010 2nd International Conference on +<br/>Software Technology and Engineering, Proceedings +<br/>Kamkar, I.,Akbarzadeh-T, M.-R.,Yaghoobi, M. +<br/>Intelligent water drops a new optimization algorithm +<br/>Hide Applications +<br/>Find related documents +<br/>In Scopus based on +<br/>References +<br/>Authors +<br/>Keywords +<br/>Cairo; 29 November 2010 through 1 December 2010; Category number CFP10394-CDR; Code +<br/>83753 +<br/>View at publisher | +<br/>A fuzzy approximator with Gaussian membership functions +<br/>to estimate a human's head pose +<br/>Baradaran-K, M.a +<br/>, Toosizadeh, S.a +<br/><b>Islamic Azad University, Mashhad Branch, Mashhad, Iran</b><br/><b>Ferdowsi University of Mashhad, Mashhad, Iran</b><br/>, Akbarzadeh-T, M.-R.b +<br/>, Shekofteh, S.K.b +</td><td></td><td></td></tr><tr><td>118ca3b2e7c08094e2a50137b1548ada7935e505</td><td>Workshop track - ICLR 2018 +<br/>A DATASET TO EVALUATE THE REPRESENTATIONS +<br/>LEARNED BY VIDEO PREDICTION MODELS +<br/><b>Toyota Research Institute, Cambridge, MA 2 University of Michigan, Ann Arbor, MI</b></td><td>('34246012', 'Ryan Szeto', 'ryan szeto')<br/>('2307158', 'Simon Stent', 'simon stent')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')</td><td>{szetor,jjcorso}@umich.edu +<br/>{simon.stent,german.ros}@tri.global +</td></tr><tr><td>11aa527c01e61ec3a7a67eef8d7ffe9d9ce63f1d</td><td>Automated measurement of mouse social behaviors +<br/>using depth sensing, video tracking, and +<br/>machine learning +<br/>and David J. Andersona,1 +<br/><b>aDivision of Biology and Biological Engineering 156-29, Howard Hughes Medical Institute, California Institute of Technology, Pasadena, CA</b><br/><b>and bDivision of Engineering and Applied Sciences 136-93, California Institute of Technology, Pasadena, CA</b><br/>Contributed by David J. Anderson, August 16, 2015 (sent for review May 20, 2015) +<br/>A lack of automated, quantitative, and accurate assessment of social +<br/>behaviors in mammalian animal models has limited progress toward +<br/>understanding mechanisms underlying social interactions and their +<br/>disorders such as autism. Here we present a new integrated hard- +<br/>ware and software system that combines video tracking, depth +<br/>sensing, and machine learning for automatic detection and quanti- +<br/>fication of social behaviors involving close and dynamic interactions +<br/>between two mice of different coat colors in their home cage. We +<br/>designed a hardware setup that integrates traditional video cameras +<br/>with a depth camera, developed computer vision tools to extract the +<br/>body “pose” of individual animals in a social context, and used a +<br/>supervised learning algorithm to classify several well-described so- +<br/>cial behaviors. We validated the robustness of the automated classi- +<br/>fiers in various experimental settings and used them to examine how +<br/>genetic background, such as that of Black and Tan Brachyury (BTBR) +<br/>mice (a previously reported autism model), influences social behavior. +<br/>Our integrated approach allows for rapid, automated measurement +<br/>of social behaviors across diverse experimental designs and also af- +<br/>fords the ability to develop new, objective behavioral metrics. +<br/>social behavior | behavioral tracking | machine vision | depth sensing | +<br/>supervised machine learning +<br/>Social behaviors are critical for animals to survive and re- +<br/>produce. Although many social behaviors are innate, they +<br/>must also be dynamic and flexible to allow adaptation to a rap- +<br/>idly changing environment. The study of social behaviors in model +<br/>organisms requires accurate detection and quantification of such +<br/>behaviors (1–3). Although automated systems for behavioral +<br/>scoring in rodents are available (4–8), they are generally limited to +<br/>single-animal assays, and their capabilities are restricted either to +<br/>simple tracking or to specific behaviors that are measured using a +<br/>dedicated apparatus (6–11) (e.g., elevated plus maze, light-dark +<br/>box, etc.). By contrast, rodent social behaviors are typically scored +<br/>manually. This is slow, highly labor-intensive, and subjective, +<br/>resulting in analysis bottlenecks as well as inconsistencies between +<br/>different human observers. These issues limit progress toward +<br/>understanding the function of neural circuits and genes controlling +<br/>social behaviors and their dysfunction in disorders such as autism +<br/>(1, 12). In principle, these obstacles could be overcome through +<br/>the development of automated systems for detecting and mea- +<br/>suring social behaviors. +<br/>Automating tracking and behavioral measurements during +<br/>social interactions pose a number of challenges not encountered +<br/>in single-animal assays, however, especially in the home cage +<br/>environment (2). During many social behaviors, such as aggression +<br/>or mating, two animals are in close proximity and often cross or +<br/>touch each other, resulting in partial occlusion. This makes track- +<br/>ing body positions, distinguishing each mouse, and detecting be- +<br/>haviors particularly difficult. This is compounded by the fact that +<br/>such social interactions are typically measured in the animals’ +<br/>home cage, where bedding, food pellets, and other moveable items +<br/>can make tracking difficult. Nevertheless a home-cage environment +<br/>is important for studying social behaviors, because it avoids the +<br/>stress imposed by an unfamiliar testing environment. +<br/>Recently several techniques have been developed to track +<br/>social behaviors in animals with rigid exoskeletons, such as the +<br/>fruit fly Drosophila, which have relatively few degrees of freedom +<br/>in their movements (13–23). These techniques have had a trans- +<br/>formative impact on the study of social behaviors in that species +<br/>(2). Accordingly, the development of similar methods for mam- +<br/>malian animal models, such as the mouse, could have a similar +<br/>impact as well. However, endoskeletal animals exhibit diverse and +<br/>flexible postures, and their actions during any one social behavior, +<br/>such as aggression, are much less stereotyped than in flies. This +<br/>presents a dual challenge to automated behavior classification: +<br/>first, to accurately extract a representation of an animal’s posture +<br/>from observed data, and second, to map that representation to the +<br/>correct behavior (24–27). Current machine vision algorithms that +<br/>track social interactions in mice mainly use the relative positions of +<br/>two animals (25, 28–30); this approach generally cannot discrimi- +<br/>nate social interactions that involve close proximity and vigorous +<br/>physical activity, or identify specific behaviors such as aggression +<br/>and mounting. In addition, existing algorithms that measure social +<br/>interactions use a set of hardcoded, “hand-crafted” (i.e., pre- +<br/>defined) parameters that make them difficult to adapt to new ex- +<br/>perimental setups and conditions (25, 31). +<br/>In this study, we combined 3D tracking and machine learning +<br/>in an integrated system that can automatically detect, classify, +<br/><b>and quantify distinct social behaviors, including those involving</b><br/>Significance +<br/>Accurate, quantitative measurement of animal social behaviors +<br/>is critical, not only for researchers in academic institutions +<br/>studying social behavior and related mental disorders, but also for +<br/>pharmaceutical companies developing drugs to treat disorders +<br/>affecting social interactions, such as autism and schizophrenia. +<br/>Here we describe an integrated hardware and software system +<br/>that combines video tracking, depth-sensing technology, machine +<br/>vision, and machine learning to automatically detect and score +<br/>innate social behaviors, such as aggression, mating, and social +<br/>investigation, between mice in a home-cage environment. This +<br/>technology has the potential to have a transformative impact on +<br/>the study of the neural mechanisms underlying social behavior +<br/>and the development of new drug therapies for psychiatric dis- +<br/>orders in humans. +<br/>Author contributions: W.H., P.P., and D.J.A. designed research; W.H. performed research; +<br/>W.H., X.P.B.-A., and S.G.N. contributed new reagents/analytic tools; W.H., A.K., M.Z., P.P., +<br/>and D.J.A. analyzed data; and W.H., A.K., M.Z., P.P., and D.J.A. wrote the paper. +<br/>The authors declare no conflict of interest. +<br/>This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. +<br/>1073/pnas.1515982112/-/DCSupplemental. +<br/>www.pnas.org/cgi/doi/10.1073/pnas.1515982112 +<br/>PNAS | Published online September 9, 2015 | E5351–E5360 +</td><td>('4502168', 'Weizhe Hong', 'weizhe hong')<br/>('6201086', 'Ann Kennedy', 'ann kennedy')<br/>('4195968', 'Moriel Zelikowsky', 'moriel zelikowsky')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>1To whom correspondence may be addressed. Email: whong@caltech.edu, perona@ +<br/>caltech.edu, or wuwei@caltech.edu. +</td></tr><tr><td>113c22eed8383c74fe6b218743395532e2897e71</td><td>MODEC: Multimodal Decomposable Models for Human Pose Estimation +<br/>Ben Sapp +<br/><b>Google, Inc</b><br/><b>University of Washington</b></td><td>('1685978', 'Ben Taskar', 'ben taskar')</td><td>bensapp@google.com +<br/>taskar@cs.washington.edu +</td></tr><tr><td>11408af8861fb0a977412e58c1a23d61b8df458c</td><td>A Robust Learning Algorithm Based on +<br/>SURF and PSM for Facial Expression Recognition +<br/><b>Graduate School of Engineering, Kobe University, Kobe, 657-8501, Japan</b><br/><b>Graduate School of System Informatics, Kobe University, Kobe, 657-8501, Japan</b></td><td>('2866465', 'Jinhui Chen', 'jinhui chen')<br/>('39484328', 'Xiaoyan Lin', 'xiaoyan lin')<br/>('1744026', 'Tetsuya Takiguchi', 'tetsuya takiguchi')<br/>('1678564', 'Yasuo Ariki', 'yasuo ariki')</td><td>ianchen@me.cs.scitec.kobe-u.ac.jp, {takigu,ariki}@kobe-u.ac.jp +</td></tr><tr><td>11cc0774365b0cc0d3fa1313bef3d32c345507b1</td><td>Face Recognition Using Active Near-IR +<br/>Illumination +<br/>Centre for Vision, Speech and Signal Processing +<br/><b>University of Surrey, United Kingdom</b><br/> +</td><td>('38746097', 'Xuan Zou', 'xuan zou')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('2173900', 'Kieron Messer', 'kieron messer')</td><td>@surrey.ac.uk +</td></tr><tr><td>11f7f939b6fcce51bdd8f3e5ecbcf5b59a0108f5</td><td>Rolling Riemannian Manifolds to Solve the Multi-class Classification Problem +<br/><b>Institute of Systems and Robotics - University of Coimbra, Portugal</b><br/><b>Portugal</b></td><td>('2117944', 'Rui Caseiro', 'rui caseiro')<br/>('39458914', 'Pedro Martins', 'pedro martins')<br/>('36478254', 'João F. Henriques', 'joão f. henriques')</td><td>{ruicaseiro, pedromartins, henriques, batista}@isr.uc.pt, fleite@mat.uc.pt +</td></tr><tr><td>11269e98f072095ff94676d3dad34658f4876e0e</td><td>Facial Expression Recognition with Multithreaded +<br/>Cascade of Rotation-invariant HOG +<br/>Graduate School of System Informatics +<br/>Graduate School of System Informatics +<br/>Graduate School of System Informatics +<br/><b>Kobe University</b><br/>Kobe, 657-8501, Japan +<br/><b>Kobe University</b><br/>Kobe, 657-8501, Japan +<br/><b>Kobe University</b><br/>Kobe, 657-8501, Japan +<br/>In this paper, we propose a novel framework that adopts +<br/>robust feature representation for training the multithreading +<br/>boosting cascade. We adopt rotation-invariant HOG (Ri-HOG) +<br/>as features, which is reminiscent of Dalal et al.’s HOG [9]. +<br/>However, in this paper, we noticeably enhance the conven- +<br/>tional HOG in rotation-invariant ability and feature extraction +<br/>speed. We carry out a detailed study of the effects of various +<br/>implementation choices in descriptor performance. We subdi- +<br/>vide the local patch into annular spatial bins to achieve spatial +<br/>binning invariance. Besides, we apply radial gradient to attain +<br/>gradient binning invariance, which is inspired by Takacs et +<br/>al.’s RGT (Radial Gradient Transform) [10]. +</td><td>('2866465', 'Jinhui Chen', 'jinhui chen')<br/>('1744026', 'Tetsuya Takiguchi', 'tetsuya takiguchi')<br/>('1678564', 'Yasuo Ariki', 'yasuo ariki')</td><td>Email: ianchen@me.cs.scitec.kobe-u.ac.jp +<br/>Email: takigu@kobe-u.ac.jp +<br/>Email: ariki@kobe-u.ac.jp +</td></tr><tr><td>113e5678ed8c0af2b100245057976baf82fcb907</td><td>Facing Imbalanced Data +<br/>Recommendations for the Use of Performance Metrics +</td><td>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td>1Carnegie Mellon University, Pittsburgh, PA, laszlo.jeni@ieee.org,ftorre@cs.cmu.edu +<br/>2University of Pittsburgh, Pittsburgh, PA, jeffcohn@cs.cmu.edu +</td></tr><tr><td>11691f1e7c9dbcbd6dfd256ba7ac710581552baa</td><td>SoccerNet: A Scalable Dataset for Action Spotting in Soccer Videos +<br/><b>King Abdullah University of Science and Technology (KAUST), Saudi Arabia</b></td><td>('22314218', 'Silvio Giancola', 'silvio giancola')<br/>('41022271', 'Mohieddine Amine', 'mohieddine amine')<br/>('41015552', 'Tarek Dghaily', 'tarek dghaily')<br/>('2931652', 'Bernard Ghanem', 'bernard ghanem')</td><td>silvio.giancola@kaust.edu.sa, maa249@mail.aub.edu, tad05@mail.aub.edu, bernard.ghanem@kaust.edu.sa +</td></tr><tr><td>11c04c4f0c234a72f94222efede9b38ba6b2306c</td><td>Real-Time Human Action Recognition by Luminance Field +<br/>Trajectory Analysis +<br/>Dept of Computing +<br/>Kowloon, Hong Kong +<br/>+852 2766-7316 +<br/><b>Hong Kong Polytechnic University</b><br/><b>University of Illinois at Urbana</b><br/><b>National University of Singapore</b><br/>Dept of ECE +<br/>Champaign, USA +<br/>+1 217-244-2960 +<br/>Dept of ECE +<br/>Singapore +<br/>+65 6516-2116 +</td><td>('2659956', 'Zhu Li', 'zhu li')<br/>('1708679', 'Yun Fu', 'yun fu')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>zhu.li@ieee.org +<br/>{yunfu2,huang}@ifp.uiuc.edu +<br/>elesyan@ece.nus.edu.sg +</td></tr><tr><td>1128a4f57148cec96c0ef4ae3b5a0fbf07efbad9</td><td>Action Recognition by Learning Deep Multi-Granular +<br/>Spatio-Temporal Video Representation∗ +<br/><b>University of Science and Technology of China, Hefei 230026, P. R. China</b><br/>2 Microsoft Research, Beijing 100080, P. R. China +<br/><b>University of Rochester, NY 14627, USA</b></td><td>('35539590', 'Qing Li', 'qing li')<br/>('3430743', 'Zhaofan Qiu', 'zhaofan qiu')<br/>('2053452', 'Ting Yao', 'ting yao')<br/>('1724211', 'Tao Mei', 'tao mei')<br/>('3663422', 'Yong Rui', 'yong rui')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')</td><td>{sealq, qiudavy}@mail.ustc.edu.cn; {tiyao, tmei, yongrui}@microsoft.com; +<br/>jluo@cs.rochester.edu +</td></tr><tr><td>1149c6ac37ae2310fe6be1feb6e7e18336552d95</td><td>Proc. Int. Conf. on Artificial Neural Networks (ICANN’05), Warsaw, LNCS 3696, vol. I, pp. 569-574, Springer Verlag 2005 +<br/>Classification of Face Images for Gender, Age, +<br/>Facial Expression, and Identity1 +<br/>Department of Neuroinformatics and Cognitive Robotics +<br/><b>Ilmenau Technical University, P.O.Box 100565, 98684 Ilmenau, Germany</b></td><td>('34420922', 'Torsten Wilhelm', 'torsten wilhelm')</td><td></td></tr><tr><td>11f17191bf74c80ad0b16b9f404df6d03f7c8814</td><td>Recognition of Visually Perceived Compositional +<br/>Human Actions by Multiple Spatio-Temporal Scales +<br/>Recurrent Neural Networks +</td><td>('1754201', 'Minju Jung', 'minju jung')<br/>('1780524', 'Jun Tani', 'jun tani')</td><td></td></tr><tr><td>11367581c308f4ba6a32aac1b4a7cdb32cd63137</td><td></td><td></td><td></td></tr><tr><td>11a47a91471f40af5cf00449954474fd6e9f7694</td><td>Article +<br/>NIRFaceNet: A Convolutional Neural Network for +<br/>Near-Infrared Face Identification +<br/><b>Southwest University, Chongqing 400715, China</b><br/>† These authors contribute equally to this work. +<br/>Academic Editor: Willy Susilo +<br/>Received: 16 July 2016; Accepted: 24 October 2016; Published: 27 October 2016 +</td><td>('34063916', 'Min Peng', 'min peng')<br/>('8206607', 'Chongyang Wang', 'chongyang wang')<br/>('34520676', 'Tong Chen', 'tong chen')<br/>('2373829', 'Guangyuan Liu', 'guangyuan liu')</td><td>peng2014m@email.swu.edu.cn (M.P.); mvrjustid520@email.swu.edu.cn (C.W.); liugy@swu.edu.cn (G.L.) +<br/>* Correspondence: c_tong@swu.edu.cn; Tel.: +86-23-6825-4309 +</td></tr><tr><td>1198572784788a6d2c44c149886d4e42858d49e4</td><td>Learning Discriminative Features using Encoder/Decoder type Deep +<br/>Neural Nets +</td><td>('2162592', 'Vishwajeet Singh', 'vishwajeet singh')<br/>('40835709', 'Killamsetti Ravi Kumar', 'killamsetti ravi kumar')</td><td>1ALPES, Bolarum, Hyderabad 500010, vsthakur@gmail.com +<br/>2ALPES, Bolarum, Hyderabad 500010, ravi.killamsetti@gmail.com +<br/>3SNIST, Ghatkesar, Hyderabad 501301, kumar.e@gmail.com +</td></tr><tr><td>11fe6d45aa2b33c2ec10d9786a71c15ec4d3dca8</td><td>970 +<br/>JUNE 2008 +<br/>Tied Factor Analysis for Face Recognition +<br/>across Large Pose Differences +</td><td>('1792404', 'James H. Elder', 'james h. elder')<br/>('1734784', 'Jonathan Warrell', 'jonathan warrell')<br/>('2338011', 'Fatima M. Felisberti', 'fatima m. felisberti')</td><td></td></tr><tr><td>1134a6be0f469ff2c8caab266bbdacf482f32179</td><td>IJRET: International Journal of Research in Engineering and Technology eISSN: 2319-1163 | pISSN: 2321-7308 +<br/>FACIAL EXPRESSION IDENTIFICATION USING FOUR-BIT CO- +<br/>OCCURRENCE MATRIXFEATURES AND K-NN CLASSIFIER +<br/><b>Aditya College of Engineering, Surampalem, East Godavari</b><br/>District, Andhra Pradesh, India +</td><td>('8118823', 'Bala Shankar', 'bala shankar')<br/>('27686729', 'S R Kumar', 's r kumar')</td><td></td></tr><tr><td>11b3877df0213271676fa8aa347046fd4b1a99ad</td><td>Unsupervised Identification of Multiple Objects of +<br/>Interest from Multiple Images: dISCOVER +<br/><b>Carnegie Mellon University</b></td><td>('1713589', 'Devi Parikh', 'devi parikh')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td>{dparikh,tsuhan}@cmu.edu +</td></tr><tr><td>112780a7fe259dc7aff2170d5beda50b2bfa7bda</td><td></td><td></td><td></td></tr><tr><td>1130c38e88108cf68b92ecc61a9fc5aeee8557c9</td><td>Dynamically Encoded Actions based on Spacetime Saliency +<br/><b>Institute of Electrical Measurement and Measurement Signal Processing, TU Graz, Austria</b><br/><b>York University, Toronto, Canada</b></td><td>('2322150', 'Christoph Feichtenhofer', 'christoph feichtenhofer')<br/>('1718587', 'Axel Pinz', 'axel pinz')<br/>('1709096', 'Richard P. Wildes', 'richard p. wildes')</td><td>{feichtenhofer, axel.pinz}@tugraz.at +<br/>wildes@cse.yorku.ca +</td></tr><tr><td>11b89011298e193d9e6a1d99302221c1d8645bda</td><td>Structured Feature Selection +<br/><b>Rensselaer Polytechnic Institute, USA</b></td><td>('39965604', 'Tian Gao', 'tian gao')<br/>('2860279', 'Ziheng Wang', 'ziheng wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>{gaot, wangz10, jiq}@rpi.edu +</td></tr><tr><td>111a9645ad0108ad472b2f3b243ed3d942e7ff16</td><td>Facial Expression Classification Using +<br/>Combined Neural Networks +<br/>DEE/PUC-Rio, Marquês de São Vicente 225, Rio de Janeiro – RJ - Brazil +</td><td>('14032279', 'Rafael V. Santos', 'rafael v. santos')<br/>('1744578', 'Marley M.B.R. Vellasco', 'marley m.b.r. vellasco')<br/>('34686777', 'Raul Q. Feitosa', 'raul q. feitosa')<br/>('1687882', 'Ricardo Tanscheit', 'ricardo tanscheit')</td><td>marley@ele.puc-rio.br +</td></tr><tr><td>1177977134f6663fff0137f11b81be9c64c1f424</td><td>Multi-Manifold Deep Metric Learning for Image Set Classification +<br/>1Advanced Digital Sciences Center, Singapore +<br/><b>School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore</b><br/><b>School of ICE, Beijing University of Posts and Telecommunications, Beijing, China</b><br/><b>University of Illinois at Urbana-Champaign, Urbana, IL, USA</b><br/><b>Tsinghua University, Beijing, China</b></td><td>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('22804340', 'Gang Wang', 'gang wang')<br/>('1774956', 'Weihong Deng', 'weihong deng')<br/>('1742248', 'Pierre Moulin', 'pierre moulin')<br/>('39491387', 'Jie Zhou', 'jie zhou')</td><td>jiwen.lu@adsc.com.sg; wanggang@ntu.edu.sg; whdeng@bupt.edu.cn; +<br/>moulin@ifp.uiuc.edu; jzhou@tsinghua.edu.cn +</td></tr><tr><td>1190cba0cae3c8bb81bf80d6a0a83ae8c41240bc</td><td>Squared Earth Mover’s Distance Loss for Training +<br/>Deep Neural Networks on Ordered-Classes +<br/>Dept. of Computer Science +<br/><b>Stony Brook University</b><br/>Chen-Ping Yu +<br/><b>Phiar Technologies, Inc</b></td><td>('2321406', 'Le Hou', 'le hou')</td><td></td></tr><tr><td>111d0b588f3abbbea85d50a28c0506f74161e091</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 134 – No.10, January 2016 +<br/>Facial Expression Recognition from Visual Information +<br/>using Curvelet Transform +<br/>Surabhi Group of Institution Bhopal +<br/>systems. Further applications +</td><td>('6837599', 'Pratiksha Singh', 'pratiksha singh')</td><td></td></tr><tr><td>11ac88aebe0230e743c7ea2c2a76b5d4acbfecd0</td><td>Hybrid Cascade Model for Face Detection in the Wild +<br/>Based on Normalized Pixel Difference and a Deep +<br/>Convolutional Neural Network +<br/>Darijan Marčetić[0000-0002-6556-665X], Martin Soldić[0000-0002-4031-0404] +<br/>and Slobodan Ribarić[0000-0002-8708-8513] +<br/><b>University of Zagreb, Faculty of Electrical Engineering and Computing, Croatia</b></td><td></td><td>{darijan.marcetic, martin.soldic, slobodan.ribaric}@fer.hr +</td></tr><tr><td>117f164f416ea68e8b88a3005e55a39dbdf32ce4</td><td>Neuroaesthetics in Fashion: Modeling the Perception of Fashionability +<br/>1Institut de Rob`otica i Inform`atica Industrial (CSIC-UPC), +<br/><b>University of Toronto</b></td><td>('3114470', 'Edgar Simo-Serra', 'edgar simo-serra')<br/>('37895334', 'Sanja Fidler', 'sanja fidler')<br/>('1994318', 'Francesc Moreno-Noguer', 'francesc moreno-noguer')<br/>('2422559', 'Raquel Urtasun', 'raquel urtasun')</td><td></td></tr><tr><td>7dda2eb0054eb1aeda576ed2b27a84ddf09b07d4</td><td>2010 The 3rd International Conference on Machine Vision (ICMV 2010) +<br/>Face Recognition and Representation by Tensor-based MPCA Approach +<br/>Dept. of Control, Instrumentation, and Robot +<br/>Engineering +<br/><b>Chosun University</b><br/>Gwangju, Korea +</td><td>('2806903', 'Yun-Hee Han', 'yun-hee han')</td><td>Yhhan1059@gmail.com +</td></tr><tr><td>7d2556d674ad119cf39df1f65aedbe7493970256</td><td>Now You Shake Me: Towards Automatic 4D Cinema +<br/><b>University of Toronto</b><br/><b>Vector Institute</b><br/>http://www.cs.toronto.edu/˜henryzhou/movie4d/ +</td><td>('2481662', 'Yuhao Zhou', 'yuhao zhou')<br/>('37895334', 'Sanja Fidler', 'sanja fidler')</td><td>{henryzhou, makarand, fidler}@cs.toronto.edu +</td></tr><tr><td>7d94fd5b0ca25dd23b2e36a2efee93244648a27b</td><td>Convolutional Network for Attribute-driven and Identity-preserving Human Face +<br/>Generation +<br/><b>The Hong Kong Polytechnic University, Hong Kong</b><br/><b>bSchool of Computer Science and Technology, Harbin Institute of Technology, Harbin 150001, China</b></td><td>('1701799', 'Mu Li', 'mu li')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('1698371', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>7d8c2d29deb80ceed3c8568100376195ce0914cb</td><td>Identity-Aware Textual-Visual Matching with Latent Co-attention +<br/><b>The Chinese University of Hong Kong</b></td><td>('1700248', 'Shuang Li', 'shuang li')<br/>('1721881', 'Tong Xiao', 'tong xiao')<br/>('1764548', 'Hongsheng Li', 'hongsheng li')<br/>('1742383', 'Wei Yang', 'wei yang')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')</td><td>{sli,xiaotong,hsli,wyang,xgwang}@ee.cuhk.edu.hk +</td></tr><tr><td>7d306512b545df98243f87cb8173df83b4672b18</td><td>Flag Manifolds for the Characterization of +<br/>Geometric Structure in Large Data Sets +<br/>T. Marrinan, J. R. Beveridge, B. Draper, M. Kirby, and C. Peterson +<br/><b>Colorado State University, Fort Collins, Colorado, USA</b></td><td></td><td>kirby@math.colostate.edu +</td></tr><tr><td>7d98dcd15e28bcc57c9c59b7401fa4a5fdaa632b</td><td>FACE APPEARANCE FACTORIZATION FOR EXPRESSION ANALYSIS AND SYNTHESIS +<br/><b>Heudiasyc Laboratory, CNRS, University of Technology of Compi`egne</b><br/>BP 20529, 60205 COMPIEGNE Cedex, FRANCE. +</td><td>('2371236', 'Bouchra Abboud', 'bouchra abboud')<br/>('1742818', 'Franck Davoine', 'franck davoine')</td><td>E-mail: Bouchra.Abboud@hds.utc.fr +</td></tr><tr><td>7d41b67a641426cb8c0f659f0ba74cdb60e7159a</td><td>Soft Biometric Retrieval to Describe and Identify Surveillance Images +<br/>School of Electronics and Computer Science, +<br/><b>University of Southampton, United Kingdom</b></td><td>('3408521', 'Daniel Martinho-Corbishley', 'daniel martinho-corbishley')<br/>('1727698', 'Mark S. Nixon', 'mark s. nixon')<br/>('3000521', 'John N. Carter', 'john n. carter')</td><td>{dmc,msn,jnc}@ecs.soton.ac.uk +</td></tr><tr><td>7d1688ce0b48096e05a66ead80e9270260cb8082</td><td>Real vs. Fake Emotion Challenge: Learning to Rank Authenticity From Facial +<br/>Activity Descriptors +<br/><b>Otto von Guericke University</b><br/>Magdeburg, Germany +</td><td>('2441656', 'Frerk Saxen', 'frerk saxen')<br/>('1783606', 'Philipp Werner', 'philipp werner')<br/>('1741165', 'Ayoub Al-Hamadi', 'ayoub al-hamadi')</td><td>{Frerk.Saxen, Philipp.Werner, Ayoub.Al-Hamadi}@ovgu.de +</td></tr><tr><td>7d53678ef6009a68009d62cd07c020706a2deac3</td><td>Facial Feature Point Extraction using +<br/>the Adaptive Mean Shape in Active Shape Model +<br/><b>Hanyang University</b><br/>Haengdang-dong, Seongdong-gu, Seoul, South Korea +<br/>Giheung-eup, Yongin-si, Gyeonggi-do, Seoul, Korea +<br/><b>Samsung Advanced Institute of Technology</b></td><td>('2771795', 'Hyoung-Joon Kim', 'hyoung-joon kim')<br/>('34600044', 'Wonjun Hwang', 'wonjun hwang')<br/>('2077154', 'Seok-Cheol Kee', 'seok-cheol kee')<br/>('2982904', 'Whoi-Yul Kim', 'whoi-yul kim')<br/>('40370422', 'Hyun-Chul Kim', 'hyun-chul kim')</td><td>{hckim, khjoon}@vision.hanyang.ac.kr, wykim@hanyang.ac.kr +<br/>{wj.hwang, sckee}@samsung.com +</td></tr><tr><td>7d7be6172fc2884e1da22d1e96d5899a29831ad2</td><td>L2GSCI: Local to Global Seam Cutting and Integrating for +<br/>Accurate Face Contour Extraction +<br/><b>South China University of China</b><br/><b>South China University of China</b><br/><b>Kitware, Inc</b><br/><b>The Education University of Hong Kong</b><br/><b>South China University of China</b></td><td>('37221211', 'Yongwei Nie', 'yongwei nie')<br/>('37579534', 'Xu Cao', 'xu cao')<br/>('2792312', 'Chengjiang Long', 'chengjiang long')<br/>('2420746', 'Ping Li', 'ping li')<br/>('4882057', 'Guiqing Li', 'guiqing li')</td><td>nieyongwei@scut.edu.cn +</td></tr><tr><td>7de6e81d775e9cd7becbfd1bd685f4e2a5eebb22</td><td>Labeled Faces in the Wild: A Survey +</td><td>('1714536', 'Erik Learned-Miller', 'erik learned-miller')<br/>('1799600', 'Gary Huang', 'gary huang')<br/>('2895705', 'Aruni RoyChowdhury', 'aruni roychowdhury')<br/>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('1745420', 'Gang Hua', 'gang hua')</td><td></td></tr><tr><td>7d73adcee255469aadc5e926066f71c93f51a1a5</td><td>978-1-4799-9988-0/16/$31.00 ©2016 IEEE +<br/>1283 +<br/>ICASSP 2016 +</td><td></td><td></td></tr><tr><td>7df4f96138a4e23492ea96cf921794fc5287ba72</td><td>A Jointly Learned Deep Architecture for Facial Attribute Analysis and Face +<br/>Detection in the Wild +<br/><b>Fudan University</b></td><td>('37391748', 'Keke He', 'keke he')<br/>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')</td><td>{kkhe15, yanweifu, xyxue}@fudan.edu.cn +</td></tr><tr><td>7d9fe410f24142d2057695ee1d6015fb1d347d4a</td><td>Facial Expression Feature Extraction Based on +<br/>FastLBP +<br/><b>Beijing, China</b><br/><b>Beijing, China</b><br/>facial expression +</td><td>('1921151', 'Ya Zheng', 'ya zheng')<br/>('2780963', 'Xiuxin Chen', 'xiuxin chen')<br/>('2671173', 'Chongchong Yu', 'chongchong yu')<br/>('39681852', 'Cheng Gao', 'cheng gao')</td><td>Email: zy_lovedabao@163.com +<br/>Email: chenxx1979@126.com, chongzhy@vip.sina.com, gcandgh@163.com +</td></tr><tr><td>7dd578878e84337d6d0f5eb593f22cabeacbb94c</td><td>Classifiers for Driver Activity Monitoring +<br/>Department of Computer Science and Engineering +<br/><b>University of Minnesota</b></td><td>('3055503', 'Harini Veeraraghavan', 'harini veeraraghavan')<br/>('32975623', 'Nathaniel Bird', 'nathaniel bird')<br/>('1734862', 'Stefan Atev', 'stefan atev')<br/>('1696163', 'Nikolaos Papanikolopoulos', 'nikolaos papanikolopoulos')</td><td>harini@cs.umn.edu bird@cs.umn.edu atev@cs.umn.edu npapas@cs.umn.edu +</td></tr><tr><td>7dffe7498c67e9451db2d04bb8408f376ae86992</td><td>LEAR-INRIA submission for the THUMOS workshop +<br/>LEAR, INRIA, France +</td><td>('40465030', 'Heng Wang', 'heng wang')</td><td>firstname.lastname@inria.fr +</td></tr><tr><td>7df268a3f4da7d747b792882dfb0cbdb7cc431bc</td><td>Semi-supervised Adversarial Learning to Generate +<br/>Photorealistic Face Images of New Identities from 3D +<br/>Morphable Model +<br/><b>Imperial College London, UK</b><br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, UK</b></td><td>('2151914', 'Baris Gecer', 'baris gecer')<br/>('48467774', 'Binod Bhattarai', 'binod bhattarai')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('1700968', 'Tae-Kyun Kim', 'tae-kyun kim')</td><td>{b.gecer,b.bhattarai,tk.kim}@imperial.ac.uk, +<br/>j.kittler@surrey.ac.uk +</td></tr><tr><td>7d3f6dd220bec883a44596ddec9b1f0ed4f6aca2</td><td>2106 +<br/>Linear Regression for Face Recognition +</td><td>('2095953', 'Imran Naseem', 'imran naseem')<br/>('2444665', 'Roberto Togneri', 'roberto togneri')<br/>('1698675', 'Mohammed Bennamoun', 'mohammed bennamoun')</td><td></td></tr><tr><td>7de386bf2a1b2436c836c0cc1f1f23fccb24aad6</td><td>Finding What the Driver Does +<br/>Final Report +<br/>Prepared by: +<br/>Artificial Intelligence, Robotics, and Vision Laboratory +<br/>Department of Computer Science and Engineering +<br/><b>University of Minnesota</b><br/>CTS 05-03 +<br/>HUMAN-CENTERED TECHNOLOGY TO ENHANCE SAFETY AND MOBILITY +</td><td>('3055503', 'Harini Veeraraghavan', 'harini veeraraghavan')<br/>('1734862', 'Stefan Atev', 'stefan atev')<br/>('32975623', 'Nathaniel Bird', 'nathaniel bird')<br/>('31791248', 'Paul Schrater', 'paul schrater')<br/>('40654170', 'Nilolaos Papanikolopoulos', 'nilolaos papanikolopoulos')</td><td></td></tr><tr><td>29ce6b54a87432dc8371f3761a9568eb3c5593b0</td><td>Kent Academic Repository +<br/>Full text document (pdf) +<br/>Citation for published version +<br/>Yassin, DK H. PHM and Hoque, Sanaul and Deravi, Farzin (2013) Age Sensitivity of Face Recognition +<br/> pp. 12-15. +<br/>DOI +<br/>https://doi.org/10.1109/EST.2013.8 +<br/>Link to record in KAR +<br/>http://kar.kent.ac.uk/43222/ +<br/>Document Version +<br/>Author's Accepted Manuscript +<br/>Copyright & reuse +<br/>Content in the Kent Academic Repository is made available for research purposes. Unless otherwise stated all +<br/>content is protected by copyright and in the absence of an open licence (eg Creative Commons), permissions +<br/>for further reuse of content should be sought from the publisher, author or other copyright holder. +<br/>Versions of research +<br/>The version in the Kent Academic Repository may differ from the final published version. +<br/>Users are advised to check http://kar.kent.ac.uk for the status of the paper. Users should always cite the +<br/>published version of record. +<br/>Enquiries +<br/>For any further enquiries regarding the licence status of this document, please contact: +<br/>If you believe this document infringes copyright then please contact the KAR admin team with the take-down +<br/>information provided at http://kar.kent.ac.uk/contact.html +</td><td></td><td>researchsupport@kent.ac.uk +</td></tr><tr><td>2914e8c62f0432f598251fae060447f98141e935</td><td><b>University of Nebraska - Lincoln</b><br/>Computer Science and Engineering: Theses, +<br/>Dissertations, and Student Research +<br/>Computer Science and Engineering, Department of +<br/>8-2016 +<br/>ACTIVITY ANALYSIS OF SPECTATOR +<br/>PERFORMER VIDEOS USING MOTION +<br/>TRAJECTORIES +<br/>Follow this and additional works at: http://digitalcommons.unl.edu/computerscidiss +<br/>Part of the Computer Engineering Commons +<br/>Timsina, Anish, "ACTIVITY ANALYSIS OF SPECTATOR PERFORMER VIDEOS USING MOTION TRAJECTORIES" (2016). +<br/>Computer Science and Engineering: Theses, Dissertations, and Student Research. Paper 107. +<br/>http://digitalcommons.unl.edu/computerscidiss/107 +<br/>Nebraska - Lincoln. It has been accepted for inclusion in Computer Science and Engineering: Theses, Dissertations, and Student Research by an +</td><td>('2404944', 'Anish Timsina', 'anish timsina')</td><td>DigitalCommons@University of Nebraska - Lincoln +<br/>University of Nebraska-Lincoln, timsina.anish@gmail.com +<br/>This Article is brought to you for free and open access by the Computer Science and Engineering, Department of at DigitalCommons@University of +<br/>authorized administrator of DigitalCommons@University of Nebraska - Lincoln. +</td></tr><tr><td>292eba47ef77495d2613373642b8372d03f7062b</td><td>Deep Secure Encoding: An Application to Face Recognition +</td><td>('39192292', 'Rohit Pandey', 'rohit pandey')<br/>('34872128', 'Yingbo Zhou', 'yingbo zhou')<br/>('1723877', 'Venu Govindaraju', 'venu govindaraju')</td><td></td></tr><tr><td>29e96ec163cb12cd5bd33bdf3d32181c136abaf9</td><td>Report No. UIUCDCS-R-2006-2748 +<br/>UILU-ENG-2006-1788 +<br/>Regularized Locality Preserving Projections with Two-Dimensional +<br/>Discretized Laplacian Smoothing +<br/>by +<br/>July 2006 +</td><td>('1724421', 'Deng Cai', 'deng cai')<br/>('3945955', 'Xiaofei He', 'xiaofei he')<br/>('39639296', 'Jiawei Han', 'jiawei han')</td><td></td></tr><tr><td>29e793271370c1f9f5ac03d7b1e70d1efa10577c</td><td>International Journal of Signal Processing, Image Processing and Pattern Recognition +<br/>Vol.6, No.5 (2013), pp.423-436 +<br/>http://dx.doi.org/10.14257/ijsip.2013.6.5.37 +<br/>Face Recognition Based on Multi-classifierWeighted Optimization +<br/>and Sparse Representation +<br/><b>Institute of control science and engineering</b><br/><b>University of Science and Technology Beijing</b><br/>1,2,330 Xueyuan Road, Haidian District, Beijing 100083 P. R.China +</td><td>('11241192', 'Deng Nan', 'deng nan')<br/>('7814565', 'Zhengguang Xu', 'zhengguang xu')</td><td>1dengnan666666@163.com, 2xzg_1@263.net, 3 xiaobian@ustb.edu.cn +</td></tr><tr><td>2902f62457fdf7e8e8ee77a9155474107a2f423e</td><td>Non-rigid 3D Shape Registration using an +<br/>Adaptive Template +<br/><b>University of York, UK</b></td><td>('1694260', 'Hang Dai', 'hang dai')<br/>('1737428', 'Nick Pears', 'nick pears')<br/>('32131827', 'William Smith', 'william smith')</td><td>{hd816,nick.pears,william.smith}@york.ac.uk +</td></tr><tr><td>29d3ed0537e9ef62fd9ccffeeb72c1beb049e1ea</td><td>Parametric Dictionaries and Feature Augmentation for +<br/>Continuous Domain Adaptation∗ +<br/>Adobe Research +<br/>Bangalore, India +<br/>Light +<br/>Paolo Alto, USA +<br/><b>University of Maryland</b><br/><b>College Park, USA</b></td><td>('35223379', 'Sumit Shekhar', 'sumit shekhar')<br/>('34711525', 'Nitesh Shroff', 'nitesh shroff')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>sshekha@umiacs.umd.edu +<br/>nshroff@umiacs.umd.edu +<br/>rama@umiacs.umd.edu +</td></tr><tr><td>29c7dfbbba7a74e9aafb6a6919629b0a7f576530</td><td>Automatic Facial Expression Analysis and Emotional +<br/>Classification +<br/>by +<br/>Submitted to the Department of Math and Natural Sciences +<br/>in partial fulfillment of the requirements for the degree of a +<br/>Diplomingenieur der Optotechnik und Bildverarbeitung (FH) +<br/>(Diplom Engineer of Photonics and Image Processing) +<br/>at the +<br/><b>UNIVERSITY OF APPLIED SCIENCE DARMSTADT (FHD</b><br/>Accomplished and written at the +<br/><b>MASSACHUSETTS INSTITUTE OF TECHNOLOGY (MIT</b><br/>October 2004 +<br/>Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Department of Math and Natural Sciences +<br/>October 30, 2004 +<br/>Certified by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Dr. Harald Scharfenberg +<br/>Professor at FHD +<br/>Thesis Supervisor +<br/>Accepted by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>visiting scientist at MIT +</td><td>('40163324', 'Robert Fischer', 'robert fischer')<br/>('1684626', 'Bernd Heisele', 'bernd heisele')</td><td></td></tr><tr><td>292c6b743ff50757b8230395c4a001f210283a34</td><td>Fast Violence Detection in Video +<br/>O. Deniz1, I. Serrano1, G. Bueno1 and T-K. Kim2 +<br/><b>VISILAB group, University of Castilla-La Mancha, E.T.S.I.Industriales, Avda. Camilo Jose Cela s.n, 13071 Spain</b><br/><b>Imperial College, South Kensington Campus, London SW7 2AZ, UK</b><br/>Keywords: +<br/>action recognition, violence detection, fight detection +</td><td></td><td>{oscar.deniz, ismael.serrano, gloria.bueno}@uclm.es, tk.kim@imperial.ac.uk +</td></tr><tr><td>29fc4de6b680733e9447240b42db13d5832e408f</td><td>International Journal of Multimedia and Ubiquitous Engineering +<br/>Vol. 10, No. 3 (2015), pp. 35-44 +<br/>http://dx.doi.org/10.14257/ijmue.2015.10.3.04 +<br/>Recognition of Facial Expressions Based on Tracking and +<br/>Selection of Discriminative Geometric Features +<br/><b>Korea Electronics Technology Institute, Jeonju-si, Jeollabuk-do 561-844, Rep. of</b><br/>Korea +<br/><b>Chonbuk National University, Jeonju-si</b><br/>Jeollabuk-do 561-756, Rep. of Korea +<br/><b>School of Computing Science, Simon Fraser University, Burnaby, B.C., Canada</b></td><td>('32322842', 'Deepak Ghimire', 'deepak ghimire')<br/>('2034182', 'Joonwhoan Lee', 'joonwhoan lee')<br/>('1689656', 'Ze-Nian Li', 'ze-nian li')<br/>('1682436', 'Sunghwan Jeong', 'sunghwan jeong')<br/>('1937680', 'Hyo Sub Choi', 'hyo sub choi')</td><td>deepak@keti.re.kr, chlee@jbnu.ac.kr, li@sfu.ca, shjeong@keti.re.kr, +<br/>shpark@keti.re.kr, hschoi@keti.re.kr +</td></tr><tr><td>29c1f733a80c1e07acfdd228b7bcfb136c1dff98</td><td></td><td></td><td></td></tr><tr><td>29f27448e8dd843e1c4d2a78e01caeaea3f46a2d</td><td></td><td></td><td></td></tr><tr><td>294d1fa4e1315e1cf7cc50be2370d24cc6363a41</td><td>2008 SPIE Digital Library -- Subscriber Archive Copy +</td><td></td><td></td></tr><tr><td>29d414bfde0dfb1478b2bdf67617597dd2d57fc6</td><td>Multidim Syst Sign Process (2010) 21:213–229 +<br/>DOI 10.1007/s11045-009-0099-y +<br/>Perfect histogram matching PCA for face recognition +<br/>Received: 10 August 2009 / Revised: 21 November 2009 / Accepted: 29 December 2009 / +<br/>Published online: 14 January 2010 +<br/>© Springer Science+Business Media, LLC 2010 +</td><td>('2413241', 'Ana-Maria Sevcenco', 'ana-maria sevcenco')</td><td></td></tr><tr><td>2912c3ea67678a1052d7d5cbe734a6ad90fc360e</td><td>Facial Feature Detection using a Virtual Structuring +<br/>Element +<br/>Intelligent Systems Lab Amsterdam, +<br/><b>University of Amsterdam</b><br/>Kruislaan 403, 1098 SJ Amsterdam, The Netherlands +<br/>Keywords: Feature Detection, Active Appearance Models +</td><td>('9301018', 'Roberto Valenti', 'roberto valenti')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td>rvalenti@science.uva.nl +<br/>nicu@science.uva.nl +<br/>gevers@science.uva.nl +</td></tr><tr><td>29f4ac49fbd6ddc82b1bb697820100f50fa98ab6</td><td>The Benefits and Challenges of Collecting Richer Object Annotations +<br/>Department of Computer Science +<br/><b>University of Illinois Urbana Champaign</b></td><td>('2831988', 'Ian Endres', 'ian endres')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('2433269', 'Derek Hoiem', 'derek hoiem')<br/>('1744452', 'David A. Forsyth', 'david a. forsyth')</td><td>{iendres2,afarhad2,dhoiem,daf}@uiuc.edu +</td></tr><tr><td>2910fcd11fafee3f9339387929221f4fc1160973</td><td>Evaluating Open-Universe Face Identification on the Web +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA</b><br/><b>Center for Research in Computer Vision, University of Central Florida, Orlando, FL</b></td><td>('16131262', 'Enrique G. Ortiz', 'enrique g. ortiz')</td><td>brian@briancbecker.com and eortiz@cs.ucf.edu +</td></tr><tr><td>29479bb4fe8c04695e6f5ae59901d15f8da6124b</td><td>Multiple Instance Learning for Labeling Faces in +<br/>Broadcasting News Video +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +</td><td>('38936351', 'Jun Yang', 'jun yang')<br/>('2005689', 'Rong Yan', 'rong yan')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td>juny@cs.cmu.edu +<br/>yanrong@cs.cmu.edu +<br/>alex+@cs.cmu.edu +</td></tr><tr><td>290136947fd44879d914085ee51d8a4f433765fa</td><td>On a Taxonomy of Facial Features +</td><td>('1817623', 'Brendan Klare', 'brendan klare')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>2957715e96a18dbb5ed5c36b92050ec375214aa6</td><td>Improving Face Attribute Detection with Race and Gender Diversity +<br/>InclusiveFaceNet: +</td><td>('3766392', 'Hee Jung Ryu', 'hee jung ryu')</td><td></td></tr><tr><td>291f527598c589fb0519f890f1beb2749082ddfd</td><td>Seeing People in Social Context: Recognizing +<br/>People and Social Relationships +<br/><b>University of Illinois at Urbana-Champaign, Urbana, IL</b><br/><b>Kodak Research Laboratories, Rochester, NY</b></td><td>('22804340', 'Gang Wang', 'gang wang')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')</td><td></td></tr><tr><td>291265db88023e92bb8c8e6390438e5da148e8f5</td><td>MS-Celeb-1M: A Dataset and Benchmark for +<br/>Large-Scale Face Recognition +<br/>Microsoft Research +</td><td>('3133575', 'Yandong Guo', 'yandong guo')<br/>('1684635', 'Lei Zhang', 'lei zhang')<br/>('1689532', 'Yuxiao Hu', 'yuxiao hu')<br/>('1722627', 'Xiaodong He', 'xiaodong he')<br/>('1800422', 'Jianfeng Gao', 'jianfeng gao')</td><td>{yandong.guo,leizhang,yuxiao.hu,xiaohe,jfgao}@microsoft.com +</td></tr><tr><td>29c340c83b3bbef9c43b0c50b4d571d5ed037cbd</td><td>Stacked Dense U-Nets with Dual +<br/>Transformers for Robust Face Alignment +<br/>https://github.com/deepinsight/insightface +<br/>https://jiankangdeng.github.io/ +<br/>https://ibug.doc.ic.ac.uk/people/nxue +<br/>Stefanos Zafeiriou2 +<br/>https://wp.doc.ic.ac.uk/szafeiri/ +<br/>1 InsightFace +<br/>Shanghai, China +<br/>2 IBUG +<br/><b>Imperial College London</b><br/>London, UK +</td><td>('3007274', 'Jia Guo', 'jia guo')<br/>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('3007274', 'Jia Guo', 'jia guo')<br/>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('4091869', 'Niannan Xue', 'niannan xue')</td><td></td></tr><tr><td>297d3df0cf84d24f7efea44f87c090c7d9be4bed</td><td>Appearance-based 3-D Face Recognition from +<br/>Video +<br/><b>University of Maryland, Center for Automation Research</b><br/>A.V. Williams Building +<br/><b>College Park, MD</b><br/><b>The Robotics Institute, Carnegie Mellon University</b><br/>5000 Forbes Avenue, Pittsburgh, PA 15213 +</td><td>('33731953', 'Ralph Gross', 'ralph gross')<br/>('40039594', 'Simon Baker', 'simon baker')</td><td></td></tr><tr><td>29b86534d4b334b670914038c801987e18eb5532</td><td>Total Cluster: A person agnostic clustering method for +<br/>broadcast videos +<br/><b>Computer Vision for Human Computer Interaction, Karlsruhe Institute of Technology, Germany</b><br/><b>Visual Geometry Group, University of Oxford, UK</b><br/><b>Center for Machine Vision Research, University of Oulu, Finland</b></td><td>('2103464', 'Makarand Tapaswi', 'makarand tapaswi')<br/>('3188342', 'Omkar M. Parkhi', 'omkar m. parkhi')<br/>('2827962', 'Esa Rahtu', 'esa rahtu')<br/>('1741116', 'Eric Sommerlade', 'eric sommerlade')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>tapaswi@kit.edu, omkar@robots.ox.ac.uk, erahtu@ee.oulu.fi +<br/>eric@robots.ox.ac.uk, rainer.stiefelhagen@kit.edu, az@robots.ox.ac.uk +</td></tr><tr><td>29631ca6cff21c9199c70bcdbbcd5f812d331a96</td><td>RESEARCH ARTICLE +<br/>Error Rates in Users of Automatic Face +<br/>Recognition Software +<br/><b>School of Psychology, The University of New South Wales, Sydney, Australia, 2 School of Psychology</b><br/><b>The University of Sydney, Sydney, Australia</b></td><td>('40404556', 'David White', 'david white')<br/>('29329747', 'James D. Dunn', 'james d. dunn')<br/>('5016966', 'Alexandra C. Schmid', 'alexandra c. schmid')<br/>('3086646', 'Richard I. Kemp', 'richard i. kemp')</td><td>* david.white@unsw.edu.au +</td></tr><tr><td>2965d092ed72822432c547830fa557794ae7e27b</td><td>Improving Representation and Classification of Image and +<br/>Video Data for Surveillance Applications +<br/>BSc(Biol), MSc(Biol), MSc(CompSc) +<br/>A thesis submitted for the degree of Doctor of Philosophy at +<br/><b>The University of Queensland in</b><br/>School of Information Technology and Electrical Engineering +</td><td>('2706642', 'Andres Sanin', 'andres sanin')</td><td></td></tr><tr><td>2983efadb1f2980ab5ef20175f488f77b6f059d7</td><td>ch04_88815.QXP 12/23/08 3:36 PM Page 53 +<br/>◆ 4 ◆ +<br/>EMOTION IN HUMAN–COMPUTER INTERACTION +<br/><b>Stanford University</b><br/>Understanding Emotion . . . . . . . . . . . . . . . . . . . . . . . . . . 54 +<br/>Distinguishing Emotion from Related Constructs . . . . 55 +<br/>Mood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 +<br/>Sentiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 +<br/>Effects of Affect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 +<br/>Attention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 +<br/>Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 +<br/>Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 +<br/>Assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 +<br/>Causes of Emotion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 +<br/>Needs and Goals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 +<br/>Appraisal Theories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 +<br/>Contagion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 +<br/>Moods and Sentiments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 +<br/>Previous Emotional State . . . . . . . . . . . . . . . . . . . . . . . . . . 59 +<br/>Causes of Mood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 +<br/>Contagion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 +<br/>Color . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 +<br/>Other Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 +<br/>Measuring Affect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 +<br/>Neurological Responses . . . . . . . . . . . . . . . . . . . . . . . . . . 61 +<br/>Autonomic Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 +<br/>Facial Expression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 +<br/>Voice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 +<br/>Self-Report Measures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 +<br/>Affect Recognition by Users . . . . . . . . . . . . . . . . . . . . . . . 63 +<br/>Open Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 +<br/>1. With which emotion should HCI designers +<br/>be most concerned? . . . . . . . . . . . . . . . . . . . . . . . . . 64 +<br/>2. When and how should interfaces attempt to +<br/>directly address users’ emotions and basic +<br/>needs (vs. application-specific goals)? . . . . . . . . . . . . 64 +<br/>3. How accurate must emotion recognition be +<br/>to be useful as an interface technique? . . . . . . . . . . . 64 +<br/>4. When and how should users be informed +<br/>that their affective states are being monitored +<br/>and adapted to? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 +<br/>5. How does emotion play out in computer- +<br/>mediated communication (CMC)? . . . . . . . . . . . . . . 64 +<br/>Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 +<br/>Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64 +<br/>References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 +<br/>53 +</td><td>('2739604', 'Scott Brave', 'scott brave')<br/>('2029850', 'Clifford Nass', 'clifford nass')</td><td></td></tr><tr><td>2911e7f0fb6803851b0eddf8067a6fc06e8eadd6</td><td>Joint Fine-Tuning in Deep Neural Networks +<br/>for Facial Expression Recognition +<br/>School of Electrical Engineering +<br/><b>Korea Advanced Institute of Science and Technology</b></td><td>('1800903', 'Heechul Jung', 'heechul jung')<br/>('3249661', 'Junho Yim', 'junho yim')</td><td>{heechul, haeng, junho.yim, sunny0414, junmo.kim}@kaist.ac.kr +</td></tr><tr><td>2921719b57544cfe5d0a1614d5ae81710ba804fa</td><td>Face Recognition Enhancement Based on Image +<br/>File Formats and Wavelet De-noising +<br/> +</td><td>('4050987', 'Jieqing Tan', 'jieqing tan')<br/>('40160496', 'Zhengfeng Hou', 'zhengfeng hou')</td><td></td></tr><tr><td>29a013b2faace976f2c532533bd6ab4178ccd348</td><td>This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. +<br/>Hierarchical Manifold Learning With Applications +<br/>to Supervised Classification for High-Resolution +<br/>Remotely Sensed Images +</td><td>('7192623', 'Hong-Bing Huang', 'hong-bing huang')<br/>('3239427', 'Hong Huo', 'hong huo')<br/>('1680725', 'Tao Fang', 'tao fang')</td><td></td></tr><tr><td>29921072d8628544114f68bdf84deaf20a8c8f91</td><td>Multi-Task Curriculum Transfer Deep Learning of Clothing Attributes +<br/><b>School of EECS, Queen Mary University of London, UK</b></td><td>('40204089', 'Qi Dong', 'qi dong')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('2171228', 'Xiatian Zhu', 'xiatian zhu')</td><td>{q.dong, s.gong, xiatian.zhu}@qmul.ac.uk +</td></tr><tr><td>2969f822b118637af29d8a3a0811ede2751897b5</td><td>Cascaded Shape Space Pruning for Robust Facial Landmark Detection +<br/>Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing 100190, China</b></td><td>('1874505', 'Xiaowei Zhao', 'xiaowei zhao')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1695600', 'Xiujuan Chai', 'xiujuan chai')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{xiaowei.zhao,shiguang.shan,xiujuan.chai,xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>29756b6b16d7b06ea211f21cdaeacad94533e8b4</td><td>Thresholding Approach based on GPU for Facial +<br/>Expression Recognition +<br/>1 Benemérita Universidad Autónoma de Puebla, Faculty of Computer Science, Puebla, México +<br/>2Instituto Tecnológico de Puebla, Puebla, México +</td><td>('4348305', 'Jesús García-Ramírez', 'jesús garcía-ramírez')<br/>('3430302', 'Adolfo Aguilar-Rico', 'adolfo aguilar-rico')</td><td>gr_jesus@outlook.com,{aolvera,iolmos}@cs.buap.mx +<br/>{kremhilda,adolforico2}@gmail.com +</td></tr><tr><td>293193d24d5c4d2975e836034bbb2329b71c4fe7</td><td>Building a Corpus of Facial Expressions +<br/>for Learning-Centered Emotions +<br/>Instituto Tecnológico de Culiacán, Culiacán, Sinaloa, +<br/>Mexico +</td><td>('1744658', 'María Lucía Barrón-Estrada', 'maría lucía barrón-estrada')<br/>('38814197', 'Bianca Giovanna Aispuro-Medina', 'bianca giovanna aispuro-medina')<br/>('38906263', 'Elvia Minerva Valencia-Rodríguez', 'elvia minerva valencia-rodríguez')<br/>('38797488', 'Ana Cecilia Lara-Barrera', 'ana cecilia lara-barrera')</td><td>{lbarron, rzatarain, m06170904, m95170906, m15171452} @itculiacan.edu.mx +</td></tr><tr><td>294bd7eb5dc24052237669cdd7b4675144e22306</td><td>International Journal of Science and Research (IJSR) +<br/>ISSN (Online): 2319-7064 +<br/>Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438 +<br/>Automatic Face Annotation +<br/> +<br/><b>M.Tech Student, Mount Zion College of Engineering, Pathanamthitta, Kerala, India</b></td><td></td><td></td></tr><tr><td>2988f24908e912259d7a34c84b0edaf7ea50e2b3</td><td>A Model of Brightness Variations Due to +<br/>Illumination Changes and Non-rigid Motion +<br/>Using Spherical Harmonics +<br/>Jos´e M. Buenaposada +<br/>Dep. Ciencias de la Computaci´on, +<br/>U. Rey Juan Carlos, Spain +<br/>http://www.dia.fi.upm.es/~pcr +<br/>Inst. for Systems and Robotics +<br/>Inst. Superior T´ecnico, Portugal +<br/>http://www.isr.ist.utl.pt/~adb +<br/>Enrique Mu˜noz +<br/>Facultad de Inform´atica, +<br/>U. Complutense de Madrid, Spain +<br/>Dep. de Inteligencia Artificial, +<br/>U. Polit´ecnica de Madrid, Spain +<br/>http://www.dia.fi.upm.es/~pcr +<br/>http://www.dia.fi.upm.es/~pcr +</td><td>('1714730', 'Alessio Del Bue', 'alessio del bue')<br/>('1778998', 'Luis Baumela', 'luis baumela')</td><td></td></tr><tr><td>29156e4fe317b61cdcc87b0226e6f09e416909e0</td><td></td><td></td><td></td></tr><tr><td>29f0414c5d566716a229ab4c5794eaf9304d78b6</td><td>Hindawi Publishing Corporation +<br/>EURASIP Journal on Advances in Signal Processing +<br/>Volume 2008, Article ID 579416, 17 pages +<br/>doi:10.1155/2008/579416 +<br/>Review Article +<br/>Biometric Template Security +<br/><b>Michigan State University, 3115 Engineering Building</b><br/>East Lansing, MI 48824, USA +<br/>Received 2 July 2007; Revised 28 September 2007; Accepted 4 December 2007 +<br/>Recommended by Arun Ross +<br/>Biometric recognition offers a reliable solution to the problem of user authentication in identity management systems. With the +<br/>widespread deployment of biometric systems in various applications, there are increasing concerns about the security and privacy +<br/>of biometric technology. Public acceptance of biometrics technology will depend on the ability of system designers to demonstrate +<br/>that these systems are robust, have low error rates, and are tamper proof. We present a high-level categorization of the various +<br/>vulnerabilities of a biometric system and discuss countermeasures that have been proposed to address these vulnerabilities. In par- +<br/>ticular, we focus on biometric template security which is an important issue because, unlike passwords and tokens, compromised +<br/>biometric templates cannot be revoked and reissued. Protecting the template is a challenging task due to intrauser variability in the +<br/>acquired biometric traits. We present an overview of various biometric template protection schemes and discuss their advantages +<br/>and limitations in terms of security, revocability, and impact on matching accuracy. A template protection scheme with provable +<br/>security and acceptable recognition performance has thus far remained elusive. Development of such a scheme is crucial as bio- +<br/>metric systems are beginning to proliferate into the core physical and information infrastructure of our society. +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>1. +<br/>INTRODUCTION +<br/>A reliable identity management system is urgently needed in +<br/>order to combat the epidemic growth in identity theft and to +<br/>meet the increased security requirements in a variety of ap- +<br/>plications ranging from international border crossings to se- +<br/>curing information in databases. Establishing the identity of +<br/>a person is a critical task in any identity management system. +<br/>Surrogate representations of identity such as passwords and +<br/>ID cards are not sufficient for reliable identity determination +<br/>because they can be easily misplaced, shared, or stolen. Bio- +<br/>metric recognition is the science of establishing the identity +<br/>of a person using his/her anatomical and behavioral traits. +<br/>Commonly used biometric traits include fingerprint, face, +<br/>iris, hand geometry, voice, palmprint, handwritten signa- +<br/>tures, and gait (see Figure 1). Biometric traits have a number +<br/>of desirable properties with respect to their use as an authen- +<br/>tication token, namely, reliability, convenience, universality, +<br/>and so forth. These characteristics have led to the widespread +<br/>deployment of biometric authentication systems. But there +<br/>are still some issues concerning the security of biometric +<br/>recognition systems that need to be addressed in order to en- +<br/>sure the integrity and public acceptance of these systems. +<br/>There are five major components in a generic biomet- +<br/>ric authentication system, namely, sensor, feature extrac- +<br/>tor, template database, matcher, and decision module (see +<br/>Figure 2). Sensor is the interface between the user and the +<br/>authentication system and its function is to scan the bio- +<br/>metric trait of the user. Feature extraction module processes +<br/>the scanned biometric data to extract the salient information +<br/>(feature set) that is useful in distinguishing between differ- +<br/>ent users. In some cases, the feature extractor is preceded +<br/>by a quality assessment module which determines whether +<br/>the scanned biometric trait is of sufficient quality for fur- +<br/>ther processing. During enrollment, the extracted feature +<br/>set is stored in a database as a template (XT) indexed by +<br/>the user’s identity information. Since the template database +<br/>could be geographically distributed and contain millions of +<br/>records (e.g., in a national identification system), maintain- +<br/>ing its security is not a trivial task. The matcher module is +<br/>usually an executable program, which accepts two biomet- +<br/>ric feature sets XT and XQ (from template and query, resp.) +<br/>as inputs, and outputs a match score (S) indicating the sim- +<br/>ilarity between the two sets. Finally, the decision module +<br/>makes the identity decision and initiates a response to the +<br/>query. +</td><td>('6680444', 'Anil K. Jain', 'anil k. jain')<br/>('34633765', 'Karthik Nandakumar', 'karthik nandakumar')<br/>('2743820', 'Abhishek Nagar', 'abhishek nagar')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>Correspondence should be addressed to Karthik Nandakumar, nandakum@cse.msu.edu +</td></tr><tr><td>293ade202109c7f23637589a637bdaed06dc37c9</td><td></td><td></td><td></td></tr><tr><td>7c61d21446679776f7bdc7afd13aedc96f9acac1</td><td>Hierarchical Label Inference for Video Classification +<br/><b>Simon Fraser University</b><br/><b>Simon Fraser University</b><br/><b>Simon Fraser University</b></td><td>('3079079', 'Nelson Nauata', 'nelson nauata')<br/>('2847110', 'Jonathan Smith', 'jonathan smith')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td>nnauata@sfu.ca +<br/>jws4@sfu.ca +<br/>mori@cs.sfu.ca +</td></tr><tr><td>7cee802e083c5e1731ee50e731f23c9b12da7d36</td><td>2B3C: 2 Box 3 Crop of Facial Image for Gender Classification with Convolutional +<br/>Networks +<br/>Department of Electronics and Communication Engineering and +<br/><b>Computer Vision Group, L. D. College of Engineering, Ahmedabad, India</b></td><td>('23922616', 'Vandit Gajjar', 'vandit gajjar')</td><td> gajjar.vandit.381@ldce.ac.in +</td></tr><tr><td>7c47da191f935811f269f9ba3c59556c48282e80</td><td>Robust Eye Centers Localization +<br/>with Zero–Crossing Encoded Image Projections +<br/>Image Processing and Analysis Laboratory +<br/><b>University Politehnica of Bucharest, Romania, Address Splaiul Independent ei</b><br/>Image Processing and Analysis Laboratory +<br/><b>University Politehnica of Bucharest, Romania, Address Splaiul Independent ei</b><br/>Image Processing and Analysis Laboratory +<br/><b>University Politehnica of Bucharest, Romania, Address Splaiul Independent ei</b></td><td>('2143956', 'Laura Florea', 'laura florea')<br/>('2760434', 'Corneliu Florea', 'corneliu florea')<br/>('2905899', 'Constantin Vertan', 'constantin vertan')</td><td>laura.florea@upb.ro +<br/>corneliu.florea@upb.ro +<br/>constantin.vertan@upb.ro +</td></tr><tr><td>7c7ab59a82b766929defd7146fd039b89d67e984</td><td>Improving Multiview Face Detection with +<br/>Multi-Task Deep Convolutional Neural Networks +<br/>Microsoft Research +<br/>One Microsoft Way, Redmond WA 98052 +</td><td>('1706673', 'Cha Zhang', 'cha zhang')<br/>('1809184', 'Zhengyou Zhang', 'zhengyou zhang')</td><td></td></tr><tr><td>7ca337735ec4c99284e7c98f8d61fb901dbc9015</td><td>Proceedings of the 8th International +<br/>IEEE Conference on Intelligent Transportation Systems +<br/>Vienna, Austria, September 13-16, 2005 +<br/>TC4.2 +<br/>Driver Activity Monitoring through Supervised and Unsupervised Learning +<br/>Harini Veeraraghavan Stefan Atev Nathaniel Bird Paul Schrater Nikolaos Papanikolopoulos† +<br/>Department of Computer Science and Engineering +<br/><b>University of Minnesota</b></td><td></td><td>{harini,atev,bird,schrater,npapas}@cs.umn.edu +</td></tr><tr><td>7c1cfab6b60466c13f07fe028e5085a949ec8b30</td><td>Deep Feature Consistent Variational Autoencoder +<br/><b>University of Nottingham, Ningbo China</b><br/><b>Shenzhen University, Shenzhen China</b><br/><b>University of Nottingham, Ningbo China</b><br/><b>University of Nottingham, Ningbo China</b></td><td>('3468964', 'Xianxu Hou', 'xianxu hou')<br/>('1687690', 'Linlin Shen', 'linlin shen')<br/>('39508183', 'Ke Sun', 'ke sun')<br/>('1698461', 'Guoping Qiu', 'guoping qiu')</td><td>xianxu.hou@nottingham.edu.cn +<br/>llshen@szu.edu.cn +<br/>ke.sun@nottingham.edu.cn +<br/>guoping.qiu@nottingham.edu.cn +</td></tr><tr><td>7c45b5824645ba6d96beec17ca8ecfb22dfcdd7f</td><td>News image annotation on a large parallel text-image corpus +<br/>Universit´e de Rennes 1/IRISA, CNRS/IRISA, INRIA Rennes-Bretagne Atlantique +<br/>Campus de Beaulieu +<br/>35042 Rennes Cedex, France +</td><td>('1694537', 'Pierre Tirilly', 'pierre tirilly')<br/>('1735666', 'Vincent Claveau', 'vincent claveau')<br/>('2436627', 'Patrick Gros', 'patrick gros')</td><td>ptirilly@irisa.fr, vclaveau@irisa.fr, pgros@inria.fr +</td></tr><tr><td>7c17280c9193da3e347416226b8713b99e7825b8</td><td>VideoCapsuleNet: A Simplified Network for Action +<br/>Detection +<br/>Kevin Duarte +<br/>Yogesh S Rawat +<br/>Center for Research in Computer Vision +<br/><b>University of Central Florida</b><br/>Orlando, FL 32816 +</td><td>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>kevin_duarte@knights.ucf.edu +<br/>yogesh@crcv.ucf.edu +<br/>shah@crcv.ucf.edu +</td></tr><tr><td>7cffcb4f24343a924a8317d560202ba9ed26cd0b</td><td>The Unconstrained Ear Recognition Challenge +<br/><b>University of Ljubljana</b><br/>Ljubljana, Slovenia +<br/>IIT Kharagpur +<br/>Kharagpur, India +<br/><b>University of Colorado Colorado Springs</b><br/>Colorado Springs, CO, USA +<br/><b>Islamic Azad University</b><br/>Qazvin, Iran +<br/><b>Imperial College London</b><br/>London, UK +<br/>ITU Department of Computer Engineering +<br/>Istanbul, Turkey +</td><td>('34862665', 'Peter Peer', 'peter peer')<br/>('3110004', 'Anjith George', 'anjith george')<br/>('2173052', 'Adil Ahmad', 'adil ahmad')<br/>('39000630', 'Elshibani Omar', 'elshibani omar')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')<br/>('3062107', 'Reza Safdari', 'reza safdari')<br/>('47943220', 'Yuxiang Zhou', 'yuxiang zhou')<br/>('23981209', 'Dogucan Yaman', 'dogucan yaman')</td><td>ziga.emersic@fri.uni-lj.si +</td></tr><tr><td>7c0a6824b556696ad7bdc6623d742687655852db</td><td>18th Telecommunications forum TELFOR 2010 +<br/>Serbia, Belgrade, November 23-25, 2010. +<br/>MPCA+DATER: A Novel Approach for Face +<br/>Recognition Based on Tensor Objects +<br/>Ali. A. Shams Baboli, Member, IEEE, G. Rezai-rad, Member, IEEE, Aref. Shams Baboli +</td><td></td><td></td></tr><tr><td>7c95449a5712aac7e8c9a66d131f83a038bb7caa</td><td>This is an author produced version of Facial first impressions from another angle: How +<br/>social judgements are influenced by changeable and invariant facial properties. +<br/>White Rose Research Online URL for this paper: +<br/>http://eprints.whiterose.ac.uk/102935/ +<br/>Article: +<br/>Rhodes (2017) Facial first impressions from another angle: How social judgements are +<br/>influenced by changeable and invariant facial properties. British journal of psychology. pp. +<br/>397-415. ISSN 0007-1269 +<br/>https://doi.org/10.1111/bjop.12206 +<br/>promoting access to +<br/>White Rose research papers +<br/>http://eprints.whiterose.ac.uk/ +</td><td>('16854522', 'Clare', 'clare')<br/>('9384336', 'Young', 'young')</td><td>eprints@whiterose.ac.uk +</td></tr><tr><td>7c4c442e9c04c6b98cd2aa221e9d7be15efd8663</td><td>Classifier Learning with Hidden Information +<br/><b>ECSE, Rensselaer Polytechnic Institute, Troy, NY</b></td><td>('2860279', 'Ziheng Wang', 'ziheng wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>wangz10@rpi.edu +<br/>jiq@rpi.edu +</td></tr><tr><td>7c3e09e0bd992d3f4670ffacb4ec3a911141c51f</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Transferring Object-Scene Convolutional Neural Networks for +<br/>Event Recognition in Still Images +<br/>Received: date / Accepted: date +</td><td>('33345248', 'Limin Wang', 'limin wang')</td><td></td></tr><tr><td>7c2ec6f4ab3eae86e0c1b4f586e9c158fb1d719d</td><td>Dissimilarity-Based Classifications in Eigenspaces(cid:63) +<br/><b>Myongji University, Yongin, 449-728 South</b><br/><b>Faculty of Electrical Engineering, Mathematics and Computer Science, Delft University of</b></td><td>('34959719', 'Sang-Woon Kim', 'sang-woon kim')<br/>('1747298', 'Robert P. W. Duin', 'robert p. w. duin')</td><td>Korea. e-mail : kimsw@mju.ac.kr +<br/>Technology, The Netherlands. e-mail : r.p.w.duin@tudelft.nl +</td></tr><tr><td>7cf8a841aad5b7bdbea46a7bb820790e9ce12d0b</td><td>SUPERVISED HEAT KERNEL LPP +<br/>METHOD FOR FACE RECOGNITION +<br/><b>Utah State University, Logan UT</b></td><td>('1725739', 'Xiaojun Qi', 'xiaojun qi')</td><td>cryshan@cc.usu.edu and xqi@cc.usu.edu +</td></tr><tr><td>7c9622ad1d8971cd74cc9e838753911fe27ccac4</td><td>Representation Learning with Smooth +<br/>Autoencoder +<br/>Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b></td><td>('2582309', 'Kongming Liang', 'kongming liang')<br/>('1783542', 'Hong Chang', 'hong chang')<br/>('10338111', 'Zhen Cui', 'zhen cui')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{kongming.liang, hong.chang, zhen.cui, shiguang.shan, xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>7c2c9b083817f7a779d819afee383599d2e97ed8</td><td>Disentangling Motion, Foreground and Background Features in Videos +<br/><b>Beihang University</b><br/>Beijing, China +<br/>V´ıctor Campos +<br/>Xavier Giro-i-Nieto +<br/>Barcelona Supercomputing Center +<br/>Universitat Politecnica de Catalunya +<br/>Barcelona, Catalonia/Spain +<br/>Barcelona, Catalonia/Spain +<br/>Barcelona Supercomputing Center +<br/>Barcelona, Catalonia/Spain +<br/>Cristian Canton Ferrer +<br/>Facebook +<br/>Seattle (WA), USA +</td><td>('10668384', 'Xunyu Lin', 'xunyu lin')<br/>('1711068', 'Jordi Torres', 'jordi torres')</td><td>xunyulin2017@outlook.com +<br/>victor.campos@bsc.es +<br/>xavier.giro@upc.edu +<br/>jordi.torres@bsc.es +<br/>ccanton@fb.com +</td></tr><tr><td>7c45339253841b6f0efb28c75f2c898c79dfd038</td><td>Unsupervised Joint Alignment of Complex Images +<br/><b>University of Massachusetts Amherst</b><br/>Amherst, MA +<br/>Erik Learned-Miller +</td><td>('3219900', 'Gary B. Huang', 'gary b. huang')<br/>('2246870', 'Vidit Jain', 'vidit jain')</td><td>fgbhuang,vidit,elmg@cs.umass.edu +</td></tr><tr><td>7c825562b3ff4683ed049a372cb6807abb09af2a</td><td>Finding Tiny Faces +<br/>Supplementary Materials +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>1. Error analysis +<br/>Quantitative analysis We plot the distribution of error modes among false positives in Fig. 1 and the impact of object +<br/>characteristics on detection performance in Fig. 2 and Fig. 3. +<br/>Qualitative analysis We show top 20 scoring false positives in Fig. 4. +<br/>2. Experimental details +<br/>Multi-scale features Inspired by the way [3] trains “FCN-8s at-once”, we scale the learning rate of predictor built on +<br/>top of each layer by a fixed constant. Specifically, we use a scaling factor of 1 for res4, 0.1 for res3, and 0.01 for res2. +<br/>One more difference between our model and [3] is that: instead of predicting at original resolution, our model predicts +<br/>at the resolution of res3 feature (downsampled by 8X comparing to input resolution). +<br/>Input sampling We first randomly re-scale the input image by 0.5X, 1X, or 2X. Then we randomly crop a 500x500 +<br/>image region out of the re-scaled input. We pad with average RGB value (prior to average subtraction) when cropping +<br/>outside image boundary. +<br/>Border cases Similar to [2], we ignore gradients coming from heatmap locations whose detection windows cross the +<br/>image boundary. The only difference is, we treat padded average pixels (as described in Input sampling) as outside +<br/>image boundary as well. +<br/>Online hard mining and balanced sampling We apply hard mining on both positive and negative examples. Our +<br/>implementation is simpler yet still effective comparing to [4]. We set a small threshold (0.03) on classification loss +<br/>to filter out easy locations. Then we sample at most 128 locations for both positive and negative (respectively) from +<br/>remaining ones whose losses are above the threshold. We compare training with and without hard mining on validation +<br/>performance in Table 1. +<br/>Loss function Our loss function is formulated in the same way as [2]. Note that we also use Huber loss as the loss +<br/>function for bounding box regression. +<br/>Bounding box regression Our bounding box regression is formulated as [2] and trained jointly with classification +<br/>using stochastic gradient descent. We compare between testing with and without regression in terms of performance +<br/>on WIDER FACE validation set. +</td><td>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td>{peiyunh,deva}@cs.cmu.edu +</td></tr><tr><td>7c7b0550ec41e97fcfc635feffe2e53624471c59</td><td>1051-4651/14 $31.00 © 2014 IEEE +<br/>DOI 10.1109/ICPR.2014.124 +<br/>660 +</td><td></td><td></td></tr><tr><td>7ce03597b703a3b6754d1adac5fbc98536994e8f</td><td></td><td></td><td></td></tr><tr><td>7c36afc9828379de97f226e131390af719dbc18d</td><td>Unsupervised Face-Name Association +<br/>via Commute Distance +<br/>1Zhejiang Provincial Key Laboratory of Service Robot +<br/><b>College of Computer Science, Zhejiang University, Hangzhou, China</b><br/><b>State Key Lab of CADandCG, College of Computer Science, Zhejiang University, Hangzhou, China</b></td><td>('4140420', 'Jiajun Bu', 'jiajun bu')<br/>('40155478', 'Bin Xu', 'bin xu')<br/>('2484982', 'Chenxia Wu', 'chenxia wu')<br/>('2588203', 'Chun Chen', 'chun chen')<br/>('1704030', 'Jianke Zhu', 'jianke zhu')<br/>('1724421', 'Deng Cai', 'deng cai')<br/>('3945955', 'Xiaofei He', 'xiaofei he')</td><td>{bjj,xbzju,chenxiawu,chenc,jkzhu}@zju.edu.cn +<br/>{dengcai,xiaofeihe}@cad.zju.edu.cn +</td></tr><tr><td>7c119e6bdada2882baca232da76c35ae9b5277f8</td><td>Facial Expression Recognition Using Embedded +<br/>Hidden Markov Model +<br/><b>Intelligence Computing Research Center</b><br/>HIT Shenzhen Graduate School +<br/>Shenzhen, China +</td><td>('24233679', 'Languang He', 'languang he')<br/>('1747105', 'Xuan Wang', 'xuan wang')<br/>('10106946', 'Chenglong Yu', 'chenglong yu')<br/>('38700402', 'Kun Wu', 'kun wu')</td><td>{telent, wangxuan, ycl, wukun} @cs.hitsz.edu.cn +</td></tr><tr><td>7ca7255c2e0c86e4adddbbff2ce74f36b1dc522d</td><td>Stereo Matching for Unconstrained Face Recognition +<br/>Ph.D. Proposal +<br/><b>University of Maryland</b><br/>Department of Computer Science +<br/><b>College Park, MD</b><br/>May 10, 2009 +</td><td>('38171682', 'Carlos D. Castillo', 'carlos d. castillo')</td><td>carlos@cs.umd.edu +</td></tr><tr><td>7c42371bae54050dbbf7ded1e7a9b4109a23a482</td><td>The International Arab Journal of Information Technology, Vol. 12, No. 2, March 2015 183 +<br/>Optimized Features Selection using Hybrid PSO- +<br/>GA for Multi-View Gender Classification +<br/><b>Foundation University Rawalpindi Campus, Pakistan</b><br/><b>University of Central Punjab, Pakistan</b><br/><b>University of Dammam, Saudi Arabia</b><br/>4Department of Computer Science, SZABIST, Pakistan +</td><td>('1723986', 'Muhammad Nazir', 'muhammad nazir')<br/>('11616523', 'Muhammad Khan', 'muhammad khan')</td><td></td></tr><tr><td>7c953868cd51f596300c8231192d57c9c514ae17</td><td>Detecting and Aligning Faces by Image Retrieval +<br/>Zhe Lin2 +<br/><b>Northwestern University</b><br/>2Adobe Research +<br/>2145 Sheridan Road, Evanston, IL 60208 +<br/>345 Park Ave, San Jose, CA 95110 +</td><td>('1720987', 'Xiaohui Shen', 'xiaohui shen')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')<br/>('1736695', 'Ying Wu', 'ying wu')</td><td>{xsh835, yingwu}@eecs.northwestern.edu +<br/>{zlin, jbrandt}@adobe.com +</td></tr><tr><td>7c6dbaebfe14878f3aee400d1378d90d61373921</td><td>A Novel Biometric Feature Extraction Algorithm using Two +<br/>Dimensional Fisherface in 2DPCA subspace for Face Recognition +<br/>School of Electrical, Electronic and Computer Engineering +<br/><b>University of Newcastle</b><br/>Newcastle upon Tyne, NE1 7RU +<br/>UNITED KINDOM +</td><td>('3156162', 'R. M. MUTELO', 'r. m. mutelo')</td><td></td></tr><tr><td>7c9a65f18f7feb473e993077d087d4806578214e</td><td>SpringerLink - Zeitschriftenbeitrag +<br/>http://www.springerlink.com/content/93hr862660nl1164/?p=abe5352... +<br/>Deutsch +<br/>Deutsch +<br/>Go +<br/>Vorherige Beitrag Nächste Beitrag +<br/>Beitrag markieren +<br/>In den Warenkorb legen +<br/>Zu gespeicherten Artikeln +<br/>hinzufügen +<br/>Permissions & Reprints +<br/>Diesen Artikel empfehlen +<br/>Ergebnisse +<br/>finden +<br/>Erweiterte Suche +<br/>Go +<br/>im gesamten Inhalt +<br/>in dieser Zeitschrift +<br/>in diesem Heft +<br/>Diesen Beitrag exportieren +<br/>Diesen Beitrag exportieren als RIS +<br/>| Text +<br/>Text +<br/>PDF +<br/>PDF ist das gebräuchliche Format +<br/>für Online Publikationen. Die Größe +<br/>dieses Dokumentes beträgt 564 +<br/>Kilobyte. Je nach Art Ihrer +<br/>Internetverbindung kann der +<br/>Download einige Zeit in Anspruch +<br/>nehmen. +<br/>öffnen: Gesamtdokument +<br/>Publikationsart Subject Collections +<br/>Zurück zu: Journal Issue +<br/>Athens Authentication Point +<br/>Zeitschriftenbeitrag +<br/>Willkommen! +<br/>Um unsere personalisierten +<br/>Angebote nutzen zu können, +<br/>müssen Sie angemeldet sein. +<br/>Login +<br/>Jetzt registrieren +<br/>Zugangsdaten vergessen? +<br/>Hilfe. +<br/>Mein Menü +<br/>Markierte Beiträge +<br/>Alerts +<br/>Meine Bestellungen +<br/>Private emotions versus social interaction: a data-driven approach towards +<br/>analysing emotion in speech +<br/>Zeitschrift +<br/>Verlag +<br/>ISSN +<br/>Heft +<br/>Kategorie +<br/>DOI +<br/>Seiten +<br/>Subject Collection +<br/>SpringerLink Date +<br/>User Modeling and User-Adapted Interaction +<br/>Springer Netherlands +<br/>0924-1868 (Print) 1573-1391 (Online) +<br/>Volume 18, Numbers 1-2 / Februar 2008 +<br/>Original Paper +<br/>10.1007/s11257-007-9039-4 +<br/>175-206 +<br/>Informatik +<br/>Freitag, 12. Oktober 2007 +<br/>Gespeicherte Beiträge +<br/>Alle +<br/>Favoriten +<br/>(1) Lehrstuhl für Mustererkennung, FAU Erlangen – Nürnberg, Martensstr. 3, 91058 Erlangen, +<br/>Germany +<br/>Received: 3 July 2006 Accepted: 14 January 2007 Published online: 12 October 2007 +</td><td>('1745089', 'Anton Batliner', 'anton batliner')<br/>('1732747', 'Stefan Steidl', 'stefan steidl')<br/>('2596771', 'Christian Hacker', 'christian hacker')<br/>('1739326', 'Elmar Nöth', 'elmar nöth')</td><td></td></tr><tr><td>7c1e1c767f7911a390d49bed4f73952df8445936</td><td>NON-RIGID OBJECT DETECTION WITH LOCAL INTERLEAVED SEQUENTIAL ALIGNMENT (LISA) +<br/>Non-Rigid Object Detection with Local +<br/>Interleaved Sequential Alignment (LISA) +<br/>and Tom´aˇs Svoboda, Member, IEEE +</td><td>('35274952', 'Karel Zimmermann', 'karel zimmermann')<br/>('2687885', 'David Hurych', 'david hurych')</td><td></td></tr><tr><td>7cf579088e0456d04b531da385002825ca6314e2</td><td>Emotion Detection on TV Show Transcripts with +<br/>Sequence-based Convolutional Neural Networks +<br/>Mathematics and Computer Science +<br/>Mathematics and Computer Science +<br/><b>Emory University</b><br/>Atlanta, GA 30322, USA +<br/><b>Emory University</b><br/>Atlanta, GA 30322, USA +</td><td>('10669356', 'Sayyed M. Zahiri', 'sayyed m. zahiri')<br/>('4724587', 'Jinho D. Choi', 'jinho d. choi')</td><td>sayyed.zahiri@emory.edu +<br/>jinho.choi@emory.edu +</td></tr><tr><td>7c80d91db5977649487388588c0c823080c9f4b4</td><td>DocFace: Matching ID Document Photos to Selfies∗ +<br/><b>Michigan State University</b><br/>East Lansing, Michigan, USA +</td><td>('9644181', 'Yichun Shi', 'yichun shi')<br/>('1739705', 'Anil K. Jain', 'anil k. jain')</td><td>shiyichu@msu.edu, jain@cse.msu.edu +</td></tr><tr><td>7c349932a3d083466da58ab1674129600b12b81c</td><td></td><td></td><td></td></tr><tr><td>7c30ea47f5ae1c5abd6981d409740544ed16ed16</td><td>ROITBERG, AL-HALAH, STIEFELHAGEN: NOVELTY DETECTION FOR ACTION RECOGNITION +<br/>Informed Democracy: Voting-based Novelty +<br/>Detection for Action Recognition +<br/><b>Karlsruhe Institute of Technology</b><br/>76131 Karlsruhe, +<br/>Germany +</td><td>('33390229', 'Alina Roitberg', 'alina roitberg')<br/>('2256981', 'Ziad Al-Halah', 'ziad al-halah')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>alina.roitberg@kit.edu +<br/>ziad.al-halah@kit.edu +<br/>rainer.stiefelhagen@kit.edu +</td></tr><tr><td>1648cf24c042122af2f429641ba9599a2187d605</td><td>Boosting Cross-Age Face Verification via Generative Age Normalization +<br/>(cid:2) Orange Labs, 4 rue Clos Courtel, 35512 Cesson-S´evign´e, France +<br/>† Eurecom, 450 route des Chappes, 06410 Biot, France +</td><td>('3116433', 'Grigory Antipov', 'grigory antipov')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')<br/>('2341854', 'Moez Baccouche', 'moez baccouche')</td><td>{grigory.antipov,moez.baccouche}@orange.com +<br/>jean-luc.dugelay@eurecom.fr +</td></tr><tr><td>162403e189d1b8463952fa4f18a291241275c354</td><td>Action Recognition with Spatio-Temporal +<br/>Visual Attention on Skeleton Image Sequences +<br/>With a strong ability of modeling sequential data, Recur- +<br/>rent Neural Networks (RNN) with Long Short-Term Memory +<br/>(LSTM) neurons outperform the previous hand-crafted feature +<br/>based methods [9], [10]. Each skeleton frame is converted into +<br/>a feature vector and the whole sequence is fed into the RNN. +<br/>Despite the strong ability in modeling temporal sequences, +<br/>RNN structures lack the ability to efficiently learn the spatial +<br/>relations between the joints. To better use spatial information, +<br/>a hierarchical structure is proposed in [11], [12] that feeds +<br/>the joints into the network as several pre-defined body part +<br/>groups. However, +<br/>limit +<br/>the effectiveness of representing spatial relations. A spatio- +<br/>temporal 2D LSTM (ST-LSTM) network [13] is proposed +<br/>to learn the spatial and temporal relations simultaneously. +<br/>Furthermore, a two-stream RNN structure [14] is proposed to +<br/>learn the spatio-temporal relations with two RNN branches. +<br/>the pre-defined body regions still +</td><td>('21518096', 'Zhengyuan Yang', 'zhengyuan yang')<br/>('3092578', 'Yuncheng Li', 'yuncheng li')<br/>('1706007', 'Jianchao Yang', 'jianchao yang')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')</td><td></td></tr><tr><td>160259f98a6ec4ec3e3557de5e6ac5fa7f2e7f2b</td><td>Discriminant Multi-Label Manifold Embedding for Facial Action Unit +<br/>Detection +<br/>Signal Procesing Laboratory (LTS5), ´Ecole Polytechnique F´ed´erale de Lausanne, Switzerland +</td><td>('1697965', 'Hua Gao', 'hua gao')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')</td><td>anil.yuce@epfl.ch, hua.gao@epfl.ch, jean-philippe.thiran@epfl.ch +</td></tr><tr><td>16671b2dc89367ce4ed2a9c241246a0cec9ec10e</td><td>2006 +<br/>Detecting the Number of Clusters +<br/>in n-Way Probabilistic Clustering +</td><td>('1788526', 'Zhaoshui He', 'zhaoshui he')<br/>('1747156', 'Andrzej Cichocki', 'andrzej cichocki')<br/>('1795838', 'Shengli Xie', 'shengli xie')<br/>('1775180', 'Kyuwan Choi', 'kyuwan choi')</td><td></td></tr><tr><td>16fdd6d842475e6fbe58fc809beabbed95f0642e</td><td>Learning Temporal Embeddings for Complex Video Analysis +<br/><b>Stanford University, 2Simon Fraser University</b></td><td>('34066479', 'Vignesh Ramanathan', 'vignesh ramanathan')<br/>('10771328', 'Greg Mori', 'greg mori')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td>{vigneshr, kdtang}@cs.stanford.edu, mori@cs.sfu.ca, feifeili@cs.stanford.edu +</td></tr><tr><td>16bce9f940bb01aa5ec961892cc021d4664eb9e4</td><td>Mutual Component Analysis for Heterogeneous Face Recognition +<br/>39 +<br/>Heterogeneous face recognition, also known as cross-modality face recognition or inter-modality face recogni- +<br/>tion, refers to matching two face images from alternative image modalities. Since face images from different +<br/>image modalities of the same person are associated with the same face object, there should be mutual com- +<br/>ponents that reflect those intrinsic face characteristics that are invariant to the image modalities. Motivated +<br/>by this rationality, we propose a novel approach called mutual component analysis (MCA) to infer the mu- +<br/>tual components for robust heterogeneous face recognition. In the MCA approach, a generative model is first +<br/>proposed to model the process of generating face images in different modalities, and then an Expectation +<br/>Maximization (EM) algorithm is designed to iteratively learn the model parameters. The learned generative +<br/>model is able to infer the mutual components (which we call the hidden factor, where hidden means the +<br/>factor is unreachable and invisible, and can only be inferred from observations) that are associated with +<br/>the person’s identity, thus enabling fast and effective matching for cross-modality face recognition. To en- +<br/>hance recognition performance, we propose an MCA-based multi-classifier framework using multiple local +<br/>features. Experimental results show that our new approach significantly outperforms the state-of-the-art +<br/>results on two typical application scenarios, sketch-to-photo and infrared-to-visible face recognition. +<br/>Categories and Subject Descriptors: I.5.1 [Pattern Recognition]: Models +<br/>General Terms: Design, Algorithms, Performance +<br/>Additional Key Words and Phrases: Face recognition, heterogeneous face recognition, mutual component +<br/>analysis (MCA) +<br/>ACM Reference Format: +<br/>Heterogeneous Face Recognition ACM Trans. Intell. Syst. Technol. 9, 4, Article 39 (July 2015), 22 pages. +<br/>DOI: http://dx.doi.org/10.1145/2807705 +<br/>This work was supported by grants from National Natural Science Foundation of China (61103164 and +<br/>61125106), Natural Science Foundation of Guangdong Province (2014A030313688), Australian Research +<br/>Council Projects (FT-130101457 and LP-140100569), Key Laboratory of Human-Machine Intelligence- +<br/><b>Synergy Systems, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Guangdong</b><br/>Innovative Research Team Program (No.201001D0104648280), the Key Research Program of the Chinese +<br/><b>Academy of Sciences (Grant No. KGZD-EW-T03), and project MMT-8115038 of the Shun Hing Institute of</b><br/><b>Advanced Engineering, The Chinese University of Hong Kong</b><br/><b>Author s addresses: Z. Li and D. Gong, Shenzhen Institutes of Advanced Technology, Chinese Academy</b><br/><b>tum Computation and Intelligent Systems, Faculty of Engineering and Information Technology, University</b><br/><b>Key Laboratory of Transient Optics and Photonics, Xi an Institute of Optics and Precision Mechanics, Chi</b><br/>Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted +<br/>without fee provided that copies are not made or distributed for profit or commercial advantage and that +<br/>copies bear this notice and the full citation on the first page. Copyrights for components of this work owned +</td><td>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('2856494', 'Dihong Gong', 'dihong gong')<br/>('20638185', 'Qiang Li', 'qiang li')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('1720243', 'Xuelong Li', 'xuelong li')<br/>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('20638185', 'Qiang Li', 'qiang li')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('1720243', 'Xuelong Li', 'xuelong li')</td><td>of Sciences, P. R. China; e-mail: {zhifeng.li, dh.gong}@siat.ac.cn; Q. Li and D. Tao, Centre for Quan- +<br/>of Technology Sydney, 81 Broadway, Ultimo, NSW 2007, Australia; e-mail: qiang.li-2@student.uts.edu.au, +<br/>dacheng.tao@uts.edu.au; X. Li, the Center for OPTical IMagery Analysis and Learning (OPTIMAL), State +<br/>nese Academy of Sciences, Xi’an 710119, Shaanxi, China; e-mail: xuelong li@opt.ac.cn. +</td></tr><tr><td>16de1324459fe8fdcdca80bba04c3c30bb789bdf</td><td></td><td></td><td></td></tr><tr><td>16892074764386b74b6040fe8d6946b67a246a0b</td><td></td><td></td><td></td></tr><tr><td>16395b40e19cbc6d5b82543039ffff2a06363845</td><td>Action Recognition in Video Using Sparse Coding and Relative Features +<br/>Anal´ı Alfaro +<br/>P. Universidad Catolica de Chile +<br/>P. Universidad Catolica de Chile +<br/>P. Universidad Catolica de Chile +<br/>Santiago, Chile +<br/>Santiago, Chile +<br/>Santiago, Chile +</td><td>('1797475', 'Domingo Mery', 'domingo mery')<br/>('7263603', 'Alvaro Soto', 'alvaro soto')</td><td>ajalfaro@uc.cl +<br/>dmery@ing.puc.cl +<br/>asoto@ing.uc.cl +</td></tr><tr><td>1677d29a108a1c0f27a6a630e74856e7bddcb70d</td><td>Efficient Misalignment-Robust Representation +<br/>for Real-Time Face Recognition +<br/><b>The Hong Kong Polytechnic University, Hong Kong</b></td><td>('5828998', 'Meng Yang', 'meng yang')<br/>('36685537', 'Lei Zhang', 'lei zhang')<br/>('1698371', 'David Zhang', 'david zhang')</td><td>{csmyang,cslzhang}@comp.polyu.edu.hk +</td></tr><tr><td>16b9d258547f1eccdb32111c9f45e2e4bbee79af</td><td>2006 Xiyuan Ave. +<br/>Chengdu, Sichuan 611731 +<br/>2006 Xiyuan Ave. +<br/>Chengdu, Sichuan 611731 +<br/><b>University of Electronic Science and Technology of China</b><br/><b>Johns Hopkins University</b><br/>3400 N. Charles St. +<br/>Baltimore, Maryland 21218 +<br/><b>Johns Hopkins University</b><br/>3400 N. Charles St. +<br/>Baltimore, Maryland 21218 +<br/>NormFace: L2 Hypersphere Embedding for Face Verification +<br/><b>University of Electronic Science and Technology of China</b></td><td>('1709439', 'Jian Cheng', 'jian cheng')<br/>('40031188', 'Xiang Xiang', 'xiang xiang')<br/>('1746141', 'Alan L. Yuille', 'alan l. yuille')<br/>('39369840', 'Feng Wang', 'feng wang')</td><td>feng.w(cid:29)@gmail.com +<br/>chengjian@uestc.edu.cn +<br/>xxiang@cs.jhu.edu +<br/>alan.yuille@jhu.edu +</td></tr><tr><td>16c884be18016cc07aec0ef7e914622a1a9fb59d</td><td>UNIVERSITÉ DE GRENOBLE +<br/>No attribué par la bibliothèque +<br/>THÈSE +<br/>pour obtenir le grade de +<br/>DOCTEUR DE L’UNIVERSITÉ DE GRENOBLE +<br/>Spécialité : Mathématiques et Informatique +<br/>préparée au Laboratoire Jean Kuntzmann +<br/>dans le cadre de l’École Doctorale Mathématiques, +<br/>Sciences et Technologies de l’Information, Informatique +<br/>présentée et soutenue publiquement +<br/>par +<br/>le 27 septembre 2010 +<br/>Exploiting Multimodal Data for Image Understanding +<br/>Données multimodales pour l’analyse d’image +<br/>Directeurs de thèse : Cordelia Schmid et Jakob Verbeek +<br/>JURY +<br/>M. Éric Gaussier +<br/>M. Antonio Torralba +<br/><b>Mme Tinne Tuytelaars Katholieke Universiteit Leuven</b><br/><b>M. Mark Everingham University of Leeds</b><br/>Mme Cordelia Schmid +<br/>M. Jakob Verbeek +<br/>Président +<br/>Université Joseph Fourier +<br/><b>Massachusetts Institute of Technology Rapporteur</b><br/>Rapporteur +<br/>Examinateur +<br/>Examinatrice +<br/>Examinateur +<br/>INRIA Grenoble +<br/>INRIA Grenoble +</td><td>('2737253', 'Matthieu Guillaumin', 'matthieu guillaumin')</td><td></td></tr><tr><td>162dfd0d2c9f3621d600e8a3790745395ab25ebc</td><td>Head Pose Estimation Based on Multivariate Label Distribution +<br/>School of Computer Science and Engineering +<br/><b>Southeast University, Nanjing, China</b></td><td>('1735299', 'Xin Geng', 'xin geng')<br/>('40228279', 'Yu Xia', 'yu xia')</td><td>{xgeng, xiayu}@seu.edu.cn +</td></tr><tr><td>16f940b4b5da79072d64a77692a876627092d39c</td><td>A Framework for Automated Measurement of the Intensity of Non-Posed Facial +<br/>Action Units +<br/><b>University of Denver, Denver, CO</b><br/><b>University of Miami, Coral Gables, FL</b><br/><b>University of Miami, Coral Gables, FL</b><br/><b>University of Pittsburgh, Pittsburgh, PA</b><br/>Emails: +</td><td>('3093835', 'Mohammad H. Mahoor', 'mohammad h. mahoor')<br/>('2897823', 'Steven Cadavid', 'steven cadavid')<br/>('1874236', 'Daniel S. Messinger', 'daniel s. messinger')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>mmahoor@du.edu, scadavid@umsis.miami.edu, dmessinger@miami.edu, and jeffcohn@pitt.edu +</td></tr><tr><td>16572c545384174f8136d761d2b0866e968120a8</td><td>Sequential Max-Margin Event Detectors +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, 15213. USA</b></td><td>('39792229', 'Dong Huang', 'dong huang')<br/>('2583890', 'Shitong Yao', 'shitong yao')<br/>('1734275', 'Yi Wang', 'yi wang')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td></td></tr><tr><td>16820ccfb626dcdc893cc7735784aed9f63cbb70</td><td>Real-time Embedded Age and Gender Classification in Unconstrained Video +<br/>School of Electrical Engineering and Computer Science +<br/><b>University of Ottawa</b><br/>Ottawa, ON K1N 6N5 Canada +<br/>CogniVue Corporation +<br/>Gatineau, QC, Canada +</td><td>('2014654', 'Ramin Azarmehr', 'ramin azarmehr')<br/>('1807494', 'Won-Sook Lee', 'won-sook lee')<br/>('2551825', 'Christina Xu', 'christina xu')<br/>('32944169', 'Daniel Laroche', 'daniel laroche')</td><td>{razar033,laganier,wslee}@uottawa.ca +<br/>{cxu,dlaroche}@cognivue.com +</td></tr><tr><td>1630e839bc23811e340bdadad3c55b6723db361d</td><td>SONG, TAN, CHEN: EXPLOITING RELATIONSHIP BETWEEN ATTRIBUTES +<br/>Exploiting Relationship between Attributes for +<br/>Improved Face Verification +<br/>Department of Computer Science and +<br/><b>Technology, Nanjing University of Aero</b><br/>nautics and Astronautics, Nanjing 210016, +<br/>P.R. China +</td><td>('3075941', 'Fengyi Song', 'fengyi song')<br/>('2248421', 'Xiaoyang Tan', 'xiaoyang tan')<br/>('1680768', 'Songcan Chen', 'songcan chen')</td><td>f.song@nuaa.edu.cn +<br/>x.tan@nuaa.edu.cn +<br/>s.chen@nuaa.edu.cn +</td></tr><tr><td>164b0e2a03a5a402f66c497e6c327edf20f8827b</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Sparse Deep Transfer Learning for +<br/>Convolutional Neural Network +<br/><b>The Chinese University of Hong Kong, Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, China</b></td><td>('2335888', 'Jiaming Liu', 'jiaming liu')<br/>('47903936', 'Yali Wang', 'yali wang')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td>jiaming.liu@email.ucr.edu, {yl.wang, yu.qiao}@siat.ac.cn +</td></tr><tr><td>16286fb0f14f6a7a1acc10fcd28b3ac43f12f3eb</td><td>J Nonverbal Behav +<br/>DOI 10.1007/s10919-008-0059-5 +<br/>O R I G I N A L P A P E R +<br/>All Smiles are Not Created Equal: Morphology +<br/>and Timing of Smiles Perceived as Amused, Polite, +<br/>and Embarrassed/Nervous +<br/>Ó Springer Science+Business Media, LLC 2008 +</td><td>('2059653', 'Zara Ambadar', 'zara ambadar')</td><td></td></tr><tr><td>1667a77db764e03a87a3fd167d88b060ef47bb56</td><td>Alternative Semantic Representations for +<br/>Zero-Shot Human Action Recognition +<br/><b>School of Computer Science, The University of Manchester</b><br/>Manchester, M13 9PL, UK +</td><td>('1729612', 'Qian Wang', 'qian wang')<br/>('32811782', 'Ke Chen', 'ke chen')</td><td>{qian.wang,ke.chen}@manchester.ac.uk +</td></tr><tr><td>169618b8dc9b348694a31c6e9e17b989735b4d39</td><td>Unsupervised Representation Learning by Sorting Sequences +<br/><b>University of California, Merced</b><br/>Maneesh Singh3 +<br/>2Virginia Tech +<br/>3Verisk Analytics +<br/>http://vllab1.ucmerced.edu/˜hylee/OPN/ +</td><td>('2837591', 'Hsin-Ying Lee', 'hsin-ying lee')<br/>('3068086', 'Jia-Bin Huang', 'jia-bin huang')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>16e95a907b016951da7c9327927bb039534151da</td><td>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 32, XXXX-XXXX (2016) +<br/>3D Face Recognition Using Spherical Vector Norms Map * +<br/>a Beijing Key Laboratory of Information Service Engineering, +<br/><b>Beijing Union University, 100101, China</b><br/><b>b Computer Technology Institute, Beijing Union University, 100101, China</b><br/>c Beijing Advanced Innovation Center for Imaging Technology, +<br/><b>Capital Normal University, 100048, China</b><br/>In this paper, we introduce a novel, automatic method for 3D face recognition. A +<br/>new feature called a spherical vector norms map of a 3D face is created using the normal +<br/>vector of each point. This feature contains more detailed information than the original +<br/>depth image in regions such as the eyes and nose. For certain flat areas of 3D face, such +<br/>as the forehead and cheeks, this map could increase the distinguishability of different +<br/>points. In addition, this feature is robust to facial expression due to an adjustment that is +<br/>made in the mouth region. Then, the facial representations, which are based on Histo- +<br/>grams of Oriented Gradients, are extracted from the spherical vector norms map and the +<br/>original depth image. A new partitioning strategy is proposed to produce the histogram +<br/>of eight patches of a given image, in which all of the pixels are binned based on the +<br/>magnitude and direction of their gradients. In this study, SVNs map and depth image are +<br/>represented compactly with two histograms of oriented gradients; this approach is com- +<br/>pleted by Linear Discriminant Analysis and a Nearest Neighbor classifier. +<br/>Keywords: spherical vector norms map, Histograms of Oriented Gradients, 3D face +<br/>recognition, Linear Discriminant Analysis, Face Recognition Grand Challenge database +<br/>1. INTRODUCTION +<br/>With the rapidly decreasing costs of 3D capturing devices, many researchers are in- +<br/>vestigating 3D face recognition systems because it could overcome limitations illumina- +<br/>tion and make-up, but still bear limitations mostly due to facial expression. We summa- +<br/>rize a smaller subset of expressive-robust methods below: +<br/>1. Deformable template-based approaches: Berretti et al. [1] proposed an approach +<br/>that describes the geometric information of a 3D facial using a surface graph form, and +<br/>the relevant information among the neighboring points could be encoded into a compact +<br/>representation. 3DWWs (3D Weighted Walkthroughs) descriptors were proposed to +<br/>demonstrate the mutual spatial displacement among pairwise arcs of points of the corre- +<br/>sponding stripes. An 81.2% verification rate at a 0.1% FAR was achieved on the all vs. +<br/>all experiment. The advantage of the method is the computational complexity is low. +<br/>Kakadiaris et al. [2] mapped 3D geometry information onto a 2D regular grid using +<br/>an elastically adapted deformable model. Then, advanced wavelet analysis was used for +<br/>recognition and get good performance. +<br/>Drira et al. [3] used radial curves emanating from the nose tips which were already +<br/>provided, and used elastic shape analysis of these curves to develop a Riemannian +<br/>framework. Finally, they analyze the shapes of full facial surfaces. +<br/>1249 +</td><td>('3282147', 'Xue-Qiao Wang', 'xue-qiao wang')<br/>('2130097', 'Jia-Zheng Yuan', 'jia-zheng yuan')<br/>('1930238', 'Qing Li', 'qing li')</td><td>E-mail: {ldxueqiao; jiazheng; liqing10}@buu.edu.cn +</td></tr><tr><td>166186e551b75c9b5adcc9218f0727b73f5de899</td><td>Volume 4, Issue 2, February 2016 +<br/>International Journal of Advance Research in +<br/>Computer Science and Management Studies +<br/>Research Article / Survey Paper / Case Study +<br/>Available online at: www.ijarcsms.com +<br/>ISSN: 2321-7782 (Online) +<br/>Automatic Age and Gender Recognition in Human Face Image +<br/>Dataset using Convolutional Neural Network System +<br/>Subhani Shaik1 +<br/>Assoc. Prof & Head of the Department +<br/>Department of CSE, +<br/>Associate Professor +<br/>Department of CSE, +<br/>St.Mary’s Group of Institutions Guntur +<br/>St.Mary’s Group of Institutions Guntur +<br/>Chebrolu(V&M),Guntur(Dt), +<br/>Andhra Pradesh - India +<br/>Chebrolu(V&M),Guntur(Dt), +<br/>Andhra Pradesh - India +</td><td>('39885231', 'Anto A. Micheal', 'anto a. micheal')</td><td></td></tr><tr><td>16d6737b50f969247339a6860da2109a8664198a</td><td>Convolutional Neural Networks +<br/>for Age and Gender Classification +<br/><b>Stanford University</b></td><td>('22241470', 'Ari Ekmekji', 'ari ekmekji')</td><td>aekmekji@stanford.edu +</td></tr><tr><td>16d9b983796ffcd151bdb8e75fc7eb2e31230809</td><td>EUROGRAPHICS 2018 / D. Gutierrez and A. Sheffer +<br/>(Guest Editors) +<br/>Volume 37 (2018), Number 2 +<br/>GazeDirector: Fully Articulated Eye Gaze Redirection in Video +<br/>ID: paper1004 +</td><td></td><td></td></tr><tr><td>1679943d22d60639b4670eba86665371295f52c3</td><td></td><td></td><td></td></tr><tr><td>162c33a2ec8ece0dc96e42d5a86dc3fedcf8cd5e</td><td>Mygdalis, V., Iosifidis, A., Tefas, A., & Pitas, I. (2016). Large-Scale +<br/>Classification by an Approximate Least Squares One-Class Support Vector +<br/>of a meeting held 20-22 August 2015, Helsinki, Finland (Vol. 2, pp. 6-10). +<br/><b>Institute of Electrical and Electronics Engineers (IEEE). DOI</b><br/>10.1109/Trustcom.2015.555 +<br/>Peer reviewed version +<br/>Link to published version (if available): +<br/>10.1109/Trustcom.2015.555 +<br/>Link to publication record in Explore Bristol Research +<br/>PDF-document +<br/><b>University of Bristol - Explore Bristol Research</b><br/>General rights +<br/>This document is made available in accordance with publisher policies. Please cite only the published +<br/>version using the reference above. Full terms of use are available: +<br/>http://www.bristol.ac.uk/pure/about/ebr-terms +<br/> </td><td></td><td></td></tr><tr><td>1610d2d4947c03a89c0fda506a74ba1ae2bc54c2</td><td>Robust Real-Time 3D Face Tracking from RGBD Videos under Extreme Pose, +<br/>Depth, and Expression Variations +<br/>Hai X. Pham +<br/><b>Rutgers University, USA</b></td><td>('1736042', 'Vladimir Pavlovic', 'vladimir pavlovic')</td><td>{hxp1,vladimir}@cs.rutgers.edu +</td></tr><tr><td>1659a8b91c3f428f1ba6aeba69660f2c9d0a85c6</td><td>Recent Developments in Social Signal Processing +<br/><b>Institute of Informatics - ISLA</b><br/><b>University of Amsterdam, Amsterdam, The Netherlands</b><br/>†Department of Computing +<br/><b>Imperial College London, London, UK</b><br/><b>EEMCS, University of Twente Enschede, The Netherlands</b><br/><b>University of Glasgow</b><br/>Glasgow, Scotland +</td><td>('1764521', 'Albert Ali Salah', 'albert ali salah')<br/>('1694605', 'Maja Pantic', 'maja pantic')<br/>('1719436', 'Alessandro Vinciarelli', 'alessandro vinciarelli')</td><td>Email: a.a.salah@uva.nl +<br/>Email: m.pantic@imperial.ac.uk +<br/>Email: vincia@dcs.gla.ac.uk +</td></tr><tr><td>169076ffe5e7a2310e98087ef7da25aceb12b62d</td><td></td><td></td><td></td></tr><tr><td>167736556bea7fd57cfabc692ec4ae40c445f144</td><td>METHODS +<br/>published: 13 January 2016 +<br/>doi: 10.3389/fict.2015.00028 +<br/>Improved Motion Description for +<br/>Action Classification +<br/>Inria, Centre Rennes – Bretagne Atlantique, Rennes, France +<br/>Even though the importance of explicitly integrating motion characteristics in video +<br/>descriptions has been demonstrated by several recent papers on action classification, our +<br/>current work concludes that adequately decomposing visual motion into dominant and +<br/>residual motions, i.e., camera and scene motion, significantly improves action recognition +<br/>algorithms. This holds true both for the extraction of the space-time trajectories and for +<br/>computation of descriptors. We designed a new motion descriptor – the DCS descriptor – +<br/>that captures additional information on local motion patterns enhancing results based on +<br/>differential motion scalar quantities, divergence, curl, and shear features. Finally, applying +<br/>the recent VLAD coding technique proposed in image retrieval provides a substantial +<br/>improvement for action recognition. These findings are complementary to each other +<br/>and they outperformed all previously reported results by a significant margin on three +<br/>challenging datasets: Hollywood 2, HMDB51, and Olympic Sports as reported in Jain +<br/>et al. (2013). These results were further improved by Oneata et al. (2013), Wang and +<br/>Schmid (2013), and Zhu et al. (2013) through the use of the Fisher vector encoding. We +<br/>therefore also employ Fisher vector in this paper, and we further enhance our approach by +<br/>combining trajectories from both optical flow and compensated flow. We as well provide +<br/><b>additional details of DCS descriptors, including visualization. For extending the evaluation</b><br/>a novel dataset with 101 action classes, UCF101, was added. +<br/>Keywords: action classification, camera motion, optical flow, motion trajectories, motion descriptors +<br/>1. INTRODUCTION +<br/>The recognition of human actions in unconstrained videos remains a challenging problem in +<br/>computer vision despite the fact that human actions are often attributed to essential meaningful +<br/>content in such videos. The field receives sustained attention due to its potential applications, +<br/>such as for designing video-surveillance systems, in providing automatic annotation of video +<br/>archives, as well as for improving human–computer interaction. The solutions that were proposed +<br/>to address the above problems were inherited from the techniques first designed for image search +<br/>and classification. +<br/>Successful local features were developed to describe image patches (Schmid and Mohr, 1997; +<br/>Lowe, 2004) and translated in the 2D + t domain as spatio-temporal local descriptors (Laptev et al., +<br/>2008; Wang et al., 2009) and now include motion clues of Wang et al. (2011). These descriptors +<br/>are often extracted from spatial–temporal interest points (Laptev and Lindeberg, 2003; Willems +<br/>et al., 2008). Furthermore, several approaches assume underlying temporal motion model involving +<br/>trajectories (Hervieu et al., 2008; Matikainen et al., 2009; Messing et al., 2009; Sun et al., 2009; +<br/>Brox and Malik, 2010; Wang et al., 2011; Wu et al., 2011; Gaidon et al., 2012; Wang and Schmid, +<br/>2013). +<br/>Edited by: +<br/>Jean-Marc Odobez, +<br/><b>Idiap Research Institute, Switzerland</b><br/>Reviewed by: +<br/>Thanh Duc Ngo, +<br/><b>Ho Chi Minh City University of</b><br/>Information Technology, Vietnam +<br/>Jean Martinet, +<br/><b>Lille 1 University, France</b><br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Computer Image Analysis, a section +<br/>of the journal Frontiers in ICT +<br/>Received: 16 April 2015 +<br/>Accepted: 22 December 2015 +<br/>Published: 13 January 2016 +<br/>Citation: +<br/>Jain M, Jégou H and Bouthemy P +<br/>(2016) Improved Motion Description +<br/>for Action Classification. +<br/>doi: 10.3389/fict.2015.00028 +<br/>Frontiers in ICT | www.frontiersin.org +<br/>January 2016 | Volume 2 | Article 28 +</td><td>('40027484', 'Mihir Jain', 'mihir jain')<br/>('1681054', 'Hervé Jégou', 'hervé jégou')<br/>('1716733', 'Patrick Bouthemy', 'patrick bouthemy')<br/>('40027484', 'Mihir Jain', 'mihir jain')</td><td>m.jain@uva.nl +</td></tr><tr><td>167ea1631476e8f9332cef98cf470cb3d4847bc6</td><td>Visual Search at Pinterest +<br/>1Visual Discovery, Pinterest +<br/><b>University of California, Berkeley</b></td><td>('39554931', 'Yushi Jing', 'yushi jing')<br/>('1911082', 'Dmitry Kislyuk', 'dmitry kislyuk')<br/>('39835325', 'Andrew Zhai', 'andrew zhai')<br/>('2560579', 'Jiajing Xu', 'jiajing xu')<br/>('7408951', 'Jeff Donahue', 'jeff donahue')<br/>('2608161', 'Sarah Tavel', 'sarah tavel')</td><td>{jing, dliu, dkislyuk, andrew, jiajing, jdonahue, sarah}@pinterest.com +</td></tr><tr><td>161eb88031f382e6a1d630cd9a1b9c4bc6b47652</td><td>1 +<br/>Automatic Facial Expression Recognition +<br/>Using Features of Salient Facial Patches +</td><td>('2680543', 'Aurobinda Routray', 'aurobinda routray')</td><td></td></tr><tr><td>420782499f38c1d114aabde7b8a8104c9e40a974</td><td>Joint Ranking and Classification using Weak Data for Feature Extraction +<br/>Fashion Style in 128 Floats: +<br/>Department of Computer Science and Engineering +<br/><b>Waseda University, Tokyo, Japan</b></td><td>('3114470', 'Edgar Simo-Serra', 'edgar simo-serra')<br/>('1692113', 'Hiroshi Ishikawa', 'hiroshi ishikawa')</td><td>esimo@aoni.waseda.jp +<br/>hfs@waseda.jp +</td></tr><tr><td>4209783b0cab1f22341f0600eed4512155b1dee6</td><td>Accurate and Efficient Similarity Search for Large Scale Face Recognition +<br/>BUPT +<br/>BUPT +<br/>BUPT +</td><td>('49712251', 'Ce Qi', 'ce qi')<br/>('35963823', 'Zhizhong Liu', 'zhizhong liu')<br/>('1684263', 'Fei Su', 'fei su')</td><td></td></tr><tr><td>42e3dac0df30d754c7c7dab9e1bb94990034a90d</td><td>PANDA: Pose Aligned Networks for Deep Attribute Modeling +<br/>2EECS, UC Berkeley +<br/>1Facebook AI Research +</td><td>('40565777', 'Ning Zhang', 'ning zhang')<br/>('2210374', 'Manohar Paluri', 'manohar paluri')<br/>('1753210', 'Trevor Darrell', 'trevor darrell')</td><td>{mano, ranzato, lubomir}@fb.com +<br/>{nzhang, trevor}@eecs.berkeley.edu +</td></tr><tr><td>4217473596b978f13a211cdf47b7d3f6588c785f</td><td>An Efficient Approach for Clustering Face Images +<br/><b>Michigan State University</b><br/>Noblis +<br/>Anil Jain +<br/>Michigan State Universtiy +</td><td>('40653304', 'Charles Otto', 'charles otto')<br/>('1817623', 'Brendan Klare', 'brendan klare')</td><td>ottochar@msu.edu +<br/>Brendan.Klare@noblis.org +<br/>jain@msu.edu +</td></tr><tr><td>4223666d1b0b1a60c74b14c2980069905088edc6</td><td>A Convergent Incoherent Dictionary Learning +<br/>Algorithm for Sparse Coding +<br/>Department of Mathematics +<br/><b>National University of Singapore</b></td><td>('3183763', 'Chenglong Bao', 'chenglong bao')<br/>('2217653', 'Yuhui Quan', 'yuhui quan')<br/>('39689301', 'Hui Ji', 'hui ji')</td><td></td></tr><tr><td>42afe6d016e52c99e2c0d876052ade9c192d91e7</td><td>Spontaneous vs. Posed Facial Behavior: +<br/>Automatic Analysis of Brow Actions +<br/><b>Imperial College London, UK</b><br/><b>Faculty of EEMCS, University of Twente, The Netherlands</b><br/><b>Psychology and Psychiatry, University of Pittsburgh, USA</b></td><td>('1795528', 'Michel F. Valstar', 'michel f. valstar')<br/>('1694605', 'Maja Pantic', 'maja pantic')<br/>('2059653', 'Zara Ambadar', 'zara ambadar')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>{michel.valstar,m.pantic}@imperial.ac.uk, {ambadar,jeffcohn}@pitt.edu, +</td></tr><tr><td>42765c170c14bd58e7200b09b2e1e17911eed42b</td><td>2 +<br/>Feature Extraction Based on Wavelet +<br/>Moments and Moment Invariants in +<br/>Machine Vision Systems +<br/>G.A. Papakostas, D.E. Koulouriotis and V.D. Tourassis +<br/><b>Democritus University of Thrace</b><br/>Department of Production Engineering and Management +<br/>Greece +<br/>1. Introduction +<br/>Recently, there has been an increasing interest on modern machine vision systems for +<br/>industrial and commercial purposes. More and more products are introduced in the market, +<br/>which are making use of visual information captured by a camera in order to perform a +<br/>specific task. Such machine vision systems are used for detecting and/or recognizing a face +<br/>in an unconstrained environment for security purposes, for analysing the emotional states of +<br/>a human by processing his facial expressions or for providing a vision based interface in the +<br/>context of the human computer interaction (HCI) etc.. +<br/>In almost all the modern machine vision systems there is a common processing procedure +<br/>called feature extraction, dealing with the appropriate representation of the visual information. +<br/>This task has two main objectives simultaneously, the compact description of the useful +<br/>information by a set of numbers (features), by keeping the dimension as low as possible. +<br/>Image moments constitute an important feature extraction method (FEM) which generates +<br/>high discriminative features, able to capture the particular characteristics of the described +<br/>pattern, which distinguish it among similar or totally different objects. Their ability to fully +<br/>describe an image by encoding its contents in a compact way makes them suitable for many +<br/>disciplines of the engineering life, such as image analysis (Sim et al., 2004), image +<br/>watermarking (Papakostas et al., 2010a) and pattern recognition (Papakostas et al., 2007, +<br/>2009a, 2010b). +<br/>Among the several moment families introduced in the past, the orthogonal moments are +<br/>the most popular moments widely used in many applications, owing to their +<br/>orthogonality property that comes from the nature of the polynomials used as kernel +<br/>functions, which they constitute an orthogonal base. As a result, the orthogonal moments +<br/>have minimum information redundancy meaning that different moment orders describe +<br/>different parts of the image. +<br/>In order to use the moments to classify visual objects, they have to ensure high recognition +<br/>rates for all possible object’s orientations. This requirement constitutes a significant +<br/>operational feature of each modern pattern recognition system and it can be satisfied during +<br/>www.intechopen.com +</td><td></td><td></td></tr><tr><td>429c3588ce54468090cc2cf56c9b328b549a86dc</td><td></td><td></td><td></td></tr><tr><td>42cc9ea3da1277b1f19dff3d8007c6cbc0bb9830</td><td>Coordinated Local Metric Learning +<br/>Inria∗ +</td><td>('2143851', 'Shreyas Saxena', 'shreyas saxena')<br/>('34602236', 'Jakob Verbeek', 'jakob verbeek')</td><td></td></tr><tr><td>42350e28d11e33641775bef4c7b41a2c3437e4fd</td><td>212 +<br/>Multilinear Discriminant Analysis +<br/>for Face Recognition +</td><td>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('38188040', 'Dong Xu', 'dong xu')<br/>('1706370', 'Qiang Yang', 'qiang yang')<br/>('39089563', 'Lei Zhang', 'lei zhang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>42e155ea109eae773dadf74d713485be83fca105</td><td></td><td></td><td></td></tr><tr><td>4223917177405eaa6bdedca061eb28f7b440ed8e</td><td>B-spline Shape from Motion & Shading: An Automatic Free-form Surface +<br/>Modeling for Face Reconstruction +<br/><b>School of Computer Science, Tianjin University</b><br/><b>School of Computer Science, Tianjin University</b><br/><b>School of Software, Tianjin University</b></td><td>('1919846', 'Weilong Peng', 'weilong peng')<br/>('1683334', 'Zhiyong Feng', 'zhiyong feng')<br/>('29962190', 'Chao Xu', 'chao xu')</td><td>wlpeng@tju.edu.cn +<br/>zyfeng@tju.edu.cn +</td></tr><tr><td>42eda7c20db9dc0f42f72bb997dd191ed8499b10</td><td>Gaze Embeddings for Zero-Shot Image Classification +<br/><b>Max Planck Institute for Informatics</b><br/>Saarland Informatics Campus +<br/>2Amsterdam Machine Learning Lab +<br/><b>University of Amsterdam</b></td><td>('7789181', 'Nour Karessli', 'nour karessli')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')</td><td></td></tr><tr><td>42c9394ca1caaa36f535721fa9a64b2c8d4e0dee</td><td>Label Efficient Learning of Transferable +<br/>Representations across Domains and Tasks +<br/><b>Stanford University</b><br/>Virginia Tech +<br/><b>University of California, Berkeley</b></td><td>('3378742', 'Zelun Luo', 'zelun luo')<br/>('8299168', 'Yuliang Zou', 'yuliang zou')<br/>('4742485', 'Judy Hoffman', 'judy hoffman')</td><td>zelunluo@stanford.edu +<br/>ylzou@vt.edu +<br/>jhoffman@eecs.berkeley.edu +</td></tr><tr><td>4270460b8bc5299bd6eaf821d5685c6442ea179a</td><td>Int J Comput Vis (2009) 84: 163–183 +<br/>DOI 10.1007/s11263-008-0147-3 +<br/>Partial Similarity of Objects, or How to Compare a Centaur +<br/>to a Horse +<br/>Received: 30 September 2007 / Accepted: 3 June 2008 / Published online: 26 July 2008 +<br/>© Springer Science+Business Media, LLC 2008 +</td><td>('1731883', 'Alexander M. Bronstein', 'alexander m. bronstein')<br/>('1692832', 'Ron Kimmel', 'ron kimmel')</td><td></td></tr><tr><td>4205cb47ba4d3c0f21840633bcd49349d1dc02c1</td><td>ACTION RECOGNITION WITH GRADIENT BOUNDARY CONVOLUTIONAL NETWORK +<br/><b>Research Institute of Shenzhen, Wuhan University, Shenzhen, China</b><br/><b>National Engineering Research Center for Multimedia Software, Wuhan University, Wuhan, China</b><br/><b>Center for Research in Computer Vision, University of Central Florida, Orlando, USA</b></td><td>('2559431', 'Huafeng Chen', 'huafeng chen')<br/>('1736897', 'Jun Chen', 'jun chen')<br/>('1732874', 'Chen Chen', 'chen chen')<br/>('37254976', 'Ruimin Hu', 'ruimin hu')</td><td></td></tr><tr><td>42ded74d4858bea1070dadb08b037115d9d15db5</td><td>Exigent: An Automatic Avatar Generation System +<br/>Computer Science and Artificial Intelligence Laboratory +<br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, Massachusetts 02139, USA +</td><td>('2852664', 'Dominic Kao', 'dominic kao')<br/>('1709421', 'D. Fox Harrell', 'd. fox harrell')</td><td>{dkao,fox.harrell}@mit.edu +</td></tr><tr><td>42ea8a96eea023361721f0ea34264d3d0fc49ebd</td><td>Parameterized Principal Component Analysis +<br/><b>Florida State University, USA</b></td><td>('2109527', 'Ajay Gupta', 'ajay gupta')<br/>('2455529', 'Adrian Barbu', 'adrian barbu')</td><td></td></tr><tr><td>42f6f5454dda99d8989f9814989efd50fe807ee8</td><td>Conditional generative adversarial nets for convolutional face generation +<br/>Symbolic Systems Program, Natural Language Processing Group +<br/><b>Stanford University</b></td><td>('24339276', 'Jon Gauthier', 'jon gauthier')</td><td>jgauthie@stanford.edu +</td></tr><tr><td>429d4848d03d2243cc6a1b03695406a6de1a7abd</td><td>Face Recognition based on Logarithmic Fusion +<br/>International Journal of Soft Computing and Engineering (IJSCE) +<br/>ISSN: 2231-2307, Volume-2, Issue-3, July 2012 +<br/>of SVD and KT +<br/>Ramachandra A C, Raja K B, Venugopal K R, L M Patnaik +<br/>to +<br/> +</td><td></td><td></td></tr><tr><td>42dc36550912bc40f7faa195c60ff6ffc04e7cd6</td><td>Hindawi Publishing Corporation +<br/>ISRN Machine Vision +<br/>Volume 2013, Article ID 579126, 10 pages +<br/>http://dx.doi.org/10.1155/2013/579126 +<br/>Research Article +<br/>Visible and Infrared Face Identification via +<br/>Sparse Representation +<br/><b>LITIS EA 4108-QuantIF Team, University of Rouen, 22 Boulevard Gambetta, 76183 Rouen Cedex, France</b><br/><b>GREYC UMR CNRS 6072 ENSICAEN-Image Team, University of Caen Basse-Normandie, 6 Boulevard Mar echal Juin</b><br/>14050 Caen, France +<br/>Received 4 April 2013; Accepted 27 April 2013 +<br/>Academic Editors: O. Ghita, D. Hernandez, Z. Hou, M. La Cascia, and J. M. Tavares +<br/>Copyright © 2013 P. Buyssens and M. Revenu. This is an open access article distributed under the Creative Commons Attribution +<br/>License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly +<br/>cited. +<br/>We present a facial recognition technique based on facial sparse representation. A dictionary is learned from data, and patches +<br/>extracted from a face are decomposed in a sparse manner onto this dictionary. We particularly focus on the design of dictionaries +<br/>that play a crucial role in the final identification rates. Applied to various databases and modalities, we show that this approach +<br/>gives interesting performances. We propose also a score fusion framework that allows quantifying the saliency classifiers outputs +<br/>and merging them according to these saliencies. +<br/>1. Introduction +<br/>Face recognition is a topic which has been of increasing inter- +<br/>est during the last two decades due to a vast number of pos- +<br/>sible applications: biometrics, video surveillance, advanced +<br/>HMI, or image/video indexation. Although considerable +<br/>progress has been made in this domain, especially with the +<br/>development of powerful methods (such as the Eigenfaces +<br/>or the Elastic Bunch Graph Matching methods), automatic +<br/>face recognition is not enough accurate in uncontrolled envi- +<br/>ronments for a large use. Many factors can degrade the per- +<br/>formances of facial biometric system: illumination variation +<br/>creates artificial shadows, changing locally the appearance of +<br/>the face; head poses modify the distance between localized +<br/>features; facial expression introduces global changes; artefacts +<br/>wearing, such as glasses or scarf, may hide parts of the face. +<br/>For the particular case of illumination, a lot of work has +<br/>been done on the preprocessing step of the images to reduce +<br/>the effect of the illumination on the face. Another approach is +<br/>to use other imagery such as infrared, which has been showed +<br/>to be a promising alternative. An infrared capture of a face is +<br/>nearly invariant to illumination changes and allows a system +<br/><b>to process in all the illumination conditions, including total</b><br/>darkness like night. +<br/>While visual cameras measure the electromagnetic +<br/>energy in the visible spectrum (0.4–0.7 𝜇m), sensors in the +<br/>IR respond to thermal radiation in the infrared spectrum +<br/>(0.7–14.0 𝜇m). The infrared spectrum can mainly be divided +<br/>into reflected IR (Figure 1(b)) and emissive IR (Figure 1(c)). +<br/>Reflected IR contains near infrared (NIR) (0.7–0.9 𝜇m) +<br/>and short-wave infrared (SWIR) (0.9–2.4 𝜇m). The ther- +<br/>mal IR band is associated with thermal radiation emitted +<br/>by the objects. It contains the midwave infrared (MWIR) +<br/>(3.0–5.0 𝜇m) and long-wave infrared (LWIR) (8.0–14.0 𝜇m). +<br/>Although the reflected IR is by far the most studied, we use +<br/>thermal long-wave IR in this study. +<br/>Despite the advantages of infrared modality, infrared im- +<br/>agery has other limitations. Since a face captured under this +<br/>modality renders its thermal patterns, a temperature screen +<br/>placed in front of the face will totally occlude it. This phe- +<br/>nomenon appears when a subject simply wears glasses. In this +<br/>case, the captured face has two black holes, corresponding to +<br/>the glasses, which is far more inconvenient than in the visible +</td><td>('2825139', 'Pierre Buyssens', 'pierre buyssens')</td><td>Correspondence should be addressed to Pierre Buyssens; pierre.buyssens@gmail.com +</td></tr><tr><td>424259e9e917c037208125ccc1a02f8276afb667</td><td></td><td></td><td></td></tr><tr><td>42ecfc3221c2e1377e6ff849afb705ecd056b6ff</td><td>Pose Invariant Face Recognition under Arbitrary +<br/>Unknown Lighting using Spherical Harmonics +<br/>Department of Computer Science, +<br/>SUNY at Stony Brook, NY, 11790 +</td><td>('38323599', 'Lei Zhang', 'lei zhang')<br/>('1686020', 'Dimitris Samaras', 'dimitris samaras')</td><td>{lzhang, samaras}@cs.sunysb.edu +</td></tr><tr><td>421955c6d2f7a5ffafaf154a329a525e21bbd6d3</td><td>570 +<br/>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 22, NO. 6, +<br/>JUNE 2000 +<br/>Evolutionary Pursuit and Its +<br/>Application to Face Recognition +</td><td>('39664966', 'Chengjun Liu', 'chengjun liu')<br/>('1781577', 'Harry Wechsler', 'harry wechsler')</td><td></td></tr><tr><td>42e0127a3fd6a96048e0bc7aab6d0ae88ba00fb0</td><td></td><td></td><td></td></tr><tr><td>42df75080e14d32332b39ee5d91e83da8a914e34</td><td>4280 +<br/>Illumination Compensation Using Oriented +<br/>Local Histogram Equalization and +<br/>Its Application to Face Recognition +</td><td>('1822733', 'Ping-Han Lee', 'ping-han lee')<br/>('2250469', 'Szu-Wei Wu', 'szu-wei wu')<br/>('1732064', 'Yi-Ping Hung', 'yi-ping hung')</td><td></td></tr><tr><td>4276eb27e2e4fc3e0ceb769eca75e3c73b7f2e99</td><td>Face Recognition From Video +<br/>1Siemens Corporate Research +<br/><b>College Road East, Princeton, NJ</b><br/>2Center for Automation Research (CfAR) and +<br/>Department of Electrical and Computer Engineering +<br/><b>University of Maryland, College Park, MD</b><br/>I. INTRODUCTION +<br/>While face recognition (FR) from a single still image has been studied extensively [13], [57], FR based on a +<br/>video sequence is an emerging topic, evidenced by the growing increase in the literature. It is predictable that with +<br/>the ubiquity of video sequences, FR based on video sequences will become more and more popular. In this chapter, +<br/>we also address FR based on a group of still images (also referred to as multiple still images). Multiple still images +<br/>are not necessarily from a video sequence; they can come from multiple independent still captures. +<br/>It is obvious that multiple still images or a video sequence can be regarded as a single still image in a degenerate +<br/>manner. More specifically, suppose that we have a group of face images {y1, . . . , yT} and a single-still-image-based +<br/>FR algorithm A (or the base algorithm), we can construct a recognition algorithm based on multiple still images +<br/>or a video sequence by fusing multiple base algorithms denoted by Ai’s. Each Ai takes a different single image +<br/>yi as input. The fusion rule can be additive, multiplicative, and so on. +<br/>Even though the fusion algorithm might work well in practice, clearly, the overall recognition performance solely +<br/>depends on the base algorithm and hence designing the base algorithm A (or the similarity function k) is of ultimate +<br/>importance. However, the fused algorithms neglect additional properties manifested in multiple still images or video +<br/>sequences. Generally speaking, algorithms that judiciously exploit these properties will perform better in terms of +<br/>recognition accuracy, computational efficiency, etc. +<br/>There are three additional properties available from multiple still images and/or video sequences: +<br/>- [P 1: Set of observations]. This property is directly exploited by the fused algorithms. One main disadvantage +<br/>may be the ad hoc nature of the combination rule. However, theoretical analysis based on a set of observations +<br/>can be performed. For example, a set of observations can be summarized using quantities like matrix, probability +<br/>density function, manifold, etc. Hence, corresponding knowledge can be utilized to match two sets. +<br/>- [P 2: Temporal continuity/Dynamics]. Successive frames in the video sequences are continuous in the +<br/>temporal dimension. Such continuity, coming from facial expression, geometric continuity related to head +<br/>July 14, 2008 +<br/>DRAFT +</td><td>('1682187', 'Shaohua Kevin Zhou', 'shaohua kevin zhou')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('1867477', 'Gaurav Aggarwal', 'gaurav aggarwal')</td><td>Email: shaohua.zhou@siemens.com, rama@cfar.umd.edu, gaurav@cs.umd.edu +</td></tr><tr><td>89945b7cd614310ebae05b8deed0533a9998d212</td><td>Divide-and-Conquer Method for L1 Norm Matrix +<br/>Factorization in the Presence of Outliers and +<br/>Missing Data +</td><td>('1803714', 'Deyu Meng', 'deyu meng')</td><td></td></tr><tr><td>89de30a75d3258816c2d4d5a733d2bef894b66b9</td><td></td><td></td><td></td></tr><tr><td>89002a64e96a82486220b1d5c3f060654b24ef2a</td><td>PIEFA: Personalized Incremental and Ensemble Face Alignment +<br/>Yang Yu⋆ +<br/><b>Rutgers University</b><br/>Piscataway, NJ, 08854 +<br/><b>The University of North Carolina at Charlotte</b><br/>Charlotte, NC, 28223 +</td><td>('4340744', 'Xi Peng', 'xi peng')<br/>('1753384', 'Shaoting Zhang', 'shaoting zhang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>xpeng.nb,yyu,dnm@cs.rutgers.edu +<br/>szhang16@uncc.edu +</td></tr><tr><td>89c84628b6f63554eec13830851a5d03d740261a</td><td>Image Enhancement and Automated Target Recognition +<br/>Techniques for Underwater Electro-Optic Imagery +<br/><b>Metron, Inc</b><br/>11911 Freedom Dr., Suite 800 +<br/>Reston, VA 20190 +<br/>Contract Number N00014-07-C-0351 +<br/>http:www.metsci.com +<br/>LONG TERM GOALS +<br/>The long-term goal of this project is to provide a flexible, accurate and extensible automated target +<br/>recognition (ATR) system for use with a variety of imaging and non-imaging sensors. Such an ATR +<br/>system, once it achieves a high level of performance, can relieve human operators from the tedious +<br/>business of pouring over vast quantities of mostly mundane data, calling the operator in only when the +<br/>computer assessment involves an unacceptable level of ambiguity. The ATR system will provide most +<br/>leading edge algorithms for detection, segmentation, and classification while incorporating many novel +<br/>algorithms that we are developing at Metron. To address one of the most critical challenges in ATR +<br/>technology, the system will also provide powerful feature extraction routines designed for specific +<br/>applications of current interest. +<br/>OBJECTIVES +<br/>The main objective of this project is to develop a complete, flexible, and extensible modular automated +<br/>target recognition (MATR) system for computer aided detection and classification (CAD/CAC) of +<br/>target objects from within cluttered and possibly noisy image data. The MATR system framework is +<br/>designed to be applicable to a wide range of situations, each with its own challenges, and so is +<br/>organized in such a way that the constituent algorithms are interchangeable and can be selected based +<br/>on their individual suitability to the particular task within the specific application. The ATR system +<br/>designer can select combinations of algorithms, many of which are being developed at Metron, to +<br/>produce a variety of systems, each tailored to specific needs. While the development of the system is +<br/>still ongoing, results for mine countermeasures (MCM) applications using electro-optical (EO) image +<br/>data have been encouraging. A brief description of the system framework, some of the novel +<br/>algorithms, and preliminary test results are provided in this interim report. +<br/>APPROACH +<br/>The MATR system is composed of several modules, as depicted in Figure 1, reflecting the sequence of +<br/>steps in the ATR process. The detection step is concerned with finding portions of an image that +<br/>contain possible objects of interest, or targets, that merit further attention. During the localization and +<br/>segmentation phase the position and approximate size and shape of the object is estimated and a +<br/>portion of the image, or “snippet,” containing the object is extracted. At this stage, image processing +<br/>may be performed on the snippet to reorient the target, mitigate noise, accentuate edge detail, etc. +<br/>1 +</td><td>('2395986', 'Thomas Giddings', 'thomas giddings')<br/>('2386585', 'Cetin Savkli', 'cetin savkli')<br/>('2632462', 'Joseph Shirron', 'joseph shirron')</td><td>phone: (703) 437-2428 fax: (703) 787-3518 email: giddings@metsci.com +</td></tr><tr><td>89c51f73ec5ebd1c2a9000123deaf628acf3cdd8</td><td>American Journal of Applied Sciences 5 (5): 574-580, 2008 +<br/>ISSN 1546-9239 +<br/>© 2008 Science Publications +<br/>Face Recognition Based on Nonlinear Feature Approach +<br/>1Eimad E.A. Abusham, 1Andrew T.B. Jin, 1Wong E. Kiong and 2G. Debashis +<br/>1Faculty of Information Science and Technology, +<br/><b>Faculty of Engineering and Technology, Multimedia University (Melaka Campus</b><br/>Jalan Ayer Keroh Lama, 75450 Bukit Beruang, Melaka, Malaysia +</td><td></td><td></td></tr><tr><td>89c73b1e7c9b5e126a26ed5b7caccd7cd30ab199</td><td>Application of an Improved Mean Shift Algorithm +<br/>in Real-time Facial Expression Recognition +<br/><b>School of Computer and Communication, Hunan University of Technology, Hunan, Zhuzhou, 412008 china</b><br/><b>School of Electrical and Information Engineering, Hunan University of Technology, Hunan, Zhuzhou, 412008 china</b><br/><b>School of Computer and Communication, Hunan University of Technology, Hunan, Zhuzhou, 412008 china</b><br/>Yan-hui ZHU +<br/><b>School of Computer and Communication, Hunan University of Technology, Hunan, Zhuzhou, 412008 china</b><br/>facial +<br/>real-time +<br/>expression +</td><td>('1719090', 'Zhao-yi Peng', 'zhao-yi peng')<br/>('1696179', 'Yu Zhou', 'yu zhou')<br/>('2276926', 'Zhi-qiang Wen', 'zhi-qiang wen')</td><td>Email:pengzhaoyi@163.com +<br/>Email:zypzy@163.com +<br/>Email: swayhzhu@163.com +<br/>Email: zhqwen20001@163.com +</td></tr><tr><td>89e7d23e0c6a1d636f2da68aaef58efee36b718b</td><td>Lucas-Kanade Scale Invariant Feature Transform for +<br/>Uncontrolled Viewpoint Face Recognition +<br/>1Division of Computer Science and Engineering, +<br/>2Center for Advanced Image and Information Technology +<br/><b>Chonbuk National University, Jeonju 561-756, Korea</b></td><td>('2642847', 'Yongbin Gao', 'yongbin gao')<br/>('4292934', 'Hyo Jong Lee', 'hyo jong lee')</td><td></td></tr><tr><td>893239f17dc2d17183410d8a98b0440d98fa2679</td><td>UvA-DARE (Digital Academic Repository) +<br/>Expression-Invariant Age Estimation +<br/>Published in: +<br/>Proceedings of the British Machine Vision Conference 2014 +<br/>DOI: +<br/>10.5244/C.28.14 +<br/>Link to publication +<br/>Citation for published version (APA): +<br/>French, & T. Pridmore (Eds.), Proceedings of the British Machine Vision Conference 2014 (pp. 14.1-14.11). +<br/>BMVA Press. DOI: 10.5244/C.28.14 +<br/>General rights +<br/>It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), +<br/>other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). +<br/>Disclaimer/Complaints regulations +<br/>If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating +<br/>your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask +<br/><b>the Library: http://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam</b><br/>The Netherlands. You will be contacted as soon as possible. +<br/>Download date: 04 Aug 2017 +<br/><b>UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl</b></td><td>('49776777', 'Alvarez Lopez', 'alvarez lopez')</td><td></td></tr><tr><td>89f4bcbfeb29966ab969682eae235066a89fc151</td><td>A Comparison of Photometric Normalisation Algorithms for Face Verification +<br/>Centre for Vision, Speech and Signal Processing +<br/><b>University of Surrey</b><br/>Guildford, Surrey, GU2 7XH, UK +</td><td>('39213687', 'James Short', 'james short')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('2173900', 'Kieron Messer', 'kieron messer')</td><td>(cid:0)j.short,j.kittler,k.messer(cid:1)@eim.surrey.ac.uk +</td></tr><tr><td>892c911ca68f5b4bad59cde7eeb6c738ec6c4586</td><td>RESEARCH ARTICLE +<br/>The Ryerson Audio-Visual Database of +<br/>Emotional Speech and Song (RAVDESS): A +<br/>dynamic, multimodal set of facial and vocal +<br/>expressions in North American English +<br/><b>Ryerson University, Toronto, Canada</b><br/><b>Information Systems, University of Wisconsin-River Falls, Wisconsin, WI, United States of America</b></td><td>('2940438', 'Frank A. Russo', 'frank a. russo')</td><td>* steven.livingstone@uwrf.edu +</td></tr><tr><td>8913a5b7ed91c5f6dec95349fbc6919deee4fc75</td><td>BigBIRD: A Large-Scale 3D Database of Object Instances +</td><td>('37248999', 'Arjun Singh', 'arjun singh')<br/>('1905626', 'James Sha', 'james sha')<br/>('39537097', 'Karthik S. Narayan', 'karthik s. narayan')<br/>('2461427', 'Tudor Achim', 'tudor achim')<br/>('1689992', 'Pieter Abbeel', 'pieter abbeel')</td><td></td></tr><tr><td>8986585975c0090e9ad97bec2ba6c4b437419dae</td><td>Unsupervised Hard Example Mining from +<br/>Videos for Improved Object Detection +<br/><b>College of Information and Computer Sciences, University of Massachusetts, Amherst</b><br/>{souyoungjin,arunirc,hzjiang,ashishsingh, +</td><td>('24525313', 'SouYoung Jin', 'souyoung jin')<br/>('2895705', 'Aruni RoyChowdhury', 'aruni roychowdhury')<br/>('40175280', 'Huaizu Jiang', 'huaizu jiang')<br/>('1785936', 'Ashish Singh', 'ashish singh')<br/>('39087749', 'Aditya Prasad', 'aditya prasad')<br/>('32315404', 'Deep Chakraborty', 'deep chakraborty')</td><td>aprasad,dchakraborty,elm}@cs.umass.edu +</td></tr><tr><td>89cabb60aa369486a1ebe586dbe09e3557615ef8</td><td>Bayesian Networks as Generative +<br/>Models for Face Recognition +<br/><b>IDIAP RESEARCH INSTITUTE</b><br/>´ECOLE POLYTECHNIQUE F´ED´ERALE DE LAUSANNE +<br/>supervised by: +<br/>Dr. S. Marcel +<br/>Prof. H. Bourlard +<br/>2009 +</td><td>('16602458', 'Guillaume Heusch', 'guillaume heusch')</td><td></td></tr><tr><td>89d3a57f663976a9ac5e9cdad01267c1fc1a7e06</td><td>Neural Class-Specific Regression for face +<br/>verification +</td><td>('38813382', 'Guanqun Cao', 'guanqun cao')<br/>('9219875', 'Moncef Gabbouj', 'moncef gabbouj')</td><td></td></tr><tr><td>8983485996d5d9d162e70d66399047c5d01ac451</td><td>Deep Feature-based Face Detection on Mobile Devices +<br/><b>Center for Automation Research, University of Maryland, College Park, MD</b><br/><b>Rutgers University, Piscataway, NJ</b></td><td>('40599829', 'Sayantan Sarkar', 'sayantan sarkar')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{ssarkar2, rama}@umiacs.umd.edu +<br/>vishal.m.patel@rutgers.edu +</td></tr><tr><td>89bc311df99ad0127383a9149d1684dfd8a5aa34</td><td>Towards ontology driven learning of +<br/>visual concept detectors +<br/><b>Dextro Robotics, Inc. 101 Avenue of the Americas, New York, USA</b></td><td>('3407640', 'Sanchit Arora', 'sanchit arora')<br/>('21781318', 'Chuck Cho', 'chuck cho')<br/>('1810102', 'Paul Fitzpatrick', 'paul fitzpatrick')</td><td></td></tr><tr><td>8981be3a69cd522b4e57e9914bf19f034d4b530c</td><td>Fast Automatic Video Retrieval using Web Images +<br/><b>Center For Automation Research, University of Maryland, College Park</b></td><td>('2257769', 'Xintong Han', 'xintong han')<br/>('47679939', 'Bharat Singh', 'bharat singh')<br/>('2852035', 'Vlad I. Morariu', 'vlad i. morariu')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td>{xintong,bharat,morariu,lsd}@umiacs.umd.edu +</td></tr><tr><td>898a66979c7e8b53a10fd58ac51fbfdb6e6e6e7c</td><td>Dynamic vs. Static Recognition of Facial +<br/>Expressions +<br/>No Author Given +<br/><b>No Institute Given</b></td><td></td><td></td></tr><tr><td>89d7cc9bbcd2fdc4f4434d153ecb83764242227b</td><td>(IJERA) ISSN: 2248-9622 www.ijera.com +<br/>Vol. 3, Issue 2, March -April 2013, pp.351-355 +<br/>Face-Name Graph Matching For The Personalities In Movie +<br/>Screen +<br/><b>VelTech HighTech Dr. Rangarajan Dr.Sakunthala Engineering College</b><br/><b>Final Year Student, M.Tech IT, Vel Tech Dr. RR andDr. SR Technical University, Chennai</b><br/>Chennai.) +</td><td></td><td></td></tr><tr><td>896f4d87257abd0f628c1ffbbfdac38c86a56f50</td><td>Action and Gesture Temporal Spotting with +<br/>Super Vector Representation +<br/><b>Southwest Jiaotong University, Chengdu, China</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Key Lab of CVPR, Shenzhen Institutes of Advanced Technology, CAS</b></td><td>('1766837', 'Xiaojiang Peng', 'xiaojiang peng')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('2985266', 'Zhuowei Cai', 'zhuowei cai')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td></td></tr><tr><td>891b10c4b3b92ca30c9b93170ec9abd71f6099c4</td><td>Facial landmark detection using structured output deep +<br/>neural networks +<br/>Soufiane Belharbi ∗1, Cl´ement Chatelain∗1, Romain H´erault∗1, and S´ebastien +<br/>1LITIS EA 4108, INSA de Rouen, Saint ´Etienne du Rouvray 76800, France +<br/>2LITIS EA 4108, UFR des Sciences, Universit´e de Rouen, France. +<br/>September 24, 2015 +</td><td>('49529671', 'Adam', 'adam')</td><td></td></tr><tr><td>451b6409565a5ad18ea49b063561a2645fa4281b</td><td>Action Sets: Weakly Supervised Action Segmentation without Ordering +<br/>Constraints +<br/><b>University of Bonn, Germany</b></td><td>('32774629', 'Alexander Richard', 'alexander richard')<br/>('51267303', 'Hilde Kuehne', 'hilde kuehne')<br/>('2946643', 'Juergen Gall', 'juergen gall')</td><td>{richard,kuehne,gall}@iai.uni-bonn.de +</td></tr><tr><td>45c340c8e79077a5340387cfff8ed7615efa20fd</td><td></td><td></td><td></td></tr><tr><td>455204fa201e9936b42756d362f62700597874c4</td><td>A REGION BASED METHODOLOGY FOR FACIAL +<br/>EXPRESSION RECOGNITION +<br/><b>Medical School, University of Ioannina, Ioannina, Greece</b><br/>Unit of Medical Technology and Intelligent Information Systems, Dept. of Computer Science +<br/><b>University of Ioannina, Ioannina, Greece</b><br/>Keywords: +<br/>Facial expression recognition, Gabor filters, filter bank, artificial neural networks, Japanese Female Facial +<br/>Expression Database (JAFFE). +</td><td>('2059518', 'Anastasios C. Koutlas', 'anastasios c. koutlas')<br/>('1692818', 'Dimitrios I. Fotiadis', 'dimitrios i. fotiadis')</td><td>me01697@cc.uoi.gr +<br/>fotiadis@cs.uoi.gr +</td></tr><tr><td>4541c9b4b7e6f7a232bdd62ae653ba5ec0f8bbf6</td><td>The role of structural facial asymmetry in asymmetry of +<br/>peak facial expressions +<br/>Karen L. Schmidt +<br/><b>University of Pittsburgh, PA, USA</b><br/><b>Carnegie Mellon University, Pittsburgh, PA, USA</b><br/>Jeffrey F. Cohn +<br/><b>University of Pittsburgh, PA, USA</b><br/>joy, anger, and disgust expressions, +<br/>Asymmetric facial expression is generally attributed to asymmetry in movement, +<br/>but structural asymmetry in the face may also affect asymmetry of expression. +<br/>Asymmetry in posed expressions was measured using image-based approaches in +<br/>digitised sequences of facial expression in 55 individuals, N /16 men, N /39 +<br/>women. Structural asymmetry (at neutral expression) was higher in men than +<br/>women and accounted for .54, .62, and .66 of the variance in asymmetry at peak +<br/>expression for +<br/>respectively. Movement +<br/>asymmetry (measured by change in pixel values over time) was found, but was +<br/>unrelated to peak asymmetry in joy or anger expressions over the whole face and in +<br/>facial subregions relevant to the expression. Movement asymmetry was negatively +<br/>related to peak asymmetry in disgust expressions. Sidedness of movement +<br/>asymmetry (defined as the ratio of summed movement on the left to movement +<br/>on the right) was consistent across emotions within individuals. Sidedness was +<br/>found only for joy expressions, which had significantly more movement on the left. +<br/>The significant role of structural asymmetry in asymmetry of emotion expression +<br/>and the exploration of facial expression asymmetry have important implications for +<br/>evolutionary interpretations of facial signalling and facial expressions in general. +<br/><b>Address correspondence to: Karen L. Schmidt, University of</b><br/>This study is part of a larger programme of research that is ongoing in the Department of +<br/><b>Psychiatry at the University of Pittsburgh</b><br/><b>Science and the Robotics Institute at Carnegie Mellon University. This study was supported in part</b><br/><b>by grants from the National Institute of Mental Health (MH 15279 and MH067976 (K. Schmidt</b><br/>and MH51435 (J. Cohn). Additional support for this project was received from Office of Naval +<br/>Research (HID 29-203). The authors acknowledge the contribution of Rebecca McNutt to this +<br/>article. A preliminary version of these results was presented at the Tenth Annual Conference: Facial +<br/>Measurement and Meaning in Rimini, Italy, September 2003. +<br/># 2006 Psychology Press, an imprint of the Taylor & Francis Group, an informa business +<br/>DOI: 10.1080/13576500600832758 +</td><td>('1689241', 'Yanxi Liu', 'yanxi liu')</td><td>Pittsburgh, 121 University Place, Pittsburgh PA 15217, USA. E-mail: kschmidt@pitt.edu +</td></tr><tr><td>4552f4d46a2cc67ccc4dd8568e5c95aa2eedb4ec</td><td>Disentangling Features in 3D Face Shapes +<br/>for Joint Face Reconstruction and Recognition∗ +<br/><b>College of Computer Science, Sichuan University</b><br/><b>Michigan State University</b></td><td>('1734409', 'Feng Liu', 'feng liu')<br/>('1778454', 'Ronghang Zhu', 'ronghang zhu')<br/>('39422721', 'Dan Zeng', 'dan zeng')<br/>('7345195', 'Qijun Zhao', 'qijun zhao')<br/>('38284381', 'Xiaoming Liu', 'xiaoming liu')</td><td></td></tr><tr><td>459960be65dd04317dd325af5b7cbb883d822ee4</td><td>The Meme Quiz: A Facial Expression Game Combining +<br/>Human Agency and Machine Involvement +<br/>Department of Computer Science and Engineering +<br/><b>University of Washington</b></td><td>('3059933', 'Kathleen Tuite', 'kathleen tuite')</td><td>{ktuite,kemelmi}@cs.washington.edu +</td></tr><tr><td>45f858f9e8d7713f60f52618e54089ba68dfcd6d</td><td>What Actions are Needed for Understanding Human Actions in Videos? +<br/><b>Carnegie Mellon University</b><br/>github.com/gsig/actions-for-actions +</td><td>('34280810', 'Gunnar A. Sigurdsson', 'gunnar a. sigurdsson')</td><td></td></tr><tr><td>45e7ddd5248977ba8ec61be111db912a4387d62f</td><td>CHEN ET AL.: ADVERSARIAL POSENET +<br/>Adversarial Learning of Structure-Aware Fully +<br/>Convolutional Networks for Landmark +<br/>Localization +</td><td>('50579509', 'Yu Chen', 'yu chen')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')<br/>('2126047', 'Xiu-Shen Wei', 'xiu-shen wei')<br/>('2161037', 'Lingqiao Liu', 'lingqiao liu')<br/>('49499405', 'Jian Yang', 'jian yang')</td><td></td></tr><tr><td>45215e330a4251801877070c85c81f42c2da60fb</td><td>Domain Adaptive Dictionary Learning +<br/><b>Center for Automation Research, UMIACS, University of Maryland, College Park</b><br/><b>Arts Media and Engineering, Arizona State University</b></td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>qiu@cs.umd.edu, {pvishalm, rama}@umiacs.umd.edu, pturaga@asu.edu +</td></tr><tr><td>457cf73263d80a1a1338dc750ce9a50313745d1d</td><td>Published as a conference paper at ICLR 2017 +<br/>DECOMPOSING MOTION AND CONTENT FOR +<br/>NATURAL VIDEO SEQUENCE PREDICTION +<br/><b>University of Michigan, Ann Arbor, USA</b><br/>2Adobe Research, San Jose, CA 95110 +<br/>3POSTECH, Pohang, Korea +<br/><b>Beihang University, Beijing, China</b><br/>5Google Brain, Mountain View, CA 94043 +</td><td>('2241528', 'Seunghoon Hong', 'seunghoon hong')<br/>('10668384', 'Xunyu Lin', 'xunyu lin')<br/>('1697141', 'Honglak Lee', 'honglak lee')<br/>('1768964', 'Jimei Yang', 'jimei yang')<br/>('1711926', 'Ruben Villegas', 'ruben villegas')</td><td></td></tr><tr><td>4526992d4de4da2c5fae7a5ceaad6b65441adf9d</td><td>System for Medical Mask Detection +<br/>in the Operating Room Through +<br/>Facial Attributes +<br/>A. Nieto-Rodr´ıguez, M. Mucientes(B), and V.M. Brea +<br/>Center for Research in Information Technologies (CiTIUS), +<br/><b>University of Santiago de Compostela, Santiago de Compostela, Spain</b></td><td></td><td>{adrian.nietorodriguez,manuel.mucientes,victor.brea}@usc.es +</td></tr><tr><td>45e616093a92e5f1e61a7c6037d5f637aa8964af</td><td>Fine-grained Evaluation on Face Detection in the Wild +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences, China</b></td><td>('1716231', 'Bin Yang', 'bin yang')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{yb.derek,yanjjie}@gmail.com +<br/>{zlei,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>45efd6c2dd4ca19eed38ceeb7c2c5568231451e1</td><td>Comparative Analysis of Statistical Approach +<br/>for Face Recognition +<br/><b>CMR Institute of Technology, Hyderabad, (India</b></td><td>('39463904', 'M.Janga Reddy', 'm.janga reddy')</td><td></td></tr><tr><td>45f3bf505f1ce9cc600c867b1fb2aa5edd5feed8</td><td></td><td></td><td></td></tr><tr><td>4560491820e0ee49736aea9b81d57c3939a69e12</td><td>Investigating the Impact of Data Volume and +<br/>Domain Similarity on Transfer Learning +<br/>Applications +<br/>State Farm Insurance, Bloomington IL 61710, USA, +</td><td>('30492517', 'Michael Bernico', 'michael bernico')<br/>('50024782', 'Yuntao Li', 'yuntao li')<br/>('41092475', 'Dingchao Zhang', 'dingchao zhang')</td><td>michael.bernico.qepz@statefarm.com +</td></tr><tr><td>4571626d4d71c0d11928eb99a3c8b10955a74afe</td><td>Geometry Guided Adversarial Facial Expression Synthesis +<br/>1National Laboratory of Pattern Recognition, CASIA +<br/>2Center for Research on Intelligent Perception and Computing, CASIA +<br/>3Center for Excellence in Brain Science and Intelligence Technology, CAS +</td><td>('3051419', 'Lingxiao Song', 'lingxiao song')<br/>('9702077', 'Zhihe Lu', 'zhihe lu')<br/>('1705643', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')<br/>('1688870', 'Tieniu Tan', 'tieniu tan')</td><td></td></tr><tr><td>4534d78f8beb8aad409f7bfcd857ec7f19247715</td><td>Under review as a conference paper at ICLR 2017 +<br/>TRANSFORMATION-BASED MODELS OF VIDEO +<br/>SEQUENCES +<br/>Facebook AI Research +</td><td>('39248118', 'Anitha Kannan', 'anitha kannan')<br/>('3149531', 'Arthur Szlam', 'arthur szlam')<br/>('1687325', 'Du Tran', 'du tran')</td><td>joost@joo.st, {akannan, ranzato, aszlam, trandu, soumith}@fb.com +</td></tr><tr><td>459e840ec58ef5ffcee60f49a94424eb503e8982</td><td>One-shot Face Recognition by Promoting Underrepresented Classes +<br/>Microsoft +<br/>One Microsoft Way, Redmond, Washington, United States +</td><td>('3133575', 'Yandong Guo', 'yandong guo')<br/>('1684635', 'Lei Zhang', 'lei zhang')</td><td>{yandong.guo, leizhang}@microsoft.com +</td></tr><tr><td>45fbeed124a8956477dbfc862c758a2ee2681278</td><td></td><td></td><td></td></tr><tr><td>451c42da244edcb1088e3c09d0f14c064ed9077e</td><td>1964 +<br/>© EURASIP, 2011 - ISSN 2076-1465 +<br/>19th European Signal Processing Conference (EUSIPCO 2011) +<br/>INTRODUCTION +</td><td></td><td></td></tr><tr><td>4568063b7efb66801e67856b3f572069e774ad33</td><td>Correspondence Driven Adaptation for Human Profile Recognition +<br/><b>NEC Laboratories America, Inc</b><br/>2Huawei Technologies (USA) +<br/>Cupertino, CA 95014 +<br/>Santa Clara, CA 95050 +</td><td>('2909406', 'Ming Yang', 'ming yang')<br/>('1682028', 'Shenghuo Zhu', 'shenghuo zhu')<br/>('39157653', 'Fengjun Lv', 'fengjun lv')<br/>('38701713', 'Kai Yu', 'kai yu')</td><td>{myang,zsh,kyu}@sv.nec-labs.com +<br/>felix.Lv@huawei.com +</td></tr><tr><td>45c31cde87258414f33412b3b12fc5bec7cb3ba9</td><td>Coding Facial Expressions with Gabor Wavelets +<br/><b>ATR Human Information Processing Research Laboratory</b><br/>2-2 Hikaridai, Seika-cho +<br/>Soraku-gun, Kyoto 619-02, Japan +<br/><b>Kyushu University</b></td><td>('34801422', 'Shigeru Akamatsu', 'shigeru akamatsu')<br/>('40533190', 'Miyuki Kamachi', 'miyuki kamachi')<br/>('8365437', 'Jiro Gyoba', 'jiro gyoba')</td><td>mlyons@hip.atr.co.jp +</td></tr><tr><td>4542273a157bfd4740645a6129d1784d1df775d2</td><td>FaceRipper +<br/>Automatic Face Indexer and Tagger for Personal +<br/>Albums and Videos +<br/>A PROJECT REPORT +<br/>SUBMITTED IN PARTIAL FULFILMENT OF THE +<br/>REQUIREMENTS FOR THE DEGREE OF +<br/>Master of Engineering +<br/>IN +<br/>COMPUTER SCIENCE AND ENGINEERING +<br/>by +<br/>Computer Science and Automation +<br/><b>Indian Institute of Science</b><br/>BANGALORE – 560 012 +<br/>July 2007 +</td><td>('2819449', 'Mehul Parsana', 'mehul parsana')</td><td></td></tr><tr><td>4511e09ee26044cb46073a8c2f6e1e0fbabe33e8</td><td></td><td></td><td></td></tr><tr><td>45513d0f2f5c0dac5b61f9ff76c7e46cce62f402</td><td>LEE,GRAUMAN:FACEDISCOVERYWITHSOCIALCONTEXT +<br/>Face Discovery with Social Context +<br/>https://webspace.utexas.edu/yl3663/~ylee/ +<br/>http://www.cs.utexas.edu/~grauman/ +<br/><b>University of Texas at Austin</b><br/>Austin, TX, USA +</td><td>('1883898', 'Yong Jae Lee', 'yong jae lee')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>45e459462a80af03e1bb51a178648c10c4250925</td><td>LCrowdV: Generating Labeled Videos for +<br/>Simulation-based Crowd Behavior Learning +<br/><b>The University of North Carolina at Chapel Hill</b></td><td>('3422427', 'Ernest Cheung', 'ernest cheung')<br/>('3422442', 'Tsan Kwong Wong', 'tsan kwong wong')<br/>('2718563', 'Aniket Bera', 'aniket bera')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1699159', 'Dinesh Manocha', 'dinesh manocha')</td><td></td></tr><tr><td>458677de7910a5455283a2be99f776a834449f61</td><td>Face Image Retrieval Using Facial Attributes By +<br/>K-Means +<br/>[1]I.Sudha, [2]V.Saradha, [3]M.Tamilselvi, [4]D.Vennila +<br/>[1]AP, Department of CSE ,[2][3][4] B.Tech(CSE) +<br/><b>Achariya college of Engineering Technology</b><br/>Puducherry +</td><td></td><td></td></tr><tr><td>45a6333fc701d14aab19f9e2efd59fe7b0e89fec</td><td>HAND POSTURE DATASET CREATION FOR GESTURE +<br/>RECOGNITION +<br/>Luis Anton-Canalis +<br/>Instituto de Sistemas Inteligentes y Aplicaciones Numericas en Ingenieria +<br/>Campus Universitario de Tafira, 35017 Gran Canaria, Spain +<br/>Elena Sanchez-Nielsen +<br/>Departamento de E.I.O. y Computacion +<br/>38271 Universidad de La Laguna, Spain +<br/>Keywords: +<br/>Image understanding, Gesture recognition, Hand dataset. +</td><td></td><td></td></tr><tr><td>450c6a57f19f5aa45626bb08d7d5d6acdb863b4b</td><td>Towards Interpretable Face Recognition +<br/><b>Michigan State University</b><br/>2 Adobe Inc. +<br/>3 Aibee +</td><td>('32032812', 'Bangjie Yin', 'bangjie yin')<br/>('1849929', 'Luan Tran', 'luan tran')<br/>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('1720987', 'Xiaohui Shen', 'xiaohui shen')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td>{yinbangj, tranluan, liuxm}@msu.edu, xshen@adobe.com, lhxustcer@gmail.com +</td></tr><tr><td>1f9b2f70c24a567207752989c5bd4907442a9d0f</td><td>Deep Representations to Model User ‘Likes’ +<br/><b>School of Computer Engineering, Nanyang Technological University, Singapore</b><br/><b>Institute for Infocomm Research, Singapore</b><br/><b>QCIS, University of Technology, Sydney</b></td><td>('2731733', 'Sharath Chandra Guntuku', 'sharath chandra guntuku')<br/>('10638646', 'Joey Tianyi Zhou', 'joey tianyi zhou')<br/>('1872875', 'Sujoy Roy', 'sujoy roy')<br/>('1807998', 'Ivor W. Tsang', 'ivor w. tsang')</td><td>sharathc001@e.ntu.edu.sg, tzhou1@ntu.edu.sg, wslin@ntu.edu.sg +<br/>sujoy@i2r.a-star.edu.sg +<br/>ivor.tsang@uts.edu.au +</td></tr><tr><td>1fe1bd6b760e3059fff73d53a57ce3a6079adea1</td><td>SINGH ET AL.: SCALING BAG-OF-VISUAL-WORDS GENERATION +<br/>Fast-BoW: Scaling Bag-of-Visual-Words +<br/>Generation +<br/>Visual Learning & Intelligence Group +<br/>Department of Computer Science and +<br/>Engineering +<br/><b>Indian Institute of Technology</b><br/>Hyderabad +<br/>Kandi, Sangareddy, Telangana, India +</td><td>('40624178', 'Dinesh Singh', 'dinesh singh')<br/>('51292354', 'Abhijeet Bhure', 'abhijeet bhure')<br/>('51305895', 'Sumit Mamtani', 'sumit mamtani')<br/>('34358756', 'C. Krishna Mohan', 'c. krishna mohan')</td><td>cs14resch11003@iith.ac.in +<br/>cs15btech11001@iith.ac.in +<br/>cs15btech11022@iith.ac.in +<br/>ckm@iith.ac.in +</td></tr><tr><td>1f05473c587e2a3b587f51eb808695a1c10bc153</td><td>Towards Good Practices for Very Deep Two-Stream ConvNets +<br/><b>The Chinese University of Hong Kong, Hong Kong</b><br/><b>Shenzhen key lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('1915826', 'Zhe Wang', 'zhe wang')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td>{07wanglimin,bitxiong,buptwangzhe2012}@gmail.com, yu.qiao@siat.ac.cn +</td></tr><tr><td>1fa3948af1c338f9ae200038c45adadd2b39a3e4</td><td>Computational Explorations of Split Architecture in Modeling Face and Object +<br/>Recognition +<br/><b>University of California San Diego</b><br/>9500 Gilman Drive #0404, La Jolla, CA 92093, USA +<br/><b>University of California San Diego</b><br/>9500 Gilman Drive #0515, La Jolla, CA 92093, USA +</td><td></td><td>Janet Hui-wen Hsiao (jhsiao@cs.ucsd.edu) +<br/>Garrison W. Cottrell (gary@ucsd.edu) +<br/>Danke Shieh (danke@ucsd.edu) +</td></tr><tr><td>1ffe20eb32dbc4fa85ac7844178937bba97f4bf0</td><td>Face Clustering: Representation and Pairwise +<br/>Constraints +</td><td>('9644181', 'Yichun Shi', 'yichun shi')<br/>('40653304', 'Charles Otto', 'charles otto')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>1f8304f4b51033d2671147b33bb4e51b9a1e16fe</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Beyond Trees: +<br/>MAP Inference in MRFs via Outer-Planar Decomposition +<br/>Received: date / Accepted: date +</td><td>('1746610', 'Dhruv Batra', 'dhruv batra')</td><td></td></tr><tr><td>1f89439524e87a6514f4fbe7ed34bda4fd1ce286</td><td><b>Carnegie Mellon University</b><br/>Department of Statistics +<br/><b>Dietrich College of Humanities and Social Sciences</b><br/>9-2005 +<br/>Devising Face Authentication System and +<br/>Performance Evaluation Based on Statistical +<br/>Models +<br/><b>Carnegie Mellon University</b><br/>Follow this and additional works at: http://repository.cmu.edu/statistics +<br/>Part of the Statistics and Probability Commons +</td><td>('2046854', 'Sinjini Mitra', 'sinjini mitra')<br/>('1680307', 'Anthony Brockwell', 'anthony brockwell')<br/>('1794486', 'Marios Savvides', 'marios savvides')<br/>('1684961', 'Stephen E. Fienberg', 'stephen e. fienberg')</td><td>Research Showcase @ CMU +<br/>Carnegie Mellon University, abrock@stat.cmu.edu +<br/>Carnegie Mellon University, msavvid@cs.cmu.edu +<br/>Carnegie Mellon University, fienberg@stat.cmu.edu +<br/>This Technical Report is brought to you for free and open access by the Dietrich College of Humanities and Social Sciences at Research Showcase @ +<br/>CMU. It has been accepted for inclusion in Department of Statistics by an authorized administrator of Research Showcase @ CMU. For more +<br/>information, please contact research-showcase@andrew.cmu.edu. +</td></tr><tr><td>1f9ae272bb4151817866511bd970bffb22981a49</td><td>An Iterative Regression Approach for Face Pose Estima- +<br/>tion from RGB Images +<br/>This paper presents a iterative optimization method, explicit shape regression, for face pose +<br/>detection and localization. The regression function is learnt to find out the entire facial shape +<br/>and minimize the alignment errors. A cascaded learning framework is employed to enhance +<br/>shape constraint during detection. A combination of a two-level boosted regression, shape +<br/>performance. In this paper, we have explain the advantage of ESR for deformable object like +<br/>face pose estimation and reveal its generic applications of the method. In the experiment, +<br/>we compare the results with different work and demonstrate the accuracy and robustness in +<br/>different scenarios. +<br/>Introduction +<br/>Pose estimation is an important problem in computer vision, and has enabled many practical ap- +<br/>plication from face expression 1 to activity tracking 2. Researchers design a new algorithm called +<br/>explicit shape regression (ESR) to find out face alignment from a picture 3. Figure 1 shows how +<br/>the system uses ESR to learn a shape of a human face image. A simple way to identify a face is to +<br/>find out facial landmarks like eyes, nose, mouth and chin. The researchers define a face shape S +<br/>and S is composed of Nf p facial landmarks. Therefore, they get S = [x1, y1, ..., xNf p, yNf p]T . The +<br/>objective of the researchers is to estimate a shape S of a face image. The way to know the accuracy +</td><td>('3988780', 'Wenye He', 'wenye he')</td><td></td></tr><tr><td>1fd6004345245daf101c98935387e6ef651cbb55</td><td>Learning Symmetry Features for Face Detection +<br/>Based on Sparse Group Lasso +<br/>Center for Research on Intelligent Perception and Computing, +<br/><b>National Laboratory of Pattern Recognition, Institute of Automation</b><br/>Chinese Academy of Sciences, Beijing, China +</td><td>('39763795', 'Qi Li', 'qi li')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')<br/>('1705643', 'Ran He', 'ran he')<br/>('1688870', 'Tieniu Tan', 'tieniu tan')</td><td>{qli,znsun,rhe,tnt}@nlpr.ia.ac.cn +</td></tr><tr><td>1fc249ec69b3e23856b42a4e591c59ac60d77118</td><td>Evaluation of a 3D-aided Pose Invariant 2D Face Recognition System +<br/>Computational Biomedicine Lab +<br/>4800 Calhoun Rd. Houston, TX, USA +</td><td>('5084124', 'Xiang Xu', 'xiang xu')<br/>('26401746', 'Ha A. Le', 'ha a. le')<br/>('39634395', 'Pengfei Dou', 'pengfei dou')<br/>('2461369', 'Yuhang Wu', 'yuhang wu')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td>{xxu18, hale4, pdou, ywu35, ikakadia}@central.uh.edu +</td></tr><tr><td>1fbde67e87890e5d45864e66edb86136fbdbe20e</td><td>The Action Similarity Labeling Challenge +</td><td>('3294355', 'Orit Kliper-Gross', 'orit kliper-gross')<br/>('1756099', 'Tal Hassner', 'tal hassner')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td></td></tr><tr><td>1f41a96589c5b5cee4a55fc7c2ce33e1854b09d6</td><td>Demographic Estimation from Face Images: +<br/>Human vs. Machine Performance +</td><td>('34393045', 'Hu Han', 'hu han')<br/>('40653304', 'Charles Otto', 'charles otto')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>1fd2ed45fb3ba77f10c83f0eef3b66955645dfe0</td><td></td><td></td><td></td></tr><tr><td>1fe59275142844ce3ade9e2aed900378dd025880</td><td>Facial Landmark Detection via Progressive Initialization +<br/><b>National University of Singapore</b><br/>Singapore 117576 +</td><td>('3124720', 'Shengtao Xiao', 'shengtao xiao')</td><td>xiao shengtao@u.nus.edu, eleyans@nus.edu.sg, ashraf@nus.edu.sg +</td></tr><tr><td>1f2d12531a1421bafafe71b3ad53cb080917b1a7</td><td></td><td></td><td></td></tr><tr><td>1fe121925668743762ce9f6e157081e087171f4c</td><td>Unsupervised Learning of Overcomplete Face Descriptors +<br/>Center for Machine Vision Research +<br/><b>University of Oulu</b></td><td>('32683737', 'Juha Ylioinas', 'juha ylioinas')<br/>('1776374', 'Juho Kannala', 'juho kannala')<br/>('1751372', 'Abdenour Hadid', 'abdenour hadid')</td><td>firstname.lastname@ee.oulu.fi +</td></tr><tr><td>1fefb2f8dd1efcdb57d5c2966d81f9ab22c1c58d</td><td>vExplorer: A Search Method to Find Relevant YouTube Videos for Health +<br/>Researchers +<br/>IBM Research, Cambridge, MA, USA +</td><td>('1764750', 'Hillol Sarker', 'hillol sarker')<br/>('3456866', 'Murtaza Dhuliawala', 'murtaza dhuliawala')<br/>('31633051', 'Nicholas Fay', 'nicholas fay')<br/>('15793829', 'Amar Das', 'amar das')</td><td></td></tr><tr><td>1fdeba9c4064b449231eac95e610f3288801fd3e</td><td>Fine-Grained Head Pose Estimation Without Keypoints +<br/><b>Georgia Institute of Technology</b></td><td>('31601235', 'Nataniel Ruiz', 'nataniel ruiz')<br/>('39832600', 'Eunji Chong', 'eunji chong')<br/>('1692956', 'James M. Rehg', 'james m. rehg')</td><td>{nataniel.ruiz, eunjichong, rehg}@gatech.edu +</td></tr><tr><td>1f8e44593eb335c2253d0f22f7f9dc1025af8c0d</td><td>Fine-tuning regression forests votes for object alignment in the wild. +<br/>Yang, H; Patras, I +<br/>© 2017 IEEE. Personal use of this material is permitted. Permission from IEEE must be +<br/><b>obtained for all other uses, in any current or future media, including reprinting/republishing</b><br/>this material for advertising or promotional purposes, creating new collective works, for resale +<br/>or redistribution to servers or lists, or reuse of any copyrighted component of this work in +<br/>other works. +<br/>For additional information about this publication click this link. +<br/>http://qmro.qmul.ac.uk/xmlui/handle/123456789/22607 +<br/>Information about this research object was correct at the time of download; we occasionally +<br/>make corrections to records, please therefore check the published record when citing. For +</td><td></td><td>more information contact scholarlycommunications@qmul.ac.uk +</td></tr><tr><td>1f94734847c15fa1da68d4222973950d6b683c9e</td><td>Embedding Label Structures for Fine-Grained Feature Representation +<br/>UNC Charlotte +<br/>Charlotte, NC 28223 +<br/>NEC Lab America +<br/>Cupertino, CA 95014 +<br/>NEC Lab America +<br/>Cupertino, CA 95014 +<br/>UNC Charlotte +<br/>Charlotte, NC 28223 +</td><td>('2739998', 'Xiaofan Zhang', 'xiaofan zhang')<br/>('1757386', 'Feng Zhou', 'feng zhou')<br/>('1695082', 'Yuanqing Lin', 'yuanqing lin')<br/>('1753384', 'Shaoting Zhang', 'shaoting zhang')</td><td>xzhang35@uncc.edu +<br/>feng@nec-labs.com +<br/>ylin@nec-labs.com +<br/>szhang16@uncc.edu +</td></tr><tr><td>1f745215cda3a9f00a65166bd744e4ec35644b02</td><td>Facial Cosmetics Database and Impact Analysis on +<br/>Automatic Face Recognition +<br/># Computer Science Department, TU Muenchen +<br/>Boltzmannstr. 3, 85748 Garching b. Muenchen, Germany +<br/>∗ Multimedia Communications Department, EURECOM +<br/>450 Route des Chappes, 06410 Biot, France +</td><td>('38996894', 'Marie-Lena Eckert', 'marie-lena eckert')<br/>('1862703', 'Neslihan Kose', 'neslihan kose')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td>1 marie-lena.eckert@mytum.de +<br/>2 kose@eurecom.fr +<br/>3 jld@eurecom.fr +</td></tr><tr><td>1fff309330f85146134e49e0022ac61ac60506a9</td><td>Data-Driven Sparse Sensor Placement for Reconstruction +</td><td>('37119658', 'Krithika Manohar', 'krithika manohar')<br/>('1824880', 'Bingni W. Brunton', 'bingni w. brunton')<br/>('1937069', 'J. Nathan Kutz', 'j. nathan kutz')<br/>('3083169', 'Steven L. Brunton', 'steven l. brunton')</td><td>∗Corresponding author: kmanohar@uw.edu +</td></tr><tr><td>1fd3dbb6e910708fa85c8a86e17ba0b6fef5617c</td><td><b>ARISTOTLE UNIVERSITY OF THESSALONIKI</b><br/>FACULTY OF SCIENCES +<br/>DEPARTMENT OF INFORMATICS +<br/>POSTGRADUATE STUDIES PROGRAMME +<br/>Age interval and gender prediction using PARAFAC2 on +<br/>speech recordings and face images +<br/>Supervisor: Professor Kotropoulos Constantine +<br/>A thesis submitted in partial fulfillment of the requirements +<br/>for the degree of Master of Science +<br/>July 2016 +</td><td></td><td></td></tr><tr><td>1f24cef78d1de5aa1eefaf344244dcd1972797e8</td><td>Outlier-Robust Tensor PCA +<br/><b>National University of Singapore, Singapore</b></td><td>('33481412', 'Pan Zhou', 'pan zhou')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')</td><td>pzhou@u.nus.edu +<br/>elefjia@nus.edu.sg +</td></tr><tr><td>1fe990ca6df273de10583860933d106298655ec8</td><td><b>College of Information Science and Engineering</b><br/><b>Hunan University</b><br/>Changsha, 410082 P.R. China +<br/>In this paper, we propose a wavelet-based illumination normalization method for +<br/>face recognition against different directions and strength of light. Here, by one-level +<br/>discrete wavelet transform, a given face image is first decomposed into low frequency +<br/>and high frequency components, respectively, and then the two components are pro- +<br/>cessed separately through contrast enhancement to eliminate the effect of illumination +<br/>variations and enhance the detailed edge information. Finally the normalized image is +<br/>obtained through the inverse discrete wavelet transform. Experimental results on the +<br/>Yale B, the extended Yale B and CMU PIE face databases show that the proposed +<br/>method can effectively reduce the effect of illumination variations on face recognition. +<br/>Keywords: face recognition, illumination normalization, discrete wavelet transform, edge +<br/>enhancement, face representation +<br/>1. INTRODUCTION +<br/>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 31, 1711-1731 (2015) +<br/>A Wavelet-Based Image Preprocessing Method +<br/>for Illumination Insensitive Face Recognition +<br/>Face recognition plays an important role in pattern recognition and computer vision +<br/>due to its wide applications in human computer interaction, information security and +<br/>access control, law enforcement and entertainment [1]. Various methods have been pro- +<br/>posed for face recognition, such as PCA [2], LDA [3], LFA [4], EBGM [5], probabilistic +<br/>and Bayesian matching [6] and SVM [7]. These methods can yield good performance +<br/>when face images are well frontally illuminated. Existing studies have proved that face +<br/>recognition for the same face with different illumination conditions is more difficult than +<br/>the perception of face identity [8, 9]. The reason is that an object's appearance largely +<br/>depends on the way in which it is viewed. Illumination variations mainly consist of the +<br/>lighting direction and the lighting intensity. Usually, slight changes in illumination pro- +<br/>duce dramatical changes in the face appearance. So, the performance of face recognition +<br/>is highly sensitive to the illumination condition. For example, the unsuitable lighting +<br/>direction and intensity may lead to underexposed or overexposed regions over the face, +<br/>and weaken the discrimination of face features such as skin texture, eye detail, etc. +<br/>Therefore, illumination normalization is a very important task for face recognition under +<br/>varying illumination. +<br/>To make face recognition relatively insensitive to illumination variations, many +<br/>methods have been proposed with the goal of illumination normalization, illumination- +<br/>invariant feature extraction or illumination variation modeling [10]. Illumination-inva- +<br/>riant approaches generally fall into three classes. The first class is to preprocess face +<br/>images by using some simply techniques, such as logarithm transform and histogram +<br/>Received March 26, 2014; revised May 26, 2014; accepted July 17, 2014. +<br/>Communicated by Chung-Lin Huang. +<br/>1711 +</td><td>('2078993', 'Xiaochao Zhao', 'xiaochao zhao')<br/>('2138422', 'Yaping Lin', 'yaping lin')<br/>('2431083', 'Bo Ou', 'bo ou')<br/>('1824216', 'Junfeng Yang', 'junfeng yang')</td><td>E-mail: {s12103017; yplin; oubo; B12100031}@hnu.edu.cn +</td></tr><tr><td>1feeab271621128fe864e4c64bab9b2e2d0ed1f1</td><td>Article +<br/>Perception-Link Behavior Model: Supporting +<br/>a Novel Operator Interface for a Customizable +<br/>Anthropomorphic Telepresence Robot +<br/><b>BeingTogether Centre, Institute for Media Innovation, Singapore 637553, Singapore</b><br/><b>Robotic Research Centre, Nanyang Technological University, Singapore 639798, Singapore</b><br/>Received: 15 May 2017; Accepted: 15 July 2017; Published: 20 July 2017 +</td><td>('1768723', 'William Gu', 'william gu')<br/>('9216152', 'Gerald Seet', 'gerald seet')<br/>('1695679', 'Nadia Magnenat-Thalmann', 'nadia magnenat-thalmann')</td><td>mglseet@ntu.edu.sg (G.S.); NADIATHALMANN@ntu.edu.sg (N.M.-T.) +<br/>* Correspondence: GUYU0007@e.ntu.edu.sg +</td></tr><tr><td>73b90573d272887a6d835ace89bfaf717747c59b</td><td>Feature Disentangling Machine - A Novel +<br/>Approach of Feature Selection and Disentangling +<br/>in Facial Expression Analysis +<br/><b>University of South Carolina, USA</b><br/><b>Center for Computational Intelligence, Nanyang Technology University, Singapore</b><br/>3 Center for Quantum Computation and Intelligent Systems, +<br/><b>University of Technology, Australia</b></td><td>('40205868', 'Ping Liu', 'ping liu')<br/>('10638646', 'Joey Tianyi Zhou', 'joey tianyi zhou')<br/>('3091647', 'Zibo Meng', 'zibo meng')<br/>('49107074', 'Shizhong Han', 'shizhong han')<br/>('1686235', 'Yan Tong', 'yan tong')</td><td></td></tr><tr><td>73f467b4358ac1cafb57f58e902c1cab5b15c590</td><td> ISSN 0976 3724 47 +<br/>Combination of Dimensionality Reduction Techniques for Face +<br/>Image Retrieval: A Review +<br/><b>M.Tech Scholar, MES College of Engineering, Kuttippuram</b><br/>Kerala +<br/><b>MES College of Engineering, Kuttippuram</b><br/>Kerala +</td><td></td><td>fousisadath@gmail.com +<br/>Jahfar.ali@gmail.com +</td></tr><tr><td>7323b594d3a8508f809e276aa2d224c4e7ec5a80</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>An Experimental Evaluation of Covariates +<br/>Effects on Unconstrained Face Verification +</td><td>('2927406', 'Boyu Lu', 'boyu lu')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>732e8d8f5717f8802426e1b9debc18a8361c1782</td><td>Unimodal Probability Distributions for Deep Ordinal Classification +</td><td>('12757989', 'Christopher Beckham', 'christopher beckham')</td><td></td></tr><tr><td>73ed64803d6f2c49f01cffef8e6be8fc9b5273b8</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Cooking in the kitchen: Recognizing and Segmenting Human +<br/>Activities in Videos +<br/>Received: date / Accepted: date +</td><td>('51267303', 'Hilde Kuehne', 'hilde kuehne')</td><td></td></tr><tr><td>7306d42ca158d40436cc5167e651d7ebfa6b89c1</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Transductive Zero-Shot Action Recognition by +<br/>Word-Vector Embedding +<br/>Received: date / Accepted: date +</td><td>('47158489', 'Xun Xu', 'xun xu')</td><td></td></tr><tr><td>734cdda4a4de2a635404e4c6b61f1b2edb3f501d</td><td>Tie and Guan EURASIP Journal on Image and Video Processing 2013, 2013:8 +<br/>http://jivp.eurasipjournals.com/content/2013/1/8 +<br/>R ES EAR CH +<br/>Open Access +<br/>Automatic landmark point detection and tracking +<br/>for human facial expressions +</td><td>('1721867', 'Ling Guan', 'ling guan')</td><td></td></tr><tr><td>739d400cb6fb730b894182b29171faaae79e3f01</td><td>A New Regularized Orthogonal Local Fisher Discriminant Analysis for Image +<br/>Feature Extraction +<br/>dept. name of organization, name of organization, City, Country +<br/><b>School of Management Engineering, Henan Institute of Engineering, Zhengzhou 451191, P.R. China</b><br/><b>Institute of Information Science, Beijing Jiaotong University, Beijing 100044, P.R. China</b></td><td>('2539310', 'ZHONGFENG WANG', 'zhongfeng wang')<br/>('2539310', 'ZHONGFENG WANG', 'zhongfeng wang')<br/>('1718667', 'Zhan WANG', 'zhan wang')</td><td></td></tr><tr><td>732e4016225280b485c557a119ec50cffb8fee98</td><td>Are all training examples equally valuable? +<br/><b>Massachusetts Institute of Technology</b><br/>Universitat Oberta de Catalunya +<br/>Agata Lapedriza +<br/>Computer Vision Center +<br/><b>Massachusetts Institute of Technology</b><br/><b>Massachusetts Institute of Technology</b><br/><b>Massachusetts Institute of Technology</b></td><td>('2367683', 'Hamed Pirsiavash', 'hamed pirsiavash')<br/>('3326347', 'Zoya Bylinskii', 'zoya bylinskii')<br/>('1690178', 'Antonio Torralba', 'antonio torralba')</td><td>hpirsiav@mit.edu +<br/>agata@mit.edu +<br/>zoya@mit.edu +<br/>torralba@mit.edu +</td></tr><tr><td>7373c4a23684e2613f441f2236ed02e3f9942dd4</td><td>This document is downloaded from DR-NTU, Nanyang Technological +<br/><b>University Library, Singapore</b><br/>Title +<br/>Feature extraction through binary pattern of phase +<br/>congruency for facial expression recognition +<br/>Author(s) +<br/>Shojaeilangari, Seyedehsamaneh; Yau, Wei-Yun; Li, Jun; +<br/>Teoh, Eam Khwang +<br/>Citation +<br/>Shojaeilangari, S., Yau, W. Y., Li, J., & Teoh, E. K. +<br/>(2012). Feature extraction through binary pattern of +<br/>phase congruency for facial expression recognition. 12th +<br/>International Conference on Control Automation Robotics +<br/>& Vision (ICARCV), 166-170. +<br/>Date +<br/>2012 +<br/>URL +<br/>http://hdl.handle.net/10220/18012 +<br/>Rights +<br/>© 2012 IEEE. Personal use of this material is permitted. +<br/>Permission from IEEE must be obtained for all other +<br/><b>uses, in any current or future media, including</b><br/>reprinting/republishing this material for advertising or +<br/>promotional purposes, creating new collective works, for +<br/>resale or redistribution to servers or lists, or reuse of any +<br/>copyrighted component of this work in other works. The +<br/>published version is available at: +<br/>[http://dx.doi.org/10.1109/ICARCV.2012.6485152]. +</td><td></td><td></td></tr><tr><td>732686d799d760ccca8ad47b49a8308b1ab381fb</td><td>Running head: TEACHERS’ DIFFERING BEHAVIORS +<br/>1 +<br/>Graduate School of Psychology +<br/>RESEARCH MASTER’S PSYCHOLOGY THESıS REPORT +<br/> +<br/>Teachers’ differing classroom behaviors: +<br/>The role of emotional sensitivity and cultural tolerance +<br/>Research Master’s, Social Psychology +<br/>Ethics Committee Reference Code: 2016-SP-7084 +</td><td>('7444483', 'Agneta Fischer', 'agneta fischer')<br/>('22253276', 'Disa Sauter', 'disa sauter')<br/>('2808612', 'Monique Volman', 'monique volman')</td><td></td></tr><tr><td>73fbdd57270b9f91f2e24989178e264f2d2eb7ae</td><td>978-1-4673-0046-9/12/$26.00 ©2012 IEEE +<br/>1945 +<br/>ICASSP 2012 +</td><td></td><td></td></tr><tr><td>738a985fba44f9f5acd516e07d0d9578f2ffaa4e</td><td>MACHINE LEARNING TECHNIQUES FOR FACE ANALYSIS +<br/>Man Machine Interaction Group +<br/><b>Delft University of Technology</b><br/>Mekelweg 4, 2628 CD Delft +<br/>The Netherlands +<br/>from +<br/>learning, pattern recognition, classifiers, face +<br/>KEYWORDS +<br/>Machine +<br/>detection, facial expression recognition. +</td><td>('2866326', 'D. Datcu', 'd. datcu')</td><td>E-mail: {D.Datcu, L.J.M.Rothkrantz}@ewi.tudelft.nl +</td></tr><tr><td>73fd7e74457e0606704c5c3d3462549f1b2de1ad</td><td>Learning Predictable and Discriminative Attributes +<br/>for Visual Recognition +<br/><b>School of Software, Tsinghua University, Beijing 100084, China</b></td><td>('34811036', 'Yuchen Guo', 'yuchen guo')<br/>('38329336', 'Guiguang Ding', 'guiguang ding')<br/>('39665252', 'Xiaoming Jin', 'xiaoming jin')<br/>('1751179', 'Jianmin Wang', 'jianmin wang')</td><td>yuchen.w.guo@gmail.com, {dinggg,xmjin,jimwang}@tsinghua.edu.cn, +</td></tr><tr><td>73c5bab5c664afa96b1c147ff21439135c7d968b</td><td>Whitened LDA for Face Recognition ∗ +<br/>Ubiquitous Computing Lab +<br/><b>Kyung Hee University</b><br/>Suwon, Korea +<br/>Ubiquitous Computing Lab +<br/><b>Kyung Hee University</b><br/>Suwon, Korea +<br/>Mobile Computing Lab +<br/><b>SungKyunKwan University</b><br/>Suwon, Korea +</td><td>('1687579', 'Vo Dinh Minh Nhat', 'vo dinh minh nhat')<br/>('1700806', 'Sungyoung Lee', 'sungyoung lee')<br/>('1718666', 'Hee Yong Youn', 'hee yong youn')</td><td>vdmnhat@oslab.khu.ac.kr +<br/>sylee@oslab.khu.ac.kr +<br/>youn@ece.skku.ac.kr +</td></tr><tr><td>73c9cbbf3f9cea1bc7dce98fce429bf0616a1a8c</td><td></td><td></td><td></td></tr><tr><td>877100f430b72c5d60de199603ab5c65f611ce17</td><td>Within-person variability in men’s facial +<br/>width-to-height ratio +<br/><b>University of York, York, United Kingdom</b></td><td>('40598264', 'Robin S.S. Kramer', 'robin s.s. kramer')</td><td></td></tr><tr><td>870433ba89d8cab1656e57ac78f1c26f4998edfb</td><td>Regressing Robust and Discriminative 3D Morphable Models +<br/>with a very Deep Neural Network +<br/><b>Institute for Robotics and Intelligent Systems, USC, CA, USA</b><br/><b>Information Sciences Institute, USC, CA, USA</b><br/><b>The Open University of Israel, Israel</b></td><td>('1756099', 'Tal Hassner', 'tal hassner')<br/>('11269472', 'Iacopo Masi', 'iacopo masi')</td><td></td></tr><tr><td>872dfdeccf99bbbed7c8f1ea08afb2d713ebe085</td><td>L2-constrained Softmax Loss for Discriminative Face Verification +<br/><b>Center for Automation Research, UMIACS, University of Maryland, College Park, MD</b></td><td>('48467498', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('38171682', 'Carlos D. Castillo', 'carlos d. castillo')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{rranjan1,carlos,rama}@umiacs.umd.edu +</td></tr><tr><td>87e6cb090aecfc6f03a3b00650a5c5f475dfebe1</td><td>KIM, BALTRUŠAITIS et al.: HOLISTICALLY CONSTRAINED LOCAL MODEL +<br/>Holistically Constrained Local Model: +<br/>Going Beyond Frontal Poses for Facial +<br/>Landmark Detection +<br/>Tadas Baltrušaitis2 +<br/>Amir Zadeh2 +<br/>Gérard Medioni1 +<br/><b>Institute for Robotics and Intelligent</b><br/>Systems +<br/><b>University of Southern California</b><br/>Los Angeles, CA, USA +<br/><b>Language Technologies Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA, USA +</td><td>('2792633', 'KangGeon Kim', 'kanggeon kim')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td>kanggeon.kim@usc.edu +<br/>tbaltrus@cs.cmu.edu +<br/>abagherz@cs.cmu.edu +<br/>morency@cs.cmu.edu +<br/>medioni@usc.edu +</td></tr><tr><td>8796f2d54afb0e5c924101f54d469a1d54d5775d</td><td>Journal of Signal and Information Processing, 2012, 3, 45-50 +<br/>http://dx.doi.org/10.4236/jsip.2012.31007 Published Online February 2012 (http://www.SciRP.org/journal/jsip) +<br/>45 +<br/>Illumination Invariant Face Recognition Using Fuzzy LDA +<br/>and FFNN +<br/><b>School of Electrical Engineering, Iran University of Science and Technology, Tehran, Iran</b><br/>Received October 20th, 2011; revised November 24th, 2011; accepted December 10th, 2011 +</td><td>('1697559', 'Behzad Bozorgtabar', 'behzad bozorgtabar')<br/>('3280435', 'Hamed Azami', 'hamed azami')<br/>('3097307', 'Farzad Noorian', 'farzad noorian')</td><td>Email: b_bozorgtabar@elec.iust.ac.ir, hmdazami@gmail.com, fnoorian@ee.iust.ac.ir +</td></tr><tr><td>87f285782d755eb85d8922840e67ed9602cfd6b9</td><td>INCORPORATING BOLTZMANN MACHINE PRIORS +<br/>FOR SEMANTIC LABELING IN IMAGES AND VIDEOS +<br/>A Dissertation Presented +<br/>by +<br/>ANDREW KAE +<br/>Submitted to the Graduate School of the +<br/><b>University of Massachusetts Amherst in partial ful llment</b><br/>of the requirements for the degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>May 2014 +<br/>Computer Science +</td><td></td><td></td></tr><tr><td>871f5f1114949e3ddb1bca0982086cc806ce84a8</td><td>Discriminative Learning of Apparel Features +<br/>1 Computer Vision Laboratory, D-ITET, ETH Z¨urich, Switzerland +<br/>2 ESAT - PSI / IBBT, K.U. Leuven, Belgium +</td><td>('2173683', 'Rasmus Rothe', 'rasmus rothe')<br/>('2113583', 'Marko Ristin', 'marko ristin')<br/>('1727791', 'Matthias Dantone', 'matthias dantone')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{rrothe,ristin,mdantone,vangool}@vision.ee.ethz.ch +<br/>luc.vangool@esat.kuleuven.be +</td></tr><tr><td>8724fc4d6b91eebb79057a7ce3e9dfffd3b1426f</td><td>Ordered Pooling of Optical Flow Sequences for Action Recognition +<br/>1Data61/CSIRO, 2 Australian Center for Robotic Vision +<br/><b>Australian National University, Canberra, Australia</b><br/>Fatih Porikli1,2,3 +</td><td>('48094509', 'Jue Wang', 'jue wang')<br/>('2691929', 'Anoop Cherian', 'anoop cherian')</td><td>jue.wang@anu.edu.au +<br/>anoop.cherian@anu.edu.au +<br/>fatih.porikli@anu.edu.au +</td></tr><tr><td>87bee0e68dfc86b714f0107860d600fffdaf7996</td><td>Automated 3D Face Reconstruction from Multiple Images +<br/>using Quality Measures +<br/><b>Institute for Vision and Graphics, University of Siegen, Germany</b></td><td>('2712313', 'Marcel Piotraschke', 'marcel piotraschke')<br/>('2880906', 'Volker Blanz', 'volker blanz')</td><td>piotraschke@nt.uni-siegen.de, blanz@informatik.uni-siegen.de +</td></tr><tr><td>87309bdb2b9d1fb8916303e3866eca6e3452c27d</td><td>Kernel Coding: General Formulation and Special Cases +<br/><b>Australian National University, Canberra, ACT 0200, Australia</b><br/>NICTA(cid:63), Locked Bag 8001, Canberra, ACT 2601, Australia +</td><td>('2862871', 'Mathieu Salzmann', 'mathieu salzmann')</td><td></td></tr><tr><td>878169be6e2c87df2d8a1266e9e37de63b524ae7</td><td>CBMM Memo No. 089 +<br/>May 10, 2018 +<br/>Image interpretation above and below the object level +</td><td>('2507298', 'Guy Ben-Yosef', 'guy ben-yosef')<br/>('1743045', 'Shimon Ullman', 'shimon ullman')</td><td></td></tr><tr><td>878301453e3d5cb1a1f7828002ea00f59cbeab06</td><td>Faceness-Net: Face Detection through +<br/>Deep Facial Part Responses +</td><td>('1692609', 'Shuo Yang', 'shuo yang')<br/>('47571885', 'Ping Luo', 'ping luo')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>87e592ee1a7e2d34e6b115da08700a1ae02e9355</td><td>Deep Pictorial Gaze Estimation +<br/>AIT Lab, Department of Computer Science, ETH Zurich +</td><td>('20466488', 'Seonwook Park', 'seonwook park')<br/>('21195502', 'Adrian Spurr', 'adrian spurr')<br/>('2531379', 'Otmar Hilliges', 'otmar hilliges')</td><td>{firstname.lastname}@inf.ethz.ch +</td></tr><tr><td>87147418f863e3d8ff8c97db0b42695a1c28195b</td><td>Attributes for Improved Attributes: A +<br/>Multi-Task Network for Attribute Classification +<br/><b>University of Maryland, College Park</b></td><td>('3351637', 'Emily M. Hand', 'emily m. hand')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>87dd3fd36bccbe1d5f1484ac05f1848b51c6eab5</td><td>SPATIO-TEMPORAL MAXIMUM AVERAGE CORRELATION +<br/>HEIGHT TEMPLATES IN ACTION RECOGNITION AND VIDEO +<br/>SUMMARIZATION +<br/>by +<br/><b>B.A. Earlham College, Richmond Indiana</b><br/><b>M.S. University of Central Florida</b><br/>A dissertation submitted in partial fulfillment of the requirements +<br/>for the degree of Doctor of Philosophy +<br/>in the School of Electrical Engineering and Computer Science +<br/><b>in the College of Engineering and Computer Science</b><br/><b>at the University of Central Florida</b><br/>Orlando, Florida +<br/>Summer Term +<br/>2010 +<br/>Major Professor: Mubarak Shah +</td><td>('35188194', 'MIKEL RODRIGUEZ', 'mikel rodriguez')</td><td></td></tr><tr><td>87bb183d8be0c2b4cfceb9ee158fee4bbf3e19fd</td><td>Craniofacial Image Analysis +</td><td>('1935115', 'Ezgi Mercan', 'ezgi mercan')<br/>('1771661', 'Indriyati Atmosukarto', 'indriyati atmosukarto')<br/>('10423763', 'Jia Wu', 'jia wu')<br/>('1744684', 'Shu Liang', 'shu liang')<br/>('1809809', 'Linda G. Shapiro', 'linda g. shapiro')</td><td></td></tr><tr><td>8006219efb6ab76754616b0e8b7778dcfb46603d</td><td>CONTRIBUTIONSTOLARGE-SCALELEARNINGFORIMAGECLASSIFICATIONZeynepAkataPhDThesisl’´EcoleDoctoraleMath´ematiques,SciencesetTechnologiesdel’Information,InformatiquedeGrenoble </td><td></td><td></td></tr><tr><td>80193dd633513c2d756c3f568ffa0ebc1bb5213e</td><td></td><td></td><td></td></tr><tr><td>808b685d09912cbef4a009e74e10476304b4cccf</td><td>From Understanding to Controlling Privacy +<br/>against Automatic Person Recognition in Social Media +<br/><b>Max Planck Institute for Informatics, Germany</b></td><td>('2390510', 'Seong Joon Oh', 'seong joon oh')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')<br/>('1739548', 'Mario Fritz', 'mario fritz')</td><td>{joon,mfritz,schiele}@mpi-inf.mpg.de +</td></tr><tr><td>804b4c1b553d9d7bae70d55bf8767c603c1a09e3</td><td>978-1-4799-9988-0/16/$31.00 ©2016 IEEE +<br/>1831 +<br/>ICASSP 2016 +</td><td></td><td></td></tr><tr><td>800cbbe16be0f7cb921842d54967c9a94eaa2a65</td><td>MULTIMODAL RECOGNITION OF +<br/>EMOTIONS +</td><td></td><td></td></tr><tr><td>80135ed7e34ac1dcc7f858f880edc699a920bf53</td><td>EFFICIENT ACTION AND EVENT RECOGNITION IN VIDEOS USING +<br/>EXTREME LEARNING MACHINES +<br/>by +<br/>G¨ul Varol +<br/><b>B.S., Computer Engineering, Bo gazi ci University</b><br/><b>Submitted to the Institute for Graduate Studies in</b><br/>Science and Engineering in partial fulfillment of +<br/>the requirements for the degree of +<br/>Master of Science +<br/>Graduate Program in Computer Engineering +<br/><b>Bo gazi ci University</b><br/>2015 +</td><td></td><td></td></tr><tr><td>803c92a3f0815dbf97e30c4ee9450fd005586e1a</td><td>Max-Mahalanobis Linear Discriminant Analysis Networks +</td><td>('19201674', 'Tianyu Pang', 'tianyu pang')</td><td></td></tr><tr><td>80277fb3a8a981933533cf478245f262652a33b5</td><td>Synergy-based Learning of Facial Identity +<br/><b>Institute for Computer Graphics and Vision</b><br/><b>Graz University of Technology, Austria</b></td><td>('1791182', 'Peter M. Roth', 'peter m. roth')<br/>('3628150', 'Horst Bischof', 'horst bischof')</td><td>{koestinger,pmroth,bischof}@icg.tugraz.at +</td></tr><tr><td>80840df0802399838fe5725cce829e1b417d7a2e</td><td>Fast Approximate L∞ Minimization: Speeding Up Robust Regression +<br/><b>School of Computer Science and Technology, Nanjing University of Science and Technology, China</b><br/><b>School of Computer Science, The University of Adelaide, Australia</b></td><td>('2731972', 'Fumin Shen', 'fumin shen')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')<br/>('26065407', 'Rhys Hill', 'rhys hill')<br/>('5546141', 'Anton van den Hengel', 'anton van den hengel')<br/>('3195119', 'Zhenmin Tang', 'zhenmin tang')</td><td></td></tr><tr><td>80c8d143e7f61761f39baec5b6dfb8faeb814be9</td><td>Local Directional Pattern based Fuzzy Co- +<br/>occurrence Matrix Features for Face recognition +<br/>Professor, CSE Dept. +<br/><b>Gokaraju Rangaraju Institute of Engineering and Technology, Hyd</b></td><td>('39121253', 'P Chandra Sekhar Reddy', 'p chandra sekhar reddy')</td><td></td></tr><tr><td>809ea255d144cff780300440d0f22c96e98abd53</td><td>ArcFace: Additive Angular Margin Loss for Deep Face Recognition +<br/><b>Imperial College London</b><br/>UK +<br/>DeepInSight +<br/>China +<br/><b>Imperial College London</b><br/>UK +</td><td>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('3007274', 'Jia Guo', 'jia guo')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>j.deng16@imperial.ac.uk +<br/>guojia@gmail.com +<br/>s.zafeiriou@imperial.ac.uk +</td></tr><tr><td>80345fbb6bb6bcc5ab1a7adcc7979a0262b8a923</td><td>Research Article +<br/>Soft Biometrics for a Socially Assistive Robotic +<br/>Platform +<br/>Open Access +</td><td>('2104853', 'Pierluigi Carcagnì', 'pierluigi carcagnì')<br/>('2417460', 'Dario Cazzato', 'dario cazzato')<br/>('33097940', 'Marco Del Coco', 'marco del coco')<br/>('35438199', 'Pier Luigi Mazzeo', 'pier luigi mazzeo')<br/>('4730472', 'Marco Leo', 'marco leo')<br/>('1741861', 'Cosimo Distante', 'cosimo distante')</td><td></td></tr><tr><td>80a6bb337b8fdc17bffb8038f3b1467d01204375</td><td>Proceedings of the International Conference on Computer and Information Science and Technology +<br/>Ottawa, Ontario, Canada, May 11 – 12, 2015 +<br/>Paper No. 126 +<br/>Subspace LDA Methods for Solving the Small Sample Size +<br/>Problem in Face Recognition +<br/><b></b><br/>101 KwanFu Rd., Sec. 2, Hsinchu, Taiwan +</td><td>('2018515', 'Ching-Ting Huang', 'ching-ting huang')<br/>('1830341', 'Chaur-Chin Chen', 'chaur-chin chen')</td><td>j60626j@gmail.com;cchen@cs.nthu.edu.tw +</td></tr><tr><td>80be8624771104ff4838dcba9629bacfe6b3ea09</td><td>Simultaneous Feature and Dictionary Learning +<br/>for Image Set Based Face Recognition +<br/>1 Advanced Digital Sciences Center, Singapore +<br/><b>Nanyang Technological University, Singapore</b><br/><b>Beijing University of Posts and Telecommunications, Beijing, China</b><br/><b>University of Illinois at Urbana-Champaign, IL USA</b></td><td>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('39209795', 'Gang Wang', 'gang wang')</td><td></td></tr><tr><td>8000c4f278e9af4d087c0d0895fff7012c5e3d78</td><td>Multi-Task Warped Gaussian Process for Personalized Age Estimation +<br/><b>Hong Kong University of Science and Technology</b></td><td>('36233573', 'Yu Zhang', 'yu zhang')</td><td>{zhangyu,dyyeung}@cse.ust.hk +</td></tr><tr><td>80097a879fceff2a9a955bf7613b0d3bfa68dc23</td><td>Active Self-Paced Learning for Cost-Effective and +<br/>Progressive Face Identification +</td><td>('1737218', 'Liang Lin', 'liang lin')<br/>('3170394', 'Keze Wang', 'keze wang')<br/>('1803714', 'Deyu Meng', 'deyu meng')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('36685537', 'Lei Zhang', 'lei zhang')</td><td></td></tr><tr><td>80bd795930837330e3ced199f5b9b75398336b87</td><td>Relative Forest for Attribute Prediction +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>CAS), Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>Graduate University of Chinese Academy of Sciences, Beijing 100049, China</b></td><td>('1688086', 'Shaoxin Li', 'shaoxin li')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{shaoxin.li, shiguang.shan, xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>74de03923a069ffc0fb79e492ee447299401001f</td><td>On Film Character Retrieval in Feature-Length Films +<br/>1 Introduction +<br/>The problem of automatic face recognition (AFR) concerns matching a detected (roughly localized) face +<br/>against a database of known faces with associated identities. This task, although very intuitive to humans +<br/>and despite the vast amounts of research behind it, still poses a significant challenge to computer-based +<br/>methods. For reviews of the literature and commercial state-of-the-art see [5, 31] and [22, 23]. Much AFR +<br/>research has concentrated on the user authentication paradigm (e.g. [2, 8, 19]). In contrast, we consider the +<br/>content-based multimedia retrieval setup: our aim is to retrieve, and rank by confidence, film shots based on +<br/>the presence of specific actors. A query to the system consists of the user choosing the person of interest in +<br/>one or more keyframes. Possible applications include: +<br/>1. DVD browsing: Current DVD technology allows users to quickly jump to the chosen part of a film +<br/>using an on-screen index. However, the available locations are predefined. AFR technology could allow +<br/>the user to rapidly browse scenes by formulating queries based on the presence of specific actors. +<br/>2. Content-based web search: Many web search engines have very popular image search features (e.g. +<br/>http://www.google.co.uk/imghp). Currently, the search is performed based on the keywords +<br/>that appear in picture filenames or in the surrounding web page content. Face recognition can make the +<br/>retrieval much more accurate by focusing on the content of images. +<br/>We proceed from the face detection stage, assuming localized faces. Face detection technology is fairly +<br/>mature and a number of reliable face detectors have been built, see [17, 21, 25, 30]. We use a local imple- +<br/>mentation of the method of Schneiderman and Kanade [25] and consider a face to be correctly detected if +<br/>both eyes and the mouth are visible, see Figure 1. In a typical feature-length film, using every 10th frame, +<br/>we obtain 2000-5000 face detections which result from a cast of 10-20 primary and secondary characters +<br/>(see §3). +<br/>Problem challenges. +<br/>A number of factors other than identity influence the way a face appears in an image. Lighting conditions, +<br/><b>and especially light angle, drastically change the appearance of a face [1]. Facial expressions, including</b><br/>closed or partially closed eyes, also complicate the problem, just as head pose does. Partial occlusions, be +<br/>they artefacts in front of a face or resulting from hair style change, or growing a beard or moustache also +</td><td>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>1 Department of Engineering, University of Cambridge, UK oa214@cam.ac.uk +<br/>2 Department of Engineering, University of Oxford, UK az@robots.ox.ac.uk +</td></tr><tr><td>74f643579949ccd566f2638b85374e7a6857a9fc</td><td>Monogenic Binary Pattern (MBP): A Novel Feature Extraction and +<br/>Representation Model for Face Recognition +<br/><b>Biometric Research Center, The Hong Kong Polytechnic University</b><br/>Different from other face recognition methods, LBP +<br/>methods use local structural information and histogram +<br/>of sub-regions to extract and describe facial features. +<br/>Following LBP, LGBPHS [6] was proposed to use +<br/>Gabor filtering to enhance the facial features and then +<br/>extract the local Gabor binary pattern histogram +<br/>sequence, which improves much LBP’s robustness to +<br/>illumination changes. The Gabor phase was also used +<br/>to improve the recognition rate [7-8], and a typical +<br/>method of this class is the HGPP [8], which captures +<br/>the Global Gabor phase and Local Gabor phase +<br/>variation. Despite the high accuracy, the expense of +<br/>the above mentioned Gabor +<br/>face +<br/>recognition methods is also very expensive: both the +<br/>computational cost and the storage space are high +<br/>because Gabor filtering is usually applied at five +<br/>different scales and along eight different orientations, +<br/>which limits the application of these methods. +<br/>filter based +<br/>is a +<br/>signal +<br/>(HMBP) +<br/>the MBP +<br/>to describe +<br/>two-dimensional +<br/>This paper presents a new local facial feature +<br/>extraction method, namely monogenic binary pattern +<br/>(MBP), based on the theory of monogenic signal +<br/>analysis [9], and then proposes to use the histogram of +<br/>features. +<br/>MBP +<br/>Monogenic +<br/>(2D) +<br/>generalization of the one-dimensional analytic signal, +<br/>through which +<br/>the multi-resolution magnitude, +<br/>orientation and phase of a 2D signal can be estimated. +<br/>The proposed MBP combines monogenic orientation +<br/>and monogenic magnitude information for face feature +<br/>extraction and description. The advantage of MBP +<br/>over other Gabor based methods [4][6][8] is that it has +<br/>much lower time and space complexity but with better +<br/>or comparable performance. This is mainly because +<br/>monogenic signal analysis +<br/>itself a compact +<br/>representation of features with little information loss. +<br/>It does not use steerable filters to create multi- +<br/>orientation features like Gabor filters do. HMBP is the +<br/>sub-region spatial histogram sequence of MBP +<br/>features, which is robust to face image variation of +<br/>is +</td><td>('5828998', 'Meng Yang', 'meng yang')<br/>('36685537', 'Lei Zhang', 'lei zhang')<br/>('40613710', 'Lin Zhang', 'lin zhang')<br/>('1698371', 'David Zhang', 'david zhang')</td><td>E-mail: {csmyang, cslzhang, cslinzhang, csdzhang}@comp.polyu.edu.hk +</td></tr><tr><td>74ce7e5e677a4925489897665c152a352c49d0a2</td><td>SONG ET AL.: SEGMENTATION-GUIDED IMAGE INPAINTING +<br/>SPG-Net: Segmentation Prediction and +<br/>Guidance Network for Image Inpainting +<br/><b>University of Southern California</b><br/>3740 McClintock Ave +<br/>Los Angeles, USA +<br/>2 Baidu Research +<br/>1195 Bordeaux Dr., +<br/>Sunnyvale, USA +</td><td>('3383051', 'Yuhang Song', 'yuhang song')<br/>('1683340', 'Chao Yang', 'chao yang')<br/>('8035191', 'Yeji Shen', 'yeji shen')<br/>('1722767', 'Peng Wang', 'peng wang')<br/>('38592052', 'Qin Huang', 'qin huang')<br/>('9363144', 'C.-C. Jay Kuo', 'c.-c. jay kuo')</td><td>yuhangso@usc.edu +<br/>chaoy@usc.edu +<br/>yejishen@usc.edu +<br/>wangpeng54@baidu.com +<br/>qinhuang@usc.edu +<br/>cckuo@sipi.usc.edu +</td></tr><tr><td>74408cfd748ad5553cba8ab64e5f83da14875ae8</td><td>Facial Expressions Tracking and Recognition: Database Protocols for Systems Validation +<br/>and Evaluation +</td><td></td><td></td></tr><tr><td>747d5fe667519acea1bee3df5cf94d9d6f874f20</td><td></td><td></td><td></td></tr><tr><td>74dbe6e0486e417a108923295c80551b6d759dbe</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 45– No.11, May 2012 +<br/>An HMM based Model for Prediction of Emotional +<br/>Composition of a Facial Expression using both +<br/>Significant and Insignificant Action Units and +<br/>Associated Gender Differences +<br/>Department of Management and Information +<br/>Department of Management and Information +<br/>Systems Science +<br/>1603-1 Kamitomioka, Nagaoka +<br/>Niigata, Japan +<br/>Systems Science +<br/>1603-1 Kamitomioka, Nagaoka +<br/>Niigata, Japan +</td><td>('2931637', 'Suvashis Das', 'suvashis das')<br/>('1808643', 'Koichi Yamada', 'koichi yamada')</td><td></td></tr><tr><td>740e095a65524d569244947f6eea3aefa3cca526</td><td>Towards Human-like Performance Face Detection: A +<br/>Convolutional Neural Network Approach +<br/><b>University of Twente</b><br/>P.O. Box 217, 7500AE Enschede +<br/>The Netherlands +</td><td>('2651432', 'Joshua van Kleef', 'joshua van kleef')</td><td>j.a.vankleef-1@student.utwente.nl +</td></tr><tr><td>74e869bc7c99093a5ff9f8cfc3f533ccf1b135d8</td><td>Context and Subcategories for +<br/>Sliding Window Object Recognition +<br/>CMU-RI-TR-12-17 +<br/>Submitted in partial fulfillment of the +<br/>requirements for the degree of +<br/>Doctor of Philosophy in Robotics +<br/><b>The Robotics Institute</b><br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, Pennsylvania 15213 +<br/>August 2012 +<br/>Thesis Committee +<br/>Martial Hebert, Co-Chair +<br/>Alexei A. Efros, Co-Chair +<br/>Takeo Kanade +<br/><b>Deva Ramanan, University of California at Irvine</b></td><td>('2038685', 'Santosh K. Divvala', 'santosh k. divvala')<br/>('2038685', 'Santosh K. Divvala', 'santosh k. divvala')</td><td></td></tr><tr><td>747c25bff37b96def96dc039cc13f8a7f42dbbc7</td><td>EmoNets: Multimodal deep learning approaches for emotion +<br/>recognition in video +</td><td>('3127597', 'Samira Ebrahimi Kahou', 'samira ebrahimi kahou')<br/>('1748421', 'Vincent Michalski', 'vincent michalski')<br/>('2488222', 'Nicolas Boulanger-Lewandowski', 'nicolas boulanger-lewandowski')<br/>('1923596', 'David Warde-Farley', 'david warde-farley')<br/>('1751762', 'Yoshua Bengio', 'yoshua bengio')</td><td></td></tr><tr><td>741485741734a99e933dd0302f457158c6842adf</td><td> A Novel Automatic Facial Expression +<br/>Recognition Method Based on AAM +<br/><b>State Key Laboratory of Robotics and System, Harbin Institute of Technology, Harbin, China</b></td><td>('1703431', 'Li Wang', 'li wang')<br/>('2677485', 'Ruifeng Li', 'ruifeng li')<br/>('1751643', 'Ke Wang', 'ke wang')</td><td>Email: wangli-hb@163.com, lrf100@ hit.edu.cn, wangke@ hit.edu.cn +</td></tr><tr><td>744fa8062d0ae1a11b79592f0cd3fef133807a03</td><td>Aalborg Universitet +<br/>Deep Pain +<br/>Rodriguez, Pau; Cucurull, Guillem; Gonzàlez, Jordi; M. Gonfaus, Josep ; Nasrollahi, Kamal; +<br/>Moeslund, Thomas B.; Xavier Roca, F. +<br/>Published in: +<br/>I E E E Transactions on Cybernetics +<br/>DOI (link to publication from Publisher): +<br/>10.1109/TCYB.2017.2662199 +<br/>Publication date: +<br/>2017 +<br/>Document Version +<br/>Accepted author manuscript, peer reviewed version +<br/><b>Link to publication from Aalborg University</b><br/>Citation for published version (APA): +<br/>Rodriguez, P., Cucurull, G., Gonzàlez, J., M. Gonfaus, J., Nasrollahi, K., Moeslund, T. B., & Xavier Roca, F. +<br/>(2017). Deep Pain: Exploiting Long Short-Term Memory Networks for Facial Expression Classification. I E E E +<br/>Transactions on Cybernetics, 1-11. DOI: 10.1109/TCYB.2017.2662199 +<br/>General rights +<br/>Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners +<br/>and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. +<br/> ? Users may download and print one copy of any publication from the public portal for the purpose of private study or research. +<br/> ? You may not further distribute the material or use it for any profit-making activity or commercial gain +<br/> ? You may freely distribute the URL identifying the publication in the public portal ? +<br/>Take down policy +<br/>the work immediately and investigate your claim. +<br/>Downloaded from vbn.aau.dk on: marts 22, 2018 +<br/> </td><td></td><td>If you believe that this document breaches copyright please contact us at vbn@aub.aau.dk providing details, and we will remove access to +</td></tr><tr><td>743e582c3e70c6ec07094887ce8dae7248b970ad</td><td>International Journal of Signal Processing, Image Processing and Pattern Recognition +<br/>Vol.8, No.10 (2015), pp.29-38 +<br/>http://dx.doi.org/10.14257/ijsip.2015.8.10.04 +<br/>Face Recognition based on Deep Neural Network +<br/><b>Shandong Women s University</b></td><td>('9094473', 'Li Xinhua', 'li xinhua')<br/>('29742002', 'Yu Qian', 'yu qian')</td><td>lixinhua@sdwu.edu.cn +</td></tr><tr><td>74b0095944c6e29837c208307a67116ebe1231c8</td><td></td><td></td><td></td></tr><tr><td>74156a11c2997517061df5629be78428e1f09cbd</td><td>Cancún Center, Cancún, México, December 4-8, 2016 +<br/>978-1-5090-4846-5/16/$31.00 ©2016 IEEE +<br/>2784 +</td><td></td><td></td></tr><tr><td>748e72af01ba4ee742df65e9c030cacec88ce506</td><td>Discriminative Regions Selection for Facial Expression +<br/>Recognition +<br/><b>MIRACL-FSEG, University of Sfax</b><br/>3018 Sfax, Tunisia +<br/><b>MIRACL-FS, University of Sfax</b><br/>3018 Sfax, Tunisia +</td><td>('2049116', 'Hazar Mliki', 'hazar mliki')<br/>('1749733', 'Mohamed Hammami', 'mohamed hammami')</td><td></td></tr><tr><td>745b42050a68a294e9300228e09b5748d2d20b81</td><td></td><td></td><td></td></tr><tr><td>749d605dd12a4af58de1fae6f5ef5e65eb06540e</td><td>Multi-Task Video Captioning with Video and Entailment Generation +<br/>UNC Chapel Hill +</td><td>('10721120', 'Ramakanth Pasunuru', 'ramakanth pasunuru')<br/>('7736730', 'Mohit Bansal', 'mohit bansal')</td><td>{ram, mbansal}@cs.unc.edu +</td></tr><tr><td>749382d19bfe9fb8d0c5e94d0c9b0a63ab531cb7</td><td>A Modular Framework to Detect and Analyze Faces for +<br/>Audience Measurement Systems +<br/><b>Fraunhofer Institute for Integrated Circuits IIS</b><br/>Department Electronic Imaging +<br/>Am Wolfsmantel 33, 91058 Erlangen, Germany +</td><td>('33046373', 'Andreas Ernst', 'andreas ernst')<br/>('27421829', 'Tobias Ruf', 'tobias ruf')</td><td>{andreas.ernst, tobias.ruf, christian.kueblbeck}@iis.fraunhofer.de +</td></tr><tr><td>74c19438c78a136677a7cb9004c53684a4ae56ff</td><td>RESOUND: Towards Action Recognition +<br/>without Representation Bias +<br/>UC San Diego +</td><td>('48513320', 'Yingwei Li', 'yingwei li')<br/>('47002970', 'Yi Li', 'yi li')<br/>('1699559', 'Nuno Vasconcelos', 'nuno vasconcelos')</td><td>{yil325,yil898,nvasconcelos}@ucsd.edu +</td></tr><tr><td>74618fb4ce8ce0209db85cc6069fe64b1f268ff4</td><td>Rendering and Animating Expressive +<br/>Caricatures +<br/>Mukundan +<br/>*HITLab New Zealand, +<br/><b>University</b><br/>of Canterbury, +<br/>Christchurch, +<br/>New Zealand +<br/>tComputer +<br/>Science +<br/>and Software Engineering +<br/>Email: {mohammad.obaid, +<br/><b>University</b><br/>of Canterbury, +<br/>New Zealand +<br/>non +<br/>stylized +<br/>and control +<br/>on the generated caricature. +<br/>A stroke-based +<br/>of the caricature, +<br/>of facial expressions. +<br/>rendering of caricatures +<br/>from a given face image, with +<br/>the facial appearance +<br/>using quadratic deformation +<br/>rendering (NPR) engine is developed to generate +<br/>that appears to be a sketch of the original +</td><td>('1761180', 'Mohammad Obaid', 'mohammad obaid')<br/>('1684805', 'Mark Billinghurst', 'mark billinghurst')</td><td>mark.billinghurst}@hitlabnz.org, +<br/>mukund@cosc.canterbury.ac.nz +</td></tr><tr><td>74875368649f52f74bfc4355689b85a724c3db47</td><td>Object Detection by Labeling Superpixels +<br/>1National Laboratory of Pattern Recognition, Chinese Academy of Sciences +<br/><b>Institute of Data Science and Technology, Alibaba Group</b><br/><b>Institute of Deep Learning, Baidu Research</b></td><td>('1721677', 'Junjie Yan', 'junjie yan')<br/>('2278628', 'Yinan Yu', 'yinan yu')<br/>('8362374', 'Xiangyu Zhu', 'xiangyu zhu')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td></td></tr><tr><td>7492c611b1df6bce895bee6ba33737e7fc7f60a6</td><td>The 3D Menpo Facial Landmark Tracking Challenge +<br/><b>Imperial College London, UK</b><br/><b>Center for Machine Vision and Signal Analysis, University of Oulu, Finland</b><br/><b>University of Exeter, UK</b></td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('34586458', 'Grigorios G. Chrysos', 'grigorios g. chrysos')<br/>('2931390', 'Anastasios Roussos', 'anastasios roussos')<br/>('31243357', 'Evangelos Ververas', 'evangelos ververas')<br/>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('2814229', 'George Trigeorgis', 'george trigeorgis')</td><td>{s.zafeiriou, g.chrysos}@imperial.ac.uk +</td></tr><tr><td>74eae724ef197f2822fb7f3029c63014625ce1ca</td><td>International Journal of Bio-Science and Bio-Technology +<br/>Vol. 5, No. 2, April, 2013 +<br/>Feature Extraction based on Local Directional Pattern with SVM +<br/>Decision-level Fusion for Facial Expression Recognition +<br/>1Key Laboratory of Education Informalization for Nationalities, Ministry of +<br/><b>Education, Yunnan Normal University, Kunming, China</b><br/><b>College of Information, Yunnan Normal University, Kunming, China</b></td><td>('2535958', 'Juxiang Zhou', 'juxiang zhou')<br/>('3305175', 'Tianwei Xu', 'tianwei xu')<br/>('2411704', 'Jianhou Gan', 'jianhou gan')</td><td>zjuxiang@126.com,xutianwei@ynnu.edu.cn,kmganjh@yahoo.com.cn +</td></tr><tr><td>7480d8739eb7ab97c12c14e75658e5444b852e9f</td><td>NEGREL ET AL.: REVISITED MLBOOST FOR FACE RETRIEVAL +<br/>MLBoost Revisited: A Faster Metric +<br/>Learning Algorithm for Identity-Based Face +<br/>Retrieval +<br/>Frederic Jurie +<br/>Normandie Univ, UNICAEN, +<br/>ENSICAEN, CNRS +<br/>France +</td><td>('2838835', 'Romain Negrel', 'romain negrel')<br/>('2504258', 'Alexis Lechervy', 'alexis lechervy')</td><td>romain.negrel@unicaen.fr +<br/>alexis.lechervy@unicaen.fr +<br/>frederic.jurie@unicaen.fr +</td></tr><tr><td>74ba4ab407b90592ffdf884a20e10006d2223015</td><td>Partial Face Detection in the Mobile Domain +</td><td>('3152615', 'Upal Mahbub', 'upal mahbub')<br/>('40599829', 'Sayantan Sarkar', 'sayantan sarkar')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>7405ed035d1a4b9787b78e5566340a98fe4b63a0</td><td>Self-Expressive Decompositions for +<br/>Matrix Approximation and Clustering +</td><td>('1746363', 'Eva L. Dyer', 'eva l. dyer')<br/>('3318961', 'Raajen Patel', 'raajen patel')<br/>('1746260', 'Richard G. Baraniuk', 'richard g. baraniuk')</td><td></td></tr><tr><td>744db9bd550bf5e109d44c2edabffec28c867b91</td><td>FX e-Makeup for Muscle Based Interaction +<br/>1 Department of Informatics, PUC-Rio, Rio de Janeiro, Brazil +<br/>2 Department of Mechanical Engineering, PUC-Rio, Rio de Janeiro, Brazil +<br/>3 Department of Administration, PUC-Rio, Rio de Janeiro, Brazil +</td><td>('21852164', 'Abel Arrieta', 'abel arrieta')<br/>('38047086', 'Felipe Esteves', 'felipe esteves')<br/>('1805792', 'Hugo Fuks', 'hugo fuks')</td><td>{kvega,hugo}@inf.puc-rio.br +<br/>abel.arrieta@aluno.puc-rio.br +<br/>felipeesteves@aluno.puc-rio.br +</td></tr><tr><td>74325f3d9aea3a810fe4eab8863d1a48c099de11</td><td>Regression-Based Image Alignment +<br/>for General Object Categories +<br/><b>Queensland University of Technology (QUT</b><br/>Brisbane QLD 4000, Australia +<br/><b>Carnegie Mellon University (CMU</b><br/>Pittsburgh PA 15289, USA +</td><td>('2266155', 'Hilton Bristow', 'hilton bristow')<br/>('1820249', 'Simon Lucey', 'simon lucey')</td><td></td></tr><tr><td>744d23991a2c48d146781405e299e9b3cc14b731</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIP.2016.2535284, IEEE +<br/>Transactions on Image Processing +<br/>Aging Face Recognition: A Hierarchical Learning +<br/>Model Based on Local Patterns Selection +</td><td>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('2856494', 'Dihong Gong', 'dihong gong')<br/>('1720243', 'Xuelong Li', 'xuelong li')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td></td></tr><tr><td>1a45ddaf43bcd49d261abb4a27977a952b5fff12</td><td>LDOP: Local Directional Order Pattern for Robust +<br/>Face Retrieval +<br/> +</td><td>('34992579', 'Shiv Ram Dubey', 'shiv ram dubey')<br/>('34356161', 'Snehasis Mukherjee', 'snehasis mukherjee')</td><td></td></tr><tr><td>1a41e5d93f1ef5b23b95b7163f5f9aedbe661394</td><td>Hindawi Publishing Corporation +<br/>e Scientific World Journal +<br/>Volume 2014, Article ID 903160, 9 pages +<br/>http://dx.doi.org/10.1155/2014/903160 +<br/>Research Article +<br/>Alignment-Free and High-Frequency Compensation in +<br/>Face Hallucination +<br/><b>College of Computer Science and Information Technology, Central South University of Forestry and Technology, Hunan 410004, China</b><br/><b>College of Information Science and Engineering, Ritsumeikan University, Shiga 525-8577, Japan</b><br/>Received 25 August 2013; Accepted 21 November 2013; Published 12 February 2014 +<br/>Academic Editors: S. Bourennane and J. Marot +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Face hallucination is one of learning-based super resolution techniques, which is focused on resolution enhancement of facial +<br/>images. Though face hallucination is a powerful and useful technique, some detailed high-frequency components cannot be +<br/>recovered. It also needs accurate alignment between training samples. In this paper, we propose a high-frequency compensation +<br/>framework based on residual images for face hallucination method in order to improve the reconstruction performance. The basic +<br/>idea of proposed framework is to reconstruct or estimate a residual image, which can be used to compensate the high-frequency +<br/>components of the reconstructed high-resolution image. Three approaches based on our proposed framework are proposed. We +<br/>also propose a patch-based alignment-free face hallucination. In the patch-based face hallucination, we first segment facial images +<br/>into overlapping patches and construct training patch pairs. For an input low-resolution (LR) image, the overlapping patches +<br/>are also used to obtain the corresponding high-resolution (HR) patches by face hallucination. The whole HR image can then be +<br/>reconstructed by combining all of the HR patches. Experimental results show that the high-resolution images obtained using our +<br/>proposed approaches can improve the quality of those obtained by conventional face hallucination method even if the training data +<br/>set is unaligned. +<br/>1. Introduction +<br/>There is a high demand for high-resolution (HR) images such +<br/>as video surveillance, remote sensing, and medical imaging +<br/>because high-resolution images can reveal more information +<br/>than low-resolution images. However, it is hard to improve +<br/>the image resolution by replacing sensors because of the +<br/>high cost, hardware physical limits. Super resolution image +<br/>reconstruction (SR) is one promising technique to solve the +<br/>problem [1, 2]. SR can be broadly classified into two families of +<br/>methods: (1) the classical multiframe super resolution [2] and +<br/>(2) the single-frame super resolution, which is also known as +<br/>example-based or learning-based super resolution [3–5]. In +<br/>the classical multiimage SR, the HR image is reconstructed +<br/>by combining subpixel-aligned multiimages (LR images). In +<br/>the learning-based SR, the HR image is reconstructed by +<br/>learning correspondence between low and high-resolution +<br/>image patches from a database. +<br/>Face hallucination is one of learning-based SR techniques +<br/>proposed by Baker and Kanade [1, 6], which is focused on +<br/>resolution enhancement of facial images. To date, a lot of +<br/>algorithms of face hallucination methods have been proposed +<br/>[7–12]. Though face hallucination is a powerful and useful +<br/>technique, some detailed high-frequency components cannot +<br/>be recovered. In this paper, we propose a high-frequency +<br/>compensation framework based on residual images for face +<br/>hallucination method in order to improve the reconstruction +<br/>performance. The basic idea of proposed framework is to +<br/>reconstruct or estimate a residual image, which can be used +<br/>to compensate the high-frequency components of the recon- +<br/>structed high-resolution image. Three approaches based on +<br/>our proposed framework are proposed. We also propose a +<br/>patch-based alignment-free face hallucination method. In the +<br/>patch-based face hallucination, we first segment facial images +<br/>into overlapping patches and construct training patch pairs. +<br/>For an input LR image, the overlapping patches are also used +<br/>to obtain the corresponding HR patches by face hallucination. +<br/>The whole HR image can then be reconstructed by combining +<br/>all of the HR patches. +</td><td>('1699766', 'Yen-Wei Chen', 'yen-wei chen')<br/>('2755407', 'So Sasatani', 'so sasatani')<br/>('1707360', 'Xian-Hua Han', 'xian-hua han')<br/>('1699766', 'Yen-Wei Chen', 'yen-wei chen')</td><td>Correspondence should be addressed to Yen-Wei Chen; chen@is.ritsumei.ac.jp +</td></tr><tr><td>1a65cc5b2abde1754b8c9b1d932a68519bcb1ada</td><td>LU, LIAN, YUILLE: PARSING SEMANTIC PARTS OF CARS +<br/>Parsing Semantic Parts of Cars Using +<br/>Graphical Models and Segment Appearance +<br/>Consistency +<br/>Alan Yuille2 +<br/>1 Department of Electrical Engineering +<br/><b>Tsinghua University</b><br/>2 Department of Statistics +<br/><b>University of California, Los Angeles</b></td><td>('2282045', 'Wenhao Lu', 'wenhao lu')<br/>('5964529', 'Xiaochen Lian', 'xiaochen lian')</td><td>yourslewis@gmail.com +<br/>lianxiaochen@gmail.com +<br/>yuille@stat.ucla.edu +</td></tr><tr><td>1aa766bbd49bac8484e2545c20788d0f86e73ec2</td><td> +<br/>Baseline Face Detection, Head Pose Estimation, and Coarse +<br/>Direction Detection for Facial Data in the SHRP2 Naturalistic +<br/>Driving Study +<br/>J. Paone, D. Bolme, R. Ferrell, Member, IEEE, D. Aykac, and +<br/>T. Karnowski, Member, IEEE +<br/>Oak Ridge National Laboratory, Oak Ridge, TN +</td><td></td><td></td></tr><tr><td>1a849b694f2d68c3536ed849ed78c82e979d64d5</td><td>This is a repository copy of Symmetric Shape Morphing for 3D Face and Head Modelling. +<br/>White Rose Research Online URL for this paper: +<br/>http://eprints.whiterose.ac.uk/131760/ +<br/>Version: Accepted Version +<br/>Proceedings Paper: +<br/>Dai, Hang, Pears, Nicholas Edwin orcid.org/0000-0001-9513-5634, Smith, William Alfred +<br/>Peter orcid.org/0000-0002-6047-0413 et al. (1 more author) (2018) Symmetric Shape +<br/>Morphing for 3D Face and Head Modelling. In: The 13th IEEE Conference on Automatic +<br/>Face and Gesture Recognition. IEEE . +<br/>Reuse +<br/>Items deposited in White Rose Research Online are protected by copyright, with all rights reserved unless +<br/>indicated otherwise. They may be downloaded and/or printed for private study, or other acts as permitted by +<br/>national copyright laws. The publisher or other rights holders may allow further reproduction and re-use of +<br/>the full text version. This is indicated by the licence information on the White Rose Research Online record +<br/>for the item. +<br/>Takedown +<br/>If you consider content in White Rose Research Online to be in breach of UK law, please notify us by +<br/>https://eprints.whiterose.ac.uk/ +</td><td></td><td>emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request. +<br/>eprints@whiterose.ac.uk +</td></tr><tr><td>1a46d3a9bc1e4aff0ccac6403b49a13c8a89fc1d</td><td>Online Robust Image Alignment via Iterative Convex Optimization +<br/>Center for Data Analytics & Biomedical Informatics, Computer & Information Science Department, +<br/><b>Temple University, Philadelphia, PA 19122, USA</b><br/><b>School of Information and Control Engineering, Nanjing University of Information Science and Technology, Nanjing, 210044, China</b><br/><b>Purdue University, West Lafayette, IN 47907, USA</b></td><td>('36578908', 'Yi Wu', 'yi wu')<br/>('39274045', 'Bin Shen', 'bin shen')<br/>('1805398', 'Haibin Ling', 'haibin ling')</td><td>fwuyi,hblingg@temple.edu, bshen@purdue.edu +</td></tr><tr><td>1a878e4667fe55170252e3f41d38ddf85c87fcaf</td><td>Discriminative Machine Learning with Structure +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2010-4 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2010/EECS-2010-4.html +<br/>January 12, 2010 +</td><td>('1685481', 'Simon Lacoste-Julien', 'simon lacoste-julien')</td><td></td></tr><tr><td>1a41831a3d7b0e0df688fb6d4f861176cef97136</td><td><b>massachusetts institute of technology artificial intelligence laboratory</b><br/>A Biological Model of Object +<br/>Recognition with Feature Learning +<br/>AI Technical Report 2003-009 +<br/>CBCL Memo 227 +<br/>June 2003 +<br/>© 2 0 0 3 m a s s a c h u s e t t s i n s t i t u t e o f +<br/>t e c h n o l o g y, c a m b r i d g e , m a 0 2 1 3 9 u s a — w w w. a i . m i t . e d u +</td><td>('1848733', 'Jennifer Louie', 'jennifer louie')</td><td>@ MIT +</td></tr><tr><td>1ac2882559a4ff552a1a9956ebeadb035cb6df5b</td><td>How much training data for facial action unit detection? +<br/><b>University of Pittsburgh, Pittsburgh, PA, USA</b><br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, USA</b></td><td>('36185909', 'Jeffrey M. Girard', 'jeffrey m. girard')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')<br/>('1820249', 'Simon Lucey', 'simon lucey')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td></td></tr><tr><td>1a7a17c4f97c68d68fbeefee1751d349b83eb14a</td><td>Iterative Hessian sketch: Fast and accurate solution +<br/>approximation for constrained least-squares +<br/>1Department of Electrical Engineering and Computer Science +<br/>2Department of Statistics +<br/><b>University of California, Berkeley</b><br/>November 4, 2014 +</td><td>('3173667', 'Mert Pilanci', 'mert pilanci')<br/>('1721860', 'Martin J. Wainwright', 'martin j. wainwright')</td><td>{mert, wainwrig}@berkeley.edu +</td></tr><tr><td>1aef6f7d2e3565f29125a4871cd60c4d86c48361</td><td>Natural Language Video Description using +<br/>Deep Recurrent Neural Networks +<br/><b>University of Texas at Austin</b><br/>Doctoral Dissertation Proposal +</td><td>('1811430', 'Subhashini Venugopalan', 'subhashini venugopalan')<br/>('1797655', 'Raymond J. Mooney', 'raymond j. mooney')</td><td>vsub@cs.utexas.edu +</td></tr><tr><td>1a6c3c37c2e62b21ebc0f3533686dde4d0103b3f</td><td>International Journal of Linguistics and Computational Applications (IJLCA) ISSN 2394-6385 (Print) +<br/>Volume 4, Issue 1, January – March 2017 ISSN 2394-6393 (Online) +<br/> Implementation of Partial Face Recognition +<br/>using Directional Binary Code +<br/>N.Pavithra #1, A.Sivapriya*2, K.Hemalatha*3 , D.Lakshmi*4 +<br/><b>Final Year, PanimalarInstitute of Technology</b><br/><b>PanimalarInstitute of Technology, Tamilnadu, India</b><br/>in +<br/>faith +<br/>is proposed. It +<br/>face alignment and +</td><td></td><td></td></tr><tr><td>1a167e10fe57f6d6eff0bb9e45c94924d9347a3e</td><td>Boosting VLAD with Double Assignment using +<br/>Deep Features for Action Recognition in Videos +<br/><b>University of Trento, Italy</b><br/>Tuan A. Nguyen +<br/><b>University of Tokyo, Japan</b><br/><b>University of Tokyo, Japan</b><br/><b>University Politehnica of Bucharest, Romania</b><br/><b>University of Trento, Italy</b></td><td>('3429470', 'Ionut C. Duta', 'ionut c. duta')<br/>('1712839', 'Kiyoharu Aizawa', 'kiyoharu aizawa')<br/>('1796198', 'Bogdan Ionescu', 'bogdan ionescu')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')</td><td>ionutcosmin.duta@unitn.it +<br/>t nguyen@hal.t.u-tokyo.ac.jp +<br/>aizawa@hal.t.u-tokyo.ac.jp +<br/>bionescu@imag.pub.ro +<br/>niculae.sebe@unitn.it +</td></tr><tr><td>1a3eee980a2252bb092666cf15dd1301fa84860e</td><td>PCA GAUSSIANIZATION FOR IMAGE PROCESSING +<br/>Image Processing Laboratory (IPL), Universitat de Val`encia +<br/>Catedr´atico A. Escardino - 46980 Paterna, Val`encia, Spain +</td><td>('2732577', 'Valero Laparra', 'valero laparra')<br/>('1684246', 'Gustavo Camps-Valls', 'gustavo camps-valls')</td><td>{lapeva,gcamps,jmalo}@uv.es +</td></tr><tr><td>1a140d9265df8cf50a3cd69074db7e20dc060d14</td><td>Face Parts Localization Using +<br/>Structured-Output Regression Forests +<br/><b>School of EECS, Queen Mary University of London</b></td><td>('2966679', 'Heng Yang', 'heng yang')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')</td><td>{heng.yang,i.patras}@eecs.qmul.ac.uk +</td></tr><tr><td>1a85956154c170daf7f15f32f29281269028ff69</td><td>Active Pictorial Structures +<br/><b>Imperial College London</b><br/>180 Queens Gate, SW7 2AZ, London, U.K. +</td><td>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('2575567', 'Joan Alabort-i-Medina', 'joan alabort-i-medina')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>{e.antonakos, ja310, s.zafeiriou}@imperial.ac.uk +</td></tr><tr><td>1a031378cf1d2b9088a200d9715d87db8a1bf041</td><td>Workshop track - ICLR 2018 +<br/>DEEP DICTIONARY LEARNING: SYNERGIZING RE- +<br/>CONSTRUCTION AND CLASSIFICATION +</td><td>('3362896', 'Shahin Mahdizadehaghdam', 'shahin mahdizadehaghdam')<br/>('1733181', 'Ashkan Panahi', 'ashkan panahi')<br/>('1769928', 'Hamid Krim', 'hamid krim')</td><td>{smahdiz,apanahi,ahk}@ncsu.edu & liyi.dai.civ@mail.mil +</td></tr><tr><td>1afd481036d57320bf52d784a22dcb07b1ca95e2</td><td>The Computer Journal Advance Access published December 6, 2012 +<br/><b>The Author 2012. Published by Oxford University Press on behalf of The British Computer Society. All rights reserved</b><br/>doi:10.1093/comjnl/bxs146 +<br/>Automated Content Metadata Extraction +<br/>Services Based on MPEG Standards +<br/>D.C. Gibbon∗, Z. Liu, A. Basso and B. Shahraray +<br/>AT&T Labs Research, Middletown, NJ, USA +<br/>This paper is concerned with the generation, acquisition, standardized representation and transport +<br/>of video metadata. The use of MPEG standards in the design and development of interoperable +<br/>media architectures and web services is discussed. A high-level discussion of several algorithms +<br/>for metadata extraction is presented. Some architectural and algorithmic issues encountered when +<br/>designing services for real-time processing of video streams, as opposed to traditional offline media +<br/>processing, are addressed. A prototype real-time video analysis system for generating MPEG-7 +<br/>Audiovisual Description Profile from MPEG-2 transport stream encapsulated video is presented. +<br/>Such a capability can enable a range of new services such as content-based personalization of live +<br/>broadcasts given that the MPEG-7 based data models fit in well with specifications for advanced +<br/>television services such as TV-Anytime andAlliance for Telecommunications Industry Solutions IPTV +<br/>Interoperability Forum. +<br/>Keywords: MPEG-7; MPEG-21; audiovisual description profile; video processing; automated metadata +<br/>extraction; video metadata, real-time media processing +<br/>Received 1 March 2012; revised 11 September 2012; accepted 9 October 2012 +<br/>Handling editor: Marios Angelides +<br/>1. +<br/>INTRODUCTION +<br/>Content descriptors have gained considerable prominence +<br/>in the content ecosystem in the last decade. This growing +<br/>significance stems from the fact that rich metadata promotes +<br/>user engagement, enables fine-grained access to content and +<br/>allows more intelligent and targeted access to content. +<br/>Effective utilization of content descriptors involves three +<br/>basic steps, namely generation, representation and transport. +<br/>In traditional broadcasting, +<br/>the generation of the content +<br/>descriptions has been a manual process in which individuals +<br/>would access the content and would index it according to +<br/>specific rules (i.e. annotation guides). While in the past this +<br/>was a viable option due to the limited amount of available +<br/>content, with the large volumes of content that are generated +<br/>today (e.g. YouTube uploads have currently surpassed 1 h of +<br/>video every second), manual indexing is no longer a viable +<br/>option. Research in multimedia content analysis has generated a +<br/>variety of algorithms for content feature extraction in the visual, +<br/>text, music and speech domains. Such algorithms provide +<br/>descriptions with different levels of confidence and are often +<br/>combined to improve their accuracy and descriptive power. +<br/>Despite the enormous progress that has been made in this area, +<br/>content description generation is not yet sufficiently advanced +<br/>to be fully automated for all applications and types of content. +<br/>However, for a subset of content types and certain applications, +<br/>the current state of the art in automated content processing has +<br/>proven sufficient. +<br/>Another important consideration in effective and widespread +<br/>utilization of content metadata is the adoption of appropriate +<br/>representations for the metadata. Historically, the represen- +<br/>tation of content metadata has been specialized to specific +<br/>representation and service needs (i.e. the asset distribution +<br/>interface from CableLabs for traditional paid video on demand +<br/>services). Recently, in the context of MPEG, a standardization +<br/>effort has been undertaken to create more general represen- +<br/>tations of content descriptors that are independent of any +<br/>particular application and to enable interoperability among +<br/>metadata generation systems and applications. +<br/>Finally, for a certain class of applications and services, +<br/>real-time delivery or transport of metadata is critical, but +<br/>is an area that is still in its infancy. For example, today’s +<br/>systems for delivering television electronic program guide +<br/>(EPG) information make efficient use of multicast delivery, +<br/>but the data are largely static (the data may only change +<br/>The Computer Journal, 2012 +</td><td></td><td>For Permissions, please email: journals.permissions@oup.com +<br/>Corresponding author: dcg@research.att.com +</td></tr><tr><td>1a9337d70a87d0e30966ecd1d7a9b0bbc7be161f</td><td></td><td></td><td></td></tr><tr><td>1a4b6ee6cd846ef5e3030a6ae59f026e5f50eda6</td><td>Deep Learning for Video Classification and Captioning +<br/><b>Fudan University, 2Microsoft Research Asia, 3University of Maryland</b><br/>1. Introduction +<br/>Today’s digital contents are inherently multimedia: text, audio, image, +<br/>video and etc. Video, in particular, becomes a new way of communication +<br/>between Internet users with the proliferation of sensor-rich mobile devices. +<br/>Accelerated by the tremendous increase in Internet bandwidth and storage +<br/>space, video data has been generated, published and spread explosively, be- +<br/>coming an indispensable part of today’s big data. This has encouraged the +<br/>development of advanced techniques for a broad range of video understand- +<br/>ing applications. A fundamental issue that underlies the success of these +<br/>technological advances is the understanding of video contents. Recent ad- +<br/>vances in deep learning in image [41, 68, 17, 50] and speech [21, 27] domain +<br/>have encouraged techniques to learn robust video feature representations to +<br/>effectively exploit abundant multimodal clues in video data. +<br/>In this paper, we focus on reviewing two lines of research aiming to stimu- +<br/>late the comprehension of videos with deep learning: video classification and +<br/>video captioning. While video classification concentrates on automatically +<br/>labeling video clips based on their semantic contents like human actions or +<br/>complex events, video captioning attempts to generate a complete and nat- +<br/>ural sentence, enriching the single label as in video classification, to capture +<br/>the most informative dynamics in videos. +<br/>There have been several efforts surveying literatures on video content +<br/>understanding. Most of the approaches surveyed in these works adopted +<br/>hand-crafted features coupled with typical machine learning pipelines for +<br/>action recognition and event detection [1, 88, 61, 35]. In contrast, this paper +<br/>focuses on discussing state-of-the-art deep learning techniques not only for +<br/>video classification but also video captioning. As deep learning for video +<br/>analysis is an emerging and vibrant field, we hope this paper could help +<br/>stimulate future research along the line. +</td><td>('3099139', 'Zuxuan Wu', 'zuxuan wu')<br/>('2053452', 'Ting Yao', 'ting yao')<br/>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')</td><td>zxwu@cs.umd.edu, tiyao@microsoft.com, {ygj, yanweifu}@fudan.edu.cn +</td></tr><tr><td>1a9a192b700c080c7887e5862c1ec578012f9ed1</td><td>IEEE TRANSACTIONS ON SYSTEM, MAN AND CYBERNETICS, PART B +<br/>Discriminant Subspace Analysis for Face +<br/>Recognition with Small Number of Training +<br/>Samples +</td><td>('1844328', 'Hui Kong', 'hui kong')<br/>('1786811', 'Xuchun Li', 'xuchun li')<br/>('1752714', 'Matthew Turk', 'matthew turk')<br/>('1708413', 'Chandra Kambhamettu', 'chandra kambhamettu')</td><td></td></tr><tr><td>1af52c853ff1d0ddb8265727c1d70d81b4f9b3a9</td><td>ARTICLE +<br/>International Journal of Advanced Robotic Systems +<br/>Face Recognition Under Illumination +<br/>Variation Using Shadow Compensation +<br/>and Pixel Selection +<br/>Regular Paper +<br/><b>Dankook University, 126 Jukjeon-dong, Suji-gu, Yongin-si, Gyeonggi-do, Korea</b><br/>Received 14 Jun 2012; Accepted 31 Aug 2012 +<br/>DOI: 10.5772/52939 +<br/>© 2012 Choi; licensee InTech. This is an open access article distributed under the terms of the Creative +<br/>Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, +<br/>distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>to other +<br/>features +<br/>for +<br/>face +<br/>retinal or +<br/>is similar +<br/>to +<br/>the +<br/>fingerprint, +<br/>image +<br/>taken with +<br/>it widely applicable +<br/>illumination variation. By using +</td><td>('1737997', 'Sang-Il Choi', 'sang-il choi')</td><td>* Corresponding author E-mail: choisi@dankook.ac.kr +</td></tr><tr><td>1a8ccc23ed73db64748e31c61c69fe23c48a2bb1</td><td>Extensive Facial Landmark Localization +<br/>with Coarse-to-fine Convolutional Network Cascade +<br/>Megvii Inc. +</td><td>('1848243', 'Erjin Zhou', 'erjin zhou')</td><td>{zej,fhq,czm,jyn,yq}@megvii.com +</td></tr><tr><td>1a40092b493c6b8840257ab7f96051d1a4dbfeb2</td><td>Sports Videos in the Wild (SVW): A Video Dataset for Sports Analysis +<br/><b>Michigan State University, East Lansing, MI, USA</b><br/>2 TechSmith Corporation, Okemos, MI, USA +</td><td>('2941187', 'Seyed Morteza Safdarnejad', 'seyed morteza safdarnejad')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('1938832', 'Lalita Udpa', 'lalita udpa')<br/>('40467330', 'Brooks Andrus', 'brooks andrus')<br/>('1678721', 'John Wood', 'john wood')<br/>('37008125', 'Dean Craven', 'dean craven')</td><td></td></tr><tr><td>1ad97cce5fa8e9c2e001f53f6f3202bddcefba22</td><td>Grassmann Averages for Scalable Robust PCA +<br/>DIKU and MPIs T¨ubingen∗ +<br/>Denmark and Germany +<br/>DTU Compute∗ +<br/>Lyngby, Denmark +</td><td>('1808965', 'Aasa Feragen', 'aasa feragen')<br/>('2142792', 'Søren Hauberg', 'søren hauberg')</td><td>aasa@diku.dk +<br/>sohau@dtu.dk +</td></tr><tr><td>1a1118cd4339553ad0544a0a131512aee50cf7de</td><td></td><td></td><td></td></tr><tr><td>1a6c9ef99bf0ab9835a91fe5f1760d98a0606243</td><td>ConceptMap: +<br/>Mining Noisy Web Data for Concept Learning +<br/><b>Bilkent University, 06800 Cankaya, Turkey</b></td><td>('2540074', 'Eren Golge', 'eren golge')</td><td></td></tr><tr><td>1afdedba774f6689eb07e048056f7844c9083be9</td><td>Markov Random Field Structures for Facial Action Unit Intensity Estimation +<br/>∗Department of Computing +<br/><b>Imperial College London</b><br/>180 Queen’s Gate +<br/>London, UK +<br/>†EEMCS +<br/><b>University of Twente</b><br/>7522 NB Enschede +<br/>Netherlands +</td><td>('3007548', 'Georgia Sandbach', 'georgia sandbach')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{gls09,s.zafeiriou,m.pantic}@imperial.ac.uk +</td></tr><tr><td>1a2b3fa1b933042687eb3d27ea0a3fcb67b66b43</td><td>WANG AND MORI: MAX-MARGIN LATENT DIRICHLET ALLOCATION +<br/>Max-Margin Latent Dirichlet Allocation for +<br/>Image Classification and Annotation +<br/><b>University</b><br/>of Illinois at Urbana Champaign +<br/>School of Computing Science, Simon +<br/><b>Fraser University</b></td><td>('40457160', 'Yang Wang', 'yang wang')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td>yangwang@uiuc.edu +<br/>mori@cs.sfu.ca +</td></tr><tr><td>1a7a2221fed183b6431e29a014539e45d95f0804</td><td>Person Identification Using Text and Image Data +<br/>David S. Bolme, J. Ross Beveridge and Adele E. Howe +<br/>Computer Science Department +<br/>Colorado State Univeristy +<br/>Fort Collins, Colorado 80523 +</td><td></td><td>[bolme,ross,howe]@cs.colostate.edu +</td></tr><tr><td>1a5b39a4b29afc5d2a3cd49087ae23c6838eca2b</td><td>Competitive Game Designs for Improving the Cost +<br/>Effectiveness of Crowdsourcing +<br/><b>L3S Research Center, Hannover, Germany</b></td><td>('2993225', 'Markus Rokicki', 'markus rokicki')<br/>('3257370', 'Sergiu Chelaru', 'sergiu chelaru')<br/>('2553718', 'Sergej Zerr', 'sergej zerr')<br/>('1745880', 'Stefan Siersdorfer', 'stefan siersdorfer')</td><td>{rokicki,chelaru,siersdorfer,zerr}@L3S.de +</td></tr><tr><td>2878b06f3c416c98496aad6fc2ddf68d2de5b8f6</td><td>Available online at www.sciencedirect.com +<br/>Computer Vision and Image Understanding 110 (2008) 91–101 +<br/>www.elsevier.com/locate/cviu +<br/>Two-stage optimal component analysis +<br/><b>Florida State University, Tallahassee, FL 32306, USA</b><br/><b>Florida State University, Tallahassee, FL 32306, USA</b><br/><b>c School of Computational Science, Florida State University, Tallahassee, FL 32306, USA</b><br/>Received 26 September 2006; accepted 30 April 2007 +<br/>Available online 8 June 2007 +</td><td>('2207859', 'Yiming Wu', 'yiming wu')<br/>('1800002', 'Xiuwen Liu', 'xiuwen liu')<br/>('2436294', 'Washington Mio', 'washington mio')</td><td></td></tr><tr><td>287795991fad3c61d6058352879c7d7ae1fdd2b6</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 66– No.8, March 2013 +<br/>Biometrics Security: Facial Marks Detection from the +<br/>Low Quality Images +<br/>and facial marks are detected using LoG with morphological +<br/>operator. This method though was not enough to detect the +<br/>facial marks from the low quality images [7]. But, facial +<br/>marks have been used to speed up the retrieval process in +<br/>order to differentiate the human faces [15]. +<br/><b>B.S.Abdur Rahman University B.S.Abdur Rahman University</b><br/> Dept. Of Information Technology Dept. Of Computer Science & Engineering +<br/> Chennai, India Chennai, India +<br/> +</td><td>('9401261', 'Ziaul Haque Choudhury', 'ziaul haque choudhury')</td><td></td></tr><tr><td>28a900a07c7cbce6b6297e4030be3229e094a950</td><td>382 The International Arab Journal of Information Technology, Vol. 9, No. 4, July 2012 +<br/>Local Directional Pattern Variance (LDPv): A +<br/>Robust Feature Descriptor for Facial +<br/>Expression Recognition +<br/><b>Kyung Hee University, South Korea</b></td><td>('3182680', 'Taskeed Jabid', 'taskeed jabid')<br/>('1685505', 'Oksam Chae', 'oksam chae')</td><td></td></tr><tr><td>282503fa0285240ef42b5b4c74ae0590fe169211</td><td>Feeding Hand-Crafted Features for Enhancing the Performance of +<br/>Convolutional Neural Networks +<br/><b>Seoul National University</b><br/>Seoul Nat’l Univ. +<br/><b>Seoul National University</b></td><td>('35453923', 'Sepidehsadat Hosseini', 'sepidehsadat hosseini')<br/>('32193683', 'Seok Hee Lee', 'seok hee lee')<br/>('1707645', 'Nam Ik Cho', 'nam ik cho')</td><td>sepid@ispl.snu.ac.kr +<br/>seokheel@snu.ac.kr +<br/>nicho@snu.ac.kr +</td></tr><tr><td>28e0ed749ebe7eb778cb13853c1456cb6817a166</td><td></td><td></td><td></td></tr><tr><td>28b9d92baea72ec665c54d9d32743cf7bc0912a7</td><td></td><td></td><td></td></tr><tr><td>283d226e346ac3e7685dd9a4ba8ae55ee4f2fe43</td><td>BAYESIAN DATA ASSOCIATION FOR TEMPORAL SCENE +<br/>UNDERSTANDING +<br/>by +<br/>A Dissertation Submitted to the Faculty of the +<br/>DEPARTMENT OF COMPUTER SCIENCE +<br/>In Partial Fulfillment of the Requirements +<br/>For the Degree of +<br/>DOCTOR OF PHILOSOHPY +<br/><b>In the Graduate College</b><br/><b>THE UNIVERSITY OF ARIZONA</b><br/>2013 +</td><td>('10399726', 'Ernesto Brau Avila', 'ernesto brau avila')</td><td></td></tr><tr><td>28d7029cfb73bcb4ad1997f3779c183972a406b4</td><td>Discriminative Nonlinear Analysis Operator +<br/>Learning: When Cosparse Model Meets Image +<br/>Classification +</td><td>('2833510', 'Zaidao Wen', 'zaidao wen')<br/>('1940528', 'Biao Hou', 'biao hou')<br/>('1734497', 'Licheng Jiao', 'licheng jiao')</td><td></td></tr><tr><td>280d59fa99ead5929ebcde85407bba34b1fcfb59</td><td>978-1-4799-9988-0/16/$31.00 ©2016 IEEE +<br/>2662 +<br/>ICASSP 2016 +</td><td></td><td></td></tr><tr><td>28f5138d63e4acafca49a94ae1dc44f7e9d84827</td><td>Journal of Machine Learning Research xx (2012) xx-xx +<br/>Submitted xx/xx; Published xx/xx +<br/>MahNMF: Manhattan Non-negative Matrix Factorization +<br/>Center for Quantum Computation and Intelligent Systems +<br/>Faculty of Engineering and Information Technology +<br/><b>University of Technology, Sydney</b><br/>Sydney, NSW 2007, Australia +<br/>Center for Quantum Computation and Intelligent Systems +<br/>Faculty of Engineering and Information Technology +<br/><b>University of Technology, Sydney</b><br/>Sydney, NSW 2007, Australia +<br/>School of Computer Science +<br/><b>National University of Defense Technology</b><br/>Changsha, Hunan 410073, China +<br/>Centre for Computational Statistics and Machine Learning (CSML) +<br/>Department of Computer Science +<br/><b>University College London</b><br/>Gower Street, London WC1E 6BT, United Kingdom +<br/>Editor: xx +</td><td>('2067095', 'Naiyang Guan', 'naiyang guan')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('1764542', 'Zhigang Luo', 'zhigang luo')<br/>('1792322', 'John Shawe-Taylor', 'john shawe-taylor')</td><td>Guan.Naiyang@uts.edu.au +<br/>dacheng.tao@uts.edu.au +<br/>zgluo@nudt.edu.cn +<br/>J.Shawe-Taylor@cs.ucl.ac.uk +</td></tr><tr><td>28e1668d7b61ce21bf306009a62b06593f1819e3</td><td>RESEARCH ARTICLE +<br/>Validation of the Amsterdam Dynamic Facial +<br/>Expression Set – Bath Intensity Variations +<br/>(ADFES-BIV): A Set of Videos Expressing Low, +<br/>Intermediate, and High Intensity Emotions +<br/><b>University of Bath, Bath, United Kingdom</b><br/>☯ These authors contributed equally to this work. +</td><td>('7249951', 'Tanja S. H. Wingenbach', 'tanja s. h. wingenbach')<br/>('2708124', 'Chris Ashwin', 'chris ashwin')<br/>('39455300', 'Mark Brosnan', 'mark brosnan')</td><td>* tshw20@bath.ac.uk +</td></tr><tr><td>28cd46a078e8fad370b1aba34762a874374513a5</td><td>CVPAPER.CHALLENGE IN 2016, JULY 2017 +<br/>cvpaper.challenge in 2016: Futuristic Computer +<br/>Vision through 1,600 Papers Survey +</td><td>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')<br/>('1713046', 'Yun He', 'yun he')<br/>('9935341', 'Shunya Ueta', 'shunya ueta')<br/>('5014206', 'Teppei Suzuki', 'teppei suzuki')<br/>('3408038', 'Kaori Abe', 'kaori abe')<br/>('2554424', 'Asako Kanezaki', 'asako kanezaki')<br/>('22219521', 'Toshiyuki Yabe', 'toshiyuki yabe')<br/>('10800402', 'Yoshihiro Kanehara', 'yoshihiro kanehara')<br/>('22174281', 'Hiroya Yatsuyanagi', 'hiroya yatsuyanagi')<br/>('1692565', 'Shinya Maruyama', 'shinya maruyama')<br/>('3217653', 'Masataka Fuchida', 'masataka fuchida')<br/>('2642022', 'Yudai Miyashita', 'yudai miyashita')<br/>('34935749', 'Kazushige Okayasu', 'kazushige okayasu')<br/>('20505300', 'Yuta Matsuzaki', 'yuta matsuzaki')</td><td></td></tr><tr><td>286adff6eff2f53e84fe5b4d4eb25837b46cae23</td><td>Single-Image Depth Perception in the Wild +<br/><b>University of Michigan, Ann Arbor</b></td><td>('1732404', 'Weifeng Chen', 'weifeng chen')<br/>('8342699', 'Jia Deng', 'jia deng')<br/>('2097755', 'Zhao Fu', 'zhao fu')<br/>('2500067', 'Dawei Yang', 'dawei yang')</td><td>{wfchen,zhaofu,ydawei,jiadeng}@umich.edu +</td></tr><tr><td>286812ade95e6f1543193918e14ba84e5f8e852e</td><td>DOU, WU, SHAH, KAKADIARIS: 3D FACE RECONSTRUCTION FROM 2D LANDMARKS +<br/>Robust 3D Face Shape Reconstruction from +<br/>Single Images via Two-Fold Coupled +<br/>Structure Learning +<br/>Computational Biomedicine Lab +<br/>Department of Computer Science +<br/><b>University of Houston</b><br/>Houston, TX, USA +</td><td>('39634395', 'Pengfei Dou', 'pengfei dou')<br/>('2461369', 'Yuhang Wu', 'yuhang wu')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td>bensondou@gmail.com +<br/>yuhang@cbl.uh.edu +<br/>sshah@central.uh.edu +<br/>ioannisk@uh.edu +</td></tr><tr><td>282a3ee79a08486f0619caf0ada210f5c3572367</td><td></td><td></td><td></td></tr><tr><td>288dbc40c027af002298b38954d648fddd4e2fd3</td><td></td><td></td><td></td></tr><tr><td>28f311b16e4fe4cc0ff6560aae3bbd0cb6782966</td><td>Learning Language from Perceptual Context +<br/>Department of Computer Science +<br/><b>University of Texas at Austin</b><br/>David L. Chen +<br/>Austin, TX 78712 +<br/>Doctoral Dissertation Proposal +</td><td>('1797655', 'Raymond J. Mooney', 'raymond j. mooney')</td><td>dlcc@cs.utexas.edu +</td></tr><tr><td>28312c3a47c1be3a67365700744d3d6665b86f22</td><td></td><td></td><td></td></tr><tr><td>28d06fd508d6f14cd15f251518b36da17909b79e</td><td>What’s in a Name? First Names as Facial Attributes +<br/><b>Stanford University</b><br/><b>Cornell University</b><br/><b>Stanford University</b></td><td>('2896700', 'Huizhong Chen', 'huizhong chen')<br/>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')<br/>('1739786', 'Bernd Girod', 'bernd girod')</td><td>hchen2@stanford.edu +<br/>andrew.c.gallagher@cornell.edu +<br/>bgirod@stanford.edu +</td></tr><tr><td>28b5b5f20ad584e560cd9fb4d81b0a22279b2e7b</td><td>A New Fuzzy Stacked Generalization Technique +<br/>and Analysis of its Performance +</td><td>('2159942', 'Mete Ozay', 'mete ozay')<br/>('7158165', 'Fatos T. Yarman Vural', 'fatos t. yarman vural')</td><td></td></tr><tr><td>281486d172cf0c78d348ce7d977a82ff763efccd</td><td>Mining a Deep And-OR Object Semantics from Web Images via Cost-Sensitive +<br/>Question-Answer-Based Active Annotations +<br/><b>Shanghai Jiao Tong University</b><br/><b>University of California, Los Angeles</b><br/><b>cid:107)Chongqing University of Posts and Telecommunications</b></td><td>('22063226', 'Quanshi Zhang', 'quanshi zhang')<br/>('39092098', 'Ying Nian Wu', 'ying nian wu')<br/>('3133970', 'Song-Chun Zhu', 'song-chun zhu')</td><td></td></tr><tr><td>288964068cd87d97a98b8bc927d6e0d2349458a2</td><td>Mean-Variance Loss for Deep Age Estimation from a Face +<br/>1Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b><br/>3CAS Center for Excellence in Brain Science and Intelligence Technology +</td><td>('34393045', 'Hu Han', 'hu han')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>hongyu.pan@vipl.ict.ac.cn, {hanhu,sgshan,xlchen}@ict.ac.cn +</td></tr><tr><td>28bc378a6b76142df8762cd3f80f737ca2b79208</td><td>Understanding Objects in Detail with Fine-grained Attributes +<br/>Ross Girshick5 +<br/>David Weiss7 +</td><td>('1687524', 'Andrea Vedaldi', 'andrea vedaldi')<br/>('2585200', 'Siddharth Mahendran', 'siddharth mahendran')<br/>('2381485', 'Stavros Tsogkas', 'stavros tsogkas')<br/>('35208858', 'Subhransu Maji', 'subhransu maji')<br/>('1776374', 'Juho Kannala', 'juho kannala')<br/>('2827962', 'Esa Rahtu', 'esa rahtu')<br/>('1758219', 'Matthew B. Blaschko', 'matthew b. blaschko')<br/>('1685978', 'Ben Taskar', 'ben taskar')<br/>('2362960', 'Naomi Saphra', 'naomi saphra')<br/>('2920190', 'Sammy Mohamed', 'sammy mohamed')<br/>('2010660', 'Iasonas Kokkinos', 'iasonas kokkinos')<br/>('34838386', 'Karen Simonyan', 'karen simonyan')</td><td></td></tr><tr><td>287900f41dd880802aa57f602e4094a8a9e5ae56</td><td></td><td></td><td></td></tr><tr><td>28c0cb56e7f97046d6f3463378d084e9ea90a89a</td><td>Automatic Face Recognition for Film Character Retrieval in Feature-Length +<br/>Films +<br/>Ognjen Arandjelovi´c +<br/><b>University of Oxford, UK</b></td><td>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>E-mail: oa214@cam.ac.uk,az@robots.ox.ac.uk +</td></tr><tr><td>28be652db01273289499bc6e56379ca0237506c0</td><td>FaLRR: A Fast Low Rank Representation Solver +<br/><b>School of Computer Engineering, Nanyang Technological University, Singapore</b><br/><b>of Engineering and Information Technology, University of Technology, Sydney, Australia</b><br/>‡Centre for Quantum Computation & Intelligent Systems and the Faculty +<br/>In this paper, we develop a fast solver of low rank representation (LRR) [3] +<br/>called FaLRR, which achieves order-of-magnitude speedup over existing +<br/>LRR solvers, and is theoretically guaranteed to obtain a global optimum. +<br/>LRR [3] has shown promising performance for various computer vision +<br/>applications such as face clustering. Let X = [x1, . . . ,xn] ∈ Rd×n be a set +<br/>of data samples drawn from a union of several subspaces, where d is the +<br/>feature dimension and n is the total number of data samples. LRR seeks +<br/>a low-rank data representation matrix Z ∈ Rn×n such that X can be self- +<br/>expressed (i.e., X = XZ) when the data is clean. Considering that input +<br/>data may contain outliers (i.e., some columns of X are corrupted), the LRR +<br/>problem can be formulated as, +<br/>(cid:107)Z(cid:107)∗ + λ(cid:107)E(cid:107)2,1 +<br/>min +<br/>Z,E +<br/>s.t. X = XZ + E, +<br/>(1) +<br/>where λ is a tradeoff parameter and E ∈ Rd×n denotes the representation +<br/>error. The nuclear norm based term (cid:107)Z(cid:107)∗ acts as an approximation of the +<br/>rank regularizer, and the (cid:96)2,1 norm based term (cid:107)E(cid:107)2,1 encourages E to be +<br/>column-sparse. +<br/>Regarding optimization, several algorithms [2, 3, 4] were proposed to +<br/>exactly solve LRR. Moreover, to efficiently obtain an approximated solution +<br/>of LRR, a distributed framework [5] was developed. However, the existing +<br/>algorithms are usually based on the original formulation in (1) or a similar +<br/>variant [4], which are two-variable problems with regard to the original data +<br/>matrix. In this paper, we develop a fast LRR solver named FaLRR, which +<br/>is based on a new reformulation of LRR as an optimization problem with +<br/>regard to factorized data (which is obtained by skinny SVD on the original +<br/>data matrix). +<br/>Reformulation. Specifically, we study a more general formulation of +<br/>LRR as follows, +<br/>min +<br/>Z∈Rn×m,E∈Rd×m +<br/>(cid:107)Z(cid:107)∗ + λ(cid:107)E(cid:107)2,1 +<br/>s.t. XD = XZ + E +<br/>(2) +<br/>rUr = V(cid:48) +<br/>which includes (1) as a special case. Let r denote the rank of X. More- +<br/>over, let us factorize X via the skinny singular value decomposition (SVD): +<br/>X = UrSrV(cid:48) +<br/>r, where Ur ∈ Rd×r and Vr ∈ Rn×r are two column-wise orthog- +<br/>onal matrices that satisfy U(cid:48) +<br/>rVr = Ir, Sr ∈ Rr×r is a diagonal matrix +<br/>defined as Sr = diag([σ1, . . . ,σr](cid:48)), in which {σi}r +<br/>i=1 are the r positive sin- +<br/>gular values of X sorted in descending order. Based on the definitions above, +<br/>we present the reformulation by the following theorem: +<br/>Theorem 1 Let W∗ denote an optimal solution of the following problem, +<br/>(3) +<br/>Then, {Z∗,E∗}, defined as Z∗ = VrW∗ and E∗ = XD− XVrW∗, is an op- +<br/>timal solution of the problem in (2). In particular, (cid:107)Z∗(cid:107)∗ = (cid:107)W∗(cid:107)∗ and +<br/>(cid:107)E∗(cid:107)2,1 = (cid:107)Sr(V(cid:48) +<br/>rD−W∗)(cid:107)2,1 always hold, implying that the two problems +<br/>in (2) and (3) have equal optimal objective values. +<br/>(cid:107)W(cid:107)∗ + λ(cid:107)Sr(V(cid:48) +<br/>rD− W)(cid:107)2,1 . +<br/>min +<br/>W∈Rr×m +<br/>Optimization. In terms of optimization, we rewrite the problem in (3) +<br/>as follows by introducing another variable Q ∈ Rr×m: +<br/>min +<br/>W,Q∈Rr×m +<br/>(cid:107)W(cid:107)∗ + λ(cid:107)SrQ(cid:107)2,1 +<br/>s.t. W + Q = V(cid:48) +<br/>rD, +<br/>(4) +<br/>and develop an efficient algorithm based on the alternating direction method +<br/>(ADM) [1, 2], in which both resultant subproblems can be solved exactly. +<br/>The corresponding augmented Lagrangian [1] w.r.t. (4) is +<br/>Lρ (W,Q,L) +<br/>= (cid:107)W(cid:107)∗ + λ(cid:107)SrQ(cid:107)2,1 +(cid:10)L,V(cid:48) +<br/>rD− W− Q(cid:11) + +<br/>(cid:107)V(cid:48) +<br/>rD− W− Q(cid:107)2 +<br/>F , +</td><td>('2518469', 'Shijie Xiao', 'shijie xiao')<br/>('12135788', 'Wen Li', 'wen li')<br/>('38188040', 'Dong Xu', 'dong xu')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td></td></tr><tr><td>28bcf31f794dc27f73eb248e5a1b2c3294b3ec9d</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 96– No.13, June 2014 +<br/>Improved Combination of LBP plus LFDA for Facial +<br/>Expression Recognition using SRC +<br/>Research Scholar, CSE Department, +<br/><b>Government College of Engineering, Aurangabad</b><br/>human +<br/>facial +<br/>expression +<br/>recognition +</td><td></td><td></td></tr><tr><td>2836d68c86f29bb87537ea6066d508fde838ad71</td><td>Personalized Age Progression with Aging Dictionary +<br/><b>School of Computer Science and Engineering, Nanjing University of Science and Technology</b><br/><b>National University of Singapore</b><br/>Figure 1. A personalized aging face by the proposed method. The personalized aging face contains the aging layer (e.g., +<br/>wrinkles) and the personalized layer (e.g., mole). The former can be seen as the corresponding face in a linear combination +<br/>of the aging patterns, while the latter is invariant in the aging process. For better view, please see ×3 original color PDF. +</td><td>('2287686', 'Xiangbo Shu', 'xiangbo shu')<br/>('8053308', 'Jinhui Tang', 'jinhui tang')<br/>('2356867', 'Hanjiang Lai', 'hanjiang lai')<br/>('1776665', 'Luoqi Liu', 'luoqi liu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>{shuxb104,laihanj}@gmail.com, jinhuitang@njust.edu.cn, {liuluoqi, eleyans}@nus.edu.sg +</td></tr><tr><td>28de411a5b3eb8411e7bcb0003c426aa91f33e97</td><td> Volume 4, Issue 4, April 2014 ISSN: 2277 128X +<br/>International Journal of Advanced Research in +<br/> Computer Science and Software Engineering +<br/> Research Paper +<br/> Available online at: www.ijarcsse.com +<br/>Emotion Detection Using Facial Expressions -A Review +<br/> +<br/>Department of computer science and Application +<br/> M Tech Student +<br/> Department of computer science and Application +<br/> Assistant professor +<br/><b>Kurukshetra University, Kurukshetra</b><br/><b>Kurukshetra University, Kurukshetra</b><br/> Haryana (India) +<br/> +<br/>Haryana (India) +</td><td>('2234813', 'Jyoti Rani', 'jyoti rani')<br/>('39608299', 'Kanwal Garg', 'kanwal garg')</td><td></td></tr><tr><td>28b26597a7237f9ea6a9255cde4e17ee18122904</td><td>Cerebral Cortex September 2015;25:2876–2882 +<br/>doi:10.1093/cercor/bhu083 +<br/>Advance Access publication April 25, 2014 +<br/>Network Interactions Explain Sensitivity to Dynamic Faces in the Superior Temporal Sulcus +<br/>1MRC Cognition and Brain Sciences Unit, Cambridge CB2 7EF, UK and 2Wellcome Centre for Imaging Neuroscience, +<br/><b>University College London, 12 Queen Square, London WC1N 3BG, UK</b><br/>The superior temporal sulcus (STS) in the human and monkey is sen- +<br/>sitive to the motion of complex forms such as facial and bodily +<br/>actions. We used functional magnetic resonance imaging (fMRI) to +<br/>explore network-level explanations for how the form and motion +<br/>information in dynamic facial expressions might be combined in the +<br/>human STS. Ventral occipitotemporal areas selective for facial form +<br/>were localized in occipital and fusiform face areas (OFA and FFA), +<br/>and motion sensitivity was localized in the more dorsal temporal +<br/>area V5. We then tested various connectivity models that modeled +<br/>communication between the ventral form and dorsal motion path- +<br/>ways. We show that facial form information modulated transmission +<br/>of motion information from V5 to the STS, and that this face- +<br/>selective modulation likely originated in OFA. This finding shows that +<br/>form-selective motion sensitivity in the STS can be explained in +<br/>terms of modulation of gain control on information flow in the motion +<br/>pathway, and provides a substantial constraint for theories of the +<br/>perception of faces and biological motion. +<br/>Keywords: biological motion, dynamic causal modeling, face perception, +<br/>functional magnetic resonance imaging, superior temporal sulcus +<br/>Introduction +<br/>Humans and other animals effortlessly recognize facial iden- +<br/>tities and actions such as emotional expressions even when +<br/>faces continuously move. Brain representations of dynamic +<br/>faces may be manifested as greater responses in the superior +<br/>temporal sulcus (STS) to facial motion than motion of nonface +<br/>objects (Pitcher et al. 2011), suggesting localized representa- +<br/>tions that combine information about motion and facial form. +<br/>This finding relates to a considerable literature on “biological +<br/>motion,” which studies how the complex forms of bodily actions +<br/>are perceived from only the motion of light points fixed to limb +<br/>joints, with form-related texture cues removed (Johansson 1973). +<br/>Perception of such stimuli has been repeatedly associated with +<br/>the human posterior STS (Vaina et al. 2001; Vaina and Gross +<br/>2004; Giese and Poggio 2003; Hein and Knight 2008; Jastorff +<br/>and Orban 2009) with similar results observed in potentially cor- +<br/>responding areas of the macaque STS (Oram and Perrett 1994; +<br/>Jastorff et al. 2012). The STS has been described as integrating +<br/>form and motion information (Vaina et al. 2001; Giese and +<br/>Poggio 2003), containing neurons that code for conjunctions of +<br/>certain forms and movements (Oram and Perrett 1996). Never- +<br/>theless, the mechanisms by which STS neurons come to be sensi- +<br/>tive to the motion of some forms, but not others, remains a +<br/>matter of speculation (Giese and Poggio 2003). +<br/>We propose that network interactions can provide a mech- +<br/>anistic explanation for STS sensitivity to motion that is selective +<br/>to certain forms, in this case, faces. Specifically, STS responses +<br/>to dynamic faces could result from communicative interactions +<br/>between pathways sensitive to motion and facial form. Such in- +<br/>teractions can occur when one pathway modulates or “gates” +<br/>the ability of the other pathway to transmit information to the +<br/>STS. Using functional magnetic resonance imaging (fMRI), we +<br/>localized face-selective motion sensitivity in the STS of the +<br/>human and then used causal connectivity analyses to model +<br/>how these STS responses are influenced by areas sensitive to +<br/>motion and areas selective to facial form. We localized ventral +<br/>occipital and fusiform face areas (OFA and FFA) (Kanwisher +<br/>et al. 1997), which selectively respond to facial form versus +<br/>other objects (Calder and Young 2005; Calder 2011). We also +<br/>localized motion sensitivity to faces and nonfaces in the more +<br/>dorsal temporal hMT+/V5 complex (hereafter, V5). Together, +<br/>these areas provide ventral and dorsal pathways to the STS. +<br/>The ventral pathway transmits facial form information, via OFA +<br/>and FFA, and the dorsal pathway transmits motion informa- +<br/>tion, via V5. We then compared combinations of bilinear and +<br/>nonlinear dynamic causal models (Friston et al. 2003) to iden- +<br/>tify connectivity models that optimally explain how interac- +<br/>tions between these form and motion pathways could generate +<br/>STS responses to dynamic faces. We found that information +<br/>about facial form, most likely originating in the OFA, gates the +<br/>transmission of information about motion from V5 to the STS. +<br/>Thus, integrated facial form and motion information in the STS +<br/>can arise due to network interactions, where form and motion +<br/>pathways play distinct roles. +<br/>Materials and Methods +<br/>Participants +<br/>fMRI data were collected from 18 healthy, right-handed participants +<br/>(over 18 years, 13 females) with normal or corrected-to-normal vision. +<br/>Experimental procedures were approved by the Cambridge Psych- +<br/>ology Research Ethics Committee. +<br/>Imaging Acquisition +<br/>A 3T Siemens Tim Trio MRI scanner with a 32-channel head coil was +<br/>used for data acquisition. We collected a structural T1-weighted MPRAGE +<br/>image (1-mm isotropic voxels). Functional data consisted of whole-brain +<br/>T2*-weighted echo-planar imaging volumes with 32 oblique axial slices +<br/>that were 3.5 mm thick, in-plane 64 × 64 matrix with resolution of 3 × 3 +<br/>mm, TR 2 s, TE 30 ms, flip angle 78°. We discarded the first 5 “dummy” +<br/>volumes to ensure magnetic equilibration. +<br/>Experimental Design +<br/>The experiment used a block design with 2 runs (229 scans per run), +<br/>which were collected as the localizer for another experiment (Furl, +<br/>Henson, et al. 2013). Note that the dynamic causal modeling (DCM) +<br/>analyses reported in Furl, Henson et al. (2013) used independent data +<br/>(from separate runs using different stimuli) to address a different phe- +<br/>nomenon than considered here. All blocks were 11 s, comprised +<br/><b>The Author 2014. Published by Oxford University Press</b><br/>This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted +<br/>reuse, distribution, and reproduction in any medium, provided the original work is properly cited. +</td><td>('3162581', 'Nicholas Furl', 'nicholas furl')<br/>('1690599', 'Richard N. Henson', 'richard n. henson')<br/>('1737497', 'Karl J. Friston', 'karl j. friston')<br/>('2825775', 'Andrew J. Calder', 'andrew j. calder')<br/>('3162581', 'Nicholas Furl', 'nicholas furl')</td><td>UK. E-mail: nick.furl@mrc-cbu.cam.ac.uk +</td></tr><tr><td>28fe6e785b32afdcd2c366c9240a661091b850cf</td><td>International Journal of Applied Information Systems (IJAIS) – ISSN : 2249-0868 +<br/>Foundation of Computer Science FCS, New York, USA +<br/>Volume 10 – No.7, March 2016 – www.ijais.org +<br/>Facial Expression Recognition using Patch based Gabor +<br/>Features +<br/>Electronics & Telecommunication Engg +<br/>Electronics & Telecommunication Engg +<br/><b>St. Francis Institute of Technology</b><br/><b>St. Francis Institute of Technology</b><br/>Department +<br/>Mumbai, India +<br/>Department +<br/>Mumbai, India +</td><td>('40187425', 'Vaqar Ansari', 'vaqar ansari')<br/>('9390824', 'Anju Chandran', 'anju chandran')</td><td></td></tr><tr><td>28c9198d30447ffe9c96176805c1cd81615d98c8</td><td>rsos.royalsocietypublishing.org +<br/>Research +<br/>Cite this article: Saunders TJ, Taylor AH, +<br/>Atkinson QD. 2016 No evidence that a range of +<br/>artificial monitoring cues influence online +<br/>donations to charity in an MTurk sample. +<br/>R. Soc. open sci. 3: 150710. +<br/>http://dx.doi.org/10.1098/rsos.150710 +<br/>Received: 22 December 2015 +<br/>Accepted: 13 September 2016 +<br/>Subject Category: +<br/>Psychology and cognitive neuroscience +<br/>Subject Areas: +<br/>behaviour/psychology/evolution +<br/>Keywords: +<br/>prosociality, eye images, charity donation, +<br/>reputation, online behaviour +<br/>Author for correspondence: +<br/>Quentin D. Atkinson +<br/>No evidence that a range of +<br/>artificial monitoring cues +<br/>influence online donations +<br/>to charity in an MTurk +<br/>sample +<br/>Timothy J. Saunders, Alex H. Taylor and +<br/>Quentin D. Atkinson +<br/><b>School of Psychology, University of Auckland, Auckland, New Zealand</b><br/>AHT, 0000-0003-3492-7667 +<br/>Monitoring cues, such as an image of a face or pair of +<br/>eyes, have been found to increase prosocial behaviour in +<br/>several studies. However, other studies have found little +<br/>or no support for this effect. Here, we examined whether +<br/>monitoring cues affect online donations to charity while +<br/>manipulating the emotion displayed, the number of watchers +<br/>and the cue type. We also include as statistical controls a +<br/>range of likely covariates of prosocial behaviour. Using the +<br/>crowdsourcing Internet marketplace, Amazon Mechanical Turk +<br/>(MTurk), 1535 participants completed our survey and were +<br/>given the opportunity to donate to charity while being shown +<br/>an image prime. None of the monitoring primes we tested +<br/>had a significant effect on charitable giving. By contrast, the +<br/>control variables of culture, age, sex and previous charity +<br/>giving frequency did predict donations. This work supports +<br/>the importance of cultural differences and enduring individual +<br/>differences in prosocial behaviour and shows that a range of +<br/>artificial monitoring cues do not reliably boost online charity +<br/>donation on MTurk. +<br/>Introduction +<br/>1. +<br/>Humans care deeply about their reputations [1]. If we know +<br/>our choices will be made public, we act more prosocially [2–6]. +<br/>Recent work has shown that simple but evolutionarily significant +<br/>artificial monitoring cues, such as an image of a pair of eyes, +<br/>can promote cooperation [7–22]. While an image alone cannot +<br/>monitor behaviour, the evolutionary legacy hypothesis holds that +<br/>humans possess an evolved proximate mechanism that causes us +<br/>to react to monitoring cues as if our reputations are at stake [9]. +<br/>Work using a range of economic games has shown that people act +<br/>2016 The Authors. Published by the Royal Society under the terms of the Creative Commons +<br/>Attribution License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted +<br/>use, provided the original author and source are credited. +</td><td></td><td>e-mail: q.atkinson@auckland.ac.nz +</td></tr><tr><td>28d4e027c7e90b51b7d8908fce68128d1964668a</td><td></td><td></td><td></td></tr><tr><td>2866cbeb25551257683cf28f33d829932be651fe</td><td>In Proceedings of the 2018 IEEE International Conference on Image Processing (ICIP) +<br/>The final publication is available at: http://dx.doi.org/10.1109/ICIP.2018.8451026 +<br/>A TWO-STEP LEARNING METHOD FOR DETECTING LANDMARKS +<br/>ON FACES FROM DIFFERENT DOMAINS +<br/>Erickson R. Nascimento +<br/>Universidade Federal de Minas Gerais (UFMG), Brazil +</td><td>('2749017', 'Bruna Vieira Frade', 'bruna vieira frade')</td><td>{brunafrade, erickson}@dcc.ufmg.br +</td></tr><tr><td>28d99dc2d673d62118658f8375b414e5192eac6f</td><td>Using Ranking-CNN for Age Estimation +<br/>1Department of Computer Science +<br/>2Department of Mathematics +<br/>3Research & Innovation Center +<br/><b>Wayne State University</b><br/><b>Wayne State University</b><br/>Ford Motor Company +</td><td>('15841224', 'Shixing Chen', 'shixing chen')<br/>('28887876', 'Jialiang Le', 'jialiang le')</td><td>{schen, czhang, mdong}@wayne.edu +<br/>{jle1, mrao}@ford.com +</td></tr><tr><td>280bc9751593897091015aaf2cab39805768b463</td><td>U.U.Tariq et al. / Carpathian Journal of Electronic and Computer Engineering 6/1 (2013) 8-15 8 +<br/>________________________________________________________________________________________________________ +<br/>Gender Perception From Faces Using Boosted LBPH +<br/>(Local Binary Patten Histograms) +<br/><b>COMSATS Institute of Information Technology</b><br/>Department of Electrical Engineering +<br/>Abbottabad, Pakistan +<br/> +</td><td></td><td>Umair_tariq29@yahoo.com +</td></tr><tr><td>28aa89b2c827e5dd65969a5930a0520fdd4a3dc7</td><td></td><td></td><td></td></tr><tr><td>28b061b5c7f88f48ca5839bc8f1c1bdb1e6adc68</td><td>Predicting User Annoyance Using Visual Attributes +<br/>Virginia Tech +<br/>Goibibo +<br/>Virginia Tech +<br/>Virginia Tech +</td><td>('1755657', 'Gordon Christie', 'gordon christie')<br/>('2076800', 'Amar Parkash', 'amar parkash')<br/>('3051209', 'Ujwal Krothapalli', 'ujwal krothapalli')<br/>('1713589', 'Devi Parikh', 'devi parikh')</td><td>gordonac@vt.edu +<br/>amar08007@iiitd.ac.in +<br/>ujjwal@vt.edu +<br/>parikh@vt.edu +</td></tr><tr><td>288d2704205d9ca68660b9f3a8fda17e18329c13</td><td>Studying Very Low Resolution Recognition Using Deep Networks +<br/><b>Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA</b></td><td>('2969311', 'Zhangyang Wang', 'zhangyang wang')<br/>('3307026', 'Shiyu Chang', 'shiyu chang')<br/>('2680237', 'Yingzhen Yang', 'yingzhen yang')<br/>('1771885', 'Ding Liu', 'ding liu')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>{zwang119, chang87, yyang58, dingliu2, t-huang1}@illinois.edu +</td></tr><tr><td>17b46e2dad927836c689d6787ddb3387c6159ece</td><td>GeoFaceExplorer: Exploring the Geo-Dependence of +<br/>Facial Attributes +<br/><b>University of Kentucky</b><br/>UNC Charlotte +<br/>UNC Charlotte +<br/><b>University of Kentucky</b></td><td>('2121759', 'Connor Greenwell', 'connor greenwell')<br/>('1690110', 'Richard Souvenir', 'richard souvenir')<br/>('1715594', 'Scott Spurlock', 'scott spurlock')<br/>('1990750', 'Nathan Jacobs', 'nathan jacobs')</td><td>csgr222@uky.edu +<br/>souvenir@uncc.edu +<br/>sspurloc@uncc.edu +<br/>jacobs@cs.uky.edu +</td></tr><tr><td>17a85799c59c13f07d4b4d7cf9d7c7986475d01c</td><td>ADVERTIMENT. La consulta d’aquesta tesi queda condicionada a l’acceptació de les següents +<br/>condicions d'ús: La difusió d’aquesta tesi per mitjà del servei TDX (www.tesisenxarxa.net) ha +<br/>estat autoritzada pels titulars dels drets de propietat intel·lectual únicament per a usos privats +<br/>emmarcats en activitats d’investigació i docència. No s’autoritza la seva reproducció amb finalitats +<br/>de lucre ni la seva difusió i posada a disposició des d’un lloc aliè al servei TDX. No s’autoritza la +<br/>presentació del seu contingut en una finestra o marc aliè a TDX (framing). Aquesta reserva de +<br/>drets afecta tant al resum de presentació de la tesi com als seus continguts. En la utilització o cita +<br/>de parts de la tesi és obligat indicar el nom de la persona autora. +<br/>ADVERTENCIA. La consulta de esta tesis queda condicionada a la aceptación de las siguientes +<br/>condiciones de uso: La difusión de esta tesis por medio del servicio TDR (www.tesisenred.net) ha +<br/>sido autorizada por los titulares de los derechos de propiedad intelectual únicamente para usos +<br/>privados enmarcados en actividades de investigación y docencia. No se autoriza su reproducción +<br/>con finalidades de lucro ni su difusión y puesta a disposición desde un sitio ajeno al servicio TDR. +<br/>No se autoriza la presentación de su contenido en una ventana o marco ajeno a TDR (framing). +<br/>Esta reserva de derechos afecta tanto al resumen de presentación de la tesis como a sus +<br/>contenidos. En la utilización o cita de partes de la tesis es obligado indicar el nombre de la +<br/>persona autora. +<br/>WARNING. On having consulted this thesis you’re accepting the following use conditions: +<br/>Spreading this thesis by the TDX (www.tesisenxarxa.net) service has been authorized by the +<br/>titular of the intellectual property rights only for private uses placed in investigation and teaching +<br/>activities. Reproduction with lucrative aims is not authorized neither its spreading and availability +<br/>from a site foreign to the TDX service. Introducing its content in a window or frame foreign to the +<br/>TDX service is not authorized (framing). This rights affect to the presentation summary of the +<br/>thesis as well as to its contents. In the using or citation of parts of the thesis it’s obliged to indicate +<br/>the name of the author +</td><td></td><td></td></tr><tr><td>1768909f779869c0e83d53f6c91764f41c338ab5</td><td>A Large-Scale Car Dataset for Fine-Grained Categorization and Verification +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Key Lab of CVPR, Shenzhen Institutes of Advanced Technology</b><br/>Chinese Academy of Sciences, Shenzhen, China +</td><td>('2889075', 'Linjie Yang', 'linjie yang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1693209', 'Ping Luo', 'ping luo')</td><td>{yl012,pluo,ccloy,xtang}@ie.cuhk.edu.hk +</td></tr><tr><td>171ca25bc2cdfc79cad63933bcdd420d35a541ab</td><td>Calibration-Free Gaze Estimation Using Human Gaze Patterns +<br/><b>University of Amsterdam</b><br/>Amsterdam, The Netherlands +</td><td>('1765602', 'Fares Alnajar', 'fares alnajar')<br/>('1695527', 'Theo Gevers', 'theo gevers')<br/>('9301018', 'Roberto Valenti', 'roberto valenti')<br/>('1682828', 'Sennay Ghebreab', 'sennay ghebreab')</td><td>{f.alnajar,th.gevers,r.valenti,s.ghebreab}@uva.nl +</td></tr><tr><td>176bd61cc843d0ed6aa5af83c22e3feb13b89fe1</td><td>14 +<br/>Investigating Spontaneous Facial Action +<br/>Recognition through +<br/>AAM Representations of the Face +<br/><b>Carnegie Mellon University</b><br/>USA +<br/>1. Introduction +<br/>The Facial Action Coding System (FACS) [Ekman et al., 2002] is the leading method for +<br/>measuring facial movement in behavioral science. FACS has been successfully applied, but +<br/>not limited to, identifying the differences between simulated and genuine pain, differences +<br/>betweenwhen people are telling the truth versus lying, and differences between suicidal and +<br/>non-suicidal patients [Ekman and Rosenberg, 2005]. Successfully recognizing facial actions +<br/>is recognized as one of the “major” hurdles to overcome, for successful automated +<br/>expression recognition. +<br/>How one should represent the face for effective action unit recognition is the main topic of +<br/>interest in this chapter. This interest is motivated by the plethora of work in existence in +<br/>other areas of face analysis, such as face recognition [Zhao et al., 2003], that demonstrate the +<br/>benefit of representation when performing recognition tasks. It is well understood in the +<br/>field of statistical pattern recognition [Duda et al., 2001] given a fixed classifier and training +<br/>set that how one represents a pattern can greatly effect recognition performance. The face +<br/>can be represented in a myriad of ways. Much work in facial action recognition has centered +<br/>solely on the appearance (i.e., pixel values) of the face given quite a basic alignment (e.g., +<br/>eyes and nose). In our work we investigate the employment of the Active Appearance +<br/>Model (AAM) framework [Cootes et al., 2001, Matthews and Baker, 2004] in order to derive +<br/>effective representations for facial action recognition. Some of the representations we will be +<br/>employing can be seen in Figure 1. +<br/>Experiments in this chapter are run across two action unit databases. The Cohn- Kanade +<br/>FACS-Coded Facial Expression Database [Kanade et al., 2000] is employed to investigate the +<br/>effect of face representation on posed facial action unit recognition. Posed facial actions are +<br/>those that have been elicited by asking subjects to deliberately make specific facial actions or +<br/>expressions. Facial actions are typically recorded under controlled circumstances that +<br/>include full-face frontal view, good lighting, constrained head movement and selectivity in +<br/>terms of the type and magnitude of facial actions. Almost all work in automatic facial +<br/>expression analysis has used posed image data and the Cohn-Kanade database may be the +<br/>database most widely used [Tian et al., 2005]. The RU-FACS Spontaneous Expression +<br/>Database is employed to investigate how these same representations affect spontaneous facial +<br/>action unit recognition. Spontaneous facial actions are representative of “real-world” facial +<br/>Source: Face Recognition, Book edited by: Kresimir Delac and Mislav Grgic, ISBN 978-3-902613-03-5, pp.558, I-Tech, Vienna, Austria, June 2007 +</td><td>('1820249', 'Simon Lucey', 'simon lucey')<br/>('2640279', 'Ahmed Bilal Ashraf', 'ahmed bilal ashraf')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td></td></tr><tr><td>17d01f34dfe2136b404e8d7f59cebfb467b72b26</td><td>Riemannian Similarity Learning +<br/><b>Bioinformatics Institute, A*STAR, Singapore</b><br/><b>School of Computing, National University of Singapore, Singapore</b></td><td>('39466179', 'Li Cheng', 'li cheng')</td><td>chengli@bii.a-star.edu.sg +</td></tr><tr><td>176f26a6a8e04567ea71677b99e9818f8a8819d0</td><td>MEG: Multi-Expert Gender classification from +<br/>face images in a demographics-balanced dataset +</td><td>('1763890', 'Maria De Marsico', 'maria de marsico')<br/>('1795333', 'Michele Nappi', 'michele nappi')<br/>('1772512', 'Daniel Riccio', 'daniel riccio')</td><td>1Universidad de Las Palmas de Gran Canaria, Spain. Email: mcastrillon@siani.es +<br/>2Sapienza University of Rome, Italy. Email: demarsico@di.uniroma1.it +<br/>3University of Salerno, Fisciano (SA), Italy. Email: mnappi@unisa.it +<br/>4University of Naples Federico II, Italy, Email: daniel.riccio@unina.it +</td></tr><tr><td>17cf838720f7892dbe567129dcf3f7a982e0b56e</td><td>Global-Local Face Upsampling Network +<br/><b>Mitsubishi Electric Research Labs (MERL), Cambridge, MA, USA</b></td><td>('2577513', 'Oncel Tuzel', 'oncel tuzel')<br/>('2066068', 'Yuichi Taguchi', 'yuichi taguchi')<br/>('2387467', 'John R. Hershey', 'john r. hershey')</td><td></td></tr><tr><td>17035089959a14fe644ab1d3b160586c67327db2</td><td></td><td></td><td></td></tr><tr><td>17370f848801871deeed22af152489e39b6e1454</td><td>UNDERSAMPLED FACE RECOGNITION WITH ONE-PASS DICTIONARY LEARNING +<br/><b>Research Center for Information Technology Innovation, Academia Sinica, Taipei, Taiwan</b></td><td>('2017922', 'Chia-Po Wei', 'chia-po wei')<br/>('2733735', 'Yu-Chiang Frank Wang', 'yu-chiang frank wang')</td><td>{cpwei, ycwang}@citi.sinica.edu.tw +</td></tr><tr><td>178a82e3a0541fa75c6a11350be5bded133a59fd</td><td>Techset Composition Ltd, Salisbury +<br/>Doc: +<br/>{IEE}BMT/Articles/Pagination/BMT20140045.3d +<br/>www.ietdl.org +<br/>Received on 15th July 2014 +<br/>Revised on 17th September 2014 +<br/>Accepted on 23rd September 2014 +<br/>doi: 10.1049/iet-bmt.2014.0045 +<br/>ISSN 2047-4938 +<br/>BioHDD: a dataset for studying biometric +<br/>identification on heavily degraded data +<br/><b>IT Instituto de Telecomunica es, University of Beira Interior, Covilh , Portugal</b><br/><b>Remote Sensing Unit Optics, Optometry and Vision Sciences Group, University of Beira Interior</b><br/>Covilhã, Portugal +</td><td>('1712429', 'Hugo Proença', 'hugo proença')</td><td>E-mail: gmelfe@ubi.pt +</td></tr><tr><td>17479e015a2dcf15d40190e06419a135b66da4e0</td><td>Predicting First Impressions with Deep Learning +<br/><b>University of Notre Dame</b><br/><b>Harvard University 3Perceptive Automata, Inc</b></td><td>('7215627', 'Mel McCurrie', 'mel mccurrie')<br/>('51174355', 'Fernando Beletti', 'fernando beletti')<br/>('51176594', 'Lucas Parzianello', 'lucas parzianello')<br/>('51176974', 'Allen Westendorp', 'allen westendorp')<br/>('2613438', 'Walter J. Scheirer', 'walter j. scheirer')</td><td></td></tr><tr><td>17fa1c2a24ba8f731c8b21f1244463bc4b465681</td><td>Published as a conference paper at ICLR 2016 +<br/>DEEP MULTI-SCALE VIDEO PREDICTION BEYOND +<br/>MEAN SQUARE ERROR +<br/><b>New York University</b><br/>2Facebook Artificial Intelligence Research +</td><td>('2341378', 'Camille Couprie', 'camille couprie')</td><td>mathieu@cs.nyu.edu, {coupriec,yann}@fb.com +</td></tr><tr><td>17579791ead67262fcfb62ed8765e115fb5eca6f</td><td>Real-Time Fashion-guided Clothing Semantic Parsing: a Lightweight Multi-Scale +<br/>Inception Neural Network and Benchmark +<br/>1School of Data and Computer Science +<br/><b>Beijing University of Posts and Telecommunications, Beijing, P.R. China</b><br/><b>Sun Yat-Sen University, Guangzhou, P.R. China</b><br/>2 PRMCT Lab +</td><td>('3079146', 'Yuhang He', 'yuhang he')</td><td></td></tr><tr><td>177d1e7bbea4318d379f46d8d17720ecef3086ac</td><td>JMLR: Workshop and Conference Proceedings 44 (2015) 60-71 +<br/>NIPS 2015 +<br/>The 1st International Workshop “Feature Extraction: Modern Questions and Challenges” +<br/>Learning Multi-channel Deep Feature Representations for +<br/>Face Recognition +<br/><b>Wayne State University, Detroit, MI 48202, USA</b><br/><b>University of Illinois at Urbana Champaign, Urbana</b><br/>IL 61801, USA +<br/>Editor: Afshin Rostamizadeh +</td><td>('2410994', 'Xue-wen Chen', 'xue-wen chen')<br/>('2708905', 'Melih S. Aslan', 'melih s. aslan')<br/>('1982110', 'Kunlei Zhang', 'kunlei zhang')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>xuewen.chen@wayne.edu +<br/>melih.aslan@wayne.edu +<br/>kunlei.zhang@wayne.edu +<br/>t-huang1@illinois.edu +</td></tr><tr><td>17a995680482183f3463d2e01dd4c113ebb31608</td><td>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. X, NO. Y, MONTH Z +<br/>Structured Label Inference for +<br/>Visual Understanding +</td><td>('3079079', 'Nelson Nauata', 'nelson nauata')<br/>('2804000', 'Hexiang Hu', 'hexiang hu')<br/>('2057809', 'Guang-Tong Zhou', 'guang-tong zhou')<br/>('47640964', 'Zhiwei Deng', 'zhiwei deng')<br/>('2928799', 'Zicheng Liao', 'zicheng liao')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td></td></tr><tr><td>17aa78bd4331ef490f24bdd4d4cd21d22a18c09c</td><td></td><td></td><td></td></tr><tr><td>170a5f5da9ac9187f1c88f21a88d35db38b4111a</td><td>Online Real-time Multiple Spatiotemporal Action Localisation and Prediction +<br/>Philip Torr2 +<br/><b>Oxford Brookes University</b><br/><b>Oxford University</b><br/>Figure 1: Online spatiotemporal action localisation in a test ‘fencing’ video from UCF-101 [39]. (a) to (c): A 3D volumetric view of +<br/>the video showing detection boxes and selected frames. At any given time, a certain portion (%) of the entire video is observed by the +<br/>system, and the detection boxes are linked up to incrementally build online space-time action tubes in real-time. Note that the proposed +<br/>method is able to detect multiple co-occurring action instances (3 action instances are shown in different colours). Note also that one of +<br/>the fencers moves out of the image boundaries between frames 114 and 145, to which our model responds by trimming action tube 01 +<br/>at frame 114, and initiating a new tube (03) at frame 146. +</td><td>('1931660', 'Gurkirt Singh', 'gurkirt singh')<br/>('3017538', 'Suman Saha', 'suman saha')<br/>('3019396', 'Michael Sapienza', 'michael sapienza')<br/>('1754181', 'Fabio Cuzzolin', 'fabio cuzzolin')</td><td>{gurkirt.singh-2015,suman.saha-2014,fabio.cuzzolin}@brookes.ac.uk +<br/>{michael.sapienza,philip.torr}@eng.ox.ac.uk +</td></tr><tr><td>17c0d99171efc957b88c31a465c59485ab033234</td><td></td><td></td><td></td></tr><tr><td>1742ffea0e1051b37f22773613f10f69d2e4ed2c</td><td></td><td></td><td></td></tr><tr><td>1791f790b99471fc48b7e9ec361dc505955ea8b1</td><td></td><td></td><td></td></tr><tr><td>17a8d1b1b4c23a630b051f35e47663fc04dcf043</td><td>Differential Angular Imaging for Material Recognition +<br/><b>Rutgers University, Piscataway, NJ</b><br/><b>Drexel University, Philadelphia, PA</b></td><td>('48181328', 'Jia Xue', 'jia xue')</td><td>{jia.xue,zhang.hang}@rutgers.edu, kdana@ece.rutgers.edu, kon@drexel.edu +</td></tr><tr><td>171d8a39b9e3d21231004f7008397d5056ff23af</td><td>Simultaneous Facial Landmark Detection, Pose and Deformation Estimation +<br/>under Facial Occlusion +<br/>ECSE Department +<br/><b>Institute of Automation</b><br/>ECSE Department +<br/><b>Rensselaer Polytechnic Institute</b><br/>Chinese Academy of Sciences +<br/><b>Rensselaer Polytechnic Institute</b></td><td>('1746738', 'Yue Wu', 'yue wu')<br/>('2864523', 'Chao Gou', 'chao gou')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>wuyuesophia@gmail.com +<br/>gouchao2012@ic.ac.cn +<br/>jiq@rpi.edu +</td></tr><tr><td>17045163860fc7c38a0f7d575f3e44aaa5fa40d7</td><td>Boosting VLAD with Supervised Dictionary +<br/>Learning and High-Order Statistics +<br/><b>Southwest Jiaotong University, Chengdu, China</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Key Lab of CVPR, Shenzhen Institutes of Advanced Technology, CAS</b><br/>Hong Kong, China +<br/><b>Hengyang Normal University, Hengyang, China</b><br/>Shenzhen, China +</td><td>('1766837', 'Xiaojiang Peng', 'xiaojiang peng')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('33427555', 'Yu Qiao', 'yu qiao')<br/>('37040717', 'Qiang Peng', 'qiang peng')</td><td></td></tr><tr><td>174930cac7174257515a189cd3ecfdd80ee7dd54</td><td>Multi-view Face Detection Using Deep Convolutional +<br/>Neural Networks +<br/>Yahoo +<br/>Mohammad Saberian +<br/>inc.com +<br/>Yahoo +<br/>Yahoo +</td><td>('2114438', 'Sachin Sudhakar Farfade', 'sachin sudhakar farfade')<br/>('33642044', 'Li-Jia Li', 'li-jia li')</td><td>fsachin@yahoo-inc.com +<br/>saberian@yahoo- +<br/>lijiali.vision@gmail.com +</td></tr><tr><td>17fad2cc826d2223e882c9fda0715fcd5475acf3</td><td></td><td></td><td></td></tr><tr><td>17e563af203d469c456bb975f3f88a741e43fb71</td><td>Naming TV Characters by Watching and Analyzing Dialogs +<br/><b>Karlsruhe Institute of Technology, 76131 Karlsruhe, Germany</b></td><td>('3408009', 'Monica-Laura Haurilet', 'monica-laura haurilet')<br/>('2103464', 'Makarand Tapaswi', 'makarand tapaswi')<br/>('2256981', 'Ziad Al-Halah', 'ziad al-halah')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>{haurilet, tapaswi, ziad.al-halah, rainer.stiefelhagen}@kit.edu +</td></tr><tr><td>171389529df11cc5a8b1fbbe659813f8c3be024d</td><td>Manifold Estimation in View-based Feature +<br/>Space for Face Synthesis across Poses +<br/>Center for Visualization and Virtual Environments +<br/><b>University of Kentucky, USA</b></td><td>('2257812', 'Xinyu Huang', 'xinyu huang')<br/>('2943451', 'Jizhou Gao', 'jizhou gao')<br/>('1772171', 'Sen-Ching S. Cheung', 'sen-ching s. cheung')<br/>('38958903', 'Ruigang Yang', 'ruigang yang')</td><td></td></tr><tr><td>17d5e5c9a9ee4cf85dfbb9d9322968a6329c3735</td><td>Study on Parameter Selection Using SampleBoost +<br/>Computer Science and Engineering Department, +<br/><b>University of North Texas, Denton, Texas, USA</b></td><td>('1898814', 'Mohamed Abouelenien', 'mohamed abouelenien')<br/>('1982703', 'Xiaohui Yuan', 'xiaohui yuan')</td><td>{mohamed, xiaohui.yuan}@unt.edu +</td></tr><tr><td>1750db78b7394b8fb6f6f949d68f7c24d28d934f</td><td>Detecting Facial Retouching Using Supervised +<br/>Deep Learning +<br/>Bowyer, Fellow, IEEE +</td><td>('5014060', 'Aparna Bharati', 'aparna bharati')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')</td><td></td></tr><tr><td>17cf6195fd2dfa42670dc7ada476e67b381b8f69</td><td>†Image Processing Laboratory, Department of Image Engineering +<br/>Graduate School of Advanced Imaging Science, Multimedia, and Film +<br/><b>Chung-Ang University, Seoul, Korea</b><br/><b>Korea Electronics Technology Institute, 203-103 B/D 192, Yakdae-Dong</b><br/>Wonmi-Gu Puchon-Si, Kyunggi-Do 420-140, Korea +<br/>‡Imaging, Robotics, and Intelligent Systems Laboratory +<br/>Department of Electrical and Computer Engineering +<br/><b>The University of Tennessee, Knoxville</b><br/>AUTOMATIC FACE REGION TRACKING FOR HIGHLY ACCURATE FACE +<br/>RECOGNITION IN UNCONSTRAINED ENVIRONMENTS +</td><td>('2243148', 'Young-Ouk Kim', 'young-ouk kim')<br/>('1684329', 'Joonki Paik', 'joonki paik')<br/>('39533703', 'Jingu Heo', 'jingu heo')</td><td></td></tr><tr><td>173657da03e3249f4e47457d360ab83b3cefbe63</td><td>HKU-Face: A Large Scale Dataset for +<br/>Deep Face Recognition +<br/>Final Report +<br/>3035140108 +<br/>COMP4801 Final Year Project +<br/>Project Code: 17007 +</td><td>('3347561', 'Haicheng Wang', 'haicheng wang')</td><td></td></tr><tr><td>174f46eccb5852c1f979d8c386e3805f7942bace</td><td>The Shape-Time Random Field for Semantic Video Labeling +<br/>School of Computer Science +<br/><b>University of Massachusetts, Amherst MA, USA</b></td><td>('2177037', 'Andrew Kae', 'andrew kae')</td><td>{akae,marlin,elm}@cs.umass.edu +</td></tr><tr><td>17670b60dcfb5cbf8fdae0b266e18cf995f6014c</td><td>Longitudinal Face Modeling via +<br/>Temporal Deep Restricted Boltzmann Machines +<br/><b>Computer Science and Software Engineering, Concordia University, Montr eal, Qu ebec, Canada</b><br/>2 CyLab Biometrics Center and the Department of Electrical and Computer Engineering, +<br/><b>Carnegie Mellon University, Pittsburgh, PA, USA</b></td><td>('1876581', 'Chi Nhan Duong', 'chi nhan duong')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('2687827', 'Kha Gia Quach', 'kha gia quach')<br/>('1699922', 'Tien D. Bui', 'tien d. bui')</td><td>1{c duon, k q, bui}@encs.concordia.ca, 2kluu@andrew.cmu.edu +</td></tr><tr><td>17027a05c1414c9a06a1c5046899abf382a1142d</td><td>Articulated Motion Discovery using Pairs of Trajectories +<br/><b>University of Edinburgh</b><br/>2Google Research +</td><td>('2059950', 'Luca Del Pero', 'luca del pero')<br/>('2262946', 'Susanna Ricco', 'susanna ricco')<br/>('1694199', 'Rahul Sukthankar', 'rahul sukthankar')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')</td><td>ldelper@inf.ed.ac.uk +<br/>ricco@google.com +<br/>sukthankar@google.com +<br/>ferrari@inf.ed.ac.uk +</td></tr><tr><td>17ded725602b4329b1c494bfa41527482bf83a6f</td><td>Compact Convolutional Neural Network Cascade for Face Detection +<br/>Kalinovskii I.A. +<br/>Spitsyn V.G. +<br/><b>Tomsk Polytechnic University</b><br/><b>Tomsk Polytechnic University</b><br/>Tomsk, Russia +<br/>Tomsk, Russia +</td><td></td><td>kua_21@mail.ru +<br/>spvg@tpu.ru +</td></tr><tr><td>177bc509dd0c7b8d388bb47403f28d6228c14b5c</td><td>Deep Learning Face Representation from Predicting 10,000 Classes +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('1681656', 'Yi Sun', 'yi sun')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>sy011@ie.cuhk.edu.hk +<br/>xgwang@ee.cuhk.edu.hk +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>7ba0bf9323c2d79300f1a433ff8b4fe0a00ad889</td><td></td><td></td><td></td></tr><tr><td>7bbaa09c9e318da4370a83b126bcdb214e7f8428</td><td>FaaSter, Better, Cheaper: The Prospect of +<br/>Serverless Scientific Computing and HPC +<br/><b>Zurich University of Applied Sciences, School of Engineering</b><br/>Service Prototyping Lab (blog.zhaw.ch/icclab/), 8401 Winterthur, Switzerland +<br/><b>ISISTAN Research Institute - CONICET - UNICEN</b><br/>Campus Universitario, Paraje Arroyo Seco, Tandil (7000), Buenos Aires, Argentina +<br/><b>ITIC Research Institute, National University of Cuyo</b><br/>Padre Jorge Contreras 1300, M5502JMA Mendoza, Argentina +</td><td>('1765470', 'Josef Spillner', 'josef spillner')<br/>('2891834', 'Cristian Mateos', 'cristian mateos')<br/>('34889755', 'David A. Monge', 'david a. monge')</td><td>josef.spillner@zhaw.ch +<br/>cristian.mateos@isistan.unicen.edu.ar +<br/>dmonge@uncu.edu.ar +</td></tr><tr><td>7b63ed54345d8c06523f6b03c41a09b5c8f227e2</td><td>Facial Expression Recognition Based on +<br/>Combination of Spatio-temporal and Spectral +<br/>Features in Local Facial Regions +<br/>Department of Electrical Engineering, +<br/><b>Najafabad Branch, Islamic Azad University</b><br/>Isfahan, Iran. +</td><td>('9337964', 'Nakisa Abounasr', 'nakisa abounasr')</td><td>n_abounasr@sel.iaun.ac.ir +</td></tr><tr><td>7bf0a1aa1d0228a51d24c0c3a83eceb937a6ae25</td><td><b>UNIVERSITY OF CALIFORNIA, SAN DIEGO</b><br/>Video-based Car Surveillance: License Plate, Make, and Model Recognition +<br/>A thesis submitted in partial satisfaction of the +<br/>requirements for the degree Masters of Science +<br/>in Computer Science +<br/>by +<br/>Louka Dlagnekov +<br/>Committee in charge: +<br/>Professor Serge J. Belongie, Chairperson +<br/>2005 +</td><td>('3520515', 'David A. Meyer', 'david a. meyer')<br/>('1765887', 'David J. Kriegman', 'david j. kriegman')</td><td></td></tr><tr><td>7b9961094d3e664fc76b12211f06e12c47a7e77d</td><td>Bridging Biometrics and Forensics +<br/><b>EECS, Syracuse University, Syracuse, NY, USA</b></td><td>('38495931', 'Yanjun Yan', 'yanjun yan')<br/>('2598035', 'Lisa Ann Osadciw', 'lisa ann osadciw')</td><td>{yayan, laosadci}@syr.edu +</td></tr><tr><td>7bfe085c10761f5b0cc7f907bdafe1ff577223e0</td><td></td><td></td><td></td></tr><tr><td>7b43326477795a772c08aee750d3e433f00f20be</td><td>Computational Methods for Behavior Analysis +<br/>Thesis by +<br/>In Partial Fulfillment of the Requirements for the +<br/>degree of +<br/>Doctor of Philosophy +<br/><b>CALIFORNIA INSTITUTE OF TECHNOLOGY</b><br/>Pasadena, California +<br/>2017 +<br/>Defended September 16, 2016 +</td><td>('2948199', 'Eyrun Eyjolfsdottir', 'eyrun eyjolfsdottir')</td><td></td></tr><tr><td>7b9b3794f79f87ca8a048d86954e0a72a5f97758</td><td>DOI 10.1515/jisys-2013-0016 Journal of Intelligent Systems 2013; 22(4): 365–415 +<br/>Passing an Enhanced Turing Test – +<br/>Interacting with Lifelike Computer +<br/>Representations of Specific Individuals +</td><td>('1708812', 'Avelino J. Gonzalez', 'avelino j. gonzalez')<br/>('1745342', 'Jason Leigh', 'jason leigh')<br/>('1727179', 'Ronald F. DeMara', 'ronald f. demara')<br/>('7777088', 'Steven Jones', 'steven jones')<br/>('1761244', 'Sangyoon Lee', 'sangyoon lee')<br/>('1917523', 'Carlos Leon-Barth', 'carlos leon-barth')<br/>('3191606', 'Miguel Elvir', 'miguel elvir')<br/>('33294824', 'James Hollister', 'james hollister')<br/>('2680448', 'Steven Kobosko', 'steven kobosko')</td><td></td></tr><tr><td>7bce4f4e85a3bfcd6bfb3b173b2769b064fce0ed</td><td>A Psychologically-Inspired Match-Score Fusion Model +<br/>for Video-Based Facial Expression Recognition +<br/><b>VISLab, EBUII-216, University of California Riverside</b><br/>Riverside, California, USA, 92521-0425 +</td><td>('1707159', 'Bir Bhanu', 'bir bhanu')<br/>('1803478', 'Songfan Yang', 'songfan yang')</td><td>{acruz, bhanu, syang}@ee.ucr.edu +</td></tr><tr><td>7b0f1fc93fb24630eb598330e13f7b839fb46cce</td><td>Learning to Find Eye Region Landmarks for Remote Gaze +<br/>Estimation in Unconstrained Settings +<br/>ETH Zurich +<br/>MPI for Informatics +<br/>MPI for Informatics +<br/>ETH Zurich +</td><td>('20466488', 'Seonwook Park', 'seonwook park')<br/>('2520795', 'Xucong Zhang', 'xucong zhang')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')<br/>('2531379', 'Otmar Hilliges', 'otmar hilliges')</td><td>spark@inf.ethz.ch +<br/>xczhang@mpi-inf.mpg.de +<br/>bulling@mpi-inf.mpg.de +<br/>otmarh@inf.ethz.ch +</td></tr><tr><td>7be60f8c34a16f30735518d240a01972f3530e00</td><td>Facial Expression Recognition with Temporal Modeling of Shapes +<br/><b></b><br/><b>The University of Texas at Austin</b></td><td>('18692590', 'Suyog Jain', 'suyog jain')<br/>('1713065', 'Changbo Hu', 'changbo hu')</td><td>suyog@cs.utexas.edu, changbo.hu@gmail.com, aggarwaljk@mail.utexas.edu +</td></tr><tr><td>7bdcd85efd1e3ce14b7934ff642b76f017419751</td><td>289 +<br/>Learning Discriminant Face Descriptor +</td><td>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td></td></tr><tr><td>7b3b7769c3ccbdf7c7e2c73db13a4d32bf93d21f</td><td>On the Design and Evaluation of Robust Head Pose for +<br/>Visual User Interfaces: Algorithms, Databases, and +<br/>Comparisons +<br/>Laboratory of Intelligent and +<br/>Safe Automobiles +<br/>UCSD - La Jolla, CA, USA +<br/>Laboratory of Intelligent and +<br/>Safe Automobiles +<br/>UCSD - La Jolla, CA, USA +<br/>Laboratory of Intelligent and +<br/>Safe Automobiles +<br/>UCSD - La Jolla, CA, USA +<br/>Laboratory of Intelligent and +<br/>Safe Automobiles +<br/>UCSD - La Jolla, CA, USA +<br/>Mohan Trivedi +<br/>Laboratory of Intelligent and +<br/>Safe Automobiles +<br/>UCSD - La Jolla, CA, USA +</td><td>('1841835', 'Sujitha Martin', 'sujitha martin')<br/>('1947383', 'Ashish Tawari', 'ashish tawari')<br/>('1780529', 'Erik Murphy-Chutorian', 'erik murphy-chutorian')<br/>('3205274', 'Shinko Y. Cheng', 'shinko y. cheng')</td><td>scmartin@ucsd.edu +<br/>atawari@ucsd.edu +<br/>erikmc@google.com +<br/>sycheng@hrl.com +<br/>mtrivedi@ucsd.edu +</td></tr><tr><td>8fe38962c24300129391f6d7ac24d7783e0fddd0</td><td><b>Center for Research in Computer Vision, University of Central Florida</b></td><td>('33209161', 'Amir Mazaheri', 'amir mazaheri')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>amirmazaheri@knights.ucf.edu +<br/>shah@crcv.ucf.edu +</td></tr><tr><td>8f6d05b8f9860c33c7b1a5d704694ed628db66c7</td><td>Non-linear dimensionality reduction and sparse +<br/>representation models for facial analysis +<br/>To cite this version: +<br/>Medical Imaging. INSA de Lyon, 2014. English. <NNT : 2014ISAL0019>. <tel-01127217> +<br/>HAL Id: tel-01127217 +<br/>https://tel.archives-ouvertes.fr/tel-01127217 +<br/>Submitted on 7 Mar 2015 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('35061362', 'Yuyao Zhang', 'yuyao zhang')<br/>('35061362', 'Yuyao Zhang', 'yuyao zhang')</td><td></td></tr><tr><td>8f772d9ce324b2ef5857d6e0b2a420bc93961196</td><td>MAHPOD et al.: CFDRNN +<br/>Facial Landmark Point Localization using +<br/>Coarse-to-Fine Deep Recurrent Neural Network +</td><td>('2748312', 'Shahar Mahpod', 'shahar mahpod')<br/>('3001038', 'Rig Das', 'rig das')<br/>('1767715', 'Emanuele Maiorana', 'emanuele maiorana')<br/>('1926432', 'Yosi Keller', 'yosi keller')<br/>('1682433', 'Patrizio Campisi', 'patrizio campisi')</td><td></td></tr><tr><td>8f3e120b030e6c1d035cb7bd9c22f6cc75782025</td><td>Bayesian Networks and the Imprecise Dirichlet +<br/>Model applied to Recognition Problems +<br/><b>Dalle Molle Institute for Arti cial Intelligence</b><br/>Galleria 2, Manno-Lugano, Switzerland +<br/><b>Rensselaer Polytechnic Institute</b><br/>110 Eighth St., Troy, NY, USA +</td><td>('1726583', 'Qiang Ji', 'qiang ji')</td><td>cassio@idsia.ch, jiq@rpi.edu +</td></tr><tr><td>8fb611aca3bd8a3a0527ac0f38561a5a9a5b8483</td><td></td><td></td><td></td></tr><tr><td>8fda2f6b85c7e34d3e23927e501a4b4f7fc15b2a</td><td>Feature Selection with Annealing for Big Data +<br/>Learning +</td><td>('2455529', 'Adrian Barbu', 'adrian barbu')<br/>('34680388', 'Yiyuan She', 'yiyuan she')<br/>('2139735', 'Liangjing Ding', 'liangjing ding')<br/>('3019469', 'Gary Gramajo', 'gary gramajo')</td><td></td></tr><tr><td>8fed5ea3b69ea441a8b02f61473eafee25fb2374</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Two-Dimensional PCA with F-Norm Minimization +<br/><b>State Key Laboratory of ISN, Xidian University</b><br/><b>State Key Laboratory of ISN, Xidian University</b><br/>Xi’an China +<br/>Xi’an China +</td><td>('38469552', 'Quanxue Gao', 'quanxue gao')<br/>('40326660', 'Qianqian Wang', 'qianqian wang')</td><td></td></tr><tr><td>8fa3478aaf8e1f94e849d7ffbd12146946badaba</td><td>Attributes for Classifier Feedback +<br/><b>Indraprastha Institute of Information Technology (Delhi, India</b><br/><b>Toyota Technological Institute (Chicago, US</b></td><td>('2076800', 'Amar Parkash', 'amar parkash')<br/>('1713589', 'Devi Parikh', 'devi parikh')</td><td></td></tr><tr><td>8f3da45ff0c3e1777c3a7830f79c10f5896bcc21</td><td>Situation Recognition with Graph Neural Networks +<br/><b>The Chinese University of Hong Kong, 2University of Toronto, 3Youtu Lab, Tencent</b><br/><b>Uber Advanced Technologies Group, 5Vector Institute</b></td><td>('8139953', 'Ruiyu Li', 'ruiyu li')<br/>('2103464', 'Makarand Tapaswi', 'makarand tapaswi')<br/>('2246396', 'Renjie Liao', 'renjie liao')<br/>('1729056', 'Jiaya Jia', 'jiaya jia')<br/>('2422559', 'Raquel Urtasun', 'raquel urtasun')<br/>('37895334', 'Sanja Fidler', 'sanja fidler')</td><td>ryli@cse.cuhk.edu.hk, {makarand,rjliao,urtasun,fidler}@cs.toronto.edu, leojia9@gmail.com +</td></tr><tr><td>8ff8c64288a2f7e4e8bf8fda865820b04ab3dbe8</td><td>Age Estimation Using Expectation of Label Distribution Learning ∗ +<br/><b>National Key Laboratory for Novel Software Technology, Nanjing University, China</b><br/><b>MOE Key Laboratory of Computer Network and Information Integration, Southeast University, China</b></td><td>('2226422', 'Bin-Bin Gao', 'bin-bin gao')<br/>('7678704', 'Hong-Yu Zhou', 'hong-yu zhou')<br/>('1808816', 'Jianxin Wu', 'jianxin wu')<br/>('1735299', 'Xin Geng', 'xin geng')</td><td>{gaobb,zhouhy,wujx}@lamda.nju.edu.cn, xgeng@seu.edu.cn +</td></tr><tr><td>8f9c37f351a91ed416baa8b6cdb4022b231b9085</td><td>Generative Adversarial Style Transfer Networks for Face Aging +<br/>Sveinn Palsson +<br/>D-ITET, ETH Zurich +<br/>Eirikur Agustsson +<br/>D-ITET, ETH Zurich +</td><td></td><td>spalsson@ethz.ch +<br/>aeirikur@ethz.ch +</td></tr><tr><td>8f8c0243816f16a21dea1c20b5c81bc223088594</td><td></td><td></td><td></td></tr><tr><td>8f08b2101d43b1c0829678d6a824f0f045d57da5</td><td>Supplementary Material for: Active Pictorial Structures +<br/><b>Imperial College London</b><br/>180 Queens Gate, SW7 2AZ, London, U.K. +<br/>In the following sections, we provide additional material for the paper “Active Pictorial Structures”. Section 1 explains in +<br/>more detail the differences between the proposed Active Pictorial Structures (APS) and Pictorial Structures (PS). Section 2 +<br/>presents the proofs about the structure of the precision matrices of the Gaussian Markov Random Filed (GMRF) (Eqs. 10 +<br/>and 12 of the main paper). Section 3 gives an analysis about the forward Gauss-Newton optimization of APS and shows that +<br/>the inverse technique with fixed Jacobian and Hessian, which is used in the main paper, is much faster. Finally, Sec. 4 shows +<br/>additional experimental results and conducts new experiments on different objects (human eyes and cars). An open-source +<br/>implementation of APS is available within the Menpo Project [1] in http://www.menpo.org/. +<br/>1. Differences between Active Pictorial Structures and Pictorial Structures +<br/>As explained in the main paper, the proposed model is partially motivated by PS [4, 8]. In the original formulation of PS, +<br/>the cost function to be optimized has the form +<br/>(cid:88) +<br/>n(cid:88) +<br/>n(cid:88) +<br/>i=1 +<br/>arg min +<br/>= arg min +<br/>i=1 +<br/>mi((cid:96)i) + +<br/>dij((cid:96)i, (cid:96)j) = +<br/>i,j:(vi,vj )∈E +<br/>[A((cid:96)i) − µa +<br/>i ]T (Σa +<br/>i )−1[A((cid:96)i) − µa +<br/>i ] + +<br/>(cid:88) +<br/>i,j:(vi,vj )∈E +<br/>[(cid:96)i − (cid:96)j − µd +<br/>ij]T (Σd +<br/>ij)−1[(cid:96)i − (cid:96)j − µd +<br/>ij] +<br/>(1) +<br/>1 , . . . , (cid:96)T +<br/>n ]T is the vector of landmark coordinates ((cid:96)i = [xi, yi]T , ∀i = 1, . . . , n), A((cid:96)i) is a feature vector +<br/>where s = [(cid:96)T +<br/>ij} denote the mean +<br/>extracted from the image location (cid:96)i and we have assumed a tree G = (V, E). {µa +<br/>and covariances of the appearance and deformation respectively. In Eq. 1, mi((cid:96)i) is a function measuring the degree of +<br/>mismatch when part vi is placed at location (cid:96)i in the image. Moreover, dij((cid:96)i, (cid:96)j) denotes a function measuring the degree +<br/>of deformation of the model when part vi is placed at location (cid:96)i and part vj is placed at location (cid:96)j. The authors show +<br/>an inference algorithm based on distance transform [3] that can find a global minimum of Eq. 1 without any initialization. +<br/>However, this algorithm imposes two important restrictions: (1) appearance of each part is independent of the rest of them +<br/>and (2) G must always be acyclic (a tree). Additionally, the computation of mi((cid:96)i) for all parts (i = 1, . . . , n) and all possible +<br/>image locations (response maps) has a high computational cost, which makes the algorithm very slow. Finally, in [8], the +<br/>authors only use a diagonal covariance for the relative locations (deformation) of each edge of the graph, which restricts the +<br/>flexibility of the model. +<br/>i } and {µd +<br/>ij, Σd +<br/>i , Σa +<br/>In the proposed APS, we aim to minimize the cost function (Eq. 19 of the main paper) +<br/>(cid:107)A(S(¯s, p)) − ¯a(cid:107)2 +<br/>[A(S(¯s, p)) − ¯a]T Qa[A(S(¯s, p)) − ¯a] + [S(¯s, p) − ¯s]T Qd[S(¯s, p) − ¯s] +<br/>Qa + (cid:107)S(¯s, p) − ¯s(cid:107)2 +<br/>Qd = +<br/>arg min +<br/>= arg min +<br/>(2) +<br/>There are two main differences between APS and PS: (1) we employ a statistical shape model and optimize with respect +<br/>to its parameters and (2) we use the efficient Gauss-Newton optimization technique. However, these differences introduce +<br/>some important advantages, as also mentioned in the main paper. The proposed formulation allows to define a graph (not +<br/>only tree) between the object’s parts. This means that we can assume dependencies between any pair of landmarks for both +</td><td>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('2575567', 'Joan Alabort-i-Medina', 'joan alabort-i-medina')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>{e.antonakos, ja310, s.zafeiriou}@imperial.ac.uk +</td></tr><tr><td>8fbec9105d346cd23d48536eb20c80b7c2bbbe30</td><td>The Effectiveness of Face Detection Algorithms in Unconstrained Crowd Scenes +<br/>Department of Computer Science and Engineering +<br/><b>University of Notre Dame</b><br/>Notre Dame, IN 46656 +</td><td>('27937356', 'Jeremiah R. Barr', 'jeremiah r. barr')<br/>('1799014', 'Kevin W. Bowyer', 'kevin w. bowyer')<br/>('1704876', 'Patrick J. Flynn', 'patrick j. flynn')</td><td>jbarr1,kwb,flynn@nd.edu +</td></tr><tr><td>8f3e3f0f97844d3bfd9e9ec566ac7a54f6931b09</td><td>Electronic Letters on Computer Vision and Image Analysis 14(2):24-44; 2015 +<br/>A Survey on Human Emotion Recognition Approaches, +<br/>Databases and Applications +<br/><b>Francis Xavier Engineering College, Tirunelveli, Tamilnadu, India</b><br/><b>P.S.R Engineering College, Sivakasi, Tamilnadu, India</b><br/>Received 7th Aug 2015; accepted 30th Nov 2015 +</td><td></td><td></td></tr><tr><td>8f8a5be9dc16d73664285a29993af7dc6a598c83</td><td>IJCSNS International Journal of Computer Science and Network Security, VOL.11 No.1, January 2011 +<br/>71 +<br/>Neural Network based Face Recognition with Gabor Filters +<br/><b>Jahangirnagar University, Savar, Dhaka 1342, Bangladesh</b><br/> +</td><td>('5463951', 'Amina Khatun', 'amina khatun')<br/>('38674112', 'Al-Amin Bhuiyan', 'al-amin bhuiyan')</td><td></td></tr><tr><td>8f5ce25e6e1047e1bf5b782d045e1dac29ca747e</td><td>A Novel Discriminant Non-negative Matrix +<br/>Factorization Algorithm with Applications to +<br/>Facial Image Characterization Problems +<br/><b>yAristotle University of Thessaloniki</b><br/>Department of Informatics +<br/>Box 451 +<br/>54124 Thessaloniki, Greece +<br/>Address for correspondence: +<br/><b>Aristotle University of Thessaloniki</b><br/>54124 Thessaloniki +<br/>GREECE +<br/>Tel. ++ 30 231 099 63 04 +<br/>Fax ++ 30 231 099 63 04 +<br/>April 18, 2007 +<br/>DRAFT +</td><td>('1754270', 'Irene Kotsia', 'irene kotsia')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>email: fekotsia, dralbert, pitasg@aiia.csd.auth.gr +</td></tr><tr><td>8f89aed13cb3555b56fccd715753f9ea72f27f05</td><td>Attended End-to-end Architecture for Age +<br/>Estimation from Facial Expression Videos +</td><td>('1678473', 'Wenjie Pei', 'wenjie pei')</td><td></td></tr><tr><td>8f92cccacf2c84f5d69db3597a7c2670d93be781</td><td>FACIAL EXPRESSION SYNTHESIS THROUGH FACIAL EXPRESSIONS +<br/>STATISTICAL ANALYSIS +<br/><b>Aristotle University of Thessaloniki</b><br/>Department of Informatics +<br/>Box 451, 54124 Thessaloniki, Greece +</td><td>('2764130', 'Stelios Krinidis', 'stelios krinidis')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>email: pitas@zeus.csd.auth.gr, stelios.krinidis@mycosmos.gr +</td></tr><tr><td>8f6263e4d3775757e804796e104631c7a2bb8679</td><td>Characterizing Visual Representations within Convolutional Neural Networks: +<br/>Toward a Quantitative Approach +<br/><b>Center for Brain Science, Harvard University, Cambridge, MA 02138 USA</b><br/><b>Center for Brain Science, Harvard University, Cambridge, MA 02138 USA</b></td><td>('1739108', 'Chuan-Yung Tsai', 'chuan-yung tsai')<br/>('2042941', 'David D. Cox', 'david d. cox')</td><td>CHUANYUNGTSAI@FAS.HARVARD.EDU +<br/>DAVIDCOX@FAS.HARVARD.EDU +</td></tr><tr><td>8f9f599c05a844206b1bd4947d0524234940803d</td><td></td><td></td><td></td></tr><tr><td>8f60c343f76913c509ce623467bf086935bcadac</td><td>Joint 3D Face Reconstruction and Dense +<br/>Alignment with Position Map Regression +<br/>Network +<br/><b>Shanghai Jiao Tong University, CloudWalk Technology</b><br/><b>Research Center for Intelligent Security Technology, CIGIT</b></td><td>('9196752', 'Yao Feng', 'yao feng')<br/>('1917608', 'Fan Wu', 'fan wu')<br/>('3492237', 'Xiaohu Shao', 'xiaohu shao')<br/>('1706354', 'Yanfeng Wang', 'yanfeng wang')<br/>('39851640', 'Xi Zhou', 'xi zhou')</td><td>fengyao@sjtu.edu.cn, wufan@cloudwalk.cn, shaoxiaohu@cigit.ac.cn +<br/>wangyanfeng@sjtu.edu.cn, zhouxi@cloudwalk.cn +</td></tr><tr><td>8fd9c22b00bd8c0bcdbd182e17694046f245335f</td><td> +<br/>Recognizing Facial Expressions in Videos +</td><td>('8502461', 'Lin Su', 'lin su')<br/>('14362431', 'Matthew Balazsi', 'matthew balazsi')</td><td></td></tr><tr><td>8f5facdc0a2a79283864aad03edc702e2a400346</td><td> +<br/> +<br/>ISSN: 2277-3754 +<br/>ISO 9001:2008 Certified +<br/>International Journal of Engineering and Innovative Technology (IJEIT) +<br/>Volume 4, Issue 7, January 2015 +<br/>Human Age Estimation Framework using +<br/>Bio-Inspired Features for Facial Image +<br/>Santhosh Kumar G, Dr. Suresh H. N. +<br/>Research scholor, BIT, under VTU, Belgaum India +<br/><b>Bangalore Institute of Technology</b><br/>Bangalore–04, Karnataka +</td><td></td><td></td></tr><tr><td>8a3c5507237957d013a0fe0f082cab7f757af6ee</td><td>Facial Landmark Detection by +<br/>Deep Multi-task Learning +<br/><b>The Chinese University of Hong Kong</b></td><td>('3152448', 'Zhanpeng Zhang', 'zhanpeng zhang')<br/>('1693209', 'Ping Luo', 'ping luo')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>8af411697e73f6cfe691fe502d4bfb42510b4835</td><td>Dynamic Local Ternary Pattern for Face Recognition and +<br/>Verification +<br/><b>Institute of Information Technology</b><br/><b>University of Dhaka, Bangladesh</b><br/>Department of Industrial and Management Engineering +<br/><b>Hankuk University of Foreign Studies, South Korea</b><br/>M. Abdullah-Al-Wadud +</td><td>('39036762', 'Mohammad Ibrahim', 'mohammad ibrahim')<br/>('31210416', 'Humayun Kayesh', 'humayun kayesh')<br/>('13193999', 'Shah', 'shah')<br/>('2233124', 'Mohammad Shoyaib', 'mohammad shoyaib')</td><td>ibrahim iit@yahoo.com, iftekhar.efat@gmail.com, hkayesh@gmail.com, khaled@univdhaka.edu, shoyaib@du.ac.bd +<br/>wadud@hufs.ac.kr +</td></tr><tr><td>8acdc4be8274e5d189fb67b841c25debf5223840</td><td>Gultepe and Makrehchi +<br/>Hum. Cent. Comput. Inf. Sci. (2018) 8:25 +<br/>https://doi.org/10.1186/s13673-018-0148-3 +<br/>RESEARCH +<br/>Improving clustering performance +<br/>using independent component analysis +<br/>and unsupervised feature learning +<br/>Open Access +<br/>*Correspondence: +<br/>Department of Electrical +<br/>and Computer Engineering, +<br/><b>University of Ontario Institute</b><br/>of Technology, 2000 Simcoe +<br/>St N, Oshawa, ON L1H 7K4, +<br/>Canada +</td><td>('2729102', 'Eren Gultepe', 'eren gultepe')<br/>('3183840', 'Masoud Makrehchi', 'masoud makrehchi')</td><td>eren.gultepe@uoit.net +</td></tr><tr><td>8a1ed5e23231e86216c9bdd62419c3b05f1e0b4d</td><td>Facial Keypoint Detection +<br/><b>Stanford University</b><br/>March 13, 2016 +</td><td>('29909347', 'Shayne Longpre', 'shayne longpre')<br/>('9928926', 'Ajay Sohmshetty', 'ajay sohmshetty')</td><td>slongpre@stanford.edu, ajay14@stanford.edu +</td></tr><tr><td>8a54f8fcaeeede72641d4b3701bab1fe3c2f730a</td><td>What do you think of my picture? Investigating factors +<br/>of influence in profile images context perception +<br/>Heynderickx +<br/>To cite this version: +<br/>think of my picture? Investigating factors of influence in profile images context perception. Human +<br/>Vision and Electronic Imaging XX, Mar 2015, San Francisco, United States. Proc. SPIE 9394, Hu- +<br/>man Vision and Electronic Imaging XX, 9394, <http://spie.org/EI/conferencedetails/human-vision- +<br/>electronic-imaging>. <10.1117/12.2082817>. <hal-01149535> +<br/>HAL Id: hal-01149535 +<br/>https://hal.archives-ouvertes.fr/hal-01149535 +<br/>Submitted on 7 May 2015 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('34678433', 'Filippo Mazza', 'filippo mazza')<br/>('40130265', 'Matthieu Perreira Da Silva', 'matthieu perreira da silva')<br/>('7591543', 'Patrick Le Callet', 'patrick le callet')<br/>('34678433', 'Filippo Mazza', 'filippo mazza')<br/>('40130265', 'Matthieu Perreira Da Silva', 'matthieu perreira da silva')<br/>('7591543', 'Patrick Le Callet', 'patrick le callet')<br/>('1728396', 'Ingrid Heynderickx', 'ingrid heynderickx')</td><td></td></tr><tr><td>8a8861ad6caedc3993e31d46e7de6c251a8cda22</td><td>StreetStyle: Exploring world-wide clothing styles from millions of photos +<br/><b>Cornell University</b><br/>Figure 1: Extracting and measuring clothing style from Internet photos at scale. (a) We apply deep learning methods to learn to extract +<br/>fashion attributes from images and create a visual embedding of clothing style. We use this embedding to analyze millions of Instagram photos +<br/>of people sampled worldwide, in order to study spatio-temporal trends in clothing around the globe. (b) Further, using our embedding, we +<br/>can cluster images to produce a global set of representative styles, from which we can (c) use temporal and geo-spatial statistics to generate +<br/>concise visual depictions of what makes clothing unique in each city versus the rest. +</td><td>('40353974', 'Kevin Matzen', 'kevin matzen')<br/>('1791337', 'Kavita Bala', 'kavita bala')<br/>('1830653', 'Noah Snavely', 'noah snavely')</td><td></td></tr><tr><td>8aae23847e1beb4a6d51881750ce36822ca7ed0b</td><td>Comparison Between Geometry-Based and Gabor-Wavelets-Based +<br/>Facial Expression Recognition Using Multi-Layer Perceptron +<br/><b>ATR Human Information Processing Research Laboratories</b><br/><b>ATR Interpreting Telecommunications Research Laboratories</b><br/>2-2 Hikaridai, Seika-cho, Soraku-gun, Kyoto 619-02, Japan +<br/>INRIA, 2004 route des Lucioles, BP 93, F-06902 Sophia-Antipolis Cedex, France +</td><td>('1809184', 'Zhengyou Zhang', 'zhengyou zhang')<br/>('34801422', 'Shigeru Akamatsu', 'shigeru akamatsu')<br/>('36206997', 'Michael Schuster', 'michael schuster')</td><td>e-mail: zzhang@sophia.inria.fr, zzhang@hip.atr.co.jp +</td></tr><tr><td>8a866bc0d925dfd8bb10769b8b87d7d0ff01774d</td><td>WikiArt Emotions: An Annotated Dataset of Emotions Evoked by Art +<br/>National Research Council Canada +</td><td>('2886725', 'Svetlana Kiritchenko', 'svetlana kiritchenko')</td><td>{saif.mohammad,svetlana.kiritchenko}@nrc-cnrc.gc.ca +</td></tr><tr><td>8a40b6c75dd6392ee0d3af73cdfc46f59337efa9</td><td></td><td></td><td></td></tr><tr><td>8a3bb63925ac2cdf7f9ecf43f71d65e210416e17</td><td>ShearFace: Efficient Extraction of Anisotropic +<br/>Features for Face Recognition +<br/>1Research Groups on Intelligent Machines, +<br/><b>University of Sfax</b><br/> Sfax 3038, Tunisia +<br/>and anisotropic +</td><td>('2791150', 'Mohamed Anouar Borgi', 'mohamed anouar borgi')<br/>('8847309', 'Demetrio Labate', 'demetrio labate')</td><td>{anoir.borgi@ieee.org; dlabate@math.uh.edu} +</td></tr><tr><td>8a0159919ee4e1a9f4cbfb652a1be212bf0554fd</td><td><b>University of Surrey</b><br/>Faculty of Engineering and Physical Sciences +<br/>Department of Computer Science +<br/>PhD Thesis +<br/>Application of Power Laws to +<br/>Biometrics, Forensics and +<br/>Network Traffic Analysis +<br/>by +<br/>Supervisor: Prof. A.T.S. Ho +<br/>Co-supervisors: Dr. N. Poh, Dr. S. Li +<br/>November, 2016 +</td><td>('2909991', 'Aamo Iorliam', 'aamo iorliam')</td><td></td></tr><tr><td>8ad0d8cf4bcb5c7eccf09f23c8b7d25439c4ae2b</td><td>Predicting the Future with Transformational +<br/>States +<br/><b>University of Pennsylvania, 2Ryerson University</b></td><td>('2689633', 'Andrew Jaegle', 'andrew jaegle')<br/>('40805511', 'Oleh Rybkin', 'oleh rybkin')<br/>('3150825', 'Konstantinos G. Derpanis', 'konstantinos g. derpanis')<br/>('1751586', 'Kostas Daniilidis', 'kostas daniilidis')</td><td>ajaegle@upenn.edu, oleh@cis.upenn.edu, +<br/>kosta@scs.ryerson.ca, kostas@cis.upenn.edu +</td></tr><tr><td>8adb2fcab20dab5232099becbd640e9c4b6a905a</td><td>Beyond Euclidean Eigenspaces: +<br/>Bayesian Matching for Visual Recognition +<br/><b>Mitsubishi Electric Research Laboratory</b><br/>MIT Media Laboratory +<br/> +<br/> +<br/>Cambridge, MA +<br/>Cambridge, MA +</td><td>('1780935', 'Baback Moghaddam', 'baback moghaddam')<br/>('1682773', 'Alex Pentland', 'alex pentland')</td><td>baback@merl.com +<br/>sandy@media.mit.edu +</td></tr><tr><td>8a0d10a7909b252d0e11bf32a7f9edd0c9a8030b</td><td>Animals on the Web +<br/><b>University of California, Berkeley</b><br/><b>University of Illinois, Urbana-Champaign</b><br/>Computer Science Division +<br/>Department of Computer Science +</td><td>('1685538', 'Tamara L. Berg', 'tamara l. berg')<br/>('1744452', 'David A. Forsyth', 'david a. forsyth')</td><td>millert@cs.berkeley.edu +<br/>daf@cs.uiuc.edu +</td></tr><tr><td>8a91ad8c46ca8f4310a442d99b98c80fb8f7625f</td><td>2592 +<br/>2D Segmentation Using a Robust Active +<br/>Shape Model With the EM Algorithm +</td><td>('38769654', 'Carlos Santiago', 'carlos santiago')<br/>('3259175', 'Jacinto C. Nascimento', 'jacinto c. nascimento')<br/>('1744810', 'Jorge S. Marques', 'jorge s. marques')</td><td></td></tr><tr><td>8aed6ec62cfccb4dba0c19ee000e6334ec585d70</td><td>Localizing and Visualizing Relative Attributes +</td><td>('2299381', 'Fanyi Xiao', 'fanyi xiao')<br/>('1883898', 'Yong Jae Lee', 'yong jae lee')</td><td></td></tr><tr><td>8a336e9a4c42384d4c505c53fb8628a040f2468e</td><td>Wang and Luo EURASIP Journal on Bioinformatics +<br/>and Systems Biology (2016) 2016:13 +<br/>DOI 10.1186/s13637-016-0048-7 +<br/>R ES EAR CH +<br/>Detecting Visually Observable Disease +<br/>Symptoms from Faces +<br/>Open Access +</td><td>('2207567', 'Kuan Wang', 'kuan wang')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')</td><td></td></tr><tr><td>7e600faee0ba11467d3f7aed57258b0db0448a72</td><td></td><td></td><td></td></tr><tr><td>7ed3b79248d92b255450c7becd32b9e5c834a31e</td><td>L1-regularized Logistic Regression Stacking and Transductive CRF Smoothing +<br/>for Action Recognition in Video +<br/><b>University of Florence</b><br/>Lorenzo Seidenari +<br/><b>University of Florence</b><br/>Andrew D. Bagdanov +<br/><b>University of Florence</b><br/><b>University of Florence</b></td><td>('2602265', 'Svebor Karaman', 'svebor karaman')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')</td><td>svebor.karaman@unifi.it +<br/>lorenzo.seidenari@unifi.it +<br/>bagdanov@dsi.unifi.it +<br/>alberto.delbimbo@unifi.it +</td></tr><tr><td>7e8016bef2c180238f00eecc6a50eac473f3f138</td><td>TECHNISCHE UNIVERSIT ¨AT M ¨UNCHEN +<br/>Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +<br/>Immersive Interactive Data Mining and Machine +<br/>Learning Algorithms for Big Data Visualization +<br/>Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +<br/>der Technischen Universit¨at M¨unchen zur Erlangung des akademischen Grades eines +<br/>Doktor-Ingenieurs (Dr.-Ing.) +<br/>genehmigten Dissertation. +<br/>Vorsitzender: +<br/>Univ.-Prof. Dr. sc.techn. Andreas Herkersdorf +<br/>Pr¨ufer der Dissertation: +<br/>1. Univ.-Prof. Dr.-Ing. habil. Gerhard Rigoll +<br/>2. Univ.-Prof. Dr.-Ing. habil. Dirk Wollherr +<br/>3. Prof. Dr. Mihai Datcu +<br/>Die Dissertation wurde am 13.08.2015 bei der Technischen Universit¨at M¨unchen eingerei- +<br/>cht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik am 16.02.2016 +<br/>angenommen. +</td><td>('2133342', 'Mohammadreza Babaee', 'mohammadreza babaee')</td><td></td></tr><tr><td>7ed2c84fdfc7d658968221d78e745dfd1def6332</td><td>May 15, 2007 6:32 +<br/>World Scientific Review Volume - 9.75in x 6.5in +<br/>ObjectRecognitionLCV2 +<br/>Chapter 1 +<br/>Evaluation of linear combination of views for object recognition +<br/>on real and synthetic datasets +<br/>Department of computer science, +<br/><b>University College London</b><br/>Malet Place, London, WC1E 6BT +<br/>In this work, we present a method for model-based recognition of 3d objects from +<br/>a small number of 2d intensity images taken from nearby, but otherwise arbitrary +<br/>viewpoints. Our method works by linearly combining images from two (or more) +<br/>viewpoints of a 3d object to synthesise novel views of the object. The object is +<br/>recognised in a target image by matching to such a synthesised, novel view. All +<br/>that is required is the recovery of the linear combination parameters, and since +<br/>we are working directly with pixel intensities, we suggest searching the parameter +<br/>space using a global, evolutionary optimisation algorithm combined with a local +<br/>search method in order efficiently to recover the optimal parameters and thus +<br/>recognise the object in the scene. We have experimented with both synthetic +<br/>data and real-image, public databases. +<br/>1.1. Introduction +<br/>Object recognition is one of the most important and basic problems in computer +<br/>vision and, for this reason, it has been studied extensively resulting in a plethora +<br/>of publications and a variety of different approachesa aiming to solve this problem. +<br/>Nevertheless accurate, robust and efficient solutions remain elusive because of the +<br/>inherent difficulties when dealing in particular with 3d objects that may be seen +<br/>from a variety of viewpoints. Variations in geometry, photometry and viewing angle, +<br/>noise, occlusions and incomplete data are some of the problems with which object +<br/>recognition systems are faced. +<br/>In this paper, we will address a particular kind of extrinsic variations: varia- +<br/>tions of the image due to changes in the viewpoint from which the object is seen. +<br/>Traditionally, methods that aimed to solve the recognition problem for objects with +<br/>varying pose relied on an explicit 3d model of the object, generating 2d projections +<br/>from that model and comparing them with the scene image. Such was the work +<br/>aFor a comprehensive review of object recognition methods and deformable templates in particular, +<br/>see Refs. 1–4. +</td><td>('1797883', 'Vasileios Zografos', 'vasileios zografos')<br/>('31557997', 'Bernard F. Buxton', 'bernard f. buxton')</td><td>{v.zografos,b.buxton}@cs.ucl.ac.uk +</td></tr><tr><td>7eaa97be59019f0d36aa7dac27407b004cad5e93</td><td>Sampling Generative Networks +<br/>School of Design +<br/><b>Victoria University of Wellington</b><br/>Wellington, New Zealand +</td><td>('40603980', 'Tom White', 'tom white')</td><td>tom.white@vuw.ac.nz +</td></tr><tr><td>7eb895e7de883d113b75eda54389460c61d63f67</td><td>Can you tell a face from a HEVC bitstream? +<br/><b>School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada</b></td><td>('3393216', 'Saeed Ranjbar Alvar', 'saeed ranjbar alvar')<br/>('3320198', 'Hyomin Choi', 'hyomin choi')</td><td>Email: {saeedr,chyomin, ibajic}@sfu.ca +</td></tr><tr><td>7e467e686f9468b826133275484e0a1ec0f5bde6</td><td>Efficient On-the-fly Category Retrieval +<br/>using ConvNets and GPUs +<br/><b>Visual Geometry Group, University of Oxford</b></td><td>('34838386', 'Karen Simonyan', 'karen simonyan')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>{ken,karen,az}@robots.ox.ac.uk +</td></tr><tr><td>7e3367b9b97f291835cfd0385f45c75ff84f4dc5</td><td>Improved Local Binary Pattern Based Action Unit Detection Using +<br/>Morphological and Bilateral Filters +<br/>1Signal Processing Laboratory (LTS5) +<br/>´Ecole Polytechnique F´ed´erale de Lausanne, +<br/>Switzerland +<br/>2nViso SA +<br/>Lausanne, Switzerland +</td><td>('2916630', 'Matteo Sorci', 'matteo sorci')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')</td><td>{anil.yuce;jean-philippe.thiran}@epfl.ch +<br/>matteo.sorci@nviso.ch +</td></tr><tr><td>7ef0cc4f3f7566f96f168123bac1e07053a939b2</td><td>Triangular Similarity Metric Learning: a Siamese +<br/>Architecture Approach +<br/>To cite this version: +<br/>puter Science [cs]. UNIVERSITE DE LYON, 2016. English. <NNT : 2016LYSEI045>. <tel- +<br/>01314392> +<br/>HAL Id: tel-01314392 +<br/>https://hal.archives-ouvertes.fr/tel-01314392 +<br/>Submitted on 11 May 2016 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('37848497', 'Lilei Zheng', 'lilei zheng')<br/>('37848497', 'Lilei Zheng', 'lilei zheng')</td><td></td></tr><tr><td>7e00fb79576fe213853aeea39a6bc51df9fdca16</td><td>Online Multi-Face Detection and Tracking +<br/>using Detector Confidence and Structured SVMs +<br/><b>Eindhoven University of Technology, The Netherlands</b><br/>2TNO Embedded Systems Innovation, Eindhoven, The Netherlands +</td><td>('3199035', 'Francesco Comaschi', 'francesco comaschi')<br/>('1679431', 'Sander Stuijk', 'sander stuijk')<br/>('1708289', 'Twan Basten', 'twan basten')<br/>('1684335', 'Henk Corporaal', 'henk corporaal')</td><td>{f.comaschi, s.stuijk, a.a.basten, h.corporaal}@tue.nl +</td></tr><tr><td>7e2cfbfd43045fbd6aabd9a45090a5716fc4e179</td><td>Global Norm-Aware Pooling for Pose-Robust Face Recognition at Low False Positive Rate +<br/>Global Norm-Aware Pooling for Pose-Robust Face Recognition at Low False +<br/>Positive Rate +<br/><b>a School of Computer and Information Technology, Beijing Jiaotong University, Beijing</b><br/>China +<br/><b>b Research Institute, Watchdata Inc., Beijing, China</b><br/>c DeepInSight, China +</td><td>('39326372', 'Sheng Chen', 'sheng chen')<br/>('3007274', 'Jia Guo', 'jia guo')<br/>('1681842', 'Yang Liu', 'yang liu')<br/>('46757550', 'Xiang Gao', 'xiang gao')<br/>('2765914', 'Zhen Han', 'zhen han')</td><td>{shengchen, zhan}@bjtu.edu.cn +<br/>{yang.liu.yj, xiang.gao}@watchdata.com +<br/>guojia@gmail.com +</td></tr><tr><td>7ee53d931668fbed1021839db4210a06e4f33190</td><td>What if we do not have multiple videos of the same action? — +<br/>Video Action Localization Using Web Images +<br/><b>Center for Research in Computer Vision (CRCV), University of Central Florida (UCF</b></td><td>('3195774', 'Waqas Sultani', 'waqas sultani')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>waqassultani@knights.ucf.edu, shah@crcv.ucf.edu +</td></tr><tr><td>7e18b5f5b678aebc8df6246716bf63ea5d8d714e</td><td>Original research +<br/>published: 15 January 2018 +<br/>doi: 10.3389/fpsyt.2017.00309 +<br/>increased loss aversion in +<br/>Unmedicated Patients with +<br/>Obsessive–compulsive Disorder +<br/>1 Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, NY, United States, 2 Fishberg Department of +<br/><b>Neuroscience, Icahn School of Medicine at Mount Sinai, Friedman Brain Institute, New York, NY, United States</b><br/><b>of Psychology, University of Michigan, Ann Arbor, MI, United States, University of Michigan, Ann</b><br/>Arbor, MI, United States +<br/>introduction: Obsessive–compulsive disorder (OCD) patients show abnormalities in +<br/>decision-making and, clinically, appear to show heightened sensitivity to potential nega- +<br/>tive outcomes. Despite the importance of these cognitive processes in OCD, few studies +<br/>have examined the disorder within an economic decision-making framework. Here, we +<br/>investigated loss aversion, a key construct in the prospect theory that describes the +<br/>tendency for individuals to be more sensitive to potential losses than gains when making +<br/>decisions. +<br/>Methods: Across two study sites, groups of unmedicated OCD patients (n = 14), medi- +<br/>cated OCD patients (n = 29), and healthy controls (n = 34) accepted or rejected a series +<br/>of 50/50 gambles containing varying loss/gain values. Loss aversion was calculated +<br/>as the ratio of the likelihood of rejecting a gamble with increasing potential losses to +<br/>the likelihood of accepting a gamble with increasing potential gains. Decision times to +<br/>accept or reject were also examined and correlated with loss aversion. +<br/>results: Unmedicated OCD patients exhibited significantly more loss aversion com- +<br/>pared to medicated OCD or controls, an effect that was replicated across both sites +<br/>and remained significant even after controlling for OCD symptom severity, trait anxiety, +<br/>and sex. Post hoc analyses further indicated that unmedicated patients’ increased +<br/>likelihood to reject a gamble as its loss value increased could not be explained solely by +<br/>greater risk aversion among patients. Unmedicated patients were also slower to accept +<br/>than reject gambles, effects that were not found in the other two groups. Loss aversion +<br/>was correlated with decision times in unmedicated patients but not in the other two +<br/>groups. +<br/>Discussion: These data identify abnormalities of decision-making in a subgroup +<br/>of OCD patients not taking psychotropic medication. The findings help elucidate +<br/>the cognitive mechanisms of the disorder and suggest that future treatments could +<br/>aim to target abnormalities of loss/gain processing during decision-making in this +<br/>population. +<br/>Keywords: decision-making, prospect theory, choice behavior, reward, obsessive–compulsive disorder +<br/>Edited by: +<br/>Qinghua He, +<br/><b>Southwest University, China</b><br/>Reviewed by: +<br/>Qiang Wang, +<br/><b>Beijing Normal University, China</b><br/>Michael Grady Wheaton, +<br/><b>Columbia University, United States</b><br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted +<br/>to Psychopathology, +<br/>a section of the journal +<br/>Frontiers in Psychiatry +<br/>Received: 08 December 2017 +<br/>Accepted: 26 December 2017 +<br/>Published: 15 January 2018 +<br/>Citation: +<br/>Sip KE, Gonzalez R, Taylor SF and +<br/>Stern ER (2018) Increased Loss +<br/>Aversion in Unmedicated Patients +<br/>with Obsessive–Compulsive Disorder. +<br/>Front. Psychiatry 8:309. +<br/>doi: 10.3389/fpsyt.2017.00309 +<br/>Frontiers in Psychiatry | www.frontiersin.org +<br/>January 2018 | Volume 8 | Article 309 +</td><td>('3592712', 'Kamila E. Sip', 'kamila e. sip')<br/>('31801083', 'Richard Gonzalez', 'richard gonzalez')<br/>('2085281', 'Stephan F. Taylor', 'stephan f. taylor')<br/>('2025121', 'Emily R. Stern', 'emily r. stern')<br/>('2025121', 'Emily R. Stern', 'emily r. stern')</td><td>emily.stern@mssm.edu, +<br/>emily.stern@nyumc.org +</td></tr><tr><td>7e9df45ece7843fe050033c81014cc30b3a8903a</td><td>AUDIO-VISUAL INTENT-TO-SPEAK DETECTION FOR HUMAN-COMPUTER +<br/>INTERACTION +<br/>Institut Eurecom +<br/> , route des Cr^etes, BP +<br/> +<br/><b>IBM T.J. Watson Research Center</b><br/>Yorktown Heights, NY +</td><td>('3163391', 'Philippe de Cuetos', 'philippe de cuetos')<br/>('2264160', 'Chalapathy Neti', 'chalapathy neti')<br/>('33666044', 'Andrew W. Senior', 'andrew w. senior')</td><td>decuetos@eurecom.fr +<br/> cneti,aws @us.ibm.com +</td></tr><tr><td>7ebd323ddfe3b6de8368c4682db6d0db7b70df62</td><td>Proceedings of the International Conference on Computer and Information Science and Technology +<br/>Ottawa, Ontario, Canada, May 11 – 12, 2015 +<br/>Paper No. 111 +<br/>Location-based Face Recognition Using Smart Mobile Device +<br/>Sensors +<br/>Department of Computer Science +<br/><b>University of Victoria, Victoria, Canada</b></td><td>('2019933', 'Nina Taherimakhsousi', 'nina taherimakhsousi')<br/>('1747880', 'Hausi A. Müller', 'hausi a. müller')</td><td>ninata@uvic.ca; hausi@uvic.ca +</td></tr><tr><td>7eb85bcb372261bad707c05e496a09609e27fdb3</td><td>A Compute-efficient Algorithm for Robust Eyebrow Detection +<br/><b>Nanyang Technological University, 2University of California San Diego</b></td><td>('36375772', 'Supriya Sathyanarayana', 'supriya sathyanarayana')<br/>('1710219', 'Ravi Kumar Satzoda', 'ravi kumar satzoda')<br/>('1924458', 'Suchitra Sathyanarayana', 'suchitra sathyanarayana')</td><td>supriya001@e.ntu.edu.sg, rsatzoda@eng.ucsd.edu, ssathyanarayana@ucsd.edu, astsrikan@ntu.edu.sg +</td></tr><tr><td>7ed6ff077422f156932fde320e6b3bd66f8ffbcb</td><td>State of 3D Face Biometrics for Homeland Security Applications +<br/>Chaudhari4 +</td><td>('2925401', 'Anshuman Razdan', 'anshuman razdan')<br/>('1693971', 'Gerald Farin', 'gerald farin')</td><td></td></tr><tr><td>7ebb153704706e457ab57b432793d2b6e5d12592</td><td>ZHONG, ARANDJELOVI ´C, ZISSERMAN: FACES IN PLACES +<br/>Faces In Places: compound query retrieval +<br/>Relja Arandjelovi´c2 +<br/>1 Visual Geometry Group +<br/>Department of Engineering Science +<br/><b>University of Oxford, UK</b><br/>2 WILLOW project +<br/>Departement d’Informatique de l’École +<br/>Normale Supérieure +<br/>ENS/INRIA/CNRS UMR 8548 +</td><td>('6730372', 'Yujie Zhong', 'yujie zhong')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>yujie@robots.ox.ac.uk +<br/>relja.arandjelovic@inria.fr +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>7ec7163ec1bc237c4c2f2841c386f2dbfd0cc922</td><td>ORIGINAL RESEARCH +<br/>published: 20 June 2018 +<br/>doi: 10.3389/fpsyg.2018.00971 +<br/>Skiing and Thinking About It: +<br/>Moment-to-Moment and +<br/>Retrospective Analysis of Emotions +<br/>in an Extreme Sport +<br/>and Tove Irene Dahl +<br/><b>UiT The Arctic University of Norway, Troms , Norway</b><br/>Happiness is typically reported as an important reason for participating in challenging +<br/>activities like extreme sport. While in the middle of the activity, however, participants +<br/>do not seem particularly happy. So where does the happiness come from? The +<br/>article proposes some answers from a study of facially expressed emotions measured +<br/>moment-by-moment during a backcountry skiing event. Self-reported emotions were +<br/>also assessed immediately after the skiing. Participants expressed lower levels of +<br/>happiness while skiing, compared to when stopping for a break. Moment-to-moment +<br/>and self-reported measures of emotions were largely unrelated. These findings are +<br/>explained with reference to the Functional Wellbeing Approach (Vittersø, 2013), which +<br/>argues that some moment-to-moment feelings are non-evaluative in the sense of being +<br/>generated directly by the difficulty of an activity. By contrast, retrospective emotional +<br/>feelings are more complex as they include an evaluation of the overall goals and values +<br/>associated with the activity as a whole. +<br/>Keywords: emotions, facial expression, moment-to-moment, functional wellbeing approach, extreme sport, +<br/>backcountry skiing +<br/>INTRODUCTION +<br/>We engage in recreational activities in order to feel good. This pursuit is not restricted to +<br/>leisure activities like sunbathing at the beach or enjoying a fine meal with friends and family. +<br/>Mountaineers, BASE jumpers, and other extreme athletes also claim that the importance of their +<br/>favorite activities is the experience of positive feelings (Brymer, 2005; Willig, 2008; Brown and +<br/>Fraser, 2009; Hetland and Vittersø, 2012). But what exactly is it that feels so good about these +<br/>vigorous and exhausting activities, often referred to as extreme sport? To explore this question, +<br/>we developed a new way of measuring emotions in real time during the activity. We equipped +<br/>the participants with a camera that captured their facially expressed emotion while skiing. These +<br/>films were then analyzed with software for automatic coding of facial expressions and compared +<br/>the participants self-reported emotions assessed in retrospect. This approach enabled us to explore +<br/>long standing questions as to how such positive experiences are created. Are they a result of a series +<br/>of online positive feelings? Or is it the impact of a few central features like intensity peaks, rapid +<br/>emotional changes, and happy endings that create them? Is it the experience of flow? Or is it the +<br/>feeling of mastery that kicks in only after the activity has been successfully accomplished? +<br/>Edited by: +<br/>Eric Brymer, +<br/><b>Leeds Beckett University</b><br/>United Kingdom +<br/>Reviewed by: +<br/>Michael Banissy, +<br/><b>Goldsmiths, University of London</b><br/>United Kingdom +<br/>Ralf Christopher Buckley, +<br/><b>Grif th University, Australia</b><br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Movement Science and Sport +<br/>Psychology, +<br/>a section of the journal +<br/>Frontiers in Psychology +<br/>Received: 26 September 2017 +<br/>Accepted: 25 May 2018 +<br/>Published: 20 June 2018 +<br/>Citation: +<br/>Hetland A, Vittersø J, Wie SOB, +<br/>Kjelstrup E, Mittner M and Dahl TI +<br/>(2018) Skiing and Thinking About It: +<br/>Moment-to-Moment +<br/>and Retrospective Analysis +<br/>of Emotions in an Extreme Sport. +<br/>Front. Psychol. 9:971. +<br/>doi: 10.3389/fpsyg.2018.00971 +<br/>Frontiers in Psychology | www.frontiersin.org +<br/>June 2018 | Volume 9 | Article 971 +</td><td>('50814786', 'Audun Hetland', 'audun hetland')<br/>('2956586', 'Joar Vittersø', 'joar vittersø')<br/>('50823709', 'Simen Oscar Bø Wie', 'simen oscar bø wie')<br/>('50829546', 'Eirik Kjelstrup', 'eirik kjelstrup')<br/>('4281140', 'Matthias Mittner', 'matthias mittner')<br/>('50814786', 'Audun Hetland', 'audun hetland')</td><td>audun.hetland@uit.no +</td></tr><tr><td>7e0c75ce731131e613544e1a85ae0f2c28ee4c1f</td><td><b>Imperial College London</b><br/>Department of Computing +<br/>Regression-based Estimation of Pain and +<br/>Facial Expression Intensity +<br/>June, 2015 +<br/>Submitted in part fulfilment of the requirements for the degree of PhD in Computing and +<br/><b>the Diploma of Imperial College London. This thesis is entirely my own work, and, except</b><br/>where otherwise indicated, describes my own research. +</td><td>('3291812', 'Sebastian Kaltwang', 'sebastian kaltwang')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>7ef44b7c2b5533d00001ae81f9293bdb592f1146</td><td>No d’ordre : 227-2013 +<br/>Anne 2013 +<br/>THESE DE L’UNIVERSITE DE LYON +<br/>Dlivre par +<br/>L’UNIVERSITE CLAUDE BERNARD - LYON 1 +<br/>Ecole Doctorale Informatique et Mathmatiques +<br/>P H D T H E S I S +<br/>D´etection des ´emotions `a partir de vid´eos dans un +<br/>environnement non contrˆol´e +<br/>Detection of emotions from video in non-controlled environment +<br/>Soutenue publiquement (Public defense) le 14/11/2013 +<br/>Composition du jury (Dissertation committee): +<br/>Rapporteurs +<br/>Mr. Renaud SEGUIER +<br/>Mr. Jean-Claude MARTIN +<br/>Examinateurs +<br/>Mr. Thomas MOESLUND +<br/>Mr. Patrick LAMBERT +<br/>Mr. Samir GARBAYA +<br/>Directeur +<br/>Mme. Saida BOUAKAZ +<br/>Co-encadrant +<br/>Mr. Alexandre MEYER +<br/>Mr. Hubert KONIK +<br/>Professor, Supelec, CNRS UMR 6164, Rennes, France +<br/>Professor, LIMSI-CNRS, Universit´e Paris-Sud, France +<br/>Professor, Department of Architecture, Design and Media Technology, +<br/><b>Aalborg University, Denmark</b><br/>Professor, LISTIC - Polytech Annecy-Chambery, France +<br/>Associate Professor, Le2i, ENSAM, Chalon sur Saone, France +<br/>Professor, LIRIS-CNRS, Universit´e Claude Bernard Lyon 1, France +<br/>Associate Professor, LIRIS, Universit´e Claude Bernard Lyon 1, France +<br/>Associate Professor, LaHC, Universit´e Jean Monnet, Saint-Etienne, France +</td><td>('1943666', 'Rizwan Ahmed Khan', 'rizwan ahmed khan')</td><td></td></tr><tr><td>7e1ea2679a110241ed0dd38ff45cd4dfeb7a8e83</td><td>Extensions of Hierarchical Slow Feature +<br/>Analysis for Efficient Classification and +<br/>Regression on High-Dimensional Data +<br/>Dissertation +<br/>Submitted to the Faculty of Electrical +<br/>Engineering and Information Technology +<br/>at the +<br/><b>Ruhr University Bochum</b><br/>for the +<br/>Degree of Doktor-Ingenieur +<br/>Alberto Nicol´as Escalante Ba˜nuelos +<br/>by +<br/>Bochum, Germany, January, 2017 +</td><td></td><td></td></tr><tr><td>7e507370124a2ac66fb7a228d75be032ddd083cc</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TAFFC.2017.2708106, IEEE +<br/>Transactions on Affective Computing +<br/>Dynamic Pose-Robust Facial Expression +<br/>Recognition by Multi-View Pairwise Conditional +<br/>Random Forests +<br/>1 Sorbonne Universit´es, UPMC Univ Paris 06 +<br/>CNRS, UMR 7222, F-75005, Paris, France +</td><td>('3190846', 'Arnaud Dapogny', 'arnaud dapogny')<br/>('2521061', 'Kevin Bailly', 'kevin bailly')</td><td></td></tr><tr><td>1056347fc5e8cd86c875a2747b5f84fd570ba232</td><td></td><td></td><td></td></tr><tr><td>10550ee13855bd7403946032354b0cd92a10d0aa</td><td>Accelerating Neuromorphic Vision Algorithms +<br/>for Recognition +<br/>Ahmed Al Maashri +<br/>Vijaykrishnan Narayanan +<br/><b>Microsystems Design Lab, The Pennsylvania State University</b><br/>†IBM System and Technology Group +<br/><b>School of Electrical, Computer and Energy Engineering, Arizona State University</b></td><td>('1723845', 'Michael DeBole', 'michael debole')<br/>('36156473', 'Matthew Cotter', 'matthew cotter')<br/>('2916636', 'Nandhini Chandramoorthy', 'nandhini chandramoorthy')<br/>('37095722', 'Yang Xiao', 'yang xiao')<br/>('1685028', 'Chaitali Chakrabarti', 'chaitali chakrabarti')</td><td>{maashri, mjcotter, nic5090, yux106, vijay}@cse.psu.edu +<br/>mvdebole@us.ibm.com +<br/>chaitali@asu.edu +</td></tr><tr><td>10e12d11cb98ffa5ae82343f8904cfe321ae8004</td><td>A New Simplex Sparse Learning Model to Measure +<br/>Data Similarity for Clustering +<br/><b>University of Texas at Arlington</b><br/>Arlington, Texas 76019, USA +</td><td>('39122448', 'Jin Huang', 'jin huang')<br/>('1688370', 'Feiping Nie', 'feiping nie')<br/>('1748032', 'Heng Huang', 'heng huang')</td><td>huangjinsuzhou@gmail.com, feipingnie@gmail.com, heng@uta.edu +</td></tr><tr><td>10e7dd3bbbfbc25661213155e0de1a9f043461a2</td><td>Cross Euclidean-to-Riemannian Metric Learning +<br/>with Application to Face Recognition from Video +</td><td>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1681236', 'Luc Van Gool', 'luc van gool')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>100105d6c97b23059f7aa70589ead2f61969fbc3</td><td>Frontal to Profile Face Verification in the Wild +<br/><b>Center for Automation Research, University of Maryland, College Park, MD 20740, USA</b><br/><b>The State University of New Jersey</b><br/>Piscataway, NJ 08854, USA. +</td><td>('2500202', 'Soumyadip Sengupta', 'soumyadip sengupta')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('34734622', 'David W. Jacobs', 'david w. jacobs')</td><td></td></tr><tr><td>100da509d4fa74afc6e86a49352751d365fceee5</td><td>Multiclass Recognition and Part Localization with Humans in the Loop +<br/>†Department of Computer Science and Engineering +<br/><b>University of California, San Diego</b><br/>Serge Belongie† +<br/>‡Department of Electrical Engineering +<br/><b>California Institute of Technology</b></td><td>('2367820', 'Catherine Wah', 'catherine wah')<br/>('3251767', 'Steve Branson', 'steve branson')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>{cwah,sbranson,sjb}@cs.ucsd.edu +<br/>perona@caltech.edu +</td></tr><tr><td>10ab1b48b2a55ec9e2920a5397febd84906a7769</td><td></td><td></td><td></td></tr><tr><td>10af69f11301679b6fbb23855bf10f6af1f3d2e6</td><td>Beyond Gaussian Pyramid: Multi-skip Feature Stacking for Action Recognition +<br/><b>School of Computer Science, Carnegie Mellon University</b></td><td>('46329993', 'Ming Lin', 'ming lin')<br/>('2314980', 'Xuanchong Li', 'xuanchong li')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')<br/>('1681921', 'Bhiksha Raj', 'bhiksha raj')</td><td>lanzhzh, minglin, xcli, alex, bhiksha@cs.cmu.edu +</td></tr><tr><td>10ce3a4724557d47df8f768670bfdd5cd5738f95</td><td>Fihe igh Fie +<br/>Ac e ad +<br/>Ra +<br/>The Rbic i e Caegie e +<br/>5000 Fbe Ave e ib gh A 15213 +<br/>Abac. ay face ecgii ak he e ad i +<br/>cdii f he be ad ga +<br/> +<br/>a di(cid:11)ee e ad de a di(cid:11)ee i +<br/>ecgii a +<br/> bjec ca ed a abiay e ad de abiay i +<br/>ad ay be f be iage agai ca ed a abiay e ad +<br/> de abiay i +<br/>Fihe +<br/>iage. achig bewee he be ad ga +<br/>he Fihe +<br/>d ci +<br/> ay face ecgii ceai he e f he be ad ga +<br/>di(cid:11)ee. The ga +<br/>The a +<br/>ga +<br/>view ca ed f a caea i he ce f he . The be f ga +<br/>ad be iage ca a +<br/>iage f each bjec a fa +<br/>yica +<br/>iage a ig +<br/>Face ecgii ac e i.e. face ecgii whee he ga +<br/>iage d have he ae e ha eceived vey +<br/>have bee ed which ca ecgize face [1] e geea +<br/>a a vaiey f e. +<br/>a evey e. A +<br/>f exa +<br/>caiig a +<br/>iai vaiai. ca be ed wih abiay ga +<br/>Afe e vaiai he ex igi(cid:12)ca fac a(cid:11)ecig he aea +<br/>ace f face i i +<br/>face ecgii ac i +<br/>face [4 5]. +<br/>i +</td><td></td><td>fgiaiibg@c.c .ed +</td></tr><tr><td>100428708e4884300e4c1ac1f84cbb16e7644ccf</td><td>REGULARIZED SHEARLET NETWORK FOR FACE RECOGNITION USING SINGLE +<br/>SAMPLE PER PERSON +<br/><b>Research Groups on Intelligent Machines, University of Sfax, Sfax 3038, Tunisia</b><br/><b>University of Houston, Houston, TX 77204, USA</b></td><td>('2791150', 'Mohamed Anouar Borgi', 'mohamed anouar borgi')<br/>('8847309', 'Demetrio Labate', 'demetrio labate')<br/>('3410172', 'Chokri Ben Amar', 'chokri ben amar')</td><td>{anoir.borgi@ieee.org; dlabate@math.uh.edu ; maher.elarbi@gmail.com ; chokri.benamar@ieee.org } +</td></tr><tr><td>102e374347698fe5404e1d83f441630b1abf62d9</td><td>Facial Image Analysis for Fully-Automatic +<br/>Prediction of Difficult Endotracheal Intubation +</td><td>('40564153', 'Patrick Schoettker', 'patrick schoettker')<br/>('2916630', 'Matteo Sorci', 'matteo sorci')<br/>('1697965', 'Hua Gao', 'hua gao')<br/>('2612411', 'Christophe Perruchoud', 'christophe perruchoud')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')</td><td></td></tr><tr><td>10f17534dba06af1ddab96c4188a9c98a020a459</td><td>People-LDA: Anchoring Topics to People using Face Recognition +<br/>Erik Learned-Miller +<br/><b>University of Massachusetts Amherst</b><br/>Amherst MA 01003 +<br/>http://vis-www.cs.umass.edu/(cid:152)vidit/peopleLDA +</td><td>('2246870', 'Vidit Jain', 'vidit jain')<br/>('1735747', 'Andrew McCallum', 'andrew mccallum')</td><td></td></tr><tr><td>10e0e6f1ec00b20bc78a5453a00c792f1334b016</td><td>Pose-Selective Max Pooling for Measuring Similarity +<br/>1Dept. of Computer Science +<br/>2Dept. of Electrical & Computer Engineering +<br/><b>Johns Hopkins University, 3400 N. Charles St, Baltimore, MD 21218, USA</b></td><td>('40031188', 'Xiang Xiang', 'xiang xiang')</td><td>xxiang@cs.jhu.edu +</td></tr><tr><td>102b968d836177f9c436141e382915a4f8549276</td><td>Affective Multimodal Human-Computer Interaction +<br/><b>Faculty of EEMCS, Delft University of Technology, The Netherlands</b><br/><b>Faculty of Science, University of Amsterdam, The Netherlands</b><br/><b>Psychology and Psychiatry, University of Pittsburgh, USA</b><br/><b>Beckman Institute, University of Illinois at Urbana-Champaign, USA</b></td><td>('1694605', 'Maja Pantic', 'maja pantic')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>mpantic@ieee.org, nicu@science.uva.nl, jeffcohn@pitt.edu, huang@ifp.uiuc.edu +</td></tr><tr><td>100641ed8a5472536dde53c1f50fa2dd2d4e9be9</td><td>Visual Attributes for Enhanced Human-Machine Communication* +</td><td>('1713589', 'Devi Parikh', 'devi parikh')</td><td></td></tr><tr><td>10195a163ab6348eef37213a46f60a3d87f289c5</td><td></td><td></td><td></td></tr><tr><td>10e704c82616fb5d9c48e0e68ee86d4f83789d96</td><td></td><td></td><td></td></tr><tr><td>101569eeef2cecc576578bd6500f1c2dcc0274e2</td><td>Multiaccuracy: Black-Box Post-Processing for Fairness in +<br/>Classification +<br/>James Zou +</td><td>('40102677', 'Michael P. Kim', 'michael p. kim')<br/>('27316199', 'Amirata Ghorbani', 'amirata ghorbani')</td><td>mpk@cs.stanford.edu +<br/>amiratag@stanford.edu +<br/>jamesz@stanford.edu +</td></tr><tr><td>106732a010b1baf13c61d0994552aee8336f8c85</td><td>Expanded Parts Model for Semantic Description +<br/>of Humans in Still Images +</td><td>('2515597', 'Gaurav Sharma', 'gaurav sharma')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>10e70a34d56258d10f468f8252a7762950830d2b</td><td></td><td></td><td></td></tr><tr><td>102b27922e9bd56667303f986404f0e1243b68ab</td><td>Wang et al. Appl Inform (2017) 4:13 +<br/>DOI 10.1186/s40535-017-0042-5 +<br/>RESEARCH +<br/>Multiscale recurrent regression networks +<br/>for face alignment +<br/>Open Access +<br/>*Correspondence: +<br/>3 State Key Lab of Intelligent +<br/>Technologies and Systems, +<br/>Beijing 100084, People’s +<br/>Republic of China +<br/>Full list of author information +<br/>is available at the end of the +<br/>article +</td><td>('27660491', 'Caixun Wang', 'caixun wang')<br/>('22192520', 'Haomiao Sun', 'haomiao sun')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('2632601', 'Jianjiang Feng', 'jianjiang feng')<br/>('25060740', 'Jie Zhou', 'jie zhou')</td><td>lujiwen@tsinghua.edu.cn +</td></tr><tr><td>10fcbf30723033a5046db791fec2d3d286e34daa</td><td>On-Line Cursive Handwriting Recognition: A Survey of Methods +<br/>and Performances +<br/>*Faculty of Computer Science & Information Systems, Universiti Teknologi Malaysia (UTM) , 81310 +<br/>Skudai, Johor, Malaysia. +</td><td>('1731121', 'Dzulkifli Mohamad', 'dzulkifli mohamad')<br/>('1921146', 'M. Othman', 'm. othman')</td><td> 1dzul@fsksm.utm.my, faisal@gmm.fsksm.utm.my, razib@fsksm.utm.my +</td></tr><tr><td>101d4cfbd6f8a7a10bd33505e2b183183f1d8770</td><td>The 2013 SESAME Multimedia Event Detection and +<br/>Recounting System +<br/>SRI International (SRI) +<br/><b>University of Amsterdam (UvA</b><br/><b>University of Southern California</b><br/>(USC) +<br/>Cees G.M. Snoek +<br/>Remi Trichet +</td><td>('1764443', 'Robert C. Bolles', 'robert c. bolles')<br/>('40560201', 'J. Brian Burns', 'j. brian burns')<br/>('48804780', 'James A. Herson', 'james a. herson')<br/>('31693932', 'Gregory K. Myers', 'gregory k. myers')<br/>('2594026', 'Stephanie Pancoast', 'stephanie pancoast')<br/>('1746492', 'Julien van Hout', 'julien van hout')<br/>('49966591', 'Julie Wong', 'julie wong')<br/>('3000952', 'AmirHossein Habibian', 'amirhossein habibian')<br/>('1769315', 'Dennis C. Koelma', 'dennis c. koelma')<br/>('3245057', 'Zhenyang Li', 'zhenyang li')<br/>('2690389', 'Masoud Mazloom', 'masoud mazloom')<br/>('37806314', 'Silvia-Laura Pintea', 'silvia-laura pintea')<br/>('1964898', 'Sung Chun Lee', 'sung chun lee')<br/>('1858100', 'Pramod Sharma', 'pramod sharma')<br/>('40559421', 'Chen Sun', 'chen sun')</td><td></td></tr><tr><td>108b2581e07c6b7ca235717c749d45a1fa15bb24</td><td>Using Stereo Matching with General Epipolar +<br/>Geometry for 2D Face Recognition +<br/>across Pose +</td><td>('38171682', 'Carlos D. Castillo', 'carlos d. castillo')<br/>('34734622', 'David W. Jacobs', 'david w. jacobs')</td><td></td></tr><tr><td>106092fafb53e36077eba88f06feecd07b9e78e7</td><td>Attend and Interact: Higher-Order Object Interactions for Video Understanding +<br/><b>Georgia Institute of Technology, 2NEC Laboratories America, 3Georgia Tech Research Institute</b></td><td>('7437104', 'Chih-Yao Ma', 'chih-yao ma')<br/>('2293919', 'Asim Kadav', 'asim kadav')<br/>('50162780', 'Iain Melvin', 'iain melvin')<br/>('1746245', 'Zsolt Kira', 'zsolt kira')<br/>('1775043', 'Hans Peter Graf', 'hans peter graf')</td><td></td></tr><tr><td>103c8eaca2a2176babab2cc6e9b25d48870d6928</td><td>FINDING RELEVANT SEMANTIC CONTENT FOR GROUNDED LANGUAGE LEARNING +<br/>PANNING FOR GOLD: +<br/><b>The University of Texas at Austin</b><br/>Department of Computer Science +<br/>Austin, TX 78712, USA +</td><td>('47514115', 'David L. Chen', 'david l. chen')<br/>('1797655', 'Raymond J. Mooney', 'raymond j. mooney')</td><td>dlcc@cs.utexas.edu and mooney@cs.utexas.edu +</td></tr><tr><td>10d334a98c1e2a9e96c6c3713aadd42a557abb8b</td><td>Scene Text Recognition using Part-based Tree-structured Character Detection +<br/>State Key Laboratory of Management and Control for Complex Systems, CASIA, Beijing, China +</td><td>('1959339', 'Cunzhao Shi', 'cunzhao shi')<br/>('1683416', 'Chunheng Wang', 'chunheng wang')<br/>('2658590', 'Baihua Xiao', 'baihua xiao')<br/>('1698138', 'Yang Zhang', 'yang zhang')<br/>('39001252', 'Song Gao', 'song gao')<br/>('34539206', 'Zhong Zhang', 'zhong zhang')</td><td>{cunzhao.shi,chunheng.wang,baihua.xiao,yang.zhang,song.gao,zhong.zhang}@ia.ac.cn +</td></tr><tr><td>10f66f6550d74b817a3fdcef7fdeba13ccdba51c</td><td>Benchmarking Face Alignment +<br/><b>Institute for Anthropomatics</b><br/><b>Karlsruhe Institute of Technology</b><br/>Karlsruhe, Germany +</td><td>('1697965', 'Hua Gao', 'hua gao')</td><td>Email: {gao, ekenel}@kit.edu +</td></tr><tr><td>107fc60a6c7d58a6e2d8572ad8c19cc321a9ef53</td><td>Hollywood in Homes: Crowdsourcing Data +<br/>Collection for Activity Understanding +<br/><b>Carnegie Mellon University</b><br/>2Inria +<br/><b>University of Washington</b><br/><b>The Allen Institute for AI</b><br/>http://allenai.org/plato/charades/ +</td><td>('34280810', 'Gunnar A. Sigurdsson', 'gunnar a. sigurdsson')<br/>('39849136', 'Xiaolong Wang', 'xiaolong wang')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('1785596', 'Ivan Laptev', 'ivan laptev')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')</td><td></td></tr><tr><td>1048c753e9488daa2441c50577fe5fdba5aa5d7c</td><td>Recognising faces in unseen modes: a tensor based approach +<br/><b>Curtin University of Technology</b><br/>GPO Box U1987, Perth, WA 6845, Australia. +</td><td>('2867032', 'Santu Rana', 'santu rana')<br/>('1713220', 'Wanquan Liu', 'wanquan liu')<br/>('1679953', 'Mihai Lazarescu', 'mihai lazarescu')<br/>('1679520', 'Svetha Venkatesh', 'svetha venkatesh')</td><td>{santu.rana, wanquan, m.lazarescu, svetha}@cs.curtin.edu.au +</td></tr><tr><td>10ca2e03ff995023a701e6d8d128455c6e8db030</td><td>Modeling Stylized Character Expressions +<br/>via Deep Learning +<br/><b>University of Washington</b><br/>Seattle, WA, USA +<br/>2 Zillow Group, Seattle, WA, USA +<br/>3 Gage Academy of Art, Seattle, WA, USA +</td><td>('2494850', 'Deepali Aneja', 'deepali aneja')<br/>('2952700', 'Alex Colburn', 'alex colburn')<br/>('9610752', 'Gary Faigin', 'gary faigin')<br/>('3349536', 'Barbara Mones', 'barbara mones')</td><td>{deepalia,shapiro,mones}@cs.washington.edu +<br/>alexco@cs.washington.edu +<br/>gary@gageacademy.org +</td></tr><tr><td>1921e0a97904bdf61e17a165ab159443414308ed</td><td><b>Bielefeld University</b><br/>Faculty of Technology +<br/>Applied Informatics +<br/>Bachelor Thesis +<br/>Retrieval of Web Images for +<br/>Computer Vision Research +<br/>September 28, 2009 +<br/>Author: +<br/>malinke techfak.uni-bielefeld.de +<br/>Supervisors: +<br/>Dipl.-Inform. Marco Kortkamp +<br/>PD Dr.-Ing. Sven Wachsmuth +</td><td></td><td></td></tr><tr><td>19841b721bfe31899e238982a22257287b9be66a</td><td>Published as a conference paper at ICLR 2018 +<br/>SKIP RNN: LEARNING TO SKIP STATE UPDATES IN +<br/>RECURRENT NEURAL NETWORKS +<br/>†Barcelona Supercomputing Center, ‡Google Inc, +<br/><b>Universitat Polit`ecnica de Catalunya, Columbia University</b></td><td>('2447185', 'Brendan Jou', 'brendan jou')<br/>('1711068', 'Jordi Torres', 'jordi torres')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>{victor.campos, jordi.torres}@bsc.es, bjou@google.com, +<br/>xavier.giro@upc.edu, shih.fu.chang@columbia.edu +</td></tr><tr><td>1922ad4978ab92ce0d23acc4c7441a8812f157e5</td><td>Face Alignment by Coarse-to-Fine Shape Searching +<br/><b>The Chinese University of Hong Kong</b><br/>2SenseTime Group +<br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('2226254', 'Shizhan Zhu', 'shizhan zhu')<br/>('40475617', 'Cheng Li', 'cheng li')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>zs014@ie.cuhk.edu.hk, chengli@sensetime.com, ccloy@ie.cuhk.edu.hk, xtang@ie.cuhk.edu.hk +</td></tr><tr><td>19e62a56b6772bbd37dfc6b8f948e260dbb474f5</td><td>Cross-Domain Metric Learning Based on Information Theory +<br/>1. State Key Laboratory of Computer Science +<br/>2. Science and Technology on Integrated Information System Laboratory +<br/><b>Institute of Software, Chinese Academy of Sciences, Beijing 100190, China</b><br/><b>University of Science and Technology of China</b></td><td>('39483391', 'Hao Wang', 'hao wang')<br/>('40451597', 'Wei Wang', 'wei wang')<br/>('1783918', 'Chen Zhang', 'chen zhang')<br/>('34532334', 'Fanjiang Xu', 'fanjiang xu')</td><td>weiwangpenny@gmail.com +</td></tr><tr><td>192723085945c1d44bdd47e516c716169c06b7c0</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation +<br/>Vision and Attention Theory Based Sampling +<br/>for Continuous Facial Emotion Recognition +<br/>Ninad S. Thakoor, Member, IEEE +<br/>10 +<br/>11 +<br/>12 +<br/>13 +<br/>14 +<br/>15 +<br/>16 +<br/>17 +<br/>18 +<br/>19 +<br/>20 +<br/>21 +<br/>22 +<br/>23 +<br/>24 +<br/>25 +<br/>26 +<br/>27 +<br/>28 +<br/>29 +<br/>30 +<br/>31 +<br/>32 +<br/>33 +<br/>34 +<br/>35 +<br/>36 +<br/>37 +</td><td>('1693314', 'Albert C. Cruz', 'albert c. cruz')<br/>('1707159', 'Bir Bhanu', 'bir bhanu')</td><td></td></tr><tr><td>19fb5e5207b4a964e5ab50d421e2549ce472baa8</td><td>International Conference on Computer Systems and Technologies - CompSysTech’14 +<br/>Online Emotional Facial Expression Dictionary +<br/>Léon Rothkrantz +</td><td></td><td></td></tr><tr><td>1989a1f9ce18d8c2a0cee3196fe6fa363aab80c2</td><td>ROBUST ONLINE FACE TRACKING-BY-DETECTION +<br/>2TNO Embedded Systems Innovation, Eindhoven, The Netherlands +<br/><b>Eindhoven University of Technology, The Netherlands</b></td><td>('3199035', 'Francesco Comaschi', 'francesco comaschi')<br/>('1679431', 'Sander Stuijk', 'sander stuijk')<br/>('1708289', 'Twan Basten', 'twan basten')<br/>('1684335', 'Henk Corporaal', 'henk corporaal')</td><td>{f.comaschi, s.stuijk, a.a.basten, h.corporaal}@tue.nl +</td></tr><tr><td>1962e4c9f60864b96c49d85eb897141486e9f6d1</td><td>Neural Comput & Applic (2011) 20:565–573 +<br/>DOI 10.1007/s00521-011-0577-7 +<br/>O R I G I N A L A R T I C L E +<br/>Locality preserving embedding for face and handwriting digital +<br/>recognition +<br/>Received: 3 December 2008 / Accepted: 11 March 2011 / Published online: 1 April 2011 +<br/>Ó Springer-Verlag London Limited 2011 +<br/>supervised manifold +<br/>the local sub-manifolds. +</td><td>('5383601', 'Zhihui Lai', 'zhihui lai')</td><td></td></tr><tr><td>193debca0be1c38dabc42dc772513e6653fd91d8</td><td>Mnemonic Descent Method: +<br/>A recurrent process applied for end-to-end face alignment +<br/><b>Imperial College London, UK</b><br/><b>Goldsmiths, University of London, UK</b><br/><b>Center for Machine Vision and Signal Analysis, University of Oulu, Finland</b></td><td>('2814229', 'George Trigeorgis', 'george trigeorgis')<br/>('2796644', 'Patrick Snape', 'patrick snape')<br/>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('1752913', 'Mihalis A. Nicolaou', 'mihalis a. nicolaou')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>(cid:63){g.trigeorgis, p.snape, e.antonakos, s.zafeiriou}@imperial.ac.uk, †m.nicolaou@gold.ac.uk +</td></tr><tr><td>191674c64f89c1b5cba19732869aa48c38698c84</td><td>International Journal of Advanced Technology in Engineering and Science www.ijates.com +<br/>Volume No.03, Issue No. 03, March 2015 ISSN (online): 2348 – 7550 +<br/>FACE IMAGE RETRIEVAL USING ATTRIBUTE - +<br/>ENHANCED SPARSE CODEWORDS +<br/>E.Sakthivel1 , M.Ashok kumar2 +<br/><b>PG scholar, Communication Systems, Adhiyamaan College of Engineeing, Hosur, (India</b><br/><b>Electronics And Communication Engg., Adhiyamaan College of Engg., Hosur, (India</b></td><td></td><td></td></tr><tr><td>190d8bd39c50b37b27b17ac1213e6dde105b21b8</td><td>This document is downloaded from DR-NTU, Nanyang Technological +<br/><b>University Library, Singapore</b><br/>Title +<br/>Mining weakly labeled web facial images for search- +<br/>based face annotation +<br/>Author(s) Wang, Dayong; Hoi, Steven C. H.; He, Ying; Zhu, Jianke +<br/>Citation +<br/>Wang, D., Hoi, S. C. H., He, Y., & Zhu, J. (2014). Mining +<br/>weakly labeled web facial images for search-based face +<br/>annotation. IEEE Transactions on Knowledge and Data +<br/>Engineering, 26(1), 166-179. +<br/>Date +<br/>2014 +<br/>URL +<br/>http://hdl.handle.net/10220/18955 +<br/>Rights +<br/>© 2014 IEEE. Personal use of this material is permitted. +<br/>Permission from IEEE must be obtained for all other +<br/><b>uses, in any current or future media, including</b><br/>reprinting/republishing this material for advertising or +<br/>promotional purposes, creating new collective works, for +<br/>resale or redistribution to servers or lists, or reuse of any +<br/>copyrighted component of this work in other works. +<br/>Published version of this article is available at [DOI: +<br/>http://dx.doi.org/10.1109/TKDE.2012.240]. +</td><td></td><td></td></tr><tr><td>19af008599fb17bbd9b12288c44f310881df951c</td><td>Discriminative Local Sparse Representations for +<br/>Robust Face Recognition +</td><td>('1719561', 'Yi Chen', 'yi chen')<br/>('35210356', 'Umamahesh Srinivas', 'umamahesh srinivas')<br/>('1694440', 'Thong T. Do', 'thong t. do')<br/>('3346079', 'Vishal Monga', 'vishal monga')<br/>('1709073', 'Trac D. Tran', 'trac d. tran')</td><td></td></tr><tr><td>19296e129c70b332a8c0a67af8990f2f4d4f44d1</td><td>Metric Learning Approaches for Face Identification +<br/>Is that you? +<br/>M. Guillaumin, J. Verbeek and C. Schmid +<br/>LEAR team, INRIA Rhˆone-Alpes, France +<br/>Supplementary Material +</td><td></td><td></td></tr><tr><td>19666b9eefcbf764df7c1f5b6938031bcf777191</td><td>Group Component Analysis for Multi-block Data: +<br/>Common and Individual Feature Extraction +</td><td>('1764724', 'Guoxu Zhou', 'guoxu zhou')<br/>('1747156', 'Andrzej Cichocki', 'andrzej cichocki')<br/>('38741479', 'Yu Zhang', 'yu zhang')</td><td></td></tr><tr><td>198b6beb53e0e61357825d57938719f614685f75</td><td>Vaulted Verification: A Scheme for Revocable Face +<br/>Recognition +<br/><b>University of Colorado, Colorado Springs</b></td><td>('3035230', 'Michael Wilber', 'michael wilber')</td><td>mwilber@uccs.edu +</td></tr><tr><td>1921795408345751791b44b379f51b7dd54ebfa2</td><td>From Face Recognition to Models of Identity: +<br/>A Bayesian Approach to Learning about +<br/>Unknown Identities from Unsupervised Data +<br/><b>Imperial College London, UK</b><br/>2 Microsoft Research, Cambridge, UK +</td><td>('2388416', 'Sebastian Nowozin', 'sebastian nowozin')</td><td>dc315@imperial.ac.uk +<br/>Sebastian.Nowozin@microsoft.com +</td></tr><tr><td>190b3caa2e1a229aa68fd6b1a360afba6f50fde4</td><td></td><td></td><td></td></tr><tr><td>19e0cc41b9f89492b6b8c2a8a58d01b8242ce00b</td><td>W. ZHANG ET AL.: IMPROVING HFR WITH CGAN +<br/>Improving Heterogeneous Face Recognition +<br/>with Conditional Adversarial Networks +<br/>1 Laboratory LIRIS +<br/>Ecole Centrale de Lyon +<br/>Ecully, France +<br/>2 Computer Vision Lab +<br/><b>Stony Brook University</b><br/>Stony Brook, NY, USA +</td><td>('2553752', 'Wuming Zhang', 'wuming zhang')<br/>('2496409', 'Zhixin Shu', 'zhixin shu')<br/>('1686020', 'Dimitris Samaras', 'dimitris samaras')<br/>('34086868', 'Liming Chen', 'liming chen')</td><td>wuming.zhang@ec-lyon.fr +<br/>zhshu@cs.stonybrook.edu +<br/>samaras@cs.stonybrook.edu +<br/>liming.chen@ec-lyon.fr +</td></tr><tr><td>19e7bdf8310f9038e1a9cf412b8dd2c77ff64c54</td><td>Facial Action Coding Using Multiple Visual Cues and a Hierarchy of Particle +<br/>Filters +<br/><b>Computer Vision and Robotics Research Laboratory</b><br/><b>University of California, San Diego</b></td><td>('32049271', 'Joel C. McCall', 'joel c. mccall')<br/>('1713989', 'Mohan M. Trivedi', 'mohan m. trivedi')</td><td>jmccall@ucsd.edu mtrivedi@ucsd.edu +</td></tr><tr><td>1938d85feafdaa8a65cb9c379c9a81a0b0dcd3c4</td><td>Monogenic Binary Coding: An Efficient Local Feature +<br/>Extraction Approach to Face Recognition +<br/><b>The Hong Kong Polytechnic University, Hong Kong, China</b></td><td>('5828998', 'Meng Yang', 'meng yang')<br/>('36685537', 'Lei Zhang', 'lei zhang')<br/>('1738911', 'Simon C. K. Shiu', 'simon c. k. shiu')<br/>('1698371', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>195d331c958f2da3431f37a344559f9bce09c0f7</td><td>Parsing Occluded People by Flexible Compositions +<br/><b>University of California, Los Angeles</b><br/>Figure 1: An illustration of the flexible compositions. Each connected sub- +<br/>tree of the full graph (include the full graph itself) is a flexible composition. +<br/>The flexible compositions that do not have certain parts are suitable for the +<br/>people with those parts occluded. +<br/>Figure 2: The absence of body parts evidence can help to predict occlusion. +<br/>However, absence of evidence is not evidence of absence. +<br/>It can fail in +<br/>some challenging scenes. The local image measurements near the occlusion +<br/>boundary (i.e., around the right elbow and left shoulder) can reliably provide +<br/>evidence of occlusion. +<br/>This paper presents an approach to parsing humans when there is signifi- +<br/>cant occlusion. We model humans using a graphical model which has a tree +<br/>structure building on recent work [1, 6] and exploit the connectivity prior +<br/>that, even in presence of occlusion, the visible nodes form a connected sub- +<br/>tree of the graphical model. We call each connected subtree a flexible com- +<br/>position of object parts. This involves a novel method for learning occlusion +<br/>cues. During inference we need to search over a mixture of different flexible +<br/>models. By exploiting part sharing, we show that this inference can be done +<br/>extremely efficiently requiring only twice as many computations as search- +<br/>ing for the entire object (i.e., not modeling occlusion). We evaluate our +<br/>model on the standard benchmarked “We Are Family" Stickmen dataset [2] +<br/>and obtain significant performance improvements over the best alternative +<br/>algorithms. +<br/>Parsing humans into parts is an important visual task with many applica- +<br/>tions such as activity recognition. A common approach is to formulate this +<br/>task in terms of graphical models where the graph nodes and edges repre- +<br/>sent human parts and their spatial relationships respectively. This approach +<br/>is becoming successful on benchmarked datasets [1, 6]. But in many real +<br/>world situations many human parts are occluded. Standard methods are par- +<br/>tially robust to occlusion by, for example, using a latent variable to indicate +<br/>whether a part is present and paying a penalty if the part is not detected, but +<br/>are not designed to deal with significant occlusion. One of these models [1] +<br/>will be used in this paper as a base model, and we will compare to it. +<br/>In this paper, we observe that part occlusions often occur in regular pat- +<br/>terns. The visible parts of a human tend to consist of a subset of connected +<br/>parts even when there is significant occlusion (see Figures 1 and 2). In the +<br/>terminology of graphical models, the visible (non-occluded) nodes form a +<br/>connected subtree of the full graphical model (following current models, for +<br/>simplicity, we assume that the graphical model is treelike). This connectiv- +<br/>ity prior is not always valid (i.e., the visible parts of humans may form two +<br/>or more connected subsets), but our analysis suggests it’s often true. In any +<br/>case, we will restrict ourselves to it in this paper, since verifying that some +<br/>isolated pieces of body parts belong to the same person is still very difficult +<br/>for vision methods, especially in challenging scenes where multiple people +<br/>occlude one another (see Figure 2). +<br/>To formulate our approach we build on the base model [1], which is the +<br/>state of the art on several benchmarked datasets [3, 4, 5], but is not designed +<br/>for dealing with significant occlusion. We explicitly model occlusions us- +<br/>ing the connectivity prior above. This means that we have a mixture of +<br/>models where the number of components equals the number of all the pos- +<br/>sible connected subtrees of the graph, which we call flexible compositions, +</td><td>('34420250', 'Xianjie Chen', 'xianjie chen')</td><td></td></tr><tr><td>199c2df5f2847f685796c2523221c6436f022464</td><td>Self Quotient Image for Face Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences; 2Miscrosoft Research Asian; 3Media School</b><br/><b>Bournemouth University</b></td><td>('29948255', 'Haitao Wang', 'haitao wang')<br/>('34679741', 'Stan Z. Li', 'stan z. li')<br/>('1744302', 'Yangsheng Wang', 'yangsheng wang')</td><td></td></tr><tr><td>19c0069f075b5b2d8ac48ad28a7409179bd08b86</td><td>Modifying the Memorability of Face Photographs +<br/><b>Massachusetts Institute of Technology</b><br/>Computer Science and Artificial Intelligence Laboratory +</td><td>('2556428', 'Aditya Khosla', 'aditya khosla')<br/>('2553201', 'Wilma A. Bainbridge', 'wilma a. bainbridge')<br/>('1690178', 'Antonio Torralba', 'antonio torralba')<br/>('31735139', 'Aude Oliva', 'aude oliva')</td><td>{khosla, wilma, torralba, oliva}@csail.mit.edu +</td></tr><tr><td>19c0c7835dba1a319b59359adaa738f0410263e8</td><td>228 +<br/>Natural Image Statistics and +<br/>Low-Complexity Feature Selection +</td><td>('30125215', 'Manuela Vasconcelos', 'manuela vasconcelos')<br/>('1699559', 'Nuno Vasconcelos', 'nuno vasconcelos')</td><td></td></tr><tr><td>19808134b780b342e21f54b60095b181dfc7a600</td><td></td><td></td><td></td></tr><tr><td>19d583bf8c5533d1261ccdc068fdc3ef53b9ffb9</td><td>FaceNet: A Unified Embedding for Face Recognition and Clustering +<br/>Google Inc. +<br/>Google Inc. +<br/>Google Inc. +</td><td>('3302320', 'Florian Schroff', 'florian schroff')<br/>('2741985', 'Dmitry Kalenichenko', 'dmitry kalenichenko')<br/>('2276542', 'James Philbin', 'james philbin')</td><td>fschroff@google.com +<br/>dkalenichenko@google.com +<br/>jphilbin@google.com +</td></tr><tr><td>1910f5f7ac81d4fcc30284e88dee3537887acdf3</td><td> Volume 6, Issue 5, May 2016 ISSN: 2277 128X +<br/>International Journal of Advanced Research in +<br/> Computer Science and Software Engineering +<br/> Research Paper +<br/> Available online at: www.ijarcsse.com +<br/>Semantic Based Hypergraph Reranking Model for Web +<br/>Image Search +<br/>1, 2, 3, 4 B. E. Dept of CSE, 5 Asst. Prof. Dept of CSE +<br/><b>Dr.D.Y.Patil College of Engineering, Pune, Maharashtra, India</b></td><td></td><td></td></tr><tr><td>19a9f658ea14701502d169dc086651b1d9b2a8ea</td><td>Structural Models for Face Detection +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences, China</b></td><td>('1721677', 'Junjie Yan', 'junjie yan')<br/>('2520795', 'Xucong Zhang', 'xucong zhang')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1716143', 'Dong Yi', 'dong yi')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jjyan,xczhang,zlei,dyi,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>197c64c36e8a9d624a05ee98b740d87f94b4040c</td><td>Regularized Greedy Column Subset Selection +<br/>aDepartment of Computer Systems, Universidad Polit´ecnica de Madrid +<br/>bDepartment of Applied Mathematics, Universidad Polit´ecnica de Madrid +</td><td>('1858768', 'Alberto Mozo', 'alberto mozo')</td><td>*bruno.ordozgoiti@upm.es +</td></tr><tr><td>19d4855f064f0d53cb851e9342025bd8503922e2</td><td>Learning SURF Cascade for Fast and Accurate Object Detection +<br/>Intel Labs China +</td><td>('35423937', 'Jianguo Li', 'jianguo li')<br/>('2470865', 'Yimin Zhang', 'yimin zhang')</td><td></td></tr><tr><td>19d3b02185ad36fb0b792f2a15a027c58ac91e8e</td><td>Im2Text: Describing Images Using 1 Million +<br/>Captioned Photographs +<br/>Tamara L Berg +<br/><b>Stony Brook University</b><br/>Stony Brook, NY 11794 +</td><td>('2004053', 'Vicente Ordonez', 'vicente ordonez')<br/>('2170826', 'Girish Kulkarni', 'girish kulkarni')</td><td>{vordonezroma or tlberg}@cs.stonybrook.edu +</td></tr><tr><td>193ec7bb21321fcf43bbe42233aed06dbdecbc5c</td><td>UC Santa Barbara +<br/>UC Santa Barbara Previously Published Works +<br/>Title +<br/>Automatic 3D facial expression analysis in videos +<br/>Permalink +<br/>https://escholarship.org/uc/item/3g44f7k8 +<br/>Authors +<br/>Chang, Y +<br/>Vieira, M +<br/>Turk, M +<br/>et al. +<br/>Publication Date +<br/>2005-01-01 +<br/>Peer reviewed +<br/>eScholarship.org +<br/>Powered by the California Digital Library +<br/><b>University of California</b></td><td></td><td></td></tr><tr><td>19da9f3532c2e525bf92668198b8afec14f9efea</td><td>Challenge: Face verification across age progression using real-world data +<br/>Video and Image Modeling and Synthesis Lab, Department of Computer Science, +<br/><b>University of Delaware, Newark, DE. USA</b><br/>1. Overview +<br/>Analysis of face images has been the topic of in-depth research with wide spread applications. Face recognition, verifi- +<br/>cation, age progression studies are some of the topics under study. In order to facilitate comparison and benchmarking of +<br/>different approaches, various datasets have been released. For the specific topics of face verification with age progression, +<br/>aging pattern extraction and age estimation, only two public datasets are currently available. The FGNET and MORPH +<br/>datasets contain a large number of subjects, but only a few images are available for each subject. We present a new dataset, +<br/>VADANA, which complements them by providing a large number of high quality digital images for each subject within and +<br/>across ages (depth vs. breadth). It provides the largest number of intra-personal pairs, essential for better training and testing. +<br/>The images also offer a natural range of pose, expression and illumination variation. We demonstrate the difference and +<br/>difficulty of VADANA by testing with state-of-the-art algorithms. Our findings from experiments show how VADANA can +<br/>aid further research on different types of verification algorithms. +<br/>The following sections provide details for the proposed challenge. The dataset details, the need and motivation for its +<br/>creation, comparison to existing benchmarks and the experiments performed on the same have been provided in the attached +<br/>paper. +<br/>2. Problem definition and challenges +<br/>There are various problems in facial image analysis, such as face detection (finding faces in a given image), face recogni- +<br/>tion (matching new image to a known dataset), face verification (determine if a given unknown pair of face images belong to +<br/>same person) and many others. In this work, we focus on face verification specifically in the case of age progression. +<br/>Problem definition: The input is a pair of facial images. The images are such that at least region from top of forehead till +<br/>the chin is covered. Though in general, the images cover from top of head region and part of neck region also. The identity +<br/>of the person(s) in the images is not known a priori. The system must determine if the two images belong to the same person +<br/>(intra-personal pair or intra-pair) or to different persons (extra-personal pair or extra-pair). The two images are taken across a +<br/>time period such that the age gap between the pair may range from 0 to 9 years. Also, the pose, expression and illumination +<br/>is uncontrolled for both images. +<br/>Training setup: During the training phase, the system is provided with pair of images (both intra-pairs and extra-pairs). +<br/>The age of the subject in a given image and thus the age gap between a pair is provided during training. A classifier is trained +<br/>using the features from the images. +<br/>Testing setup: During the testing phase, the input is a pair of images. The subjects in these pairs are different from those +<br/>in the training, i.e, the training and testing subjects are non-overlapping. There is no explicit age (or age-gap) information +<br/>provided at this stage. The system must classify the pair as either intra-personal or extra-personal. +<br/>Applications: The above problem definition closely resembles various real-world application scenarios such as passport +<br/>verification, security and surveillance matching in videos/image captured over a period of time, clustering of people in large +<br/>datasets where identities are unknown and many others. +<br/>Challenges: The challenges stem from various aspects of the above problem definition: (1) The subject identities are not +<br/>known, the system must therefore only rely on the information from the pair of images to determine the final classification. +<br/>(2) The images are taken at different times, ranging from a gap of few months to up to 9 years (as in the case of passport +<br/>verification). The effects due to aging thus contribute to shape and appearance changes even for an intra-pair (same person). +</td><td>('1692539', 'Gowri Somanath', 'gowri somanath')<br/>('1708413', 'Chandra Kambhamettu', 'chandra kambhamettu')</td><td>somanath,chandra@cis.udel.edu +</td></tr><tr><td>19868a469dc25ee0db00947e06c804b88ea94fd0</td><td>SP-SVM: Large Margin Classifier for Data on Multiple Manifolds +<br/><b>Purdue University, West Lafayette, IN. 47907, USA</b><br/><b>College of Information and Control Engineering, China University of Petroleum, Qingdao 266580, China</b><br/><b>Santa Clara University, Santa Clara, CA. 95053, USA</b><br/><b>cid:5)School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN. 47907, USA</b></td><td>('39274045', 'Bin Shen', 'bin shen')<br/>('1678435', 'Bao-Di Liu', 'bao-di liu')<br/>('34913796', 'Qifan Wang', 'qifan wang')<br/>('3047254', 'Yi Fang', 'yi fang')<br/>('1741931', 'Jan P. Allebach', 'jan p. allebach')</td><td>bshen@purdue.edu, thu.liubaodi@gmail.com, wang868@purdue.edu, yfang@scu.edu, allebach@ecn.purdue.edu +</td></tr><tr><td>192235f5a9e4c9d6a28ec0d333e36f294b32f764</td><td>Reconfiguring the Imaging Pipeline for Computer Vision +<br/><b>Cornell University</b><br/><b>Carnegie Mellon University</b><br/><b>Cornell University</b></td><td>('2328520', 'Mark Buckler', 'mark buckler')<br/>('39131476', 'Suren Jayasuriya', 'suren jayasuriya')<br/>('2138184', 'Adrian Sampson', 'adrian sampson')</td><td></td></tr><tr><td>19878141fbb3117d411599b1a74a44fc3daf296d</td><td>Eye-State Action Unit Detection by Gabor Wavelets +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA</b><br/><b>University of Pittsburgh, Pittsburgh, PA</b><br/>http://www.cs.cmu.edu/face +</td><td>('40383812', 'Ying-li Tian', 'ying-li tian')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>Email: fyltian, tkg@cs.cmu.edu +<br/>jeffcohn@pitt.edu +</td></tr><tr><td>19f076998ba757602c8fec04ce6a4ca674de0e25</td><td>Turk J Elec Eng & Comp Sci +<br/>(2016) 24: 219 { 233 +<br/>c⃝ T (cid:127)UB_ITAK +<br/>doi:10.3906/elk-1304-139 +<br/>Fast and de-noise support vector machine training method based on fuzzy +<br/>clustering method for large real world datasets +<br/>(cid:3) +<br/><b>Islamic Azad University, Gonabad, Iran</b><br/>Received: 15.04.2013 +<br/>(cid:15) +<br/>Accepted/Published Online: 30.10.2013 +<br/>(cid:15) +<br/>Final Version: 01.01.2016 +</td><td>('9437627', 'Omid Naghash ALMASI', 'omid naghash almasi')<br/>('4945660', 'Modjtaba ROUHANI', 'modjtaba rouhani')</td><td></td></tr><tr><td>19eb486dcfa1963c6404a9f146c378fc7ae3a1df</td><td></td><td></td><td></td></tr><tr><td>4c6daffd092d02574efbf746d086e6dc0d3b1e91</td><td></td><td></td><td></td></tr><tr><td>4cb8a691a15e050756640c0a35880cdd418e2b87</td><td>Class-based matching of object parts +<br/>Department of Computer Science and Applied Mathematics +<br/><b>Weizmann Institute of Science</b><br/>Rehovot, ISRAEL 76100 +</td><td>('1938475', 'Evgeniy Bart', 'evgeniy bart')<br/>('1743045', 'Shimon Ullman', 'shimon ullman')</td><td>(cid:0)evgeniy.bart, shimon.ullman(cid:1)@weizmann.ac.il +</td></tr><tr><td>4cc681239c8fda3fb04ba7ac6a1b9d85b68af31d</td><td>Mining Spatial and Spatio-Temporal ROIs for Action Recognition +<br/>Jiang Wang2 Alan Yuille1,3 +<br/><b>University of California, Los Angeles</b><br/><b>Baidu Research, USA 3John Hopkins University</b></td><td>('5964529', 'Xiaochen Lian', 'xiaochen lian')</td><td>{lianxiaochen@,yuille@stat.}ucla.edu +<br/>{chenzhuoyuan,yangyi05,wangjiang03}@baidu.com +</td></tr><tr><td>4c6e1840451e1f86af3ef1cb551259cb259493ba</td><td>HAND POSTURE DATASET CREATION FOR GESTURE +<br/>RECOGNITION +<br/>Instituto de Sistemas Inteligentes y Aplicaciones Numericas en Ingenieria +<br/>Campus Universitario de Tafira, 35017 Gran Canaria, Spain +<br/>Departamento de E.I.O. y Computacion +<br/>38271 Universidad de La Laguna, Spain +<br/>Keywords: +<br/>Image understanding, Gesture recognition, Hand dataset. +</td><td>('1706692', 'Luis Anton-Canalis', 'luis anton-canalis')<br/>('1797958', 'Elena Sanchez-Nielsen', 'elena sanchez-nielsen')</td><td>lanton@iusiani.ulpgc.es +<br/>enielsen@ull.es +</td></tr><tr><td>4c87aafa779747828054cffee3125fcea332364d</td><td>View-Constrained Latent Variable Model +<br/>for Multi-view Facial Expression Classification +<br/><b>Imperial College London, UK</b><br/><b>EEMCS, University of Twente, The Netherlands</b></td><td>('2308430', 'Stefanos Eleftheriadis', 'stefanos eleftheriadis')<br/>('1729713', 'Ognjen Rudovic', 'ognjen rudovic')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{s.eleftheriadis,o.rudovic,m.pantic}@imperial.ac.uk +</td></tr><tr><td>4c29e1f31660ba33e46d7e4ffdebb9b8c6bd5adc</td><td></td><td></td><td></td></tr><tr><td>4cdae53cebaeeebc3d07cf6cd36fecb2946f3e56</td><td>Photorealistic Facial Texture Inference Using Deep Neural Networks +<br/>*Pinscreen +<br/><b>University of Southern California</b><br/><b>USC Institute for Creative Technologies</b><br/>Figure 1: We present an inference framework based on deep neural networks for synthesizing photorealistic facial texture +<br/>along with 3D geometry from a single unconstrained image. We can successfully digitize historic figures that are no longer +<br/>available for scanning and produce high-fidelity facial texture maps with mesoscopic skin details. +</td><td>('2059597', 'Shunsuke Saito', 'shunsuke saito')<br/>('1792471', 'Lingyu Wei', 'lingyu wei')<br/>('1808579', 'Liwen Hu', 'liwen hu')<br/>('1897417', 'Koki Nagano', 'koki nagano')<br/>('1706574', 'Hao Li', 'hao li')</td><td></td></tr><tr><td>4c8e5fc0877d066516bb63e6c31eb1b8b5f967eb</td><td>MODI, KOVASHKA: CONFIDENCE AND DIVERSITY FOR ACTIVE SELECTION +<br/>Confidence and Diversity for Active +<br/>Selection of Feedback in Image Retrieval +<br/>Department of Computer Science +<br/><b>University of Pittsburgh</b><br/>Pittsburgh, PA, USA +</td><td>('1770205', 'Adriana Kovashka', 'adriana kovashka')</td><td>bhavin_modi@hotmail.com +<br/>kovashka@cs.pitt.edu +</td></tr><tr><td>4c8ef4f98c6c8d340b011cfa0bb65a9377107970</td><td>Sentiment Recognition in Egocentric +<br/>Photostreams +<br/><b>Intelligent Systems Group, University of Groningen, The Netherlands</b><br/><b>University of Barcelona, Spain</b><br/>3 Computer Vision Center, Barcelona, Spain +</td><td>('1742086', 'Nicola Strisciuglio', 'nicola strisciuglio')<br/>('1730388', 'Nicolai Petkov', 'nicolai petkov')<br/>('1724155', 'Petia Radeva', 'petia radeva')</td><td>e.talavera.martinez@rug.nl, +</td></tr><tr><td>4c822785c29ceaf67a0de9c699716c94fefbd37d</td><td>A Key Volume Mining Deep Framework for Action Recognition +<br/>2 SenseTime Group Limited +<br/><b>Tsinghua University</b><br/><b>Shenzhen Institutes of Advanced Technology, CAS, China</b><br/>Figure 1. Key volumes detected by our key volume mining deep framework. A volume is a spatial-temporal video clip. The top row shows +<br/>key volumes are very sparse among the whole video, and the second row shows that key volumes may come from different modalities +<br/>(different motion patterns here). Note that frames are sampled with fixed time interval. +</td><td>('2121584', 'Wangjiang Zhu', 'wangjiang zhu')<br/>('1748341', 'Jie Hu', 'jie hu')<br/>('1687740', 'Gang Sun', 'gang sun')<br/>('2032273', 'Xudong Cao', 'xudong cao')<br/>('40612284', 'Yu Qiao', 'yu qiao')</td><td></td></tr><tr><td>4c815f367213cc0fb8c61773cd04a5ca8be2c959</td><td>978-1-4244-4296-6/10/$25.00 ©2010 IEEE +<br/>2470 +<br/>ICASSP 2010 +</td><td></td><td></td></tr><tr><td>4ccf64fc1c9ca71d6aefdf912caf8fea048fb211</td><td>Light-weight Head Pose Invariant Gaze Tracking +<br/><b>University of Maryland</b><br/>NVIDIA +<br/>NVIDIA +</td><td>('48467498', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('24817039', 'Shalini De Mello', 'shalini de mello')<br/>('1690538', 'Jan Kautz', 'jan kautz')</td><td>rranjan1@umiacs.umd.edu +<br/>shalinig@nvidia.com +<br/>jkautz@nvidia.com +</td></tr><tr><td>4cdb6144d56098b819076a8572a664a2c2d27f72</td><td>Face Synthesis for Eyeglass-Robust Face +<br/>Recognition +<br/><b>CBSRandNLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, China</b></td><td>('46220439', 'Jianzhu Guo', 'jianzhu guo')<br/>('8362374', 'Xiangyu Zhu', 'xiangyu zhu')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jianzhu.guo,xiangyu.zhu,zlei,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>4c4e49033737467e28aa2bb32f6c21000deda2ef</td><td>Improving Landmark Localization with Semi-Supervised Learning +<br/><b>MILA-University of Montreal, 2NVIDIA, 3Ecole Polytechnique of Montreal, 4CIFAR, 5Facebook AI Research</b></td><td>('25056820', 'Sina Honari', 'sina honari')<br/>('2824500', 'Pavlo Molchanov', 'pavlo molchanov')<br/>('2342481', 'Stephen Tyree', 'stephen tyree')<br/>('1707326', 'Pascal Vincent', 'pascal vincent')<br/>('1690538', 'Jan Kautz', 'jan kautz')</td><td>1{honaris, vincentp}@iro.umontreal.ca, +<br/>2{pmolchanov, styree, jkautz}@nvidia.com, 3christopher.pal@polymtl.ca +</td></tr><tr><td>4c6233765b5f83333f6c675d3389bbbf503805e3</td><td>Real-time High Performance Deformable Model for Face Detection in the Wild +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences, China</b></td><td>('1721677', 'Junjie Yan', 'junjie yan')<br/>('2520795', 'Xucong Zhang', 'xucong zhang')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jjyan,xczhang,zlei,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>4c078c2919c7bdc26ca2238fa1a79e0331898b56</td><td>Unconstrained Facial Landmark Localization with Backbone-Branches +<br/>Fully-Convolutional Networks +<br/><b>Sun Yat-Sen University</b><br/>Guangzhou Higher Education Mega Center, Guangzhou 510006, PR China +</td><td>('1742286', 'Zhujin Liang', 'zhujin liang')<br/>('2442939', 'Shengyong Ding', 'shengyong ding')<br/>('1737218', 'Liang Lin', 'liang lin')</td><td>alfredtofu@gmail.com, marcding@163.com, linliang@ieee.org +</td></tr><tr><td>4cfa8755fe23a8a0b19909fa4dec54ce6c1bd2f7</td><td>EFFICIENT LIKELIHOOD BAYESIAN CONSTRAINED LOCAL MODEL +<br/><b>The Hong Kong Polytechnic University</b><br/><b>Hong Kong Applied Science and Technology Research Institute Company Limited, Hong Kong, China</b></td><td>('2116302', 'Hailiang Li', 'hailiang li')<br/>('1703078', 'Kin-Man Lam', 'kin-man lam')<br/>('3280193', 'Man-Yau Chiu', 'man-yau chiu')<br/>('2233216', 'Kangheng Wu', 'kangheng wu')<br/>('1982263', 'Zhibin Lei', 'zhibin lei')</td><td>harley.li@connect.polyu.hk,{harleyli, edmondchiu, khwu, lei}@astri.org, enkmlam@polyu.edu.hk +</td></tr><tr><td>4cac9eda716a0addb73bd7ffea2a5fb0e6ec2367</td><td>Representing Videos based on Scene Layouts +<br/>for Recognizing Agent-in-Place Actions +<br/><b>University of Maryland, College Park</b><br/>2Comcast Applied AI Research +<br/>3DeepMind +<br/>4Adobe Research +</td><td>('2180291', 'Ruichi Yu', 'ruichi yu')<br/>('3254319', 'Hongcheng Wang', 'hongcheng wang')<br/>('7674316', 'Jingxiao Zheng', 'jingxiao zheng')</td><td>{richyu, jxzheng, lsd}@umiacs.umd.edu +<br/>anglili@google.com morariu@adobe.com +</td></tr><tr><td>4c4236b62302957052f1bbfbd34dbf71ac1650ec</td><td>SEMI-SUPERVISED FACE RECOGNITION WITH LDA SELF-TRAINING +<br/>Multimedia Communications Department, EURECOM +<br/>2229 Route des Crêtes , BP 193, F-06560 Sophia-Antipolis Cedex, France +</td><td>('37560971', 'Xuran Zhao', 'xuran zhao')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td>{zhaox, evans, dugelay}@eurecom.fr +</td></tr><tr><td>4cd0da974af9356027a31b8485a34a24b57b8b90</td><td>Binarized Convolutional Landmark Localizers for Human Pose Estimation and +<br/>Face Alignment with Limited Resources +<br/><b>Computer Vision Laboratory, The University of Nottingham</b><br/>Nottingham, United Kingdom +</td><td>('3458121', 'Adrian Bulat', 'adrian bulat')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>{adrian.bulat, yorgos.tzimiropoulos}@nottingham.ac.uk +</td></tr><tr><td>4c170a0dcc8de75587dae21ca508dab2f9343974</td><td>FaceTracer: A Search Engine for +<br/>Large Collections of Images with Faces +<br/><b>Columbia University</b></td><td>('40631426', 'Neeraj Kumar', 'neeraj kumar')</td><td></td></tr><tr><td>4c81c76f799c48c33bb63b9369d013f51eaf5ada</td><td>Multi-modal Score Fusion and Decision Trees for Explainable Automatic Job +<br/>Candidate Screening from Video CVs +<br/><b>Nam k Kemal University, Tekirda g, Turkey</b><br/><b>Bo gazic i University, Istanbul, Turkey</b></td><td>('38007788', 'Heysem Kaya', 'heysem kaya')<br/>('1764521', 'Albert Ali Salah', 'albert ali salah')</td><td>hkaya@nku.edu.tr, furkan.gurpinar@boun.edu.tr,salah@boun.edu.tr +</td></tr><tr><td>4c1ce6bced30f5114f135cacf1a37b69bb709ea1</td><td>Gaze Direction Estimation by Component Separation for +<br/>Recognition of Eye Accessing Cues +<br/>Ruxandra Vrˆanceanu +<br/>Image Processing and Analysis Laboratory +<br/><b>University Politehnica of Bucharest, Romania, Address Splaiul Independent ei</b><br/>Image Processing and Analysis Laboratory +<br/><b>University Politehnica of Bucharest, Romania, Address Splaiul Independent ei</b><br/>Image Processing and Analysis Laboratory +<br/><b>University Politehnica of Bucharest, Romania, Address Splaiul Independent ei</b><br/>Image Processing and Analysis Laboratory +<br/><b>University Politehnica of Bucharest, Romania, Address Splaiul Independent ei</b></td><td>('2760434', 'Corneliu Florea', 'corneliu florea')<br/>('2143956', 'Laura Florea', 'laura florea')<br/>('2905899', 'Constantin Vertan', 'constantin vertan')</td><td>rvranceanu@imag.pub.ro +<br/>corneliu.florea@upb.ro +<br/>laura.florea@upb.ro +<br/>constantin.vertan@upb.ro +</td></tr><tr><td>4c5b38ac5d60ab0272145a5a4d50872c7b89fe1b</td><td>Facial Expression Recognition with Emotion-Based +<br/>Feature Fusion +<br/><b>The Hong Kong Polytechnic University, Hong Kong, SAR, 2University of Technology Sydney, Australia</b></td><td>('13671251', 'Cigdem Turan', 'cigdem turan')<br/>('1703078', 'Kin-Man Lam', 'kin-man lam')<br/>('1706670', 'Xiangjian He', 'xiangjian he')</td><td>E-mail: cigdem.turan@connect.polyu.hk, enkmlam@polyu.edu.hk, Xiangjian.He@uts.edu.au +</td></tr><tr><td>4c523db33c56759255b2c58c024eb6112542014e</td><td>Patch-based Within-Object Classification∗ +<br/><b>University College London</b><br/><b>MRC Laboratory For Molecular Cell Biology, University College London</b></td><td>('1904148', 'Jania Aghajanian', 'jania aghajanian')<br/>('1734784', 'Jonathan Warrell', 'jonathan warrell')<br/>('1695363', 'Peng Li', 'peng li')<br/>('32948556', 'Jennifer L. Rohn', 'jennifer l. rohn')<br/>('31046411', 'Buzz Baum', 'buzz baum')</td><td>1{j.aghajanian, j.warrell, s.prince, p.li}@cs.ucl.ac.uk 2{j.rohn, b.baum}@ucl.ac.uk +</td></tr><tr><td>261c3e30bae8b8bdc83541ffa9331b52fcf015e6</td><td>PATEL, SMITH: SFS+3DMM FOR FACE RECOGNITION +<br/>Shape-from-shading driven 3D Morphable +<br/>Models for Illumination Insensitive Face +<br/>Recognition +<br/>William A.P. Smith +<br/>Department of Computer Science, +<br/><b>The University of York</b></td><td>('37519514', 'Ankur Patel', 'ankur patel')</td><td>ankur@cs.york.ac.uk +<br/>wsmith@cs.york.ac.uk +</td></tr><tr><td>26f03693c50eb50a42c9117f107af488865f3dc1</td><td>Eigenhill vs. Eigenface and Eigenedge +<br/><b>Istanbul Technical University</b><br/>Department of Computer Science +</td><td>('1858702', 'Alper Yilmaz', 'alper yilmaz')<br/>('1766445', 'Muhittin Gökmen', 'muhittin gökmen')</td><td>yilmaz@cs.ucf.edu +<br/>gokmen@cs.itu.edu.tr +</td></tr><tr><td>2661f38aaa0ceb424c70a6258f7695c28b97238a</td><td>IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 42, NO. 4, AUGUST 2012 +<br/>1027 +<br/>Multilayer Architectures for Facial +<br/>Action Unit Recognition +</td><td>('4072965', 'Tingfan Wu', 'tingfan wu')<br/>('2593137', 'Nicholas J. Butko', 'nicholas j. butko')<br/>('12114845', 'Paul Ruvolo', 'paul ruvolo')<br/>('1775637', 'Jacob Whitehill', 'jacob whitehill')<br/>('1741200', 'Javier R. Movellan', 'javier r. movellan')</td><td></td></tr><tr><td>2609079d682998da2bc4315b55a29bafe4df414e</td><td>ON RANK AGGREGATION FOR FACE RECOGNITION FROM VIDEOS +<br/>IIIT-Delhi, India +</td><td>('2559473', 'Himanshu S. Bhatt', 'himanshu s. bhatt')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')</td><td></td></tr><tr><td>264a84f4d27cd4bca94270620907cffcb889075c</td><td>Deep Motion Features for Visual Tracking +<br/><b>Computer Vision Laboratory, Link oping University, Sweden</b></td><td>('8161428', 'Susanna Gladh', 'susanna gladh')<br/>('2488938', 'Martin Danelljan', 'martin danelljan')<br/>('2358803', 'Fahad Shahbaz Khan', 'fahad shahbaz khan')<br/>('2228323', 'Michael Felsberg', 'michael felsberg')</td><td></td></tr><tr><td>26d407b911d1234e8e3601e586b49316f0818c95</td><td>[POSTER] Feasibility of Corneal Imaging for Handheld Augmented Reality +<br/><b>Coburg University</b></td><td>('37101400', 'Daniel Schneider', 'daniel schneider')<br/>('2708269', 'Jens Grubert', 'jens grubert')</td><td></td></tr><tr><td>26a44feb7a64db7986473ca801c251aa88748477</td><td>Journal of Machine Learning Research 1 () +<br/>Submitted ; Published +<br/>Unsupervised Learning of Gaussian Mixture Models with a +<br/>Uniform Background Component +<br/>Department of Statistics +<br/><b>Florida State University</b><br/>Tallahassee, FL 32306-4330, USA +<br/>Department of Statistics +<br/><b>Florida State University</b><br/>Tallahassee, FL 32306-4330, USA +<br/>Editor: +</td><td>('2761870', 'Sida Liu', 'sida liu')<br/>('2455529', 'Adrian Barbu', 'adrian barbu')</td><td>sida.liu@stat.fsu.edu +<br/>abarbu@stat.fsu.edu +</td></tr><tr><td>264f7ab36ff2e23a1514577a6404229d7fe1242b</td><td>Facial Expression Recognition by De-expression Residue Learning +<br/>Department of Computer Science +<br/><b>State University of New York at Binghamton, USA</b></td><td>('2671017', 'Huiyuan Yang', 'huiyuan yang')<br/>('8072251', 'Lijun Yin', 'lijun yin')</td><td>{hyang51, uciftci}@binghamton.edu; lijun@cs.binghamton.edu +</td></tr><tr><td>26a72e9dd444d2861298d9df9df9f7d147186bcd</td><td>DOI 10.1007/s00138-016-0768-4 +<br/>ORIGINAL PAPER +<br/>Collecting and annotating the large continuous action dataset +<br/>Received: 18 June 2015 / Revised: 18 April 2016 / Accepted: 22 April 2016 / Published online: 21 May 2016 +<br/>© The Author(s) 2016. This article is published with open access at Springerlink.com +</td><td>('2089428', 'Daniel Paul Barrett', 'daniel paul barrett')</td><td></td></tr><tr><td>266766818dbc5a4ca1161ae2bc14c9e269ddc490</td><td>Article +<br/>Boosting a Low-Cost Smart Home Environment with +<br/>Usage and Access Control Rules +<br/><b>Institute of Information Science and Technologies of CNR (CNR-ISTI)-Italy, 56124 Pisa, Italy</b><br/>Received: 27 April 2018; Accepted: 31 May 2018; Published: 8 June 2018 +</td><td>('1773887', 'Paolo Barsocchi', 'paolo barsocchi')<br/>('38567341', 'Antonello Calabrò', 'antonello calabrò')<br/>('1693901', 'Erina Ferro', 'erina ferro')<br/>('2209975', 'Claudio Gennaro', 'claudio gennaro')<br/>('1709783', 'Eda Marchetti', 'eda marchetti')<br/>('2508924', 'Claudio Vairo', 'claudio vairo')</td><td>antonello.calabro@isti.cnr.it (A.C.); erina.ferro@isti.cnr.it (E.F.); claudio.gennaro@isti.cnr.it (C.G.); +<br/>eda.marchetti@isti.cnr.it (E.M.); claudio.vairo@isti.cnr.it (C.V.) +<br/>* Correspondence: paolo.barsocchi@isti.cnr.it; Tel.: +39-050-315-2965 +</td></tr><tr><td>265af79627a3d7ccf64e9fe51c10e5268fee2aae</td><td>1817 +<br/>A Mixture of Transformed Hidden Markov +<br/>Models for Elastic Motion Estimation +</td><td>('1932096', 'Huijun Di', 'huijun di')<br/>('3265275', 'Linmi Tao', 'linmi tao')<br/>('1797002', 'Guangyou Xu', 'guangyou xu')</td><td></td></tr><tr><td>267c6e8af71bab68547d17966adfaab3b4711e6b</td><td></td><td></td><td></td></tr><tr><td>26af867977f90342c9648ccf7e30f94470d40a73</td><td>IJIRST –International Journal for Innovative Research in Science & Technology| Volume 3 | Issue 04 | September 2016 +<br/>ISSN (online): 2349-6010 +<br/>Joint Gender and Face Recognition System for +<br/>RGB-D Images with Texture and DCT Features +<br/>PG Student +<br/>Department of Computer Science & Information Systems +<br/><b>Federal Institute of Science and Technology, Mookkannoor</b><br/>PO, Angamaly, Ernakulam, Kerala 683577, India +<br/>Prasad J. C. +<br/>Associate Professor +<br/>Department of Computer Science & Engineering +<br/><b>Federal Institute of Science and Technology, Mookkannoor</b><br/>PO, Angamaly, Ernakulam, Kerala 683577, India +</td><td></td><td></td></tr><tr><td>26a89701f4d41806ce8dbc8ca00d901b68442d45</td><td></td><td></td><td></td></tr><tr><td>26c884829897b3035702800937d4d15fef7010e4</td><td>IEICE TRANS. INF. & SYST., VOL.Exx–??, NO.xx XXXX 200x +<br/>PAPER +<br/>Facial Expression Recognition by Supervised Independent +<br/>Component Analysis using MAP Estimation +<br/>, Member +<br/>SUMMARY Permutation ambiguity of the classical Inde- +<br/>pendent Component Analysis (ICA) may cause problems in fea- +<br/>ture extraction for pattern classification. Especially when only a +<br/>small subset of components is derived from data, these compo- +<br/>nents may not be most distinctive for classification, because ICA +<br/>is an unsupervised method. We include a selective prior for de- +<br/>mixing coefficients into the classical ICA to alleviate the problem. +<br/>Since the prior is constructed upon the classification information +<br/>from the training data, we refer to the proposed ICA model with +<br/>a selective prior as a supervised ICA (sICA). We formulated the +<br/>learning rule for sICA by taking a Maximum a Posteriori (MAP) +<br/>scheme and further derived a fixed point algorithm for learning +<br/>the de-mixing matrix. We investigate the performance of sICA +<br/>in facial expression recognition from the aspects of both correct +<br/>rate of recognition and robustness even with few independent +<br/>components. +<br/>key words: +<br/>dent component analysis, fixed-point algorithm +<br/>facial expression recognition, supervised indepen- +<br/>1. +<br/>Introduction +<br/>Various methods have been proposed for auto- +<br/>matic recognition of facial expression in the past several +<br/>decades, which could be roughly classified into three +<br/>categories: 1) Appearance-based method, represented +<br/>by eigenfaces, fisherfaces and other methods using +<br/>machine-learning techniques, such as neural networks +<br/>and Support Vector Machine (SVM); 2) Model-based +<br/><b>methods, including graph matching, optical- ow-based</b><br/>method and others; and 3) Hybrids of appearance based +<br/>and model-based methods, such as Active Appearance +<br/>Model (AAM). Detailed review of these methods could +<br/>be found in two surveys in Refs.[1][2]. Appearance- +<br/>based methods are superior to model-based methods +<br/>in system complexity and performance reproducibil- +<br/>ity. Further, appearance-based methods allow efficient +<br/>characterization of a low-dimensional subspace within +<br/>the overall space of raw image measurement, which +<br/>deepen our understanding of facial expressions from +<br/>their manifolds in subspace, and provide a statistical +<br/>framework for the theoretical analysis of system per- +<br/>formance. ICA, a powerful technique for blind source +<br/>separation, was applied to facial expression recognition +<br/>by Bartlett et al. for feature extraction.[3] They argued +<br/>that facial expression consists of those features standing +<br/>for minor, non-rigid, local variations of faces[3]. Struc- +<br/>Manuscript received January 1, 200x. +<br/>Manuscript revised January 1, 200x. +<br/>Final manuscript received January 1, 200x. +<br/>The author is with the school of information science, +<br/><b>Japan Advanced Institute of Science and Technology</b><br/>tural information for those local variations are related +<br/>to higher-order statistics, which could be well extracted +<br/>by ICA.[5] The efficiency of ICA in extracting features +<br/>for facial expression recognition has been verified by +<br/>many previous works.[4][6][7] +<br/>The major purpose of the present work is to im- +<br/>prove the performance of ICA in facial expression recog- +<br/>nition. In the classical ICA, the derived independent +<br/>components are in random order, i.e., permutation am- +<br/>biguity, where the original order provides no informa- +<br/>tion on the significance of components in discrimina- +<br/>tion.[8] As a result, the derived independent compo- +<br/>nents may not be most distinctive for the classification +<br/>task, especially when only a small subset of compo- +<br/>nents is derived. Feature selection must be performed +<br/>along with the feature extraction. The selection can +<br/>be applied before, during or after ICA. In Ref.[4], Best +<br/>Individual Feature (BIF) selection was adopted, where +<br/>features were chosen according to some defined criteria +<br/>individually. Methods by means of Sequential Forward +<br/>Selection (SFS) and Sequential Floating Forward Se- +<br/>lection (SFFS) were also proposed. [9] Since the selec- +<br/>tion is performed after ICA, the features are limited to +<br/>those chosen from the set of independent components +<br/>obtained. To create a candidate set with enough repre- +<br/>sentative features for discrimination, a large number of +<br/>independent components should be learned, which may +<br/>be expensive in computational cost. It is meaningful +<br/>to search for a way to affect the selection of features +<br/>before or during ICA. GEMC [10] makes a selection +<br/>before ICA by heuristically replacing PCA with a dis- +<br/>criminant analysis as the pre-processing to ICA, which +<br/>still lacks a mathematical explanation. ICA in a local +<br/>facial residue space is also proposed for face recognition, +<br/>which can be regarded as using the pre-specified residue +<br/>space to limit the selection of independent components +<br/>before applying ICA. [11] +<br/>We propose an approach to implement the feature +<br/>selection during the learning of independent compo- +<br/>nents. A constraint ICA has been proposed for the +<br/>analysis of EEG signals, where all components should +<br/>be sparse and close to a supplied reference signal by +<br/>including a correlation term. [12] In our case, we try to +<br/>design a method to let those components with higher +<br/>degrees of class separation emerge easier than others. +<br/>The classical ICA in Ref.[13] was shown to be deriv- +<br/>able under the scheme of Maximum Log-Likelihood +</td><td>('1753878', 'Fan Chen', 'fan chen')<br/>('1791753', 'Kazunori Kotani', 'kazunori kotani')</td><td></td></tr><tr><td>266ed43dcea2e7db9f968b164ca08897539ca8dd</td><td>Beyond Principal Components: Deep Boltzmann Machines for Face Modeling +<br/><b>Concordia University, Computer Science and Software Engineering, Montr eal, Qu ebec, Canada</b><br/><b>Carnegie Mellon University, CyLab Biometrics Center, Pittsburgh, PA, USA</b></td><td>('1876581', 'Chi Nhan Duong', 'chi nhan duong')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('2687827', 'Kha Gia Quach', 'kha gia quach')<br/>('1699922', 'Tien D. Bui', 'tien d. bui')</td><td>1 {c duon, k q, bui}@encs.concordia.ca, 2 kluu@andrew.cmu.edu +</td></tr><tr><td>26ad6ceb07a1dc265d405e47a36570cb69b2ace6</td><td>RESEARCH AND EXPLOR ATORY +<br/>DEVELOPMENT DEPARTMENT +<br/>REDD-2015-384 +<br/>Neural Correlates of Cross-Cultural +<br/>How to Improve the Training and Selection for +<br/>Military Personnel Involved in Cross-Cultural +<br/>Operating Under Grant #N00014-12-1-0629/113056 +<br/>Adaptation +<br/>September, 2015 +<br/>Interactions +<br/>Prepared for: +<br/>Office of Naval Research +</td><td>('20444535', 'Jonathon Kopecky', 'jonathon kopecky')<br/>('29125372', 'Alice Jackson', 'alice jackson')</td><td></td></tr><tr><td>2642810e6c74d900f653f9a800c0e6a14ca2e1c7</td><td>Projection Bank: From High-dimensional Data to Medium-length Binary Codes +<br/>Department of Computer Science and Digital Technologies +<br/><b>Northumbria University, Newcastle upon Tyne, NE1 8ST, UK</b></td><td>('40017778', 'Li Liu', 'li liu')<br/>('9452165', 'Mengyang Yu', 'mengyang yu')<br/>('40799321', 'Ling Shao', 'ling shao')</td><td>li2.liu@northumbria.ac.uk, m.y.yu@ieee.org, ling.shao@ieee.org +</td></tr><tr><td>26437fb289cd7caeb3834361f0cc933a02267766</td><td>2012 International Conference on Management and Education Innovation +<br/>IPEDR vol.37 (2012) © (2012) IACSIT Press, Singapore +<br/>Innovative Assessment Technologies: Comparing ‘Face-to-Face’ and +<br/>Game-Based Development of Thinking Skills in Classroom Settings +<br/><b>University of Szeged, 2 E tv s Lor nd University</b></td><td>('39201903', 'Gyöngyvér Molnár', 'gyöngyvér molnár')<br/>('32197908', 'András Lőrincz', 'andrás lőrincz')</td><td></td></tr><tr><td>26e570049aaedcfa420fc8c7b761bc70a195657c</td><td>J Sign Process Syst +<br/>DOI 10.1007/s11265-017-1276-0 +<br/>Hybrid Facial Regions Extraction for Micro-expression +<br/>Recognition System +<br/>Received: 2 February 2016 / Revised: 20 October 2016 / Accepted: 10 August 2017 +<br/>© Springer Science+Business Media, LLC 2017 +</td><td>('39888137', 'Sze-Teng Liong', 'sze-teng liong')<br/>('2339975', 'John See', 'john see')<br/>('37809010', 'Su-Wei Tan', 'su-wei tan')</td><td></td></tr><tr><td>2654ef92491cebeef0997fd4b599ac903e48d07a</td><td>Facial Expression Recognition from Near-Infrared Video Sequences +<br/>1. Machine Vision Group, Infotech Oulu and Department of Electrical and Information +<br/>Engineering, +<br/><b>P. O. Box 4500 FI-90014 University of Oulu, Finland</b><br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/>P. O. Box 95 Zhongguancun Donglu, Beijing 100080, China +</td><td>('2021982', 'Matti Taini', 'matti taini')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')<br/>('34679741', 'Stan Z. Li', 'stan z. li')<br/>('1714724', 'Matti Pietikäinen', 'matti pietikäinen')</td><td>E-mail: {matti.taini,gyzhao,mkp}@ee.oulu.fi +<br/>E-mail: szli@cbsr.ia.ac.cn +</td></tr><tr><td>2679e4f84c5e773cae31cef158eb358af475e22f</td><td>Adaptive Deep Metric Learning for Identity-Aware Facial Expression Recognition +<br/><b>Carnegie Mellon University, Pittsburgh, PA</b><br/><b>Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Science</b><br/><b>The Hong Kong Polytechnic University, Hong Kong, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, China</b></td><td>('1790207', 'Xiaofeng Liu', 'xiaofeng liu')<br/>('1748883', 'Jane You', 'jane you')<br/>('37774211', 'Ping Jia', 'ping jia')</td><td>liuxiaofeng@cmu.edu, kumar@ece.cmu.edu, csyjia@comp.polyu.edu.hk, jiap@ciomp.ac.cn +</td></tr><tr><td>21ef129c063bad970b309a24a6a18cbcdfb3aff5</td><td>POUR L'OBTENTION DU GRADE DE DOCTEUR ÈS SCIENCESacceptée sur proposition du jury:Dr J.-M. Vesin, président du juryProf. J.-Ph. Thiran, Prof. D. Sander, directeurs de thèseProf. M. F. Valstar, rapporteurProf. H. K. Ekenel, rapporteurDr S. Marcel, rapporteurIndividual and Inter-related Action Unit Detection in Videos for Affect RecognitionTHÈSE NO 6837 (2016)ÉCOLE POLYTECHNIQUE FÉDÉRALE DE LAUSANNEPRÉSENTÉE LE 19 FÉVRIER 2016À LA FACULTÉ DES SCIENCES ET TECHNIQUES DE L'INGÉNIEURLABORATOIRE DE TRAITEMENT DES SIGNAUX 5PROGRAMME DOCTORAL EN GÉNIE ÉLECTRIQUE Suisse2016PARAnıl YÜCE </td><td></td><td></td></tr><tr><td>218b2c5c9d011eb4432be4728b54e39f366354c1</td><td>Enhancing Training Collections for Image +<br/>Annotation: An Instance-Weighted Mixture +<br/>Modeling Approach +</td><td>('1793498', 'Neela Sawant', 'neela sawant')<br/>('40116905', 'Jia Li', 'jia li')</td><td></td></tr><tr><td>217a21d60bb777d15cd9328970cab563d70b5d23</td><td>Hidden Factor Analysis for Age Invariant Face Recognition +<br/>1Shenzhen Key Lab of Computer Vision and Pattern Recognition +<br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, China</b><br/><b>Toyota Technological Institute at Chicago</b><br/><b>The Chinese University of Hong Kong</b><br/>4Media Lab, Huawei Technologies Co. Ltd., China +</td><td>('2856494', 'Dihong Gong', 'dihong gong')<br/>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('1807606', 'Dahua Lin', 'dahua lin')<br/>('7137861', 'Jianzhuang Liu', 'jianzhuang liu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>dh.gong@siat.ac.cn +<br/>zhifeng.li@siat.ac.cn +<br/>dhlin@ttic.edu +<br/>liu.jianzhuang@huawei.com +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>21e828071249d25e2edaca0596e27dcd63237346</td><td></td><td></td><td></td></tr><tr><td>21a2f67b21905ff6e0afa762937427e92dc5aa0b</td><td>Hindawi +<br/>Computational Intelligence and Neuroscience +<br/>Volume 2017, Article ID 8710492, 13 pages +<br/>https://doi.org/10.1155/2017/8710492 +<br/>Research Article +<br/>Extra Facial Landmark Localization via +<br/>Global Shape Reconstruction +<br/><b>School of Automation Engineering, University of Electronic Science and Technology of China, No. 2006, Xiyuan Ave</b><br/>West Hi-Tech Zone, Chengdu 611731, China +<br/>Received 4 January 2017; Revised 26 March 2017; Accepted 4 April 2017; Published 23 April 2017 +<br/>Academic Editor: Elio Masciari +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Localizing facial landmarks is a popular topic in the field of face analysis. However, problems arose in practical applications such +<br/>as handling pose variations and partial occlusions while maintaining moderate training model size and computational efficiency +<br/>still challenges current solutions. In this paper, we present a global shape reconstruction method for locating extra facial landmarks +<br/>comparing to facial landmarks used in the training phase. In the proposed method, the reduced configuration of facial landmarks +<br/>is first decomposed into corresponding sparse coefficients. Then explicit face shape correlations are exploited to regress between +<br/>sparse coefficients of different facial landmark configurations. Finally extra facial landmarks are reconstructed by combining the +<br/>pretrained shape dictionary and the approximation of sparse coefficients. By applying the proposed method, both the training +<br/>time and the model size of a class of methods which stack local evidences as an appearance descriptor can be scaled down with +<br/>only a minor compromise in detection accuracy. Extensive experiments prove that the proposed method is feasible and is able to +<br/>reconstruct extra facial landmarks even under very asymmetrical face poses. +<br/>1. Introduction +<br/>Facial landmark localization is the first and a crucial step for +<br/>many face analysis tasks such as face recognition [1], cartoon +<br/>facial animation [2, 3], and facial expression understanding +<br/>[4, 5]. Most facial landmarks are located along the dominant +<br/>contours around facial features like eyebrows, nose, and +<br/>mouth. Therefore facial landmarks on a face image jointly +<br/>describe a face shape which lies in the shape space [6]. +<br/>For the last ten years remarkable progress has been +<br/>made in the field of facial +<br/>landmark localization [7, 8]. +<br/>Among a large number of proposed methods, the most +<br/>popular solution is to treat the facial landmark localiza- +<br/>tion problem as a holistic shape regression process and +<br/>to learn a general regression model from labeled training +<br/>images [9, 10]. Following this shape regression idea, various +<br/>methods try to model a regression function that directly +<br/>maps the appearance of images to landmark coordinates +<br/>without the need of computing a parametric model. All +<br/>facial landmarks in a shape are iterated collectively and the +<br/>relationship between facial landmarks is flexibly embedded +<br/>into the iteration process. On the other hand, to generate +<br/>enough description of face images, multiscale local feature +<br/>descriptors are typically adopted in most shape regression +<br/>methods. For example, cascaded pose regression (CPR) [7] +<br/>was first proposed to estimate general object poses with pose- +<br/>indexed features and then extended for the problem of face +<br/>alignment in explicit shape regression (ESR) [11] method. +<br/>ESR combines two-level boosting regression, shape-indexed +<br/>features, and correlation-based feature selection. As another +<br/>example, supervised descent method (SDM) [12] and its +<br/>extensions also have shown an impressive performance in the +<br/>field of facial landmark localization. These kinds of methods +<br/>stack shape-indexed high dimension feature descriptors and +<br/>train regression functions from a supervised gradient descent +<br/>view. +<br/>However, facial landmark localization still meets great +<br/>challenges in practical applications, such as handling pose +<br/>variations and partial occlusion while maintaining moderate +<br/>training model size and computational efficiency. In SDM +<br/>and its improved methods, the dimension of regression +</td><td>('9684590', 'Shuqiu Tan', 'shuqiu tan')<br/>('2915473', 'Dongyi Chen', 'dongyi chen')<br/>('9486108', 'Chenggang Guo', 'chenggang guo')<br/>('2122143', 'Zhiqi Huang', 'zhiqi huang')<br/>('9684590', 'Shuqiu Tan', 'shuqiu tan')</td><td>Correspondence should be addressed to Dongyi Chen; dychen@uestc.edu.cn +</td></tr><tr><td>2162654cb02bcd10794ae7e7d610c011ce0fb51b</td><td>4697 +<br/>978-1-4799-5751-4/14/$31.00 ©2014 IEEE +<br/>1http://www.skype.com/ +<br/>2http://www.google.com/hangouts/ +<br/>tification, sparse coding +</td><td></td><td></td></tr><tr><td>21258aa3c48437a2831191b71cd069c05fb84cf7</td><td>A Robust and E(cid:14)cient Doubly Regularized +<br/>Metric Learning Approach +<br/>Siemens Corporate Research, Princeton, NJ, 08540 +<br/><b>CISE, University of Florida, Gainesville, FL</b></td><td>('35582088', 'Meizhu Liu', 'meizhu liu')<br/>('1733005', 'Baba C. Vemuri', 'baba c. vemuri')</td><td></td></tr><tr><td>21f3c5b173503185c1e02a3eb4e76e13d7e9c5bc</td><td>m a s s a c h u s e t t s i n s t i t u t e o f +<br/>t e c h n o l o g y — a r t i f i c i a l i n t e l l i g e n c e l a b o r a t o r y +<br/>Rotation Invariant Real-time +<br/>Face Detection and +<br/>Recognition System +<br/>AI Memo 2001-010 +<br/>CBCL Memo 197 +<br/>May 31, 2001 +<br/>© 2 0 0 1 m a s s a c h u s e t t s i n s t i t u t e o f +<br/>t e c h n o l o g y, c a m b r i d g e , m a 0 2 1 3 9 u s a — w w w. a i . m i t . e d u +</td><td>('35541734', 'Purdy Ho', 'purdy ho')</td><td></td></tr><tr><td>21bd9374c211749104232db33f0f71eab4df35d5</td><td>Integrating Facial Makeup Detection Into +<br/>Multimodal Biometric User Verification System +<br/>CuteSafe Technology Inc. +<br/>Gebze, Kocaeli, Turkey +<br/>Eurecom Digital Security Department +<br/>06410 Biot, France +</td><td>('39935459', 'Ekberjan Derman', 'ekberjan derman')<br/>('3179061', 'Chiara Galdi', 'chiara galdi')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td>ekberjan.derman@cutesafe.com +<br/>{chiara.galdi, jean-luc.dugelay}@eurecom.fr +</td></tr><tr><td>214db8a5872f7be48cdb8876e0233efecdcb6061</td><td>Semantic-aware Co-indexing for Image Retrieval +<br/><b>NEC Laboratories America, Inc</b><br/>2Dept. of CS, Univ. of Texas at San Antonio +<br/>Cupertino, CA 95014 +<br/>San Antonio, TX 78249 +</td><td>('1776581', 'Shiliang Zhang', 'shiliang zhang')<br/>('2909406', 'Ming Yang', 'ming yang')<br/>('3991189', 'Xiaoyu Wang', 'xiaoyu wang')<br/>('1695082', 'Yuanqing Lin', 'yuanqing lin')<br/>('1713616', 'Qi Tian', 'qi tian')</td><td>{myang,xwang,ylin}@nec-labs.com +<br/>slzhang.jdl@gmail.com qitian@cs.utsa.edu +</td></tr><tr><td>21104bcf07ef0269ab133471a3200b9bf94b2948</td><td>Beyond Comparing Image Pairs: Setwise Active Learning for Relative Attributes +<br/><b>University of Texas at Austin</b></td><td>('2548555', 'Lucy Liang', 'lucy liang')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>214ac8196d8061981bef271b37a279526aab5024</td><td>Face Recognition Using Smoothed High-Dimensional +<br/>Representation +<br/>Center for Machine Vision Research, PO Box 4500, +<br/><b>FI-90014 University of Oulu, Finland</b></td><td>('32683737', 'Juha Ylioinas', 'juha ylioinas')<br/>('1776374', 'Juho Kannala', 'juho kannala')<br/>('1751372', 'Abdenour Hadid', 'abdenour hadid')</td><td></td></tr><tr><td>213a579af9e4f57f071b884aa872651372b661fd</td><td>Int J Comput Vis +<br/>DOI 10.1007/s11263-013-0672-6 +<br/>Automatic and Efficient Human Pose Estimation for Sign +<br/>Language Videos +<br/>Received: 4 February 2013 / Accepted: 29 October 2013 +<br/>© Springer Science+Business Media New York 2013 +</td><td>('36326860', 'James Charles', 'james charles')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>21626caa46cbf2ae9e43dbc0c8e789b3dbb420f1</td><td>978-1-4673-2533-2/12/$26.00 ©2012 IEEE +<br/>1437 +<br/>ICIP 2012 +</td><td></td><td></td></tr><tr><td>217de4ff802d4904d3f90d2e24a29371307942fe</td><td>POOF: Part-Based One-vs-One Features for Fine-Grained Categorization, Face +<br/>Verification, and Attribute Estimation +<br/><b>Columbia University</b><br/><b>Columbia University</b></td><td>('1778562', 'Thomas Berg', 'thomas berg')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')</td><td>tberg@cs.columbia.edu +<br/>belhumeur@cs.columbia.edu +</td></tr><tr><td>2135a3d9f4b8f5771fa5fc7c7794abf8c2840c44</td><td>Lessons from Collecting a Million Biometric Samples +<br/><b>University of Notre Dame</b><br/>Notre Dame, IN 46556, USA +<br/><b>National Institute of Standards and Technology</b><br/>Gaithersburg, MD 20899, USA +</td><td>('1704876', 'Patrick J. Flynn', 'patrick j. flynn')<br/>('1799014', 'Kevin W. Bowyer', 'kevin w. bowyer')<br/>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')</td><td>flynn@cse.nd.edu +<br/>kwb@cse.nd.edu +<br/>jonathon@nist.gov +</td></tr><tr><td>210b98394c3be96e7fd75d3eb11a391da1b3a6ca</td><td>Spatiotemporal Derivative Pattern: A Dynamic +<br/>Texture Descriptor for Video Matching +<br/>Saeed Mian3 +<br/><b>Tafresh University, Tafresh, Iran</b><br/><b>Electrical Eng. Dep., Central Tehran Branch, Islamic Azad University, Tehran, Iran</b><br/><b>Computer Science and Software Engineering, The University of Western Australia</b><br/>WA 6009, Australia +</td><td>('3046235', 'Farshid Hajati', 'farshid hajati')<br/>('2014145', 'Mohammad Tavakolian', 'mohammad tavakolian')<br/>('2997971', 'Soheila Gheisari', 'soheila gheisari')</td><td>{hajati,m_tavakolian}@tafreshu.ac.ir +<br/>s.gheisari@iauctb.ac.ir +<br/>ajmal.mian@uwa.edu.au +</td></tr><tr><td>21765df4c0224afcc25eb780bef654cbe6f0bc3a</td><td>Multi-Channel Correlation Filters +<br/><b>National University of Singapore</b><br/><b>National University of Singapore</b><br/>Singapore +<br/>Singapore +<br/>CSIRO +<br/>Australia +</td><td>('2860592', 'Hamed Kiani Galoogahi', 'hamed kiani galoogahi')<br/>('1715286', 'Terence Sim', 'terence sim')<br/>('1820249', 'Simon Lucey', 'simon lucey')</td><td>hkiani@comp.nus.edu.sg +<br/>tsim@comp.nus.edu.sg +<br/>simon.lucey@csiro.au +</td></tr><tr><td>21b16df93f0fab4864816f35ccb3207778a51952</td><td>Recognition of Static Gestures applied to Brazilian Sign Language (Libras) +<br/><b>Math Institute</b><br/>Department of Technology, Department of Exact Sciences +<br/><b>Federal University of Bahia (UFBA</b><br/><b>State University of Feira de Santana (UEFS</b><br/>Salvador, Brazil +<br/>Feira de Santana, Brazil +</td><td>('2009399', 'Igor L. O. Bastos', 'igor l. o. bastos')<br/>('3057269', 'Michele F. Angelo', 'michele f. angelo')<br/>('2563043', 'Angelo C. Loula', 'angelo c. loula')</td><td>igorcrexito@gmail.com +<br/>mfangelo@uefs.ecomp.br, angelocl@gmail.com +</td></tr><tr><td>212608e00fc1e8912ff845ee7a4a67f88ba938fc</td><td>Coupled Deep Learning for Heterogeneous Face Recognition +<br/>Center for Research on Intelligent Perception and Computing (CRIPAC), +<br/>National Laboratory of Pattern Recognition (NLPR), +<br/><b>Institute of Automation, Chinese Academy of Sciences, Beijing, P. R. China</b></td><td>('2225749', 'Xiang Wu', 'xiang wu')<br/>('3051419', 'Lingxiao Song', 'lingxiao song')<br/>('1705643', 'Ran He', 'ran he')<br/>('1688870', 'Tieniu Tan', 'tieniu tan')</td><td>alfredxiangwu@gmail.com, {lingxiao.song, rhe, tnt}@nlpr.ia.ac.cn +</td></tr><tr><td>4d49c6cff198cccb21f4fa35fd75cbe99cfcbf27</td><td>Topological Principal Component Analysis for +<br/>face encoding and recognition +<br/>Juan J. Villanueva +<br/>Computer Vision Center and Departament d’Inform(cid:18)atica, Edi(cid:12)ci O, Universitat +<br/>Aut(cid:18)onoma de Barcelona +</td><td>('38034605', 'Albert Pujol', 'albert pujol')<br/>('2997661', 'Felipe Lumbreras', 'felipe lumbreras')</td><td></td></tr><tr><td>4d625677469be99e0a765a750f88cfb85c522cce</td><td>Understanding Hand-Object Manipulation +<br/>with Grasp Types and Object Attributes +<br/><b>Institute of Industrial Science</b><br/><b>The University of Tokyo, Japan</b><br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University, USA</b><br/><b>Institute of Industrial Science</b><br/><b>The University of Tokyo, Japan</b></td><td>('3172280', 'Minjie Cai', 'minjie cai')<br/>('37991449', 'Kris M. Kitani', 'kris m. kitani')<br/>('9467266', 'Yoichi Sato', 'yoichi sato')</td><td>cai-mj@iis.u-tokyo.ac.jp +<br/>kkitani@cs.cmu.edu +<br/>ysato@iis.u-tokyo.ac.jp +</td></tr><tr><td>4da735d2ed0deeb0cae4a9d4394449275e316df2</td><td>Gothenburg, Sweden, June 19-22, 2016 +<br/>978-1-5090-1820-8/16/$31.00 ©2016 IEEE +<br/>1410 +</td><td></td><td></td></tr><tr><td>4d15254f6f31356963cc70319ce416d28d8924a3</td><td>Quo vadis Face Recognition? +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>Department of Psychology +<br/><b>University of Pittsburgh</b><br/>Pittsburgh, PA 15260 +</td><td>('33731953', 'Ralph Gross', 'ralph gross')<br/>('1838212', 'Jianbo Shi', 'jianbo shi')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>frgross,jshig@cs.cmu.edu +<br/>jeffcohn@pitt.edu +</td></tr><tr><td>4d530a4629671939d9ded1f294b0183b56a513ef</td><td>International Journal of Machine Learning and Computing, Vol. 2, No. 4, August 2012 +<br/>Facial Expression Classification Method Based on Pseudo +<br/>Zernike Moment and Radial Basis Function Network +<br/> +</td><td>('2009230', 'Tran Binh Long', 'tran binh long')<br/>('2710459', 'Le Hoang Thai', 'le hoang thai')<br/>('1971778', 'Tran Hanh', 'tran hanh')</td><td></td></tr><tr><td>4d2975445007405f8cdcd74b7fd1dd547066f9b8</td><td>Image and Video Processing +<br/>for Affective Applications +</td><td>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>4df889b10a13021928007ef32dc3f38548e5ee56</td><td></td><td></td><td></td></tr><tr><td>4d6462fb78db88afff44561d06dd52227190689c</td><td>Face-to-Face Social Activity Detection Using +<br/>Data Collected with a Wearable Device +<br/>1 Computer Vision Center, Campus UAB, Edifici O, Bellaterra, Barcelona, Spain +<br/><b>Dep. of Applied Mathematics and Analysis, University of Barcelona, Spain</b><br/>http://www.cvc.uab.es, http://www.maia.ub.es +</td><td>('7629833', 'Pierluigi Casale', 'pierluigi casale')<br/>('9783922', 'Oriol Pujol', 'oriol pujol')<br/>('1724155', 'Petia Radeva', 'petia radeva')</td><td>pierluigi@cvc.uab.es +</td></tr><tr><td>4d423acc78273b75134e2afd1777ba6d3a398973</td><td></td><td></td><td></td></tr><tr><td>4db9e5f19366fe5d6a98ca43c1d113dac823a14d</td><td>Combining Crowdsourcing and Face Recognition to Identify Civil War Soldiers +<br/>Are 1,000 Features Worth A Picture? +<br/>Department of Computer Science and Center for Human-Computer Interaction +<br/>Virginia Tech, Arlington, VA, USA +</td><td>('32698591', 'Vikram Mohanty', 'vikram mohanty')<br/>('51219402', 'David Thames', 'david thames')<br/>('2427623', 'Kurt Luther', 'kurt luther')</td><td></td></tr><tr><td>4dd6d511a8bbc4d9965d22d79ae6714ba48c8e41</td><td></td><td></td><td></td></tr><tr><td>4de757faa69c1632066391158648f8611889d862</td><td>International Journal of Advanced Engineering Research and Science (IJAERS) Vol-3, Issue-3 , March- 2016] +<br/>ISSN: 2349-6495 +<br/>Review of Face Recognition Technology Using +<br/>Feature Fusion Vector +<br/><b>S.R.C.E.M, Banmore, RGPV, University, Bhopal, Madhya Pradesh, India</b><br/> +</td><td></td><td></td></tr><tr><td>4dd71a097e6b3cd379d8c802460667ee0cbc8463</td><td>Real-time Multi-view Facial Landmark Detector +<br/>Learned by the Structured Output SVM +<br/>1 Center for Machine Perception, Department of Cybernetics, Faculty of Electrical Engineering, Czech +<br/><b>Technical University in Prague, 166 27 Prague 6, Technick a 2 Czech Republic</b><br/><b>National Institute of Informatics, Tokyo, Japan</b></td><td>('39492787', 'Diego Thomas', 'diego thomas')<br/>('1691286', 'Akihiro Sugimoto', 'akihiro sugimoto')</td><td></td></tr><tr><td>4db0968270f4e7b3fa73e41c50d13d48e20687be</td><td>Fashion Forward: Forecasting Visual Style in Fashion +<br/><b>Karlsruhe Institute of Technology, 76131 Karlsruhe, Germany</b><br/><b>The University of Texas at Austin, 78701 Austin, USA</b></td><td>('2256981', 'Ziad Al-Halah', 'ziad al-halah')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>{ziad.al-halah, rainer.stiefelhagen}@kit.edu, grauman@cs.utexas.edu +</td></tr><tr><td>4d9c02567e7b9e065108eb83ea3f03fcff880462</td><td>Towards Facial Expression Recognition in the Wild: A New Database and Deep +<br/>Recognition System +<br/><b>School of Electronics and Information, Northwestern Polytechnical University, China</b></td><td>('3411701', 'Xianlin Peng', 'xianlin peng')<br/>('1917901', 'Zhaoqiang Xia', 'zhaoqiang xia')<br/>('2871379', 'Lei Li', 'lei li')<br/>('4729239', 'Xiaoyi Feng', 'xiaoyi feng')</td><td>pengxl515@163.com, zxia@nwpu.edu.cn, li lei 08@163.com, fengxiao@nwpu.edu.cn +</td></tr><tr><td>4d7e1eb5d1afecb4e238ba05d4f7f487dff96c11</td><td>978-1-5090-4117-6/17/$31.00 ©2017 IEEE +<br/>2352 +<br/>ICASSP 2017 +</td><td></td><td></td></tr><tr><td>4d90bab42806d082e3d8729067122a35bbc15e8d</td><td></td><td></td><td></td></tr><tr><td>4d3c4c3fe8742821242368e87cd72da0bd7d3783</td><td>Hybrid Deep Learning for Face Verification +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('1681656', 'Yi Sun', 'yi sun')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>sy011@ie.cuhk.edu.hk +<br/>xgwang@ee.cuhk.edu.hk +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>4d01d78544ae0de3075304ff0efa51a077c903b7</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 77– No.13, September 2013 +<br/>ART Network based Face Recognition with Gabor Filters +<br/>Dept. of Computer Science & Engineering +<br/>Dept. of Computer Science & Engineering +<br/><b>Jahangirnagar University</b><br/>Savar, Dhaka – 1342, Bangladesh. +</td><td>('5380965', 'Md. Mozammel Haque', 'md. mozammel haque')<br/>('39604645', 'Md. Al-amin Bhuiyan', 'md. al-amin bhuiyan')</td><td></td></tr><tr><td>4dd2be07b4f0393995b57196f8fc79d666b3aec5</td><td>3572 +<br/>2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +<br/>978-1-4799-2893-4/14/$31.00 ©2014 IEEE +<br/>EXPRESSION RECOGNITION +<br/>Dept. of Electronic Engineering +<br/><b>Yeungnam University</b><br/>Gyeongsan, Korea +<br/>1. INTRODUCTION +</td><td>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('1685841', 'Chan-Su Lee', 'chan-su lee')</td><td></td></tr><tr><td>4d8ce7669d0346f63b20393ffaa438493e7adfec</td><td>Similarity Features for Facial Event Analysis +<br/><b>Rutgers University, Piscataway NJ 08854, USA</b><br/>2 National Laboratory of Pattern Recognition, Chinese Academy of Sciences +<br/>Beijing, 100080, China +</td><td>('39606160', 'Peng Yang', 'peng yang')<br/>('1734954', 'Qingshan Liu', 'qingshan liu')</td><td>peyang@cs.rutgers.edu +</td></tr><tr><td>4d6ad0c7b3cf74adb0507dc886993e603c863e8c</td><td>Human Activity Recognition Based on Wearable +<br/>Sensor Data: A Standardization of the +<br/>State-of-the-Art +<br/>Smart Surveillance Interest Group, Computer Science Department +<br/>Universidade Federal de Minas Gerais, Brazil +</td><td>('2954974', 'Antonio C. Nazare', 'antonio c. nazare')</td><td>Email: {arturjordao, antonio.nazare, jessicasena, william}@dcc.ufmg.br +</td></tr><tr><td>4d16337cc0431cd43043dfef839ce5f0717c3483</td><td>A Scalable and Privacy-Aware IoT Service for Live Video Analytics +<br/><b>Carnegie Mellon University</b><br/><b>Carnegie Mellon University</b><br/>Intel Labs +<br/>Norman Sadeh +<br/><b>Carnegie Mellon University</b><br/><b>Carnegie Mellon University</b><br/><b>Carnegie Mellon University</b></td><td>('3196473', 'Junjue Wang', 'junjue wang')<br/>('1773498', 'Brandon Amos', 'brandon amos')<br/>('1802347', 'Padmanabhan Pillai', 'padmanabhan pillai')<br/>('1732721', 'Anupam Das', 'anupam das')<br/>('1747303', 'Mahadev Satyanarayanan', 'mahadev satyanarayanan')</td><td>junjuew@cs.cmu.edu +<br/>bamos@cs.cmu.edu +<br/>padmanabhan.s.pillai@intel.com +<br/>sadeh@cs.cmu.edu +<br/>anupamd@cs.cmu.edu +<br/>satya@cs.cmu.edu +</td></tr><tr><td>4d0b3921345ae373a4e04f068867181647d57d7d</td><td>Learning attributes from human gaze +<br/>Department of Computer Science +<br/><b>University of Pittsburgh</b><br/>IEEE 2017 Winter +<br/>Conference on +<br/>Applications of +<br/>Computer Vision +</td><td>('1916866', 'Nils Murrugarra-Llerena', 'nils murrugarra-llerena')<br/>('1770205', 'Adriana Kovashka', 'adriana kovashka')</td><td></td></tr><tr><td>4dca3d6341e1d991c902492952e726dc2a443d1c</td><td>Learning towards Minimum Hyperspherical Energy +<br/><b>Georgia Institute of Technology 2Emory University</b><br/><b>South China University of Technology 4NVIDIA 5Google Brain 6Ant Financial</b></td><td>('36326884', 'Weiyang Liu', 'weiyang liu')<br/>('10035476', 'Rongmei Lin', 'rongmei lin')<br/>('46270580', 'Zhen Liu', 'zhen liu')<br/>('47968201', 'Lixin Liu', 'lixin liu')<br/>('1751019', 'Zhiding Yu', 'zhiding yu')<br/>('47175326', 'Bo Dai', 'bo dai')<br/>('1779453', 'Le Song', 'le song')</td><td></td></tr><tr><td>4d0ef449de476631a8d107c8ec225628a67c87f9</td><td>© 2010 IEEE. Personal use of this material is permitted. Permission from IEEE +<br/>must be obtained for all other uses, in any current or future media, including +<br/>reprinting/republishing this material for advertising or promotional purposes, +<br/>creating new collective works, for resale or redistribution to servers or lists, or +<br/>reuse of any copyrighted component of this work in other works. +<br/>Pre-print of article that appeared at BTAS 2010. +<br/>The published article can be accessed from: +<br/>http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5634517 +</td><td></td><td></td></tr><tr><td>4d47261b2f52c361c09f7ab96fcb3f5c22cafb9f</td><td>Deep multi-frame face super-resolution +<br/>Evgeniya Ustinova, Victor Lempitsky +<br/>October 17, 2017 +</td><td></td><td></td></tr><tr><td>4df3143922bcdf7db78eb91e6b5359d6ada004d2</td><td>Behav Res (2015) 47:1122–1135 +<br/>DOI 10.3758/s13428-014-0532-5 +<br/>The Chicago face database: A free stimulus set of faces +<br/>and norming data +<br/>Published online: 13 January 2015 +<br/><b>Psychonomic Society, Inc</b></td><td>('2428798', 'Joshua Correll', 'joshua correll')</td><td></td></tr><tr><td>75879ab7a77318bbe506cb9df309d99205862f6c</td><td>Analysis Of Emotion Recognition From Facial +<br/>Expressions Using Spatial And Transform Domain +<br/>Methods +</td><td>('2855399', 'P. Suja', 'p. suja')<br/>('2510426', 'Shikha Tripathi', 'shikha tripathi')</td><td></td></tr><tr><td>7574f999d2325803f88c4915ba8f304cccc232d1</td><td>Transfer Learning For Cross-Dataset Recognition: A Survey +<br/>This paper summarises and analyses the cross-dataset recognition transfer learning techniques with the +<br/>emphasis on what kinds of methods can be used when the available source and target data are presented +<br/>in different forms for boosting the target task. This paper for the first time summarises several transferring +<br/>criteria in details from the concept level, which are the key bases to guide what kind of knowledge to transfer +<br/>between datasets. In addition, a taxonomy of cross-dataset scenarios and problems is proposed according the +<br/>properties of data that define how different datasets are diverged, thereby review the recent advances on +<br/>each specific problem under different scenarios. Moreover, some real world applications and corresponding +<br/>commonly used benchmarks of cross-dataset recognition are reviewed. Lastly, several future directions are +<br/>identified. +<br/>Additional Key Words and Phrases: Cross-dataset, transfer learning, domain adaptation +<br/>1. INTRODUCTION +<br/>It has been explored how human would transfer learning in one context to another +<br/>similar context [Woodworth and Thorndike 1901; Perkins et al. 1992] in the field of +<br/>Psychology and Education. For example, learning to drive a car helps a person later +<br/>to learn more quickly to drive a truck, and learning mathematics prepares students to +<br/>study physics. The machine learning algorithms are mostly inspired by human brains. +<br/>However, most of them require a huge amount of training examples to learn a new +<br/>model from scratch and fail to apply knowledge learned from previous domains or +<br/>tasks. This may be due to that a basic assumption of statistical learning theory is +<br/>that the training and test data are drawn from the same distribution and belong to +<br/>the same task. Intuitively, learning from scratch is not realistic and practical, because +<br/>it violates how human learn things. In addition, manually labelling a large amount +<br/>of data for new domain or task is labour extensive, especially for the modern “data- +<br/>hungry” and “data-driven” learning techniques (i.e. deep learning). However, the big +<br/>data era provides a huge amount available data collected for other domains and tasks. +<br/>Hence, how to use the previously available data smartly for the current task with +<br/>scarce data will be beneficial for real world applications. +<br/>To reuse the previous knowledge for current tasks, the differences between old data +<br/>and new data need to be taken into account. Take the object recognition as an ex- +<br/>ample. As claimed by Torralba and Efros [2011], despite the great efforts of object +<br/>datasets creators, the datasets appear to have strong build-in bias caused by various +<br/>factors, such as selection bias, capture bias, category or label bias, and negative set +<br/>bias. This suggests that no matter how big the dataset is, it is impossible to cover +<br/>the complexity of the real visual world. Hence, the dataset bias needs to be consid- +<br/>ered before reusing data from previous datasets. Pan and Yang [2010] summarise that +<br/>the differences between different datasets can be caused by domain divergence (i.e. +<br/>distribution shift or feature space difference) or task divergence (i.e. conditional dis- +<br/>tribution shift or label space difference), or both. For example, in visual recognition, +<br/>the distributions between the previous and current data can be discrepant due to the +<br/>different environments, lighting, background, sensor types, resolutions, view angles, +<br/>and post-processing. Those external factors may cause the distribution divergence or +<br/>even feature space divergence between different domains. On the other hand, the task +<br/>divergence between current and previous data is also ubiquitous. For example, it is +<br/>highly possible that an animal species that we want to recognize have not been seen +<br/>ACM Journal Name, Vol. V, No. N, Article A, Publication date: January YYYY. +</td><td>('47539715', 'Jing Zhang', 'jing zhang')<br/>('40508657', 'Wanqing Li', 'wanqing li')<br/>('1719314', 'Philip Ogunbona', 'philip ogunbona')</td><td></td></tr><tr><td>75fcbb01bc7e53e9de89cb1857a527f97ea532ce</td><td>Detection of Facial Landmarks from Neutral, Happy, +<br/>and Disgust Facial Images +<br/>Research Group for Emotions, Sociality, and Computing +<br/>Tampere Unit for Computer-Human Interaction +<br/>Department of Computer Sciences +<br/><b>University of Tampere</b><br/>FIN-33014 Tampere, Finland +</td><td>('2396729', 'Ioulia Guizatdinova', 'ioulia guizatdinova')<br/>('1718377', 'Veikko Surakka', 'veikko surakka')</td><td>ig74400@cs.uta.fi +<br/>Veikko.Surakka@uta.fi +</td></tr><tr><td>757e4cb981e807d83539d9982ad325331cb59b16</td><td>Demographics versus Biometric Automatic +<br/>Interoperability +<br/><b>Sapienza University of Rome, Italy</b><br/><b>Biometric and Image Processing Lab, University of Salerno, Italy</b><br/><b>George Mason University, Fairfax Virginia, USA</b></td><td>('1763890', 'Maria De Marsico', 'maria de marsico')<br/>('1795333', 'Michele Nappi', 'michele nappi')<br/>('1772512', 'Daniel Riccio', 'daniel riccio')<br/>('1781577', 'Harry Wechsler', 'harry wechsler')</td><td>demarsico@di.uniroma1.it +<br/>{mnappi,driccio}@unisa.it +<br/>wechsler@cs.gmu.edu +</td></tr><tr><td>75e9a141b85d902224f849ea61ab135ae98e7bfb</td><td></td><td></td><td></td></tr><tr><td>75503aff70a61ff4810e85838a214be484a674ba</td><td>Improved Facial Expression Recognition via Uni-Hyperplane Classification +<br/>S.W. Chew∗, S. Lucey†, P. Lucey‡, S. Sridharan∗, and J.F. Cohn‡ +</td><td></td><td></td></tr><tr><td>75fd9acf5e5b7ed17c658cc84090c4659e5de01d</td><td>Project-Out Cascaded Regression with an application to Face Alignment +<br/><b>School of Computer Science, University of Nottingham</b><br/>Contributions. Cascaded regression approaches [2] have been recently +<br/>shown to achieve state-of-the-art performance for many computer vision +<br/>tasks. Beyond its connection to boosting, cascaded regression has been in- +<br/>terpreted as a learning-based approach to iterative optimization methods like +<br/>the Newton’s method. However, in prior work [1],[4], the connection to op- +<br/>timization theory is limited only in learning a mapping from image features +<br/>to problem parameters. +<br/>In this paper, we consider the problem of facial deformable model fit- +<br/>ting using cascaded regression and make the following contributions: (a) We +<br/>propose regression to learn a sequence of averaged Jacobian and Hessian +<br/>matrices from data, and from them descent directions in a fashion inspired +<br/>by Gauss-Newton optimization. (b) We show that the optimization problem +<br/>in hand has structure and devise a learning strategy for a cascaded regres- +<br/>sion approach that takes the problem structure into account. By doing so, the +<br/>proposed method learns and employs a sequence of averaged Jacobians and +<br/>descent directions in a subspace orthogonal to the facial appearance varia- +<br/>tion; hence, we call it Project-Out Cascaded Regression (PO-CR). (c) Based +<br/>on the principles of PO-CR, we built a face alignment system that produces +<br/>remarkably accurate results on the challenging iBUG data set outperform- +<br/>ing previously proposed systems by a large margin. Code for our system is +<br/>available from http://www.cs.nott.ac.uk/~yzt/. +<br/>Shape and appearance models. We use parametric shape and appearance +<br/>models. An instance of the shape model is given by s(p) = s0 + Sp. An +<br/>instance of the appearance model is given by A(c) = A0 + Ac. +<br/>Face alignment via Gauss-Newton optimization. In this section, we for- +<br/>mulate and solve the non-linear least squares optimization problem for face +<br/>alignment using Gauss-Newton optimization. This will provide the basis for +<br/>learning and fitting in PO-CR in the next section. In particular, to localize +<br/>the landmarks in a new image, we would like to find p and c such that [3] +<br/>||I(s(p))− A(c)||2. +<br/>argmin +<br/>p,c +<br/>An update for p and c can be found by solving the following problem +<br/>arg min +<br/>∆p,∆c +<br/>||I(s(p)) + JI∆p− A0 − Ac− A∆c||2. +<br/>(1) +<br/>(2) +<br/>By exploiting the problem structure, the calculation for the optimal ∆c at +<br/>each iteration is not necessary. We end up with the following problem [3] +<br/>||I(s(p)) + JI∆p− A0||2 +<br/>P, +<br/>argmin +<br/>∆p +<br/>(3) +<br/>where P = E − AAT is a projection operator that projects out the facial +<br/>appearance variation from the image Jacobian JI. The solution to the above +<br/>problem is readily given by +<br/>∆p = −H−1 +<br/>P JT +<br/>P (I(s(p))− A0). +<br/>(4) +<br/>Face alignment via Project-Out Cascaded Regression. Based on Eqs. (3) +<br/>and (4), the key idea in PO-CR is to compute from a set of training examples +<br/>a sequence of averaged Jacobians(cid:98)J(k) from which the facial appearance +<br/>variation is projected-out and from them and descent directions: +<br/>Step I. Starting from the ground truth shape parameters p∗ +<br/>i for each +<br/>training image Ii, i = 1, . . . ,H, we generate a set of K perturbed shape pa- +<br/>rameters for iteration 1 pi, j(1), j = 1, . . . ,K that capture the statistics of the +<br/>PO-CR learns the averaged projected-out Jacobian(cid:98)JP(1) = P(cid:98)J(1) for itera- +<br/>face detection initialization process. Using the set ∆pi, j(1) = p∗ +<br/>i − pi, j(1), +<br/>tion 1 by solving the following weighted least squares problem +<br/>||I(s(pi, j(1))) + J(1)∆pi, j(1)− A0||2 +<br/>P, +<br/>arg min(cid:98)JP(1) +<br/>i=1 +<br/>j=1 +<br/>Step II. Having computed(cid:98)JP(1), we compute(cid:98)HP(1) =(cid:98)JP(1)T(cid:98)JP(1) . +<br/>Step III. The descent directions R(1) for iteration 1 are given by +<br/>R(1) =(cid:98)HP(1)−1(cid:98)JP(1)T . +<br/>(6) +<br/>Step IV. For each training sample, a new estimate for its shape parame- +<br/>ters (to be used at the next iteration) is obtained from +<br/>pi, j(2) = pi, j(1) + R(1)(I(s(pi, j(1)))− A0). +<br/>(7) +<br/>Finally, Steps I-IV are sequentially repeated until convergence and the whole +<br/>process produces a set of L regressor matrices R(l), l = 1, . . . ,L. +<br/>During testing,we extract image features I(s(p(k))) and then compute +<br/>an update for the shape parameters from +<br/>∆p(k) = R(k)(I(s(p(k)))− A0). +<br/>(8) +<br/>Results. We conducted a large number of experiments on LFPW, Helen, +<br/>AFW and iBUG data sets. In the following figure, we show fiiting results +<br/>from the challenging iBUG data set. +<br/>Figure 1: Application of PO-CR to the alignment of the iBUG data set. +<br/>[1] T.F. Cootes, G.J. Edwards, and C.J. Taylor. Active appearance models. +<br/>TPAMI, 23(6):681–685, 2001. +<br/>[2] Piotr Dollár, Peter Welinder, and Pietro Perona. Cascaded pose regres- +<br/>sion. In CVPR, 2010. +</td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td></td></tr><tr><td>75308067ddd3c53721430d7984295838c81d4106</td><td>Article +<br/>Rapid Facial Reactions +<br/>in Response to Facial +<br/>Expressions of Emotion +<br/>Displayed by Real Versus +<br/>Virtual Faces +<br/>i-Perception +<br/>2018 Vol. 9(4), 1–18 +<br/>! The Author(s) 2018 +<br/>DOI: 10.1177/2041669518786527 +<br/>journals.sagepub.com/home/ipe +<br/><b>LIMSI, CNRS, University of Paris-Sud, Orsay, France</b></td><td>('28174013', 'Jean-Claude Martin', 'jean-claude martin')</td><td></td></tr><tr><td>75cd81d2513b7e41ac971be08bbb25c63c37029a</td><td></td><td></td><td></td></tr><tr><td>75bf3b6109d7a685236c8589f8ead7d769ea863f</td><td>Model Selection with Nonlinear Embedding for Unsupervised Domain Adaptation +<br/><b>Center for Cognitive Ubiquitous Computing, Arizona State University, Tempe, AZ, USA</b></td><td>('3151995', 'Hemanth Venkateswara', 'hemanth venkateswara')<br/>('2471253', 'Shayok Chakraborty', 'shayok chakraborty')<br/>('1743991', 'Sethuraman Panchanathan', 'sethuraman panchanathan')</td><td>{hemanthv, shayok.chakraborty, troy.mcdaniel, panch}@asu.edu +</td></tr><tr><td>759cf57215fcfdd8f59c97d14e7f3f62fafa2b30</td><td>Real-time Distracted Driver Posture Classification +<br/>Department of Computer Science and Engineering, School of Sciences and Engineering +<br/><b>The American University in Cairo, New Cairo 11835, Egypt</b></td><td>('3434212', 'Yehya Abouelnaga', 'yehya abouelnaga')<br/>('2150605', 'Hesham M. Eraqi', 'hesham m. eraqi')<br/>('2233511', 'Mohamed N. Moustafa', 'mohamed n. moustafa')</td><td>{devyhia,heraqi,m.moustafa}@aucegypt.edu +</td></tr><tr><td>751970d4fb6f61d1b94ca82682984fd03c74f127</td><td>Array-based Electromyographic Silent Speech Interface +<br/><b>Cognitive Systems Lab, Karlsruhe Institute of Technology, Karlsruhe, Germany</b><br/>Keywords: +<br/>EMG, EMG-based Speech Recognition, Silent Speech Interface, Electrode Array +</td><td>('1723149', 'Michael Wand', 'michael wand')<br/>('2289793', 'Christopher Schulte', 'christopher schulte')<br/>('1684236', 'Matthias Janke', 'matthias janke')<br/>('1713194', 'Tanja Schultz', 'tanja schultz')</td><td>{michael.wand, matthias.janke, tanja.schultz}@kit.edu, christopher.schulte@student.kit.edu +</td></tr><tr><td>759a3b3821d9f0e08e0b0a62c8b693230afc3f8d</td><td>Attribute and Simile Classifiers for Face Verification +<br/><b>Columbia University</b></td><td>('40631426', 'Neeraj Kumar', 'neeraj kumar')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')<br/>('1750470', 'Shree K. Nayar', 'shree k. nayar')</td><td></td></tr><tr><td>75ebe1e0ae9d42732e31948e2e9c03d680235c39</td><td>“Hello! My name is... Buffy” – Automatic +<br/>Naming of Characters in TV Video +<br/><b>University of Oxford</b></td><td>('3056091', 'Mark Everingham', 'mark everingham')<br/>('1782755', 'Josef Sivic', 'josef sivic')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>{me,josef,az}@robots.ox.ac.uk +</td></tr><tr><td>75e5ba7621935b57b2be7bf4a10cad66a9c445b9</td><td></td><td></td><td></td></tr><tr><td>75859ac30f5444f0d9acfeff618444ae280d661d</td><td>Multibiometric Cryptosystems based on Feature +<br/>Level Fusion +</td><td>('2743820', 'Abhishek Nagar', 'abhishek nagar')<br/>('34633765', 'Karthik Nandakumar', 'karthik nandakumar')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>758d7e1be64cc668c59ef33ba8882c8597406e53</td><td>IEEE TRANSACTIONS ON AFFECTIVE COMPUTING +<br/>AffectNet: A Database for Facial Expression, +<br/>Valence, and Arousal Computing in the Wild +</td><td>('2314025', 'Ali Mollahosseini', 'ali mollahosseini')<br/>('3093835', 'Mohammad H. Mahoor', 'mohammad h. mahoor')</td><td></td></tr><tr><td>7553fba5c7f73098524fbb58ca534a65f08e91e7</td><td>Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/>A Monthly Journal of Computer Science and Information Technology +<br/>ISSN 2320–088X +<br/>IJCSMC, Vol. 3, Issue. 6, June 2014, pg.816 – 824 +<br/> RESEARCH ARTICLE +<br/>A Practical Approach for Determination +<br/>of Human Gender & Age +<br/> +<br/><b>India</b><br/><b>India</b></td><td>('1802780', 'Harpreet Kaur', 'harpreet kaur')<br/>('1802780', 'Harpreet Kaur', 'harpreet kaur')<br/>('38968310', 'Ahsan Hussain', 'ahsan hussain')</td><td>1 hkaur_bhatia23@yahoo.com, 2 ahsanhbaba@gmail.com +</td></tr><tr><td>751b26e7791b29e4e53ab915bfd263f96f531f56</td><td>Mood Meter: Counting Smiles in the Wild +<br/>Mohammed (Ehsan) Hoque * +<br/>Media Lab +<br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, MA, USA +</td><td>('2806721', 'Will Drevo', 'will drevo')<br/>('1719389', 'Rosalind W. Picard', 'rosalind w. picard')<br/>('15977480', 'Javier Hernandez', 'javier hernandez')</td><td>{javierhr, mehoque, drevo, picard}@mit.edu +</td></tr><tr><td>75da1df4ed319926c544eefe17ec8d720feef8c0</td><td>FDDB: A Benchmark for Face Detection in Unconstrained Settings +<br/><b>University of Massachusetts Amherst</b><br/><b>University of Massachusetts Amherst</b><br/>Amherst MA 01003 +<br/>Amherst MA 01003 +</td><td>('1714536', 'Erik Learned-Miller', 'erik learned-miller')<br/>('2246870', 'Vidit Jain', 'vidit jain')</td><td>elm@cs.umass.edu +<br/>vidit@cs.umass.edu +</td></tr><tr><td>75259a613285bdb339556ae30897cb7e628209fa</td><td>Unsupervised Domain Adaptation for Zero-Shot Learning +<br/><b>Queen Mary University of London, London E1 4NS, UK</b></td><td>('2999293', 'Elyor Kodirov', 'elyor kodirov')<br/>('1700927', 'Tao Xiang', 'tao xiang')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')</td><td>{e.kodirov, t.xiang, z.fu, s.gong}@qmul.ac.uk +</td></tr><tr><td>754f7f3e9a44506b814bf9dc06e44fecde599878</td><td>Quantized Densely Connected U-Nets for +<br/>Efficient Landmark Localization +</td><td>('2986505', 'Zhiqiang Tang', 'zhiqiang tang')<br/>('4340744', 'Xi Peng', 'xi peng')<br/>('1947101', 'Shijie Geng', 'shijie geng')<br/>('3008832', 'Lingfei Wu', 'lingfei wu')<br/>('1753384', 'Shaoting Zhang', 'shaoting zhang')</td><td>1Rutgers University, {zt53, sg1309, dnm}@rutgers.edu +<br/>2Binghamton University, xpeng@binghamton.edu +<br/>3IBM T. J. Watson, lwu@email.wm.edu +<br/>4SenseTime, zhangshaoting@sensetime.com +</td></tr><tr><td>75249ebb85b74e8932496272f38af274fbcfd696</td><td>Face Identification in Large Galleries +<br/>Smart Surveillance Interest Group, Department of Computer Science +<br/>Universidade Federal de Minas Gerais, Belo Horizonte, Brazil +</td><td>('1679142', 'William Robson Schwartz', 'william robson schwartz')</td><td>rafaelvareto@dcc.ufmg.br, filipe.oc87@gmail.com, william@dcc.ufmg.br +</td></tr><tr><td>75d2ecbbcc934563dff6b39821605dc6f2d5ffcc</td><td>Capturing Subtle Facial Motions in 3D Face Tracking +<br/><b>Beckman Institute</b><br/><b>University of Illinois at Urbana-Champaign</b><br/>Urbana, IL 61801 +</td><td>('1735018', 'Zhen Wen', 'zhen wen')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>{zhenwen, huang}@ifp.uiuc.edu +</td></tr><tr><td>81a142c751bf0b23315fb6717bc467aa4fdfbc92</td><td>978-1-5090-4117-6/17/$31.00 ©2017 IEEE +<br/>1767 +<br/>ICASSP 2017 +</td><td></td><td></td></tr><tr><td>81bfe562e42f2eab3ae117c46c2e07b3d142dade</td><td>A Hajj And Umrah Location Classification System For Video +<br/>Crowded Scenes +<br/>Adnan A. Gutub† +<br/><b>Center of Research Excellence in Hajj and Umrah, Umm Al-Qura University, Makkah, KSA</b><br/><b>College of Computers and Information Systems, Umm Al-Qura University, Makkah, KSA</b></td><td>('2872536', 'Hossam M. Zawbaa', 'hossam m. zawbaa')<br/>('1977955', 'Salah A. Aly', 'salah a. aly')</td><td></td></tr><tr><td>81695fbbbea2972d7ab1bfb1f3a6a0dbd3475c0f</td><td><b>UNIVERSITY OF TARTU</b><br/>FACULTY OF SCIENCE AND TECHNOLOGY +<br/><b>Institute of Computer Science</b><br/>Computer Science +<br/>Comparison of Face Recognition +<br/>Neural Networks +<br/>Bachelor's thesis (6 ECST) +<br/>Supervisor: Tambet Matiisen +<br/>Tartu 2016 +</td><td></td><td></td></tr><tr><td>8147ee02ec5ff3a585dddcd000974896cb2edc53</td><td>Angular Embedding: +<br/>A Robust Quadratic Criterion +<br/>Stella X. Yu, Member, +<br/>IEEE +</td><td></td><td></td></tr><tr><td>8199803f476c12c7f6c0124d55d156b5d91314b6</td><td>The iNaturalist Species Classification and Detection Dataset +<br/>1Caltech +<br/>2Google +<br/>3Cornell Tech +<br/>4iNaturalist +</td><td>('2996914', 'Grant Van Horn', 'grant van horn')<br/>('13412044', 'Alex Shepard', 'alex shepard')<br/>('1690922', 'Pietro Perona', 'pietro perona')<br/>('50172592', 'Serge Belongie', 'serge belongie')</td><td></td></tr><tr><td>816bd8a7f91824097f098e4f3e0f4b69f481689d</td><td>Latent Semantic Analysis of Facial Action Codes +<br/>for Automatic Facial Expression Recognition +<br/>D-ITET/BIWI +<br/>ETH Zurich +<br/>Zurich, Switzerland +<br/><b>IDIAP Research Institute</b><br/>Martigny, Switzerland +<br/><b>IDIAP Research Institute</b><br/>Martigny, Switzerland +</td><td>('8745904', 'Beat Fasel', 'beat fasel')<br/>('1824057', 'Florent Monay', 'florent monay')<br/>('1698682', 'Daniel Gatica-Perez', 'daniel gatica-perez')</td><td>bfasel@vision.ee.ethz.ch +<br/>monay@idiap.ch +<br/>gatica@idiap.ch +</td></tr><tr><td>81706277ed180a92d2eeb94ac0560f7dc591ee13</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 55– No.15, October 2012 +<br/>Emotion based Contextual Semantic Relevance +<br/>Feedback in Multimedia Information Retrieval +<br/>Department of Computer Engineering, Indian +<br/><b>Institute of Technology, Banaras Hindu</b><br/><b>University, Varanasi, 221005, India</b><br/>Anil K. Tripathi +<br/>Department of Computer Engineering, Indian +<br/><b>Institute of Technology, Banaras Hindu</b><br/><b>University, Varanasi, 221005, India</b><br/>to +<br/>find some +<br/>issued by a user +</td><td>('41132883', 'Karm Veer Singh', 'karm veer singh')</td><td></td></tr><tr><td>81831ed8e5b304e9d28d2d8524d952b12b4cbf55</td><td></td><td></td><td></td></tr><tr><td>81b2a541d6c42679e946a5281b4b9dc603bc171c</td><td>Universit¨at Ulm | 89069 Ulm | Deutschland +<br/>Fakult¨at f¨ur Ingenieurwissenschaften und Informatik +<br/>Institut f¨ur Neuroinformatik +<br/>Direktor: Prof. Dr. G¨unther Palm +<br/>Semi-Supervised Learning with Committees: +<br/>Exploiting Unlabeled Data Using Ensemble +<br/>Learning Algorithms +<br/>Dissertation zur Erlangung des Doktorgrades +<br/>Doktor der Naturwissenschaften (Dr. rer. nat.) +<br/>der Fakult¨at f¨ur Ingenieurwissenschaften und Informatik +<br/>der Universit¨at Ulm +<br/>vorgelegt von +<br/>aus Kairo, ¨Agypten +<br/>Ulm, Deutschland +<br/>2010 +</td><td>('1799097', 'Mohamed Farouk Abdel Hady', 'mohamed farouk abdel hady')</td><td></td></tr><tr><td>81e11e33fc5785090e2d459da3ac3d3db5e43f65</td><td>International Journal of Advances in Engineering & Technology, March 2012. +<br/>©IJAET ISSN: 2231-1963 +<br/>A NOVEL FACE RECOGNITION APPROACH USING A +<br/>MULTIMODAL FEATURE VECTOR +<br/><b>Central Mechanical Engineering Research Institute, Durgapur, West Bengal, India</b><br/><b>National Institute of Technology, Durgapur, West Bengal, India</b></td><td>('9155672', 'Jhilik Bhattacharya', 'jhilik bhattacharya')<br/>('40301536', 'Nattami Sekhar', 'nattami sekhar')<br/>('1872045', 'Somajyoti Majumder', 'somajyoti majumder')<br/>('33606010', 'Gautam Sanyal', 'gautam sanyal')</td><td></td></tr><tr><td>81e366ed1834a8d01c4457eccae4d57d169cb932</td><td>Pose-Configurable Generic Tracking of Elongated Objects +<br/>Multimedia Systems Department +<br/><b>Gdansk University of Technology</b><br/>Departement Electronique et Physique +<br/>Institut Mines-Telecom / Telecom SudParis +</td><td>('2120042', 'Daniel Wesierski', 'daniel wesierski')<br/>('2603633', 'Patrick Horain', 'patrick horain')</td><td>daniel.wesierski@pg.gda.pl +<br/>patrick.horain@telecom-sudaris.eu +</td></tr><tr><td>8164ebc07f51c9e0db4902980b5ac3f5a8d8d48c</td><td>Shuffle-Then-Assemble: Learning +<br/>Object-Agnostic Visual Relationship Features +<br/>School of Computer Science and Engineering, +<br/><b>Nanyang Technological University</b></td><td>('47008946', 'Xu Yang', 'xu yang')<br/>('5462268', 'Hanwang Zhang', 'hanwang zhang')<br/>('1688642', 'Jianfei Cai', 'jianfei cai')</td><td>s170018@e.ntu.edu.sg,{hanwangzhang,asjfcai}@ntu.edu.sg +</td></tr><tr><td>81fc86e86980a32c47410f0ba7b17665048141ec</td><td>Segment-based Methods for Facial Attribute +<br/>Detection from Partial Faces +<br/>Department of Electrical and Computer Engineering and the Center for Automation Research, +<br/><b>UMIACS, University of Maryland, College Park, MD</b></td><td>('3152615', 'Upal Mahbub', 'upal mahbub')</td><td>{umahbub, ssarkar2, rama}@umiacs.umd.edu +</td></tr><tr><td>8160b3b5f07deaa104769a2abb7017e9c031f1c1</td><td>683 +<br/>Exploiting Discriminant Information in Nonnegative +<br/>Matrix Factorization With Application +<br/>to Frontal Face Verification +</td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')<br/>('2336758', 'Ioan Buciu', 'ioan buciu')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td></td></tr><tr><td>814d091c973ff6033a83d4e44ab3b6a88cc1cb66</td><td>Behav Res (2016) 48:567–576 +<br/>DOI 10.3758/s13428-015-0601-4 +<br/>The EU-Emotion Stimulus Set: A validation study +<br/>Published online: 30 September 2015 +<br/><b>Psychonomic Society, Inc</b></td><td>('2625704', 'Delia Pigat', 'delia pigat')<br/>('2391819', 'Shahar Tal', 'shahar tal')<br/>('2100443', 'Ofer Golan', 'ofer golan')<br/>('1884685', 'Simon Baron-Cohen', 'simon baron-cohen')<br/>('3343472', 'Daniel Lundqvist', 'daniel lundqvist')</td><td></td></tr><tr><td>816eff5e92a6326a8ab50c4c50450a6d02047b5e</td><td>fLRR: Fast Low-Rank Representation Using +<br/>Frobenius Norm +<br/>Low Rank Representation (LRR) intends to find the representation +<br/>with lowest-rank of a given data set, which can be formulated as a +<br/>rank minimization problem. Since the rank operator is non-convex and +<br/>discontinuous, most of the recent works use the nuclear norm as a convex +<br/>relaxation. This letter theoretically shows that under some conditions, +<br/>Frobenius-norm-based optimization problem has an unique solution that +<br/>is also a solution of the original LRR optimization problem. In other +<br/>words, it is feasible to apply Frobenius-norm as a surrogate of the +<br/>nonconvex matrix rank function. This replacement will largely reduce the +<br/>time-costs for obtaining the lowest-rank solution. Experimental results +<br/>show that our method (i.e., fast Low Rank Representation, fLRR), +<br/>performs well in terms of accuracy and computation speed in image +<br/>clustering and motion segmentation compared with nuclear-norm-based +<br/>LRR algorithm. +<br/>Introduction: Given a data set X ∈ Rm×n(m < n) composed of column +<br/>vectors, let A be a data set composed of vectors with the same dimension +<br/>as those in X. Both X and A can be considered as matrices. A linear +<br/>representation of X with respect to A is a matrix Z that satisfies the +<br/>equation X = AZ. The data set A is called a dictionary. In general, this +<br/>linear matrix equation will have infinite solutions, and any solution can be +<br/>considered to be a representation of X associated with the dictionary A. To +<br/>obtain an unique Z and explore the latent structure of the given data set, +<br/>various assumptions could be enforced over Z. +<br/>Liu et al. recently proposed Low Rank Representation (LRR) [1] by +<br/>assuming that data are approximately sampled from an union of low-rank +<br/>subspaces. Mathematically, LRR aims at solving +<br/>min rank(Z) +<br/>s.t. X = AZ, +<br/>(1) +<br/>where rank(Z) could be defined as the number of nonzero eigenvalues of +<br/>the matrix Z. Clearly, (1) is non-convex and discontinuous, whose convex +<br/>relaxation is as follows, +<br/>min kZk∗ +<br/>s.t. X = AZ, +<br/>(2) +<br/>where kZk∗ is the nuclear norm, which is a convex and continuous +<br/>optimization problem. +<br/>Considering the possible corruptions, the objective function of LRR is +<br/>min kZk∗ + λkEkp +<br/>s.t. X = AZ + E, +<br/>(3) +<br/>where k · kp could be ℓ1-norm for describing sparse corruption or ℓ2,1- +<br/>norm for characterizing sample-specified corruption. +<br/>The above nuclear-norm-based optimization problems are generally +<br/>solved using Augmented Lagrange Multiplier algorithm (ALM) [2] which +<br/>requires repeatedly performing Single Value Decomposition (SVD) over +<br/>Z. Hence, this optimization program is inefficient. +<br/>Beyond the nuclear-norm, do other norms exist that can be used as +<br/>a surrogates for rank-minimization problem in LRR? Can we develop +<br/>a fast algorithm to calculate LRR? This letter addresses these problems +<br/>by theoretically showing the equivalence between the solutions of a +<br/>Frobenius-norm-based problem and the original LRR problem. And we +<br/>further develop fast Low Rank Representation (fLRR) based on the +<br/>theoretical results. +<br/>Theoretical Analysis: In the following analyses, Theorem 1 and +<br/>Theorem 3 prove that Frobenius-norm-based problem is a surrogate of +<br/>the rank-minimization problem of LRR in the case of clean data and +<br/>corrupted ones, respectively. Theorem 2 shows that our Frobenius-norm- +<br/>based method could produce a block-diagonal Z under some conditions. +<br/>This property is helpful to subspace clustering. +<br/>Let A ∈ Rm×n be a matrix with rank r. The full SVD and skinny +<br/>SVD of A are A = U ΣV T and A = UrΣrV T +<br/>r , where U and V are two +<br/>orthogonal matrices with the size of m × m and n × n, respectively. In +<br/>addition, Σ is an m × n rectangular diagonal matrix, its diagonal elements +<br/>are nonnegative real numbers. Σr is a r × r diagonal matrix with singular +<br/>values located on the diagonal in decreasing order, Ur and Vr consist of the +<br/>first r columns of U and V , respectively. Clearly, Ur and Vr are column +<br/>orthogonal matrices, i.e., U T +<br/>r Vr = Ir, where Ir denotes the +<br/>r Ur = Ir, V T +<br/>identity matrix with the size of r × r. The pseudoinverse of A is defined +<br/>by A† = VrΣ−1 +<br/>r U T +<br/>r . +<br/>Given a matrix M ∈ Rm×n, the Frobenius norm of M is defined by +<br/>kM kF =ptrace (M T M ) =qPmin{m,n} +<br/>value of M . Clearly, kM kF = 0 if and only if M = 0. +<br/>i=1 +<br/>σ2 +<br/>i , where σi is a singular +<br/>Lemma 1: Suppose P is a column orthogonal matrix, i.e., P T P = I. Then, +<br/>kP M kF = kM kF . +<br/>Lemma 2: For the matrices M and N with same number of columns, it +<br/>holds that +<br/>= kM k2 +<br/>F + kN k2 +<br/>F . +<br/>(4) +<br/>N (cid:21)(cid:13)(cid:13)(cid:13)(cid:13) +<br/>(cid:13)(cid:13)(cid:13)(cid:13) +<br/>(cid:20) M +<br/>The proofs of the above two lemmas are trivial. +<br/>Theorem 1: +<br/>minimization problem +<br/>Suppose +<br/>that X ∈ span{A}, +<br/>the Frobenius norm +<br/>min kZkF +<br/>s.t. X = AZ, +<br/>(5) +<br/>has an unique solution Z ∗ = A†X which is also the lowest-rank solution +<br/>of LRR in terms of (1). +<br/>Proof: Let the full and skinny SVDs of A be A = U ΣV T and A = +<br/>r U T +<br/>UrΣrV T +<br/>r . +<br/>r , respectively. Then, the pseudoinverse of A is A† = VrΣ−1 +<br/>Defining Vc by V T =(cid:20) V T +<br/>V T +<br/>(cid:21) and V T +<br/>c Vr = 0. Moreover, it can be easily +<br/>checked that Z ∗ satisfies X = AZ ∗ owing to X ∈ span{A}. +<br/>To prove that Z ∗ is the unique solution of the optimization problem +<br/>(5), two steps are required. First, we will prove that, for any solution Z of +<br/>X = AZ, it must hold that kZkF ≥ kZ ∗kF . Using Lemma 1, we have +<br/>kZkF = (cid:13)(cid:13)(cid:13)(cid:13) +<br/>= (cid:13)(cid:13)(cid:13)(cid:13) +<br/>V T +<br/>(cid:20) V T +<br/>(cid:20) V T +<br/>(cid:21) [Z ∗ + (Z − Z ∗)](cid:13)(cid:13)(cid:13)(cid:13)F +<br/>c (Z − Z ∗) (cid:21)(cid:13)(cid:13)(cid:13)(cid:13)F +<br/>r (Z − Z ∗) +<br/>r Z ∗ + V T +<br/>c Z ∗ + V T +<br/>V T +<br/>As A (Z − Z ∗) = 0, +<br/>r (Z − Z ∗) = 0. Denote B = Σ−1 +<br/>V T +<br/>V T +<br/>c Vr = 0, we have V T +<br/>i.e., UrΣrV T +<br/>r U T +<br/>c VrB = 0. Then, +<br/>r (Z − Z ∗) = 0, +<br/>r X, +<br/>follows that +<br/>then Z ∗ = VrB. Because +<br/>it +<br/>c Z ∗ = V T +<br/>(cid:20) +<br/>kZkF =(cid:13)(cid:13)(cid:13)(cid:13) +<br/>V T +<br/>c (Z − Z ∗) (cid:21)(cid:13)(cid:13)(cid:13)(cid:13)F +<br/>By Lemma 2, +<br/>kZk2 +<br/>F = kBk2 +<br/>F + kV T +<br/>c (Z − Z ∗)k2 +<br/>F , +<br/>then, kZkF ≥ kBkF . +<br/>By Lemma 1, +<br/>kBkF = kVrBkF = kZ ∗kF , +<br/>(6) +<br/>(7) +<br/>(8) +<br/>thus, kZkF ≥ kZ ∗kF for any solution Z of X = AZ. +<br/>In the second step, we will prove that if there exists another solution Z +<br/>of (5), Z = Z ∗ must hold. Clearly, Z is a solution of (5) which implies that +<br/>X = AZ and kZkF = kZ ∗kF . From (7) and (8), +<br/>kZk2 +<br/>F + kV T +<br/>F = kZ ∗k2 +<br/>Since kZkF = kZ ∗kF , +<br/>c (Z − Z ∗) k2 +<br/>F . +<br/>c (Z − Z ∗) kF = 0, +<br/>r (Z − Z ∗) = 0, this gives +<br/>and so V T +<br/>V T (Z − Z ∗) = 0. Because V is an orthogonal matrix, it must hold +<br/>that Z = Z ∗. The above proves that Z ∗ is the unique solution of the +<br/>optimization problem (5). +<br/>c (Z − Z ∗) = 0. Together with V T +<br/>it must hold that kV T +<br/>(9) +<br/>Next, we prove that Z ∗ is also a solution of the LRR optimization +<br/>problem (1). Clearly, for any solution Z of X = AZ, +<br/>it holds that +<br/>rank(Z) ≥ rank(AZ) = rank(X). On the other hand, rank(Z ∗) = +<br/>rank(A†X) ≤ rank(X). Thus, rank(Z ∗) = rank(X). This shows that +<br/>Z ∗ is the lowest-rank solution of the LRR optimization problem (1). The +<br/>proof is complete. +<br/>(cid:4) +<br/>In the following, Theorem 2 will show that the optimal Z of (5) will +<br/>be block-diagonal if the data are sampled from a set of independent +<br/>subspaces {S1, S2, · · · , Sk}, where the dimensionality of Si is ri and +<br/>i = {1, 2, · · · , k}. Note that, {S1, S2, · · · , Sk} are independent if and +<br/>only if SiTPj6=i Sj = {0}. Suppose that X = [X1, X2, · · · , Xk] and +<br/>A = [A1, A2, · · · , Ak], where Ai and Xi contain mi and ni data points +<br/>ELECTRONICS LETTERS 12th December 2011 Vol. 00 No. 00 +</td><td>('2235162', 'Haixian Zhang', 'haixian zhang')<br/>('4340744', 'Xi Peng', 'xi peng')</td><td></td></tr><tr><td>8149c30a86e1a7db4b11965fe209fe0b75446a8c</td><td>Semi-Supervised Multiple Instance Learning based +<br/>Domain Adaptation for Object Detection +<br/>Siemens Corporate Research +<br/>Siemens Corporate Research +<br/>Siemens Corporate Research +<br/>Amit Kale +<br/>Bangalore +<br/>Bangalore +<br/>{chhaya.methani, +<br/>Bangalore +<br/>rahul.thota, +</td><td>('2970569', 'Chhaya Methani', 'chhaya methani')<br/>('31516659', 'Rahul Thota', 'rahul thota')</td><td>kale.amit}@siemens.com +</td></tr><tr><td>81da427270c100241c07143885ba3051ec4a2ecb</td><td>Learning the Synthesizability of Dynamic Texture Samples∗ +<br/><b>State Key Lab. LIESMARS, Wuhan University, China</b><br/>2Computer Vision Lab., ETH Zurich, Switzerland +<br/>February 6, 2018 +</td><td>('1706687', 'Feng Yang', 'feng yang')<br/>('39943835', 'Gui-Song Xia', 'gui-song xia')<br/>('1778526', 'Dengxin Dai', 'dengxin dai')<br/>('1733213', 'Liangpei Zhang', 'liangpei zhang')</td><td>{guisong.xia, fengyang, zlp62}@whu.edu.cn +<br/>dai@vision.ee.ethz.ch +</td></tr><tr><td>861c650f403834163a2c27467a50713ceca37a3e</td><td>Probabilistic Elastic Part Model for Unsupervised Face Detector Adaptation +<br/><b>Stevens Institute of Technology</b><br/>Hoboken, NJ 07030 +<br/>Adobe Systems Inc. +<br/>San Jose, CA 95110 +</td><td>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('1745420', 'Gang Hua', 'gang hua')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')<br/>('1706007', 'Jianchao Yang', 'jianchao yang')</td><td>{hli18, ghua}@stevens.edu +<br/>{zlin, jbrandt, jiayang}@adobe.com +</td></tr><tr><td>86614c2d2f6ebcb9c600d4aef85fd6bf6eab6663</td><td>Benchmarks for Cloud Robotics +<br/>Arjun Singh +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2016-142 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-142.html +<br/>August 12, 2016 +</td><td></td><td></td></tr><tr><td>86b69b3718b9350c9d2008880ce88cd035828432</td><td>Improving Face Image Extraction by Using Deep Learning Technique +<br/>National Library of Medicine, NIH, Bethesda, MD +</td><td>('1726787', 'Zhiyun Xue', 'zhiyun xue')<br/>('1721328', 'Sameer Antani', 'sameer antani')<br/>('1691151', 'L. Rodney Long', 'l. rodney long')<br/>('1705831', 'Dina Demner-Fushman', 'dina demner-fushman')<br/>('1692057', 'George R. Thoma', 'george r. thoma')</td><td></td></tr><tr><td>86904aee566716d9bef508aa9f0255dc18be3960</td><td>Learning Anonymized Representations with +<br/>Adversarial Neural Networks +</td><td>('1743922', 'Pablo Piantanida', 'pablo piantanida')<br/>('1751762', 'Yoshua Bengio', 'yoshua bengio')<br/>('1694313', 'Pierre Duhamel', 'pierre duhamel')</td><td></td></tr><tr><td>86f191616423efab8c0d352d986126a964983219</td><td>Visual to Sound: Generating Natural Sound for Videos in the Wild +<br/><b>University of North Carolina at Chapel Hill, 2Adobe Research</b></td><td>('49455017', 'Yipin Zhou', 'yipin zhou')<br/>('8056043', 'Zhaowen Wang', 'zhaowen wang')<br/>('2442612', 'Chen Fang', 'chen fang')<br/>('30190128', 'Trung Bui', 'trung bui')<br/>('1685538', 'Tamara L. Berg', 'tamara l. berg')</td><td></td></tr><tr><td>867e709a298024a3c9777145e037e239385c0129</td><td> INTERNATIONAL JOURNAL +<br/> OF PROFESSIONAL ENGINEERING STUDIES Volume VIII /Issue 2 / FEB 2017 +<br/>ANALYTICAL REPRESENTATION OF UNDERSAMPLED FACE +<br/>RECOGNITION APPROACH BASED ON DICTIONARY LEARNING +<br/>AND SPARSE REPRESENTATION +<br/>(M.Tech)1, Assistant Professor2, Assistant Professor3, HOD of CSE Department4 +</td><td>('32628937', 'Murala Sandeep', 'murala sandeep')<br/>('1702980', 'Ranga Reddy', 'ranga reddy')</td><td></td></tr><tr><td>869a2fbe42d3fdf40ed8b768edbf54137be7ac71</td><td>Relative Attributes for Enhanced Human-Machine Communication +<br/><b>Toyota Technological Institute, Chicago</b><br/><b>Indraprastha Institute of Information Technology, Delhi</b><br/><b>University of Texas, Austin</b></td><td>('1713589', 'Devi Parikh', 'devi parikh')<br/>('1770205', 'Adriana Kovashka', 'adriana kovashka')<br/>('2076800', 'Amar Parkash', 'amar parkash')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>86c5478f21c4a9f9de71b5ffa90f2a483ba5c497</td><td>Kernel Selection using Multiple Kernel Learning and Domain +<br/>Adaptation in Reproducing Kernel Hilbert Space, for Face +<br/>Recognition under Surveillance Scenario +<br/><b>Indian Institute of Technology, Madras, Chennai 600036, INDIA</b><br/>Face Recognition (FR) has been the interest to several researchers over the past few decades due to its passive nature of biometric +<br/>authentication. Despite high accuracy achieved by face recognition algorithms under controlled conditions, achieving the same +<br/>performance for face images obtained in surveillance scenarios, is a major hurdle. Some attempts have been made to super-resolve +<br/>the low-resolution face images and improve the contrast, without considerable degree of success. The proposed technique in this +<br/>paper tries to cope with the very low resolution and low contrast face images obtained from surveillance cameras, for FR under +<br/>surveillance conditions. For Support Vector Machine classification, the selection of appropriate kernel has been a widely discussed +<br/>issue in the research community. In this paper, we propose a novel kernel selection technique termed as MFKL (Multi-Feature +<br/>Kernel Learning) to obtain the best feature-kernel pairing. Our proposed technique employs a effective kernel selection by Multiple +<br/>Kernel Learning (MKL) method, to choose the optimal kernel to be used along with unsupervised domain adaptation method in the +<br/>Reproducing Kernel Hilbert Space (RKHS), for a solution to the problem. Rigorous experimentation has been performed on three +<br/>real-world surveillance face datasets : FR SURV [33], SCface [20] and ChokePoint [44]. Results have been shown using Rank-1 +<br/>Recognition Accuracy, ROC and CMC measures. Our proposed method outperforms all other recent state-of-the-art techniques by +<br/>a considerable margin. +<br/>Index Terms—Kernel Selection, Surveillance, Multiple Kernel Learning, Domain Adaptation, RKHS, Hallucination +<br/>I. INTRODUCTION +<br/>Face Recognition (FR) is a classical problem which is far +<br/>from being solved. Face Recognition has a clear advantage +<br/>of being natural and passive over other biometric techniques +<br/>requiring co-operative subjects. Most face recognition algo- +<br/>rithms perform well under a controlled environment. A face +<br/>recognition system trained at a certain resolution, illumination +<br/>and pose, recognizes faces under similar conditions with very +<br/>high accuracy. In contrary, if the face of the same subject is +<br/>presented with considerable change in environmental condi- +<br/>tions, then such a face recognition system fails to achieve a +<br/>desired level of accuracy. So, we aim to find a solution to the +<br/>face recognition under unconstrained environment. +<br/>Face images obtained by an outdoor panoramic surveillance +<br/>camera, are often confronted with severe degradations (e.g., +<br/>low-resolution, blur, low-contrast, interlacing and noise). This +<br/>significantly limits the performance of face recognition sys- +<br/>tems used for binding “security with surveillance” applica- +<br/>tions. Here, images used for training are usually available be- +<br/>forehand which are taken under a well controlled environment +<br/>in an indoor setup (laboratory, control room), whereas the +<br/>images used for testing are captured when a subject comes +<br/>under a surveillance scene. With ever increasing demands +<br/>to combine “security with surveillance” in an integrated and +<br/>automated framework, it is necessary to analyze samples of +<br/>face images of subjects acquired by a surveillance camera +<br/>from a long distance. Hence the subject must be accurately +<br/>recognized from a low resolution, blurred and degraded (low +<br/>contrast, aliasing, noise) face image, as obtained from the +<br/>surveillance camera. These face images are difficult to match +<br/>because they are often captured under non-ideal conditions. +<br/>Thus, face recognition in surveillance scenario is an impor- +<br/>tant and emerging research area which motivates the work +<br/>presented in this paper. +<br/>Performance of most classifiers degrade when both the +<br/>resolution and contrast of face templates used for recognition +<br/>are low. There have been many advancement in this area +<br/>during the past decade, where attempts have been made to +<br/>deal with this problem under an unconstrained environment. +<br/>For surveillance applications, a face recognition system must +<br/>recognize a face in an unconstrained environment without the +<br/>notice of the subject. Degradation of faces is quite evident in +<br/>the surveillance scenario due to low-resolution and camera- +<br/>blur. Variations in the illuminating conditions of the faces +<br/>not only reduces the recognition accuracy but occasionally +<br/>degrades the performance of face detection which is the first +<br/>step of face recognition. The work presented in this paper deals +<br/>with such issues involved in FR under surveillance conditions. +<br/>In the work presented in this paper, the face samples from +<br/>both gallery and probe are initially passed through a robust +<br/>face detector, the Chehra face tracker, to find a tightly cropped +<br/>face image. A domain adaptation (DA) based algorithm, +<br/>formulated using eigen-domain transformation is designed to +<br/>bridge the gap between the features obtained from the gallery +<br/>and the probe samples. A novel Multiple kernel Learning +<br/>(MKL) based learning method, termed MFKL (Multi-Feature +<br/>Kernel Learning), is then used to obtain an optimal combi- +<br/>nation (pairing) of the feature and the kernel for FR. The +<br/>novelty of the work presented in this paper is the optimal +<br/>pairing of feature and kernel to provide best performance with +<br/>DA based learning for FR. Results of performance analysis on +</td><td>('2643208', 'Samik Banerjee', 'samik banerjee')<br/>('1680398', 'Sukhendu Das', 'sukhendu das')</td><td></td></tr><tr><td>86c053c162c08bc3fe093cc10398b9e64367a100</td><td>Cascade of Forests for Face Alignment +</td><td>('2966679', 'Heng Yang', 'heng yang')<br/>('2876552', 'Changqing Zou', 'changqing zou')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')</td><td></td></tr><tr><td>86b985b285c0982046650e8d9cf09565a939e4f9</td><td></td><td></td><td></td></tr><tr><td>861802ac19653a7831b314cd751fd8e89494ab12</td><td>Time-of-Flight and Depth Imaging. Sensors, Algorithms +<br/>and Applications: Dagstuhl Seminar 2012 and GCPR +<br/>Workshop on Imaging New Modalities (Lecture ... Vision, +<br/>Pattern Recognition, and Graphics) +<br/>Publisher: Springer; 2013 edition +<br/>(November 8, 2013) +<br/>Language: English +<br/>Pages: 320 +<br/>ISBN: 978-3642449635 +<br/>Size: 20.46 MB +<br/>Format: PDF / ePub / Kindle +<br/>Cameras for 3D depth imaging, using +<br/>either time-of-flight (ToF) or +<br/>structured light sensors, have received +<br/>a lot of attention recently and have +<br/>been improved considerably over the +<br/>last few years. The present +<br/>techniques... +</td><td>('1727057', 'Marcin Grzegorzek', 'marcin grzegorzek')<br/>('1680185', 'Christian Theobalt', 'christian theobalt')<br/>('39897382', 'Reinhard Koch', 'reinhard koch')<br/>('1758212', 'Andreas Kolb', 'andreas kolb')</td><td></td></tr><tr><td>86ed5b9121c02bcf26900913f2b5ea58ba23508f</td><td>Actions ⇠ Transformations +<br/><b>Carnegie Mellon University</b><br/><b>University of Washington</b><br/><b>The Allen Institute for AI</b></td><td>('39849136', 'Xiaolong Wang', 'xiaolong wang')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')</td><td></td></tr><tr><td>861b12f405c464b3ffa2af7408bff0698c6c9bf0</td><td>International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 +<br/>Volume: 3 Issue: 5 +<br/> 3337 - 3342 +<br/>_______________________________________________________________________________________________ +<br/>An Effective Technique for Removal of Facial Dupilcation by SBFA +<br/>Computer Department, +<br/>GHRCEM, +<br/>Pune, India +<br/>Computer Department, +<br/>GHRCEM, +<br/> Pune, India +</td><td>('2947776', 'Ayesha Butalia', 'ayesha butalia')</td><td>deepikapatil941@gmail.com +<br/>ayeshabutalia@gmail.com +</td></tr><tr><td>86b6afc667bb14ff4d69e7a5e8bb2454a6bbd2cd</td><td>YUE et al.: ATTENTIONAL ALIGNMENT NETWORK +<br/>Attentional Alignment Network +<br/><b>Beihang University, Beijing, China</b><br/>2 The Key Laboratory of Advanced +<br/>Technologies for Near Space +<br/>Information Systems +<br/>Ministry of +<br/>Technology of China +<br/><b>University of Texas at Arlington</b><br/>TX, USA +<br/><b>Shanghai Jiao Tong University</b><br/>Shanghai, China +<br/>Industry and Information +</td><td>('35310815', 'Lei Yue', 'lei yue')<br/>('6050999', 'Xin Miao', 'xin miao')<br/>('3127895', 'Pengbo Wang', 'pengbo wang')<br/>('1740430', 'Baochang Zhang', 'baochang zhang')<br/>('34798935', 'Xiantong Zhen', 'xiantong zhen')<br/>('40916581', 'Xianbin Cao', 'xianbin cao')</td><td>yuelei@buaa.edu.cn +<br/>xin.miao@mavs.uta.edu +<br/>wangpengbo_vincent@sjtu.edu.cn +<br/>bczhang@buaa.edu.cn +<br/>zhenxt@buaa.edu.cn +<br/>xbcao@buaa.edu.cn +</td></tr><tr><td>862d17895fe822f7111e737cbcdd042ba04377e8</td><td>Semi-Latent GAN: Learning to generate and modify facial images from +<br/>attributes +<br/><b>The school of Data Science, Fudan University</b><br/>† Disney Research, +</td><td>('11740128', 'Weidong Yin', 'weidong yin')<br/>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('14517812', 'Leonid Sigal', 'leonid sigal')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')</td><td>yanweifu@fudan.edu.cn +</td></tr><tr><td>86d0127e1fd04c3d8ea78401c838af621647dc95</td><td>Facial Attribute Prediction +<br/><b>College of Information and Engineering, Hunan University, Changsha, China</b><br/><b>School of Computer Science, National University of Defense Technology, Changsha, China</b><br/><b>University of Texas at San Antonio, USA</b></td><td>('48664471', 'Mingxing Duan', 'mingxing duan')<br/>('50842217', 'Qi Tian', 'qi tian')</td><td>duanmingxing16@nudt.edu.cn, lkl@hnu.edu.cn, qi.tian@utsa.edu +</td></tr><tr><td>86e1bdbfd13b9ed137e4c4b8b459a3980eb257f6</td><td>The Kinetics Human Action Video Dataset +<br/>Jo˜ao Carreira +<br/>Paul Natsev +</td><td>('21028601', 'Will Kay', 'will kay')<br/>('34838386', 'Karen Simonyan', 'karen simonyan')<br/>('11809518', 'Brian Zhang', 'brian zhang')<br/>('38961760', 'Chloe Hillier', 'chloe hillier')<br/>('2259154', 'Sudheendra Vijayanarasimhan', 'sudheendra vijayanarasimhan')<br/>('39045746', 'Fabio Viola', 'fabio viola')<br/>('1691808', 'Tim Green', 'tim green')<br/>('2830305', 'Trevor Back', 'trevor back')<br/>('2573615', 'Mustafa Suleyman', 'mustafa suleyman')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>wkay@google.com +<br/>joaoluis@google.com +<br/>simonyan@google.com +<br/>brianzhang@google.com +<br/>chillier@google.com +<br/>svnaras@google.com +<br/>fviola@google.com +<br/>tfgg@google.com +<br/>back@google.com +<br/>natsev@google.com +<br/>mustafasul@google.com +<br/>zisserman@google.com +</td></tr><tr><td>86b6de59f17187f6c238853810e01596d37f63cd</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 7, No. 3, 2016 +<br/>Competitive Representation Based Classification +<br/>Using Facial Noise Detection +<br/>Chongqing Key Laboratory of Computational Intelligence +<br/><b>College of Computer Science and Technology, Chongqing</b><br/>Chongqing Key Laboratory of Computational Intelligence +<br/><b>College of Computer Science and Technology, Chongqing</b><br/><b>University of Posts and Telecommunications</b><br/><b>University of Posts and Telecommunications</b><br/>Chongqing, China +<br/>Chongqing, China +<br/>Chongqing Key Laboratory of Computational Intelligence +<br/><b>College of Computer Science and Technology, Chongqing</b><br/>Chongqing Key Laboratory of Computational Intelligence +<br/><b>College of Computer Science and Technology, Chongqing</b><br/><b>University of Posts and Telecommunications</b><br/><b>University of Posts and Telecommunications</b><br/>Chongqing, China +<br/>Chongqing, China +</td><td>('1779859', 'Tao Liu', 'tao liu')<br/>('32611393', 'Ying Liu', 'ying liu')<br/>('38837555', 'Cong Li', 'cong li')<br/>('40032263', 'Chao Li', 'chao li')</td><td></td></tr><tr><td>86b105c3619a433b6f9632adcf9b253ff98aee87</td><td>1424403677/06/$20.00 ©2006 IEEE +<br/>1013 +<br/>ICME 2006 +</td><td></td><td></td></tr><tr><td>86f3552b822f6af56cb5079cc31616b4035ccc4e</td><td>Towards Miss Universe Automatic Prediction: The Evening Gown Competition +<br/><b>University of Queensland, Brisbane, Australia</b><br/>(cid:5) Data61, CSIRO, Australia +</td><td>('1850202', 'Johanna Carvajal', 'johanna carvajal')<br/>('2331880', 'Arnold Wiliem', 'arnold wiliem')<br/>('1781182', 'Conrad Sanderson', 'conrad sanderson')</td><td></td></tr><tr><td>86a8b3d0f753cb49ac3250fa14d277983e30a4b7</td><td>Exploiting Unlabeled Ages for Aging Pattern Analysis on A Large Database +<br/><b>West Virginia University, Morgantown, WV</b></td><td>('1720735', 'Chao Zhang', 'chao zhang')<br/>('1822413', 'Guodong Guo', 'guodong guo')</td><td>cazhang@mix.wvu.edu, guodong.guo@mail.wvu.edu +</td></tr><tr><td>860588fafcc80c823e66429fadd7e816721da42a</td><td>Unsupervised Discovery of Object Landmarks as Structural Representations +<br/><b>University of Michigan, Ann Arbor</b><br/>2Google Brain +</td><td>('1692992', 'Yuting Zhang', 'yuting zhang')<br/>('1857914', 'Yijie Guo', 'yijie guo')<br/>('50442731', 'Yixin Jin', 'yixin jin')<br/>('49513553', 'Yijun Luo', 'yijun luo')<br/>('46915665', 'Zhiyuan He', 'zhiyuan he')<br/>('1697141', 'Honglak Lee', 'honglak lee')</td><td>{yutingzh, guoyijie, jinyixin, lyjtour, zhiyuan, honglak}@umich.edu +<br/>honglak@google.com +</td></tr><tr><td>86b51bd0c80eecd6acce9fc538f284b2ded5bcdd</td><td></td><td></td><td></td></tr><tr><td>8699268ee81a7472a0807c1d3b1db0d0ab05f40d</td><td></td><td></td><td></td></tr><tr><td>86374bb8d309ad4dbde65c21c6fda6586ae4147a</td><td>Detect-and-Track: Efficient Pose Estimation in Videos +<br/><b>The Robotics Institute, Carnegie Mellon University</b><br/><b>Dartmouth College</b><br/>2Facebook +<br/>https://rohitgirdhar.github.io/DetectAndTrack +</td><td>('3102850', 'Rohit Girdhar', 'rohit girdhar')<br/>('2082991', 'Georgia Gkioxari', 'georgia gkioxari')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')<br/>('2210374', 'Manohar Paluri', 'manohar paluri')</td><td></td></tr><tr><td>869583b700ecf33a9987447aee9444abfe23f343</td><td></td><td></td><td></td></tr><tr><td>72282287f25c5419dc6fd9e89ec9d86d660dc0b5</td><td>A Rotation Invariant Latent Factor Model for +<br/>Moveme Discovery from Static Poses +<br/><b>California Institute of Technology, Pasadena, CA, USA</b></td><td>('3339867', 'Matteo Ruggero Ronchi', 'matteo ruggero ronchi')<br/>('14834454', 'Joon Sik Kim', 'joon sik kim')<br/>('1740159', 'Yisong Yue', 'yisong yue')</td><td>{mronchi, jkim5, yyue}@caltech.edu +</td></tr><tr><td>72a87f509817b3369f2accd7024b2e4b30a1f588</td><td>Fault diagnosis of a railway device using semi-supervised +<br/>independent factor analysis with mixing constraints +<br/>To cite this version: +<br/>using semi-supervised independent factor analysis with mixing constraints. Pattern Analysis and +<br/>Applications, Springer Verlag, 2012, 15 (3), pp.313-326. <hal-00750589> +<br/>HAL Id: hal-00750589 +<br/>https://hal.archives-ouvertes.fr/hal-00750589 +<br/>Submitted on 11 Nov 2012 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('3202810', 'Etienne Côme', 'etienne côme')<br/>('1707103', 'Latifa Oukhellou', 'latifa oukhellou')<br/>('1710347', 'Thierry Denoeux', 'thierry denoeux')<br/>('2688359', 'Patrice Aknin', 'patrice aknin')<br/>('3202810', 'Etienne Côme', 'etienne côme')<br/>('1707103', 'Latifa Oukhellou', 'latifa oukhellou')<br/>('1710347', 'Thierry Denoeux', 'thierry denoeux')<br/>('2688359', 'Patrice Aknin', 'patrice aknin')</td><td></td></tr><tr><td>72a00953f3f60a792de019a948174bf680cd6c9f</td><td>Stat Comput (2007) 17:57–70 +<br/>DOI 10.1007/s11222-006-9004-9 +<br/>Understanding the role of facial asymmetry in human face +<br/>identification +<br/>Received: May 2005 / Accepted: September 2006 / Published online: 30 January 2007 +<br/>C(cid:1) Springer Science + Business Media, LLC 2007 +</td><td>('2046854', 'Sinjini Mitra', 'sinjini mitra')</td><td></td></tr><tr><td>726b8aba2095eef076922351e9d3a724bb71cb51</td><td></td><td></td><td></td></tr><tr><td>721b109970bf5f1862767a1bec3f9a79e815f79a</td><td></td><td></td><td></td></tr><tr><td>727ecf8c839c9b5f7b6c7afffe219e8b270e7e15</td><td>LEVERAGING GEO-REFERENCED DIGITAL PHOTOGRAPHS +<br/>A DISSERTATION +<br/>SUBMITTED TO THE DEPARTMENT OF COMPUTER SCIENCE +<br/>AND THE COMMITTEE ON GRADUATE STUDIES +<br/><b>OF STANFORD UNIVERSITY</b><br/>IN PARTIAL FULFILLMENT OF THE REQUIREMENTS +<br/>FOR THE DEGREE OF +<br/>DOCTOR OF PHILOSOPHY +<br/>July 2005 +</td><td>('1687465', 'Mor Naaman', 'mor naaman')</td><td></td></tr><tr><td>72a5e181ee8f71b0b153369963ff9bfec1c6b5b0</td><td>Expression recognition in videos using a weighted +<br/>component-based feature descriptor +<br/>1. Machine Vision Group, Department of Electrical and Information Engineering, +<br/><b>University of Oulu, Finland</b><br/><b>Research Center for Learning Science, Southeast University, China</b><br/>http://www.ee.oulu.fi/mvg +</td><td>('18780812', 'Xiaohua Huang', 'xiaohua huang')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')<br/>('40608983', 'Wenming Zheng', 'wenming zheng')</td><td>{huang.xiaohua,gyzhao,mkp}@ee.oulu.fi +<br/>wenming_zheng@seu.edu.cn +</td></tr><tr><td>72ecaff8b57023f9fbf8b5b2588f3c7019010ca7</td><td>Facial Keypoints Detection +</td><td>('27744156', 'Shenghao Shi', 'shenghao shi')</td><td></td></tr><tr><td>72591a75469321074b072daff80477d8911c3af3</td><td>Group Component Analysis for Multi-block Data: +<br/>Common and Individual Feature Extraction +</td><td>('1764724', 'Guoxu Zhou', 'guoxu zhou')<br/>('1747156', 'Andrzej Cichocki', 'andrzej cichocki')<br/>('38741479', 'Yu Zhang', 'yu zhang')</td><td></td></tr><tr><td>7224d58a7e1f02b84994b60dc3b84d9fe6941ff5</td><td>When Face Recognition Meets with Deep Learning: an Evaluation of +<br/>Convolutional Neural Networks for Face Recognition +<br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, UK</b><br/><b>Electronic Engineering and Computer Science, Queen Mary University of London, UK</b><br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition, Chinese Academy of Sciences, China♠ +</td><td>('38819702', 'Guosheng Hu', 'guosheng hu')<br/>('2653152', 'Yongxin Yang', 'yongxin yang')<br/>('1716143', 'Dong Yi', 'dong yi')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{g.hu,j.kittler,w.christmas}@surrey.ac.uk,{yongxin.yang,t.hospedales}@qmul.ac.uk, {szli,dyi}@cbsr.ia.ac.cn +</td></tr><tr><td>729dbe38538fbf2664bc79847601f00593474b05</td><td></td><td></td><td></td></tr><tr><td>729a9d35bc291cc7117b924219bef89a864ce62c</td><td>Recognizing Material Properties from Images +</td><td>('40116153', 'Gabriel Schwartz', 'gabriel schwartz')<br/>('1708819', 'Ko Nishino', 'ko nishino')</td><td></td></tr><tr><td>72e10a2a7a65db7ecdc7d9bd3b95a4160fab4114</td><td>Face Alignment using Cascade Gaussian Process Regression Trees +<br/><b>Korea Advanced institute of Science and Technology</b><br/>Face alignment is a task to locate fiducial facial landmark points, such as eye +<br/>corners, nose tip, mouth corners, and chin, in a face image. Shape regression +<br/>has become an accurate, robust, and fast framework for face alignment [2, +<br/>In shape regression, face shape s = (x1,y1,··· ,xp,yp)(cid:62), that is a +<br/>4, 5]. +<br/>concatenation of p facial landmark coordinates {(xi,yi)}p +<br/>i=1, is initialized +<br/>and iteratively updated through a cascade regression trees (CRT) as shown +<br/>in Figure 1. Each tree estimates the shape increment from the current shape +<br/>estimate, and the final shape estimate is given by a cumulated sum of the +<br/>outputs of the trees to the initial estimate as follows: +<br/>ˆsT = ˆs0 + +<br/>t=1 +<br/>f t (xt;θ t ), +<br/>(1) +<br/>where T is the number of stages, t is an index that denotes the stage, ˆst is a +<br/>shape estimate, xt is a feature vector that is extracted from an input image +<br/>I, and f t (·;·) is a tree that is parameterized by θ t. Starting from the rough +<br/>initial shape estimate ˆs0, each stage iteratively updates the shape estimate +<br/>by ˆst = ˆst−1 + f t (xt;θ t ). +<br/>The two key elements of CRT-based shape regression that impact to the +<br/>prediction performance are gradient boosting [3] for learning the CRT and +<br/>the shape-indexed features [2] which the trees are based. In gradient boost- +<br/>ing, each stage iteratively fits training data in a greedy stage-wise manner by +<br/>reducing the regression residuals that are defined as the differences between +<br/>the ground truth shapes and shape estimates. The shape-indexed features +<br/>are extracted from the pixel coordinates referenced by the shape estimate. +<br/>The shape-indexed features are extremely cheap to compute and are robust +<br/>against geometric variations. +<br/>Instead of using gradient boosting, we propose cascade Gaussian pro- +<br/>cess regression trees (cGPRT) that can be incorporated as a learning method +<br/>for a CRT prediction framework. The cGPRT is constructed by combining +<br/>Gaussian process regression trees (GPRT) in a cascade stage-wise manner. +<br/>Given training samples S = (s1,··· ,sN )(cid:62) and Xt = (x1,··· ,xN )(cid:62), GPRT +<br/>models the relationship between inputs and outputs by a regression function +<br/>f (x) drawn from a Gaussian process with independent additive noise εi, +<br/>i = 1,··· ,N, +<br/>si = f (xi) + εi, +<br/>f (x) ∼ GP(0,k(x,x(cid:48))), +<br/>εi ∼ N (0,σ 2 +<br/>n ). +<br/>A kernel k(x,x(cid:48)) in GPRT is defined by a set of M number of trees: +<br/>k(x,x(cid:48)) = σ 2 +<br/>κm(x,x(cid:48)) = +<br/>m=1 +<br/>(cid:26) 1 +<br/>κm(x,x(cid:48)), +<br/>if τm(x) = τm(x(cid:48)) +<br/>otherwise, +<br/>(2) +<br/>(3) +<br/>(4) +<br/>(5) +<br/>(6) +<br/>where σ 2 +<br/>k is the scaling parameter that represents the kernel power, and τ is +<br/>a split function takes an input x and computes the leaf index b ∈ {1,··· ,B}. +<br/>Given an input x∗, distribution over its predictive variable f∗ is given as +<br/>¯f∗ = +<br/>i=1 +<br/>αik(xi,x∗), +<br/>(7) +<br/>where α = (α1,··· ,αN )(cid:62) is given by K−1 +<br/>n IN, +<br/>and K is a covariance matrix of which K(i, j) is computed from the i-th and +<br/>j-th row vector of X. Computation of Equation (7) is in O(N); however, this +<br/>can be more efficient as follows: +<br/>s S. Here, Ks is given by K+σ 2 +<br/>¯f∗ = +<br/>m=1 +<br/>¯αm,τm(x∗), +<br/>(8) +</td><td>('2350325', 'Donghoon Lee', 'donghoon lee')<br/>('2857402', 'Hyunsin Park', 'hyunsin park')</td><td></td></tr><tr><td>72160aae43cd9b2c3aae5574acc0d00ea0993b9e</td><td>Boosting Facial Expression Recognition in a Noisy Environment +<br/>Using LDSP-Local Distinctive Star Pattern +<br/> 1 Department of Computer Science and Engineering +<br/><b>Stamford University Bangladesh, Dhaka-1209, Bangladesh</b><br/>2 Department of Computer Science and Engineering +<br/><b>Stamford University Bangladesh, Dhaka-1209, Bangladesh</b><br/>3 Department of Computer Science and Engineering +<br/><b>Stamford University Bangladesh, Dhaka-1209, Bangladesh</b></td><td>('7484236', 'Mohammad Shahidul Islam', 'mohammad shahidul islam')<br/>('7497618', 'Tarin Kazi', 'tarin kazi')</td><td></td></tr><tr><td>72cbbdee4f6eeee8b7dd22cea6092c532271009f</td><td>Adversarial Occlusion-aware Face Detection +<br/>1National Laboratory of Pattern Recognition, CASIA +<br/>2Center for Research on Intelligent Perception and Computing, CASIA +<br/><b>University of Chinese Academy of Sciences, Beijing 100190, China</b></td><td>('3065234', 'Yujia Chen', 'yujia chen')<br/>('3051419', 'Lingxiao Song', 'lingxiao song')<br/>('1705643', 'Ran He', 'ran he')</td><td></td></tr><tr><td>721d9c387ed382988fce6fa864446fed5fb23173</td><td></td><td></td><td></td></tr><tr><td>72c0c8deb9ea6f59fde4f5043bff67366b86bd66</td><td>Age progression in Human Faces : A Survey +</td><td>('34713849', 'Narayanan Ramanathan', 'narayanan ramanathan')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>721e5ba3383b05a78ef1dfe85bf38efa7e2d611d</td><td>BULAT, TZIMIROPOULOS: CONVOLUTIONAL AGGREGATION OF LOCAL EVIDENCE +<br/>Convolutional aggregation of local evidence +<br/>for large pose face alignment +<br/>Computer Vision Laboratory +<br/><b>University of Nottingham</b><br/>Nottingham, UK +</td><td>('3458121', 'Adrian Bulat', 'adrian bulat')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>adrian.bulat@nottingham.ac.uk +<br/>yorgos.tzimiropoulos@nottingham.ac.uk +</td></tr><tr><td>72f4aaf7e2e3f215cd8762ce283988220f182a5b</td><td>Turk J Elec Eng & Comp Sci, Vol.18, No.4, 2010, c(cid:2) T ¨UB˙ITAK +<br/>doi:10.3906/elk-0906-48 +<br/>Active illumination and appearance model for face +<br/>alignment +<br/><b>Institute of Informatics, Istanbul Technical University, Istanbul, 34469, TURKEY</b><br/><b>Istanbul Technical University, Istanbul, 34469, TURKEY</b><br/><b>DTU Informatics, Technical University of Denmark, DK-2800 Kgs. Lyngby, DENMARK</b></td><td>('2061450', 'Fatih KAHRAMAN', 'fatih kahraman')<br/>('1762901', 'Sune DARKNER', 'sune darkner')<br/>('2134834', 'Rasmus LARSEN', 'rasmus larsen')</td><td>e-mail: kahraman@be.itu.edu.tr +<br/>e-mail: gokmen@itu.edu.tr +<br/>e-mail: {sda, rl}@imm.dtu.dk +</td></tr><tr><td>72a55554b816b66a865a1ec1b4a5b17b5d3ba784</td><td>Real-Time Face Identification +<br/>via CNN +<br/>and Boosted Hashing Forest +<br/><b>State Research Institute of Aviation Systems (GosNIIAS), Moscow, Russia</b><br/>IEEE Computer Society Workshop on Biometrics +<br/>In conjunction with CVPR 2016, June 26, 2016 +</td><td>('2966131', 'Yury Vizilter', 'yury vizilter')<br/>('5669812', 'Vladimir Gorbatsevich', 'vladimir gorbatsevich')<br/>('34296728', 'Andrey Vorotnikov', 'andrey vorotnikov')<br/>('7729536', 'Nikita Kostromov', 'nikita kostromov')</td><td>viz@gosniias.ru, gvs@gosniias.ru, vorotnikov@gosniias.ru, nikita-kostromov@yandex.ru +</td></tr><tr><td>72450d7e5cbe79b05839c30a4f0284af5aa80053</td><td>Natural Facial Expression Recognition Using Dynamic +<br/>and Static Schemes +<br/>1 Computer Vision Center, 08193 Bellaterra, Barcelona, Spain +<br/>2 IKERBASQUE, Basque Foundation for Science +<br/><b>University of the Basque Country, San Sebastian, Spain</b></td><td>('3262395', 'Bogdan Raducanu', 'bogdan raducanu')<br/>('1803584', 'Fadi Dornaika', 'fadi dornaika')</td><td>bogdan@cvc.uab.es +<br/>fadi dornaika@ehu.es +</td></tr><tr><td>72bf9c5787d7ff56a1697a3389f11d14654b4fcf</td><td>RobustFaceRecognitionUsing +<br/>SymmetricShape-from-Shading +<br/>W.Zhao +<br/>RamaChellappa +<br/>CenterforAutomationResearchand +<br/>ElectricalandComputerEngineeringDepartment +<br/><b>UniversityofMaryland</b><br/><b>CollegePark, MD</b><br/>ThesupportoftheO(cid:14)ceofNavalResearchunderGrantN +</td><td></td><td>Email:fwyzhao,ramag@cfar.umd.edu +</td></tr><tr><td>725c3605c2d26d113637097358cd4c08c19ff9e1</td><td>Deep Reasoning with Knowledge Graph for Social Relationship Understanding +<br/><b>School of Data and Computer Science, Sun Yat-sen University, China</b><br/>2 SenseTime Research, China +</td><td>('29988001', 'Zhouxia Wang', 'zhouxia wang')<br/>('1765674', 'Tianshui Chen', 'tianshui chen')<br/>('12254824', 'Weihao Yu', 'weihao yu')<br/>('47413456', 'Hui Cheng', 'hui cheng')<br/>('1737218', 'Liang Lin', 'liang lin')</td><td>zhouzi1212,tianshuichen,jimmy.sj.ren,weihaoyu6@gmail.com, +<br/>chengh9@mail.sysu.edu.cn, linliang@ieee.org +</td></tr><tr><td>445461a34adc4bcdccac2e3c374f5921c93750f8</td><td>Emotional Expression Classification using Time-Series Kernels∗ +</td><td>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')<br/>('1733113', 'Takeo Kanade', 'takeo kanade')</td><td>1E¨otv¨os Lor´and University, Budapest, Hungary, {andras.lorincz,szzoli}@elte.hu +<br/>2Carnegie Mellon University, Pittsburgh, PA, laszlo.jeni@ieee.org,tk@cs.cmu.edu +<br/>3University of Pittsburgh, Pittsburgh, PA, jeffcohn@cs.cmu.edu +</td></tr><tr><td>4414a328466db1e8ab9651bf4e0f9f1fe1a163e4</td><td>1164 +<br/>© EURASIP, 2010 ISSN 2076-1465 +<br/>18th European Signal Processing Conference (EUSIPCO-2010) +<br/>INTRODUCTION +</td><td></td><td></td></tr><tr><td>442f09ddb5bb7ba4e824c0795e37cad754967208</td><td></td><td></td><td></td></tr><tr><td>443acd268126c777bc7194e185bec0984c3d1ae7</td><td>Retrieving Relative Soft Biometrics +<br/>for Semantic Identification +<br/>School of Electronics and Computer Science, +<br/><b>University of Southampton, United Kingdom</b></td><td>('3408521', 'Daniel Martinho-Corbishley', 'daniel martinho-corbishley')<br/>('1727698', 'Mark S. Nixon', 'mark s. nixon')<br/>('3000521', 'John N. Carter', 'john n. carter')</td><td>{dmc,msn,jnc}@ecs.soton.ac.uk +</td></tr><tr><td>44f23600671473c3ddb65a308ca97657bc92e527</td><td>Convolutional Two-Stream Network Fusion for Video Action Recognition +<br/><b>Graz University of Technology</b><br/><b>Graz University of Technology</b><br/><b>University of Oxford</b></td><td>('2322150', 'Christoph Feichtenhofer', 'christoph feichtenhofer')<br/>('1718587', 'Axel Pinz', 'axel pinz')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>feichtenhofer@tugraz.at +<br/>axel.pinz@tugraz.at +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>4439746eeb7c7328beba3f3ef47dc67fbb52bcb3</td><td>An Efficient Face Detection Method Using Adaboost and Facial Parts +<br/>Computer, IT and Electronic department +<br/><b>Azad University of Qazvin</b><br/>Tehran, Iran +</td><td>('2514753', 'Yasaman Heydarzadeh', 'yasaman heydarzadeh')<br/>('2514753', 'Yasaman Heydarzadeh', 'yasaman heydarzadeh')<br/>('1681854', 'Abolfazl Toroghi Haghighat', 'abolfazl toroghi haghighat')</td><td>heydarzadeh@ qiau.ac.ir , haghighat@qiau.ac.ir +</td></tr><tr><td>446a99fdedd5bb32d4970842b3ce0fc4f5e5fa03</td><td>A Pose-Adaptive Constrained Local Model For +<br/>Accurate Head Pose Tracking +<br/>Eikeo +<br/>11 rue Leon Jouhaux, +<br/>F-75010, Paris, France +<br/>Sorbonne Universit´es +<br/>UPMC Univ Paris 06 +<br/>CNRS UMR 7222, ISIR +<br/>F-75005, Paris, France +<br/>Eikeo +<br/>11 rue Leon Jouhaux, +<br/>F-75010, Paris, France +</td><td>('2416620', 'Lucas Zamuner', 'lucas zamuner')<br/>('2521061', 'Kevin Bailly', 'kevin bailly')<br/>('2254216', 'Erwan Bigorgne', 'erwan bigorgne')</td><td>lucas.zamuner@eikeo.com +<br/>kevin.bailly@upmc.fr +<br/>erwan.bigorgne@eikeo.com +</td></tr><tr><td>4467a1ae8ddf0bc0e970c18a0cdd67eb83c8fd6f</td><td>Learning features from Improved Dense Trajectories using deep convolutional +<br/>networks for Human Activity Recognition +<br/><b>University Drive</b><br/>Burnaby, BC +<br/>Canada V5A 1S6 +<br/>Sportlogiq Inc. +<br/>780 Avenue Brewster, +<br/>Montreal QC, +<br/>Canada H4C 1A8 +<br/><b>University Drive</b><br/>Burnaby, BC +<br/>Canada V5A 1S6 +</td><td>('2716937', 'Srikanth Muralidharan', 'srikanth muralidharan')<br/>('2190580', 'Simon Fraser', 'simon fraser')<br/>('15695326', 'Mehrsan Javan', 'mehrsan javan')<br/>('10771328', 'Greg Mori', 'greg mori')<br/>('2190580', 'Simon Fraser', 'simon fraser')</td><td>smuralid@sfu.ca +<br/>mehrsan@sportlogiq.com +<br/>mori@cs.sfu.ca +</td></tr><tr><td>44b1399e8569a29eed0d22d88767b1891dbcf987</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. +<br/>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +<br/>Learning Multi-modal Latent Attributes +</td><td>('1697755', 'Timothy M. Hospedales', 'timothy m. hospedales')<br/>('1700927', 'Tao Xiang', 'tao xiang')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')</td><td></td></tr><tr><td>44f48a4b1ef94a9104d063e53bf88a69ff0f55f3</td><td>Automatically Building Face Datasets of New Domains +<br/>from Weakly Labeled Data with Pretrained Models +<br/><b>Sun Yat-sen University</b></td><td>('2442939', 'Shengyong Ding', 'shengyong ding')<br/>('4080607', 'Junyu Wu', 'junyu wu')<br/>('1723992', 'Wei Xu', 'wei xu')<br/>('38255852', 'Hongyang Chao', 'hongyang chao')</td><td></td></tr><tr><td>446dc1413e1cfaee0030dc74a3cee49a47386355</td><td>Recent Advances in Zero-shot Recognition +</td><td>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('1700927', 'Tao Xiang', 'tao xiang')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')<br/>('14517812', 'Leonid Sigal', 'leonid sigal')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')</td><td></td></tr><tr><td>44a3ec27f92c344a15deb8e5dc3a5b3797505c06</td><td>A Taxonomy of Part and Attribute Discovery +<br/>Techniques +</td><td>('35208858', 'Subhransu Maji', 'subhransu maji')</td><td></td></tr><tr><td>44aeda8493ad0d44ca1304756cc0126a2720f07b</td><td>Face Alive Icons +</td><td>('1685323', 'Xin Li', 'xin li')<br/>('2304980', 'Chieh-Chih Chang', 'chieh-chih chang')<br/>('1679040', 'Shi-Kuo Chang', 'shi-kuo chang')</td><td>1University of Pittsburgh, USA,{flying, chang}@cs.pitt.edu +<br/>2Industrial Technology Research Institute, Taiwan, chieh@itri.org.tw +</td></tr><tr><td>449b1b91029e84dab14b80852e35387a9275870e</td><td></td><td></td><td></td></tr><tr><td>44078d0daed8b13114cffb15b368acc467f96351</td><td></td><td></td><td></td></tr><tr><td>44d23df380af207f5ac5b41459c722c87283e1eb</td><td>Human Attribute Recognition by Deep +<br/>Hierarchical Contexts +<br/><b>The Chinese University of Hong Kong</b></td><td>('47002704', 'Yining Li', 'yining li')<br/>('2000034', 'Chen Huang', 'chen huang')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{ly015,chuang,ccloy,xtang}@ie.cuhk.edu.hk +</td></tr><tr><td>44c9b5c55ca27a4313daf3760a3f24a440ce17ad</td><td>Revisiting hand-crafted feature for action recognition: +<br/>a set of improved dense trajectories +<br/><b>Hiroshima University, Japan</b><br/>ENSICAEN, France +<br/><b>Hiroshima University, Japan</b></td><td>('2223849', 'Kenji Matsui', 'kenji matsui')<br/>('1744862', 'Toru Tamaki', 'toru tamaki')<br/>('30171131', 'Gwladys Auffret', 'gwladys auffret')<br/>('1688940', 'Bisser Raytchev', 'bisser raytchev')<br/>('1686272', 'Kazufumi Kaneda', 'kazufumi kaneda')</td><td></td></tr><tr><td>44dd150b9020b2253107b4a4af3644f0a51718a3</td><td>An Analysis of the Sensitivity of Active Shape +<br/>Models to Initialization when Applied to Automatic +<br/>Facial Landmarking +</td><td>('2363348', 'Keshav Seshadri', 'keshav seshadri')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td></td></tr><tr><td>447d8893a4bdc29fa1214e53499ffe67b28a6db5</td><td></td><td>('35734434', 'Maxime BERTHE', 'maxime berthe')</td><td></td></tr><tr><td>44f65e3304bdde4be04823fd7ca770c1c05c2cef</td><td>SIViP +<br/>DOI 10.1007/s11760-009-0125-4 +<br/>ORIGINAL PAPER +<br/>On the use of phase of the Fourier transform for face recognition +<br/>under variations in illumination +<br/>Received: 17 November 2008 / Revised: 20 February 2009 / Accepted: 7 July 2009 +<br/>© Springer-Verlag London Limited 2009 +</td><td>('2627097', 'Anil Kumar Sao', 'anil kumar sao')</td><td></td></tr><tr><td>44fbbaea6271e47ace47c27701ed05e15da8f7cf</td><td>588306 PSSXXX10.1177/0956797615588306Kret et al.Effect of Pupil Size on Trust +<br/>research-article2015 +<br/>Research Article +<br/>Pupil Mimicry Correlates With Trust in +<br/>In-Group Partners With Dilating Pupils +<br/> 1 –10 +<br/>© The Author(s) 2015 +<br/>Reprints and permissions: +<br/>sagepub.com/journalsPermissions.nav +<br/>DOI: 10.1177/0956797615588306 +<br/>pss.sagepub.com +<br/>M. E. Kret1,2, A. H. Fischer1,2, and C. K. W. De Dreu1,2,3 +<br/><b>University of Amsterdam; 2Amsterdam Brain and Cognition Center, University of</b><br/><b>Amsterdam; and 3Center for Experimental Economics and Political Decision Making, University of Amsterdam</b></td><td></td><td></td></tr><tr><td>44eb4d128b60485377e74ffb5facc0bf4ddeb022</td><td></td><td></td><td></td></tr><tr><td>448ed201f6fceaa6533d88b0b29da3f36235e131</td><td></td><td></td><td></td></tr><tr><td>441bf5f7fe7d1a3939d8b200eca9b4bb619449a9</td><td>Head Pose Estimation in the Wild using Approximate View Manifolds +<br/><b>University of Florida</b><br/>Gainesville, FL, USA +<br/><b>University of Florida</b><br/>Gainesville, FL, USA +</td><td>('30900274', 'Kalaivani Sundararajan', 'kalaivani sundararajan')<br/>('2171076', 'Damon L. Woodard', 'damon l. woodard')</td><td>kalaivani.s@ufl.edu +<br/>dwoodard@ufl.edu +</td></tr><tr><td>447a5e1caf847952d2bb526ab2fb75898466d1bc</td><td>Under review as a conference paper at ICLR 2018 +<br/>LEARNING NON-LINEAR TRANSFORM WITH DISCRIM- +<br/>INATIVE AND MINIMUM INFORMATION LOSS PRIORS +<br/>Anonymous authors +<br/>Paper under double-blind review +</td><td></td><td></td></tr><tr><td>449808b7aa9ee6b13ad1a21d9f058efaa400639a</td><td>Recovering 3D Facial Shape via Coupled 2D/3D Space Learning +<br/>1Key Lab of Intelligent Information Processing of CAS, +<br/><b>Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>Graduate University of CAS, 100190, Beijing, China</b><br/><b>System Research Center, NOKIA Research Center, Beijing, 100176, China</b><br/><b>Institute of Digital Media, Peking University, Beijing, 100871, China</b></td><td>('3079475', 'Annan Li', 'annan li')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')<br/>('1695600', 'Xiujuan Chai', 'xiujuan chai')<br/>('1698902', 'Wen Gao', 'wen gao')</td><td>{anli,sgshan,xlchen,wgao}@jdl.ac.cn +<br/>ext-xiujuan.chai@nokia.com +</td></tr><tr><td>2a7bca56e2539c8cf1ae4e9da521879b7951872d</td><td>Exploiting Unrelated Tasks in Multi-Task Learning +<br/>Anonymous Author 1 +<br/>Unknown Institution 1 +<br/>Anonymous Author 2 +<br/>Unknown Institution 2 +<br/>Anonymous Author 3 +<br/>Unknown Institution 3 +</td><td></td><td></td></tr><tr><td>2a65d7d5336b377b7f5a98855767dd48fa516c0f</td><td>Fast Supervised LDA for Discovering Micro-Events in +<br/>Large-Scale Video Datasets +<br/>Multimedia Understanding Group +<br/><b>Aristotle University of Thessaloniki, Greece</b></td><td>('3493855', 'Angelos Katharopoulos', 'angelos katharopoulos')<br/>('3493472', 'Despoina Paschalidou', 'despoina paschalidou')<br/>('1789830', 'Christos Diou', 'christos diou')<br/>('1708199', 'Anastasios Delopoulos', 'anastasios delopoulos')</td><td>{katharas, pdespoin}@auth.gr; diou@mug.ee.auth.gr; adelo@eng.auth.gr +</td></tr><tr><td>2af2b74c3462ccff3a6881ff7cf4f321b3242fa9</td><td>Chen ZN, Ngo CW, Zhang W et al. Name-face association in Web videos: A large-scale dataset, baselines, and open issues. +<br/>1468-z +<br/>Name-Face Association in Web Videos: A Large-Scale Dataset, +<br/>Baselines, and Open Issues +<br/><b>Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China</b><br/><b>City University of Hong Kong, Hong Kong, China</b><br/><b>Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190, China</b><br/><b>School of Computer Science, Fudan University, Shanghai 200433, China</b><br/>Received February 24, 2014; revised July 3, 2014. +</td><td>('1751681', 'Chong-Wah Ngo', 'chong-wah ngo')<br/>('40538946', 'Wei Zhang', 'wei zhang')<br/>('1778024', 'Juan Cao', 'juan cao')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')</td><td>E-mail: zhineng.chen@ia.ac.cn; cscwngo@cityu.edu.hk; wzhang34-c@my.cityu.edu.hk; caojuan@ict.ac.cn; ygj@fudan.edu.cn +</td></tr><tr><td>2aaa6969c03f435b3ea8431574a91a0843bd320b</td><td></td><td></td><td></td></tr><tr><td>2af620e17d0ed67d9ccbca624250989ce372e255</td><td>Meta-Class Features for Large-Scale Object Categorization on a Budget +<br/><b>Dartmouth College</b><br/>Hanover, NH, U.S.A. +</td><td>('34338883', 'Alessandro Bergamo', 'alessandro bergamo')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')</td><td>{aleb, lorenzo}@cs.dartmouth.edu +</td></tr><tr><td>2a35d20b2c0a045ea84723f328321c18be6f555c</td><td>on Converting Supervised Classification to Semi-supervised Classification +<br/>Boost Picking: A Universal Method +<br/><b>Beijing Institute of Technology, Beijing 100081 CHINA</b><br/><b>North China University of Technology, Beijing 100144 CHINA</b><br/><b>Beijing Institute of Technology, Beijing 100081 CHINA</b><br/><b>Beijing Institute of Technology, Beijing 100081 CHINA</b></td><td>('1742846', 'Fuqiang Liu', 'fuqiang liu')<br/>('33179404', 'Fukun Bi', 'fukun bi')<br/>('3148439', 'Yiding Yang', 'yiding yang')<br/>('36522003', 'Liang Chen', 'liang chen')</td><td></td></tr><tr><td>2ad7cef781f98fd66101fa4a78e012369d064830</td><td></td><td></td><td></td></tr><tr><td>2ad29b2921aba7738c51d9025b342a0ec770c6ea</td><td></td><td></td><td></td></tr><tr><td>2a9b398d358cf04dc608a298d36d305659e8f607</td><td>Facial Action Unit Recognition with Sparse Representation +<br/><b>University of Denver, Denver, CO</b><br/><b>University of Pittsburgh, Pittsburgh, PA</b><br/>facial +<br/>image exhibiting +</td><td>('3093835', 'Mohammad H. Mahoor', 'mohammad h. mahoor')<br/>('5510802', 'Mu Zhou', 'mu zhou')<br/>('1837267', 'Kevin L. Veon', 'kevin l. veon')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>Emails: mmahoor@du.edu, mu.zhou09fall@gmail.com, kevin.veon@du.edu, seyedmohammad.mavadati@du.edu, and jeffcohn@pitt.edu +</td></tr><tr><td>2a0efb1c17fbe78470acf01e4601a75735a805cc</td><td>Illumination-InsensitiveFaceRecognitionUsing +<br/>SymmetricShape-from-Shading +<br/>WenYiZhao +<br/>RamaChellappa +<br/>CenterforAutomationResearch +<br/><b>UniversityofMaryland, CollegePark, MD</b></td><td></td><td>Email:fwyzhao,ramag@cfar.umd.edu +</td></tr><tr><td>2a6bba2e81d5fb3c0fd0e6b757cf50ba7bf8e924</td><td></td><td></td><td></td></tr><tr><td>2ac21d663c25d11cda48381fb204a37a47d2a574</td><td>Interpreting Hand-Over-Face Gestures +<br/><b>University of Cambridge</b></td><td>('2022940', 'Marwa Mahmoud', 'marwa mahmoud')<br/>('39626495', 'Peter Robinson', 'peter robinson')</td><td></td></tr><tr><td>2a4153655ad1169d482e22c468d67f3bc2c49f12</td><td>Face Alignment Across Large Poses: A 3D Solution +<br/>1 Center for Biometrics and Security Research & National Laboratory of Pattern Recognition, +<br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/><b>Michigan State University</b></td><td>('8362374', 'Xiangyu Zhu', 'xiangyu zhu')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('1704812', 'Hailin Shi', 'hailin shi')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{xiangyu.zhu,zlei,hailin.shi,szli}@nlpr.ia.ac.cn +<br/>liuxm@msu.edu +</td></tr><tr><td>2aa2b312da1554a7f3e48f71f2fce7ade6d5bf40</td><td>Estimating Sheep Pain Level Using Facial Action Unit Detection +<br/><b>Computer Laboratory, University of Cambridge, Cambridge, UK</b></td><td>('9871228', 'Yiting Lu', 'yiting lu')<br/>('2022940', 'Marwa Mahmoud', 'marwa mahmoud')<br/>('39626495', 'Peter Robinson', 'peter robinson')</td><td></td></tr><tr><td>2aec012bb6dcaacd9d7a1e45bc5204fac7b63b3c</td><td>Robust Registration and Geometry Estimation from Unstructured +<br/>Facial Scans +</td><td>('19214361', 'Maxim Bazik', 'maxim bazik')</td><td></td></tr><tr><td>2ae139b247057c02cda352f6661f46f7feb38e45</td><td>Combining Modality Specific Deep Neural Networks for +<br/>Emotion Recognition in Video +<br/>1École Polytechique de Montréal, Université de Montréal, Montréal, Canada +<br/>2Laboratoire d’Informatique des Systèmes Adaptatifs, Université de Montréal, Montréal, Canada +</td><td>('3127597', 'Samira Ebrahimi Kahou', 'samira ebrahimi kahou')<br/>('2900675', 'Xavier Bouthillier', 'xavier bouthillier')<br/>('2558801', 'Pierre Froumenty', 'pierre froumenty')<br/>('1710604', 'Roland Memisevic', 'roland memisevic')<br/>('1724875', 'Pascal Vincent', 'pascal vincent')<br/>('1751762', 'Yoshua Bengio', 'yoshua bengio')</td><td>{samira.ebrahimi-kahou, christopher.pal, pierre.froumenty}@polymtl.ca +<br/>{bouthilx, gulcehrc, memisevr, vincentp, courvila, bengioy}@iro.umontreal.ca +</td></tr><tr><td>2a3e19d7c54cba3805115497c69069dd5a91da65</td><td>Looking at Hands in Autonomous Vehicles: +<br/>A ConvNet Approach using Part Affinity Fields +<br/>LISA: Laboratory for Intelligent & Safe Automobiles +<br/><b>University of California San Diego</b></td><td>('2812409', 'Kevan Yuen', 'kevan yuen')<br/>('1713989', 'Mohan M. Trivedi', 'mohan m. trivedi')</td><td>kcyuen@eng.ucsd.edu, mtrivedi@eng.ucsd.edu +</td></tr><tr><td>2af19b5ff2ca428fa42ef4b85ddbb576b5d9a5cc</td><td>Multi-Region Probabilistic Histograms +<br/>for Robust and Scalable Identity Inference (cid:63) +<br/>NICTA, PO Box 6020, St Lucia, QLD 4067, Australia +<br/><b>University of Queensland, School of ITEE, QLD 4072, Australia</b></td><td>('1781182', 'Conrad Sanderson', 'conrad sanderson')<br/>('2270092', 'Brian C. Lovell', 'brian c. lovell')</td><td></td></tr><tr><td>2a14b6d9f688714dc60876816c4b7cf763c029a9</td><td>Combining Multiple Sources of Knowledge in Deep CNNs for Action Recognition +<br/><b>University of North Carolina at Chapel Hill</b></td><td>('2155311', 'Eunbyung Park', 'eunbyung park')<br/>('1682965', 'Xufeng Han', 'xufeng han')<br/>('1685538', 'Tamara L. Berg', 'tamara l. berg')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')</td><td>{eunbyung,xufeng,tlberg,aberg}@cs.unc.edu +</td></tr><tr><td>2a88541448be2eb1b953ac2c0c54da240b47dd8a</td><td>Discrete Graph Hashing +<br/><b>IBM T. J. Watson Research Center</b><br/><b>Columbia University</b><br/>(cid:2)Google Research +</td><td>('39059457', 'Wei Liu', 'wei liu')<br/>('2794322', 'Sanjiv Kumar', 'sanjiv kumar')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>weiliu@us.ibm.com +<br/>cm3052@columbia.edu +<br/>sfchang@ee.columbia.edu +<br/>sanjivk@google.com +</td></tr><tr><td>2a5903bdb3fdfb4d51f70b77f16852df3b8e5f83</td><td>121 +<br/>The Effect of Computer-Generated Descriptions +<br/>on Photo-Sharing Experiences of People With +<br/>Visual Impairments +<br/>Like sighted people, visually impaired people want to share photographs on social networking services, but +<br/>find it difficult to identify and select photos from their albums. We aimed to address this problem by +<br/>incorporating state-of-the-art computer-generated descriptions into Facebook’s photo-sharing feature. We +<br/>interviewed 12 visually impaired participants to understand their photo-sharing experiences and designed a +<br/>photo description feature for the Facebook mobile application. We evaluated this feature with six +<br/>participants in a seven-day diary study. We found that participants used the descriptions to recall and +<br/>organize their photos, but they hesitated to upload photos without a sighted person’s input. In addition to +<br/>basic information about photo content, participants wanted to know more details about salient objects and +<br/>people, and whether the photos reflected their personal aesthetic. We discuss these findings from the lens of +<br/>self-disclosure and self-presentation theories and propose new computer vision research directions that will +<br/>better support visual content sharing by visually impaired people. +<br/>CCS Concepts: • Information interfaces and presentations → Multimedia and information systems; • +<br/>Social and professional topics → People with disabilities +<br/>KEYWORDS +<br/>Visual impairments; computer-generated descriptions; SNSs; photo sharing; self-disclosure; self-presentation +<br/>ACM Reference format: +<br/>The Effect of Computer-Generated Descriptions On Photo-Sharing Experiences of People With Visual +<br/>Impairments. Proc. ACM Hum.-Comput. Interact. 1, CSCW. 121 (November 2017), 22 pages. +<br/>DOI: 10.1145/3134756 +<br/>1 INTRODUCTION +<br/>Sharing memories and experiences via photos is a common way to engage with others on social networking +<br/>services (SNSs) [39,46,51]. For instance, Facebook users uploaded more than 350 million photos a day [24] +<br/>and Twitter, which initially supported only text in tweets, now has more than 28.4% of tweets containing +<br/>images [39]. Visually impaired people (both blind and low vision) have a strong presence on SNS and are +<br/>interested in sharing photos [50]. They take photos for the same reasons that sighted people do: sharing +<br/>daily moments with their sighted friends and family [30,32]. A prior study showed that visually impaired +<br/>people shared a relatively large number of photos on Facebook—only slightly less than their sighted +<br/>counterparts [50]. +<br/> +<br/> PACM on Human-Computer Interaction, Vol. 1, No. 2, Article 121. Publication date: November 2017 +</td><td>('2582568', 'Yuhang Zhao', 'yuhang zhao')<br/>('1968133', 'Shaomei Wu', 'shaomei wu')<br/>('39685591', 'Lindsay Reynolds', 'lindsay reynolds')<br/>('3283573', 'Shiri Azenkot', 'shiri azenkot')</td><td></td></tr><tr><td>2a02355c1155f2d2e0cf7a8e197e0d0075437b19</td><td></td><td></td><td></td></tr><tr><td>2a171f8d14b6b8735001a11c217af9587d095848</td><td>Learning Social Relation Traits from Face Images +<br/><b>The Chinese University of Hong Kong</b></td><td>('3152448', 'Zhanpeng Zhang', 'zhanpeng zhang')<br/>('1693209', 'Ping Luo', 'ping luo')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>zz013@ie.cuhk.edu.hk, pluo@ie.cuhk.edu.hk, ccloy@ie.cuhk.edu.hk, xtang@ie.cuhk.edu.hk +</td></tr><tr><td>2aea27352406a2066ddae5fad6f3f13afdc90be9</td><td></td><td></td><td></td></tr><tr><td>2a0623ae989f2236f5e1fe3db25ab708f5d02955</td><td>3D Face Modelling for 2D+3D Face Recognition +<br/>J.R. Tena Rodr´ıguez +<br/>Submitted for the Degree of +<br/>Doctor of Philosophy +<br/>from the +<br/><b>University of Surrey</b><br/>Centre for Vision, Speech and Signal Processing +<br/>School of Electronics and Physical Sciences +<br/><b>University of Surrey</b><br/>Guildford, Surrey GU2 7XH, U.K. +<br/>November 2007 +<br/>c(cid:13) J.R. Tena Rodr´ıguez 2007 +</td><td></td><td></td></tr><tr><td>2ad0ee93d029e790ebb50574f403a09854b65b7e</td><td>Acquiring Linear Subspaces for Face +<br/>Recognition under Variable Lighting +<br/>David Kriegman, Senior Member, IEEE +</td><td>('2457452', 'Kuang-chih Lee', 'kuang-chih lee')<br/>('1788818', 'Jeffrey Ho', 'jeffrey ho')</td><td></td></tr><tr><td>2afdda6fb85732d830cea242c1ff84497cd5f3cb</td><td>Face Image Retrieval by Using Haar Features +<br/><b>Institute ofInformation Science, Academia Sinica, Taipei, Taiwan</b><br/><b>Graduate Institute ofNetworking and Multimedia, National Taiwan University, Taipei, Taiwan</b><br/><b>Tamkang University, Taipei, Taiwan</b></td><td>('2609751', 'Bau-Cheng Shen', 'bau-cheng shen')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')<br/>('1679560', 'Hui-Huang Hsu', 'hui-huang hsu')</td><td>{bcshen, song} @ iis.sinica. edu. tw, h_hsu@mail. tku. edu. tw +</td></tr><tr><td>2ab034e1f54c37bfc8ae93f7320160748310dc73</td><td>Siamese Capsule Networks +<br/>James O’ Neill +<br/>Department of Computer Science +<br/><b>University of Liverpool</b><br/>Liverpool, L69 3BX +</td><td></td><td>james.o-neill@liverpool.ac.uk +</td></tr><tr><td>2ff9618ea521df3c916abc88e7c85220d9f0ff06</td><td>Facial Tic Detection Using Computer Vision +<br/>Christopher D. Leveille +<br/>March 20, 2014 +</td><td>('40579411', 'Aaron Cass', 'aaron cass')</td><td></td></tr><tr><td>2fda461869f84a9298a0e93ef280f79b9fb76f94</td><td>OpenFace: an open source facial behavior analysis toolkit +<br/>Tadas Baltruˇsaitis +</td><td>('39626495', 'Peter Robinson', 'peter robinson')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td>Tadas.Baltrusaitis@cl.cam.ac.uk +<br/>Peter.Robinson@cl.cam.ac.uk +<br/>morency@cs.cmu.edu +</td></tr><tr><td>2ff9ffedfc59422a8c7dac418a02d1415eec92f1</td><td>Face Verification Using Boosted Cross-Image Features +<br/><b>University of Central Florida</b><br/><b>University of California, Berkeley</b><br/>Orlando, FL +<br/>Berkeley, CA +<br/><b>University of Central Florida</b><br/>Orlando, FL +</td><td>('1720307', 'Dong Zhang', 'dong zhang')<br/>('2405613', 'Omar Oreifej', 'omar oreifej')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>dzhang@cs.ucf.edu +<br/>oreifej@eecs.berkeley.edu +<br/>shah@crcv.ucf.edu +</td></tr><tr><td>2fdce3228d384456ea9faff108b9c6d0cf39e7c7</td><td></td><td></td><td></td></tr><tr><td>2ffcd35d9b8867a42be23978079f5f24be8d3e35</td><td> +<br/>ISSN XXXX XXXX © 2018 IJESC +<br/> +<br/> +<br/>Research Article Volume 8 Issue No.6 +<br/>Satellite based Image Processing using Data mining +<br/>E.Malleshwari1, S.Nirmal Kumar2, J.Dhinesh3 +<br/>Professor1, Assistant Professor2, PG Scholar3 +<br/>Department of Information Technology1, 2, Master of Computer Applications3 +<br/><b>Vel Tech High Tech Dr Rangarajan Dr Sakunthala Engineering College, Avadi, Chennai, India</b></td><td></td><td></td></tr><tr><td>2f7e9b45255c9029d2ae97bbb004d6072e70fa79</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>cvpaper.challenge in 2015 +<br/>A review of CVPR2015 and DeepSurvey +<br/>Nakamura +<br/>Received: date / Accepted: date +</td><td>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')<br/>('29998543', 'Hironori Hoshino', 'hironori hoshino')<br/>('3407486', 'Takaaki Imanari', 'takaaki imanari')</td><td></td></tr><tr><td>2f53b97f0de2194d588bc7fb920b89cd7bcf7663</td><td>Facial Expression Recognition Using Sparse +<br/>Gaussian Conditional Random Field +<br/>School of Electrical and Computer Engineering +<br/>School of Electrical and Computer Engineering +<br/><b>Shiraz University</b><br/>Shiraz, Iran +<br/><b>Shiraz University</b><br/>Shiraz, Iran +</td><td>('37514045', 'Mohammadamin Abbasnejad', 'mohammadamin abbasnejad')<br/>('2229932', 'Mohammad Ali Masnadi-Shirazi', 'mohammad ali masnadi-shirazi')</td><td>Email: amin.abbasnejad@gmail.com +<br/>Email: mmasnadi@shirazu.ac.ir +</td></tr><tr><td>2f16baddac6af536451b3216b02d3480fc361ef4</td><td>Web-Scale Training for Face +<br/>Identification +<br/>1 Facebook AI Research +<br/><b>Tel Aviv University</b></td><td>('2909406', 'Ming Yang', 'ming yang')<br/>('2188620', 'Yaniv Taigman', 'yaniv taigman')</td><td></td></tr><tr><td>2f489bd9bfb61a7d7165a2f05c03377a00072477</td><td>JIA, YANG: STRUCTURED SEMI-SUPERVISED FOREST +<br/>Structured Semi-supervised Forest for +<br/>Facial Landmarks Localization with Face +<br/>Mask Reasoning +<br/>1 Department of Computer Science +<br/>The Univ. of Hong Kong, HK +<br/>2 School of EECS +<br/>Queen Mary Univ. of London, UK +<br/>Angran Lin1 +</td><td>('34760532', 'Xuhui Jia', 'xuhui jia')<br/>('2966679', 'Heng Yang', 'heng yang')<br/>('40392393', 'Kwok-Ping Chan', 'kwok-ping chan')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')</td><td>xhjia@cs.hku.hk +<br/>heng.yang@qmul.ac.uk +<br/>arlin@cs.hku.hk +<br/>kpchan@cs.hku.hk +<br/>i.patras@qmul.ac.uk +</td></tr><tr><td>2f2aa67c5d6dbfaf218c104184a8c807e8b29286</td><td>Video Analytics for Surveillance Camera Networks +<br/>(Invited Paper) +<br/><b>Interactive and Digital Media Institute</b><br/><b>National University of Singapore, Singapore</b></td><td>('1986874', 'Lekha Chaisorn', 'lekha chaisorn')<br/>('3026404', 'Yongkang Wong', 'yongkang wong')</td><td></td></tr><tr><td>2f16459e2e24dc91b3b4cac7c6294387d4a0eacf</td><td></td><td></td><td></td></tr><tr><td>2f59f28a1ca3130d413e8e8b59fb30d50ac020e2</td><td>Children Gender Recognition Under Unconstrained +<br/>Conditions Based on Contextual Information +<br/>Joint Research Centre, European Commission, Ispra, Italy +</td><td>('3309307', 'Riccardo Satta', 'riccardo satta')<br/>('1907426', 'Javier Galbally', 'javier galbally')<br/>('2730666', 'Laurent Beslay', 'laurent beslay')</td><td>Email: {riccardo.satta,javier.galbally,laurent.beslay}@jrc.ec.europa.eu +</td></tr><tr><td>2f78e471d2ec66057b7b718fab8bfd8e5183d8f4</td><td>SOFTWARE ENGINEERING +<br/>VOLUME: 14 | NUMBER: 5 | 2016 | DECEMBER +<br/>An Investigation of a New Social Networks +<br/>Contact Suggestion Based on Face Recognition +<br/>Algorithm +<br/>1Modeling Evolutionary Algorithms Simulation and Artificial Intelligence, Faculty of Electrical & Electronics +<br/><b>Engineering, Ton Duc Thang University, 19 Nguyen Huu Tho Street, Ho Chi Minh City, Vietman</b><br/>2Department of Computer Science, Faculty of Electrical Engineering and Computer Science, +<br/><b>VSB Technical University of Ostrava, 17. listopadu 15, 708 33 Ostrava, Czech Republic</b><br/>DOI: 10.15598/aeee.v14i5.1116 +</td><td>('1681072', 'Ivan ZELINKA', 'ivan zelinka')<br/>('1856530', 'Petr SALOUN', 'petr saloun')<br/>('2053234', 'Jakub STONAWSKI', 'jakub stonawski')<br/>('2356663', 'Adam ONDREJKA', 'adam ondrejka')</td><td>ivan.zelinka@tdt.edu.vn, petr.saloun@vsb.cz, stonawski.jakub@gmail.com, adam.ondrejka@gmail.com +</td></tr><tr><td>2fc43c2c3f7ad1ca7a1ce32c5a9a98432725fb9a</td><td>Hierarchical Video Generation from Orthogonal +<br/>Information: Optical Flow and Texture +<br/><b>The University of Tokyo</b><br/><b>The University of Tokyo</b><br/><b>The University of Tokyo</b><br/><b>The University of Tokyo / RIKEN</b></td><td>('8197937', 'Katsunori Ohnishi', 'katsunori ohnishi')<br/>('48333400', 'Shohei Yamamoto', 'shohei yamamoto')<br/>('3250559', 'Yoshitaka Ushiku', 'yoshitaka ushiku')<br/>('1790553', 'Tatsuya Harada', 'tatsuya harada')</td><td>ohnishi@mi.t.u-tokyo.ac.jp +<br/>yamamoto@mi.t.u-tokyo.ac.jp +<br/>ushiku@mi.t.u-tokyo.ac.jp +<br/>harada@mi.t.u-tokyo.ac.jp +</td></tr><tr><td>2f88d3189723669f957d83ad542ac5c2341c37a5</td><td>Downloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Electronic-Imaging on 9/13/2018 +<br/>Terms of Use: https://www.spiedigitallibrary.org/terms-of-use +<br/>Attribute-correlatedlocalregionsfordeeprelativeattributeslearningFenZhangXiangweiKongZeJiaFenZhang,XiangweiKong,ZeJia,“Attribute-correlatedlocalregionsfordeeprelativeattributeslearning,”J.Electron.Imaging27(4),043021(2018),doi:10.1117/1.JEI.27.4.043021. </td><td></td><td></td></tr><tr><td>2fda164863a06a92d3a910b96eef927269aeb730</td><td>Names and Faces in the News +<br/>Computer Science Division +<br/>U.C. Berkeley +<br/>Berkeley, CA 94720 +</td><td>('1685538', 'Tamara L. Berg', 'tamara l. berg')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('34497462', 'Jaety Edwards', 'jaety edwards')<br/>('1965929', 'Michael Maire', 'michael maire')<br/>('6714943', 'Ryan White', 'ryan white')</td><td>daf@cs.berkeley.edu +</td></tr><tr><td>2fa057a20a2b4a4f344988fee0a49fce85b0dc33</td><td></td><td></td><td></td></tr><tr><td>2f8ef26bfecaaa102a55b752860dbb92f1a11dc6</td><td>A Graph Based Approach to Speaker Retrieval in Talk +<br/>Show Videos with Transcript-Based Supervision +</td><td>('1859487', 'Yina Han', 'yina han')<br/>('1774346', 'Guizhong Liu', 'guizhong liu')<br/>('1692389', 'Hichem Sahbi', 'hichem sahbi')<br/>('1693574', 'Gérard Chollet', 'gérard chollet')</td><td></td></tr><tr><td>2f17f6c460e02bd105dcbf14c9b73f34c5fb59bd</td><td>Article +<br/>Robust Face Recognition Using the Deep C2D-CNN +<br/>Model Based on Decision-Level Fusion +<br/><b>School of Electronic and Information, Yangtze University, Jingzhou 434023, China</b><br/><b>National Demonstration Center for Experimental Electrical and Electronic Education, Yangtze University</b><br/>Jingzhou 434023, China +<br/>† These authors contributed equally to this work. +<br/>Received: 20 May 2018; Accepted: 25 June 2018; Published: 28 June 2018 +</td><td>('1723081', 'Jing Li', 'jing li')<br/>('48216473', 'Tao Qiu', 'tao qiu')<br/>('41208300', 'Chang Wen', 'chang wen')<br/>('36203475', 'Kai Xie', 'kai xie')</td><td>201501479@yangtzeu.edu.cn (J.L.); 500646@yangtzeu.edu.cn (K.X.); wenfangqing@yangtzeu.edu.cn (F-Q.W.) +<br/>School of Computer Science, Yangtze University, Jingzhou 434023, China; 201603441@yangtzeu.edu.cn +<br/>* Correspondence: 400100@yangtzeu.edu.cn; Tel.: +86-136-9731-5482 +</td></tr><tr><td>2f184c6e2c31d23ef083c881de36b9b9b6997ce9</td><td>Polichotomies on Imbalanced Domains +<br/>by One-per-Class Compensated Reconstruction Rule +<br/>Integrated Research Centre, Universit´a Campus Bio-Medico of Rome, Rome, Italy +</td><td>('1720099', 'Paolo Soda', 'paolo soda')</td><td>{r.dambrosio,p.soda}@unicampus.it +</td></tr><tr><td>2f9c173ccd8c1e6b88d7fb95d6679838bc9ca51d</td><td></td><td></td><td></td></tr><tr><td>2f8183b549ec51b67f7dad717f0db6bf342c9d02</td><td></td><td></td><td></td></tr><tr><td>2f13dd8c82f8efb25057de1517746373e05b04c4</td><td>EVALUATION OF STATE-OF-THE-ART ALGORITHMS FOR REMOTE FACE +<br/>RECOGNITION +<br/><b>University</b><br/><b>of Maryland, College Park, MD 20742, USA</b></td><td>('38811046', 'Jie Ni', 'jie ni')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>2fa1fc116731b2b5bb97f06d2ac494cb2b2fe475</td><td>A novel approach to personal photo album representation +<br/>and management +<br/>Universit`a di Palermo - Dipartimento di Ingegneria Informatica +<br/>Viale delle Scienze, 90128, Palermo, Italy +</td><td>('1762753', 'Edoardo Ardizzone', 'edoardo ardizzone')<br/>('9127836', 'Marco La Cascia', 'marco la cascia')<br/>('1698741', 'Filippo Vella', 'filippo vella')</td><td></td></tr><tr><td>2f2406551c693d616a840719ae1e6ea448e2f5d3</td><td>Age Estimation from Face Images: +<br/>Human vs. Machine Performance +<br/>Pattern Recognition & Image Processing Laboratory +<br/><b>Michigan State University</b></td><td>('34393045', 'Hu Han', 'hu han')<br/>('40653304', 'Charles Otto', 'charles otto')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>2f882ceaaf110046e63123b495212d7d4e99f33d</td><td>High Frequency Component Compensation based Super-resolution +<br/>Algorithm for Face Video Enhancement +<br/>CVRR Lab, UC San Diego, La Jolla, CA 92093, USA +</td><td>('1807917', 'Junwen Wu', 'junwen wu')</td><td></td></tr><tr><td>2f95340b01cfa48b867f336185e89acfedfa4d92</td><td>Face Expression Recognition with a 2-Channel +<br/>Convolutional Neural Network +<br/><b></b><br/>Vogt-K¨olln-Straße 30, 22527 Hamburg, Germany +<br/>http://www.informatik.uni-hamburg.de/WTM/ +</td><td>('2283866', 'Dennis Hamester', 'dennis hamester')<br/>('1736513', 'Stefan Wermter', 'stefan wermter')</td><td>{hamester,barros,wermter}@informatik.uni-hamburg.de +</td></tr><tr><td>2f7fc778e3dec2300b4081ba2a1e52f669094fcd</td><td>Action Representation Using Classifier Decision Boundaries +<br/>3 Fatih Porikli1 +<br/>1Data61/CSIRO, +<br/>2Australian Centre for Robotic Vision +<br/><b>The Australian National University, Canberra, Australia</b></td><td>('36541522', 'Jue Wang', 'jue wang')<br/>('2691929', 'Anoop Cherian', 'anoop cherian')<br/>('2377076', 'Stephen Gould', 'stephen gould')</td><td>firstname.lastname@anu.edu.au +</td></tr><tr><td>2fea258320c50f36408032c05c54ba455d575809</td><td></td><td></td><td></td></tr><tr><td>2f0e5a4b0ef89dd2cf55a4ef65b5c78101c8bfa1</td><td>Facial Expression Recognition Using a Hybrid CNN–SIFT Aggregator +<br/>Mundher Ahmed Al-Shabi +<br/>Tee Connie +<br/>Faculty of Information Science and Technology (FIST) +<br/><b>Multimedia University</b><br/>Melaka, Malaysia +</td><td>('1700590', 'Wooi Ping Cheah', 'wooi ping cheah')</td><td></td></tr><tr><td>2faa09413162b0a7629db93fbb27eda5aeac54ca</td><td>NISTIR 7674 +<br/>Quantifying How Lighting and Focus +<br/>Affect Face Recognition Performance +<br/>Phillips, P. J. +<br/>Beveridge, J. R. +<br/>Draper, B. +<br/>Bolme, D. +<br/>Givens, G. H. +<br/>Lui, Y. M. +<br/>1 +</td><td></td><td></td></tr><tr><td>2f5e057e35a97278a9d824545d7196c301072ebf</td><td>Capturing long-tail distributions of object subcategories +<br/><b>University of California, Irvine</b><br/>Google Inc. +<br/><b>University of California, Irvine</b></td><td>('32542103', 'Xiangxin Zhu', 'xiangxin zhu')<br/>('1838674', 'Dragomir Anguelov', 'dragomir anguelov')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td>xzhu@ics.uci.edu +<br/>dragomir@google.com +<br/>dramanan@ics.uci.edu +</td></tr><tr><td>2f04ba0f74df046b0080ca78e56898bd4847898b</td><td>Aggregate Channel Features for Multi-view Face Detection +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences, China</b></td><td>('1716231', 'Bin Yang', 'bin yang')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jjyan,zlei,szli}@nlpr.ia.ac.cn +<br/>yb.derek@gmail.com +</td></tr><tr><td>433bb1eaa3751519c2e5f17f47f8532322abbe6d</td><td></td><td></td><td></td></tr><tr><td>4300fa1221beb9dc81a496cd2f645c990a7ede53</td><td></td><td></td><td></td></tr><tr><td>43010792bf5cdb536a95fba16b8841c534ded316</td><td>Towards General Motion-Based Face Recognition +<br/><b>School of Computing, National University of Singapore, Singapore</b></td><td>('2268503', 'Ning Ye', 'ning ye')<br/>('1715286', 'Terence Sim', 'terence sim')</td><td>{yening,tsim}@comp.nus.edu.sg +</td></tr><tr><td>43bb20ccfda7b111850743a80a5929792cb031f0</td><td>PhD Dissertation +<br/>International Doctorate School in Information and +<br/>Communication Technologies +<br/><b>DISI - University of Trento</b><br/>Discrimination of Computer Generated +<br/>versus Natural Human Faces +<br/>Advisor: +<br/>Prof. Giulia Boato +<br/>Universit`a degli Studi di Trento +<br/>Co-Advisor: +<br/>Prof. Francesco G. B. De Natale +<br/>Universit`a degli Studi di Trento +<br/>February 2014 +</td><td>('2598811', 'Duc-Tien Dang-Nguyen', 'duc-tien dang-nguyen')</td><td></td></tr><tr><td>438c4b320b9a94a939af21061b4502f4a86960e3</td><td>Reconstruction-Based Disentanglement for Pose-invariant Face Recognition +<br/><b>Rutgers, The State University of New Jersey</b><br/><b>University of California, San Diego</b><br/>‡ NEC Laboratories America +</td><td>('4340744', 'Xi Peng', 'xi peng')<br/>('39960064', 'Xiang Yu', 'xiang yu')<br/>('1729571', 'Kihyuk Sohn', 'kihyuk sohn')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>{xipeng.cs, dnm}@rutgers.edu, {xiangyu,ksohn,manu}@nec-labs.com +</td></tr><tr><td>439ac8edfa1e7cbc65474cab544a5b8c4c65d5db</td><td>SIViP (2011) 5:401–413 +<br/>DOI 10.1007/s11760-011-0244-6 +<br/>ORIGINAL PAPER +<br/>Face authentication with undercontrolled pose and illumination +<br/>Received: 15 September 2010 / Revised: 14 December 2010 / Accepted: 17 February 2011 / Published online: 7 August 2011 +<br/>© Springer-Verlag London Limited 2011 +</td><td>('1763890', 'Maria De Marsico', 'maria de marsico')</td><td></td></tr><tr><td>43f6953804964037ff91a4f45d5b5d2f8edfe4d5</td><td>Multi-Feature Fusion in Advanced Robotics Applications +<br/>Institut für Informatik +<br/>Technische Universität München +<br/>D-85748 Garching, Germany +</td><td>('1725709', 'Zahid Riaz', 'zahid riaz')<br/>('1685773', 'Christoph Mayer', 'christoph mayer')<br/>('1746229', 'Michael Beetz', 'michael beetz')<br/>('1699132', 'Bernd Radig', 'bernd radig')</td><td>{riaz,mayerc,beetz,radig}@in.tum.de +</td></tr><tr><td>439ec47725ae4a3660e509d32828599a495559bf</td><td>Facial Expressions Tracking and Recognition: Database Protocols for Systems Validation +<br/>and Evaluation +</td><td></td><td></td></tr><tr><td>43e99b76ca8e31765d4571d609679a689afdc99e</td><td>Learning Dense Facial Correspondences in Unconstrained Images +<br/><b>University of Southern California</b><br/>2Adobe Research +<br/>3Pinscreen +<br/><b>USC Institute for Creative Technologies</b></td><td>('9965153', 'Ronald Yu', 'ronald yu')<br/>('2059597', 'Shunsuke Saito', 'shunsuke saito')<br/>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('39686979', 'Duygu Ceylan', 'duygu ceylan')<br/>('1706574', 'Hao Li', 'hao li')</td><td></td></tr><tr><td>4377b03bbee1f2cf99950019a8d4111f8de9c34a</td><td>Selective Encoding for Recognizing Unreliably Localized Faces +<br/><b>Institute for Advanced Computer Studies</b><br/><b>University of Maryland, College Park, MD</b></td><td>('40592297', 'Ang Li', 'ang li')<br/>('2852035', 'Vlad I. Morariu', 'vlad i. morariu')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td>{angli, morariu, lsd}@umiacs.umd.edu +</td></tr><tr><td>43a03cbe8b704f31046a5aba05153eb3d6de4142</td><td>Towards Robust Face Recognition from Video +<br/>Image Science and Machine Vision Group +<br/>Oak Ridge National Laboratory +<br/>Oak Ridge, TN 37831-6010 +</td><td>('3211433', 'Jeffery R. Price', 'jeffery r. price')<br/>('2743462', 'Timothy F. Gee', 'timothy f. gee')</td><td>{pricejr, geetf}@ornl.gov +</td></tr><tr><td>434bf475addfb580707208618f99c8be0c55cf95</td><td>UNDER CONSIDERATION FOR PUBLICATION IN PATTERN RECOGNITION LETTERS +<br/>DeXpression: Deep Convolutional Neural +<br/>Network for Expression Recognition +<br/><b>German Research Center for Arti cial Intelligence (DFKI), Kaiserslautern, Germany</b><br/><b>University of Kaiserslautern, Gottlieb-Daimler-Str., Kaiserslautern 67663, Germany</b></td><td>('20651567', 'Peter Burkert', 'peter burkert')<br/>('3026604', 'Felix Trier', 'felix trier')<br/>('6149779', 'Muhammad Zeshan Afzal', 'muhammad zeshan afzal')<br/>('1703343', 'Andreas Dengel', 'andreas dengel')<br/>('1743758', 'Marcus Liwicki', 'marcus liwicki')</td><td>p burkert11@cs.uni-kl.de, f +<br/>trier10@cs.uni-kl.de, afzal@iupr.com, andreas.dengel@dfki.de, +<br/>liwicki@dfki.uni-kl.de +</td></tr><tr><td>43836d69f00275ba2f3d135f0ca9cf88d1209a87</td><td>Ozaki et al. IPSJ Transactions on Computer Vision and +<br/>Applications (2017) 9:20 +<br/>DOI 10.1186/s41074-017-0030-7 +<br/>IPSJ Transactions on Computer +<br/>Vision and Applications +<br/>RESEARCH PAPER +<br/>Open Access +<br/>Effective hyperparameter optimization +<br/>using Nelder-Mead method in deep learning +</td><td>('2167404', 'Yoshihiko Ozaki', 'yoshihiko ozaki')<br/>('30735847', 'Masaki Yano', 'masaki yano')<br/>('1703823', 'Masaki Onishi', 'masaki onishi')</td><td></td></tr><tr><td>4307e8f33f9e6c07c8fc2aeafc30b22836649d8c</td><td>Supervised Earth Mover’s Distance Learning +<br/>and its Computer Vision Applications +<br/><b>Stanford University, CA, United States</b></td><td>('1716453', 'Fan Wang', 'fan wang')<br/>('1744254', 'Leonidas J. Guibas', 'leonidas j. guibas')</td><td></td></tr><tr><td>435642641312364e45f4989fac0901b205c49d53</td><td>Face Model Compression +<br/>by Distilling Knowledge from Neurons +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Key Lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('1693209', 'Ping Luo', 'ping luo')<br/>('2042558', 'Zhenyao Zhu', 'zhenyao zhu')<br/>('3243969', 'Ziwei Liu', 'ziwei liu')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{pluo,zz012,lz013,xtang}@ie.cuhk.edu.hk, {xgwang}@ee.cuhk.edu.hk +</td></tr><tr><td>43aa40eaa59244c233f83d81f86e12eba8d74b59</td><td></td><td></td><td></td></tr><tr><td>4362368dae29cc66a47114d5ffeaf0534bf0159c</td><td>UACEE International Journal of Artificial Intelligence and Neural Networks ISSN:- 2250-3749 (online) +<br/>Performance Analysis of FDA Based Face +<br/>Recognition Using Correlation, ANN and SVM +<br/>Department of Computer Engineering +<br/>Department of Computer Engineering +<br/>Department of Computer Engineering +<br/>Anand, INDIA +<br/>Anand, INDIA +<br/>Anand, INDIA +</td><td>('9318822', 'Mahesh Goyani', 'mahesh goyani')<br/>('40632096', 'Ronak Paun', 'ronak paun')<br/>('40803051', 'Sardar Patel', 'sardar patel')<br/>('40803051', 'Sardar Patel', 'sardar patel')<br/>('40803051', 'Sardar Patel', 'sardar patel')</td><td>e- mail : mgoyani@gmail.com +<br/>e- mail : akashdhorajiya@gmail.com +<br/>e- mail : ronak_paun@yahoo.com +</td></tr><tr><td>43e268c118ac25f1f0e984b57bc54f0119ded520</td><td></td><td></td><td></td></tr><tr><td>4350bb360797a4ade4faf616ed2ac8e27315968e</td><td><b>MITSUBISHI ELECTRIC RESEARCH LABORATORIES</b><br/>http://www.merl.com +<br/>Edge Suppression by Gradient Field +<br/>Transformation using Cross-Projection +<br/>Tensors +<br/>TR2006-058 +<br/>June 2006 +</td><td>('1717566', 'Ramesh Raskar', 'ramesh raskar')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>43476cbf2a109f8381b398e7a1ddd794b29a9a16</td><td>A Practical Transfer Learning Algorithm for Face Verification +<br/>David Wipf +</td><td>('2032273', 'Xudong Cao', 'xudong cao')<br/>('1716835', 'Fang Wen', 'fang wen')<br/>('3168114', 'Genquan Duan', 'genquan duan')<br/>('40055995', 'Jian Sun', 'jian sun')</td><td>{xudongca,davidwip,fangwen,genduan,jiansun}@microsoft.com +</td></tr><tr><td>4353d0dcaf450743e9eddd2aeedee4d01a1be78b</td><td>Learning Discriminative LBP-Histogram Bins +<br/>for Facial Expression Recognition +<br/>Philips Research, High Tech Campus 36, Eindhoven 5656 AE, The Netherlands +</td><td>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('3006670', 'Tommaso Gritti', 'tommaso gritti')</td><td>{caifeng.shan, tommaso.gritti}@philips.com +</td></tr><tr><td>437a720c6f6fc1959ba95e48e487eb3767b4e508</td><td></td><td></td><td></td></tr><tr><td>436d80cc1b52365ed7b2477c0b385b6fbbb51d3b</td><td></td><td></td><td></td></tr><tr><td>434d6726229c0f556841fad20391c18316806f73</td><td>Detecting Visual Relationships with Deep Relational Networks +<br/><b>The Chinese University of Hong Kong</b></td><td>('38222190', 'Bo Dai', 'bo dai')<br/>('2617419', 'Yuqi Zhang', 'yuqi zhang')<br/>('1807606', 'Dahua Lin', 'dahua lin')</td><td>db014@ie.cuhk.edu.hk +<br/>zy016@ie.cuhk.edu.hk +<br/>dhlin@ie.cuhk.edu.hk +</td></tr><tr><td>43b8b5eeb4869372ef896ca2d1e6010552cdc4d4</td><td>Large-scale Supervised Hierarchical Feature Learning for Face Recognition +<br/>Intel Labs China +</td><td>('35423937', 'Jianguo Li', 'jianguo li')<br/>('6060281', 'Yurong Chen', 'yurong chen')</td><td></td></tr><tr><td>43ae4867d058453e9abce760ff0f9427789bab3a</td><td>951 +<br/>Graph Embedded Nonparametric Mutual +<br/>Information For Supervised +<br/>Dimensionality Reduction +</td><td>('2784463', 'Dimitrios Bouzas', 'dimitrios bouzas')<br/>('2965236', 'Nikolaos Arvanitopoulos', 'nikolaos arvanitopoulos')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')</td><td></td></tr><tr><td>435dc062d565ce87c6c20a5f49430eb9a4b573c4</td><td>to appear. +<br/>Lighting Condition Adaptation +<br/>for Perceived Age Estimation +<br/>NEC Soft, Ltd., Japan +<br/><b>Tokyo Institute of Technology, Japan</b><br/>NEC Soft, Ltd., Japan +</td><td>('2163491', 'Kazuya Ueki', 'kazuya ueki')<br/>('1719221', 'Masashi Sugiyama', 'masashi sugiyama')<br/>('1853974', 'Yasuyuki Ihara', 'yasuyuki ihara')</td><td></td></tr><tr><td>430c4d7ad76e51d83bbd7ec9d3f856043f054915</td><td></td><td></td><td></td></tr><tr><td>438b88fe40a6f9b5dcf08e64e27b2719940995e0</td><td>Building a Classi(cid:2)cation Cascade for Visual Identi(cid:2)cation from One Example +<br/>Computer Science, U.C. Berkeley +<br/>Computer Science, UMass Amherst +<br/>Computer Science, U.C. Berkeley +</td><td>('3236352', 'Andras Ferencz', 'andras ferencz')<br/>('1714536', 'Erik G. Learned-Miller', 'erik g. learned-miller')<br/>('1689212', 'Jitendra Malik', 'jitendra malik')</td><td>ferencz@cs.berkeley.edu +<br/>elm@cs.umass.edu +<br/>malik@cs.berkeley.edu +</td></tr><tr><td>433a6d6d2a3ed8a6502982dccc992f91d665b9b3</td><td>Transferring Landmark Annotations for +<br/>Cross-Dataset Face Alignment +<br/><b>The Chinese University of Hong Kong</b><br/><b>Tsinghua University</b></td><td>('2226254', 'Shizhan Zhu', 'shizhan zhu')<br/>('40475617', 'Cheng Li', 'cheng li')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>438e7999c937b94f0f6384dbeaa3febff6d283b6</td><td>Face Detection, Bounding Box Aggregation and Pose Estimation for Robust +<br/>Facial Landmark Localisation in the Wild +<br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, Guildford GU2 7XH, UK</b><br/><b>School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China</b></td><td>('2976854', 'Zhen-Hua Feng', 'zhen-hua feng')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td>{z.feng, j.kittler, m.a.rana, p.huber}@surrey.ac.uk, wu xiaojun@jiangnan.edu.cn +</td></tr><tr><td>43776d1bfa531e66d5e9826ff5529345b792def7</td><td>Automatic Critical Event Extraction +<br/>and Semantic Interpretation +<br/>by Looking-Inside +<br/>Laboratory for Intelligent and Safe Automobiles +<br/><b>University of California, San Diego</b><br/>Sept 17th, 2015 +</td><td>('1841835', 'Sujitha Martin', 'sujitha martin')<br/>('1802326', 'Eshed Ohn-Bar', 'eshed ohn-bar')<br/>('1713989', 'Mohan M. Trivedi', 'mohan m. trivedi')</td><td></td></tr><tr><td>43fb9efa79178cb6f481387b7c6e9b0ca3761da8</td><td>Mixture of Parts Revisited: Expressive Part Interactions for Pose Estimation +<br/>Anoop R Katti +<br/>IIT Madras +<br/>Chennai, India +<br/>IIT Madras +<br/>Chennai, India +</td><td>('1717115', 'Anurag Mittal', 'anurag mittal')</td><td>akatti@cse.iitm.ac.in +<br/>amittal@cse.iitm.ac.in +</td></tr><tr><td>432d8cba544bf7b09b0455561fea098177a85db1</td><td>Published as a conference paper at ICLR 2017 +<br/>TOWARDS A NEURAL STATISTICIAN +<br/>Harrison Edwards +<br/>School of Informatics +<br/><b>University of Edinburgh</b><br/>Edinburgh, UK +<br/>Amos Storkey +<br/>School of Informatics +<br/><b>University of Edinburgh</b><br/>Edinburgh, UK +</td><td></td><td>H.L.Edwards@sms.ed.ac.uk +<br/>A.Storkey@ed.ac.uk +</td></tr><tr><td>43ed518e466ff13118385f4e5d039ae4d1c000fb</td><td>Classification of Occluded Objects using Fast Recurrent +<br/>Processing +<br/>Ozgur Yilmaza,∗ +<br/><b>aTurgut Ozal University, Ankara Turkey</b></td><td></td><td></td></tr><tr><td>439647914236431c858535a2354988dde042ef4d</td><td>Face Illumination Normalization on Large and Small Scale Features +<br/><b>School of Mathematics and Computational Science, Sun Yat-sen University, China</b><br/><b>School of Information Science and Technology, Sun Yat-sen University, China</b><br/>3 Guangdong Province Key Laboratory of Information Security, China, +<br/><b>Hong Kong Baptist University</b></td><td>('2002129', 'Xiaohua Xie', 'xiaohua xie')<br/>('3333315', 'Wei-Shi Zheng', 'wei-shi zheng')<br/>('1768574', 'Pong C. Yuen', 'pong c. yuen')</td><td>Email: sysuxiexh@gmail.com, wszheng@ieee.org, stsljh@mail.sysu.edu.cn, pcyuen@comp.hkbu.edu.hk +</td></tr><tr><td>43d7d0d0d0e2d6cf5355e60c4fe5b715f0a1101a</td><td>Pobrane z czasopisma Annales AI- Informatica http://ai.annales.umcs.pl +<br/>Data: 04/05/2018 16:53:32 +<br/>U M CS +</td><td></td><td></td></tr><tr><td>439ca6ded75dffa5ddea203dde5e621dc4a88c3e</td><td>Robust Real-time Performance-driven 3D Face Tracking +<br/><b>School of Computer Science and Engineering, Nanyang Technological University, Singapore</b><br/><b>Rutgers University, USA</b></td><td>('1736042', 'Vladimir Pavlovic', 'vladimir pavlovic')<br/>('1688642', 'Jianfei Cai', 'jianfei cai')<br/>('1775268', 'Tat-Jen Cham', 'tat-jen cham')</td><td>{hxp1,vladimir}@cs.rutgers.edu +<br/>{asjfcai,astfcham}@ntu.edu.sg +</td></tr><tr><td>88e090ffc1f75eed720b5afb167523eb2e316f7f</td><td>Attribute-Based Transfer Learning for Object +<br/>Categorization with Zero/One Training Example +<br/><b>University of Maryland, College Park, MD, USA</b></td><td>('3099583', 'Xiaodong Yu', 'xiaodong yu')<br/>('1697493', 'Yiannis Aloimonos', 'yiannis aloimonos')</td><td>xdyu@umiacs.umd.edu, yiannis@cs.umd.edu +</td></tr><tr><td>8877e0b2dc3d2e8538c0cfee86b4e8657499a7c4</td><td>AUTOMATIC FACIAL EXPRESSION RECOGNITION FOR AFFECTIVE COMPUTING +<br/>BASED ON BAG OF DISTANCES +<br/><b>National Chung Cheng University, Chiayi, Taiwan, R.O.C</b><br/>E-mail: {hfs95p,wylin}cs.ccu.edu.tw +<br/><b>National Taichung University of Science and Technology, Taichung, Taiwan, R.O.C</b></td><td>('2240934', 'Fu-Song Hsu', 'fu-song hsu')<br/>('1682393', 'Wei-Yang Lin', 'wei-yang lin')<br/>('2080026', 'Tzu-Wei Tsai', 'tzu-wei tsai')</td><td>E-mail: wei@nutc.edu.tw +</td></tr><tr><td>88c6d4b73bd36e7b5a72f3c61536c8c93f8d2320</td><td>Image patch modeling in a light field +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2014-81 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2014/EECS-2014-81.html +<br/>May 15, 2014 +</td><td>('2040369', 'Zeyu Li', 'zeyu li')</td><td></td></tr><tr><td>889bc64c7da8e2a85ae6af320ae10e05c4cd6ce7</td><td>174 +<br/>Using Support Vector Machines to Enhance the +<br/>Performance of Bayesian Face Recognition +</td><td>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>88a898592b4c1dfd707f04f09ca58ec769a257de</td><td>MobileFace: 3D Face Reconstruction +<br/>with Efficient CNN Regression +<br/>1 VisionLabs, Amsterdam, The Netherlands +<br/>2 Inria, WILLOW, Departement d’Informatique de l’Ecole Normale Superieure, PSL +<br/><b>Research University, ENS/INRIA/CNRS UMR 8548, Paris, France</b></td><td>('51318557', 'Nikolai Chinaev', 'nikolai chinaev')<br/>('2564281', 'Alexander Chigorin', 'alexander chigorin')<br/>('1785596', 'Ivan Laptev', 'ivan laptev')</td><td>{n.chinaev, a.chigorin}@visionlabs.ru +<br/>ivan.laptev@inria.fr +</td></tr><tr><td>88f7a3d6f0521803ca59fde45601e94c3a34a403</td><td>Semantic Aware Video Transcription +<br/>Using Random Forest Classifiers +<br/><b>University of Southern California, Institute for Robotics and Intelligent Systems</b><br/>Los Angeles, CA 90089, USA +</td><td>('1726241', 'Chen Sun', 'chen sun')</td><td></td></tr><tr><td>8812aef6bdac056b00525f0642702ecf8d57790b</td><td>A Unified Features Approach to Human Face Image +<br/>Analysis and Interpretation +<br/>Department of Informatics, +<br/>Technische Universit¨at M¨unchen +<br/>85748 Garching, Germany +</td><td>('1725709', 'Zahid Riaz', 'zahid riaz')<br/>('2110952', 'Suat Gedikli', 'suat gedikli')<br/>('1699132', 'Bernd Radig', 'bernd radig')</td><td>{riaz|gedikli|beetz|radig}@in.tum.de +</td></tr><tr><td>881066ec43bcf7476479a4146568414e419da804</td><td>From Traditional to Modern : Domain Adaptation for +<br/>Action Classification in Short Social Video Clips +<br/>Center for Visual Information Technology, IIIT Hyderabad, India +</td><td>('2461059', 'Aditya Singh', 'aditya singh')<br/>('3448416', 'Saurabh Saini', 'saurabh saini')<br/>('1962817', 'Rajvi Shah', 'rajvi shah')</td><td></td></tr><tr><td>8813368c6c14552539137aba2b6f8c55f561b75f</td><td>Trunk-Branch Ensemble Convolutional Neural +<br/>Networks for Video-based Face Recognition +</td><td>('37990555', 'Changxing Ding', 'changxing ding')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td></td></tr><tr><td>88e2574af83db7281c2064e5194c7d5dfa649846</td><td>Hindawi Publishing Corporation +<br/>Computational Intelligence and Neuroscience +<br/>Volume 2017, Article ID 4579398, 11 pages +<br/>http://dx.doi.org/10.1155/2017/4579398 +<br/>Research Article +<br/>A Robust Shape Reconstruction Method for Facial Feature +<br/>Point Detection +<br/><b>School of Automation Engineering, University of Electronic Science and Technology of China, No. 2006, Xiyuan Ave</b><br/>West Hi-Tech Zone, Chengdu 611731, China +<br/>Received 24 October 2016; Revised 18 January 2017; Accepted 30 January 2017; Published 19 February 2017 +<br/>Academic Editor: Ezequiel L´opez-Rubio +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Facial feature point detection has been receiving great research advances in recent years. Numerous methods have been developed +<br/>and applied in practical face analysis systems. However, it is still a quite challenging task because of the large variability in expression +<br/>and gestures and the existence of occlusions in real-world photo shoot. In this paper, we present a robust sparse reconstruction +<br/>method for the face alignment problems. Instead of a direct regression between the feature space and the shape space, the concept +<br/>of shape increment reconstruction is introduced. Moreover, a set of coupled overcomplete dictionaries termed the shape increment +<br/>dictionary and the local appearance dictionary are learned in a regressive manner to select robust features and fit shape increments. +<br/>Additionally, to make the learned model more generalized, we select the best matched parameter set through extensive validation +<br/>tests. Experimental results on three public datasets demonstrate that the proposed method achieves a better robustness over the +<br/>state-of-the-art methods. +<br/>1. Introduction +<br/>In most literatures, facial feature points are also referred to +<br/>facial landmarks or facial fiducial points. These points mainly +<br/>locate around edges or corners of facial components such as +<br/>eyebrows, eyes, mouth, nose, and jaw (see Figure 1). Existing +<br/>databases for method comparison are labeled with different +<br/>number of feature points, varying from the minimum 5-point +<br/>configuration [1] to the maximal 194-point configuration +<br/>[2]. Generally facial feature point detection is a supervised +<br/>or semisupervised learning process that trains model on +<br/>a large number of labeled facial images. It starts from a +<br/>face detection process and then predicts facial landmarks +<br/>inside the detected face bounding box. The localized facial +<br/>feature points can be utilized for various face analysis +<br/>tasks, for example, face recognition [3], facial animation +<br/>[4], facial expression detection [5], and head pose tracking +<br/>[6]. +<br/>In recent years, regression-based methods have gained +<br/>increasing attention for robust facial feature point detection. +<br/>Among these methods, a cascade framework is adopted to +<br/>recursively estimate the face shape 𝑆 of an input image, +<br/>which is the concatenation of facial feature point coordinates. +<br/>Beginning with an initial shape 𝑆(1), 𝑆 is updated by inferring +<br/>a shape increment Δ𝑆 from the previous shape: +<br/>Δ𝑆(𝑡) = 𝑊(𝑡)Φ(𝑡) (𝐼, 𝑆(𝑡)) , +<br/>(1) +<br/>where Δ𝑆(𝑡) and 𝑊(𝑡) are the shape increment and linear +<br/>regression matrix after 𝑡 iterations, respectively. As the input +<br/>variable of the mapping function Φ(𝑡), 𝐼 denotes the image +<br/>appearance and 𝑆(𝑡) denotes the corresponding face shape. +<br/>The regression goes to the next iteration by the additive +<br/>formula: +<br/>𝑆(𝑡) = 𝑆(𝑡−1) + Δ𝑆(𝑡−1). +<br/>(2) +<br/>In this paper, we propose a sparse reconstruction method +<br/>that embeds sparse coding in the reconstruction of shape +<br/>increment. As a very popular signal coding algorithm, sparse +<br/>coding has been recently successfully applied to the fields +<br/>of computer vision and machine learning, such as feature +<br/>selection and clustering analysis, image classification, and +<br/>face recognition [7–11]. In our method, sparse overcomplete +<br/>dictionaries are learned to encode various facial poses and +<br/>local textures considering the complex nature of imaging +</td><td>('9684590', 'Shuqiu Tan', 'shuqiu tan')<br/>('2915473', 'Dongyi Chen', 'dongyi chen')<br/>('9486108', 'Chenggang Guo', 'chenggang guo')<br/>('2122143', 'Zhiqi Huang', 'zhiqi huang')<br/>('9684590', 'Shuqiu Tan', 'shuqiu tan')</td><td>Correspondence should be addressed to Shuqiu Tan; tanshuqiu123136@hotmail.com and Dongyi Chen; dychen@uestc.edu.cn +</td></tr><tr><td>88bef50410cea3c749c61ed68808fcff84840c37</td><td>Sparse Representations of Image Gradient Orientations for Visual Recognition +<br/>and Tracking +<br/><b>Imperial College London</b><br/><b>EEMCS, University of Twente</b><br/>180 Queen’s Gate, London SW7 2AZ, U.K. +<br/>Drienerlolaan 5, 7522 NB Enschede, +<br/>The Netherlands +</td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{gt204,s.zafeiriou,m.pantic}@imperial.ac.uk +<br/>PanticM@cs.utwente.nl +</td></tr><tr><td>883006c0f76cf348a5f8339bfcb649a3e46e2690</td><td>Weakly Supervised Pain Localization using Multiple Instance Learning +</td><td>('39707211', 'Karan Sikka', 'karan sikka')<br/>('1735697', 'Abhinav Dhall', 'abhinav dhall')</td><td></td></tr><tr><td>88850b73449973a34fefe491f8836293fc208580</td><td>www.ijaret.org Vol. 2, Issue I, Jan. 2014 +<br/> ISSN 2320-6802 +<br/>INTERNATIONAL JOURNAL FOR ADVANCE RESEARCH IN +<br/>ENGINEERING AND TECHNOLOGY +<br/>WINGS TO YOUR THOUGHTS….. +<br/>XBeats-An Emotion Based Music Player +<br/>1U.G. Student, Dept. of Computer Engineering, +<br/><b>D.J. Sanghvi College of Engineering</b><br/>Vile Parle (W), Mumbai-400056. +<br/>2 U.G. Student, Dept. of Computer Engineering, +<br/><b>D.J. Sanghvi College of Engineering</b><br/>Vile Parle (W), Mumbai-400056. +<br/>3 U.G. Student, Dept. of Computer Engineering, +<br/><b>D.J. Sanghvi College of Engineering</b><br/>Vile Parle (W), Mumbai-400056. +<br/>4 Assistant Professor, Dept. of Computer Engineering, +<br/><b>D.J. Sanghvi College of Engineering</b><br/>Vile Parle (W), Mumbai-400056. +</td><td>('40770722', 'Sayali Chavan', 'sayali chavan')<br/>('2122358', 'Dipali Bhatt', 'dipali bhatt')</td><td>sayalichavan17@gmail.com +<br/>ekta.malkan27@yahoo.in +<br/>dipupb1392@gmail.com +<br/>prakashparanjape2012@gmail.com +</td></tr><tr><td>8820d1d3fa73cde623662d92ecf2e3faf1e3f328</td><td>Continuous Video to Simple Signals for Swimming Stroke Detection with +<br/>Convolutional Neural Networks +<br/><b>La Trobe University, Australia</b><br/><b>Australian Institute of Sport</b></td><td>('38689120', 'Brandon Victor', 'brandon victor')<br/>('1787185', 'Zhen He', 'zhen he')<br/>('31548192', 'Stuart Morgan', 'stuart morgan')<br/>('2874225', 'Dino Miniutti', 'dino miniutti')</td><td>{b.victor,z.he,s.morgan}@latrobe.edu.au +<br/>Dino.Miniutti@ausport.gov.au +</td></tr><tr><td>88f2952535df5859c8f60026f08b71976f8e19ec</td><td>A neural network framework for face +<br/>recognition by elastic bunch graph matching +</td><td>('37048377', 'Francisco A. Pujol López', 'francisco a. pujol lópez')<br/>('3144590', 'Higinio Mora Mora', 'higinio mora mora')<br/>('2260459', 'José A. Girona Selva', 'josé a. girona selva')</td><td></td></tr><tr><td>8818b12aa0ff3bf0b20f9caa250395cbea0e8769</td><td>Fashion Conversation Data on Instagram +<br/>∗Graduate School of Culture Technology, KAIST, South Korea +<br/>†Department of Communication Studies, UCLA, USA +</td><td>('3459091', 'Yu-i Ha', 'yu-i ha')<br/>('2399803', 'Sejeong Kwon', 'sejeong kwon')<br/>('1775511', 'Meeyoung Cha', 'meeyoung cha')<br/>('1834047', 'Jungseock Joo', 'jungseock joo')</td><td></td></tr><tr><td>8862a573a42bbaedd392e9e634c1ccbfd177a01d</td><td>3D Face Tracking and Texture Fusion in the Wild +<br/>Centre for Vision, Speech and Signal Processing +<br/>Image Understanding and Interactive Robotics +<br/><b>University of Surrey</b><br/>Guildford, GU2 7XH, United Kingdom +<br/>Contact: http://www.patrikhuber.ch +<br/><b>Reutlingen University</b><br/>D-72762 Reutlingen, Germany +</td><td>('39976184', 'Patrik Huber', 'patrik huber')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('16764402', 'Philipp Kopp', 'philipp kopp')</td><td></td></tr><tr><td>887b7676a4efde616d13f38fcbfe322a791d1413</td><td>Deep Temporal Appearance-Geometry Network +<br/>for Facial Expression Recognition +<br/><b>Korea Advanced Institute of Science and Technology</b><br/><b>Electronics and Telecommunications Research Institute</b></td><td>('8271137', 'Injae Lee', 'injae lee')<br/>('1769295', 'Junmo Kim', 'junmo kim')<br/>('1800903', 'Heechul Jung', 'heechul jung')</td><td>{heechul, haeng, sunny0414, junmo.kim}@kaist.ac.kr†, {ninja, hyun}@etri.re.kr‡ +</td></tr><tr><td>8878871ec2763f912102eeaff4b5a2febfc22fbe</td><td>3781 +<br/>Human Action Recognition in Unconstrained +<br/>Videos by Explicit Motion Modeling +</td><td>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('9227981', 'Qi Dai', 'qi dai')<br/>('39059457', 'Wei Liu', 'wei liu')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')<br/>('1751681', 'Chong-Wah Ngo', 'chong-wah ngo')</td><td></td></tr><tr><td>8855d6161d7e5b35f6c59e15b94db9fa5bbf2912</td><td>COGNITION IN PREGNANCY AND THE POSTPARTUM PERIOD +</td><td></td><td></td></tr><tr><td>8895d6ae9f095a8413f663cc83f5b7634b3dc805</td><td>BEHL ET AL: INCREMENTAL TUBE CONSTRUCTION FOR HUMAN ACTION DETECTION 1 +<br/>Incremental Tube Construction for Human +<br/>Action Detection +<br/>Harkirat Singh Behl1 +<br/>1 Department of Engineering Science +<br/><b>University of Oxford</b><br/>Oxford, UK +<br/>2 Think Tank Team +<br/>Samsung Research America +<br/>Mountain View, CA +<br/>3 Dept. of Computing and +<br/>Communication Technologies +<br/><b>Oxford Brookes University</b><br/>Oxford, UK +<br/>(a) Illustrative results on a video sequence from the LIRIS-HARL dataset [23]. Two people enter a room +<br/>Figure 1: +<br/>and put/take an object from a box (frame 150). They then shake hands (frame 175) and start having a discussion +<br/>(frame 350). In frame 450, another person enters the room, shakes hands, and then joins the discussion. Each +<br/>action tube instance is numbered and coloured according to its action category. We selected this video to show that +<br/>our tube construction algorithm can handle very complex situations in which multiple distinct action categories +<br/>occur in sequence and at concurrent times. (b) Action tubes drawn as viewed from above, compared to (c) the +<br/>ground truth action tubes. +</td><td>('3019396', 'Michael Sapienza', 'michael sapienza')<br/>('1931660', 'Gurkirt Singh', 'gurkirt singh')<br/>('49348905', 'Suman Saha', 'suman saha')<br/>('1754181', 'Fabio Cuzzolin', 'fabio cuzzolin')<br/>('1730268', 'Philip H. S. Torr', 'philip h. s. torr')</td><td>harkirat@robots.ox.ac.uk +<br/>m.sapienza@samsung.com +<br/>gurkirt.singh-2015@brookes.ac.uk +<br/>suman.saha-2014@brookes.ac.uk +<br/>fabio.cuzzolin@brookes.ac.uk +<br/>phst@robots.ox.ac.uk +</td></tr><tr><td>88bee9733e96958444dc9e6bef191baba4fa6efa</td><td>Extending Face Identification to +<br/>Open-Set Face Recognition +<br/>Department of Computer Science +<br/>Universidade Federal de Minas Gerais +<br/>Belo Horizonte, Brazil +</td><td>('2823797', 'Cassio E. dos Santos', 'cassio e. dos santos')<br/>('1679142', 'William Robson Schwartz', 'william robson schwartz')</td><td>{cass,william}@dcc.ufmg.br +</td></tr><tr><td>88fd4d1d0f4014f2b2e343c83d8c7e46d198cc79</td><td>978-1-4799-9988-0/16/$31.00 ©2016 IEEE +<br/>2697 +<br/>ICASSP 2016 +</td><td></td><td></td></tr><tr><td>887745c282edf9af40d38425d5fdc9b3fe139c08</td><td>FAME: +<br/>Face Association through Model Evolution +<br/><b>Bilkent University</b><br/>06800 Ankara/Turkey +<br/>Pinar Duygulu +<br/><b>Bilkent University</b><br/>06800 Ankara/Turkey +</td><td>('2540074', 'Eren Golge', 'eren golge')</td><td>eren.golge@bilkent.edu.tr +<br/>pinar.duygulu@gmail.com +</td></tr><tr><td>9f6d04ce617d24c8001a9a31f11a594bd6fe3510</td><td>Personality and Individual Differences 52 (2012) 61–66 +<br/>Contents lists available at SciVerse ScienceDirect +<br/>Personality and Individual Differences +<br/>j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / p a i d +<br/>Attentional bias towards angry faces in trait-reappraisal +<br/><b>1E1 WC Mackenzie Health Sciences Centre, University of Alberta, Edmonton, AB, Canada T6G 2R</b><br/>a r t i c l e +<br/>i n f o +<br/>a b s t r a c t +<br/>Article history: +<br/>Received 31 May 2011 +<br/>Received in revised form 26 August 2011 +<br/>Accepted 31 August 2011 +<br/>Available online 2 October 2011 +<br/>Keywords: +<br/>Trait emotion regulation +<br/>Reappraisal +<br/>Attention +<br/>Individual differences +<br/>Dot-probe +<br/>Emotion regulation (ER) strategies differ in when and how they influence emotion experience, expres- +<br/>sion, and concomitant cognition. However, no study to date has directly compared cognition in individ- +<br/>uals who have a clear disposition for either cognitive or behavioural ER strategies. The present study +<br/>compared selective attention to angry faces in groups of high trait-suppressors (people who are hiding +<br/>emotional reactions in response to emotional challenge) and high trait-reappraisers (people who cogni- +<br/>tively reinterpret emotional events). Since reappraisers are also low trait-anxious and suppressors are +<br/>high trait-anxious, high and low anxious control groups, both being low in trait-ER, were also included. +<br/>Attention to angry faces was assessed using an emotional dot-probe task. Trait-reappraisers and high- +<br/>anxious individuals both showed attentional biases towards angry faces. Trait-reappraisers’ vigilance +<br/>for angry faces was significantly more pronounced compared to both trait-suppressors and low anxious +<br/>controls. We suggest that threat prioritization in high trait-reappraisal may allow deeper cognitive pro- +<br/>cessing of threat information without being associated with psychological maladjustment. +<br/>Ó 2011 Elsevier Ltd. All rights reserved. +<br/>1. Introduction +<br/>An extensive literature suggests that cognition is influenced by +<br/>the emotional connotation of to-be-processed information. Emo- +<br/>tional events, especially negative emotional events, orient, attract +<br/>and/or capture attention more so than neutral events. Evidence +<br/>comes from studies using the emotional dot-probe paradigm +<br/>(MacLeod & Mathews, 1988). This task measures selective atten- +<br/>tion biases towards or away from emotional relative to neutral +<br/>stimuli (see Methods for details). Several person variables influ- +<br/>ence such biases. For example, high trait anxious individuals are +<br/>more likely than low trait anxious individuals to show an atten- +<br/>tional bias towards threatening stimuli (Frewen, Dozois, Joanisse, +<br/>& Neufeld, 2008). Interestingly, trait anxiety seems to modify the +<br/>ability to disengage attentional resources from the location of a +<br/>threatening stimulus more so than the speed of orienting attention +<br/>toward the stimulus location. For example, Fox, Russo, Bowles, and +<br/>Dutton (2001) found that high anxious, but not low anxious indi- +<br/>viduals responded slower to a dot-probe when an angry face, as +<br/>opposed to a happy or a neutral face, appeared in a different screen +<br/>location just prior. However, high anxious participants were not +<br/>faster to respond to the dot-probe when it followed in the same +<br/>location as the angry faces compared to happy or neutral faces +<br/>(attentional orienting). Hence, trait anxiety seems associated with +<br/><b>University of</b><br/><b>Calgary, 2500 University Dr., N.W. Calgary, AB, Canada T2N 1N4. Tel</b><br/>4667; fax: +1 403 282 8249. +<br/>0191-8869/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. +<br/>doi:10.1016/j.paid.2011.08.030 +<br/>a tendency to dwell on (i.e., difficulty in disengaging attention), +<br/>rather than to quickly orient toward, threatening stimuli such as +<br/>angry facial expressions. +<br/>information, +<br/>Although it is relatively well-established that individual differ- +<br/>ences in trait emotionality (i.e., high versus low trait anxiety) influ- +<br/>ence attentional processing of emotional +<br/>little is +<br/>known about how attentional biases may interact with a person’s +<br/>attempt to modulate their emotional responses. Recent findings +<br/>in emotion regulation (ER) suggest that emotion regulative strate- +<br/>gies differ in their consequences for the emotional response and +<br/>concomitant cognition. To date, most studies of ER have compared +<br/>cognitive and behavioural forms of ER, with the two most com- +<br/>monly studied ER strategies being cognitive reappraisal and +<br/>expressive suppression (Gross, 1998; Richards & Gross, 2000). +<br/>According to Gross (1998), reappraisal involves cognitively chang- +<br/>ing our appraisal of the emotional meaning of a stimulus in order +<br/>to render it less emotional, and in so doing, down-regulating our +<br/>own emotional response. In contrast, suppression involves the +<br/>behavioural inhibition of overt reactions to emotional experiences +<br/>(e.g., frowning) without changing the evaluation of the emotional +<br/>stimulus itself. +<br/>1.1. Instructed emotion regulation +<br/>To examine the consequences of ER, researchers have tradition- +<br/>ally exposed participants to an emotion-eliciting stimulus with an +<br/>instruction to use a specific ER strategy to down-regulate (or more +<br/>rarely, up-regulate) the resulting emotion. Because participants are +</td><td>('6027810', 'Jody E. Arndt', 'jody e. arndt')<br/>('2726268', 'Esther Fujiwara', 'esther fujiwara')</td><td>E-mail address: jearndt@ucalgary.ca (J.E. Arndt). +</td></tr><tr><td>9f499948121abb47b31ca904030243e924585d5f</td><td>Hierarchical Attention Network for Action +<br/>Recognition in Videos +<br/><b>Arizona State University</b><br/><b>Arizona State University</b><br/>Yahoo Research +<br/>Neil O’Hare +<br/>Yahoo Research +<br/>Yahoo Research +<br/><b>Arizona State University</b></td><td>('33513248', 'Yilin Wang', 'yilin wang')<br/>('2893721', 'Suhang Wang', 'suhang wang')<br/>('1736632', 'Jiliang Tang', 'jiliang tang')<br/>('1787097', 'Yi Chang', 'yi chang')<br/>('2913552', 'Baoxin Li', 'baoxin li')</td><td>ywang370@asu.edu +<br/>suhang.wang@asu.edu +<br/>jlt@yahoo-inc.com +<br/>nohare@yahoo-inc.com +<br/>yichang@yahoo-inc.com +<br/>baoxin.li@asu.edu +</td></tr><tr><td>9fc04a13eef99851136eadff52e98eb9caac919d</td><td>Rethinking the Camera Pipeline for Computer Vision +<br/><b>Cornell University</b><br/><b>Carnegie Mellon University</b><br/><b>Cornell University</b></td><td>('2328520', 'Mark Buckler', 'mark buckler')<br/>('39131476', 'Suren Jayasuriya', 'suren jayasuriya')<br/>('2138184', 'Adrian Sampson', 'adrian sampson')</td><td>mab598@cornell.edu +<br/>sjayasur@andrew.cmu.edu +<br/>asampson@cs.cornell.edu +</td></tr><tr><td>9f4078773c8ea3f37951bf617dbce1d4b3795839</td><td>Leveraging Inexpensive Supervision Signals +<br/>for Visual Learning +<br/>Technical Report Number: +<br/>CMU-RI-TR-17-13 +<br/>a dissertation presented +<br/>by +<br/>to +<br/><b>The Robotics Institute</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>Master of Science +<br/>in the subject of +<br/>Robotics +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, Pennsylvania +<br/>May 2017 +<br/>All rights reserved. +</td><td>('3234247', 'Senthil Purushwalkam', 'senthil purushwalkam')<br/>('3234247', 'Senthil Purushwalkam', 'senthil purushwalkam')</td><td></td></tr><tr><td>9f65319b8a33c8ec11da2f034731d928bf92e29d</td><td>TAKING ROLL: A PIPELINE FOR FACE RECOGNITION +<br/>Dip. di Scienze Teoriche e Applicate +<br/><b>University of Insubria</b><br/>21100, Varese, Italy +<br/><b>Louisiana State University</b><br/>2222 Business Education Complex South, +<br/>LA, 70803, USA +</td><td>('39149650', 'I. Gallo', 'i. gallo')<br/>('1876793', 'S. Nawaz', 's. nawaz')<br/>('3457883', 'A. Calefati', 'a. calefati')<br/>('2398301', 'G. Piccoli', 'g. piccoli')</td><td></td></tr><tr><td>9fa1be81d31fba07a1bde0275b9d35c528f4d0b8</td><td>Identifying Persons by Pictorial and +<br/>Contextual Cues +<br/>Nicholas Leonard Pi¨el +<br/>Thesis submitted for the degree of Master of Science +<br/>Supervisor: +<br/>April 2009 +</td><td>('1695527', 'Theo Gevers', 'theo gevers')</td><td></td></tr><tr><td>9f094341bea610a10346f072bf865cb550a1f1c1</td><td>Recognition and Volume Estimation of Food Intake using a Mobile Device +<br/>Sarnoff Corporation +<br/>201 Washington Rd, +<br/>Princeton, NJ, 08540 +</td><td>('1981308', 'Manika Puri', 'manika puri')</td><td>{mpuri, zzhu, qyu, adivakaran, hsawhney}@sarnoff.com +</td></tr><tr><td>9fdfe1695adac2380f99d3d5cb6879f0ac7f2bfd</td><td>EURASIP Journal on Applied Signal Processing 2005:13, 2091–2100 +<br/>c(cid:1) 2005 Hindawi Publishing Corporation +<br/>Spatio-Temporal Graphical-Model-Based +<br/>Multiple Facial Feature Tracking +<br/>Congyong Su +<br/><b>College of Computer Science, Zhejiang University, Hangzhou 310027, China</b><br/>Li Huang +<br/><b>College of Computer Science, Zhejiang University, Hangzhou 310027, China</b><br/>Received 1 January 2004; Revised 20 February 2005 +<br/>It is challenging to track multiple facial features simultaneously when rich expressions are presented on a face. We propose a two- +<br/>step solution. In the first step, several independent condensation-style particle filters are utilized to track each facial feature in the +<br/>temporal domain. Particle filters are very effective for visual tracking problems; however multiple independent trackers ignore +<br/>the spatial constraints and the natural relationships among facial features. In the second step, we use Bayesian inference—belief +<br/>propagation—to infer each facial feature’s contour in the spatial domain, in which we learn the relationships among contours of +<br/>facial features beforehand with the help of a large facial expression database. The experimental results show that our algorithm +<br/>can robustly track multiple facial features simultaneously, while there are large interframe motions with expression changes. +<br/>Keywords and phrases: facial feature tracking, particle filter, belief propagation, graphical model. +<br/>1. +<br/>INTRODUCTION +<br/>Multiple facial feature tracking is very important in the com- +<br/>puter vision field: it needs to be carried out before video- +<br/>based facial expression analysis and expression cloning. Mul- +<br/>tiple facial feature tracking is also very challenging be- +<br/>cause there are plentiful nonrigid motions in facial fea- +<br/>tures besides rigid motions in faces. Nonrigid facial fea- +<br/>ture motions are usually very rapid and often form dense +<br/>clutter by facial features themselves. Only using traditional +<br/>Kalman filter is inadequate because it is based on Gaus- +<br/>sian density, and works relatively poorly in clutter, which +<br/>causes the density for facial feature’s contour to be multi- +<br/>modal and therefore non-Gaussian. Isard and Blake [1] firstly +<br/>proposed a face tracker by particle filters—condensation— +<br/>which is more effective in clutter than comparable Kalman +<br/>filter. +<br/>Although particle filters are often very effective for visual +<br/>tracking problems, they are specialized to temporal problems +<br/>whose corresponding graphs are simple Markov chains (see +<br/>Figure 1). There is often structure within each time instant +<br/>that is ignored by particle filters. For example, in multiple +<br/>facial feature tracking, the expressions of each facial feature +<br/>(such as eyes, brows, lips) are closely related; therefore a more +<br/>complex graph should be formulated. +<br/>The contribution of this paper is extending particle filters +<br/>to track multiple facial features simultaneously. The straight- +<br/>forward approach of tracking each facial feature by one in- +<br/>dependent particle filter is questionable, because influences +<br/>and actions among facial features are not taken into account. +<br/>In this paper, we propose a spatio-temporal graphical +<br/>model for multiple facial feature tracking (see Figure 2). Here +<br/>the graphical model is not a 2D or a 3D facial mesh model. +<br/>In the spatial domain, the model is shown in Figure 3, where +<br/>xi is a hidden random variable and yi is a noisy local ob- +<br/>servation. Nonparametric belief propagation is used to infer +<br/>facial feature’s interrelationships in a part-based face model, +<br/>allowing positions and states of some features in clutter to +<br/>be recovered. Facial structure is also taken into account, be- +<br/>cause facial features have spatial position constraints [2]. In +<br/>the temporal domain, every facial feature forms a Markov +<br/>chain (see Figure 1). +<br/>After briefly reviewing related work in Section 2, we +<br/>introduce the details of our algorithm in Sections 3 and +<br/>4. Many convincing experimental results are shown in +<br/>Section 5. Conclusions are given in Section 6. +<br/>2. RELATED WORK +<br/>After the pioneering work of Isard and Blake [1] who +<br/>creatively used particle filters for visual tracking, many +</td><td></td><td>Email: su@cs.zju.edu.cn +<br/>Email: lihuang@cs.zju.edu.cn +</td></tr><tr><td>6b333b2c6311e36c2bde920ab5813f8cfcf2b67b</td><td></td><td></td><td></td></tr><tr><td>6b3e360b80268fda4e37ff39b7f303e3684e8719</td><td>FACE RECOGNITION FROM SKETCHES USING ADVANCED +<br/>CORRELATION FILTERS USING HYBRID EIGENANALYSIS +<br/>FOR FACE SYNTHESIS +<br/><b>Language Technology Institute, Carnegie Mellon Universty</b><br/><b>Carnegie Mellon University</b><br/>Keywords: +<br/>Face from sketch synthesis, face recognition, eigenface, advanced correlation filters, OTSDF. +</td><td>('3036546', 'Yung-hui Li', 'yung-hui li')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td></td></tr><tr><td>6b9aa288ce7740ec5ce9826c66d059ddcfd8dba9</td><td></td><td></td><td></td></tr><tr><td>6bcfcc4a0af2bf2729b5bc38f500cfaab2e653f0</td><td>Facial expression recognition in the wild using improved dense trajectories and +<br/>Fisher vector encoding +<br/><b>Computational Science and Engineering Program, Bo gazic i University, Istanbul, Turkey</b><br/><b>Bo gazic i University, Istanbul, Turkey</b></td><td>('2471932', 'Sadaf Afshar', 'sadaf afshar')<br/>('1764521', 'Albert Ali Salah', 'albert ali salah')</td><td>{sadaf.afshar, salah}@boun.edu.tr +</td></tr><tr><td>6bca0d1f46b0f7546ad4846e89b6b842d538ee4e</td><td>FACE RECOGNITION FROM SURVEILLANCE-QUALITY VIDEO +<br/>A Dissertation +<br/>Submitted to the Graduate School +<br/><b>of the University of Notre Dame</b><br/>in Partial Fulfillment of the Requirements +<br/>for the Degree of +<br/>Doctor of Philosophy +<br/>by +<br/>Patrick J. Flynn, Co-Director +<br/>Graduate Program in Computer Science and Engineering +<br/>Notre Dame, Indiana +<br/>July 2010 +</td><td>('30042752', 'Deborah Thomas', 'deborah thomas')<br/>('1799014', 'Kevin W. Bowyer', 'kevin w. bowyer')</td><td></td></tr><tr><td>6b089627a4ea24bff193611e68390d1a4c3b3644</td><td>CROSS-POLLINATION OF NORMALISATION +<br/>TECHNIQUES FROM SPEAKER TO FACE +<br/>AUTHENTICATION USING GAUSSIAN +<br/>MIXTURE MODELS +<br/>Idiap-RR-03-2012 +<br/>JANUARY 2012 +<br/>Centre du Parc, Rue Marconi 19, P.O. Box 592, CH - 1920 Martigny +</td><td>('1843477', 'Roy Wallace', 'roy wallace')<br/>('1698382', 'Sébastien Marcel', 'sébastien marcel')</td><td>T +41 27 721 77 11 F +41 27 721 77 12 info@idiap.ch www.idiap.ch +</td></tr><tr><td>6b8d0569fffce5cc221560d459d6aa10c4db2f03</td><td>Interlinked Convolutional Neural Networks for +<br/>Face Parsing +<br/>State Key Laboratory of Intelligent Technology and Systems +<br/>Tsinghua National Laboratory for Information Science and Technology (TNList) +<br/>Department of Computer Science and Technology +<br/><b>Tsinghua University, Beijing 100084, China</b></td><td>('1879713', 'Yisu Zhou', 'yisu zhou')<br/>('1705418', 'Xiaolin Hu', 'xiaolin hu')<br/>('49846744', 'Bo Zhang', 'bo zhang')</td><td></td></tr><tr><td>6be0ab66c31023762e26d309a4a9d0096f72a7f0</td><td>Enhance Visual Recognition under Adverse +<br/>Conditions via Deep Networks +</td><td>('1771885', 'Ding Liu', 'ding liu')<br/>('2392101', 'Bowen Cheng', 'bowen cheng')<br/>('2969311', 'Zhangyang Wang', 'zhangyang wang')<br/>('40479011', 'Haichao Zhang', 'haichao zhang')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td></td></tr><tr><td>6bcee7dba5ed67b3f9926d2ae49f9a54dee64643</td><td>Assessment of Time Dependency in Face Recognition: +<br/>An Initial Study +<br/>IDept of Computer Science and Engineering +<br/><b>University of Notre Dame. Notre Dame, IN 46556.USA</b><br/><b>Nqtional Institute of Standards and Technology</b><br/>100 Bureau Dr.• Stop 8940, Gaithersburg, MD 20899 USA +<br/>Performance +<br/>and products +<br/>research matures +<br/>factors of strong practical +<br/>the performance of such syslemsis +<br/>are +</td><td>('1704876', 'Patrick J. Flynn', 'patrick j. flynn')<br/>('1799014', 'Kevin W. Bowyer', 'kevin w. bowyer')<br/>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')</td><td>{flynn,kwb}@nd.edu +<br/>jonathon@nist.gov +</td></tr><tr><td>6b18628cc8829c3bf851ea3ee3bcff8543391819</td><td>Face recognition based on subset selection via metric learning on manifold. +<br/>1058. [doi:10.1631/FITEE.1500085] +<br/>Face recognition based on subset +<br/>selection via metric learning on manifold +<br/>Key words: Face recognition, Sparse representation, Manifold structure, +<br/>Metric learning, Subset selection +<br/> ORCID: http://orcid.org/0000-0001-7441-4749 +<br/>Front Inform Technol & Electron Eng </td><td>('2684160', 'Hong Shao', 'hong shao')<br/>('1752664', 'Shuang Chen', 'shuang chen')<br/>('1941366', 'Wen-cheng Cui', 'wen-cheng cui')<br/>('1752664', 'Shuang Chen', 'shuang chen')</td><td>E-mail: chenshuang19891129@gmail.com +</td></tr><tr><td>6b7f7817b2e5a7e7d409af2254a903fc0d6e02b6</td><td>April 13, 2009 +<br/>14:13 WSPC/INSTRUCTION FILE +<br/>International Journal of Pattern Recognition and Artificial Intelligence +<br/>c(cid:13) World Scientific Publishing Company +<br/>Feature extraction through cross - phase congruency for facial +<br/>expression analysis +<br/>Electronics Department +<br/>Faculty of Electrical Engineering and Information Technology +<br/><b>University of Oradea</b><br/>410087, Universitatii 1, +<br/>Romania +<br/>http://webhost.uoradea.ro/ibuciu +<br/>Electronics and Communications Faculty +<br/><b>Politehnica University of Timisoara</b><br/>Bd. Vasile Parvan, no.2 +<br/>300223 Timisoara +<br/>Romania +<br/>http://hermes.etc.upt.ro +<br/>Human face analysis has attracted a large number of researchers from various fields, +<br/>such as computer vision, image processing, neurophysiology or psychology. One of the +<br/>particular aspects of human face analysis is encompassed by facial expression recognition +<br/>task. A novel method based on phase congruency for extracting the facial features used +<br/>in the facial expression classification procedure is developed. Considering a set of image +<br/>samples comprising humans expressing various expressions, this new approach computes +<br/>the phase congruency map between the samples. The analysis is performed in the fre- +<br/>quency space where the similarity (or dissimilarity) between sample phases is measured +<br/>to form discriminant features. The experiments were run using samples from two facial +<br/>expression databases. To assess the method’s performance, the technique is compared to +<br/>the state-of-the art techniques utilized for classifying facial expressions, such as Principal +<br/>Component Analysis (PCA), Independent Component Analysis (ICA), Linear Discrim- +<br/>inant Analysis (LDA), and Gabor jets. The features extracted by the aforementioned +<br/>techniques are further classified using two classifiers: a distance-based classifier and a +<br/>Support Vector Machine - based classifier. Experiments reveal superior facial expression +<br/>recognition performance for the proposed approach with respect to other techniques. +<br/>Keywords: feature extraction; phase congruency; facial expression analysis. +<br/>1. Feature Extraction for Facial Expression Recognition +<br/>Facial expression analysis is a concern of several disciplinary scientific fields, such +<br/>as computer vision, image processing, neurophysiology and psychology. The large +<br/>interest for this analysis is motivated by an impressive area of applications. These +</td><td>('2336758', 'Ioan Buciu', 'ioan buciu')<br/>('2526319', 'Ioan Nafornita', 'ioan nafornita')</td><td>ibuciu@uoradea.ro +<br/>ioan.nafornita@etc.upt.ro +</td></tr><tr><td>6b1b43d58faed7b457b1d4e8c16f5f7e7d819239</td><td></td><td></td><td></td></tr><tr><td>6bb0425baac448297fbd29a00e9c9b9926ce8870</td><td>INTERNATIONAL CONFERENCE ON COMMUNICATION, COMPUTER AND POWER (ICCCP’09) +<br/>MUSCAT, FEBRUARY 15-18, 2009 +<br/>Facial Expression Recognition Using Log-Gabor +<br/>Filters and Local Binary Pattern Operators +<br/><b>School of Electrical and Computer Engineering, RMIT University, Melbourne, Australia</b></td><td>('1857490', 'Seyed Mehdi Lajevardi', 'seyed mehdi lajevardi')<br/>('1749220', 'Zahir M. Hussain', 'zahir m. hussain')</td><td>seyed.lajevardi@rmit.edu.au, zmhussain@ieee.org +</td></tr><tr><td>6b35b15ceba2f26cf949f23347ec95bbbf7bed64</td><td></td><td></td><td></td></tr><tr><td>6b6493551017819a3d1f12bbf922a8a8c8cc2a03</td><td>Pose Normalization for Local Appearance-Based +<br/>Face Recognition +<br/>Computer Science Department, Universit¨at Karlsruhe (TH) +<br/>Am Fasanengarten 5, Karlsruhe 76131, Germany +<br/>http://isl.ira.uka.de/cvhci +</td><td>('1697965', 'Hua Gao', 'hua gao')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>{hua.gao,ekenel,stiefel}@ira.uka.de +</td></tr><tr><td>6b17b219bd1a718b5cd63427032d93c603fcf24f</td><td><b>Carnegie Mellon University</b><br/><b>Language Technologies Institute</b><br/>School of Computer Science +<br/>10-1-2016 +<br/>Videos from the 2013 Boston Marathon: An Event +<br/>Reconstruction Dataset for Synchronization and +<br/>Localization +<br/><b>Carnegie Mellon University</b><br/><b>Carnegie Mellon University</b><br/><b>Carnegie Mellon University</b><br/><b>Carnegie Mellon University</b><br/>Follow this and additional works at: http://repository.cmu.edu/lti +<br/>Part of the Computer Sciences Commons +</td><td>('3175807', 'Jia Chen', 'jia chen')<br/>('1915796', 'Junwei Liang', 'junwei liang')<br/>('2075232', 'Han Lu', 'han lu')<br/>('2927024', 'Shoou-I Yu', 'shoou-i yu')<br/>('7661726', 'Alexander Hauptmann', 'alexander hauptmann')</td><td>Research Showcase @ CMU +<br/>Carnegie Mellon University, alex@cs.cmu.edu +<br/>This Technical Report is brought to you for free and open access by the School of Computer Science at Research Showcase @ CMU. It has been +<br/>accepted for inclusion in Language Technologies Institute by an authorized administrator of Research Showcase @ CMU. For more information, please +<br/>contact research-showcase@andrew.cmu.edu. +</td></tr><tr><td>6bb630dfa797168e6627d972560c3d438f71ea99</td><td></td><td></td><td></td></tr><tr><td>6b6ff9d55e1df06f8b3e6f257e23557a73b2df96</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 61– No.17, January 2013 +<br/>Survey of Threats to the Biometric Authentication +<br/>Systems and Solutions +<br/>Research Scholor,Mewar +<br/><b>University, Chitorgarh. (INDIA</b><br/>P.C.Gupta +<br/><b>Kota University, Kota(INDIA</b><br/>Khushboo Mantri +<br/><b>M.tech.student, Arya College of</b><br/>engineering ,Jaipur(INDIA) +</td><td>('2875951', 'Sarika Khandelwal', 'sarika khandelwal')</td><td></td></tr><tr><td>07377c375ac76a34331c660fe87ebd7f9b3d74c4</td><td>Detailed Human Avatars from Monocular Video +<br/>1Computer Graphics Lab, TU Braunschweig, Germany +<br/><b>Max Planck Institute for Informatics, Saarland Informatics Campus, Germany</b><br/>Figure 1: Our method creates a detailed avatar from a monocular video of a person turning around. Based on the SMPL +<br/>model, we first compute a medium-level avatar, then add subject-specific details and finally generate a seamless texture. +</td><td>('1914886', 'Thiemo Alldieck', 'thiemo alldieck')<br/>('9765909', 'Weipeng Xu', 'weipeng xu')</td><td>{alldieck,magnor}@cg.cs.tu-bs.de {wxu,theobalt,gpons}@mpi-inf.mpg.de +</td></tr><tr><td>0729628db4bb99f1f70dd6cb2353d7b76a9fce47</td><td>Separating Pose and Expression in Face Images: +<br/>A Manifold Learning Approach +<br/><b>University of Pennsylvania</b><br/>Moore Bldg, 200 South 33rd St, Philadelphia, PA 19104, USA +<br/>(Submitted on December 27, 2006) +</td><td>('1732066', 'Daniel D. Lee', 'daniel d. lee')</td><td>E-mail: {jhham,ddlee}@seas.upenn.edu +</td></tr><tr><td>0728f788107122d76dfafa4fb0c45c20dcf523ca</td><td>The Best of Both Worlds: Combining Data-independent and Data-driven +<br/>Approaches for Action Recognition +</td><td>('1711953', 'Dezhong Yao', 'dezhong yao')<br/>('2735055', 'Ming Lin', 'ming lin')<br/>('2927024', 'Shoou-I Yu', 'shoou-i yu')</td><td>{lanzhzh, minglin, iyu, alex@cs.cmu.edu}, dyao@hust.edu.cn +</td></tr><tr><td>07c90e85ac0f74b977babe245dea0f0abcf177e3</td><td>Appeared in the 4th International Conference on Audio- and Video-Based +<br/>Biometric Person Authentication, pp 10{18, June 9 - 11, 2003, Guildford, UK +<br/>An Image Preprocessing Algorithm for +<br/>Illumination Invariant Face Recognition +<br/><b>The Robotics Institute, Carnegie Mellon University</b><br/>5000 Forbes Avenue, Pittsburgh, PA 15213 +</td><td>('33731953', 'Ralph Gross', 'ralph gross')<br/>('2407094', 'Vladimir Brajovic', 'vladimir brajovic')</td><td>frgross,brajovicg@cs.cmu.edu +</td></tr><tr><td>07ea3dd22d1ecc013b6649c9846d67f2bf697008</td><td>HUMAN-CENTRIC VIDEO UNDERSTANDING WITH WEAK +<br/>SUPERVISION +<br/>A DISSERTATION +<br/>SUBMITTED TO THE DEPARTMENT OF COMPUTER SCIENCE +<br/>AND THE COMMITTEE ON GRADUATE STUDIES +<br/><b>OF STANFORD UNIVERSITY</b><br/>IN PARTIAL FULFILLMENT OF THE REQUIREMENTS +<br/>FOR THE DEGREE OF +<br/>DOCTOR OF PHILOSOPHY +<br/>June 2016 +</td><td>('34066479', 'Vignesh Ramanathan', 'vignesh ramanathan')</td><td></td></tr><tr><td>071099a4c3eed464388c8d1bff7b0538c7322422</td><td>FACIAL EXPRESSION RECOGNITION IN THE WILD USING RICH DEEP FEATURES +<br/>Microsoft Advanced Technology labs, Microsoft Technology and Research, Cairo, Egypt +<br/> +</td><td>('34828041', 'Abubakrelsedik Karali', 'abubakrelsedik karali')<br/>('2376438', 'Ahmad Bassiouny', 'ahmad bassiouny')<br/>('3144122', 'Motaz El-Saban', 'motaz el-saban')</td><td></td></tr><tr><td>07fcbae86f7a3ad3ea1cf95178459ee9eaf77cb1</td><td>Large Scale Unconstrained Open Set Face Database +<br/><b>University of Colorado at Colorado Springs</b><br/>2Securics Inc, Colorado Springs +</td><td>('27469806', 'Archana Sapkota', 'archana sapkota')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')</td><td>asapkota@vast.uccs.edu +<br/>tboult@vast.uccs.edu +</td></tr><tr><td>076d3fc800d882445c11b9af466c3af7d2afc64f</td><td>FACE ATTRIBUTE CLASSIFICATION USING ATTRIBUTE-AWARE CORRELATION MAP +<br/>AND GATED CONVOLUTIONAL NEURAL NETWORKS +<br/><b>Korea Advanced institute of Science and Technology</b><br/>Department of Electrical Engineering +<br/>291 Daehak-ro, Yuseong-gu, Daejeon, Korea +</td><td>('3315036', 'Sunghun Kang', 'sunghun kang')<br/>('2350325', 'Donghoon Lee', 'donghoon lee')</td><td></td></tr><tr><td>071af21377cc76d5c05100a745fb13cb2e40500f</td><td></td><td></td><td></td></tr><tr><td>070ab604c3ced2c23cce2259043446c5ee342fd6</td><td>AnActiveIlluminationandAppearance(AIA)ModelforFaceAlignment +<br/>FatihKahraman,MuhittinGokmen +<br/><b>IstanbulTechnicalUniversity</b><br/>ComputerScienceDept.,Turkey +<br/>InformaticsandMathematicalModelling,Denmark +<br/>SuneDarkner,RasmusLarsen +<br/><b>TechnicalUniversityofDenmark</b></td><td></td><td>{fkahraman, gokmen}@itu.edu.tr +<br/>{sda,rl}@imm.dtu.dk +</td></tr><tr><td>071135dfb342bff884ddb9a4d8af0e70055c22a1</td><td>New Architecture and Transfer Learning for Video Classification +<br/>Temporal 3D ConvNets: +<br/><b>ESAT-PSI, KU Leuven, 2University of Bonn, 3CV:HCI, KIT, Karlsruhe, 4Sensifai</b></td><td>('3310120', 'Ali Diba', 'ali diba')<br/>('3169187', 'Mohsen Fayyaz', 'mohsen fayyaz')<br/>('38035876', 'Vivek Sharma', 'vivek sharma')<br/>('31493847', 'Amir Hossein Karami', 'amir hossein karami')<br/>('2713759', 'Mohammad Mahdi Arzani', 'mohammad mahdi arzani')<br/>('9456273', 'Rahman Yousefzadeh', 'rahman yousefzadeh')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{firstname.lastname}@esat.kuleuven.be, {lastname}@sensifai.com, +<br/>fayyaz@iai.uni-bonn.de, vivek.sharma@kit.edu +</td></tr><tr><td>0754e769eb613fd3968b6e267a301728f52358be</td><td>Towards a Watson That Sees: Language-Guided Action Recognition for +<br/>Robots +</td><td>('7607499', 'Yezhou Yang', 'yezhou yang')<br/>('1697493', 'Yiannis Aloimonos', 'yiannis aloimonos')</td><td></td></tr><tr><td>07c83f544d0604e6bab5d741b0bf9a3621d133da</td><td>Learning Spatio-Temporal Features with 3D Residual Networks +<br/>for Action Recognition +<br/><b>National Institute of Advanced Industrial Science and Technology (AIST</b><br/>Tsukuba, Ibaraki, Japan +</td><td>('2199251', 'Kensho Hara', 'kensho hara')<br/>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')<br/>('1732705', 'Yutaka Satoh', 'yutaka satoh')</td><td>{kensho.hara, hirokatsu.kataoka, yu.satou}@aist.go.jp +</td></tr><tr><td>0773c320713dae62848fceac5a0ac346ba224eca</td><td>Digital Facial Augmentation for Interactive +<br/>Entertainment +<br/>Centre for Intelligent Machines +<br/><b>McGill University</b><br/>Montreal, Quebec, Canada +</td><td>('2726121', 'Naoto Hieda', 'naoto hieda')<br/>('2242019', 'Jeremy R. Cooperstock', 'jeremy r. cooperstock')</td><td>Email: {nhieda, jer}@cim.mcgill.ca +</td></tr><tr><td>070de852bc6eb275d7ca3a9cdde8f6be8795d1a3</td><td>A FACS Valid 3D Dynamic Action Unit Database with Applications to 3D +<br/>Dynamic Morphable Facial Modeling +<br/>Department of Computer Science +<br/>School of Humanities and Social Sciences +<br/><b>University of Bath</b><br/><b>Jacobs University</b><br/>Centre for Vision, Speech and Signal Processing +<br/><b>University of Surrey</b></td><td>('1792288', 'Darren Cosker', 'darren cosker')<br/>('2035177', 'Eva Krumhuber', 'eva krumhuber')<br/>('1695085', 'Adrian Hilton', 'adrian hilton')</td><td>dpc@cs.bath.ac.uk +<br/>e.krumhuber@jacobs-university.de +<br/>a.hilton@surrey.ac.uk +</td></tr><tr><td>07a472ea4b5a28b93678a2dcf89028b086e481a2</td><td>Head Dynamic Analysis: A Multi-view +<br/>Framework +<br/><b>University of California, San Diego, USA</b></td><td>('1947383', 'Ashish Tawari', 'ashish tawari')</td><td>{atawari,mtrivedi}@ucsd.edu +</td></tr><tr><td>0717b47ab84b848de37dbefd81cf8bf512b544ac</td><td>International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 +<br/>International Conference on Humming Bird ( 01st March 2014) +<br/>RESEARCH ARTICLE +<br/> OPEN ACCESS +<br/>Robust Face Recognition and Tagging in Visual Surveillance +<br/>System +</td><td>('21008397', 'Kavitha MS', 'kavitha ms')<br/>('39546266', 'Siva Pradeepa', 'siva pradeepa')<br/>('21008397', 'Kavitha MS', 'kavitha ms')<br/>('39546266', 'Siva Pradeepa', 'siva pradeepa')</td><td>e-mail:kavithams999@gmail.com +</td></tr><tr><td>07fa153b8e6196ee6ef6efd8b743de8485a07453</td><td>Action Prediction from Videos via Memorizing Hard-to-Predict Samples +<br/><b>Northeastern University, Boston, MA, USA</b><br/><b>College of Engineering, Northeastern University, Boston, MA, USA</b><br/><b>College of Computer and Information Science, Northeastern University, Boston, MA, USA</b></td><td>('48901920', 'Yu Kong', 'yu kong')<br/>('9355577', 'Shangqian Gao', 'shangqian gao')<br/>('47935056', 'Bin Sun', 'bin sun')<br/>('1708679', 'Yun Fu', 'yun fu')</td><td>{yukong,yunfu}@ece.neu.edu, {gao.sh,sun.bi}@husky.neu.edu +</td></tr><tr><td>0708059e3bedbea1cbfae1c8cd6b7259d4b56b5b</td><td>Graph-regularized Multi-class Support Vector +<br/>Machines for Face and Action Recognition +<br/><b>Tampere University of Technology, Tampere, Finland</b></td><td>('9219875', 'Moncef Gabbouj', 'moncef gabbouj')</td><td>Email: {alexandros.iosifidis,moncef.gabbouj}@tut.fi +</td></tr><tr><td>074af31bd9caa61fea3c4216731420bd7c08b96a</td><td>Face Verification Using Sparse Representations +<br/><b>Institute for Advanced Computer Studies, University of Maryland, College Park, MD</b><br/><b>TNLIST, Tsinghua University, Beijing, 100084, China</b></td><td>('2723427', 'Huimin Guo', 'huimin guo')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('3826759', 'Jonghyun Choi', 'jonghyun choi')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td>{hmguo, jhchoi, lsd}@umiacs.umd.edu, rpwang@tsinghua.edu.cn +</td></tr><tr><td>0750a816858b601c0dbf4cfb68066ae7e788f05d</td><td>CosFace: Large Margin Cosine Loss for Deep Face Recognition +<br/>Tencent AI Lab +</td><td>('39049654', 'Hao Wang', 'hao wang')<br/>('1996677', 'Yitong Wang', 'yitong wang')<br/>('48741267', 'Zheng Zhou', 'zheng zhou')<br/>('3478009', 'Xing Ji', 'xing ji')<br/>('2856494', 'Dihong Gong', 'dihong gong')<br/>('2263912', 'Jingchao Zhou', 'jingchao zhou')<br/>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('46641573', 'Wei Liu', 'wei liu')</td><td>{hawelwang,yitongwang,encorezhou,denisji,sagazhou,michaelzfli}@tencent.com +<br/>gongdihong@gmail.com wliu@ee.columbia.edu +</td></tr><tr><td>078d507703fc0ac4bf8ca758be101e75ea286c80</td><td> ISSN: 2321-8169 +<br/>International Journal on Recent and Innovation Trends in Computing and Communication +<br/>Volume: 3 Issue: 8 +<br/> 5287 - 5296 +<br/> ________________________________________________________________________________________________________________________________ +<br/>Large- Scale Content Based Face Image Retrieval using Attribute Enhanced +<br/>Sparse Codewords. +<br/>Chaitra R, +<br/>Mtech Digital Coomunication Engineering +<br/><b>Acharya Institute Of Technology</b><br/>Bangalore +</td><td></td><td></td></tr><tr><td>0716e1ad868f5f446b1c367721418ffadfcf0519</td><td>Interactively Guiding Semi-Supervised +<br/>Clustering via Attribute-Based Explanations +<br/>Virginia Tech, Blacksburg, VA, USA +</td><td>('9276834', 'Shrenik Lad', 'shrenik lad')<br/>('1713589', 'Devi Parikh', 'devi parikh')</td><td></td></tr><tr><td>073eaa49ccde15b62425cda1d9feab0fea03a842</td><td></td><td></td><td></td></tr><tr><td>07f31bef7a7035792e3791473b3c58d03928abbf</td><td>Lessons from Collecting a Million +<br/>Biometric Samples +<br/><b>University of Notre Dame</b><br/><b>National Institute of Standards and Technology</b></td><td>('1704876', 'Patrick J. Flynn', 'patrick j. flynn')<br/>('1799014', 'Kevin W. Bowyer', 'kevin w. bowyer')<br/>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')</td><td></td></tr><tr><td>0726a45eb129eed88915aa5a86df2af16a09bcc1</td><td>Introspective Perception: Learning to Predict Failures in Vision Systems +</td><td>('2739544', 'Shreyansh Daftry', 'shreyansh daftry')<br/>('3308210', 'Sam Zeng', 'sam zeng')<br/>('1756566', 'J. Andrew Bagnell', 'j. andrew bagnell')<br/>('1709305', 'Martial Hebert', 'martial hebert')</td><td></td></tr><tr><td>07de8371ad4901356145722aa29abaeafd0986b9</td><td>April 13, 2017 +<br/>DRAFT +<br/>Towards Usable Multimedia Event Detection +<br/>February, 2017 +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>Thesis Committee: +<br/>Alexander G. Hauptmann (Chair) +<br/>Submitted in partial fulfillment of the requirements +<br/>for the degree of Doctor of Philosophy. +</td><td>('34692532', 'Zhenzhong Lan', 'zhenzhong lan')<br/>('1880336', 'Bhiksha Raj Ramakrishnan', 'bhiksha raj ramakrishnan')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')<br/>('14517812', 'Leonid Sigal', 'leonid sigal')<br/>('34692532', 'Zhenzhong Lan', 'zhenzhong lan')</td><td></td></tr><tr><td>07e639abf1621ceff27c9e3f548fadfa2052c912</td><td>RESEARCH ARTICLE +<br/>5-HTTLPR Expression Outside the Skin: An +<br/>Experimental Test of the Emotional +<br/>Reactivity Hypothesis in Children +<br/><b>Utrecht Centre for Child and Adolescent Studies, Utrecht University, Utrecht, The Netherlands</b><br/><b>Research Institute of Child Development and Education, University of Amsterdam, Utrecht, The</b><br/><b>Netherlands, Utrecht University, Utrecht, The Netherlands</b><br/><b>Current Address: Research Institute of Child Development and Education, University of Amsterdam</b><br/>Amsterdam,The Netherlands +</td><td>('4594074', 'Joyce Weeland', 'joyce weeland')<br/>('6811600', 'Meike Slagt', 'meike slagt')<br/>('5859538', 'Eddie Brummelman', 'eddie brummelman')<br/>('3935697', 'Walter Matthys', 'walter matthys')<br/>('4441681', 'Geertjan Overbeek', 'geertjan overbeek')</td><td>* j.weeland@uva.nl +</td></tr><tr><td>07da958db2e561cc7c24e334b543d49084dd1809</td><td>Dictionary Learning Based Dimensionality +<br/>Reduction for Classification +<br/>Karin Schnass and Pierre Vandergheynst +<br/><b>Signal Processing Institute</b><br/><b>Swiss Federal Institute of Technology</b><br/>Lausanne, Switzerland +<br/>EPFL-STI-ITS-LTS2 +<br/>CH-1015 Lausanne +<br/>Tel: +41 21 693 2657 +<br/>Fax: +41 21 693 7600 +<br/>EDICS: SPC-CODC +</td><td></td><td>{karin.schnass, pierre.vandergheynst}@epfl.ch +</td></tr><tr><td>0742d051caebf8a5d452c03c5d55dfb02f84baab</td><td>Real-Time Geometric Motion Blur for a Deforming Polygonal Mesh +<br/>Nathan Jones +<br/><b>Formerly: Texas AandM University</b><br/>Currently: The Software Group +</td><td></td><td>nathan.jones@tylertechnologies.com +</td></tr><tr><td>07d986b1005593eda1aeb3b1d24078db864f8f6a</td><td>International Journal of Industrial Electronics and Electrical Engineering, ISSN: 2347-6982 +<br/>Volume-3, Issue-11, Nov.-2015 +<br/>FACIAL EXPRESSION RECOGNITION USING LOCAL FACIAL +<br/>FEATURES +<br/><b>National University of Kaohsiung, 811 Kaohsiung, Taiwan</b><br/><b>National University of Kaohsiung, 811 Kaohsiung, Taiwan</b><br/><b>National Sun Yat Sen University, 804 Kaohsiung, Taiwan</b><br/>followed by +<br/>communications +<br/>[1]. Automatic +</td><td></td><td>E-mail: abc3329797@gmail.com, {cclai, johnw, stpan}@nuk.edu.tw, leesj@mail.ee.nsysu.edu.tw +</td></tr><tr><td>38d56ddcea01ce99902dd75ad162213cbe4eaab7</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>2648 +</td><td></td><td></td></tr><tr><td>389334e9a0d84bc54bcd5b94b4ce4c5d9d6a2f26</td><td>FACIAL PARAMETER EXTRACTION SYSTEM BASED ON ACTIVE CONTOURS +<br/>Universitat Politècnica de Catalunya, Barcelona, Spain +</td><td>('1767549', 'Montse Pardàs', 'montse pardàs')<br/>('1820469', 'Marcos Losada', 'marcos losada')</td><td></td></tr><tr><td>38f7f3c72e582e116f6f079ec9ae738894785b96</td><td>IJARCCE +<br/>ISSN (Online) 2278-1021 +<br/>ISSN (Print) 2319 5940 +<br/>International Journal of Advanced Research in Computer and Communication Engineering +<br/>Vol. 4, Issue 11, November 2015 +<br/>A New Technique for Face Matching after +<br/>Plastic Surgery in Forensics +<br/><b>Student, Amal Jyothi College of Engineering, Kanjirappally, India</b><br/><b>Amal Jyothi College of Engineering, Kanjirappally, India</b><br/>I. INTRODUCTION +<br/>Facial recognition is one of the most important task that +<br/>forensic examiners execute +<br/>their +<br/>investigation. This work focuses on analysing the effect of +<br/>plastic surgery in face recognition algorithms. It is +<br/>imperative for the subsequent facial recognition systems to +<br/>be capable of addressing this significant issue and +<br/>accordingly there is a need for more research in this +<br/>important area. +</td><td>('32764403', 'Anju Joseph', 'anju joseph')<br/>('16501589', 'Nilu Tressa Thomas', 'nilu tressa thomas')<br/>('40864737', 'Neethu C. Sekhar', 'neethu c. sekhar')</td><td></td></tr><tr><td>380dd0ddd5d69adc52defc095570d1c22952f5cc</td><td></td><td></td><td></td></tr><tr><td>38679355d4cfea3a791005f211aa16e76b2eaa8d</td><td>Title +<br/>Evolutionary cross-domain discriminative Hessian Eigenmaps +<br/>Author(s) +<br/>Si, S; Tao, D; Chan, KP +<br/>Citation +<br/>1086 +<br/>Issued Date +<br/>2010 +<br/>URL +<br/>http://hdl.handle.net/10722/127357 +<br/>Rights +<br/>This work is licensed under a Creative Commons Attribution- +<br/>NonCommercial-NoDerivatives 4.0 International License.; ©2010 +<br/>IEEE. Personal use of this material is permitted. However, +<br/>permission to reprint/republish this material for advertising or +<br/>promotional purposes or for creating new collective works for +<br/>resale or redistribution to servers or lists, or to reuse any +<br/>copyrighted component of this work in other works must be +<br/>obtained from the IEEE. +</td><td></td><td></td></tr><tr><td>3802c97f925cb03bac91d9db13d8b777dfd29dcc</td><td>Non-Parametric Bayesian Constrained Local Models +<br/><b>Institute of Systems and Robotics, University of Coimbra, Portugal</b></td><td>('39458914', 'Pedro Martins', 'pedro martins')<br/>('2117944', 'Rui Caseiro', 'rui caseiro')<br/>('1678231', 'Jorge Batista', 'jorge batista')</td><td>{pedromartins,ruicaseiro,batista}@isr.uc.pt +</td></tr><tr><td>38a2661b6b995a3c4d69e7d5160b7596f89ce0e6</td><td>Randomized Intraclass-Distance Minimizing Binary Codes for Face Recognition +<br/><b>Colorado State University</b><br/>Fort Collins, CO 80523 +<br/><b>National Institute of Standards and Technology</b></td><td>('40370804', 'Hao Zhang', 'hao zhang')<br/>('1757322', 'J. Ross Beveridge', 'j. ross beveridge')<br/>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')</td><td>{zhangh, ross, qmo, draper}@cs.colostate.edu +<br/>jonathon.phillips@nist.gov +</td></tr><tr><td>38682c7b19831e5d4f58e9bce9716f9c2c29c4e7</td><td>International Journal of Computer Trends and Technology (IJCTT) – Volume 18 Number 5 – Dec 2014 +<br/>Movie Character Identification Using Graph Matching +<br/>Algorithm +<br/>M.Tech Scholar, Dept of CSE, QISCET, ONGOLE, Dist: Prakasam, AP, India. +<br/>Associate Professor, Department of CSE, QISCET, ONGOLE, Dist: Prakasam, AP, India +</td><td></td><td></td></tr><tr><td>38787338ba659f0bfbeba11ec5b7748ffdbb1c3d</td><td>EVALUATION OF THE DISCRIMINATION POWER OF FEATURES EXTRACTED +<br/>FROM 2-D AND 3-D FACIAL IMAGES FOR FACIAL EXPRESSION ANALYSIS +<br/><b>University of Piraeus</b><br/>Karaoli & Dimitriou 80, Piraeus 185 34 +<br/>GREECE +</td><td>('2828175', 'Ioanna-Ourania Stathopoulou', 'ioanna-ourania stathopoulou')<br/>('1802584', 'George A. Tsihrintzis', 'george a. tsihrintzis')</td><td>phone: + 30 210 4142322, fax: + 30 210 4142264, email: {iostath, geoatsi}@unipi.gr +</td></tr><tr><td>3803b91e784922a2dacd6a18f61b3100629df932</td><td>Temporal Multimodal Fusion +<br/>for Video Emotion Classification in the Wild +<br/>Orange Labs +<br/>Cesson-Sévigné, France +<br/>Orange Labs +<br/>Cesson-Sévigné, France +<br/>Normandie Univ., UNICAEN, +<br/>ENSICAEN, CNRS +<br/>Caen, France +</td><td>('26339425', 'Valentin Vielzeuf', 'valentin vielzeuf')<br/>('2642628', 'Stéphane Pateux', 'stéphane pateux')<br/>('1801809', 'Frédéric Jurie', 'frédéric jurie')</td><td>valentin.vielzeuf@orange.com +<br/>stephane.pateux@orange.com +<br/>frederic.jurie@unicaen.fr +</td></tr><tr><td>38eea307445a39ee7902c1ecf8cea7e3dcb7c0e7</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Multi-distance Support Matrix Machine +<br/>Received: date / Accepted: date +</td><td>('34679353', 'Yunfei Ye', 'yunfei ye')<br/>('49405675', 'Dong Han', 'dong han')</td><td></td></tr><tr><td>38c901a58244be9a2644d486f9a1284dc0edbf8a</td><td>Multi-Camera Action Dataset for Cross-Camera Action Recognition +<br/>Benchmarking +<br/><b>School of Electronic Information Engineering, Tianjin University, China</b><br/><b>Interactive and Digital Media Institute, National University of Singapore, Singapore</b><br/><b>School of Computing, National University of Singapore, Singapore</b></td><td>('1803305', 'Wenhui Li', 'wenhui li')<br/>('3026404', 'Yongkang Wong', 'yongkang wong')<br/>('1678662', 'Yang Li', 'yang li')</td><td></td></tr><tr><td>385750bcf95036c808d63db0e0b14768463ff4c6</td><td></td><td></td><td></td></tr><tr><td>3852968082a16db8be19b4cb04fb44820ae823d4</td><td>Unsupervised Learning of Long-Term Motion Dynamics for Videos +<br/><b>Stanford University</b></td><td>('3378742', 'Zelun Luo', 'zelun luo')<br/>('3378457', 'Boya Peng', 'boya peng')<br/>('38485317', 'De-An Huang', 'de-an huang')<br/>('3304525', 'Alexandre Alahi', 'alexandre alahi')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td>{zelunluo,boya,dahuang,alahi,feifeili}@cs.stanford.edu +</td></tr><tr><td>38cc2f1c13420170c7adac30f9dfac69b297fb76</td><td><b>Rochester Institute of Technology</b><br/>RIT Scholar Works +<br/>Theses +<br/>7-1-2009 +<br/>Thesis/Dissertation Collections +<br/>Recognition of human activities and expressions in +<br/>video sequences using shape context descriptor +<br/>Follow this and additional works at: http://scholarworks.rit.edu/theses +<br/>Recommended Citation +<br/>Kholgade, Natasha Prashant, "Recognition of human activities and expressions in video sequences using shape context descriptor" +<br/><b>Thesis. Rochester Institute of Technology. Accessed from</b><br/>This Thesis is brought to you for free and open access by the Thesis/Dissertation Collections at RIT Scholar Works. It has been accepted for inclusion +</td><td>('2201569', 'Natasha Prashant Kholgade', 'natasha prashant kholgade')</td><td>in Theses by an authorized administrator of RIT Scholar Works. For more information, please contact ritscholarworks@rit.edu. +</td></tr><tr><td>38cbb500823057613494bacd0078aa0e57b30af8</td><td>2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops +<br/>Deep Face Deblurring +<br/><b>Imperial College London</b><br/><b>Imperial College London</b></td><td>('34586458', 'Grigorios G. Chrysos', 'grigorios g. chrysos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>g.chrysos@imperial.ac.uk +<br/>s.zafeiriou@imperial.ac.uk +</td></tr><tr><td>384f972c81c52fe36849600728865ea50a0c4670</td><td>1 +<br/>Multi-Fold Gabor, PCA and ICA Filter +<br/>Convolution Descriptor for Face Recognition +<br/> +</td><td>('1801904', 'Andrew Beng Jin Teoh', 'andrew beng jin teoh')<br/>('3326176', 'Cong Jie Ng', 'cong jie ng')</td><td></td></tr><tr><td>38f1fac3ed0fd054e009515e7bbc72cdd4cf801a</td><td>Finding Person Relations in Image Data of the +<br/>Internet Archive +<br/>Eric M¨uller-Budack1,2[0000−0002−6802−1241], +<br/>1 Leibniz Information Centre for Science and Technology (TIB), Hannover, Germany +<br/><b>L3S Research Center, Leibniz Universit at Hannover, Germany</b></td><td>('51008013', 'Kader Pustu-Iren', 'kader pustu-iren')<br/>('50983345', 'Sebastian Diering', 'sebastian diering')<br/>('1738703', 'Ralph Ewerth', 'ralph ewerth')</td><td></td></tr><tr><td>38f06a75eb0519ae1d4582a86ef4730cc8fb8d7f</td><td>Shrinkage Expansion Adaptive Metric Learning +<br/>1 School of Information and Communications Engineering, +<br/><b>Dalian University of Technology, China</b><br/><b>School of Computer Science and Technology, Harbin Institute of Technology, China</b><br/><b>Hong Kong Polytechnic University, Hong Kong</b></td><td>('2769011', 'Qilong Wang', 'qilong wang')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('36685537', 'Lei Zhang', 'lei zhang')<br/>('40426020', 'Peihua Li', 'peihua li')</td><td>{csqlwang,cswmzuo}@gmail.com, cslzhang@comp.polyu.edu.hk, +<br/>peihuali@dlut.edu.cn +</td></tr><tr><td>380d5138cadccc9b5b91c707ba0a9220b0f39271</td><td>Deep Imbalanced Learning for Face Recognition +<br/>and Attribute Prediction +</td><td>('2000034', 'Chen Huang', 'chen huang')<br/>('47002704', 'Yining Li', 'yining li')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>384945abd53f6a6af51faf254ba8ef0f0fb3f338</td><td>Visual Recognition with Humans in the Loop +<br/><b>University of California, San Diego</b><br/><b>California Institute of Technology</b></td><td>('3251767', 'Steve Branson', 'steve branson')<br/>('2367820', 'Catherine Wah', 'catherine wah')<br/>('2490700', 'Boris Babenko', 'boris babenko')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>{sbranson,cwah,gschroff,bbabenko,sjb}@cs.ucsd.edu +<br/>{welinder,perona}@caltech.edu +</td></tr><tr><td>38215c283ce4bf2c8edd597ab21410f99dc9b094</td><td>The SEMAINE Database: Annotated Multimodal Records of +<br/>Emotionally Colored Conversations between a Person and a Limited +<br/>Agent +<br/>McKeown, G., Valstar, M., Cowie, R., Pantic, M., & Schröder, M. (2012). The SEMAINE Database: Annotated +<br/>Multimodal Records of Emotionally Colored Conversations between a Person and a Limited Agent. IEEE +<br/>Transactions on Affective Computing, 3(1), 5-17. DOI: 10.1109/T-AFFC.2011.20 +<br/>Published in: +<br/>Document Version: +<br/>Peer reviewed version +<br/><b>Queen's University Belfast - Research Portal</b><br/><b>Link to publication record in Queen's University Belfast Research Portal</b><br/>General rights +<br/><b>Copyright for the publications made accessible via the Queen's University Belfast Research Portal is retained by the author(s) and / or other</b><br/>copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated +<br/>with these rights. +<br/>Take down policy +<br/>The Research Portal is Queen's institutional repository that provides access to Queen's research output. Every effort has been made to +<br/>ensure that content in the Research Portal does not infringe any person's rights, or applicable UK laws. If you discover content in the +<br/>Download date:05. Nov. 2018 +</td><td></td><td>Research Portal that you believe breaches copyright or violates any law, please contact openaccess@qub.ac.uk. +</td></tr><tr><td>38861d0d3a0292c1f54153b303b0d791cbba1d50</td><td></td><td></td><td></td></tr><tr><td>38d8ff137ff753f04689e6b76119a44588e143f3</td><td>When 3D-Aided 2D Face Recognition Meets Deep Learning: +<br/>An extended UR2D for Pose-Invariant Face Recognition +<br/>Computational Biomedicine Lab +<br/><b>University of Houston</b><br/>4800 Calhoun Rd. Houston, TX, USA +</td><td>('5084124', 'Xiang Xu', 'xiang xu')<br/>('39634395', 'Pengfei Dou', 'pengfei dou')<br/>('26401746', 'Ha A. Le', 'ha a. le')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td></td></tr><tr><td>3896c62af5b65d7ba9e52f87505841341bb3e8df</td><td>Face Recognition from Still Images and Video +<br/>Department of Electrical and Computer Engineering +<br/>Center for Automation Research +<br/><b>University of Maryland, College Park</b><br/>Related concepts Biometric identification, verification. +<br/>Definition Face recognition is concerned with identifying or verifying one or more persons from still +<br/>images or video sequences using a stored database of faces. +<br/>Background The earliest work on face recognition started as early as 1950’s in psychology and in the +<br/>1960’s in engineering, but research on automatic face recognition practically started in the 1970’s after the +<br/>seminal work of Kanade [1] and Kelly [2]. +<br/>Application Face recognition has wide range of applications in many different areas ranging from +<br/>law enforcement and surveillance, information security to human-computer interaction, virtual reality and +<br/>computer entertainment. +<br/>1 Introduction +<br/>Face recognition with its wide range of commercial and law enforcement applications has been one of the +<br/>most active areas of research in the field of computer vision and pattern recognition. Personal identification +<br/>systems based on faces have the advantage that facial images can be obtained from a distance without requir- +<br/>ing cooperation of the subject, as compared to other biometrics such as fingerprint, iris, etc. Face recognition +<br/>is concerned with identifying or verifying one or more persons from still images or video sequences using +<br/>a stored database of faces. Depending on the particular application, there can be different scenarios, rang- +<br/>ing from controlled still images to uncontrolled videos. Since face recognition is essentially the problem of +<br/>recognizing a 3D object from its 2D image or a video sequence, it has to deal with significant appearance +<br/>changes due to illumination and pose variations. Current algorithms perform well in controlled scenarios, +<br/>but their performance is far from satisfactory in uncontrolled scenarios. Most of the current research in this +<br/>area is focused toward recognizing faces in uncontrolled scenarios. This chapter is broadly divided into two +<br/>sections. The first section discusses the approaches proposed for recognizing faces from still images and the +<br/>second section deals with face recognition from video sequences. +<br/>2 Still image face recognition +<br/>This section discusses some of the early subspace and feature-based approaches, followed by those which +<br/>address the problem of appearance change due to illumination variations and approaches that can handle both +<br/>illumination and pose variations. +<br/>2.1 Early approaches +<br/>Among the early subspace-based holistic approaches, eigenfaces [3] and Fisherfaces [4][5] have proved to be +<br/>very effective for the task of face recognition. Since human faces have similar overall configuration, the facial +<br/>images can be described by a relatively low-dimensional subspace. Principal Component Analysis (PCA) [3] +<br/>has been used for finding those vectors which can best account for the distribution of facial images within the +<br/>whole image space. These vectors are eigenvectors of the covariance matrix computed from the aligned face +<br/>images in the training set and are thus known as ’eigenfaces’. Given the eigenfaces, every face in the gallery +<br/>database is represented as a vector of weights obtained by projecting the image onto the eigenfaces using +</td><td>('2642508', 'Soma Biswas', 'soma biswas')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>38192a0f9261d9727b119e294a65f2e25f72d7e6</td><td></td><td></td><td></td></tr><tr><td>38bbca5f94d4494494860c5fe8ca8862dcf9676e</td><td>Probabilistic, Features-based Object Recognition +<br/>Thesis by +<br/>In Partial Ful(cid:2)llment of the Requirements +<br/>for the Degree of +<br/>Doctor of Philosophy +<br/><b>California Institute of Technology</b><br/>Pasadena, California +<br/>2008 +<br/>(Defended October 12, 2007) +</td><td>('2462051', 'Pierre Moreels', 'pierre moreels')</td><td></td></tr><tr><td>38183fe28add21693729ddeaf3c8a90a2d5caea3</td><td>Scale-Aware Face Detection +<br/><b>SenseTime, 2Tsinghua University</b></td><td>('19235216', 'Zekun Hao', 'zekun hao')<br/>('1715752', 'Yu Liu', 'yu liu')<br/>('2137185', 'Hongwei Qin', 'hongwei qin')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('2693308', 'Xiu Li', 'xiu li')<br/>('1705418', 'Xiaolin Hu', 'xiaolin hu')</td><td>{haozekun, yanjunjie}@outlook.com, liuyuisanai@gmail.com, +<br/>{qhw12@mails., xlhu@, li.xiu@sz.}tsinghua.edu.cn +</td></tr><tr><td>38a9ca2c49a77b540be52377784b9f734e0417e4</td><td>Face Verification using Large Feature Sets and One Shot Similarity +<br/>1Department of Computer Science +<br/><b>University of Maryland</b><br/><b>College Park, MD, 20740, USA</b><br/><b>Institute of Computing</b><br/><b>University of Campinas</b><br/>Campinas, SP, 13084-971, Brazil +</td><td>('2723427', 'Huimin Guo', 'huimin guo')<br/>('1679142', 'William Robson Schwartz', 'william robson schwartz')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td>hmguo@cs.umd.edu +<br/>schwartz@ic.unicamp.br +<br/>lsd@umiacs.umd.edu +</td></tr><tr><td>3802da31c6d33d71b839e260f4022ec4fbd88e2d</td><td>Deep Attributes for One-Shot Face Recognition +<br/><b>Xerox Research Center India</b><br/>3Department of Electrical Engineering, IIT Kanpur +</td><td>('5060928', 'Aishwarya Jadhav', 'aishwarya jadhav')<br/>('1744135', 'Vinay P. Namboodiri', 'vinay p. namboodiri')<br/>('1797662', 'K. S. Venkatesh', 'k. s. venkatesh')</td><td>aishwaryauj@gmail.com, vinaypn@iitk.ac.in, venkats@iitk.ac.in +</td></tr><tr><td>00fb2836068042c19b5197d0999e8e93b920eb9c</td><td></td><td></td><td></td></tr><tr><td>00f7f7b72a92939c36e2ef9be97397d8796ee07c</td><td>3D ConvNets with Optical Flow Based Regularization +<br/><b>Stanford University</b><br/>Stanford, CA +</td><td>('35627656', 'Kevin Chavez', 'kevin chavez')</td><td>kjchavez@stanford.edu +</td></tr><tr><td>0021f46bda27ea105d722d19690f5564f2b8869e</td><td>Deep Region and Multi-label Learning for Facial Action Unit Detection +<br/><b>School of Comm. and Info. Engineering, Beijing University of Posts and Telecom., Beijing China</b><br/><b>Robotics Institute, Carnegie Mellon University, USA</b></td><td>('2393320', 'Kaili Zhao', 'kaili zhao')</td><td></td></tr><tr><td>0081e2188c8f34fcea3e23c49fb3e17883b33551</td><td>Training Deep Face Recognition Systems +<br/>with Synthetic Data +<br/>Department of Mathematics and Computer Science +<br/><b>University of Basel</b></td><td>('2780587', 'Adam Kortylewski', 'adam kortylewski')<br/>('1801001', 'Andreas Schneider', 'andreas schneider')<br/>('3277377', 'Thomas Gerig', 'thomas gerig')<br/>('34460642', 'Bernhard Egger', 'bernhard egger')<br/>('31540387', 'Andreas Morel-Forster', 'andreas morel-forster')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td></td></tr><tr><td>00dc942f23f2d52ab8c8b76b6016d9deed8c468d</td><td>Advanced Correlation-Based Character Recognition Applied to +<br/>the Archimedes Palimpsest +<br/>by +<br/><b>B. S. Rochester Institute of Technology</b><br/>A dissertation submitted in partial fulfillment of the +<br/>requirements for the degree of Doctor of Philosophy +<br/>in the Chester F. Carlson Center for Imaging Science +<br/><b>Rochester Institute of Technology</b><br/>May 2008 +<br/>Signature of the Author +<br/>Accepted by +<br/>Coordinator, Ph.D. Degree Program +<br/>Date +</td><td>('31960835', 'Derek J. Walvoord', 'derek j. walvoord')</td><td></td></tr><tr><td>0077cd8f97cafd2b389783858a6e4ab7887b0b6b</td><td>MAI et al.: ON THE RECONSTRUCTION OF DEEP FACE TEMPLATES +<br/>On the Reconstruction of Deep Face Templates +</td><td>('3391550', 'Guangcan Mai', 'guangcan mai')<br/>('1684684', 'Kai Cao', 'kai cao')<br/>('1768574', 'Pong C. Yuen', 'pong c. yuen')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>0055c7f32fa6d4b1ad586d5211a7afb030ca08cc</td><td>SAHAet al.: DEEPLEARNINGFORDETECTINGSPACE-TIMEACTIONTUBES +<br/>Deep Learning for Detecting Multiple +<br/>Space-Time Action Tubes in Videos +<br/>1 Dept. of Computing and +<br/>Communication Technologies +<br/><b>Oxford Brookes University</b><br/>Oxford, UK +<br/>2 Department of Engineering Science +<br/><b>University of Oxford</b><br/>Oxford, UK +</td><td>('3017538', 'Suman Saha', 'suman saha')<br/>('1931660', 'Gurkirt Singh', 'gurkirt singh')<br/>('3019396', 'Michael Sapienza', 'michael sapienza')<br/>('1730268', 'Philip H. S. Torr', 'philip h. s. torr')<br/>('1754181', 'Fabio Cuzzolin', 'fabio cuzzolin')</td><td>suman.saha-2014@brookes.ac.uk +<br/>gurkirt.singh-2015@brookes.ac.uk +<br/>michael.sapienza@eng.ox.ac.uk +<br/>philip.torr@eng.ox.ac.uk +<br/>fabio.cuzzolin@brookes.ac.uk +</td></tr><tr><td>009cd18ff06ff91c8c9a08a91d2516b264eee48e</td><td>8 +<br/>Face and Automatic Target Recognition Based +<br/>on Super-Resolved Discriminant Subspace +<br/><b>Chulalongkorn University, Bangkok</b><br/>Thailand +<br/>1. Introduction +<br/>Recently, super-resolution reconstruction (SRR) method of low-dimensional face subspaces +<br/>has been proposed for face recognition. This face subspace, also known as eigenface, is +<br/>extracted using principal component analysis (PCA). One of the disadvantages of the +<br/>reconstructed features obtained from the super-resolution face subspace is that no class +<br/>information is included. To remedy the mentioned problem, at first, this chapter will be +<br/>discussed about two novel methods for super-resolution reconstruction of discriminative +<br/>features, i.e., class-specific and discriminant analysis of principal components; that aims on +<br/>improving the discriminant power of the recognition systems. Next, we discuss about two- +<br/>dimensional principal component analysis (2DPCA), also refered to as image PCA. We suggest +<br/>new reconstruction algorithm based on the replacement of PCA with 2DPCA in extracting +<br/>super-resolution subspace for face and automatic target recognition. Our experimental +<br/>results on Yale and ORL face databases are very encouraging. Furthermore, the performance +<br/>of our proposed approach on the MSTAR database is also tested. +<br/>In general, the fidelity of data, feature extraction, discriminant analysis, and classification +<br/>rule are four basic elements in face and target recognition systems. One of the efficacies of +<br/>recognition systems could be improved by enhancing the fidelity of the noisy, blurred, and +<br/>undersampled images that are captured by the surveillance imagers. Regarding to the +<br/>fidelity of data, when the resolution of the captured image is too small, the quality of the +<br/>detail information becomes too limited, leading to severely poor decisions in most of the +<br/>existing recognition systems. Having used super-resolution reconstruction algorithms (Park +<br/>et al., 2003), it is fortunately to learn that a high-resolution (HR) image can be reconstructed +<br/>from an undersampled image sequence obtained from the original scene with pixel +<br/>displacements among images. This HR image is then used to input to the recognition system +<br/>in order to improve the recognition performance. In fact, super-resolution can be considered +<br/>as the numerical and regularization study of the ill-conditioned large scale problem given to +<br/>describe the relationship between low-resolution (LR) and HR pixels (Nguyen et al., 2001). +<br/>On the one hand, feature extraction aims at reducing the dimensionality of face or target +<br/>image so that the extracted feature is as representative as possible. On the other hand, +<br/>super-resolution aims at visually increasing the dimensionality of face or target image. +<br/>Having applied super-resolution methods at pixel domain (Lin et al., 2005; Wagner et al., +<br/>2004), the performance of face and target recognition applicably increases. However, with +<br/>the emphases on improving computational complexity and robustness to registration error +<br/>www.intechopen.com +</td><td>('2874330', 'Widhyakorn Asdornwised', 'widhyakorn asdornwised')</td><td></td></tr><tr><td>00214fe1319113e6649435cae386019235474789</td><td>Bachelorarbeit im Fach Informatik +<br/>Face Recognition using +<br/>Distortion Models +<br/>Mathematik, Informatik und Naturwissenschaften der +<br/>RHEINISCH-WESTFÄLISCHEN TECHNISCHEN HOCHSCHULE AACHEN +<br/>Der Fakultät für +<br/>Lehrstuhl für Informatik VI +<br/>Prof. Dr.-Ing. H. Ney +<br/>vorgelegt von: +<br/>Matrikelnummer 252400 +<br/>Gutachter: +<br/>Prof. Dr.-Ing. H. Ney +<br/>Prof. Dr. B. Leibe +<br/>Betreuer: +<br/>September 2009 +</td><td>('1804963', 'Harald Hanselmann', 'harald hanselmann')<br/>('1967060', 'Philippe Dreuw', 'philippe dreuw')</td><td></td></tr><tr><td>004e3292885463f97a70e1f511dc476289451ed5</td><td>Quadruplet-wise Image Similarity Learning +<br/>Marc T. Law +<br/><b>LIP6, UPMC - Sorbonne University, Paris, France</b></td><td>('1728523', 'Nicolas Thome', 'nicolas thome')<br/>('1702233', 'Matthieu Cord', 'matthieu cord')</td><td>{Marc.Law, Nicolas.Thome, Matthieu.Cord}@lip6.fr +</td></tr><tr><td>0004f72a00096fa410b179ad12aa3a0d10fc853c</td><td></td><td></td><td></td></tr><tr><td>00b08d22abc85361e1c781d969a1b09b97bc7010</td><td>Who is the Hero? − Semi-Supervised Person Re-Identification in Videos +<br/><b>Tampere University of Technology, Tampere, Finland</b><br/><b>Nokia Research Center, Tampere, Finland</b><br/>Keywords: +<br/>Semi-supervised person re-identification, Important person detection, Face tracks, Clustering +</td><td>('13413642', 'Umar Iqbal', 'umar iqbal')<br/>('9219875', 'Moncef Gabbouj', 'moncef gabbouj')</td><td>{umar.iqbal, moncef.gabbouj}@tut.fi, igor.curcio@nokia.com +</td></tr><tr><td>007250c2dce81dd839a55f9108677b4f13f2640a</td><td>Advances in Component Based Face Detection +<br/>S. M. Bileschi +<br/>B. Heisele +<br/>Center for Biological And Computational Learning +<br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, MA. +<br/>Honda Research and Development +<br/>Boston, MA. +</td><td></td><td></td></tr><tr><td>00e3957212517a252258baef833833921dd308d4</td><td>Adaptively Weighted Multi-task Deep Network for Person +<br/>Aribute Classification +<br/><b>Shanghai Key Lab of Intelligent Information Processing, School of Computer Science, Fudan University, China</b><br/><b>School of Data Science, Fudan University, China</b></td><td>('37391748', 'Keke He', 'keke he')<br/>('11032846', 'Zhanxiong Wang', 'zhanxiong wang')<br/>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('6260277', 'Rui Feng', 'rui feng')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')</td><td>{kkhe15,15210240046,yanweifu,fengrui,ygj,xyxue}@fudan.edu.cn +</td></tr><tr><td>00f0ed04defec19b4843b5b16557d8d0ccc5bb42</td><td></td><td></td><td></td></tr><tr><td>0037bff7be6d463785d4e5b2671da664cd7ef746</td><td>Author manuscript, published in "European Conference on Computer Vision (ECCV '10) 6311 (2010) 634--647" +<br/> DOI : 10.1007/978-3-642-15549-9_46 +</td><td></td><td></td></tr><tr><td>009a18d04a5e3ec23f8ffcfc940402fd8ec9488f</td><td>BOYRAZ ET AL. : WEAKLY-SUPERVISED ACTION RECOGNITION BY LOCALIZATION +<br/>Action Recognition by Weakly-Supervised +<br/>Discriminative Region Localization +<br/>Marshall Tappen12 +<br/>1 Department of EECS +<br/><b>University of Central Florida</b><br/>Orlando, FL USA +<br/><b>Amazon, Inc</b><br/>Seattle, WA USA +<br/><b>Sighthound, Inc</b><br/>Orlando, FL USA +</td><td>('3174233', 'Hakan Boyraz', 'hakan boyraz')<br/>('2234898', 'Syed Zain Masood', 'syed zain masood')<br/>('6312216', 'Baoyuan Liu', 'baoyuan liu')<br/>('1691260', 'Hassan Foroosh', 'hassan foroosh')</td><td>hakanb@amazon.com +<br/>zainmasood@sighthound.com +<br/>bliu@cs.ucf.edu +<br/>tappenm@amazon.com +<br/>foroosh@cs.ucf.edu +</td></tr><tr><td>0066caed1238de95a431d836d8e6e551b3cde391</td><td>Filtered Component Analysis to Increase Robustness +<br/>to Local Minima in Appearance Models +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, Pennsylvania</b><br/><b>Pennsylvania</b></td><td>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td>ftorre@cs.cmu.edu acollet@cs.cmu.edu mquero@andrew.cmu.edu +<br/>tk@cs.cmu.edu +<br/>jeffcohn@pitt.edu +</td></tr><tr><td>00075519a794ea546b2ca3ca105e2f65e2f5f471</td><td>Generating a Large, Freely-Available Dataset for +<br/>Face-Related Algorithms +<br/><b>Amherst College</b></td><td>('40175953', 'Benjamin Mears', 'benjamin mears')</td><td></td></tr><tr><td>0019925779bff96448f0c75492717e4473f88377</td><td>Deep Heterogeneous Face Recognition Networks based on Cross-modal +<br/>Distillation and an Equitable Distance Metric +<br/><b>U.S. Army Research Laboratory</b><br/><b>University of Maryland, College Park</b><br/>3Booz Allen Hamilton Inc. +</td><td>('39412489', 'Christopher Reale', 'christopher reale')<br/>('2445131', 'Hyungtae Lee', 'hyungtae lee')<br/>('1688527', 'Heesung Kwon', 'heesung kwon')</td><td>reale@umiacs.umd.edu +<br/>lee hyungtae@bah.com +<br/>heesung.kwon.civ@mail.mil +</td></tr><tr><td>00e9011f58a561500a2910a4013e6334627dee60</td><td>FACIAL EXPRESSION RECOGNITION USING ANGLE-RELATED INFORMATION +<br/>FROM FACIAL MESHES +<br/>1Computer Science Department, Aristotle +<br/><b>University of Thessaloniki</b><br/><b>University Campus, 54124, Thessaloniki, Greece</b><br/>phone: (+30) 2310 996361, fax: (+30) 2310 996304, +<br/>web: www.aiia.csd.auth.gr +</td><td>('1738865', 'Nicholas Vretos', 'nicholas vretos')<br/>('1681629', 'Vassilios Solachidis', 'vassilios solachidis')<br/>('3176394', 'Petr Somol', 'petr somol')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>email: vretos,vasilis,pitas@aiia.csd.auth.gr +</td></tr><tr><td>00d9d88bb1bdca35663946a76d807fff3dc1c15f</td><td>Subjects and Their Objects: Localizing Interactees for a +<br/>Person-Centric View of Importance +</td><td>('3197570', 'Chao-Yeh Chen', 'chao-yeh chen')</td><td></td></tr><tr><td>00a967cb2d18e1394226ad37930524a31351f6cf</td><td>Fully-adaptive Feature Sharing in Multi-Task Networks with Applications in +<br/>Person Attribute Classification +<br/>UC San Diego +<br/>IBM Research +<br/>IBM Research +<br/>Binghamton Univeristy, SUNY +<br/>UC San Diego +<br/>Rogerio Feris +<br/>IBM Research +</td><td>('2325498', 'Yongxi Lu', 'yongxi lu')<br/>('8991006', 'Yu Cheng', 'yu cheng')<br/>('40632040', 'Abhishek Kumar', 'abhishek kumar')<br/>('2443456', 'Shuangfei Zhai', 'shuangfei zhai')<br/>('1737723', 'Tara Javidi', 'tara javidi')</td><td>yol070@ucsd.edu +<br/>abhishk@us.ibm.com +<br/>szhai2@binghamton.edu +<br/>chengyu@us.ibm.com +<br/>tjavidi@eng.ucsd.edu +<br/>rsferis@us.ibm.com +</td></tr><tr><td>00f1e5e954f9eb7ffde3ca74009a8c3c27358b58</td><td>Unsupervised Clustering for Google Searches of Celebrity Images +<br/><b>California Institute of Technology, Pasadena, CA</b><br/>* These authors contributed equally in this work +</td><td>('3075121', 'Alex Holub', 'alex holub')<br/>('2462051', 'Pierre Moreels', 'pierre moreels')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>holub@vision.caltech.edu, pmoreels@vision.caltech.edu, perona@vision.caltech.edu +</td></tr><tr><td>00a3cfe3ce35a7ffb8214f6db15366f4e79761e3</td><td>Kinect for real-time emotion recognition via facial expressions. Frontiers of +<br/>Information Technology & Electronic Engineering, 16(4):272-282. +<br/>[doi:10.1631/FITEE.1400209] +<br/>Using Kinect for real-time emotion +<br/>recognition via facial expressions +<br/>Key words: Kinect, Emotion recognition, Facial expression, Real-time +<br/>classification, Fusion algorithm, Support vector machine (SVM) +<br/> ORCID: http://orcid.org/0000-0002-5021-9057 +<br/>Front Inform Technol & Electron Eng </td><td>('2566775', 'Qi-rong Mao', 'qi-rong mao')<br/>('2016065', 'Xin-yu Pan', 'xin-yu pan')<br/>('20342486', 'Yong-zhao Zhan', 'yong-zhao zhan')<br/>('2800876', 'Xiang-jun Shen', 'xiang-jun shen')<br/>('2566775', 'Qi-rong Mao', 'qi-rong mao')</td><td>E-mail: mao_qr@ujs.edu.cn +</td></tr><tr><td>0058cbe110933f73c21fa6cc9ae0cd23e974a9c7</td><td>BISWAS, JACOBS: AN EFFICIENT ALGORITHM FOR LEARNING DISTANCES +<br/>An Efficient Algorithm for Learning +<br/>Distances that Obey the Triangle Inequality +<br/>http://www.xrci.xerox.com/profile-main/67 +<br/>http://www.cs.umd.edu/~djacobs/ +<br/>Xerox Research Centre India +<br/>Bangalore, India +<br/>Computer Science Department +<br/><b>University of Maryland</b><br/><b>College Park, USA</b></td><td>('2221075', 'Arijit Biswas', 'arijit biswas')<br/>('1682573', 'David Jacobs', 'david jacobs')</td><td></td></tr><tr><td>004a1bb1a2c93b4f379468cca6b6cfc6d8746cc4</td><td>Balanced k-Means and Min-Cut Clustering +</td><td>('1729163', 'Xiaojun Chang', 'xiaojun chang')<br/>('1688370', 'Feiping Nie', 'feiping nie')<br/>('1727419', 'Zhigang Ma', 'zhigang ma')<br/>('39033919', 'Yi Yang', 'yi yang')</td><td></td></tr><tr><td>00d94b35ffd6cabfb70b9a1d220b6823ae9154ee</td><td>Discriminative Bayesian Dictionary Learning +<br/>for Classification +</td><td>('2941543', 'Naveed Akhtar', 'naveed akhtar')<br/>('1688013', 'Faisal Shafait', 'faisal shafait')</td><td></td></tr><tr><td>00ebc3fa871933265711558fa9486057937c416e</td><td>Collaborative Representation based Classification +<br/>for Face Recognition +<br/><b>The Hong Kong Polytechnic University, Hong Kong, China</b><br/><b>b School of Applied Mathematics, Xidian University, Xi an, China</b><br/>c Principal Researcher, Microsoft Research Asia, Beijing, China +</td><td>('36685537', 'Lei Zhang', 'lei zhang')<br/>('5828998', 'Meng Yang', 'meng yang')<br/>('2340559', 'Xiangchu Feng', 'xiangchu feng')<br/>('1700297', 'Yi Ma', 'yi ma')<br/>('1698371', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>006f283a50d325840433f4cf6d15876d475bba77</td><td>756 +<br/>Preserving Structure in Model-Free Tracking +</td><td>('2883723', 'Lu Zhang', 'lu zhang')<br/>('1803520', 'Laurens van der Maaten', 'laurens van der maaten')</td><td></td></tr><tr><td>00b29e319ff8b3a521b1320cb8ab5e39d7f42281</td><td>Towards Transparent Systems: Semantic +<br/>Characterization of Failure Modes +<br/><b>Carnegie Mellon University, Pittsburgh, USA</b><br/><b>University of Washington, Seattle, USA</b><br/>3 Virginia Tech, Blacksburg, USA +</td><td>('3294630', 'Aayush Bansal', 'aayush bansal')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('1713589', 'Devi Parikh', 'devi parikh')</td><td></td></tr><tr><td>00d931eccab929be33caea207547989ae7c1ef39</td><td>The Natural Input Memory Model +<br/><b>IKAT, Universiteit Maastricht, St. Jacobsstraat 6, 6211 LB Maastricht, The Netherlands</b><br/><b>Universiteit van Amsterdam, Roeterstraat 15, 1018 WB Amsterdam, The Netherlands</b><br/><b>IKAT, Universiteit Maastricht, St. Jacobsstraat 6, 6211 LB Maastricht, The Netherlands</b></td><td></td><td>Joyca P.W. Lacroix (j.lacroix@cs.unimaas.nl) +<br/>Jaap M.J. Murre (jaap@murre.com) +<br/>Eric O. Postma (postma@cs.unimaas.nl) +<br/>H. Jaap van den Herik (herik@cs.unimaas.nl) +</td></tr><tr><td>0059b3dfc7056f26de1eabaafd1ad542e34c2c2e</td><td></td><td></td><td></td></tr><tr><td>0052de4885916cf6949a6904d02336e59d98544c</td><td><b>Springer Science + Business Media, Inc. Manufactured in The Netherlands</b><br/>DOI: 10.1007/s10994-005-3561-6 +<br/>Generalized Low Rank Approximations of Matrices +<br/><b>University of Minnesota-Twin Cities, Minneapolis</b><br/>MN 55455, USA +<br/>Editor: +<br/>Peter Flach +<br/>Published online: 12 August 2005 +</td><td>('37513601', 'Jieping Ye', 'jieping ye')</td><td>jieping@cs.umn.edu +</td></tr><tr><td>6e60536c847ac25dba4c1c071e0355e5537fe061</td><td>Computer Vision and Natural Language Processing: Recent +<br/>Approaches in Multimedia and Robotics +<br/>71 +<br/>Integrating computer vision and natural language processing is a novel interdisciplinary field that has +<br/>received a lot of attention recently. In this survey, we provide a comprehensive introduction of the integration +<br/>of computer vision and natural language processing in multimedia and robotics applications with more than +<br/>200 key references. The tasks that we survey include visual attributes, image captioning, video captioning, +<br/>visual question answering, visual retrieval, human-robot interaction, robotic actions, and robot navigation. +<br/>We also emphasize strategies to integrate computer vision and natural language processing models as a +<br/>unified theme of distributional semantics. We make an analog of distributional semantics in computer vision +<br/>and natural language processing as image embedding and word embedding, respectively. We also present a +<br/>unified view for the field and propose possible future directions. +<br/>Categories and Subject Descriptors: I.2.0 [Artificial Intelligence]: General; I.2.7 [Artificial Intelligence]: +<br/>Natural Language Processing; I.2.9 [Artificial Intelligence]: Robotics; I.2.10 [Artificial Intelligence]: +<br/>Vision and Scene Understanding; I.4.9 [Image Processing and Computer Vision]: Applications; I.5.4 +<br/>[Pattern Recognition]: Applications +<br/>General Terms: Computer Vision, Natural Language Processing, Robotics +<br/>Additional Key Words and Phrases: Language and vision, survey, multimedia, robotics, symbol grounding, +<br/>distributional semantics, computer vision, natural language processing, visual attribute, image captioning, +<br/>imitation learning, word2vec, word embedding, image embedding, semantic parsing, lexical semantics +<br/>ACM Reference Format: +<br/>Computer vision and natural language processing: Recent approaches in multimedia and robotics. ACM +<br/>Comput. Surv. 49, 4, Article 71 (December 2016), 44 pages. +<br/>DOI: http://dx.doi.org/10.1145/3009906 +<br/>1. INTRODUCTION +<br/>We have many ways to describe the world for communication between people: texts, +<br/>gestures, sign languages, and face expressions are all ways of sharing meaning. Lan- +<br/>guage is unique among communication systems in that its compositionality through +<br/>syntax allows a limitless number of meanings to be expressed. Such meaning ulti- +<br/>mately must be tied to perception of the world. This is usually referred to as the symbol +<br/>An earlier version of this article appeared as “Computer Vision and Natural Language Processing: Re- +<br/>cent Approaches in Multimedia and Robotics,” Scholarly Paper Archive, Department of Computer Science, +<br/><b>University of Maryland, College Park, MD</b><br/>Authors’ addresses: P. Wiriyathammabhum, C. Ferm ¨uller, and Y. Aloimonos, Computer Vision Lab, Uni- +<br/>Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted +<br/>without fee provided that copies are not made or distributed for profit or commercial advantage and that +<br/>copies show this notice on the first page or initial screen of a display along with the full citation. Copyrights for +</td><td>('2862582', 'Peratham Wiriyathammabhum', 'peratham wiriyathammabhum')<br/>('1937719', 'Douglas Summers-Stay', 'douglas summers-stay')<br/>('1697493', 'Yiannis Aloimonos', 'yiannis aloimonos')<br/>('2862582', 'Peratham Wiriyathammabhum', 'peratham wiriyathammabhum')<br/>('1937719', 'Douglas Summers-Stay', 'douglas summers-stay')<br/>('1697493', 'Yiannis Aloimonos', 'yiannis aloimonos')</td><td>versity of Maryland College Park, MD 20742-3275; email: {peratham@cs.umd.edu, fer@umiacs.umd.edu, +<br/>yiannis@cs.umd.edu}. D. Summers-Stay, U.S. Army Research Laboratory, Adelphi, MD 20783; email: +<br/>{douglas.a.summers-stay.civ@mail.mil}. +</td></tr><tr><td>6e198f6cc4199e1c4173944e3df6f39a302cf787</td><td>MORPH-II: Inconsistencies and Cleaning Whitepaper +<br/>NSF-REU Site at UNC Wilmington, Summer 2017 +</td><td>('39845059', 'G. Bingham', 'g. bingham')<br/>('1693470', 'B. Yip', 'b. yip')<br/>('1833570', 'M. Ferguson', 'm. ferguson')<br/>('1693283', 'C. Chen', 'c. chen')<br/>('11134292', 'Y. Wang', 'y. wang')<br/>('3369885', 'T. Kling', 't. kling')</td><td></td></tr><tr><td>6eaf446dec00536858548fe7cc66025b70ce20eb</td><td></td><td></td><td></td></tr><tr><td>6e173ad91b288418c290aa8891193873933423b3</td><td>Are you from North or South India? A hard race classification task reveals +<br/>systematic representational differences between humans and machines +<br/><b>aCentre for Neuroscience, Indian Institute of Science, Bangalore, India</b></td><td>('2478739', 'Harish Katti', 'harish katti')</td><td></td></tr><tr><td>6e91be2ad74cf7c5969314b2327b513532b1be09</td><td>Dimensionality Reduction with Subspace Structure +<br/>Preservation +<br/>Department of Computer Science +<br/>SUNY Buffalo +<br/>Buffalo, NY 14260 +</td><td>('2309967', 'Devansh Arpit', 'devansh arpit')<br/>('1841118', 'Ifeoma Nwogu', 'ifeoma nwogu')<br/>('1723877', 'Venu Govindaraju', 'venu govindaraju')</td><td>{devansh,inwogua,govind}@buffalo.edu +</td></tr><tr><td>6eba25166fe461dc388805cc2452d49f5d1cdadd</td><td>Pages 122.1-122.12 +<br/>DOI: https://dx.doi.org/10.5244/C.30.122 +</td><td></td><td></td></tr><tr><td>6e8a81d452a91f5231443ac83e4c0a0db4579974</td><td>Illumination robust face representation based on intrinsic geometrical +<br/>information +<br/>Soyel, H; Ozmen, B; McOwan, PW +<br/>This is a pre-copyedited, author-produced PDF of an article accepted for publication in IET +<br/>Conference on Image Processing (IPR 2012). The version of record is available +<br/>http://ieeexplore.ieee.org/document/6290632/?arnumber=6290632&tag=1 +<br/>For additional information about this publication click this link. +<br/>http://qmro.qmul.ac.uk/xmlui/handle/123456789/16147 +<br/>Information about this research object was correct at the time of download; we occasionally +<br/>make corrections to records, please therefore check the published record when citing. For +</td><td></td><td>more information contact scholarlycommunications@qmul.ac.uk +</td></tr><tr><td>6ed738ff03fd9042965abdfaa3ed8322de15c116</td><td>This document is downloaded from DR-NTU, Nanyang Technological +<br/><b>University Library, Singapore</b><br/>Title +<br/>K-MEAP: Generating Specified K Clusters with Multiple +<br/>Exemplars by Efficient Affinity Propagation +<br/>Author(s) Wang, Yangtao; Chen, Lihui +<br/>Citation +<br/>Wang, Y & Chen, L. (2014). K-MEAP: Generating +<br/>Specified K Clusters with Multiple Exemplars by Efficient +<br/>Affinity Propagation. 2014 IEEE International Conference +<br/>on Data Mining (ICDM), 1091-1096. +<br/>Date +<br/>2014 +<br/>URL +<br/>http://hdl.handle.net/10220/39690 +<br/>Rights +<br/>© 2014 IEEE. Personal use of this material is permitted. +<br/>Permission from IEEE must be obtained for all other +<br/><b>uses, in any current or future media, including</b><br/>reprinting/republishing this material for advertising or +<br/>promotional purposes, creating new collective works, for +<br/>resale or redistribution to servers or lists, or reuse of any +<br/>copyrighted component of this work in other works. The +<br/>published version is available at: +<br/>[http://dx.doi.org/10.1109/ICDM.2014.54]. +</td><td></td><td></td></tr><tr><td>6ecd4025b7b5f4894c990614a9a65e3a1ac347b2</td><td>International Journal on Recent and Innovation Trends in Computing and Communication +<br/> +<br/> ISSN: 2321-8169 +<br/>Volume: 2 Issue: 5 +<br/> 1275– 1281 +<br/>_______________________________________________________________________________________________ +<br/>Automatic Naming of Character using Video Streaming for Face +<br/>Recognition with Graph Matching +<br/>Nivedita.R.Pandey +<br/>Ranjan.P.Dahake +<br/>PG Student at MET’s IOE Bhujbal Knowledge City, +<br/>PG Student at MET’s IOE Bhujbal Knowledge City, +<br/>Nasik, Maharashtra, India, +<br/>Nasik, Maharashtra, India, +</td><td></td><td>pandeynivedita7@gmail.com +<br/>dahakeranjan@gmail.com +</td></tr><tr><td>6eddea1d991e81c1c3024a6cea422bc59b10a1dc</td><td>Towards automatic analysis of gestures and body +<br/>expressions in depression +<br/><b>University of Cambridge</b><br/>Computer Laboratory +<br/>Cambridge, UK +<br/><b>University of Cambridge</b><br/>Computer Laboratory +<br/>Cambridge, UK +</td><td>('2022940', 'Marwa Mahmoud', 'marwa mahmoud')<br/>('39840677', 'Peter Robinson', 'peter robinson')</td><td>marwa.mahmoud@cl.cam.ac.uk +<br/>peter.robinson@cl.cam.ac.uk +</td></tr><tr><td>6eaeac9ae2a1697fa0aa8e394edc64f32762f578</td><td></td><td></td><td></td></tr><tr><td>6ee2ea416382d659a0dddc7a88fc093accc2f8ee</td><td></td><td></td><td></td></tr><tr><td>6e97a99b2879634ecae962ddb8af7c1a0a653a82</td><td>Towards Context-aware Interaction Recognition∗ +<br/><b>School of Computer Science, University of Adelaide, Australia</b><br/>Contents +<br/>1. Introduction +<br/>2. Related work +<br/>3. Methods +<br/>3.1. Context-aware interaction classification +<br/>framework . . . . . . . . . . . . . . . . . +<br/>3.2. Feature representations for interactions +<br/>recognition . . . . . . . . . . . . . . . . +<br/>3.2.1 +<br/>Spatial feature representation . . +<br/>3.2.2 Appearance feature representation +<br/>Improving appearance representation +<br/>with attention and context-aware atten- +<br/>tion . . . . . . . . . . . . . . . . . . . . +<br/>3.4. Implementation details . . . . . . . . . . +<br/>3.3. +<br/>4. Experiments +<br/>4.1. Evaluation on the Visual Relationship +<br/>dataset . . . . . . . . . . . . . . . . . . . +<br/>4.1.1 Detection results comparison . . +<br/>Zero-shot learning performance +<br/>4.1.2 +<br/>evaluation . . . . . . . . . . . . . +<br/>4.1.3 Extensions and comparison with +<br/>the state-of-the-art methods . . . +<br/>4.2. Evaluation on the Visual Phrase dataset +<br/>5. Conclusion +</td><td>('3194022', 'Bohan Zhuang', 'bohan zhuang')<br/>('2161037', 'Lingqiao Liu', 'lingqiao liu')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')</td><td></td></tr><tr><td>6e9a8a34ab5b7cdc12ea52d94e3462225af2c32c</td><td>Fusing Aligned and Non-Aligned Face Information +<br/>for Automatic Affect Recognition in the Wild: A Deep Learning Approach +<br/>Computational NeuroSystems Laboratory (CNSL) +<br/><b>Korea Advanced Institute of Science and Technology (KAIST</b></td><td>('3918690', 'Bo-Kyeong Kim', 'bo-kyeong kim')<br/>('2527421', 'Suh-Yeon Dong', 'suh-yeon dong')<br/>('3294960', 'Jihyeon Roh', 'jihyeon roh')<br/>('34577016', 'Soo-Young Lee', 'soo-young lee')</td><td>{bokyeong1015, suhyeon.dong}@gmail.com, {rohleejh, gmkim90, sylee}@kaist.ac.kr +</td></tr><tr><td>6e3a181bf388dd503c83dc324561701b19d37df1</td><td>Finding a low-rank basis in a matrix subspace +<br/>Andr´e Uschmajew +</td><td>('2391697', 'Yuji Nakatsukasa', 'yuji nakatsukasa')</td><td></td></tr><tr><td>6ef1996563835b4dfb7fda1d14abe01c8bd24a05</td><td>Nonparametric Part Transfer for Fine-grained Recognition +<br/><b>Computer Vision Group, Friedrich Schiller University Jena</b><br/>www.inf-cv.uni-jena.de +</td><td>('1679449', 'Erik Rodner', 'erik rodner')<br/>('1720839', 'Alexander Freytag', 'alexander freytag')<br/>('1728382', 'Joachim Denzler', 'joachim denzler')</td><td></td></tr><tr><td>6e8c3b7d25e6530a631ea01fbbb93ac1e8b69d2f</td><td>Deep Episodic Memory: Encoding, Recalling, and Predicting +<br/>Episodic Experiences for Robot Action Execution +</td><td>('35309584', 'Jonas Rothfuss', 'jonas rothfuss')<br/>('2128564', 'Fabio Ferreira', 'fabio ferreira')<br/>('34876449', 'Eren Erdal Aksoy', 'eren erdal aksoy')<br/>('46432716', 'You Zhou', 'you zhou')<br/>('1722677', 'Tamim Asfour', 'tamim asfour')</td><td></td></tr><tr><td>6e911227e893d0eecb363015754824bf4366bdb7</td><td>Wasserstein Divergence for GANs +<br/>1 Computer Vision Lab, ETH Zurich, Switzerland +<br/>2 VISICS, KU Leuven, Belgium +</td><td>('1839268', 'Jiqing Wu', 'jiqing wu')<br/>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('30691454', 'Janine Thoma', 'janine thoma')<br/>('32610154', 'Dinesh Acharya', 'dinesh acharya')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{jwu,zhiwu.huang,jthoma,vangool}@vision.ee.ethz.ch, +<br/>acharyad@student.ethz.ch +</td></tr><tr><td>6ee8a94ccba10062172e5b31ee097c846821a822</td><td>Submitted 3/13; Revised 10/13; Published 12/13 +<br/>How to Solve Classification and Regression Problems on +<br/>High-Dimensional Data with a Supervised +<br/>Extension of Slow Feature Analysis +<br/>Institut f¨ur Neuroinformatik +<br/>Ruhr-Universit¨at Bochum +<br/>Bochum D-44801, Germany +<br/>Editor: David Dunson +</td><td>('2366497', 'Alberto N. Escalante', 'alberto n. escalante')<br/>('1736245', 'Laurenz Wiskott', 'laurenz wiskott')</td><td>ALBERTO.ESCALANTE@INI.RUB.DE +<br/>LAURENZ.WISKOTT@INI.RUB.DE +</td></tr><tr><td>6ee64c19efa89f955011531cde03822c2d1787b8</td><td>Table S1: Review of existing facial expression databases that are often used in social +<br/>psycholgy. +<br/>Author +<br/>Face +<br/>name +<br/>database +<br/>Expressions1 +<br/>Format +<br/>Short summary +<br/>[1] +<br/>GEMEP Corpus +<br/>Mind Reading: the +<br/>interactive +<br/>guide +<br/>to emotions +<br/>audio +<br/>and +<br/>video +<br/>record- +<br/>ings +<br/>Videos +<br/>anger, +<br/>amuse- +<br/>admiration, +<br/>ment, +<br/>tender- +<br/>ness, disgust, despair, +<br/>pride, +<br/>shame, anxiety +<br/>(worry), +<br/>interest, +<br/>irritation, joy (elation), +<br/>contempt, panic fear, +<br/>pleasure +<br/>(sensual), +<br/>relief, surprise, sadness +<br/>expressions +<br/>groups +<br/>:afraid, angry, bored, +<br/>bothered, disbelieving, +<br/>disgust, excited, +<br/>fond, +<br/>happy, hurt, interested, +<br/>kind, +<br/>romantic, +<br/>sad, sneaky, sorry, sure, +<br/>thinking, +<br/>surprised, +<br/>touched, +<br/>unfriendly, +<br/>unsure, wanting +<br/>liked, +<br/>RU-FACS Sponta- +<br/>neous Expression +<br/>Database +<br/>spontaneous facial ac- +<br/>tions +<br/>Videos +<br/>This database contains more than 7000 clips of the six basic +<br/>emotions as well as subtle emotions. For the recordings 10 +<br/>professional actors (5 female) were coached by a professional +<br/>director. The actors received a list of the emotions together +<br/>with short definitions and brief scenarios. The recordings are +<br/>available in different intensity levels and part of the database +<br/>has been validated. +<br/>The database contains over 400 videos of facial expressions +<br/>that are summarized in 24 groups. Each group consists of dif- +<br/>ferent subordinate expressions. Each expression is displayed +<br/>by 6 models ranging in age. +<br/>100 participants were asked for recording the database. There- +<br/>fore, a false option paradigm was used which is though to elicit +<br/>spontaneous facial expressions. Here, participants fill out a +<br/>questionnaire regarding their opinions about particular social +<br/>or political issues. Participants are then asked about their +<br/>answer by an interviewer. Either participants are asked to +<br/>tell the truth or to fool the interviewer. Moreover, partici- +<br/>pants were financially rewarded. The expressions were video +<br/>captured by four synchronized cameras and clips of 33 partic- +<br/>ipants have been FACS coded (onset, apex, and offset of the +<br/>face action). +<br/>Comprises 1008 short videos of expressions produced by 8 Ital- +<br/>ian professional actors. Each expression was recorded in three +<br/>intensities (low, medium, and high) and in two different condi- +<br/>tions: (1) Utterance condition in which actors spoke additional +<br/>sentences and (2) Non-Utterance condition. Here, actors were +<br/>additionally given scenarios according to the expressions to be +<br/>produced. +<br/>The expressions are taken from 12 participants (European, +<br/>Asian and African). Each expression was created using a di- +<br/>rect facial action task and all expressions were FACS coded. +<br/>Moreover, the expressions have been morphed into 5 different +<br/>levels of intensity. +<br/>It contains 165 greyscale images of 15 individuals one per +<br/>different facial expression or configuration (with or without +<br/>glasses, different camera perspectives). +<br/>This database contains two sets of facial expressions: (1) The +<br/>laboratory set, that includes 40 participants (varied in culture, +<br/>race, and appearance) displaying their own choice of expres- +<br/>sions. Participants were allowed to move their head without +<br/>going into profile view. Moreover, they were asked to avoid +<br/>speech. Each video sequence contains 1-3 expressions. +<br/>(2) +<br/>video recordings from TV that also contained speech. +<br/>The database contains videos of one actor performing approx- +<br/>imately 45 action units which were recorded from six different +<br/>viewpoints simultaneously. +<br/>For this database, between 19 and 97 different action units +<br/>were recorded form 10 participants. Action unit sequences +<br/>contain single and combined action units. The peak of each ex- +<br/>pression has been manually coded by certified FACS experts. +<br/>Moreover, a framework is proposed that allows to build dy- +<br/>namic 3D morphable models for the first time. +<br/>[2] +<br/>[3] +<br/>[5] +<br/>[6] +<br/>[7] +<br/>Breidt2 +<br/>[8] +<br/>Chen3 , 2007 +<br/>[4] +<br/>DaFEx +<br/>happiness, +<br/>fear, +<br/>and disgust +<br/>sadness, +<br/>surprise, +<br/>anger +<br/>Videos +<br/>Images +<br/>Images +<br/>Videos +<br/>happiness, +<br/>anger, +<br/>and embarrassment +<br/>fear, +<br/>sadness, +<br/>disgust, +<br/>happiness, +<br/>sleepy, +<br/>wink +<br/>surprise, +<br/>sadness, +<br/>and +<br/>happiness, +<br/>fear, +<br/>and disgust +<br/>sadness, +<br/>surprise, +<br/>anger +<br/>Facial action units +<br/>Videos +<br/>Facial action units +<br/>Videos +<br/>Montreal Set +<br/>of +<br/>facial displays of +<br/>emotion (MSFDE) +<br/>The Yale +<br/>Database +<br/>Face +<br/><b>University</b><br/>of +<br/>Database +<br/>Maryland +<br/>Face +<br/>Database +<br/>MPI +<br/>Video +<br/>the +<br/>of +<br/>Dynamic +<br/>FACS +<br/>(D3DFACS) +<br/>data +<br/>3D +<br/>set +<br/>Fa- +<br/>Taiwanese +<br/>cial +<br/>Expression +<br/>Database (TFEID) +<br/>anger, contempt, dis- +<br/>gust, +<br/>fear, happiness, +<br/>sadness and surprise +<br/>Images +<br/>The database consists of 7200 images captured from 40 indi- +<br/>viduals. The expressions are displayed in two (high and low) +<br/>intensities and two viewing angles (0◦ and 45◦ ) simultane- +<br/>ously. +<br/>[9] +<br/>CAFE Database +<br/>anger, disgust, happy, +<br/>maudlin (for sad), fear, +<br/>surprise +<br/>Images +<br/>The database consists of two normalized versions (one gamma +<br/>corrected and the other histogram equalized) of the faces. +<br/>1Neutral expression is not included. +<br/>2Please see http://vdb.kyb.tuebingen.mpg.de/. +<br/>3Please see http://bml.ym.edu.tw/ download/html/news.htm. +</td><td></td><td></td></tr><tr><td>6e00a406edb508312108f683effe6d3c1db020fb</td><td>Faces as Lighting Probes via Unsupervised Deep +<br/>Highlight Extraction +<br/><b>Simon Fraser University, Burnaby, Canada</b><br/><b>National University of Defense Technology, Changsha, China</b><br/>3 Microsoft Research, Beijing, China +</td><td>('2693616', 'Renjiao Yi', 'renjiao yi')<br/>('2041096', 'Chenyang Zhu', 'chenyang zhu')<br/>('37291674', 'Ping Tan', 'ping tan')<br/>('1686911', 'Stephen Lin', 'stephen lin')</td><td>{renjiaoy, cza68, pingtan}@sfu.ca +<br/>stevelin@microsoft.com +</td></tr><tr><td>6e94c579097922f4bc659dd5d6c6238a428c4d22</td><td>Graph Based Multi-class Semi-supervised +<br/>Learning Using Gaussian Process +<br/>State Key Laboratory of Intelligent Technology and Systems, +<br/><b>Tsinghua University, Beijing, China</b></td><td>('1809614', 'Yangqiu Song', 'yangqiu song')<br/>('1700883', 'Changshui Zhang', 'changshui zhang')<br/>('1760678', 'Jianguo Lee', 'jianguo lee')</td><td>{songyq99, lijg01}@mails.tsinghua.edu.cn, zcs@mail.tsinghua.edu.cn +</td></tr><tr><td>6e379f2d34e14efd85ae51875a4fa7d7ae63a662</td><td>A NEW MULTI-MODAL BIOMETRIC SYSTEM +<br/>BASED ON FINGERPRINT AND FINGER +<br/>VEIN RECOGNITION +<br/>Master's Thesis +<br/>Department of Software Engineering +<br/>JULY-2014 +<br/>I +</td><td>('37171106', 'Naveed AHMED', 'naveed ahmed')<br/>('1987743', 'Asaf VAROL', 'asaf varol')</td><td></td></tr><tr><td>6eb1e006b7758b636a569ca9e15aafd038d2c1b1</td><td>Human Capabilities on Video-based Facial +<br/>Expression Recognition +<br/><b>Faculty of Science and Engineering, Waseda University, Tokyo, Japan</b><br/>2 Institut f¨ur Informatik, Technische Universit¨at M¨unchen, Germany +</td><td>('32131501', 'Matthias Wimmer', 'matthias wimmer')<br/>('1989987', 'Ursula Zucker', 'ursula zucker')<br/>('1699132', 'Bernd Radig', 'bernd radig')</td><td></td></tr><tr><td>6eece104e430829741677cadc1dfacd0e058d60f</td><td>Automated Facial Image Analysis 1 +<br/>To appear in J. A. Coan & J. B. Allen (Eds.), The handbook of emotion elicitation and assess- +<br/><b>ment. Oxford University Press Series in Affective Science. New York: Oxford</b><br/>Use of Automated Facial Image Analysis for Measurement of Emotion Expression +<br/>Department of Psychology +<br/><b>University of Pittsburgh</b><br/>Takeo Kanade +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Facial expressions are a key index of emotion. They have consistent correlation with +<br/>self-reported emotion (Keltner, 1995; Rosenberg & Ekman, 1994; Ekman & Rosenberg, in press) +<br/>and emotion-related central and peripheral physiology (Davidson, Ekman, Saron, Senulis, & +<br/>Friesen, 1990; Fox & Davidson, 1988; Levenson, Ekman, & Friesen, 1990). They putatively +<br/>share similar underlying dimensions with self-reported emotion (e.g., positive and negative +<br/>affect) (Bullock & Russell, 1984; Gross & John, 1997; Watson & Tellegen, 1985). Facial +<br/>expressions serve interpersonal functions of emotion by conveying communicative intent, +<br/>signaling affective information in social referencing (Campos, Bertenthal, & Kermoian, 1992), +<br/>and more generally contributing to the regulation of social interaction (Cohn & Elmore, 1988; +<br/>Fridlund, 1994; Schmidt & Cohn, 2001). As a measure of trait affect, stability in facial +<br/>expression emerges early in life (Cohn & Campbell, 1992; Malatesta, Culver, Tesman, & +<br/>Shephard, 1989). By adulthood, stability is moderately strong, comparable to that for self- +<br/>reported emotion (Cohn, Schmidt, Gross, & Ekman, 2002), and predictive of favorable outcomes +<br/>in emotion-related domains including marriage and personal well-being over periods as long as +<br/>30 years (Harker & Keltner, 2001). Expressive changes in the face are a rich source of cues +<br/>about intra- and interpersonal functions of emotion (cf. Keltner & Haitd, 1999). +<br/>clinical practice, reliable, valid, and efficient methods of measurement are critical. Until recently, +<br/>selecting a measurement method meant choosing among one or another human-observer-based +<br/>coding system (e.g., Ekman & Friesen, 1978 and Izard, 1983) or facial electromyography +<br/>(EMG). While each of these approaches has advantages, they are not without costs. Human- +<br/>observer-based methods are time consuming to learn and use, and they are difficult to +<br/>standardize, especially across laboratories and over time (Bakeman & Gottman, 1986; Martin & +<br/>Bateson, 1986). Facial EMG requires placement of sensors on the face, which may inhibit facial +<br/>action and which rules out its use for naturalistic observation. +<br/>computer vision. Computer vision is the science of extracting and representing meaningful +<br/>information from digitized video and recognizing perceptually meaningful patterns. An early +<br/>focus in automated face image analysis by computer vision was face recognition (Kanade, 1973, +<br/>To make use of the information afforded by facial expression for emotion science and +<br/>An emerging alternative to these methods is automated facial image analysis using +</td><td>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td></td></tr><tr><td>6e0a05d87b3cc7e16b4b2870ca24cf5e806c0a94</td><td>RANDOM GRAPHS FOR STRUCTURE +<br/>DISCOVERY IN HIGH-DIMENSIONAL DATA +<br/>by +<br/>Jos¶e Ant¶onio O. Costa +<br/>A dissertation submitted in partial fulflllment +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>(Electrical Engineering: Systems) +<br/><b>in The University of Michigan</b><br/>2005 +<br/>Doctoral Committee: +<br/>Professor Alfred O. Hero III, Chair +<br/>Professor Jefirey A. Fessler +<br/>Professor David L. Neuhofi +</td><td>('1703616', 'Susan A. Murphy', 'susan a. murphy')</td><td></td></tr><tr><td>6e1802874ead801a7e1072aa870681aa2f555f35</td><td>1424407281/07/$20.00 ©2007 IEEE +<br/>I 629 +<br/>ICASSP 2007 +</td><td></td><td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a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td></tr><tr><td>6ed22b934e382c6f72402747d51aa50994cfd97b</td><td>Customized Expression Recognition for Performance-Driven +<br/>Cutout Character Animation +<br/>†NEC Laboratories America +<br/>‡Snapchat +</td><td>('39960064', 'Xiang Yu', 'xiang yu')<br/>('1706007', 'Jianchao Yang', 'jianchao yang')</td><td></td></tr><tr><td>6e93fd7400585f5df57b5343699cb7cda20cfcc2</td><td>http://journalofvision.org/9/2/22/ +<br/>Comparing a novel model based on the transferable +<br/>belief model with humans during the recognition of +<br/>partially occluded facial expressions +<br/>Département de Psychologie, Université de Montréal, +<br/>Canada +<br/>Département de Psychologie, Université de Montréal, +<br/>Canada +<br/>Département de Psychologie, Université de Montréal, +<br/>Canada +<br/>Humans recognize basic facial expressions effortlessly. Yet, despite a considerable amount of research, this task remains +<br/>elusive for computer vision systems. Here, we compared the behavior of one of the best computer models of facial +<br/>expression recognition (Z. Hammal, L. Couvreur, A. Caplier, & M. Rombaut, 2007) with the behavior of human observers +<br/>during the M. Smith, G. Cottrell, F. Gosselin, and P. G. Schyns (2005) facial expression recognition task performed on +<br/>stimuli randomly sampled using Gaussian apertures. The modelVwhich we had to significantly modify in order to give the +<br/>ability to deal with partially occluded stimuliVclassifies the six basic facial expressions (Happiness, Fear, Sadness, +<br/>Surprise, Anger, and Disgust) plus Neutral from static images based on the permanent facial feature deformations and the +<br/>Transferable Belief Model (TBM). Three simulations demonstrated the suitability of the TBM-based model to deal with +<br/>partially occluded facial parts and revealed the differences between the facial information used by humans and by the +<br/>model. This opens promising perspectives for the future development of the model. +<br/>Keywords: facial features behavior, facial expressions classification, Transferable Belief Model, Bubbles +<br/>Citation: Hammal, Z., Arguin, M., & Gosselin, F. (2009). Comparing a novel model based on the transferable belief +<br/>http://journalofvision.org/9/2/22/, doi:10.1167/9.2.22. +<br/>Introduction +<br/>Facial expressions communicate information from +<br/>which we can quickly infer the state of mind of our peers +<br/>and adjust our behavior accordingly (Darwin, 1872). To +<br/>illustrate, take a person like patient SM with complete +<br/>bilateral damage to the amygdala nuclei that prevents her +<br/>from recognizing facial expressions of fear. SM would be +<br/>incapable of interpreting the fearful expression on the face +<br/>of a bystander, who has encountered a furious Grizzly +<br/>bear, as a sign of potential +<br/>threat (Adolphs, Tranel, +<br/>Damasio, & Damasio, 1994). +<br/>Facial expressions are typically arranged into six +<br/>universally recognized basic categories Happiness, Sur- +<br/>prise, Disgust, Anger, Sadness, and Fear that are similarly +<br/>expressed across different backgrounds and cultures +<br/>(Cohn, 2006; Ekman, 1999; Izard, 1971, 1994). Facial +<br/>expressions result +<br/>from the precisely choreographed +<br/>deformation of facial features, which are often described +<br/>using the 46 Action Units (AUs; Ekman & Friesen, +<br/>1978). +<br/>Facial expression recognition and computer +<br/>vision +<br/>The study of human facial expressions has an impact in +<br/>several areas of life such as art, social interaction, cognitive +<br/>science, medicine, security, affective computing, and +<br/>human-computer interaction (HCI). An automatic facial +<br/>expressions classification system may contribute signifi- +<br/>cantly to the development of all these disciplines. However, +<br/>the development of such a system constitutes a significant +<br/>challenge because of the many constraints that are imposed +<br/>by its application in a real-world context (Pantic & Bartlett, +<br/>2007; Pantic & Patras, 2006). In particular, such systems +<br/>need to provide great accuracy and robustness without +<br/>demanding too many interventions from the user. +<br/>There have been major advances in computer vision +<br/>over the past 15 years for the recognition of the six basic +<br/>facial expressions (for reviews, see Fasel & Luettin, 2003; +<br/>Pantic & Rothkrantz, 2000b). The main approaches can be +<br/>divided in two classes: Model-based and fiducial points +<br/>approaches. The model-based approach requires the +<br/>design of a deterministic physical model that can represent +<br/>doi: 10.1167/9.2.22 +<br/>Received January 28, 2008; published February 26, 2009 +<br/>ISSN 1534-7362 * ARVO +</td><td>('1785007', 'Zakia Hammal', 'zakia hammal')<br/>('3005969', 'Martin Arguin', 'martin arguin')<br/>('2074568', 'Frédéric Gosselin', 'frédéric gosselin')</td><td></td></tr><tr><td>6eb1b5935b0613a41b72fd9e7e53a3c0b32651e9</td><td>LEGO Pictorial Scales for Assessing Affective Responses +<br/><b>t2i Lab, Chalmers University of Technology, Gothenburg, Sweden</b><br/>2Digital Productivity, CSIRO, Australia +<br/><b>University of Canterbury, New Zealand</b><br/><b>Texas AandM University, College Station, TX, USA</b><br/><b>Human Centered Multimedia, Augsburg University, Germany</b><br/><b>Human Interface Technology Lab New Zealand, University of Canterbury, New Zealand</b></td><td>('1761180', 'Mohammad Obaid', 'mohammad obaid')<br/>('39191121', 'Andreas Dünser', 'andreas dünser')<br/>('1719307', 'Elena Moltchanova', 'elena moltchanova')<br/>('33096182', 'Danielle Cummings', 'danielle cummings')<br/>('1728894', 'Christoph Bartneck', 'christoph bartneck')</td><td>mobaid@chalmers.se +</td></tr><tr><td>6e12ba518816cbc2d987200c461dc907fd19f533</td><td></td><td></td><td></td></tr><tr><td>6e782073a013ce3dbc5b9b56087fd0300c510f67</td><td>IOSR Journal of Computer Engineering (IOSR-JCE) +<br/>e-ISSN: 2278-0661,p-ISSN: 2278-8727, Volume 17, Issue 3, Ver. II (May – Jun. 2015), PP 61-68 +<br/>www.iosrjournals.org +<br/>Real Time Facial Emotion Recognition using Kinect V2 Sensor +<br/><b>Doctoral School of Automatic Control and Computers, University POLITEHNICA of Bucharest, Romania</b><br/><b>Ministry of Higher Education and Scientific Research / The University of Mustsnsiriyah/Baghdad IRAQ</b><br/>2(Department of Computers/Faculty of Automatic Control and ComputersPOLITEHNICA of Bucharest +<br/>3(Department of Computers/Faculty of Automatic Control and ComputersPOLITEHNICA of Bucharest +<br/>ROMANIA) +<br/>ROMANIA) +</td><td>('9384437', 'Hesham A. Alabbasi', 'hesham a. alabbasi')<br/>('3088730', 'Alin Moldoveanu', 'alin moldoveanu')</td><td></td></tr><tr><td>9ab463d117219ed51f602ff0ddbd3414217e3166</td><td>Weighted Transmedia +<br/>Relevance Feedback for +<br/>Image Retrieval and +<br/>Auto-annotation +<br/>TECHNICAL +<br/>REPORT +<br/>N° 0415 +<br/>December 2011 +<br/>Project-Teams LEAR - INRIA +<br/>and TVPA - XRCE +</td><td>('1722052', 'Thomas Mensink', 'thomas mensink')<br/>('34602236', 'Jakob Verbeek', 'jakob verbeek')<br/>('1808423', 'Gabriela Csurka', 'gabriela csurka')</td><td></td></tr><tr><td>9ac82909d76b4c902e5dde5838130de6ce838c16</td><td>Recognizing Facial Expressions Automatically +<br/>from Video +<br/>1 Introduction +<br/>Facial expressions, resulting from movements of the facial muscles, are the face +<br/>changes in response to a person’s internal emotional states, intentions, or social +<br/>communications. There is a considerable history associated with the study on fa- +<br/>cial expressions. Darwin (1872) was the first to describe in details the specific fa- +<br/>cial expressions associated with emotions in animals and humans, who argued that +<br/>all mammals show emotions reliably in their faces. Since that, facial expression +<br/>analysis has been a area of great research interest for behavioral scientists (Ekman, +<br/>Friesen, and Hager, 2002). Psychological studies (Mehrabian, 1968; Ambady and +<br/>Rosenthal, 1992) suggest that facial expressions, as the main mode for non-verbal +<br/>communication, play a vital role in human face-to-face communication. For illus- +<br/>tration, we show some examples of facial expressions in Fig. 1. +<br/>Computer recognition of facial expressions has many important applications in +<br/>intelligent human-computer interaction, computer animation, surveillance and se- +<br/>curity, medical diagnosis, law enforcement, and awareness systems (Shan, 2007). +<br/>Therefore, it has been an active research topic in multiple disciplines such as psy- +<br/>chology, cognitive science, human-computer interaction, and pattern recognition. +<br/>Meanwhile, as a promising unobtrusive solution, automatic facial expression analy- +<br/>sis from video or images has received much attention in last two decades (Pantic and +<br/>Rothkrantz, 2000a; Fasel and Luettin, 2003; Tian, Kanade, and Cohn, 2005; Pantic +<br/>and Bartlett, 2007). +<br/>This chapter introduces recent advances in computer recognition of facial expres- +<br/>sions. Firstly, we describe the problem space, which includes multiple dimensions: +<br/>level of description, static versus dynamic expression, facial feature extraction and +</td><td>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('3297850', 'Ralph Braspenning', 'ralph braspenning')<br/>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('3297850', 'Ralph Braspenning', 'ralph braspenning')</td><td>Philips Research, Eindhoven, The Netherlands, e-mail: caifeng.shan@philips.com +<br/>Philips Research, Eindhoven, The Netherlands, e-mail: ralph.braspenning@philips.com +</td></tr><tr><td>9a0c7a4652c49a177460b5d2fbbe1b2e6535e50a</td><td>Automatic and Quantitative evaluation of attribute discovery methods +<br/><b>The University of Queensland, School of ITEE</b><br/>QLD 4072, Australia +</td><td>('2499431', 'Liangchen Liu', 'liangchen liu')<br/>('2331880', 'Arnold Wiliem', 'arnold wiliem')<br/>('3104113', 'Shaokang Chen', 'shaokang chen')<br/>('2270092', 'Brian C. Lovell', 'brian c. lovell')</td><td>l.liu9@uq.edu.au +<br/>a.wiliem@uq.edu.au +<br/>shaokangchenuq@gmail.com +<br/>lovell@itee.uq.edu.au +</td></tr><tr><td>9ac15845defcd0d6b611ecd609c740d41f0c341d</td><td>Copyright +<br/>by +<br/>2011 +</td><td>('1926834', 'Juhyun Lee', 'juhyun lee')</td><td></td></tr><tr><td>9ac43a98fe6fde668afb4fcc115e4ee353a6732d</td><td>Survey of Face Detection on Low-quality Images +<br/><b>Beckmann Institute, University of Illinois at Urbana-Champaign, USA</b></td><td>('1698743', 'Yuqian Zhou', 'yuqian zhou')<br/>('1771885', 'Ding Liu', 'ding liu')</td><td>{yuqian2, dingliu2}@illinois.edu +<br/>huang@ifp.uiuc.edu +</td></tr><tr><td>9af1cf562377b307580ca214ecd2c556e20df000</td><td>Feb. 28 +<br/> International Journal of Advanced Studies in Computer Science and Engineering +<br/>IJASCSE, Volume 4, Issue 2, 2015 +<br/> Video-Based Facial Expression Recognition +<br/>Using Local Directional Binary Pattern +<br/>Electrical Engineering Dept., AmirKabir Univarsity of Technology +<br/>Tehran, Iran +</td><td>('38519671', 'Sahar Hooshmand', 'sahar hooshmand')<br/>('3232144', 'Ali Jamali Avilaq', 'ali jamali avilaq')<br/>('3293075', 'Amir Hossein Rezaie', 'amir hossein rezaie')</td><td></td></tr><tr><td>9a23a0402ae68cc6ea2fe0092b6ec2d40f667adb</td><td>High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs +<br/>1NVIDIA Corporation +<br/>2UC Berkeley +<br/>Figure 1: We propose a generative adversarial framework for synthesizing 2048 × 1024 images from semantic label maps +<br/>(lower left corner in (a)). Compared to previous work [5], our results express more natural textures and details. (b) We can +<br/>change labels in the original label map to create new scenes, like replacing trees with buildings. (c) Our framework also +<br/>allows a user to edit the appearance of individual objects in the scene, e.g. changing the color of a car or the texture of a road. +<br/>Please visit our website for more side-by-side comparisons as well as interactive editing demos. +</td><td>('2195314', 'Ting-Chun Wang', 'ting-chun wang')<br/>('2436356', 'Jun-Yan Zhu', 'jun-yan zhu')<br/>('1690538', 'Jan Kautz', 'jan kautz')</td><td></td></tr><tr><td>9a4c45e5c6e4f616771a7325629d167a38508691</td><td>A Facial Features Detector Integrating Holistic Facial Information and +<br/>Part-based Model +<br/>Eslam Mostafa1,2 +<br/>Aly Farag1 +<br/><b>CVIP Lab, University of Louisville, Louisville, KY 40292, USA</b><br/><b>Alexandria University, Alexandria, Egypt</b><br/><b>Assiut University, Assiut 71515, Egypt</b><br/>4Kentucky Imaging Technology (KIT), Louisville, KY 40245, USA. +</td><td>('28453046', 'Asem A. Ali', 'asem a. ali')<br/>('2239392', 'Ahmed Shalaby', 'ahmed shalaby')</td><td></td></tr><tr><td>9af9a88c60d9e4b53e759823c439fc590a4b5bc5</td><td>Learning Deep Convolutional Embeddings for Face Representation Using Joint +<br/>Sample- and Set-based Supervision +<br/>Department of Electrical and Electronic Engineering, +<br/><b>Imperial College London</b></td><td>('2151914', 'Baris Gecer', 'baris gecer')<br/>('3288623', 'Vassileios Balntas', 'vassileios balntas')<br/>('1700968', 'Tae-Kyun Kim', 'tae-kyun kim')</td><td>{b.gecer,v.balntas15,tk.kim}@imperial.ac.uk +</td></tr><tr><td>9a7858eda9b40b16002c6003b6db19828f94a6c6</td><td>MOONEY FACE CLASSIFICATION AND PREDICTION BY LEARNING ACROSS TONE +<br/>(cid:63) UC Berkeley / †ICSI +</td><td>('2301765', 'Tsung-Wei Ke', 'tsung-wei ke')<br/>('2251428', 'Stella X. Yu', 'stella x. yu')<br/>('1821337', 'David Whitney', 'david whitney')</td><td></td></tr><tr><td>9a3535cabf5d0f662bff1d897fb5b777a412d82e</td><td><b>University of Kentucky</b><br/>UKnowledge +<br/>Computer Science +<br/>Computer Science Faculty Publications +<br/>6-10-2015 +<br/>Large-Scale Geo-Facial Image Analysis +<br/>Mohammed T. Islam +<br/><b>University of Kentucky</b><br/><b>University of North Carolina at Charlotte</b><br/>Click here to let us know how access to this document benefits you. +<br/>Follow this and additional works at: https://uknowledge.uky.edu/cs_facpub +<br/>Part of the Computer Sciences Commons +<br/>Repository Citation +<br/>Islam, Mohammed T.; Greenwell, Connor; Souvenir, Richard; and Jacobs, Nathan, "Large-Scale Geo-Facial Image Analysis" (2015). +<br/>Computer Science Faculty Publications. 7. +<br/>https://uknowledge.uky.edu/cs_facpub/7 +<br/>This Article is brought to you for free and open access by the Computer Science at UKnowledge. It has been accepted for inclusion in Computer +</td><td>('2121759', 'Connor Greenwell', 'connor greenwell')<br/>('1690110', 'Richard Souvenir', 'richard souvenir')<br/>('1990750', 'Nathan Jacobs', 'nathan jacobs')</td><td>University of Kentucky, connor.greenwell@uky.edu +<br/>University of Kentucky, nathan.jacobs@uky.edu +<br/>Science Faculty Publications by an authorized administrator of UKnowledge. For more information, please contact UKnowledge@lsv.uky.edu. +</td></tr><tr><td>9abd35b37a49ee1295e8197aac59bde802a934f3</td><td>Depth2Action: Exploring Embedded Depth for +<br/>Large-Scale Action Recognition +<br/><b>University of California, Merced</b></td><td>('1749901', 'Yi Zhu', 'yi zhu')</td><td>{yzhu25,snewsam}@ucmerced.edu +</td></tr><tr><td>9a276c72acdb83660557489114a494b86a39f6ff</td><td>Emotion Classification through Lower Facial Expressions using Adaptive +<br/>Support Vector Machines +<br/>Department of Information Technology, Faculty of Industrial Technology and Management, +</td><td>('2621463', 'Porawat Visutsak', 'porawat visutsak')</td><td>King Mongkut’s University of Technology North Bangkok, porawatv@kmutnb.ac.th +</td></tr><tr><td>9a1a9dd3c471bba17e5ce80a53e52fcaaad4373e</td><td>Automatic Recognition of Spontaneous Facial +<br/>Actions +<br/><b>Institute for Neural Computation, University of California, San Diego</b><br/><b>University at Buffalo, State University of New York</b></td><td>('2218905', 'Marian Stewart Bartlett', 'marian stewart bartlett')<br/>('21751782', 'Gwen C. Littlewort', 'gwen c. littlewort')<br/>('2639526', 'Mark G. Frank', 'mark g. frank')<br/>('2767464', 'Claudia Lainscsek', 'claudia lainscsek')<br/>('2039025', 'Ian R. Fasel', 'ian r. fasel')<br/>('1741200', 'Javier R. Movellan', 'javier r. movellan')</td><td>mbartlet@ucsd.edu, gwen@mplab.ucsd.edu, clainscsek@ucsd.edu, ianfasel@cogsci.ucsd.edu, +<br/>movellan@mplab.ucsd.edu +<br/>mfrank83@buffalo.edu +</td></tr><tr><td>9a42c519f0aaa68debbe9df00b090ca446d25bc4</td><td>Face Recognition via Centralized Coordinate +<br/>Learning +</td><td>('2689287', 'Xianbiao Qi', 'xianbiao qi')<br/>('1684635', 'Lei Zhang', 'lei zhang')</td><td></td></tr><tr><td>9aad8e52aff12bd822f0011e6ef85dfc22fe8466</td><td>Temporal-Spatial Mapping for Action Recognition +</td><td>('3865974', 'Xiaolin Song', 'xiaolin song')<br/>('40093162', 'Cuiling Lan', 'cuiling lan')<br/>('8434337', 'Wenjun Zeng', 'wenjun zeng')<br/>('1757173', 'Junliang Xing', 'junliang xing')<br/>('1759461', 'Jingyu Yang', 'jingyu yang')<br/>('1692735', 'Xiaoyan Sun', 'xiaoyan sun')</td><td></td></tr><tr><td>36b40c75a3e53c633c4afb5a9309d10e12c292c7</td><td></td><td></td><td></td></tr><tr><td>363ca0a3f908859b1b55c2ff77cc900957653748</td><td>International Journal of Computer Trends and Technology (IJCTT) – volume 1 Issue 3 Number 4 – Aug 2011 +<br/> Local Binary Patterns and Linear Programming using +<br/>Facial Expression +<br/>Ms.P.Jennifer +<br/><b>Bharath Institute of Science and Technology</b><br/><b>B.Tech (C.S.E), Bharath University, Chennai</b><br/>Dr. A. Muthu kumaravel +<br/><b>Bharath Institute of Science and Technology</b><br/><b>B.Tech (C.S.E), Bharath University, Chennai</b><br/> +</td><td></td><td></td></tr><tr><td>36939e6a365e9db904d81325212177c9e9e76c54</td><td>Assessing the Accuracy of Four Popular Face Recognition Tools for +<br/>Inferring Gender, Age, and Race +<br/><b>Qatar Computing Research Institute, HBKU</b><br/>HBKU Research Complex, Doha, P.O. Box 34110, Qatar +</td><td>('1861541', 'Soon-Gyo Jung', 'soon-gyo jung')<br/>('40660541', 'Jisun An', 'jisun an')<br/>('2592694', 'Haewoon Kwak', 'haewoon kwak')<br/>('2734912', 'Joni Salminen', 'joni salminen')</td><td>{sjung,jan,hkwak,jsalminen,bjansen}@hbku.edu.qa +</td></tr><tr><td>3646b42511a6a0df5470408bc9a7a69bb3c5d742</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Applications of Computers and Electronics for the Welfare of Rural Masses (ACEWRM) 2015 +<br/>Detection of Facial Parts based on ABLATA +<br/>Technical Campus, Bhilai +<br/>Vikas Singh +<br/>Technical Campus, Bhilai +<br/>Abha Choubey +<br/>Technical Campus, Bhilai +</td><td>('9173769', 'Siddhartha Choubey', 'siddhartha choubey')</td><td></td></tr><tr><td>365f67fe670bf55dc9ccdcd6888115264b2a2c56</td><td></td><td></td><td></td></tr><tr><td>36fe39ed69a5c7ff9650fd5f4fe950b5880760b0</td><td>Tracking von Gesichtsmimik +<br/>mit Hilfe von Gitterstrukturen +<br/>zur Klassifikation von schmerzrelevanten Action +<br/>Units +<br/>1Fraunhofer-Institut f¨ur Integrierte Schaltungen IIS, Erlangen, +<br/>2Otto-Friedrich-Universit¨at Bamberg, 3Universit¨atsklinkum Erlangen +<br/>Kurzfassung. In der Schmerzforschung werden schmerzrelevante Mi- +<br/>mikbewegungen von Probanden mittels des Facial Action Coding System +<br/>klassifiziert. Die manuelle Klassifikation hierbei ist aufw¨andig und eine +<br/>automatische (Vor-)klassifikation k¨onnte den diagnostischen Wert dieser +<br/>Analysen erh¨ohen sowie den klinischen Workflow unterst¨utzen. Der hier +<br/>vorgestellte regelbasierte Ansatz erm¨oglicht eine automatische Klassifika- +<br/>tion ohne große Trainingsmengen vorklassifizierter Daten. Das Verfahren +<br/>erkennt und verfolgt Mimikbewegungen, unterst¨utzt durch ein Gitter, +<br/>und ordnet diese Bewegungen bestimmten Gesichtsarealen zu. Mit die- +<br/>sem Wissen kann aus den Bewegungen auf die zugeh¨origen Action Units +<br/>geschlossen werden. +<br/>1 Einleitung +<br/>Menschliche Empfindungen wie Emotionen oder Schmerz l¨osen spezifische Mu- +<br/>ster von Kontraktionen der Gesichtsmuskulatur aus, die Grundlage dessen sind, +<br/>was wir Mimik nennen. Aus der Beobachtung der Mimik kann wiederum auf +<br/>menschliche Empfindungen r¨uckgeschlossen werden. Im Rahmen der Schmerz- +<br/>forschung werden Videoaufnahmen von Probanden hinsichtlich des mimischen +<br/>Schmerzausdrucks analysiert. Zur Beschreibung des mimischen Ausdrucks und +<br/>dessen Ver¨anderungen wird das Facial Action Coding System (FACS) [1] verwen- +<br/>det, das anatomisch begr¨undet, kleinste sichtbare Muskelbewegungen im Gesicht +<br/>beschreibt und als einzelne Action Units (AUs) kategorisiert. Eine Vielzahl von +<br/>Untersuchungen hat gezeigt, dass spezifische Muster von Action Units auftre- +<br/>ten, wenn Probanden Schmerzen angeben [2]. Die manuelle Klassifikation und +<br/>Markierung der Action Units von Probanden in Videosequenzen bedarf einer +<br/>langwierigen Beobachtung durch ausgebildete FACS-Coder. Eine automatische +<br/>(Vor-)klassifikation kann hierbei den klinischen Workflow unterst¨utzen und dieses +<br/>Verfahren zum brauchbaren diagnostischen Instrument machen. Bisher realisier- +<br/>te Ans¨atze zum Erkennen von Gesichtsausdr¨ucken basieren auf der Klassifikation +</td><td>('31431972', 'Christine Barthold', 'christine barthold')<br/>('2009811', 'Anton Papst', 'anton papst')<br/>('1773752', 'Thomas Wittenberg', 'thomas wittenberg')<br/>('1793798', 'Stefan Lautenbacher', 'stefan lautenbacher')<br/>('1727734', 'Ute Schmid', 'ute schmid')<br/>('2500903', 'Sven Friedl', 'sven friedl')</td><td>sven.friedl@iis.fraunhofer.de +</td></tr><tr><td>36a3a96ef54000a0cd63de867a5eb7e84396de09</td><td>Automatic Photo Orientation Detection with Convolutional Neural Networks +<br/>Dept. of Computer Science +<br/><b>University of Toronto</b><br/>Toronto, Ontario, Canada +</td><td>('40121109', 'Ujash Joshi', 'ujash joshi')<br/>('1959343', 'Michael Guerzhoy', 'michael guerzhoy')</td><td>ujash.joshi@utoronto.ca, guerzhoy@cs.toronto.edu +</td></tr><tr><td>36fc4120fc0638b97c23f97b53e2184107c52233</td><td>National Conference on Innovative Paradigms in Engineering & Technology (NCIPET-2013) +<br/>Proceedings published by International Journal of Computer Applications® (IJCA) +<br/>Introducing Celebrities in an Images using HAAR +<br/>Cascade algorithm +<br/>Asst. Professor +<br/><b>PES Modern College of Engg</b><br/><b>PES Modern College of Engg</b><br/><b>PES Modern College of Engg</b><br/>Shivaji Nagar, Pune +<br/>Shivaji Nagar, Pune +<br/>Shivaji Nagar, Pune +</td><td>('12682677', 'Deipali V. Gore', 'deipali v. gore')</td><td></td></tr><tr><td>36ce0b68a01b4c96af6ad8c26e55e5a30446f360</td><td>Multimed Tools Appl +<br/>DOI 10.1007/s11042-014-2322-6 +<br/>Facial expression recognition based on a mlp neural +<br/>network using constructive training algorithm +<br/>Received: 5 February 2014 / Revised: 22 August 2014 / Accepted: 13 October 2014 +<br/>© Springer Science+Business Media New York 2014 +</td><td>('1746834', 'Hayet Boughrara', 'hayet boughrara')<br/>('3410172', 'Chokri Ben Amar', 'chokri ben amar')</td><td></td></tr><tr><td>3674f3597bbca3ce05e4423611d871d09882043b</td><td>ISSN 1796-2048 +<br/>Volume 7, Number 4, August 2012 +<br/>Contents +<br/>Special Issue: Multimedia Contents Security in Social Networks Applications +<br/>Guest Editors: Zhiyong Zhang and Muthucumaru Maheswaran +<br/>Guest Editorial +<br/>Zhiyong Zhang and Muthucumaru Maheswaran +<br/>SPECIAL ISSUE PAPERS +<br/>DRTEMBB: Dynamic Recommendation Trust Evaluation Model Based on Bidding +<br/>Gang Wang and Xiao-lin Gui +<br/>Block-Based Parallel Intra Prediction Scheme for HEVC +<br/>Jie Jiang, Baolong, Wei Mo, and Kefeng Fan +<br/>Optimized LSB Matching Steganography Based on Fisher Information +<br/>Yi-feng Sun, Dan-mei Niu, Guang-ming Tang, and Zhan-zhan Gao +<br/>A Novel Robust Zero-Watermarking Scheme Based on Discrete Wavelet Transform +<br/>Yu Yang, Min Lei, Huaqun Liu, Yajian Zhou, and Qun Luo +<br/>Stego Key Estimation in LSB Steganography +<br/>Jing Liu and Guangming Tang +<br/>REGULAR PAPERS +<br/>Facial Expression Spacial Charts for Describing Dynamic Diversity of Facial Expressions +<br/>277 +<br/>279 +<br/>289 +<br/>295 +<br/>303 +<br/>309 +<br/>314 +</td><td>('46575279', 'H. Madokoro', 'h. madokoro')</td><td></td></tr><tr><td>362bfeb28adac5f45b6ef46c07c59744b4ed6a52</td><td>INCORPORATING SCALABILITY IN UNSUPERVISED SPATIO-TEMPORAL FEATURE +<br/>LEARNING +<br/><b>University of California, Riverside, CA</b></td><td>('49616225', 'Sujoy Paul', 'sujoy paul')<br/>('2177805', 'Sourya Roy', 'sourya roy')<br/>('1688416', 'Amit K. Roy-Chowdhury', 'amit k. roy-chowdhury')</td><td></td></tr><tr><td>360d66e210f7011423364327b7eccdf758b5fdd2</td><td>17th European Signal Processing Conference (EUSIPCO 2009) +<br/>Glasgow, Scotland, August 24-28, 2009 +<br/>LOCAL FEATURE EXTRACTION METHODS FOR FACIAL EXPRESSION +<br/>RECOGNITION +<br/><b>School of Electrical and Computer Engineering, RMIT University</b><br/>City Campus, Swanston St., Melbourne, Australia +<br/>http://www.rmit.edu.au +</td><td>('1857490', 'Seyed Mehdi Lajevardi', 'seyed mehdi lajevardi')<br/>('1749220', 'Zahir M. Hussain', 'zahir m. hussain')</td><td>seyed.lajevardi@rmit.edu.au, zmhussain@ieee.org +</td></tr><tr><td>365866dc937529c3079a962408bffaa9b87c1f06</td><td> IJISET - International Journal of Innovative Science, Engineering & Technology, Vol. 1 Issue 3, May 2014. +<br/>www.ijiset.com +<br/>ISSN 2348 – 7968 +<br/>Facial Feature Expression Based Approach for Human Face +<br/>Recognition: A Review +<br/><b>SSESA, Science College, Congress Nagar, Nagpur, (MS)-India</b><br/><b>RTM Nagpur University, Campus Nagpur, (MS)-India</b><br/>for +<br/>face +<br/>task +<br/>required +<br/>extraction of +</td><td></td><td></td></tr><tr><td>361c9ba853c7d69058ddc0f32cdbe94fbc2166d5</td><td>Deep Reinforcement Learning of +<br/>Video Games +<br/>s2098407 +<br/>September 29, 2017 +<br/>MSc. Project +<br/>Arti(cid:12)cial Intelligence +<br/><b>University of Groningen, The Netherlands</b><br/>Supervisors +<br/>Dr. M.A. (Marco) Wiering +<br/>Prof. dr. L.R.B. (Lambert) Schomaker +<br/><b>ALICE Institute</b><br/><b>University of Groningen</b><br/>Nijenborgh 9, 9747 AG, Groningen, The Netherlands +</td><td>('3405120', 'Jos van de Wolfshaar', 'jos van de wolfshaar')</td><td></td></tr><tr><td>368e99f669ea5fd395b3193cd75b301a76150f9d</td><td>One-to-many face recognition with bilinear CNNs +<br/>Aruni RoyChowdhury +<br/><b>University of Massachusetts, Amherst</b><br/>Erik Learned-Miller +</td><td>('2144284', 'Tsung-Yu Lin', 'tsung-yu lin')<br/>('35208858', 'Subhransu Maji', 'subhransu maji')</td><td>{arunirc,tsungyulin,smaji,elm}@cs.umass.edu +</td></tr><tr><td>362a70b6e7d55a777feb7b9fc8bc4d40a57cde8c</td><td>978-1-4799-9988-0/16/$31.00 ©2016 IEEE +<br/>2792 +<br/>ICASSP 2016 +</td><td></td><td></td></tr><tr><td>36df81e82ea5c1e5edac40b60b374979a43668a5</td><td>ON-THE-FLY SPECIFIC PERSON RETRIEVAL +<br/><b>University of Oxford, United Kingdom</b></td><td>('3188342', 'Omkar M. Parkhi', 'omkar m. parkhi')<br/>('1687524', 'Andrea Vedaldi', 'andrea vedaldi')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>{omkar,vedaldi,az}@robots.ox.ac.uk +</td></tr><tr><td>3619a9b46ad4779d0a63b20f7a6a8d3d49530339</td><td>SIMONYAN et al.: FISHER VECTOR FACES IN THE WILD +<br/>Fisher Vector Faces in the Wild +<br/>Visual Geometry Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b></td><td>('34838386', 'Karen Simonyan', 'karen simonyan')<br/>('3188342', 'Omkar M. Parkhi', 'omkar m. parkhi')<br/>('1687524', 'Andrea Vedaldi', 'andrea vedaldi')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>karen@robots.ox.ac.uk +<br/>omkar@robots.ox.ac.uk +<br/>vedaldi@robots.ox.ac.uk +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>366d20f8fd25b4fe4f7dc95068abc6c6cabe1194</td><td></td><td></td><td></td></tr><tr><td>3630324c2af04fd90f8668f9ee9709604fe980fd</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TCSVT.2016.2607345, IEEE +<br/>Transactions on Circuits and Systems for Video Technology +<br/>Image Classification with Tailored Fine-Grained +<br/>Dictionaries +</td><td>('2287686', 'Xiangbo Shu', 'xiangbo shu')<br/>('8053308', 'Jinhui Tang', 'jinhui tang')<br/>('2272096', 'Guo-Jun Qi', 'guo-jun qi')<br/>('3233021', 'Zechao Li', 'zechao li')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>362ba8317aba71c78dafca023be60fb71320381d</td><td></td><td></td><td></td></tr><tr><td>36cf96fe11a2c1ea4d999a7f86ffef6eea7b5958</td><td>RGB-D Face Recognition with Texture and +<br/>Attribute Features +<br/>Member, IEEE +</td><td>('1931069', 'Gaurav Goswami', 'gaurav goswami')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('39129417', 'Richa Singh', 'richa singh')</td><td></td></tr><tr><td>36e8ef2e5d52a78dddf0002e03918b101dcdb326</td><td>Multiview Active Shape Models with SIFT Descriptors +<br/>for the 300-W Face Landmark Challenge +<br/><b>University of Cape Town</b><br/>Anthropics Technology Ltd. +<br/><b>University of Cape Town</b></td><td>('2822258', 'Stephen Milborrow', 'stephen milborrow')<br/>('1823550', 'Tom E. Bishop', 'tom e. bishop')<br/>('2537623', 'Fred Nicolls', 'fred nicolls')</td><td>milbo@sonic.net +<br/>t.e.bishop@gmail.com +<br/>fred.nicolls@uct.ac.za +</td></tr><tr><td>36018404263b9bb44d1fddaddd9ee9af9d46e560</td><td>OCCLUDED FACE RECOGNITION BY USING GABOR +<br/>FEATURES +<br/>1 Department of Electrical And Electronics Engineering, METU, Ankara, Turkey +<br/>2 7h%ł7$.(cid:3)%ł/7(1(cid:15)(cid:3)$QNDUD(cid:15)(cid:3)7XUNH\ +</td><td>('2920043', 'Burcu Kepenekci', 'burcu kepenekci')<br/>('3110567', 'F. Boray Tek', 'f. boray tek')<br/>('1929001', 'Gozde Bozdagi Akar', 'gozde bozdagi akar')</td><td></td></tr><tr><td>367f2668b215e32aff9d5122ce1f1207c20336c8</td><td>Proceedings of the Pakistan Academy of Sciences 52 (1): 15–25 (2015) +<br/>Copyright © Pakistan Academy of Sciences +<br/>ISSN: 0377 - 2969 (print), 2306 - 1448 (online) +<br/> Pakistan Academy of Sciences +<br/>Research Article +<br/>Speaker-Dependent Human Emotion Recognition in +<br/>Unimodal and Bimodal Scenarios +<br/><b>University of Peshawar, Pakistan</b><br/><b>University of Engineering and Technology</b><br/><b>Sarhad University of Science and Information Technology</b><br/><b>University of Peshawar, Peshawar, Pakistan</b><br/>Peshawar, Pakistan +<br/>Peshawar, Pakistan +</td><td>('34267835', 'Sanaul Haq', 'sanaul haq')<br/>('3124216', 'Tariqullah Jan', 'tariqullah jan')<br/>('1766329', 'Muhammad Asif', 'muhammad asif')<br/>('1710701', 'Amjad Ali', 'amjad ali')<br/>('40332145', 'Naveed Ahmad', 'naveed ahmad')</td><td></td></tr><tr><td>36c2db5ff76864d289781f93cbb3e6351f11984c</td><td>17th European Signal Processing Conference (EUSIPCO 2009) +<br/>Glasgow, Scotland, August 24-28, 2009 +<br/>ONE COLORED IMAGE BASED 2.5D HUMAN FACE RECONSTRUCTION +<br/>School of Electrical, Electronic and Computer Engineering +<br/><b>Newcastle University, Newcastle upon Tyne</b><br/>England, United Kingdom +</td><td>('1687577', 'Peng Liu', 'peng liu')</td><td>Email: peng.liu2@ncl.ac.uk, w.l.woo@ncl.ac.uk, s.s.dlay@ncl.ac.uk +</td></tr><tr><td>3661a34f302883c759b9fa2ce03de0c7173d2bb2</td><td>Peak-Piloted Deep Network for Facial Expression +<br/>Recognition +<br/><b>University of California, San Diego 2 Carnegie Mellon University</b><br/><b>AI Institute</b><br/><b>National University of Singapore</b><br/><b>Institute of Automation, Chinese Academy of Sciences</b></td><td>('8343585', 'Xiangyun Zhao', 'xiangyun zhao')<br/>('1776665', 'Luoqi Liu', 'luoqi liu')<br/>('1699559', 'Nuno Vasconcelos', 'nuno vasconcelos')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('1743598', 'Teng Li', 'teng li')</td><td>xiz019@ucsd.edu xdliang328@gmail.com liuluoqi@360.cn +<br/>tenglwy@gmail.com nvasconcelos@ucsd.edu eleyans@nus.edu.sg +</td></tr><tr><td>36c473fc0bf3cee5fdd49a13cf122de8be736977</td><td>Temporal Segment Networks: Towards Good +<br/>Practices for Deep Action Recognition +<br/>1Computer Vision Lab, ETH Zurich, Switzerland +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('1915826', 'Zhe Wang', 'zhe wang')<br/>('33427555', 'Yu Qiao', 'yu qiao')<br/>('1807606', 'Dahua Lin', 'dahua lin')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>368d59cf1733af511ed8abbcbeb4fb47afd4da1c</td><td>To Frontalize or Not To Frontalize: A Study of Face Pre-Processing Techniques +<br/>and Their Impact on Recognition +<br/>RichardWebster1, Vitomir ˇStruc2, Patrick J. Flynn1 and Walter J. Scheirer1 +<br/><b>University of Notre Dame, USA</b><br/><b>Faculty of Electrical Engineering, University of Ljubljana, Slovenia</b></td><td>('40061203', 'Sandipan Banerjee', 'sandipan banerjee')<br/>('6846673', 'Joel Brogan', 'joel brogan')<br/>('5014060', 'Aparna Bharati', 'aparna bharati')</td><td>{janez.krizaj, vitomir.struc}@fe.uni-lj.si +<br/>{sbanerj1, jbrogan4, abharati, brichar1, flynn, wscheire}@nd.edu +</td></tr><tr><td>366595171c9f4696ec5eef7c3686114fd3f116ad</td><td>Algorithms and Representations for Visual +<br/>Recognition +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2012-53 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2012/EECS-2012-53.html +<br/>May 1, 2012 +</td><td>('35208858', 'Subhransu Maji', 'subhransu maji')</td><td></td></tr><tr><td>36b9f46c12240898bafa10b0026a3fb5239f72f3</td><td>Collaborative Deep Reinforcement Learning for Joint Object Search +<br/><b>Peking University</b><br/>Microsoft Research +<br/><b>Peking University</b><br/>Microsoft Research +</td><td>('2045334', 'Xiangyu Kong', 'xiangyu kong')<br/>('1894653', 'Bo Xin', 'bo xin')<br/>('36637369', 'Yizhou Wang', 'yizhou wang')<br/>('1745420', 'Gang Hua', 'gang hua')</td><td>kong@pku.edu.cn +<br/>boxin@microsoft.com +<br/>yizhou.wang@pku.edu.cn +<br/>ganghua@microsoft.com +</td></tr><tr><td>361d6345919c2edc5c3ce49bb4915ed2b4ee49be</td><td><b>Delft University of Technology</b><br/>Models for supervised learning in sequence data +<br/>Pei, Wenjie +<br/>DOI +<br/>10.4233/uuid:fff15717-71ec-402d-96e6-773884659f2c +<br/>Publication date +<br/>2018 +<br/>Document Version +<br/>Publisher's PDF, also known as Version of record +<br/>Citation (APA) +<br/>Pei, W. (2018). Models for supervised learning in sequence data DOI: 10.4233/uuid:fff15717-71ec-402d- +<br/>96e6-773884659f2c +<br/>Important note +<br/>To cite this publication, please use the final published version (if applicable). +<br/>Please check the document version above. +<br/>Copyright +<br/>Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent +<br/>of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons. +<br/>Takedown policy +<br/>Please contact us and provide details if you believe this document breaches copyrights. +<br/>We will remove access to the work immediately and investigate your claim. +<br/><b>This work is downloaded from Delft University of Technology</b><br/>For technical reasons the number of authors shown on this cover page is limited to a maximum of 10. +<br/> </td><td></td><td></td></tr><tr><td>3634b4dd263c0f330245c086ce646c9bb748cd6b</td><td>Temporal Localization of Fine-Grained Actions in Videos +<br/>by Domain Transfer from Web Images +<br/><b>University of Southern California</b><br/><b>Google, Inc</b></td><td>('1726241', 'Chen Sun', 'chen sun')</td><td>{chensun,nevatia}@usc.edu +<br/>{sanketh,sukthankar}@google.com +</td></tr><tr><td>367a786cfe930455cd3f6bd2492c304d38f6f488</td><td>A Training Assistant Tool for the Automated Visual +<br/>Inspection System +<br/>A Thesis +<br/>Presented to +<br/>the Graduate School of +<br/><b>Clemson University</b><br/>In Partial Fulfillment +<br/>of the Requirements for the Degree +<br/>Master of Science +<br/>Electrical Engineering +<br/>by +<br/>December 2015 +<br/>Accepted by: +<br/>Dr. Adam W. Hoover, Committee Chair +<br/>Dr. Richard E. Groff +<br/>Dr. Yongqiang Wang +</td><td>('4154752', 'Mohan Karthik Ramaraj', 'mohan karthik ramaraj')</td><td></td></tr><tr><td>5c4ce36063dd3496a5926afd301e562899ff53ea</td><td></td><td></td><td></td></tr><tr><td>5c6de2d9f93b90034f07860ae485a2accf529285</td><td>Int. J. Biometrics, Vol. X, No. Y, xxxx +<br/>Compensating for pose and illumination in +<br/>unconstrained periocular biometrics +<br/>Department of Computer Science, +<br/>IT – Instituto de Telecomunicações, +<br/><b>University of Beira Interior</b><br/>6200-Covilhã, Portugal +<br/>Fax: +351-275-319899 +<br/>*Corresponding author +</td><td>('1678263', 'Chandrashekhar N. Padole', 'chandrashekhar n. padole')<br/>('1712429', 'Hugo Proença', 'hugo proença')</td><td>E-mail: chandupadole@ubi.pt +<br/>E-mail: hugomcp@di.ubi.pt +</td></tr><tr><td>5cbe1445d683d605b31377881ac8540e1d17adf0</td><td>On 3D Face Reconstruction via Cascaded Regression in Shape Space +<br/><b>College of Computer Science, Sichuan University, Chengdu, China</b></td><td>('50207647', 'Feng Liu', 'feng liu')<br/>('39422721', 'Dan Zeng', 'dan zeng')<br/>('1723081', 'Jing Li', 'jing li')<br/>('7345195', 'Qijun Zhao', 'qijun zhao')</td><td>qjzhao@scu.edu.cn +</td></tr><tr><td>5ca23ceb0636dfc34c114d4af7276a588e0e8dac</td><td>Texture Representation in AAM using Gabor Wavelet +<br/>and Local Binary Patterns +<br/>School of Electronic Engineering, +<br/><b>Xidian University</b><br/>Xi’an 710071, China +<br/>School of Computer Science and Information Systems, +<br/><b>Birkbeck College, University of London</b><br/>London WC1E 7HX, U.K. +<br/>School of Computer Engineering, +<br/><b>Nanyang Technological University</b><br/>50 Nanyang Avenue, Singapore 639798 +<br/>School of Electronic Engineering, +<br/><b>Xidian University</b><br/>Xi’an 710071, China +</td><td>('5452186', 'Ya Su', 'ya su')<br/>('1720243', 'Xuelong Li', 'xuelong li')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('10699750', 'Xinbo Gao', 'xinbo gao')</td><td>su1981ya@gmail.com +<br/>xuelong@dcs.bbk.ac.uk +<br/>dacheng.tao@gmail.com +<br/>xbgao@mail.xidian.edu.cn +</td></tr><tr><td>5c2a7518fb26a37139cebff76753d83e4da25159</td><td></td><td></td><td></td></tr><tr><td>5c493c42bfd93e4d08517438983e3af65e023a87</td><td>The Thirty-Second AAAI Conference +<br/>on Artificial Intelligence (AAAI-18) +<br/>Multimodal Keyless Attention +<br/>Fusion for Video Classification +<br/><b>Tsinghua University, 2Rutgers University, 3Baidu IDL</b></td><td>('1716690', 'Xiang Long', 'xiang long')<br/>('2551285', 'Chuang Gan', 'chuang gan')<br/>('1732213', 'Gerard de Melo', 'gerard de melo')<br/>('48033101', 'Xiao Liu', 'xiao liu')<br/>('48515099', 'Yandong Li', 'yandong li')<br/>('9921390', 'Fu Li', 'fu li')<br/>('35247507', 'Shilei Wen', 'shilei wen')</td><td>{longx13, ganc13}@mails.tsinghua.edu.cn, gdm@demelo.org, {liuxiao12, liyandong, lifu, wenshilei}@baidu.com +</td></tr><tr><td>5cb83eba8d265afd4eac49eb6b91cdae47def26d</td><td>Face Recognition with Local Line Binary Pattern +<br/><b>Mahanakorn University of Technology</b><br/>51 Cheum-Sampan Rd., Nong Chok, Bangkok, THAILAND 10530 +</td><td>('2337544', 'Amnart Petpon', 'amnart petpon')<br/>('1805935', 'Sanun Srisuk', 'sanun srisuk')</td><td>ta tee473@hotmail.com, sanun@mut.ac.th +</td></tr><tr><td>5c8672c0d2f28fd5d2d2c4b9818fcff43fb01a48</td><td>Robust Face Detection by Simple Means +<br/><b>Institute for Computer Graphics and Vision</b><br/><b>Graz University of Technology, Austria</b><br/>1 Motivation +<br/>Face detection is still one of the core problems in computer vision, especially in +<br/>unconstrained real-world situations where variations in face pose or bad imaging +<br/>conditions have to be handled. These problems are covered by recent benchmarks +<br/>such as Face Detection Dataset and Benchmark (FDDB) [2], which reveals that +<br/>established methods, e.g, Viola and Jones [8] suffer a drop in performance. More +<br/>effective approaches exist, but are closed source and not publicly available. Thus, +<br/>we propose a simple but effective detector that is available to the public. It +<br/>combines Histograms of Orientated Gradient (HOG) [1] features with linear +<br/>Support Vector Machine (SVM) classification. +<br/>2 Technical Details +<br/>One important aspect in the training of our face detector is bootstrapping. Thus, +<br/>we rely on iterative training. In particular, each iteration consists of first describ- +<br/>ing the face patches by HOGs [1] and then learning a linear SVM. At the end +<br/>of each iteration we bootstrap with the preliminary detector hard examples to +<br/>enrich the training set. We perform several bootstrapping rounds to improve the +<br/>detector until the desired false positive per window rate is reached. Interestingly, +<br/>we found out that picking up false positives at multiple scales in a sliding win- +<br/>dow fashion yields better results than just at a single scale. Testing several patch +<br/>sizes and HOG layouts revealed that a patch size of 36 by 36 delivers the best +<br/>results. For the HOG descriptor we ended up with a block size of 12x12, 4x4 for +<br/>the cells. Prior to the actual training we gathered face crops of the Annotated +<br/>facial landmarks in the wild (AFLW) dataset [4]. As AFLW includes the coarse +<br/>face pose we are able to retrieve about 28k frontal faces by limiting the yaw angle +<br/>between ± π +<br/>6 and mirroring them. For each face we crop a square region between +<br/>forehead and chin. The non-face patches are obtained by randomly sampling at +<br/>multiple scales of the PASCAL VOC 2007 dataset, excluding the persons subset. +<br/>3 Results +<br/>In Figure 1 we report the performance of our final detector on the challenging +<br/>FDDB benchmark compared to state-of-the-art methods. Despite the simplicity +<br/>of our detector it is able to improve considerably over the boosted classifier cas- +<br/>cade of Viola and Jones [8] and even outperforms the recent work of Jain and +</td><td>('3202367', 'Paul Wohlhart', 'paul wohlhart')<br/>('1791182', 'Peter M. Roth', 'peter m. roth')<br/>('3628150', 'Horst Bischof', 'horst bischof')</td><td>{koestinger,wohlhart,pmroth,bischof}@icg.tugraz.at +</td></tr><tr><td>5c3dce55c61ee86073575ac75cc882a215cb49e6</td><td>Neural Codes for Image Retrieval +<br/>Alexandr Chigorin1, and Victor Lempitsky2 +<br/>1 Yandex, Russia +<br/><b>Skolkovo Institute of Science and Technology (Skoltech), Russia</b><br/><b>Moscow Institute of Physics and Technology, Russia</b></td><td>('2412441', 'Artem Babenko', 'artem babenko')<br/>('32829387', 'Anton Slesarev', 'anton slesarev')</td><td></td></tr><tr><td>5c2e264d6ac253693469bd190f323622c457ca05</td><td>978-1-4799-2341-0/13/$31.00 ©2013 IEEE +<br/>4367 +<br/>ICIP 2013 +</td><td></td><td></td></tr><tr><td>5c473cfda1d7c384724fbb139dfe8cb39f79f626</td><td></td><td></td><td></td></tr><tr><td>5c820e47981d21c9dddde8d2f8020146e600368f</td><td>Extended Supervised Descent Method for +<br/>Robust Face Alignment +<br/><b>Beijing University of Posts and Telecommunications, Beijing, China</b></td><td>('9120475', 'Liu Liu', 'liu liu')<br/>('23224233', 'Jiani Hu', 'jiani hu')<br/>('1678529', 'Shuo Zhang', 'shuo zhang')<br/>('1774956', 'Weihong Deng', 'weihong deng')</td><td></td></tr><tr><td>5c5e1f367e8768a9fb0f1b2f9dbfa060a22e75c0</td><td>2132 +<br/>Reference Face Graph for Face Recognition +</td><td>('1784929', 'Mehran Kafai', 'mehran kafai')<br/>('39776603', 'Le An', 'le an')<br/>('1707159', 'Bir Bhanu', 'bir bhanu')</td><td></td></tr><tr><td>5c35ac04260e281141b3aaa7bbb147032c887f0c</td><td>Face Detection and Tracking Control with Omni Car +<br/>CS 231A Final Report +<br/>June 31, 2016 +</td><td>('2645488', 'Tung-Yu Wu', 'tung-yu wu')</td><td></td></tr><tr><td>5c435c4bc9c9667f968f891e207d241c3e45757a</td><td>RUIZ-HERNANDEZ, CROWLEY, LUX: HOW OLD ARE YOU? +<br/>"How old are you?" : Age Estimation with +<br/>Tensors of Binary Gaussian Receptive Maps +<br/>INRIA Grenoble Rhones-Alpes +<br/><b>Research Center and Laboratoire</b><br/>d’Informatique de Grenoble (LIG) +<br/>655 avenue de l’Europe +<br/>38 334 Saint Ismier Cedex, France +</td><td>('2291512', 'John A. Ruiz-Hernandez', 'john a. ruiz-hernandez')<br/>('34740185', 'James L. Crowley', 'james l. crowley')<br/>('2599357', 'Augustin Lux', 'augustin lux')</td><td>john-alexander.ruiz-hernandez@inrialpes.fr +<br/>james.crowley@inrialpes.fr +<br/>augustin.lux@inrialpes.fr +</td></tr><tr><td>5c7adde982efb24c3786fa2d1f65f40a64e2afbf</td><td>Ranking Domain-Specific Highlights +<br/>by Analyzing Edited Videos +<br/><b>University of Washington, Seattle, WA, USA</b></td><td>('1711801', 'Min Sun', 'min sun')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')</td><td></td></tr><tr><td>5c36d8bb0815fd4ff5daa8351df4a7e2d1b32934</td><td>GeePS: Scalable deep learning on distributed GPUs +<br/>with a GPU-specialized parameter server +<br/><b>Carnegie Mellon University</b></td><td>('1874200', 'Henggang Cui', 'henggang cui')<br/>('1682058', 'Hao Zhang', 'hao zhang')<br/>('1707164', 'Gregory R. Ganger', 'gregory r. ganger')<br/>('1974678', 'Phillip B. Gibbons', 'phillip b. gibbons')<br/>('1752601', 'Eric P. Xing', 'eric p. xing')</td><td></td></tr><tr><td>5cfbeae360398de9e20e4165485837bd42b93217</td><td>Cengil Emine, Cınars Ahmet, International Journal of Advance Research, Ideas and Innovations in Technology. +<br/>ISSN: 2454-132X +<br/>Impact factor: 4.295 +<br/>(Volume3, Issue5) +<br/>Available online at www.ijariit.com +<br/>Comparison Of Hog (Histogram of Oriented Gradients) and +<br/>Haar Cascade Algorithms with a Convolutional Neural Network +<br/>Based Face Detection Approaches +<br/>Computer Engineering Department +<br/><b>Firat University</b><br/>Computer Engineering Department +<br/><b>Firat University</b></td><td>('27758959', 'Emine Cengil', 'emine cengil')</td><td>ecengil@firat.edu.tr +<br/>acinar@firat.edu.tr +</td></tr><tr><td>5ca14fa73da37855bfa880b549483ee2aba26669</td><td>ISSN (e): 2250 – 3005 || Volume, 07 || Issue, 07|| June – 2017 || +<br/>International Journal of Computational Engineering Research (IJCER) +<br/>Face Recognition under Varying Illuminations Using Local +<br/>Binary Pattern And Local Ternary Pattern Fusion +<br/><b>Punjabi University Patiala</b><br/><b>Punjabi University Patiala</b></td><td>('2029759', 'Reecha Sharma', 'reecha sharma')</td><td></td></tr><tr><td>5c02bd53c0a6eb361972e8a4df60cdb30c6e3930</td><td>Multimedia stimuli databases usage patterns: a +<br/>survey report +<br/>M. Horvat1, S. Popović1 and K. Ćosić1 +<br/><b>University of Zagreb, Faculty of Electrical Engineering and Computing</b><br/>Department of Electric Machines, Drives and Automation +<br/>Zagreb, Croatia +</td><td></td><td>marko.horvat2@fer.hr +</td></tr><tr><td>5c8ae37d532c7bb8d7f00dfde84df4ba63f46297</td><td>DiscrimNet: Semi-Supervised Action Recognition from Videos using Generative +<br/>Adversarial Networks +<br/><b>Georgia Institute of Technology</b><br/>Google +<br/>Irfan Essa +<br/><b>Georgia Institute of Technology</b></td><td>('2308598', 'Unaiza Ahsan', 'unaiza ahsan')<br/>('1726241', 'Chen Sun', 'chen sun')</td><td>uahsan3@gatech.edu +<br/>chensun@google.com +<br/>irfan@gatech.edu +</td></tr><tr><td>5c717afc5a9a8ccb1767d87b79851de8d3016294</td><td>978-1-4673-0046-9/12/$26.00 ©2012 IEEE +<br/>1845 +<br/>ICASSP 2012 +</td><td></td><td></td></tr><tr><td>5ce2cb4c76b0cdffe135cf24b9cda7ae841c8d49</td><td>Facial Expression Intensity Estimation Using Ordinal Information +<br/><b>Computer and Systems Engineering, Rensselaer Polytechnic Institute</b><br/><b>School of Computer Science and Technology, University of Science and Technology of China</b></td><td>('1746803', 'Rui Zhao', 'rui zhao')<br/>('2316359', 'Quan Gan', 'quan gan')<br/>('1791319', 'Shangfei Wang', 'shangfei wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>1{zhaor,jiq}@rpi.edu, 2{gqquan@mail.,sfwang@}ustc.edu.cn +</td></tr><tr><td>5c4d4fd37e8c80ae95c00973531f34a6d810ea3a</td><td>The Open World of Micro-Videos +<br/><b>UC Irvine1, INRIA2, Carnegie Mellon University</b></td><td>('1879100', 'Phuc Xuan Nguyen', 'phuc xuan nguyen')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td></td></tr><tr><td>09b80d8eea809529b08a8b0ff3417950c048d474</td><td>Adding Unlabeled Samples to Categories by Learned Attributes +<br/><b>University of Maryland, College Park</b><br/><b>University of Washington</b></td><td>('3826759', 'Jonghyun Choi', 'jonghyun choi')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td>{jhchoi,mrastega,lsd}@umiacs.umd.edu +<br/>ali@cs.uw.edu +</td></tr><tr><td>09f58353e48780c707cf24a0074e4d353da18934</td><td>To appear in Proc. IEEE IJCB, 2014 +<br/>Unconstrained Face Recognition: Establishing Baseline +<br/>Human Performance via Crowdsourcing +<br/><b>Michigan State University, East Lansing, MI, U.S.A</b><br/><b>Cornell University, Ithaca, NY, U.S.A</b><br/>3Noblis, Falls Church, VA, U.S.A. +</td><td>('2180413', 'Lacey Best-Rowden', 'lacey best-rowden')<br/>('2339748', 'Shiwani Bisht', 'shiwani bisht')<br/>('2619953', 'Joshua C. Klontz', 'joshua c. klontz')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>bestrow1@cse.msu.edu;sb854@cornell.edu;joshua.klontz@noblis.org;jain@cse.msu.edu +</td></tr><tr><td>096eb8b4b977aaf274c271058feff14c99d46af3</td><td>REPORT DOCUMENTATION PAGE +<br/>Form Approved OMB NO. 0704-0188 +<br/>including +<br/>the +<br/>time +<br/>for reviewing +<br/>for +<br/>information, +<br/>for +<br/>this collection of +<br/>information +<br/>is estimated +<br/>to average 1 hour per response, +<br/>the data needed, and completing and reviewing +<br/>this collection of +<br/>instructions, +<br/>The public reporting burden +<br/> Send comments +<br/>searching existing data sources, gathering and maintaining +<br/>to Washington +<br/>regarding +<br/>this burden estimate or any other aspect of +<br/>Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA, 22202-4302. +<br/>Headquarters Services, Directorate +<br/>Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any oenalty for failing to comply with a collection of +<br/>information if it does not display a currently valid OMB control number. +<br/>PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. +<br/>1. REPORT DATE (DD-MM-YYYY) +<br/>05-10-2012 +<br/>4. TITLE AND SUBTITLE +<br/>Multi-observation visual recognition via joint dynamic sparse +<br/>representation +<br/>5a. CONTRACT NUMBER +<br/>W911NF-09-1-0383 +<br/>5b. GRANT NUMBER +<br/>2. REPORT TYPE +<br/>Conference Proceeding +<br/>3. DATES COVERED (From - To) +<br/>the collection of +<br/>reducing +<br/>for +<br/>information. +<br/>this burden, +<br/>including suggesstions +<br/>6. AUTHORS +<br/>Huang +<br/>7. PERFORMING ORGANIZATION NAMES AND ADDRESSES +<br/><b>William Marsh Rice University</b><br/>Office of Sponsored Research +<br/><b>William Marsh Rice University</b><br/>Houston, TX +<br/>9. SPONSORING/MONITORING AGENCY NAME(S) AND +<br/>ADDRESS(ES) +<br/>77005 - +<br/>U.S. Army Research Office +<br/> P.O. Box 12211 +<br/> Research Triangle Park, NC 27709-2211 +<br/>5c. PROGRAM ELEMENT NUMBER +<br/>611103 +<br/>5d. PROJECT NUMBER +<br/>5e. TASK NUMBER +<br/>5f. WORK UNIT NUMBER +<br/>8. PERFORMING ORGANIZATION REPORT +<br/>NUMBER +<br/>10. SPONSOR/MONITOR'S ACRONYM(S) +<br/> ARO +<br/>11. SPONSOR/MONITOR'S REPORT +<br/>NUMBER(S) +<br/>56177-CS-MUR.84 +<br/>12. DISTRIBUTION AVAILIBILITY STATEMENT +<br/>Approved for public release; distribution is unlimited. +<br/>13. SUPPLEMENTARY NOTES +<br/>The views, opinions and/or findings contained in this report are those of the author(s) and should not contrued as an official Department +<br/>of the Army position, policy or decision, unless so designated by other documentation. +</td><td>('40479011', 'Haichao Zhang', 'haichao zhang')<br/>('8147588', 'Nasser M. Nasrabadi', 'nasser m. nasrabadi')<br/>('1801395', 'Yanning Zhang', 'yanning zhang')</td><td></td></tr><tr><td>0952ac6ce94c98049d518d29c18d136b1f04b0c0</td><td></td><td></td><td></td></tr><tr><td>0969e0dc05fca21ff572ada75cb4b703c8212e80</td><td>Article +<br/>Semi-Supervised Classification Based on +<br/>Low Rank Representation +<br/><b>College of Computer and Information Science, Southwest University, Chongqing 400715, China</b><br/>Academic Editor: Javier Del Ser Lorente +<br/>Received: 1 June 2016; Accepted: 20 July 2016; Published: 22 July 2016 +</td><td>('40290479', 'Xuan Hou', 'xuan hou')<br/>('3439025', 'Guangjun Yao', 'guangjun yao')<br/>('40362316', 'Jun Wang', 'jun wang')</td><td>hx1995@email.swu.edu.cn (X.H.); guangjunyao@email.swu.edu.cn (G.Y.) +<br/>* Correspondence: kingjun@swu.edu.cn; Tel.: +86-23-6825-4396 +</td></tr><tr><td>09137e3c267a3414314d1e7e4b0e3a4cae801f45</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Two Birds with One Stone: Transforming and Generating +<br/>Facial Images with Iterative GAN +<br/>Received: date / Accepted: date +</td><td>('49626434', 'Dan Ma', 'dan ma')</td><td></td></tr><tr><td>09dd01e19b247a33162d71f07491781bdf4bfd00</td><td>Efficiently Scaling Up Video Annotation +<br/>with Crowdsourced Marketplaces +<br/>Department of Computer Science +<br/><b>University of California, Irvine, USA</b></td><td>('1856025', 'Carl Vondrick', 'carl vondrick')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td>{cvondric,dramanan,djp3}@ics.uci.edu +</td></tr><tr><td>09cf3f1764ab1029f3a7d57b70ae5d5954486d69</td><td>Comparison of ICA approaches for facial +<br/>expression recognition +<br/>I. Buciu 1,2 C. Kotropoulos 1 +<br/>I. Pitas 1 +<br/><b>Aristotle University of Thessaloniki</b><br/>GR-541 24, Thessaloniki, Box 451, Greece +<br/>2 Electronics Department +<br/>Faculty of Electrical Engineering and Information Technology +<br/><b>University of Oradea 410087, Universitatii 1, Romania</b><br/>August 18, 2008 +<br/>DRAFT +</td><td></td><td>costas,pitas@aiia.csd.auth.gr +<br/>ibuciu@uoradea.ro +</td></tr><tr><td>09fa54f1ab7aaa83124d2415bfc6eb51e4b1f081</td><td>Where to Buy It: Matching Street Clothing Photos in Online Shops +<br/><b>University of North Carolina at Chapel Hill</b><br/><b>University of Illinois at Urbana-Champaign</b></td><td>('1772294', 'M. Hadi Kiapour', 'm. hadi kiapour')<br/>('1682965', 'Xufeng Han', 'xufeng han')<br/>('1749609', 'Svetlana Lazebnik', 'svetlana lazebnik')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('1685538', 'Tamara L. Berg', 'tamara l. berg')</td><td>{hadi,xufeng,tlberg,aberg}@cs.unc.edu +<br/>slazebni@illinois.edu +</td></tr><tr><td>09926ed62511c340f4540b5bc53cf2480e8063f8</td><td>Action Tubelet Detector for Spatio-Temporal Action Localization +</td><td>('1881509', 'Vicky Kalogeiton', 'vicky kalogeiton')<br/>('2492127', 'Philippe Weinzaepfel', 'philippe weinzaepfel')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>0951f42abbf649bb564a21d4ff5dddf9a5ea54d9</td><td>Joint Estimation of Age and Gender from Unconstrained Face Images +<br/>using Lightweight Multi-task CNN for Mobile Applications +<br/><b>Institute of Information Science, Academia Sinica, Taipei</b></td><td>('1781429', 'Jia-Hong Lee', 'jia-hong lee')<br/>('2679814', 'Yi-Ming Chan', 'yi-ming chan')<br/>('2329177', 'Ting-Yen Chen', 'ting-yen chen')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')</td><td>{honghenry.lee, yiming, timh20022002, song}@iis.sinica.edu.tw +</td></tr><tr><td>09628e9116e7890bc65ebeabaaa5f607c9847bae</td><td>Semantically Consistent Regularization for Zero-Shot Recognition +<br/>Department of Electrical and Computer Engineering +<br/><b>University of California, San Diego</b></td><td>('1797523', 'Pedro Morgado', 'pedro morgado')<br/>('1699559', 'Nuno Vasconcelos', 'nuno vasconcelos')</td><td>{pmaravil,nuno}@ucsd.edu +</td></tr><tr><td>09733129161ca7d65cf56a7ad63c17f493386027</td><td>Face Recognition under Varying Illumination +<br/><b>Vienna University of Technology</b><br/>Inst. of Computer Graphics and +<br/>Algorithms +<br/>Vienna, Austria +<br/><b>Istanbul Technical University</b><br/>Department of Computer +<br/>Engineering +<br/>Istanbul, Turkey +<br/><b>Vienna University of Technology</b><br/>Inst. of Computer Graphics and +<br/>Algorithms +<br/>Vienna, Austria +</td><td>('1968256', 'Erald VUÇINI', 'erald vuçini')<br/>('1766445', 'Muhittin GÖKMEN', 'muhittin gökmen')<br/>('1725803', 'Eduard GRÖLLER', 'eduard gröller')</td><td>vucini@cg.tuwien.ac.at +<br/> gokmen@cs.itu.edu.tr +<br/>groeller@cg.tuwien.ac.at +</td></tr><tr><td>097340d3ac939ce181c829afb6b6faff946cdce0</td><td>Adding New Tasks to a Single Network with +<br/>Weight Transformations using Binary Masks +<br/><b>Sapienza University of Rome, 2Fondazione Bruno Kessler, 3University of Trento</b><br/><b>Italian Institute of Technology, 5Mapillary Research</b></td><td>('38286801', 'Massimiliano Mancini', 'massimiliano mancini')<br/>('40811261', 'Elisa Ricci', 'elisa ricci')<br/>('3033284', 'Barbara Caputo', 'barbara caputo')</td><td>{mancini,caputo}@diag.uniroma1.it,eliricci@fbk.eu,samuel@mapillary.com +</td></tr><tr><td>09507f1f1253101d04a975fc5600952eac868602</td><td>Motion Feature Network: Fixed Motion Filter +<br/>for Action Recognition +<br/><b>Seoul National University, Seoul, South Korea</b><br/>2 V.DO Inc., Suwon, Korea +</td><td>('2647624', 'Myunggi Lee', 'myunggi lee')<br/>('51151436', 'Seungeui Lee', 'seungeui lee')<br/>('51136389', 'Gyutae Park', 'gyutae park')<br/>('3160425', 'Nojun Kwak', 'nojun kwak')</td><td>{myunggi89, dehlix, sjson, pgt4861, nojunk}@snu.ac.kr +</td></tr><tr><td>09718bf335b926907ded5cb4c94784fd20e5ccd8</td><td>875 +<br/>Recognizing Partially Occluded, Expression Variant +<br/>Faces From Single Training Image per Person +<br/>With SOM and Soft k-NN Ensemble +</td><td>('2248421', 'Xiaoyang Tan', 'xiaoyang tan')<br/>('1680768', 'Songcan Chen', 'songcan chen')<br/>('1692625', 'Zhi-Hua Zhou', 'zhi-hua zhou')<br/>('2375371', 'Fuyan Zhang', 'fuyan zhang')</td><td></td></tr><tr><td>098a1ccc13b8d6409aa333c8a1079b2c9824705b</td><td>Attribute Pivots for Guiding Relevance Feedback in Image Search +<br/><b>The University of Texas at Austin</b></td><td>('1770205', 'Adriana Kovashka', 'adriana kovashka')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>{adriana, grauman}@cs.utexas.edu +</td></tr><tr><td>0903bb001c263e3c9a40f430116d1e629eaa616f</td><td>CVPR +<br/>#987 +<br/>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>045 +<br/>046 +<br/>047 +<br/>048 +<br/>049 +<br/>050 +<br/>051 +<br/>052 +<br/>053 +<br/>CVPR 2009 Submission #987. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +<br/>An Empirical Study of Context in Object Detection +<br/>Anonymous CVPR submission +<br/>Paper ID 987 +</td><td></td><td></td></tr><tr><td>090ff8f992dc71a1125636c1adffc0634155b450</td><td>Topic-aware Deep Auto-encoders (TDA) +<br/>for Face Alignment +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>CAS), Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b><br/><b>Imperial College London, London, UK</b></td><td>('1698586', 'Jie Zhang', 'jie zhang')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1874505', 'Xiaowei Zhao', 'xiaowei zhao')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>09b43b59879d59493df2a93c216746f2cf50f4ac</td><td>Deep Transfer Metric Learning +<br/><b>School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore. 2Advanced Digital Sciences Center, Singapore</b><br/>How to design a good similarity function plays an important role in many +<br/>visual recognition tasks. Recent advances have shown that learning a dis- +<br/>tance metric directly from a set of training examples can usually achieve +<br/>proposing performance than hand-crafted distance metrics [2, 3]. While +<br/>many metric learning algorithms have been presented in recent years, there +<br/>are still two shortcomings: 1) most of them usually seek a single linear dis- +<br/>tance to transform sample into a linear feature space, so that the nonlinear +<br/>relationship of samples cannot be well exploited. Even if the kernel trick +<br/>can be employed to addressed the nonlinearity issue, these methods still +<br/>suffer from the scalability problem because they cannot obtain the explicit +<br/>nonlinear mapping functions; 2) most of them assume that the training and +<br/>test samples are captured in similar scenarios so that their distributions are +<br/>assumed to be the same. This assumption doesn’t hold in many real visual +<br/>recognition applications, when samples are captured across datasets. +<br/>We propose a deep transfer metric learning (DTML) method for cross- +<br/>dataset visual recognition. Our method learns a set of hierarchical nonlinear +<br/>transformations by transferring discriminative knowledge from the labeled +<br/>source domain to the unlabeled target domain, under which the inter-class +<br/>variations are maximized and the intra-class variations are minimized, and +<br/>the distribution divergence between the source domain and the target do- +<br/>main at the top layer of the network is minimized, simultaneously. Figure 1 +<br/>illustrates the basic idea of the proposed method. +<br/>Figure 1: The basic idea of the proposed DTML method. For each sample +<br/>in the training sets from the source domain and the target domain, we pass +<br/>it to the developed deep neural network. We enforce two constraints on +<br/>the outputs of all training samples at the top of the network: 1) the inter- +<br/>class variations are maximized and the intra-class variations are minimized, +<br/>and 2) the distribution divergence between the source domain and the target +<br/>domain at the top layer of the network is minimized. +<br/>Deep Metric Learning. We construct a deep neural network to compute +<br/>the representations of each sample x. Assume there are M + 1 layers of the +<br/>network and p(m) units in the mth layer, where m = 1,2,··· ,M. The output +<br/>of x at the mth layer is computed as: +<br/>(cid:16) +<br/>W(m)h(m−1) + b(m)(cid:17) ∈ Rp(m) +<br/>(1) +<br/>f (m)(x) = h(m) = ϕ +<br/>where W(m) ∈ Rp(m)×p(m−1) and b(m) ∈ Rp(m) are the weight matrix and bias +<br/>of the parameters in this layer; and ϕ is a nonlinear activation function which +<br/>operates component-wisely, e.g., tanh or sigmoid functions. The nonlinear +<br/>mapping f (m) : Rd (cid:55)→ Rp(m) is a function parameterized by {W(i)}m +<br/>i=1 and +<br/>{b(i)}m +<br/>i=1. For the first layer, we assume h(0) = x. +<br/>For each pair of samples xi and x j, they can be finally represented as +<br/>f (m)(xi) and f (m)(x j) at the mth layer of our designed network, and their +<br/>distance metric can be measured by computing the squared Euclidean dis- +<br/>tance between f (m)(xi) and f (m)(x j) at the mth layer: +<br/>where Pi j is set as one if x j is one of k1-intra-class nearest neighbors of xi, +<br/>and zero otherwise; and Qi j is set as one if x j is one of k2-interclass nearest +<br/>neighbors of xi, and zero otherwise. +<br/>Deep Transfer Metric Learning. Given target domain data Xt and source +<br/>domain data Xs, their probability distributions are usually different in the o- +<br/>riginal feature space when they are captured from different datasets. To +<br/>reduce the distribution difference, we apply the Maximum Mean Discrep- +<br/>ancy (MMD) criterion [1] to measure their distribution difference at the mth +<br/>layer, which is defined as as follows: +<br/>ts (Xt ,Xs) = +<br/>D(m) +<br/>Nt ∑Nt +<br/>i=1 f (m)(xti)− 1 +<br/>Ns ∑Ns +<br/>i=1 f (m)(xsi) +<br/>(6) +<br/>By combining (3) and (6), we formulate DTML as the following opti- +<br/>mization problem: +<br/>(cid:13)(cid:13)(cid:13)(cid:13) 1 +<br/>(cid:13)(cid:13)(cid:13)(cid:13)2 +<br/>d2 +<br/>f (m) (xi,x j) = +<br/>(2) +<br/>min +<br/>f (M) +<br/>(cid:13)(cid:13)(cid:13) f (m)(xi)− f (m)(x j) +<br/>(cid:13)(cid:13)(cid:13)2 +<br/>(cid:16)(cid:13)(cid:13)W(m)(cid:13)(cid:13)2 +<br/>Following the graph embedding framework, we enforce the marginal +<br/>fisher analysis criterion [4] on the output of all training samples at the top +<br/>layer and formulate a strongly-supervised deep metric learning method: +<br/>F +(cid:13)(cid:13)b(m)(cid:13)(cid:13)2 +<br/>(cid:17) +<br/>(3) +<br/>J = S(M) +<br/>c − α S(M) +<br/>b + γ ∑M +<br/>m=1 +<br/>min +<br/>f (M) +<br/>where α (α > 0) is a free parameter which balances the important between +<br/>intra-class compactness and interclass separability; (cid:107)Z(cid:107)F denotes the Frobe- +<br/>nius norm of the matrix Z; γ (γ > 0) is a tunable positive regularization pa- +<br/>rameter; S(m) +<br/>define the intra-class compactness and the interclass +<br/>separability, which are defined as follows: +<br/>and S(m) +<br/>S(m) +<br/>c = +<br/>S(m) +<br/>b = +<br/>Nk1 +<br/>Nk2 +<br/>i=1∑N +<br/>∑N +<br/>i=1∑N +<br/>∑N +<br/>j=1 Pi j d2 +<br/>f (m) (xi,x j), +<br/>j=1 Qi j d2 +<br/>f (m) (xi,x j), +<br/>(4) +<br/>(5) +</td><td>('34651153', 'Junlin Hu', 'junlin hu')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('1689805', 'Yap-Peng Tan', 'yap-peng tan')</td><td></td></tr><tr><td>09df62fd17d3d833ea6b5a52a232fc052d4da3f5</td><td>ISSN: 1405-5546 +<br/>Instituto Politécnico Nacional +<br/>México +<br/> +<br/>Rivas Araiza, Edgar A.; Mendiola Santibañez, Jorge D.; Herrera Ruiz, Gilberto; González Gutiérrez, +<br/>Carlos A.; Trejo Perea, Mario; Ríos Moreno, G. J. +<br/>Mejora de Contraste y Compensación en Cambios de la Iluminación +<br/>Instituto Politécnico Nacional +<br/>Distrito Federal, México +<br/>Disponible en: http://www.redalyc.org/articulo.oa?id=61509703 +<br/> Cómo citar el artículo +<br/> Número completo +<br/> Más información del artículo +<br/> Página de la revista en redalyc.org +<br/>Sistema de Información Científica +<br/>Red de Revistas Científicas de América Latina, el Caribe, España y Portugal +<br/>Proyecto académico sin fines de lucro, desarrollado bajo la iniciativa de acceso abierto +</td><td></td><td>computacion-y-sistemas@cic.ipn.mx +</td></tr><tr><td>09b0ef3248ff8f1a05b8704a1b4cf64951575be9</td><td>Recognizing Activities of Daily Living with a Wrist-mounted Camera +<br/><b>Graduate School of Information Science and Technology, The University of Tokyo</b></td><td>('8197937', 'Katsunori Ohnishi', 'katsunori ohnishi')<br/>('2551640', 'Atsushi Kanehira', 'atsushi kanehira')<br/>('2554424', 'Asako Kanezaki', 'asako kanezaki')<br/>('1790553', 'Tatsuya Harada', 'tatsuya harada')</td><td>{ohnishi, kanehira, kanezaki, harada}@mi.t.u-tokyo.ac.jp +</td></tr><tr><td>097104fc731a15fad07479f4f2c4be2e071054a2</td><td></td><td></td><td></td></tr><tr><td>094357c1a2ba3fda22aa6dd9e496530d784e1721</td><td>A Unified Probabilistic Approach Modeling Relationships +<br/>between Attributes and Objects +<br/><b>Rensselaer Polytechnic Institute</b><br/>110 Eighth Street, Troy, NY USA 12180 +</td><td>('40066738', 'Xiaoyang Wang', 'xiaoyang wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>{wangx16,jiq}@rpi.edu +</td></tr><tr><td>09f853ce12f7361c4b50c494df7ce3b9fad1d221</td><td>myjournal manuscript No. +<br/>(will be inserted by the editor) +<br/>Random forests for real time 3D face analysis +<br/>Received: date / Accepted: date +</td><td>('3092828', 'Gabriele Fanelli', 'gabriele fanelli')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>09111da0aedb231c8484601444296c50ca0b5388</td><td></td><td></td><td></td></tr><tr><td>09750c9bbb074bbc4eb66586b20822d1812cdb20</td><td>978-1-4673-0046-9/12/$26.00 ©2012 IEEE +<br/>1385 +<br/>ICASSP 2012 +</td><td></td><td></td></tr><tr><td>09ce14b84af2dc2f76ae1cf227356fa0ba337d07</td><td>Face Reconstruction in the Wild +<br/><b>University of Washington</b><br/><b>University of Washington and Google Inc</b></td><td>('2419955', 'Ira Kemelmacher-Shlizerman', 'ira kemelmacher-shlizerman')<br/>('1679223', 'Steven M. Seitz', 'steven m. seitz')</td><td>kemelmi@cs.washington.edu +<br/>seitz@cs.washington.edu +</td></tr><tr><td>090e4713bcccff52dcd0c01169591affd2af7e76</td><td>What Do You Do? Occupation Recognition +<br/>in a Photo via Social Context +<br/><b>College of Computer and Information Science, Northeastern University, MA, USA</b><br/><b>Northeastern University, MA, USA</b></td><td>('2025056', 'Ming Shao', 'ming shao')<br/>('2897748', 'Liangyue Li', 'liangyue li')</td><td>mingshao@ccs.neu.edu, {liangyue, yunfu}@ece.neu.edu +</td></tr><tr><td>097f674aa9e91135151c480734dda54af5bc4240</td><td>Proc. VIIth Digital Image Computing: Techniques and Applications, Sun C., Talbot H., Ourselin S. and Adriaansen T. (Eds.), 10-12 Dec. 2003, Sydney +<br/>Face Recognition Based on Multiple Region Features +<br/>CSIRO Telecommunications & Industrial Physics +<br/>Australia +<br/>Tel: 612 9372 4104, Fax: 612 9372 4411, Email: +</td><td>('40833472', 'Jiaming Li', 'jiaming li')<br/>('1751724', 'Ying Guo', 'ying guo')<br/>('39877973', 'Rong-yu Qiao', 'rong-yu qiao')</td><td>jiaming.li@csiro.au +</td></tr><tr><td>5d485501f9c2030ab33f97972aa7585d3a0d59a7</td><td></td><td></td><td></td></tr><tr><td>5da740682f080a70a30dc46b0fc66616884463ec</td><td>Real-Time Head Pose Estimation Using +<br/>Multi-Variate RVM on Faces in the Wild +<br/>Augmented Vision Research Group, +<br/><b>German Research Center for Arti cial Intelligence (DFKI</b><br/>Tripstaddterstr. 122, 67663 Kaiserslautern, Germany +<br/><b>Technical University of Kaiserslautern</b><br/>http://www.av.dfki.de +</td><td>('2585383', 'Mohamed Selim', 'mohamed selim')<br/>('1771057', 'Alain Pagani', 'alain pagani')<br/>('1807169', 'Didier Stricker', 'didier stricker')</td><td>{mohamed.selim,alain.pagani,didier.stricker}@dfki.de +</td></tr><tr><td>5de5848dc3fc35e40420ffec70a407e4770e3a8d</td><td>WebVision Database: Visual Learning and Understanding from Web Data +<br/>1 Computer Vision Laboratory, ETH Zurich +<br/>2 Google Switzerland +</td><td>('1702619', 'Wen Li', 'wen li')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('1688012', 'Wei Li', 'wei li')<br/>('2794259', 'Eirikur Agustsson', 'eirikur agustsson')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>5da139fc43216c86d779938d1c219b950dd82a4c</td><td>1-4244-1437-7/07/$20.00 ©2007 IEEE +<br/>II - 205 +<br/>ICIP 2007 +</td><td></td><td></td></tr><tr><td>5dc056fe911a3e34a932513abe637076250d96da</td><td></td><td></td><td></td></tr><tr><td>5d185d82832acd430981ffed3de055db34e3c653</td><td>A Fuzzy Reasoning Model for Recognition +<br/>of Facial Expressions +<br/><b>Research Center CENTIA, Electronics and Mechatronics</b><br/>Universidad de las Américas, 72820, Puebla, Mexico +<br/>{oleg.starostenko; renan.contrerasgz; vicente.alarcon; leticia.florespo; +<br/><b>Engineering Institute, Autonomous University of Baja California, Blvd. Benito Ju rez</b><br/>Insurgentes Este, 21280, Mexicali, Baja California, Mexico +<br/>3 Universidad Politécnica de Baja California, Mexicali, Baja California, Mexico +</td><td>('1956337', 'Oleg Starostenko', 'oleg starostenko')<br/>('20083621', 'Renan Contreras', 'renan contreras')<br/>('1690236', 'Vicente Alarcón Aquino', 'vicente alarcón aquino')<br/>('2069473', 'Oleg Sergiyenko', 'oleg sergiyenko')</td><td>jorge.rodriguez}@udlap.mx +<br/>srgnk@iing.mxl.uabc.mx +<br/>vera-tyrsa@yandex.ru +</td></tr><tr><td>5d233e6f23b1c306cf62af49ce66faac2078f967</td><td>RESEARCH ARTICLE +<br/>Optimal Geometrical Set for Automated +<br/>Marker Placement to Virtualized Real-Time +<br/>Facial Emotions +<br/>School of Mechatronic Engineering, Universiti Malaysia Perlis, 02600, Ulu Pauh, Arau, Perlis, West Malaysia +</td><td>('6962924', 'Vasanthan Maruthapillai', 'vasanthan maruthapillai')<br/>('32588646', 'Murugappan Murugappan', 'murugappan murugappan')</td><td>* murugappan@unimap.edu.my +</td></tr><tr><td>5dd496e58cfedfc11b4b43c4ffe44ac72493bf55</td><td>Discriminative convolutional Fisher vector network for action recognition +<br/>School of Electrical Engineering and Computer Science +<br/><b>Queen Mary University of London</b><br/>London E1 4NS, United Kingdom +</td><td>('2685285', 'Petar Palasek', 'petar palasek')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')</td><td>p.palasek@qmul.ac.uk, i.patras@qmul.ac.uk +</td></tr><tr><td>5db075a308350c083c3fa6722af4c9765c4b8fef</td><td>The Novel Method of Moving Target Tracking Eyes +<br/>Location based on SIFT Feature Matching and Gabor +<br/>Wavelet Algorithm +<br/><b>College of Computer and Information Engineering, Nanyang Institute of Technology</b><br/>Henan Nanyang, 473004, China +<br/>* Tel.: 0086+13838972861 +<br/>Sensors & Transducers, Vol. 154, Issue 7, July 2013, pp. 129-137 +<br/> +<br/>SSSeeennnsssooorrrsss &&& TTTrrraaannnsssddduuuccceeerrrsss +<br/>© 2013 by IFSA +<br/>http://www.sensorsportal.com +<br/>Received: 28 April 2013 /Accepted: 19 July 2013 /Published: 31 July 2013 +</td><td>('2266189', 'Jing Zhang', 'jing zhang')<br/>('2732767', 'Caixia Yang', 'caixia yang')<br/>('1809507', 'Kecheng Liu', 'kecheng liu')</td><td>* E-mail: eduzhangjing@163.com +</td></tr><tr><td>5d7f8eb73b6a84eb1d27d1138965eb7aef7ba5cf</td><td>Robust Registration of Dynamic Facial Sequences +</td><td>('2046537', 'Evangelos Sariyanidi', 'evangelos sariyanidi')<br/>('1781916', 'Hatice Gunes', 'hatice gunes')<br/>('1713138', 'Andrea Cavallaro', 'andrea cavallaro')</td><td></td></tr><tr><td>5dcf78de4d3d867d0fd4a3105f0defae2234b9cb</td><td></td><td></td><td></td></tr><tr><td>5db4fe0ce9e9227042144758cf6c4c2de2042435</td><td>INTERNATIONAL JOURNAL OF ELECTRICAL AND ELECTRONIC SYSTEMS RESEARCH, VOL.3, JUNE 2010 +<br/>Recognition of Facial Expression Using Haar +<br/>Wavelet Transform +<br/>for +<br/>paper +<br/>features +<br/>investigates +<br/> +</td><td>('2254697', 'M. Satiyan', 'm. satiyan')</td><td></td></tr><tr><td>5d88702cdc879396b8b2cc674e233895de99666b</td><td>Exploiting Feature Hierarchies with Convolutional Neural Networks +<br/>for Cultural Event Recognition +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>School of Computer Science, Carnegie Mellon University, 15213, USA</b></td><td>('1730228', 'Mengyi Liu', 'mengyi liu')<br/>('1731144', 'Xin Liu', 'xin liu')<br/>('38751558', 'Yan Li', 'yan li')<br/>('1710220', 'Xilin Chen', 'xilin chen')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')</td><td>{mengyi.liu, xin.liu, yan.li}@vipl.ict.ac.cn, {xlchen, sgshan}@ict.ac.cn, alex@cs.cmu.edu +</td></tr><tr><td>5d5cd6fa5c41eb9d3d2bab3359b3e5eb60ae194e</td><td>Face Recognition Algorithms +<br/>June 16, 2010 +<br/>Ion Marqu´es +<br/>Supervisor: +<br/>Manuel Gra˜na +</td><td></td><td></td></tr><tr><td>5d09d5257139b563bd3149cfd5e6f9eae3c34776</td><td>Optics Communications 338 (2015) 77–89 +<br/>Contents lists available at ScienceDirect +<br/>Optics Communications +<br/>journal homepage: www.elsevier.com/locate/optcom +<br/>Pattern recognition with composite correlation filters designed with +<br/>multi-objective combinatorial optimization +<br/>a Instituto Politécnico Nacional – CITEDI, Ave. del Parque 1310, Mesade Otay, Tijuana B.C. 22510, México +<br/>b Department of Computer Science, CICESE, Carretera Ensenada-Tijuana 3918, Ensenada B.C. 22860, México +<br/>c Instituto Tecnológico de Tijuana, Blvd. Industrial y Ave. ITR TijuanaS/N, Mesa de Otay, Tijuana B.C. 22500, México +<br/>d National Ignition Facility, Lawrence Livermore National Laboratory, Livermore, CA 94551, USA +<br/>a r t i c l e i n f o +<br/>a b s t r a c t +<br/>Article history: +<br/>Received 12 July 2014 +<br/>Accepted 16 November 2014 +<br/>Available online 23 October 2014 +<br/>Keywords: +<br/>Object recognition +<br/>Composite correlation filters +<br/>Multi-objective evolutionary algorithm +<br/>Combinatorial optimization +<br/>Composite correlation filters are used for solving a wide variety of pattern recognition problems. These +<br/>filters are given by a combination of several training templates chosen by a designer in an ad hoc manner. +<br/>In this work, we present a new approach for the design of composite filters based on multi-objective +<br/>combinatorial optimization. Given a vast search space of training templates, an iterative algorithm is used +<br/>to synthesize a filter with an optimized performance in terms of several competing criteria. Moreover, by +<br/>employing a suggested binary-search procedure a filter bank with a minimum number of filters can be +<br/>constructed, for a prespecified trade-off of performance metrics. Computer simulation results obtained +<br/>with the proposed method in recognizing geometrically distorted versions of a target in cluttered and +<br/>noisy scenes are discussed and compared in terms of recognition performance and complexity with +<br/>existing state-of-the-art filters. +<br/>& Elsevier B.V. All rights reserved. +<br/>1. +<br/>Introduction +<br/>Nowadays, object recognition receives much research interest +<br/>due to its high impact in real-life activities, such as robotics, bio- +<br/>metrics, and target tracking [1,2]. Object recognition consists in +<br/>solving two essential tasks: detection of a target within an ob- +<br/>served scene and determination of the exact position of the de- +<br/>tected object. Different approaches can be utilized to address these +<br/>tasks, that is feature-based methods [3–6] and template matching +<br/>algorithms [7,8]. In feature-based methods the observed scene is +<br/>processed to extract relevant features of potential targets within +<br/>the scene. Next, the extracted features are processed and analyzed +<br/>to make decisions. Feature-based methods yield good results in +<br/>many applications. However, they depend on several subjective +<br/>decisions which often require optimization [9,10]. On the other +<br/>hand, correlation filtering is a template matching processing. In +<br/>this approach, the coordinates of the maximum of the filter output +<br/>are taken as estimates of the target coordinates in the observed +<br/>scene. Correlation filters possess a good mathematical basis and +<br/>they can be implemented by exploiting massive parallelism either +<br/>in hybrid opto-digital correlators [11,12] or in high-performance +<br/>n Corresponding author. Tel.: þ52 664 623 1344x82856. +<br/>http://dx.doi.org/10.1016/j.optcom.2014.10.038 +<br/>0030-4018/& Elsevier B.V. All rights reserved. +<br/>hardware such as graphics processing units (GPUs) [13] or field +<br/>programmable gate arrays (FPGAs) [14] at high rate. Additionally, +<br/>these filters are capable to reliably recognize a target in highly +<br/>cluttered and noisy environments [8,15,16]. Moreover, they are +<br/>able to estimate very accurately the position of the target within +<br/>the scene [17]. Correlation filters are usually designed by a opti- +<br/>mization of various criteria [18,19]. The filters can be broadly +<br/>classified in to two main categories: analytical and composite fil- +<br/>ters. Analytical filters optimize a performance criterion using +<br/>mathematical models of signals and noise [20,21]. Composite fil- +<br/>ters are constructed by combination of several training templates, +<br/>each of them representing an expected target view in the observed +<br/>scene [22,21]. In practice, composite filters are effective for real- +<br/>life degradations of targets such as rotations and scaling. Compo- +<br/>site filters are synthesized by means of a supervised training +<br/>process. Thus, the performance of the filters highly depends on a +<br/>proper selection of image templates used for training [20,23]. +<br/>Normally, the training templates are chosen by a designer in an ad +<br/>hoc manner. Such a subjective procedure is not optimal. In addi- +<br/>tion, Kumar and Pochavsky [24] showed that the signal to noise +<br/>ratio of a composite filter gradually reduces when the number of +<br/>training templates increases. In order to synthesize composite +<br/>filters with improved performance in terms of several competing +<br/>metrics, a search and optimization strategy is required to auto- +<br/>matically choose the set of training templates. +</td><td>('1908859', 'Victor H. Diaz-Ramirez', 'victor h. diaz-ramirez')<br/>('14245397', 'Andres Cuevas', 'andres cuevas')<br/>('1684262', 'Vitaly Kober', 'vitaly kober')<br/>('2166904', 'Leonardo Trujillo', 'leonardo trujillo')<br/>('37615801', 'Abdul Awwal', 'abdul awwal')</td><td>E-mail address: vdiazr@ipn.mx (V.H. Diaz-Ramirez). +</td></tr><tr><td>5d479f77ecccfac9f47d91544fd67df642dfab3c</td><td>Linking People in Videos with “Their” Names +<br/>Using Coreference Resolution +<br/><b>Stanford University, USA</b><br/><b>Stanford University, USA</b></td><td>('34066479', 'Vignesh Ramanathan', 'vignesh ramanathan')<br/>('2319608', 'Armand Joulin', 'armand joulin')<br/>('40085065', 'Percy Liang', 'percy liang')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td>{vigneshr,ajoulin,pliang,feifeili}@cs.stanford.edu +</td></tr><tr><td>5d01283474b73a46d80745ad0cc0c4da14aae194</td><td></td><td></td><td></td></tr><tr><td>5d197c8cd34473eb6cde6b65ced1be82a3a1ed14</td><td><b>AFaceImageDatabaseforEvaluatingOut-of-FocusBlurQiHan,QiongLiandXiamuNiuHarbinInstituteofTechnologyChina1.IntroductionFacerecognitionisoneofthemostpopularresearchfieldsofcomputervisionandmachinelearning(Tores(2004);Zhaoetal.(2003)).Alongwithinvestigationoffacerecognitionalgorithmsandsystems,manyfaceimagedatabaseshavebeencollected(Gross(2005)).Facedatabasesareimportantfortheadvancementoftheresearchfield.Becauseofthenonrigidityandcomplex3Dstructureofface,manyfactorsinfluencetheperformanceoffacedetectionandrecognitionalgorithmssuchaspose,expression,age,brightness,contrast,noise,blurandetc.Someearlyfacedatabasesgatheredunderstrictlycontrolledenvironment(Belhumeuretal.(1997);Samaria&Harter(1994);Turk&Pentland(1991))onlyallowslightexpressionvariation.Toinvestigatetherelationshipsbetweenalgorithms’performanceandtheabovefactors,morefacedatabaseswithlargerscaleandvariouscharacterswerebuiltinthepastyears(Bailly-Bailliereetal.(2003);Flynnetal.(2003);Gaoetal.(2008);Georghiadesetal.(2001);Hallinan(1995);Phillipsetal.(2000);Simetal.(2003)).Forinstance,The"CAS-PEAL","FERET","CMUPIE",and"YaleB"databasesincludevariousposes(Gaoetal.(2008);Georghiadesetal.(2001);Phillipsetal.(2000);Simetal.(2003));The"HarvardRL","CMUPIE"and"YaleB"databasesinvolvemorethan40differentconditionsinillumination(Georghiadesetal.(2001);Hallinan(1995);Simetal.(2003));Andthe"BANCA",and"NDHID"databasescontainover10timesgathering(Bailly-Bailliereetal.(2003);Flynnetal.(2003)).Thesedatabaseshelpresearcherstoevaluateandimprovetheiralgorithmsaboutfacedetection,recognition,andotherpurposes.Blurisnotthemostimportantbutstillanotablefactoraffectingtheperformanceofabiometricsystem(Fronthaleretal.(2006);Zamanietal.(2007)).Themainreasonsleadingblurconsistinout-of-focusofcameraandmotionofobject,andtheout-of-focusblurismoresignificantintheapplicationenvironmentoffacerecognition(Eskicioglu&Fisher(1995);Kimetal.(1998);Tanakaetal.(2007);Yitzhaky&Kopeika(1996)).Toinvestigatetheinfluenceofbluronafacerecognitionsystem,afaceimagedatabasewithdifferentconditionsofclarityandefficientblurevaluatingalgorithmsareneeded.Thischapterintroducesanewfacedatabasebuiltforthepurposeofblurevaluation.Theapplicationenvironmentsoffacerecognitionareanalyzedfirstly,thenaimagegatheringschemeisdesigned.Twotypicalgatheringfacilitiesareusedandthefocusstatusaredividedinto11steps.Further,theblurassessmentalgorithmsaresummarizedandthecomparisonbetweenthemisraisedonthevarious-claritydatabase.The7www.intechopen.com</b></td><td></td><td></td></tr><tr><td>5da2ae30e5ee22d00f87ebba8cd44a6d55c6855e</td><td><b>This is an Open Access document downloaded from ORCA, Cardiff University's institutional</b><br/>repository: http://orca.cf.ac.uk/111659/ +<br/>This is the author’s version of a work that was submitted to / accepted for publication. +<br/>Citation for final published version: +<br/>Krumhuber, Eva G, Lai, Yukun, Rosin, Paul and Hugenberg, Kurt 2018. When facial expressions +<br/>Publishers page: +<br/>Please note: +<br/>Changes made as a result of publishing processes such as copy-editing, formatting and page +<br/>numbers may not be reflected in this version. For the definitive version of this publication, please +<br/>refer to the published source. You are advised to consult the publisher’s version if you wish to cite +<br/>this paper. +<br/>This version is being made available in accordance with publisher policies. See +<br/>http://orca.cf.ac.uk/policies.html for usage policies. Copyright and moral rights for publications +<br/>made available in ORCA are retained by the copyright holders. +</td><td></td><td></td></tr><tr><td>5df376748fe5ccd87a724ef31d4fdb579dab693f</td><td>A Dashboard for Affective E-learning: +<br/>Data Visualization for Monitoring Online Learner Emotions +<br/>School of Computer Science +<br/><b>Carleton University</b><br/>Canada +</td><td>('2625368', 'Reza GhasemAghaei', 'reza ghasemaghaei')<br/>('40230630', 'Ali Arya', 'ali arya')<br/>('8547603', 'Robert Biddle', 'robert biddle')</td><td>Reza.GhasemAghaei@carleton.ca +</td></tr><tr><td>31aa20911cc7a2b556e7d273f0bdd5a2f0671e0a</td><td></td><td></td><td></td></tr><tr><td>31b05f65405534a696a847dd19c621b7b8588263</td><td></td><td></td><td></td></tr><tr><td>31625522950e82ad4dffef7ed0df00fdd2401436</td><td>Motion Representation with Acceleration Images +<br/><b>National Institute of Advanced Industrial Science and Technology (AIST</b><br/>Tsukuba, Ibaraki, Japan +</td><td>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')<br/>('1713046', 'Yun He', 'yun he')<br/>('3393640', 'Soma Shirakabe', 'soma shirakabe')<br/>('1732705', 'Yutaka Satoh', 'yutaka satoh')</td><td>{hirokatsu.kataoka, yun.he, shirakabe-s, yu.satou}@aist.go.jp +</td></tr><tr><td>3167f415a861f19747ab5e749e78000179d685bc</td><td>RankBoost with l1 regularization for Facial Expression Recognition and +<br/>Intensity Estimation +<br/><b>Rutgers University, Piscataway NJ 08854, USA</b><br/>2National Laboratory of Pattern Recognition, Chinese Academy of Sciences Beijing, 100080, China +</td><td>('39606160', 'Peng Yang', 'peng yang')<br/>('1734954', 'Qingshan Liu', 'qingshan liu')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td></td></tr><tr><td>3107316f243233d45e3c7e5972517d1ed4991f91</td><td>CVAE-GAN: Fine-Grained Image Generation through Asymmetric Training +<br/><b>University of Science and Technology of China</b><br/>2Microsoft Research Asia, +</td><td>('3093568', 'Jianmin Bao', 'jianmin bao')<br/>('39447786', 'Dong Chen', 'dong chen')<br/>('1716835', 'Fang Wen', 'fang wen')<br/>('7179232', 'Houqiang Li', 'houqiang li')<br/>('1745420', 'Gang Hua', 'gang hua')</td><td>jmbao@mail.ustc.edu.cn, lihq@ustc.edu.cn +<br/>{doch,fangwen,ganghua}@microsoft.com +</td></tr><tr><td>318e7e6daa0a799c83a9fdf7dd6bc0b3e89ab24a</td><td>Sparsity in Dynamics of Spontaneous +<br/>Subtle Emotions: Analysis & Application +</td><td>('35256518', 'Anh Cat Le Ngo', 'anh cat le ngo')<br/>('2339975', 'John See', 'john see')<br/>('6633183', 'Raphael C.-W. Phan', 'raphael c.-w. phan')</td><td></td></tr><tr><td>31c0968fb5f587918f1c49bf7fa51453b3e89cf7</td><td>Deep Transfer Learning for Person Re-identification +</td><td>('3447059', 'Mengyue Geng', 'mengyue geng')<br/>('5765799', 'Yaowei Wang', 'yaowei wang')<br/>('1700927', 'Tao Xiang', 'tao xiang')<br/>('1705972', 'Yonghong Tian', 'yonghong tian')</td><td></td></tr><tr><td>313d5eba97fe064bdc1f00b7587a4b3543ef712a</td><td>Compact Deep Aggregation for Set Retrieval +<br/><b>Visual Geometry Group, University of Oxford, UK</b><br/>2 DeepMind +</td><td>('6730372', 'Yujie Zhong', 'yujie zhong')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>{yujie,az}@robots.ox.ac.uk +<br/>relja@google.com +</td></tr><tr><td>31e57fa83ac60c03d884774d2b515813493977b9</td><td></td><td></td><td></td></tr><tr><td>3137a3fedf23717c411483c7b4bd2ed646258401</td><td>Joint Learning of Discriminative Prototypes +<br/>and Large Margin Nearest Neighbor Classifiers +<br/><b>Institute for Computer Graphics and Vision, Graz University of Technology</b></td><td>('3202367', 'Paul Wohlhart', 'paul wohlhart')<br/>('1791182', 'Peter M. Roth', 'peter m. roth')<br/>('3628150', 'Horst Bischof', 'horst bischof')</td><td>{koestinger,wohlhart,pmroth,bischof}@icg.tugraz.at +</td></tr><tr><td>31c34a5b42a640b824fa4e3d6187e3675226143e</td><td>Shape and Texture based Facial Action and Emotion +<br/>Recognition +<br/>(Demonstration) +<br/>Department of Computer Science and Digital Technologies +<br/><b>Northumbria University</b><br/>Newcastle, NE1 8ST, UK +</td><td>('1712838', 'Li Zhang', 'li zhang')<br/>('2801063', 'Kamlesh Mistry', 'kamlesh mistry')</td><td>{li.zhang, kamlesh.mistry, alamgir.hossain}@northumbria.ac.uk +</td></tr><tr><td>316e67550fbf0ba54f103b5924e6537712f06bee</td><td>Multimodal semi-supervised learning +<br/>for image classification +<br/>LEAR team, INRIA Grenoble, France +</td><td>('2737253', 'Matthieu Guillaumin', 'matthieu guillaumin')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>31ef5419e026ef57ff20de537d82fe3cfa9ee741</td><td>Facial Expression Analysis Based on +<br/>High Dimensional Binary Features +<br/>´Ecole Polytechique de Montr´eal, Universit´e de Montr´eal, Montr´eal, Canada +</td><td>('3127597', 'Samira Ebrahimi Kahou', 'samira ebrahimi kahou')<br/>('2558801', 'Pierre Froumenty', 'pierre froumenty')</td><td>{samira.ebrahimi-kahou, pierre.froumenty, christopher.pal}@polymtl.ca +</td></tr><tr><td>31ea88f29e7f01a9801648d808f90862e066f9ea</td><td>Published as a conference paper at ICLR 2017 +<br/>DEEP MULTI-TASK REPRESENTATION LEARNING: +<br/>A TENSOR FACTORISATION APPROACH +<br/><b>Queen Mary, University of London</b></td><td>('2653152', 'Yongxin Yang', 'yongxin yang')<br/>('1697755', 'Timothy M. Hospedales', 'timothy m. hospedales')</td><td>{yongxin.yang, t.hospedales}@qmul.ac.uk +</td></tr><tr><td>3176ee88d1bb137d0b561ee63edf10876f805cf0</td><td>Recombinator Networks: Learning Coarse-to-Fine Feature Aggregation +<br/><b>University of Montreal, 2Cornell University, 3Ecole Polytechnique of Montreal, 4CIFAR</b></td><td>('25056820', 'Sina Honari', 'sina honari')<br/>('2965424', 'Jason Yosinski', 'jason yosinski')<br/>('1707326', 'Pascal Vincent', 'pascal vincent')</td><td>1{honaris, vincentp}@iro.umontreal.ca, 2yosinski@cs.cornell.edu, 3christopher.pal@polymtl.ca +</td></tr><tr><td>31b58ced31f22eab10bd3ee2d9174e7c14c27c01</td><td></td><td></td><td></td></tr><tr><td>31835472821c7e3090abb42e57c38f7043dc3636</td><td>Flow Counting Using Realboosted +<br/>Multi-sized Window Detectors +<br/><b>Lund University, Cognimatics AB</b></td><td>('38481779', 'Mikael Nilsson', 'mikael nilsson')<br/>('3181258', 'Rikard Berthilsson', 'rikard berthilsson')</td><td></td></tr><tr><td>312b2566e315dd6e65bd42cfcbe4d919159de8a1</td><td>An Accurate Algorithm for Generating a Music Playlist +<br/>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 100– No.9, August 2014 +<br/>based on Facial Expressions +<br/>Computer Science and Engineering Department +<br/>Amity School of Engineering & Technology, +<br/><b>Amity University, Noida, India</b></td><td></td><td></td></tr><tr><td>3152e89963b8a4028c4abf6e1dc19e91c4c5a8f4</td><td>Exploring Stereotypes and Biased Data with the Crowd +<br/>Department of Computer Science +<br/><b>The University of Texas at Austin</b><br/>Department of Computer Science +<br/><b>The University of Texas at Austin</b><br/>Introduction +<br/>In 2016, Baidu and Google spent somewhere between +<br/>twenty and thirty billion dollars developing and acquir- +<br/>ing artificial intelligence and machine learning technolo- +<br/><b>gies (Bughin et al. 2017). A range of other sectors, includ</b><br/>ing health care, education, and manufacturing, are also pre- +<br/>dicted to adopt these technologies at increasing rates. Ma- +<br/>chine learning and AI are proven to have the capacity to +<br/>greatly improve lives and spur innovation. However, as soci- +<br/>ety becomes increasingly dependent on these technologies, +<br/>it is crucial that we acknowledge some of the dangers, in- +<br/>cluding the capacity for these algorithms to absorb and am- +<br/>plify harmful cultural biases. +<br/>Algorithms are often praised for their objectivity, but ma- +<br/>chine learning algorithms have increasingly made news for a +<br/>number of problematic outcomes, ranging from Google Pho- +<br/>the judicial system using algorithms that are biased against +<br/>African Americans (Dougherty 2015; Angwin et al. 2016). +<br/>These harmful outcomes can be traced back to the data that +<br/>was used to train the models. +<br/>Machine learning applications put a heavy premium on +<br/>data quantity. Research communities generally believe that +<br/>the more training data there is, the better the learning out- +<br/>come of the models will be (Halevy, Norvig, and Pereira +<br/>2009). This has led to large scale data collection. How- +<br/>ever, unless extra care is taken by the researchers, these +<br/>large data sets will often contain bias that can profoundly +<br/>change the learning outcome. Even minimal bias within +<br/>a data set can end up being amplified by machine learn- +<br/>ing models, leading to skewed results. Researchers have +<br/>found that widely used image data sets imSitu and MS- +<br/>COCO, along with textual data sets mined from Google +<br/>News, contain significant gender bias (Zhao et al. 2017; +<br/>Bolukbasi et al. 2016). This research also found that train- +<br/>ing models with this data amplified the bias in the final out- +<br/>comes. +<br/>Once these algorithms have been improperly trained they +<br/>can then be implemented into feedback loops where systems +<br/>“define their own reality and use it to justify their results” as +<br/>Copyright c(cid:13) 2018 is held by the authors. Copies may be freely +<br/>made and distributed by others. Presented at the 2016 AAAI Con- +<br/>ference on Human Computation and Crowdsourcing (HCOMP). +<br/>Cathy O’Neil describes in her book Weapons of Math De- +<br/>struction. O’Neil discusses problematic systems like Pred- +<br/>Pol, a program that predicts where crimes are most likely to +<br/>occur based on past crime reports, which may unfairly target +<br/>poor communities. +<br/>It therefore becomes necessary to consider the bias that +<br/>may be introduced as a data set is being collected and to +<br/>attempt to prevent that bias from being absorbed by an al- +<br/>gorithm. We propose using the crowd to help uncover what +<br/>bias may reside in a specific data set. +<br/>The crowd has potential to be useful for this task. One +<br/>of the key difficulties in preventing bias is knowing what +<br/>to look for. The varied demographics of crowd workers pro- +<br/>vide an extended range of perspectives that can help uncover +<br/>stereotypes that may go unnoticed by a small group of re- +<br/>searchers. Some work has already been conducted in this +<br/>area, and Bolukbasi et al. (2016) found that the crowd was +<br/>useful in determining the level of stereotype associated with +<br/>ased words by asking the crowd to rate analogies such as +<br/>“she is to sewing as he is to carpentry”. We want to extend +<br/><b>our analysis to stereotypes beyond gender, including those</b><br/>surrounding race and class. +<br/>The goal of our research is to contribute information about +<br/>how useful the crowd is at anticipating stereotypes that may +<br/>be biasing a data set without a researcher’s knowledge. The +<br/>results of the crowd’s prediction can potentially be used dur- +<br/>ing data collection to help prevent the suspected stereotypes +<br/>from introducing bias to the dataset. We conduct our re- +<br/>search by asking the crowd on Amazon’s Mechanical Turk +<br/>(AMT) to complete two similar Human Intelligence Tasks +<br/>(HITs) by suggesting stereotypes relating to their personal +<br/>experience. Our analysis of these responses focuses on de- +<br/>termining the level of diversity in the workers’ suggestions +<br/>and their demographics. Through this process we begin a +<br/>discussion on how useful the crowd can be in tackling this +<br/>difficult problem within machine learning data collection. +<br/>2 Related Work +<br/>2.1 Work on bias in data sets and amplification +<br/>As biased data sets get more coverage in the news, an in- +<br/>creasing amount of research has been conducted around de- +<br/>termining if data sets are biased and trying to mitigate the +</td><td>('32193161', 'Zeyuan Hu', 'zeyuan hu')<br/>('40410119', 'Julia Strout', 'julia strout')</td><td>iamzeyuanhu@utexas.edu +<br/>jstrout@utexas.edu +</td></tr><tr><td>31ace8c9d0e4550a233b904a0e2aabefcc90b0e3</td><td>Learning Deep Face Representation +<br/>Megvii Inc. +<br/>Megvii Inc. +<br/>Megvii Inc. +<br/>Megvii Inc. +<br/>Megvii Inc. +</td><td>('1934546', 'Haoqiang Fan', 'haoqiang fan')<br/>('2695115', 'Zhimin Cao', 'zhimin cao')<br/>('1691963', 'Yuning Jiang', 'yuning jiang')<br/>('2274228', 'Qi Yin', 'qi yin')<br/>('2479859', 'Chinchilla Doudou', 'chinchilla doudou')</td><td>fhq@megvii.com +<br/>czm@megvii.com +<br/>jyn@megvii.com +<br/>yq@megvii.com +<br/>doudou@megvii.com +</td></tr><tr><td>316d51aaa37891d730ffded7b9d42946abea837f</td><td>CBMM Memo No. 23 +<br/>April 27, 2015 +<br/>Unsupervised learning of clutter-resistant visual +<br/>representations from natural videos +<br/>by +<br/><b>MIT, McGovern Institute, Center for Brains, Minds and Machines</b></td><td>('1694846', 'Qianli Liao', 'qianli liao')</td><td></td></tr><tr><td>31afdb6fa95ded37e5871587df38976fdb8c0d67</td><td>QUANTIZED FUZZY LBP FOR FACE RECOGNITION +<br/>Jianfeng +<br/>Ren +<br/>Junsong +<br/>Yuan +<br/>BeingThere +<br/>Centre +<br/><b>Institute</b><br/>of Media Innovation +<br/>Nanyang +<br/>50 Nanyang +<br/>Technological +<br/>Singapore +<br/>Drive, +<br/>637553. +<br/><b>University</b><br/>School of Electrical +<br/>& Electronics +<br/>Engineering +<br/>Nanyang +<br/>50 Nanyang +<br/>Technological +<br/>Singapore +<br/>Avenue, +<br/>639798 +<br/><b>University</b></td><td>('3307580', 'Xudong Jiang', 'xudong jiang')</td><td></td></tr><tr><td>31d60b2af2c0e172c1a6a124718e99075818c408</td><td>Robust Facial Expression Recognition using Near Infrared Cameras +<br/>Paper: jc*-**-**-**** +<br/>Robust Facial Expression Recognition using Near Infrared +<br/>Cameras +<br/><b>The University of Tokyo</b><br/><b>Electronics and Communication Engineering, Chuo University</b><br/>[Received 00/00/00; accepted 00/00/00] +</td><td>('34415055', 'Hideki Hashimoto', 'hideki hashimoto')<br/>('9181040', 'Takashi Kubota', 'takashi kubota')</td><td></td></tr><tr><td>31f1e711fcf82c855f27396f181bf5e565a2f58d</td><td>Unconstrained Age Estimation with Deep Convolutional Neural Networks +<br/>Jun Cheng Chen1 +<br/><b>University of Maryland</b><br/>2Montgomery Blair High School +<br/><b>Rutgers University</b></td><td>('26988560', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('2349530', 'Sabrina Zhou', 'sabrina zhou')<br/>('40080979', 'Amit Kumar', 'amit kumar')<br/>('2943431', 'Azadeh Alavi', 'azadeh alavi')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>rranjan1@.umiacs.umd.edu, sabrina.zhou.m@gmail.com, {pullpull,akumar14,azadeh}@umiacs.umd.edu, +<br/>vishal.m.patel@rutgers.edu, Rama@umiacs.umd.edu +</td></tr><tr><td>312afff739d1e0fcd3410adf78be1c66b3480396</td><td></td><td></td><td></td></tr><tr><td>3107085973617bbfc434c6cb82c87f2a952021b7</td><td>Spatio-temporal Human Action Localisation and +<br/>Instance Segmentation in Temporally Untrimmed Videos +<br/><b>Oxford Brookes University</b><br/><b>University of Oxford</b><br/>Figure 1: A video sequence taken from the LIRIS-HARL dataset plotted in space-and time. (a) A top down view of the +<br/>video plotted with the detected action tubes of class ‘handshaking’ in green, and ‘person leaves baggage unattended’ in +<br/>red. Each action is located to be within a space-time tube. (b) A side view of the same space-time detections. Note that +<br/>no action is detected at the beginning of the video when there is human motion present in the video. (c) The detection +<br/>and instance segmentation result of two actions occurring simultaneously in a single frame. +</td><td>('3017538', 'Suman Saha', 'suman saha')<br/>('1931660', 'Gurkirt Singh', 'gurkirt singh')<br/>('3019396', 'Michael Sapienza', 'michael sapienza')<br/>('1730268', 'Philip H. S. Torr', 'philip h. s. torr')<br/>('1754181', 'Fabio Cuzzolin', 'fabio cuzzolin')</td><td>{suman.saha-2014, gurkirt.singh-2015, fabio.cuzzolin}@brookes.ac.uk +<br/>{michael.sapienza, philip.torr}@eng.ox.ac.uk +</td></tr><tr><td>31182c5ffc8c5d8772b6db01ec98144cd6e4e897</td><td>3D Face Reconstruction with Region Based Best Fit Blending Using +<br/>Mobile Phone for Virtual Reality Based Social Media +<br/>VALGMA 1∗ +<br/><b>iCV Research Group, Institute of Technology, University of Tartu, Tartu 50411, Estonia</b><br/><b>Hasan Kalyoncu University, Gaziantep, Turkey</b></td><td>('3087532', 'Gholamreza Anbarjafari', 'gholamreza anbarjafari')<br/>('35447268', 'Rain Eric Haamer', 'rain eric haamer')<br/>('7296001', 'Iiris Lüsi', 'iiris lüsi')<br/>('12602781', 'Toomas Tikk', 'toomas tikk')</td><td></td></tr><tr><td>31bb49ba7df94b88add9e3c2db72a4a98927bb05</td><td></td><td></td><td></td></tr><tr><td>3146fabd5631a7d1387327918b184103d06c2211</td><td>Person-independent 3D Gaze Estimation using Face Frontalization +<br/>L´aszl´o A. Jeni +<br/><b>Carnegie Mellon University</b><br/><b>University of Pittsburgh</b><br/>Pittsburgh, PA, USA +<br/>Pittsburgh, PA, USA +<br/>Figure 1: From a 2D image of a person’s face (a) a dense, part-based 3D deformable model is aligned (b) to reconstruct a partial frontal +<br/>view of the face (c). Binary features are extracted around eye and pupil markers (d) for the 3D gaze calculation (e). +</td><td>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>laszlojeni@cmu.edu +<br/>jeffcohn@pitt.edu +</td></tr><tr><td>91811203c2511e919b047ebc86edad87d985a4fa</td><td>Expression Subspace Projection for Face +<br/>Recognition from Single Sample per Person +</td><td>('1782221', 'Hoda Mohammadzade', 'hoda mohammadzade')</td><td></td></tr><tr><td>91495c689e6e614247495c3f322d400d8098de43</td><td>A Deep-Learning Approach to Facial Expression Recognition +<br/>with Candid Images +<br/>Wei Li +<br/><b>CUNY City College</b><br/>Min Li +<br/>Alibaba. Inc +<br/>Zhong Su +<br/><b>IBM China Research Lab</b><br/>Zhigang Zhu +<br/><b>CUNY Graduate Center and City College</b></td><td></td><td>lwei000@citymail.cuny.edu +<br/>mushi.lm@alibaba.inc +<br/>suzhong@cn.ibm.com +<br/>zhu@cs.ccny.cuny.edu +</td></tr><tr><td>910524c0d0fe062bf806bb545627bf2c9a236a03</td><td>Master Thesis +<br/>Improvement of Facial Expression Recognition through the +<br/>Evaluation of Dynamic and Static Features in Video Sequences +<br/>Submitted by: +<br/>Dated: +<br/> 24th June, 2008 +<br/>Supervisors: +<br/><b>Otto-von-Guericke University Magdeburg</b><br/>Faculty of Computer Science +<br/>Department of Simulation und Graphics +<br/><b>Otto-von-Guericke University Magdeburg</b><br/>Faculty of Electrical Engineering and Information Technology +<br/><b>Institute for Electronics, Signal Processing and Communications</b><br/><b></b></td><td>('1692049', 'Klaus Toennies', 'klaus toennies')<br/>('1741165', 'Ayoub Al-Hamadi', 'ayoub al-hamadi')</td><td></td></tr><tr><td>9117fd5695582961a456bd72b157d4386ca6a174</td><td>Facial Expression +<br/>n Recognition Using Dee +<br/>ep Neural +<br/>Networks +<br/>Departm +<br/>ment of Electrical and Electronic Engineering +<br/><b>he University of Hong Kong, Pokfulam</b><br/>Hong Kong +</td><td>('8550244', 'Junnan Li', 'junnan li')<br/>('1725389', 'Edmund Y. Lam', 'edmund y. lam')</td><td></td></tr><tr><td>91df860368cbcebebd83d59ae1670c0f47de171d</td><td>COCO Attributes: +<br/>Attributes for People, Animals, and Objects +<br/>Microsoft Research +<br/><b>Georgia Institute of Technology</b></td><td>('40541456', 'Genevieve Patterson', 'genevieve patterson')<br/>('12532254', 'James Hays', 'james hays')</td><td>gen@microsoft.com +<br/>hays@gatech.edu +</td></tr><tr><td>91067f298e1ece33c47df65236853704f6700a0b</td><td>IJSTE - International Journal of Science Technology & Engineering | Volume 2 | Issue 11 | May 2016 +<br/>ISSN (online): 2349-784X +<br/>Local Binary Pattern and Local Linear +<br/>Regression for Pose Invariant Face Recognition +<br/>M. Tech Student +<br/> +<br/>Shreekumar T +<br/>Associate Professor +<br/>Department of Computer Science & Engineering +<br/>Department of Computer Science & Engineering +<br/><b>Mangalore Institute of Engineering and Technology, Badaga</b><br/><b>Mangalore Institute of Engineering and Technology, Badaga</b><br/>Mijar, Moodbidri, Mangalore +<br/>Mijar, Moodbidri, Mangalore +<br/>Karunakara K +<br/>Professor & Head of Dept. +<br/>Department of Information Science & Engineering +<br/><b>Sri SidarthaInstitute of Technology, Tumkur</b></td><td></td><td></td></tr><tr><td>919d3067bce76009ce07b070a13728f549ebba49</td><td>International Journal of Scientific and Research Publications, Volume 4, Issue 6, June 2014 +<br/>ISSN 2250-3153 +<br/>1 +<br/>Time Based Re-ranking for Web Image Search +<br/>Ms. A.Udhayabharadhi *, Mr. R.Ramachandran ** +<br/><b>MCA Student, Sri Manakula Vinayagar Engineering College, Pondicherry</b><br/><b>Sri Manakula Vinayagar Engineering College, Pondicherry</b></td><td></td><td></td></tr><tr><td>9110c589c6e78daf4affd8e318d843dc750fb71a</td><td>Chapter 6 +<br/>Facial Expression Synthesis Based on Emotion +<br/>Dimensions for Affective Talking Avatar +<br/>1 Key Laboratory of Pervasive Computing, Ministry of Education +<br/>Tsinghua National Laboratory for Information Science and Technology +<br/>Department of Computer Science and Technology, +<br/><b>Tsinghua University, Beijing 100084, China</b><br/><b>Tsinghua-CUHK Joint Research Center for Media Sciences</b><br/>Technologies and Systems, +<br/><b>Graduate School at Shenzhen, Tsinghua University, Shenzhen</b><br/>3 Department of Systems Engineering and Engineering Management +<br/><b>The Chinese University of Hong Kong, HKSAR, China</b></td><td>('2180849', 'Shen Zhang', 'shen zhang')<br/>('3860920', 'Zhiyong Wu', 'zhiyong wu')<br/>('1702243', 'Helen M. Meng', 'helen m. meng')<br/>('7239047', 'Lianhong Cai', 'lianhong cai')</td><td>zhangshen05@mails.tsinghua.edu.cn, john.zy.wu@gmail.com, +<br/>hmmeng@se.cuhk.edu.hk, clh-dcs@tsinghua.edu.cn +</td></tr><tr><td>91e57667b6fad7a996b24367119f4b22b6892eca</td><td>Probabilistic Corner Detection for Facial Feature +<br/>Extraction +<br/>Article +<br/>Accepted version +<br/>E. Ardizzone, M. La Cascia, M. Morana +<br/>In Lecture Notes in Computer Science Volume 5716, 2009 +<br/>It is advisable to refer to the publisher's version if you intend to cite +<br/>from the work. +<br/>Publisher: Springer +<br/>http://link.springer.com/content/pdf/10.1007%2F978-3- +<br/>642-04146-4_50.pdf +</td><td></td><td></td></tr><tr><td>91883dabc11245e393786d85941fb99a6248c1fb</td><td></td><td></td><td></td></tr><tr><td>917bea27af1846b649e2bced624e8df1d9b79d6f</td><td>Ultra Power-Efficient CNN Domain Specific Accelerator with 9.3TOPS/Watt for +<br/>Mobile and Embedded Applications +<br/>Gyrfalcon Technology Inc. +<br/>1900 McCarthy Blvd. Milpitas, CA 95035 +</td><td>('47935028', 'Baohua Sun', 'baohua sun')<br/>('49576071', 'Lin Yang', 'lin yang')<br/>('46195424', 'Patrick Dong', 'patrick dong')<br/>('49039276', 'Wenhan Zhang', 'wenhan zhang')<br/>('35287113', 'Jason Dong', 'jason dong')<br/>('48990565', 'Charles Young', 'charles young')</td><td>{baohua.sun,lin.yang,patrick.dong,wenhan.zhang,jason.dong,charles.yang}@gyrfalcontech.com +</td></tr><tr><td>91b1a59b9e0e7f4db0828bf36654b84ba53b0557</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI +<br/>> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < +<br/> +<br/>Simultaneous Hallucination and Recognition of +<br/>Low-Resolution Faces Based on Singular Value +<br/>Decomposition +<br/>(SVD) +<br/>for performing both +</td><td>('1783889', 'Muwei Jian', 'muwei jian')<br/>('1703078', 'Kin-Man Lam', 'kin-man lam')</td><td></td></tr><tr><td>911bef7465665d8b194b6b0370b2b2389dfda1a1</td><td>RANJAN, ROMERO, BLACK: LEARNING HUMAN OPTICAL FLOW +<br/>Learning Human Optical Flow +<br/>1 MPI for Intelligent Systems +<br/>Tübingen, Germany +<br/>2 Amazon Inc. +</td><td>('1952002', 'Anurag Ranjan', 'anurag ranjan')<br/>('39040964', 'Javier Romero', 'javier romero')<br/>('2105795', 'Michael J. Black', 'michael j. black')</td><td>aranjan@tuebingen.mpg.de +<br/>javier@amazon.com +<br/>black@tuebingen.mpg.de +</td></tr><tr><td>91ead35d1d2ff2ea7cf35d15b14996471404f68d</td><td>Combining and Steganography of 3D Face Textures +</td><td>('38478675', 'Mohsen Moradi', 'mohsen moradi')</td><td></td></tr><tr><td>919d0e681c4ef687bf0b89fe7c0615221e9a1d30</td><td></td><td></td><td></td></tr><tr><td>912a6a97af390d009773452814a401e258b77640</td><td></td><td></td><td></td></tr><tr><td>91d513af1f667f64c9afc55ea1f45b0be7ba08d4</td><td>Automatic Face Image Quality Prediction +</td><td>('2180413', 'Lacey Best-Rowden', 'lacey best-rowden')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>91e507d2d8375bf474f6ffa87788aa3e742333ce</td><td>Robust Face Recognition Using Probabilistic +<br/>Facial Trait Code +<br/>†Department of Computer Science and Information Engineering, National Taiwan +<br/><b>Graduate Institute of Networking and Multimedia, National Taiwan University</b><br/><b>National Taiwan University of Science and</b><br/><b>University</b><br/>Technology +</td><td>('1822733', 'Ping-Han Lee', 'ping-han lee')<br/>('38801529', 'Gee-Sern Hsu', 'gee-sern hsu')<br/>('2250469', 'Szu-Wei Wu', 'szu-wei wu')<br/>('1732064', 'Yi-Ping Hung', 'yi-ping hung')</td><td></td></tr><tr><td>918b72a47b7f378bde0ba29c908babf6dab6f833</td><td></td><td></td><td></td></tr><tr><td>91e58c39608c6eb97b314b0c581ddaf7daac075e</td><td>Pixel-wise Ear Detection with Convolutional +<br/>Encoder-Decoder Networks +</td><td>('31834768', 'Luka Lan Gabriel', 'luka lan gabriel')<br/>('34862665', 'Peter Peer', 'peter peer')</td><td></td></tr><tr><td>91d2fe6fdf180e8427c65ffb3d895bf9f0ec4fa0</td><td></td><td></td><td></td></tr><tr><td>9103148dd87e6ff9fba28509f3b265e1873166c9</td><td>Face Analysis using 3D Morphable Models +<br/>Submitted for the Degree of +<br/>Doctor of Philosophy +<br/>from the +<br/><b>University of Surrey</b><br/>Centre for Vision, Speech and Signal Processing +<br/>Faculty of Engineering and Physical Sciences +<br/><b>University of Surrey</b><br/>Guildford, Surrey GU2 7XH, U.K. +<br/>April 2015 +</td><td>('38819702', 'Guosheng Hu', 'guosheng hu')<br/>('38819702', 'Guosheng Hu', 'guosheng hu')</td><td></td></tr><tr><td>9131c990fad219726eb38384976868b968ee9d9c</td><td>Deep Facial Expression Recognition: A Survey +</td><td>('39433609', 'Shan Li', 'shan li')<br/>('1774956', 'Weihong Deng', 'weihong deng')</td><td></td></tr><tr><td>911505a4242da555c6828509d1b47ba7854abb7a</td><td>IMPROVED ACTIVE SHAPE MODEL FOR FACIAL FEATURE LOCALIZATION +<br/><b>National Formosa University, Taiwan</b></td><td>('1711364', 'Hui-Yu Huang', 'hui-yu huang')<br/>('2782376', 'Shih-Hang Hsu', 'shih-hang hsu')</td><td>Email: hyhuang@nfu.edu.tw +</td></tr><tr><td>915d4a0fb523249ecbc88eb62cb150a60cf60fa0</td><td>Comparison of Feature Extraction Techniques in Automatic +<br/>Face Recognition Systems for Security Applications +<br/>S . Cruz-Llanas, J. Ortega-Garcia, E. Martinez-Torrico, J. Gonzalez-Rodriguez +<br/>Dpto. Ingenieria Audiovisual y Comunicaciones, EUIT Telecomunicacion, Univ. PolitCcnica de Madrid, Spain +<br/>http://www.atvs.diac.upm.es +</td><td></td><td>{cruzll, jortega, etorrico, jgonzalz}@atvs.diac.upm.es. +</td></tr><tr><td>65126e0b1161fc8212643b8ff39c1d71d262fbc1</td><td>Occlusion Coherence: Localizing Occluded Faces with a +<br/>Hierarchical Deformable Part Model +<br/><b>University of California, Irvine</b></td><td>('1898210', 'Golnaz Ghiasi', 'golnaz ghiasi')</td><td>{gghiasi,fowlkes}@ics.uci.edu +</td></tr><tr><td>65b737e5cc4a565011a895c460ed8fd07b333600</td><td>Transfer Learning For Cross-Dataset Recognition: A Survey +<br/>This paper summarises and analyses the cross-dataset recognition transfer learning techniques with the +<br/>emphasis on what kinds of methods can be used when the available source and target data are presented +<br/>in different forms for boosting the target task. This paper for the first time summarises several transferring +<br/>criteria in details from the concept level, which are the key bases to guide what kind of knowledge to transfer +<br/>between datasets. In addition, a taxonomy of cross-dataset scenarios and problems is proposed according the +<br/>properties of data that define how different datasets are diverged, thereby review the recent advances on +<br/>each specific problem under different scenarios. Moreover, some real world applications and corresponding +<br/>commonly used benchmarks of cross-dataset recognition are reviewed. Lastly, several future directions are +<br/>identified. +<br/>Additional Key Words and Phrases: Cross-dataset, transfer learning, domain adaptation +<br/>1. INTRODUCTION +<br/>It has been explored how human would transfer learning in one context to another +<br/>similar context [Woodworth and Thorndike 1901; Perkins et al. 1992] in the field of +<br/>Psychology and Education. For example, learning to drive a car helps a person later +<br/>to learn more quickly to drive a truck, and learning mathematics prepares students to +<br/>study physics. The machine learning algorithms are mostly inspired by human brains. +<br/>However, most of them require a huge amount of training examples to learn a new +<br/>model from scratch and fail to apply knowledge learned from previous domains or +<br/>tasks. This may be due to that a basic assumption of statistical learning theory is +<br/>that the training and test data are drawn from the same distribution and belong to +<br/>the same task. Intuitively, learning from scratch is not realistic and practical, because +<br/>it violates how human learn things. In addition, manually labelling a large amount +<br/>of data for new domain or task is labour extensive, especially for the modern “data- +<br/>hungry” and “data-driven” learning techniques (i.e. deep learning). However, the big +<br/>data era provides a huge amount available data collected for other domains and tasks. +<br/>Hence, how to use the previously available data smartly for the current task with +<br/>scarce data will be beneficial for real world applications. +<br/>To reuse the previous knowledge for current tasks, the differences between old data +<br/>and new data need to be taken into account. Take the object recognition as an ex- +<br/>ample. As claimed by Torralba and Efros [2011], despite the great efforts of object +<br/>datasets creators, the datasets appear to have strong build-in bias caused by various +<br/>factors, such as selection bias, capture bias, category or label bias, and negative set +<br/>bias. This suggests that no matter how big the dataset is, it is impossible to cover +<br/>the complexity of the real visual world. Hence, the dataset bias needs to be consid- +<br/>ered before reusing data from previous datasets. Pan and Yang [2010] summarise that +<br/>the differences between different datasets can be caused by domain divergence (i.e. +<br/>distribution shift or feature space difference) or task divergence (i.e. conditional dis- +<br/>tribution shift or label space difference), or both. For example, in visual recognition, +<br/>the distributions between the previous and current data can be discrepant due to the +<br/>different environments, lighting, background, sensor types, resolutions, view angles, +<br/>and post-processing. Those external factors may cause the distribution divergence or +<br/>even feature space divergence between different domains. On the other hand, the task +<br/>divergence between current and previous data is also ubiquitous. For example, it is +<br/>highly possible that an animal species that we want to recognize have not been seen +<br/>ACM Journal Name, Vol. V, No. N, Article A, Publication date: January YYYY. +</td><td>('38791459', 'Jing Zhang', 'jing zhang')<br/>('1685696', 'Wanqing Li', 'wanqing li')<br/>('1719314', 'Philip Ogunbona', 'philip ogunbona')</td><td></td></tr><tr><td>6582f4ec2815d2106957215ca2fa298396dde274</td><td>JUNE 2007 +<br/>1005 +<br/>Discriminative Learning and Recognition +<br/>of Image Set Classes Using +<br/>Canonical Correlations +</td><td>('1700968', 'Tae-Kyun Kim', 'tae-kyun kim')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('1745672', 'Roberto Cipolla', 'roberto cipolla')</td><td></td></tr><tr><td>65b1760d9b1541241c6c0222cc4ee9df078b593a</td><td>Enhanced Pictorial Structures for Precise Eye Localization +<br/>Under Uncontrolled Conditions +<br/>1Department of Computer Science and Engineering +<br/><b>Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China</b><br/>2National Key Laboratory for Novel Software Technology +<br/><b>Nanjing University, Nanjing 210093, China</b></td><td>('2248421', 'Xiaoyang Tan', 'xiaoyang tan')<br/>('3075941', 'Fengyi Song', 'fengyi song')<br/>('1692625', 'Zhi-Hua Zhou', 'zhi-hua zhou')<br/>('1680768', 'Songcan Chen', 'songcan chen')</td><td>{x.tan, f.song, s.chen}@nuaa.edu.cn +<br/>zhouzh@lamda.nju.edu.cn +</td></tr><tr><td>65d7f95fcbabcc3cdafc0ad38e81d1f473bb6220</td><td>Face Recognition for the Visually Impaired +<br/><b>King Saud University, Riyadh, Saudi Arabia</b><br/>2ISM-TEC LLC, Wilmington, Delaware, U.S.A +<br/><b>University of Georgia, Athens, GA, U.S.A</b></td><td>('2278811', 'Rabia Jafri', 'rabia jafri')<br/>('2227653', 'Syed Abid Ali', 'syed abid ali')<br/>('1712033', 'Hamid R. Arabnia', 'hamid r. arabnia')</td><td></td></tr><tr><td>65bba9fba03e420c96ec432a2a82521ddd848c09</td><td>Connectionist Temporal Modeling for Weakly +<br/>Supervised Action Labeling +<br/><b>Stanford University</b></td><td>('38485317', 'De-An Huang', 'de-an huang')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')<br/>('9200530', 'Juan Carlos Niebles', 'juan carlos niebles')</td><td>{dahuang,feifeili,jniebles}@cs.stanford.edu +</td></tr><tr><td>656531036cee6b2c2c71954bb6540ef6b2e016d0</td><td>W. LIU ET AL.: JOINTLY LEARNING NON-NEGATIVE PROJECTION AND DICTIONARY 1 +<br/>Jointly Learning Non-negative Projection +<br/>and Dictionary with Discriminative Graph +<br/>Constraints for Classification +<br/>Yandong Wen3 +<br/>Rongmei Lin4 +<br/>Meng Yang*1 +<br/><b>College of Computer Science</b><br/>Software Engineering, +<br/><b>Shenzhen University, China</b><br/>2 School of ECE, +<br/><b>Peking University, China</b><br/>3 Dept. of ECE, +<br/><b>Carnegie Mellon University, USA</b><br/>4 Dept. of Math & Computer Science, +<br/><b>Emory University, USA</b></td><td>('36326884', 'Weiyang Liu', 'weiyang liu')<br/>('1751019', 'Zhiding Yu', 'zhiding yu')</td><td>wyliu@pku.edu.cn +<br/>yzhiding@andrew.cmu.edu +<br/>yandongw@andrew.cmu.edu +<br/>rongmei.lin@emory.edu +<br/>yang.meng@szu.edu.cn +</td></tr><tr><td>65b1209d38c259fe9ca17b537f3fb4d1857580ae</td><td>Information Constraints on Auto-Encoding Variational Bayes +<br/><b>University of California, Berkeley</b><br/><b>University of California, Berkeley</b><br/><b>Ragon Institute of MGH, MIT and Harvard</b><br/>4Chan-Zuckerberg Biohub +</td><td>('39848341', 'Romain Lopez', 'romain lopez')<br/>('39967607', 'Jeffrey Regier', 'jeffrey regier')<br/>('1694621', 'Michael I. Jordan', 'michael i. jordan')<br/>('2163873', 'Nir Yosef', 'nir yosef')</td><td>{romain_lopez, regier, niryosef}@berkeley.edu +<br/>jordan@cs.berkeley.edu +</td></tr><tr><td>655d9ba828eeff47c600240e0327c3102b9aba7c</td><td>IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 35, NO. 3, JUNE 2005 +<br/>489 +<br/>Kernel Pooled Local Subspaces for Classification +</td><td>('40409453', 'Peng Zhang', 'peng zhang')<br/>('1708023', 'Jing Peng', 'jing peng')<br/>('1741392', 'Carlotta Domeniconi', 'carlotta domeniconi')</td><td></td></tr><tr><td>656a59954de3c9fcf82ffcef926af6ade2f3fdb5</td><td>Convolutional Network Representation +<br/>for Visual Recognition +<br/>Doctoral Thesis +<br/>Stockholm, Sweden, 2017 +</td><td>('2835963', 'Ali Sharif Razavian', 'ali sharif razavian')</td><td></td></tr><tr><td>652aac54a3caf6570b1c10c993a5af7fa2ef31ff</td><td><b>CARNEGIE MELLON UNIVERSITY</b><br/>STATISTICAL MODELING FOR NETWORKED VIDEO: +<br/>CODING OPTIMIZATION, ERROR CONCEALMENT AND +<br/>TRAFFIC ANALYSIS +<br/>A DISSERTATION +<br/>SUBMITTED TO THE GRADUATE SCHOOL +<br/>IN PARTIAL FULFILLMENT OF THE REQUIREMENTS +<br/>for the degree +<br/>DOCTOR OF PHILOSOPHY +<br/>in +<br/>ELECTRICAL AND COMPUTER ENGINEERING +<br/>by +<br/>Pittsburgh, Pennsylvania +<br/>July, 2001 +</td><td>('1727257', 'Deepak Srinivas Turaga', 'deepak srinivas turaga')</td><td></td></tr><tr><td>656ef752b363a24f84cc1aeba91e4fa3d5dd66ba</td><td>Robust Open-Set Face Recognition for +<br/>Small-Scale Convenience Applications +<br/><b>Institute for Anthropomatics</b><br/><b>Karlsruhe Institute of Technology</b><br/>Karlsruhe, Germany +</td><td>('1697965', 'Hua Gao', 'hua gao')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>Email: {hua.gao, ekenel, rainer.stiefelhagen}@kit.edu +</td></tr><tr><td>656aeb92e4f0e280576cbac57d4abbfe6f9439ea</td><td>Journal of Engineering Science and Technology +<br/>Vol. 12, No. 1 (2017) 155 - 167 +<br/><b>School of Engineering, Taylor s University</b><br/>USE OF IMAGE ENHANCEMENT TECHNIQUES +<br/>FOR IMPROVING REAL TIME FACE RECOGNITION EFFICIENCY +<br/>ON WEARABLE GADGETS +<br/><b>Asia Pacific University of Technology and Innovation, Kuala Lumpur 57000, Malaysia</b><br/><b>Staffordshire University, Beaconside Stafford ST18 0AB, United Kingdom</b></td><td>('22422404', 'MUHAMMAD EHSAN RANA', 'muhammad ehsan rana')</td><td>*Corresponding Author: muhd_ehsanrana@apu.edu.my +</td></tr><tr><td>656f05741c402ba43bb1b9a58bcc5f7ce2403d9a</td><td></td><td>('2319574', 'Danila Potapov', 'danila potapov')</td><td></td></tr><tr><td>6577c76395896dd4d352f7b1ee8b705b1a45fa90</td><td>TOWARDS COMPUTATIONAL MODELS OF KINSHIP VERIFICATION +<br/><b>Cornell University</b><br/><b>Cornell University</b></td><td>('2666471', 'Ruogu Fang', 'ruogu fang')<br/>('1830653', 'Noah Snavely', 'noah snavely')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td></td></tr><tr><td>650bfe7acc3f03eb4ba91d9f93da8ef0ae8ba772</td><td>A Deep Learning Approach for Subject Independent Emotion +<br/>Recognition from Facial Expressions +<br/>*Faculty of Electronics, Telecommunications & Information Technology +<br/><b>Polytechnic University of Bucharest</b><br/>Splaiul Independentei No. 313, Sector 6, Bucharest, +<br/>ROMANIA +<br/>**Department of Information Engineering and Computer Science +<br/><b>University of Trento</b><br/>ITALY +</td><td>('3178525', 'VICTOR-EMIL NEAGOE', 'victor-emil neagoe')</td><td>victoremil@gmail.com, andreibarar@gmail.com, robitupaul@gmail.com +<br/>sebe@disi.unitn.it +</td></tr><tr><td>65293ecf6a4c5ab037a2afb4a9a1def95e194e5f</td><td>Face, Age and Gender Recognition +<br/>using Local Descriptors +<br/>by +<br/>Thesis submitted to the +<br/>Faculty of Graduate and Postdoctoral Studies +<br/>In partial fulfillment of the requirements +<br/>For the M.A.Sc. degree in +<br/>Electrical and Computer Engineering +<br/>School of Electrical Engineering and Computer Science +<br/>Faculty of Engineering +<br/><b>University of Ottawa</b></td><td>('15604275', 'Mohammad Esmaeel Mousa Pasandi', 'mohammad esmaeel mousa pasandi')<br/>('15604275', 'Mohammad Esmaeel Mousa Pasandi', 'mohammad esmaeel mousa pasandi')</td><td></td></tr><tr><td>65817963194702f059bae07eadbf6486f18f4a0a</td><td>http://dx.doi.org/10.1007/s11263-015-0814-0 +<br/>WhittleSearch: Interactive Image Search with Relative Attribute +<br/>Feedback +<br/>Received: date / Accepted: date +</td><td>('1770205', 'Adriana Kovashka', 'adriana kovashka')</td><td></td></tr><tr><td>6581c5b17db7006f4cc3575d04bfc6546854a785</td><td>Contextual Person Identification +<br/>in Multimedia Data +<br/>zur Erlangung des akademischen Grades eines +<br/>Doktors der Ingenieurwissenschaften +<br/>der Fakultät für Informatik +<br/>des Karlsruher Instituts für Technologie (KIT) +<br/>genehmigte +<br/>Dissertation +<br/>von +<br/>aus Erlangen +<br/>Tag der mündlichen Prüfung: +<br/>18. November 2014 +<br/>Hauptreferent: +<br/>Korreferent: +<br/>Prof. Dr. Rainer Stiefelhagen +<br/>Karlsruher Institut für Technologie +<br/>Prof. Dr. Gerhard Rigoll +<br/>Technische Universität München +<br/>KIT – Universität des Landes Baden-Württemberg und nationales Forschungszentrum in der Helmholtz-Gemeinschaft +<br/>www.kit.edu +</td><td>('1931707', 'Martin Bäuml', 'martin bäuml')</td><td></td></tr><tr><td>6515fe829d0b31a5e1f4dc2970a78684237f6edb</td><td>Constrained Maximum Likelihood Learning of +<br/>Bayesian Networks for Facial Action Recognition +<br/>1 Electrical, Computer and Systems Eng. Dept. +<br/><b>Rensselaer Polytechnic Institute</b><br/>Troy, NY, USA +<br/>2 Visualization and Computer Vision Lab +<br/><b>GE Global Research Center</b><br/>Niskayuna, NY, USA +</td><td>('1686235', 'Yan Tong', 'yan tong')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td></td></tr><tr><td>653d19e64bd75648cdb149f755d59e583b8367e3</td><td>Decoupling “when to update” from “how to +<br/>update” +<br/><b>School of Computer Science, The Hebrew University, Israel</b></td><td>('19201820', 'Eran Malach', 'eran malach')<br/>('2554670', 'Shai Shalev-Shwartz', 'shai shalev-shwartz')</td><td></td></tr><tr><td>65babb10e727382b31ca5479b452ee725917c739</td><td>Label Distribution Learning +</td><td>('1735299', 'Xin Geng', 'xin geng')</td><td></td></tr><tr><td>62dccab9ab715f33761a5315746ed02e48eed2a0</td><td>A Short Note about Kinetics-600 +<br/>Jo˜ao Carreira +</td><td>('51210148', 'Eric Noland', 'eric noland')<br/>('51215438', 'Andras Banki-Horvath', 'andras banki-horvath')<br/>('38961760', 'Chloe Hillier', 'chloe hillier')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>joaoluis@google.com +<br/>enoland@google.com +<br/>bhandras@google.com +<br/>chillier@google.com +<br/>zisserman@google.com +</td></tr><tr><td>62d1a31b8acd2141d3a994f2d2ec7a3baf0e6dc4</td><td>Ding et al. EURASIP Journal on Image and Video Processing (2017) 2017:43 +<br/>DOI 10.1186/s13640-017-0188-z +<br/>EURASIP Journal on Image +<br/>and Video Processing +<br/>R ES EAR CH +<br/>Noise-resistant network: a deep-learning +<br/>method for face recognition under noise +<br/>Open Access +</td><td>('3012331', 'Yuanyuan Ding', 'yuanyuan ding')<br/>('1976669', 'Yongbo Cheng', 'yongbo cheng')<br/>('1847689', 'Xiaoliu Cheng', 'xiaoliu cheng')<br/>('4869582', 'Baoqing Li', 'baoqing li')<br/>('2757480', 'Xing You', 'xing you')<br/>('38334864', 'Xiaobing Yuan', 'xiaobing yuan')</td><td></td></tr><tr><td>62694828c716af44c300f9ec0c3236e98770d7cf</td><td>Padrón-Rivera, G., Rebolledo-Mendez, G., Parra, P. P., & Huerta-Pacheco, N. S. (2016). Identification of Action Units Related to +<br/>Identification of Action Units Related to Affective States in a Tutoring System +<br/>1Facultad de Estadística e Informática, Universidad Veracruzana, Mexico // 2Universidad Juárez Autónoma de +<br/>for Mathematics +<br/>Huerta-Pacheco1 +<br/>*Corresponding author +</td><td>('2221778', 'Gustavo Padrón-Rivera', 'gustavo padrón-rivera')<br/>('1731562', 'Genaro Rebolledo-Mendez', 'genaro rebolledo-mendez')</td><td>Tabasco, Mexico // zS12020111@estudiantes.uv.mx // grebolledo@uv.mx // pilar.pozos@ujat.mx // +<br/>nehuerta@uv.mx +</td></tr><tr><td>6261eb75066f779e75b02209fbd3d0f02d3e1e45</td><td>Fudan-Huawei at MediaEval 2015: Detecting Violent +<br/>Scenes and Affective Impact in Movies with Deep Learning +<br/><b>School of Computer Science, Fudan University, Shanghai, China</b><br/>2Media Lab, Huawei Technologies Co. Ltd., China +</td><td>('9227981', 'Qi Dai', 'qi dai')<br/>('3066866', 'Rui-Wei Zhao', 'rui-wei zhao')<br/>('3099139', 'Zuxuan Wu', 'zuxuan wu')<br/>('31825486', 'Xi Wang', 'xi wang')<br/>('2650085', 'Zichen Gu', 'zichen gu')<br/>('2273062', 'Wenhai Wu', 'wenhai wu')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')</td><td></td></tr><tr><td>622daa25b5e6af69f0dac3a3eaf4050aa0860396</td><td>Greedy Feature Selection for Subspace Clustering +<br/>Greedy Feature Selection for Subspace Clustering +<br/>Department of Electrical & Computer Engineering +<br/><b>Rice University, Houston, TX, 77005, USA</b><br/>Department of Electrical & Computer Engineering +<br/><b>Carnegie Mellon University, Pittsburgh, PA, 15213, USA</b><br/>Department of Electrical & Computer Engineering +<br/><b>Rice University, Houston, TX, 77005, USA</b><br/>Editor: +</td><td>('1746363', 'Eva L. Dyer', 'eva l. dyer')<br/>('1745861', 'Aswin C. Sankaranarayanan', 'aswin c. sankaranarayanan')<br/>('1746260', 'Richard G. Baraniuk', 'richard g. baraniuk')</td><td>e.dyer@rice.edu +<br/>saswin@ece.cmu.edu +<br/>richb@rice.edu +</td></tr><tr><td>620339aef06aed07a78f9ed1a057a25433faa58b</td><td></td><td></td><td></td></tr><tr><td>62b3598b401c807288a113796f424612cc5833ca</td><td></td><td></td><td></td></tr><tr><td>62f0d8446adee6a5e8102053a63a61af07ac4098</td><td>FACIAL POINT DETECTION USING CONVOLUTIONAL NEURAL NETWORK +<br/>TRANSFERRED FROM A HETEROGENEOUS TASK +<br/>**Tome R&D +<br/><b>Chubu University</b><br/>1200, Matsumoto-cho, Kasugai, AICHI +</td><td>('1687819', 'Takayoshi Yamashita', 'takayoshi yamashita')</td><td></td></tr><tr><td>628a3f027b7646f398c68a680add48c7969ab1d9</td><td>Plan for Final Year Project: +<br/>HKU-Face: A Large Scale Dataset for Deep Face +<br/>Recognition +<br/>3035140108 +<br/>3035141841 +<br/>Introduction +<br/>Face recognition has been one of the most successful techniques in the field of artificial intelligence +<br/>because of its surpassing human-level performance in academic experiments and broad application in +<br/>the industrial world. Gaussian-face[1] and Facenet[2] hold state-of-the-art record using statistical +<br/>method and deep-learning method respectively. What’s more, face recognition has been applied +<br/>in various areas like authority checking and recording, fostering a large number of start-ups like +<br/>Face++. +<br/>Our final year project will deal with the face recognition task by building a large-scaled and carefully- +<br/>filtered dataset. Our project plan specifies our roadmap and current research process. This plan first +<br/>illustrates the significance and potential enhancement in constructing large-scale face dataset for +<br/>both academics and companies. Then objectives to accomplish and related literature review will be +<br/>expressed in detail. Next, methodologies used, scope of our project and challenges faced by us are +<br/>described. The detailed timeline for this project follows as well as a small summary. +<br/>2 Motivation +<br/>Nowadays most of the face recognition tasks are supervised learning tasks which use dataset annotated +<br/>by human beings. This contains mainly two drawbacks: (1) limited size of dataset due to limited +<br/>human effort; (2) accuracy problem resulted from human perceptual bias. +<br/>Parkhi et al.[3] discuss the first problem, showing that giant companies hold private face databases +<br/>with larger size of data (See the comparison in Table 1). Other research institution could only get +<br/>access to public but smaller databases like LFW[4, 5], which acts like a barricade to even higher +<br/>performance. +<br/>Dataset +<br/>IJB-A [6] +<br/>LFW [4, 5] +<br/>YFD [7] +<br/>CelebFaces [8] +<br/>CASIA-WebFace [9] +<br/>MS-Celeb-1M [10] +<br/>Facebook +<br/>Google +<br/>Availability +<br/>public +<br/>public +<br/>public +<br/>public +<br/>public +<br/>public +<br/>private +<br/>private +<br/>identities +<br/>500 +<br/>5K +<br/>1595 +<br/>10K +<br/>10K +<br/>100K +<br/>4K +<br/>8M +<br/>images +<br/>5712 +<br/>13K +<br/>3425 videos +<br/>202K +<br/>500K +<br/>about 10M +<br/>4400K +<br/>100-200M +<br/>Table 1: Face recognition datasets +</td><td>('3347561', 'Haicheng Wang', 'haicheng wang')<br/>('40456402', 'Haoyu Li', 'haoyu li')</td><td></td></tr><tr><td>626913b8fcbbaee8932997d6c4a78fe1ce646127</td><td>Learning from Millions of 3D Scans for Large-scale 3D Face Recognition +<br/>(This the preprint of the paper published in CVPR 2018) +<br/>School of Computer Science and Software Engineering, +<br/><b>The University of Western Australia</b></td><td>('1746166', 'Syed Zulqarnain Gilani', 'syed zulqarnain gilani')<br/>('46332747', 'Ajmal Mian', 'ajmal mian')</td><td>{zulqarnain.gilani,ajmal.mian}@uwa.edu.au +</td></tr><tr><td>62374b9e0e814e672db75c2c00f0023f58ef442c</td><td>Frontalfaceauthenticationusingdiscriminatinggridswith +<br/>morphologicalfeaturevectors +<br/>A.Tefas +<br/>C.Kotropoulos +<br/>I.Pitas +<br/><b>AristotleUniversityofThessaloniki</b><br/>Box,Thessaloniki +<br/>EDICSnumbers:-KNOWContentRecognitionandUnderstanding +<br/>-MODAMultimodalandMultimediaEnvironments +<br/>Anovelelasticgraphmatchingprocedurebasedonmultiscalemorphologicaloperations,thesocalled +<br/>morphologicaldynamiclinkarchitecture,isdevelopedforfrontalfaceauthentication.Fastalgorithms +<br/>forimplementingmathematicalmorphologyoperationsarepresented.Featureselectionbyemploying +<br/>linearprojectionalgorithmsisproposed.Discriminatorypowercoe(cid:14)cientsthatweighthematching +<br/>errorateachgridnodearederived.Theperformanceofmorphologicaldynamiclinkarchitecturein +<br/>frontalfaceauthenticationisevaluatedintermsofthereceiveroperatingcharacteristicontheMVTS +<br/>faceimagedatabase.Preliminaryresultsforfacerecognitionusingtheproposedtechniquearealso +<br/>presented. +<br/>Correspondingauthor:I.Pitas +<br/>DRAFT +<br/>September +</td><td></td><td>E-mail:fcostas,tefas,pitasg@zeus.csd.auth.gr +</td></tr><tr><td>6257a622ed6bd1b8759ae837b50580657e676192</td><td></td><td></td><td></td></tr><tr><td>6226f2ea345f5f4716ac4ddca6715a47162d5b92</td><td>PERSPECTIVE +<br/>published: 19 November 2015 +<br/>doi: 10.3389/frobt.2015.00029 +<br/>Object Detection: Current and +<br/>Future Directions +<br/>1 Advanced Mining Technology Center, Universidad de Chile, Santiago, Chile, 2 Department of Electrical Engineering, +<br/>Universidad de Chile, Santiago, Chile +<br/>Object detection is a key ability required by most computer and robot vision systems. +<br/>The latest research on this area has been making great progress in many directions. In +<br/>the current manuscript, we give an overview of past research on object detection, outline +<br/>the current main research directions, and discuss open problems and possible future +<br/>directions. +<br/>Keywords: object detection, perspective, mini review, current directions, open problems +<br/>1. INTRODUCTION +<br/>During the last years, there has been a rapid and successful expansion on computer vision research. +<br/>Parts of this success have come from adopting and adapting machine learning methods, while others +<br/>from the development of new representations and models for specific computer vision problems +<br/>or from the development of efficient solutions. One area that has attained great progress is object +<br/>detection. The present works gives a perspective on object detection research. +<br/>Given a set of object classes, object detection consists in determining the location and scale of all +<br/>object instances, if any, that are present in an image. Thus, the objective of an object detector is to find +<br/>all object instances of one or more given object classes regardless of scale, location, pose, view with +<br/>respect to the camera, partial occlusions, and illumination conditions. +<br/>In many computer vision systems, object detection is the first task being performed as it allows +<br/>to obtain further information regarding the detected object and about the scene. Once an object +<br/><b>instance has been detected (e.g., a face), it is be possible to obtain further information, including: (i</b><br/>to recognize the specific instance (e.g., to identify the subject’s face), (ii) to track the object over an +<br/>image sequence (e.g., to track the face in a video), and (iii) to extract further information about the +<br/>object (e.g., to determine the subject’s gender), while it is also possible to (a) infer the presence or +<br/>location of other objects in the scene (e.g., a hand may be near a face and at a similar scale) and (b) to +<br/>better estimate further information about the scene (e.g., the type of scene, indoor versus outdoor, +<br/>etc.), among other contextual information. +<br/>Object detection has been used in many applications, with the most popular ones being: (i) +<br/>human-computer interaction (HCI), (ii) robotics (e.g., service robots), (iii) consumer electronics +<br/>(e.g., smart-phones), (iv) security (e.g., recognition, tracking), (v) retrieval (e.g., search engines, +<br/>photo management), and (vi) transportation (e.g., autonomous and assisted driving). Each of these +<br/><b>applications has different requirements, including: processing time (off-line, on-line, or real-time</b><br/>robustness to occlusions, invariance to rotations (e.g., in-plane rotations), and detection under pose +<br/>changes. While many applications consider the detection of a single object class (e.g., faces) and from +<br/>a single view (e.g., frontal faces), others require the detection of multiple object classes (humans, +<br/>vehicles, etc.), or of a single class from multiple views (e.g., side and frontal view of vehicles). +<br/>In general, most systems can detect only a single object class from a restricted set of views and +<br/>poses. +<br/>Edited by: +<br/>Venkatesh Babu Radhakrishnan, +<br/><b>Indian Institute of Science Bangalore</b><br/>India +<br/>Reviewed by: +<br/>Juxi Leitner, +<br/><b>Queensland University of Technology</b><br/>Australia +<br/>George Azzopardi, +<br/><b>University of Groningen, Netherlands</b><br/>Soma Biswas, +<br/><b>Indian Institute of Science Bangalore</b><br/>India +<br/>*Correspondence: +<br/>†Present address: +<br/>Graduate School of Informatics, +<br/><b>Kyoto University, Kyoto, Japan</b><br/>Specialty section: +<br/>This article was submitted to Vision +<br/>Systems Theory, Tools and +<br/>Applications, a section of the +<br/>journal Frontiers in Robotics and AI +<br/>Received: 20 July 2015 +<br/>Accepted: 04 November 2015 +<br/>Published: 19 November 2015 +<br/>Citation: +<br/>Verschae R and Ruiz-del-Solar J +<br/>(2015) Object Detection: Current and +<br/>Future Directions. +<br/>Front. Robot. AI 2:29. +<br/>doi: 10.3389/frobt.2015.00029 +<br/>Frontiers in Robotics and AI | www.frontiersin.org +<br/>November 2015 | Volume 2 | Article 29 +</td><td>('1689681', 'Rodrigo Verschae', 'rodrigo verschae')<br/>('1737300', 'Javier Ruiz-del-Solar', 'javier ruiz-del-solar')<br/>('1689681', 'Rodrigo Verschae', 'rodrigo verschae')<br/>('1689681', 'Rodrigo Verschae', 'rodrigo verschae')</td><td>rodrigo@verschae.org +</td></tr><tr><td>62e913431bcef5983955e9ca160b91bb19d9de42</td><td>Facial Landmark Detection with Tweaked Convolutional Neural Networks +<br/><b>USC Information Sciences Institute</b><br/><b>The Open University of Israel</b></td><td>('1746738', 'Yue Wu', 'yue wu')<br/>('1756099', 'Tal Hassner', 'tal hassner')</td><td></td></tr><tr><td>626859fe8cafd25da13b19d44d8d9eb6f0918647</td><td>Activity Recognition based on a +<br/>Magnitude-Orientation Stream Network +<br/>Smart Surveillance Interest Group, Department of Computer Science +<br/>Universidade Federal de Minas Gerais, Belo Horizonte, Brazil +</td><td>('2119408', 'Carlos Caetano', 'carlos caetano')<br/>('1679142', 'William Robson Schwartz', 'william robson schwartz')</td><td>{carlos.caetano,victorhcmelo,jefersson,william}@dcc.ufmg.br +</td></tr><tr><td>624e9d9d3d941bab6aaccdd93432fc45cac28d4b</td><td>Object-Scene Convolutional Neural Networks for Event Recognition in Images +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen key lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b></td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('1915826', 'Zhe Wang', 'zhe wang')<br/>('35031371', 'Wenbin Du', 'wenbin du')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td>07wanglimin@gmail.com, buptwangzhe2012@gmail.com, wb.du@siat.ac.cn, yu.qiao@siat.ac.cn +</td></tr><tr><td>620e1dbf88069408b008347cd563e16aeeebeb83</td><td></td><td></td><td></td></tr><tr><td>624496296af19243d5f05e7505fd927db02fd0ce</td><td>Gauss-Newton Deformable Part Models for Face Alignment in-the-Wild +<br/>1. School of Computer Science +<br/><b>University of Lincoln, U.K</b><br/>2. Department of Computing +<br/><b>Imperial College London, U.K</b></td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>gtzimiropoulos@lincoln.ac.uk +</td></tr><tr><td>62fd622b3ca97eb5577fd423fb9efde9a849cbef</td><td>Turning a Blind Eye: Explicit Removal of Biases and +<br/>Variation from Deep Neural Network Embeddings +<br/><b>Visual Geometry Group, University of Oxford</b><br/><b>University of Oxford</b><br/><b>Big Data Institute, University of Oxford</b></td><td>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>621ff353960d5d9320242f39f85921f72be69dc8</td><td>Explicit Occlusion Detection based Deformable Fitting for +<br/>Facial Landmark Localization +<br/>1Department of Computer Science +<br/><b>Rutgers University</b><br/>617 Bowser Road, Piscataway, N.J, USA +</td><td>('39960064', 'Xiang Yu', 'xiang yu')<br/>('1684164', 'Fei Yang', 'fei yang')<br/>('1768190', 'Junzhou Huang', 'junzhou huang')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')</td><td>{xiangyu,feiyang,dnm}@cs.rutgers.edu +</td></tr><tr><td>62007c30f148334fb4d8975f80afe76e5aef8c7f</td><td>Eye In-Painting with Exemplar Generative Adversarial Networks +<br/>Facebook Inc. +<br/>1 Hacker Way, Menlo Park (CA), USA +</td><td>('8277405', 'Brian Dolhansky', 'brian dolhansky')</td><td>{bdol, ccanton}@fb.com +</td></tr><tr><td>62a30f1b149843860938de6dd6d1874954de24b7</td><td>418 +<br/>Fast Algorithm for Updating the Discriminant Vectors +<br/>of Dual-Space LDA +</td><td>('40608983', 'Wenming Zheng', 'wenming zheng')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>621e8882c41cdaf03a2c4a986a6404f0272ba511</td><td>On Robust Biometric Identity Verification via +<br/>Sparse Encoding of Faces: Holistic vs Local Approaches +<br/><b>The University of Queensland, School of ITEE, QLD 4072, Australia</b></td><td>('3026404', 'Yongkang Wong', 'yongkang wong')<br/>('1781182', 'Conrad Sanderson', 'conrad sanderson')<br/>('2270092', 'Brian C. Lovell', 'brian c. lovell')</td><td></td></tr><tr><td>62e0380a86e92709fe2c64e6a71ed94d152c6643</td><td>Facial Emotion Recognition With Expression Energy +<br/>Albert Cruz +<br/>Center for Research in +<br/>Intelligent Systems +<br/>216 Winston Chung Hall +<br/>Center for Research in +<br/>Intelligent Systems +<br/>216 Winston Chung Hall +<br/>Center for Research in +<br/>Intelligent Systems +<br/>216 Winston Chung Hall +<br/>Riverside, CA, 92521-0425, +<br/>Riverside, CA, 92521-0425, +<br/>Riverside, CA, 92521-0425, +<br/>USA +<br/>USA +<br/>USA +</td><td>('1707159', 'Bir Bhanu', 'bir bhanu')<br/>('3254753', 'Ninad Thakoor', 'ninad thakoor')</td><td>acruz006@student.ucr.edu +<br/>bhanu@ee.ucr.edu +<br/>ninadt@ee.ucr.edu +</td></tr><tr><td>621f656fedda378ceaa9c0096ebb1556a42e5e0f</td><td>Single Sample Face Recognition from Video via +<br/>Stacked Supervised Auto-encoder +<br/><b>Ponti cal Catholic University of Rio de Janeiro, Brazil</b><br/><b>Rio de Janeiro State University, Brazil</b></td><td>('8730918', 'Pedro J. Soto Vega', 'pedro j. soto vega')<br/>('2017816', 'Raul Queiroz Feitosa', 'raul queiroz feitosa')<br/>('2222679', 'Patrick Nigri Happ', 'patrick nigri happ')</td><td>{psoto, raul, vhaymaq, patrick}@ele.puc-rio.br +</td></tr><tr><td>965f8bb9a467ce9538dec6bef57438964976d6d9</td><td>Recognizing Human Faces under Disguise and Makeup +<br/><b>The Hong Kong Polytechnic University</b><br/>Hung Hom, Kowloon, Hong Kong +</td><td>('17671202', 'Tsung Ying Wang', 'tsung ying wang')<br/>('35680604', 'Ajay Kumar', 'ajay kumar')</td><td>cstywang@comp.polyu.edu.hk, csajaykr@comp.polyu.edu.hk +</td></tr><tr><td>961a5d5750f18e91e28a767b3cb234a77aac8305</td><td>Face Detection without Bells and Whistles +<br/>1 ESAT-PSI/VISICS, iMinds, KU Leuven, Belgium +<br/>2 MPI Informatics, Saarbrücken, Germany +<br/>3 D-ITET/CVL, ETH Zürich, Switzerland +</td><td>('11983029', 'Markus Mathias', 'markus mathias')<br/>('1798000', 'Rodrigo Benenson', 'rodrigo benenson')<br/>('3048367', 'Marco Pedersoli', 'marco pedersoli')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>96f0e7416994035c91f4e0dfa40fd45090debfc5</td><td>Unsupervised Learning of Face Representations +<br/><b>Georgia Institute of Technology, CVIT, IIIT Hyderabad, IIT Kanpur</b></td><td>('19200118', 'Samyak Datta', 'samyak datta')<br/>('39396475', 'Gaurav Sharma', 'gaurav sharma')</td><td></td></tr><tr><td>9626bcb3fc7c7df2c5a423ae8d0a046b2f69180c</td><td>UPTEC STS 17033 +<br/>Examensarbete 30 hp +<br/>November 2017 +<br/>A deep learning approach for +<br/>action classification in American +<br/>football video sequences +</td><td>('5845058', 'Jacob Westerberg', 'jacob westerberg')</td><td></td></tr><tr><td>963d0d40de8780161b70d28d2b125b5222e75596</td><td>Convolutional Experts Network for Facial Landmark Detection +<br/><b>Carnegie Mellon University</b><br/>Tadas Baltruˇsaitis∗ +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +<br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +</td><td>('1783029', 'Amir Zadeh', 'amir zadeh')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td>abagherz@cs.cmu.edu +<br/>tbaltrus@cs.cmu.edu +<br/>morency@cs.cmu.edu +</td></tr><tr><td>968b983fa9967ff82e0798a5967920188a3590a8</td><td>2013, Vol. 139, No. 2, 271–299 +<br/>© 2013 American Psychological Association +<br/>0033-2909/13/$12.00 DOI: 10.1037/a0031640 +<br/>Children’s Recognition of Disgust in Others +<br/>Sherri C. Widen and James A. Russell +<br/><b>Boston College</b><br/>Disgust has been theorized to be a basic emotion with a facial signal that is easily, universally, +<br/>automatically, and perhaps innately recognized by observers from an early age. This article questions one +<br/>key part of that theory: the hypothesis that children recognize disgust from its purported facial signal. +<br/>Over the first 5 years, children experience disgust, produce facial expressions of disgust, develop a +<br/>concept of disgust, understand and produce the word disgust or a synonym, know about disgust’s causes +<br/>and consequences, and infer disgust in others from a situation or a behavior. Yet, only gradually do these +<br/>children come to “recognize” disgust specifically from the “disgust face” found in standardized sets of +<br/>the facial expressions of basic emotions. Improvement is gradual, with more than half of children +<br/>matching the standard disgust face to disgust only at around 9 years of age and with subsequent +<br/>improvement continuing gradually until the late teens or early adulthood. Up to age 8, a majority of +<br/>children studied believe that the standard disgust face indicates anger. Rather than relying on an already +<br/>known signal value, children may be actively learning to interpret the expression. +<br/>Keywords: facial expression, disgust, anger, emotion recognition, disgust face +<br/>Disgust has been theorized to be important for many reasons: its +<br/>status as one of only a handful of basic human emotions and hence +<br/>as a building block of other emotions (Rozin, Haidt, & McCauley, +<br/>2008); its role in avoidance of poisons, parasites, disease, and +<br/>contaminants (Curtis, De Barra, & Aunger, 2011; Hart, 1990; +<br/>Oaten, Stevenson, & Case, 2009; Schaller & Park, 2011); its role +<br/>in determining food preferences (Rozin & Fallon, 1987); its rela- +<br/>tion to psychiatric disorders, especially obsessive-compulsive dis- +<br/>order, phobias, and other anxiety disorders (Olatunji & McKay, +<br/>2007; Phillips, Fahy, David, & Senior, 1998); its diagnostic role in +<br/>neurological disorders such as Huntington’s disease (Spren- +<br/><b>gelmeyer et al., 1996); and, increasingly, its role in reactions to</b><br/>cheating and other social and moral infractions (Haidt, 2003; +<br/>Prinz, 2007). According to Giner-Sorolla, Bosson, Caswell, and +<br/>Hettinger (2012), disgust plays “a powerful role in shaping cultural +<br/>attitudes, policy, and law” (p. 1). Articles, books, and conferences +<br/>demonstrate a surge of vigorous scientific theorizing and research +<br/>on disgust. One result of this surge of research is that the idea of +<br/>disgust as a simple reaction is giving way to a more complex story. +<br/>As Herz (2012) summarized, “Our age, our personality, our cul- +<br/>ture, our thoughts and beliefs, our mood, our morals, whom we’re +<br/>with, where we are, and which of our senses is giving us the +<br/>Sherri C. Widen and James A. Russell, Department of Psychology, +<br/><b>Boston College</b><br/>This article was funded by Grant 1025563 from the National Science +<br/>Foundation. +<br/>We thank Nicole Nelson, Mary Kayyal, Joe Pochedly, Alyssa McCarthy, +<br/>Nicole Trauffer, Cara D’Arcy, Marissa DiGirolamo, Anne Yoder, and Erin +<br/>Heitzman for their comments on a draft of this article. +<br/>Correspondence concerning this article should be addressed to Sherri C. +<br/>Widen, Department of Psychology, McGuinn Hall, 140 Commonwealth +<br/>bc.edu +<br/>271 +<br/>feeling, all shape whether and how strongly we are able to feel +<br/>disgusted” (p. 57). +<br/>Much of the theorizing and research on disgust to date have +<br/>been guided, explicitly or implicitly, by a research program cen- +<br/>tered on the concept of basic emotions—indeed, that research +<br/>program has provided the standard account of disgust. Theories +<br/>within this research program (Ekman & Cordaro, 2011; Izard, +<br/>1971, 1994; Tomkins, 1962) place facial expressions at the center +<br/>of emotion. In this article, we question one key part of the standard +<br/>account of disgust: the hypothesis that, from an early age, a child +<br/>recognizes disgust in others from their facial expressions. Our +<br/>review finds evidence that is inconsistent with this hypothesis, and +<br/>we suggest that the field examine alternative accounts. To place +<br/>this evidence in a broader context, we also review evidence on +<br/>closely related topics, such as children’s disgust reactions, their +<br/>acquisition of a word for disgust, their inference of disgust from +<br/>nonfacial cues, and adults’ recognition of disgust from facial +<br/>expressions. +<br/>The Standard Account +<br/>The widely assumed standard account of disgust stems from the +<br/>classic work of Allport (1924) and Tomkins (1962) and those they +<br/>influenced (Ekman & Cordaro, 2011; Izard, 2011; Levenson, +<br/>2011). In this simple, elegant, and plausible account, so-called +<br/>basic emotions—including disgust— have dedicated neural cir- +<br/>cuitry, are triggered by specific releasing stimuli, and produce a +<br/>coordinated response pattern that includes specific autonomic ner- +<br/>vous system activation, a behavioral tendency, and a facial expres- +<br/>sion. Ekman, Friesen, and Ellsworth (1972) described this last +<br/>aspect of their theory as follows: +<br/>Regardless of the language, of whether the culture is Western or +<br/>Eastern, industrialized or preliterate, [certain] facial expressions are +<br/>labeled with the same emotion terms . . . Our neuro-cultural theory +<br/>postulates a facial affect program, located within the nervous system +</td><td></td><td>Avenue, Boston College, Chestnut Hill, MA 02467. E-mail: widensh@ +</td></tr><tr><td>969fd48e1a668ab5d3c6a80a3d2aeab77067c6ce</td><td>End-to-End Spatial Transform Face Detection and Recognition +<br/><b>Zhejiang University</b><br/><b>Zhejiang University</b><br/>Rokid.inc +</td><td>('39106061', 'Liying Chi', 'liying chi')<br/>('35028106', 'Hongxin Zhang', 'hongxin zhang')<br/>('9932177', 'Mingxiu Chen', 'mingxiu chen')</td><td>charrin0531@gmail.com +<br/>zhx@cad.zju.edu.cn +<br/>cmxnono@rokid.com +</td></tr><tr><td>96a9ca7a8366ae0efe6b58a515d15b44776faf6e</td><td>Grid Loss: Detecting Occluded Faces +<br/><b>Institute for Computer Graphics and Vision</b><br/><b>Graz University of Technology</b></td><td>('34847524', 'Michael Opitz', 'michael opitz')<br/>('1903921', 'Georg Waltner', 'georg waltner')<br/>('1762885', 'Georg Poier', 'georg poier')<br/>('1720811', 'Horst Possegger', 'horst possegger')<br/>('3628150', 'Horst Bischof', 'horst bischof')</td><td>{michael.opitz,waltner,poier,possegger,bischof}@icg.tugraz.at +</td></tr><tr><td>9696b172d66e402a2e9d0a8d2b3f204ad8b98cc4</td><td>J Inf Process Syst, Vol.9, No.1, March 2013 +<br/>pISSN 1976-913X +<br/>eISSN 2092-805X +<br/>Region-Based Facial Expression Recognition in +<br/>Still Images +</td><td>('2648759', 'Gawed M. Nagi', 'gawed m. nagi')<br/>('2057896', 'Fatimah Khalid', 'fatimah khalid')</td><td></td></tr><tr><td>964a3196d44f0fefa7de3403849d22bbafa73886</td><td></td><td></td><td></td></tr><tr><td>96e1ccfe96566e3c96d7b86e134fa698c01f2289</td><td>Published in Proc. of 11th IAPR International Conference on Biometrics (ICB 2018). Gold Coast, Australia, Feb. 2018 +<br/>Semi-Adversarial Networks: Convolutional Autoencoders for Imparting Privacy +<br/>to Face Images +<br/>Anoop Namboodiri 2 +<br/><b>Michigan State University, East Lansing, USA</b><br/><b>International Institute of Information Technology, Hyderabad, India</b></td><td>('5456235', 'Vahid Mirjalili', 'vahid mirjalili')<br/>('2562040', 'Sebastian Raschka', 'sebastian raschka')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td>mirjalil@msu.edu +<br/>raschkas@msu.edu +<br/>anoop@iiit.ac.in +<br/>rossarun@cse.msu.edu +</td></tr><tr><td>96f4a1dd1146064d1586ebe86293d02e8480d181</td><td>COMPARATIVE ANALYSIS OF RERANKING +<br/>TECHNIQUES FOR WEB IMAGE SEARCH +<br/><b>Pune Institute of Computer Technology, Pune, ( India</b></td><td></td><td></td></tr><tr><td>9606b1c88b891d433927b1f841dce44b8d3af066</td><td>Principal Component Analysis with Tensor Train +<br/>Subspace +</td><td>('2329741', 'Wenqi Wang', 'wenqi wang')<br/>('1732805', 'Vaneet Aggarwal', 'vaneet aggarwal')<br/>('1980683', 'Shuchin Aeron', 'shuchin aeron')</td><td></td></tr><tr><td>9627f28ea5f4c389350572b15968386d7ce3fe49</td><td>Load Balanced GANs for Multi-view Face Image Synthesis +<br/>1National Laboratory of Pattern Recognition, CASIA +<br/>2Center for Research on Intelligent Perception and Computing, CASIA +<br/>3Center for Excellence in Brain Science and Intelligence Technology, CAS +<br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b><br/>5Noah’s Ark Lab of Huawei Technologies +</td><td>('1680853', 'Jie Cao', 'jie cao')<br/>('49995036', 'Yibo Hu', 'yibo hu')<br/>('49828394', 'Bing Yu', 'bing yu')<br/>('1705643', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>{jie.cao,yibo.hu}@cripac.ia.ac.cn, yubing5@huawei.com, {rhe, znsun}@nlpr.ia.ac.cn +</td></tr><tr><td>966e36f15b05ef8436afecf57a97b73d6dcada94</td><td>Dimensionality Reduction using Relative +<br/>Attributes +<br/><b>Institute for Human-Machine Communication, Technische Universit at M unchen</b><br/><b>Iran</b><br/><b>The Remote Sensing Technology Institute (IMF), German Aerospace Center</b><br/>1 Introduction +<br/>Visual attributes are high-level semantic description of visual data that are close +<br/>to the language of human. They have been intensively used in various appli- +<br/>cations such as image classification [1,2], active learning [3,4], and interactive +<br/>search [5]. However, the usage of attributes in dimensionality reduction has not +<br/>been considered yet. In this work, we propose to utilize relative attributes as +<br/>semantic cues in dimensionality reduction. To this end, we employ Non-negative +<br/>Matrix Factorization (NMF) [6] constrained by embedded relative attributes to +<br/>come up with a new algorithm for dimensionality reduction, namely attribute +<br/>regularized NMF (ANMF). +<br/>2 Approach +<br/>We assume that X ∈ RD×N denotes N data points (e.g., images) represented by +<br/>D dimensional low-level feature vectors. The NMF decomposes the non-negative +<br/>matrix X into two non-negative matrices U ∈ RD×K and V ∈ RN×K such that +<br/>the multiplication of U and V approximates the original matrix X. Here, U +<br/>represents the bases and V contains the coefficients, which are considered as +<br/>new representation of the original data. The NMF objective function is: +<br/>F =(cid:13)(cid:13)X − U V T(cid:13)(cid:13)2 +<br/>s.t. U = [uik] ≥ 0 +<br/>V = [vjk] ≥ 0. +<br/>(1) +<br/>Additionally, we assume that M semantic attributes have been predefined +<br/>for the data and the relative attributes of each image are available. Precisely, +<br/>the matrix of relative attributes, Q ∈ RM×N , has been learned by some ranking +<br/>function (e,.g, rankSVM). Intuitively, those images which own similar relative +<br/>attributes have similar semantic contents and therefore belong to the same se- +<br/>mantic class. This concept can be formulated as a regularizer to be added to the +</td><td>('2133342', 'Mohammadreza Babaee', 'mohammadreza babaee')<br/>('2165157', 'Stefanos Tsoukalas', 'stefanos tsoukalas')<br/>('3281049', 'Maryam Babaee', 'maryam babaee')<br/>('1705843', 'Gerhard Rigoll', 'gerhard rigoll')<br/>('1777167', 'Mihai Datcu', 'mihai datcu')</td><td>{reza.babaee,rigoll}@tum.de, s.tsoukalas@mytum.de +<br/>babaee@eng.ui.ac.ir +<br/>mihai.datcu@dlr.de +</td></tr><tr><td>96b1000031c53cd4c1c154013bb722ffd87fa7da</td><td>ContextVP: Fully Context-Aware Video +<br/>Prediction +<br/>1 NVIDIA, Santa Clara, CA, USA +<br/>2 ETH Zurich, Zurich, Switzerland +<br/>3 The Swiss AI Lab IDSIA, Manno, Switzerland +<br/>4 NNAISENSE, Lugano, Switzerland +</td><td>('2387035', 'Wonmin Byeon', 'wonmin byeon')<br/>('1794816', 'Qin Wang', 'qin wang')<br/>('2100612', 'Rupesh Kumar Srivastava', 'rupesh kumar srivastava')<br/>('1802604', 'Petros Koumoutsakos', 'petros koumoutsakos')</td><td>wbyeon@nvidia.com +</td></tr><tr><td>96578785836d7416bf2e9c154f687eed8f93b1e4</td><td>Automated video-based facial expression analysis +<br/>of neuropsychiatric disorders +<br/><b>a Section of Biomedical Image Analysis, University of Pennsylvania, 3600 Market, Suite 380, Philadelphia, PA 19104, USA</b><br/><b>b Brain Behavior Center, University of Pennsylvania Medical Center, Hospital of the University of Pennsylvania</b><br/>3400 Spruce Street, 10th Floor Gates Building Philadelphia, PA 19104, USA +<br/><b>c School of Arts and Sciences, University of Pennsylvania Medical Center, Hospital of the University of Pennsylvania</b><br/><b>University of Pennsylvania Medical Center, Hospital of the University of Pennsylvania</b><br/>3400 Spruce Street, 10th Floor Gates Building Philadelphia, PA 19104, USA +<br/>3400 Spruce Street, 10th Floor Gates Building Philadelphia, PA 19104, USA +<br/><b>University of Pennsylvania Medical Center, Hospital of the University of Pennsylvania</b><br/><b>f Neuropsychiatry Section, University of Pennsylvania Medical Center, Hospital of the University of Pennsylvania</b><br/>3400 Spruce Street, 10th Floor Gates Building Philadelphia, PA 19104, USA +<br/>3400 Spruce Street, 10th Floor Gates Building Philadelphia, PA 19104, USA +<br/>Received 16 July 2007; received in revised form 20 September 2007; accepted 20 September 2007 +</td><td>('37761073', 'Peng Wang', 'peng wang')<br/>('28501509', 'Frederick Barrett', 'frederick barrett')<br/>('2953329', 'Elizabeth Martin', 'elizabeth martin')<br/>('5747394', 'Marina Milonova', 'marina milonova')<br/>('1826037', 'Christian Kohler', 'christian kohler')<br/>('7467718', 'Ragini Verma', 'ragini verma')</td><td></td></tr><tr><td>96e0cfcd81cdeb8282e29ef9ec9962b125f379b0</td><td>The MegaFace Benchmark: 1 Million Faces for Recognition at Scale +<br/>Department of Computer Science and Engineering +<br/><b>University of Washington</b><br/>(a) FaceScrub + MegaFace +<br/>(b) FGNET + MegaFace +<br/>Figure 1. The MegaFace challenge evaluates identification and verification as a function of increasing number of gallery distractors (going +<br/>from 10 to 1 Million). We use two different probe sets (a) FaceScrub–photos of celebrities, (b) FGNET–photos with a large variation in +<br/>age per person. We present rank-1 identification of state of the art algorithms that participated in our challenge. On the left side of each +<br/>plot is current major benchmark LFW scale (i.e., 10 distractors, see how all the top algorithms are clustered above 95%). On the right is +<br/>mega-scale (with a million distractors). Observe that rates drop with increasing numbers of distractors, even though the probe set is fixed, +<br/>and that algorithms trained on larger sets (dashed lines) generally perform better. Participate at: http://megaface.cs.washington.edu. +</td><td>('2419955', 'Ira Kemelmacher-Shlizerman', 'ira kemelmacher-shlizerman')<br/>('1679223', 'Steven M. Seitz', 'steven m. seitz')<br/>('2721528', 'Evan Brossard', 'evan brossard')</td><td></td></tr><tr><td>968f472477a8afbadb5d92ff1b9c7fdc89f0c009</td><td>Firefly-based Facial Expression Recognition +</td><td></td><td></td></tr><tr><td>96c6f50ce8e1b9e8215b8791dabd78b2bbd5f28d</td><td>Dynamic Attention-controlled Cascaded Shape Regression Exploiting Training +<br/>Data Augmentation and Fuzzy-set Sample Weighting +<br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, Guildford GU2 7XH, UK</b><br/><b>School of IoT Engineering, Jiangnan University, Wuxi 214122, China</b></td><td>('2976854', 'Zhen-Hua Feng', 'zhen-hua feng')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td>{z.feng, j.kittler, w.christmas, p.huber}@surrey.ac.uk, wu xiaojun@jiangnan.edu.cn +</td></tr><tr><td>96e731e82b817c95d4ce48b9e6b08d2394937cf8</td><td>Unconstrained Face Verification using Deep CNN Features +<br/><b>University of Maryland, College Park</b><br/><b>Rutgers, The State University of New Jersey</b></td><td>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>pullpull@cs.umd.edu, vishal.m.patel@rutgers.edu, rama@umiacs.umd.edu +</td></tr><tr><td>9686dcf40e6fdc4152f38bd12b929bcd4f3bbbcc</td><td>International Journal of Engineering Research and General Science Volume 3, Issue 1, January-February, 2015 +<br/>ISSN 2091-2730 +<br/>Emotion Based Music Player +<br/>1Department of Computer Science and Engineering +<br/>2Department of Computer Science and Engineering +<br/>3Department of Computer Science and Engineering +<br/>4Asst. Professor, Department of Computer Science and Engineering +<br/><b>M.H Saboo Siddik College of Engineering, University of Mumbai, India</b></td><td>('9928295', 'Sharik Khan', 'sharik khan')<br/>('1762886', 'Omar Khan', 'omar khan')<br/>('16079307', 'Shabana Tadvi', 'shabana tadvi')</td><td>Email:-kabani152@gmail.com +</td></tr><tr><td>9636c7d3643fc598dacb83d71f199f1d2cc34415</td><td></td><td></td><td></td></tr><tr><td>3a27d164e931c422d16481916a2fa6401b74bcef</td><td>Anti-Makeup: Learning A Bi-Level Adversarial Network for Makeup-Invariant +<br/>Face Verification +<br/>National Laboratory of Pattern Recognition, CASIA +<br/>Center for Research on Intelligent Perception and Computing, CASIA +<br/>Center for Excellence in Brain Science and Intelligence Technology, CAS +<br/><b>University of Chinese Academy of Sciences, Beijing 100190, China</b></td><td>('2496686', 'Yi Li', 'yi li')<br/>('3051419', 'Lingxiao Song', 'lingxiao song')<br/>('2225749', 'Xiang Wu', 'xiang wu')<br/>('1705643', 'Ran He', 'ran he')<br/>('1688870', 'Tieniu Tan', 'tieniu tan')</td><td>yi.li@cripac.ia.ac.cn, {lingxiao.song, rhe, tnt}@nlpr.ia.ac.cn, alfredxiangwu@gmail.com +</td></tr><tr><td>3af8d38469fb21368ee947d53746ea68cd64eeae</td><td>Multimodal Intelligent Affect Detection with Kinect +<br/>(Doctoral Consortium) +<br/><b>Northumbria University</b><br/>United Kingdom +<br/><b>Northumbria University</b><br/>United Kingdom +<br/><b>Northumbria University</b><br/>United Kingdom +</td><td>('1886853', 'Li Zhang', 'li zhang')<br/>('2004913', 'Alamgir Hossain', 'alamgir hossain')<br/>('39617655', 'Yang Zhang', 'yang zhang')</td><td>li.zhang@northumbria.ac.uk +<br/>Yang4.zhang@northumbria.ac.uk +</td></tr><tr><td>3a2fc58222870d8bed62442c00341e8c0a39ec87</td><td>Probabilistic Local Variation +<br/>Segmentation +<br/>Technion - Computer Science Department - M.Sc. Thesis MSC-2014-02 - 2014 </td><td>('3139600', 'Michael Baltaxe', 'michael baltaxe')</td><td></td></tr><tr><td>3a3f75e0ffdc0eef07c42b470593827fcd4020b4</td><td>NORMAL SIMILARITY NETWORK FOR GENERATIVE MODELLING +<br/><b>School of Computing, National University of Singapore</b></td><td>('40456486', 'Jay Nandy', 'jay nandy')<br/>('1725063', 'Wynne Hsu', 'wynne hsu')</td><td></td></tr><tr><td>3a76e9fc2e89bdd10a9818f7249fbf61d216efc4</td><td>Face Sketch Matching via Coupled Deep Transform Learning +<br/><b>IIIT-Delhi, India, 2West Virginia University</b></td><td>('1925017', 'Shruti Nagpal', 'shruti nagpal')<br/>('2220719', 'Maneet Singh', 'maneet singh')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('2487227', 'Afzel Noore', 'afzel noore')<br/>('2641605', 'Angshul Majumdar', 'angshul majumdar')</td><td>{shrutin, maneets, rsingh, mayank, angshul}@iiitd.ac.in, afzel.noore@mail.wvu.edu +</td></tr><tr><td>3a2c90e0963bfb07fc7cd1b5061383e9a99c39d2</td><td>End-to-End Deep Learning for Steering Autonomous +<br/>Vehicles Considering Temporal Dependencies +<br/><b>The American University in Cairo, Egypt</b><br/>2Valeo Schalter und Sensoren GmbH, Germany +</td><td>('2150605', 'Hesham M. Eraqi', 'hesham m. eraqi')<br/>('2233511', 'Mohamed N. Moustafa', 'mohamed n. moustafa')<br/>('11300101', 'Jens Honer', 'jens honer')</td><td></td></tr><tr><td>3a0ea368d7606030a94eb5527a12e6789f727994</td><td>Categorization by Learning +<br/>and Combining Object Parts +<br/> +<br/>Tomaso Poggio +<br/><b>Honda RandD Americas, Inc., Boston, MA, USA</b><br/><b>University of Siena, Siena, Italy</b><br/><b>Computer Graphics Research Group, University of Freiburg, Freiburg, Germany</b><br/> heisele,serre,tp +</td><td>('1684626', 'Bernd Heisele', 'bernd heisele')</td><td>@ai.mit.edu pontil@ing.unisi.it +<br/>vetter@informatik.uni-freiburg.de +</td></tr><tr><td>3a804cbf004f6d4e0b041873290ac8e07082b61f</td><td>Language-Action Tools for Cognitive Artificial Agents: Papers from the 2011 AAAI Workshop (WS-11-14) +<br/>A Corpus-Guided Framework for Robotic Visual Perception +<br/><b>University of Maryland Institute for Advanced Computer Studies, College Park, MD</b></td><td>('7607499', 'Yezhou Yang', 'yezhou yang')<br/>('1697493', 'Yiannis Aloimonos', 'yiannis aloimonos')</td><td>{cteo, yzyang, hal, fer, yiannis}@umiacs.umd.edu +</td></tr><tr><td>3a04eb72aa64760dccd73e68a3b2301822e4cdc3</td><td>Scalable Sparse Subspace Clustering +<br/><b>Machine Intelligence Laboratory, College of Computer Science, Sichuan University</b><br/>Chengdu, 610065, China. +</td><td>('8249791', 'Xi Peng', 'xi peng')<br/>('36794849', 'Lei Zhang', 'lei zhang')<br/>('9276020', 'Zhang Yi', 'zhang yi')</td><td>pangsaai@gmail.com, {leizhang, zhangyi}@scu.edu.cn +</td></tr><tr><td>3af130e2fd41143d5fc49503830bbd7bafd01f8b</td><td>How Do We Evaluate the Quality of Computational Editing Systems? +<br/>1 Inria, Univ. Grenoble Alpes & CNRS (LJK), Grenoble, France +<br/><b>University of Wisconsin-Madison, Madison, WI, USA</b></td><td>('2869929', 'Christophe Lino', 'christophe lino')<br/>('1810286', 'Quentin Galvane', 'quentin galvane')<br/>('1776507', 'Michael Gleicher', 'michael gleicher')</td><td></td></tr><tr><td>3a2cf589f5e11ca886417b72c2592975ff1d8472</td><td>Spontaneously Emerging Object Part Segmentation +<br/>Machine Learning Department +<br/><b>Carnegie Mellon University</b><br/>Machine Learning Department +<br/><b>Carnegie Mellon University</b></td><td>('1696365', 'Yijie Wang', 'yijie wang')<br/>('1705557', 'Katerina Fragkiadaki', 'katerina fragkiadaki')</td><td>yijiewang@cmu.edu +<br/>katef@cs.cmu.edu +</td></tr><tr><td>3ada7640b1c525056e6fcd37eea26cd638815cd6</td><td>Abnormal Object Recognition: +<br/>A Comprehensive Study +<br/><b>Rutgers University</b><br/><b>University of Washington</b></td><td>('3139794', 'Babak Saleh', 'babak saleh')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')</td><td></td></tr><tr><td>3abc833f4d689f37cc8a28f47fb42e32deaa4b17</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Large Scale Retrieval and Generation of Image Descriptions +<br/>Received: date / Accepted: date +</td><td>('2004053', 'Vicente Ordonez', 'vicente ordonez')<br/>('38390487', 'Margaret Mitchell', 'margaret mitchell')<br/>('34176020', 'Jesse Dodge', 'jesse dodge')<br/>('1699545', 'Yejin Choi', 'yejin choi')</td><td></td></tr><tr><td>3acb6b3e3f09f528c88d5dd765fee6131de931ea</td><td>(cid:49)(cid:50)(cid:57)(cid:40)(cid:47)(cid:3)(cid:53)(cid:40)(cid:51)(cid:53)(cid:40)(cid:54)(cid:40)(cid:49)(cid:55)(cid:36)(cid:55)(cid:44)(cid:50)(cid:49)(cid:3)(cid:41)(cid:50)(cid:53)(cid:3)(cid:39)(cid:53)(cid:44)(cid:57)(cid:40)(cid:53)(cid:3)(cid:40)(cid:48)(cid:50)(cid:55)(cid:44)(cid:50)(cid:49)(cid:3)(cid:53)(cid:40)(cid:38)(cid:50)(cid:42)(cid:49)(cid:44)(cid:55)(cid:44)(cid:50)(cid:49)(cid:3)(cid:3) +<br/>(cid:44)(cid:49)(cid:3)(cid:48)(cid:50)(cid:55)(cid:50)(cid:53)(cid:3)(cid:57)(cid:40)(cid:43)(cid:44)(cid:38)(cid:47)(cid:40)(cid:3)(cid:57)(cid:44)(cid:39)(cid:40)(cid:50)(cid:54)(cid:3) +<br/>(cid:53)(cid:68)(cid:77)(cid:78)(cid:88)(cid:80)(cid:68)(cid:85)(cid:3)(cid:55)(cid:75)(cid:72)(cid:68)(cid:74)(cid:68)(cid:85)(cid:68)(cid:77)(cid:68)(cid:81)(cid:13)(cid:15)(cid:3)(cid:37)(cid:76)(cid:85)(cid:3)(cid:37)(cid:75)(cid:68)(cid:81)(cid:88)(cid:13)(cid:15)(cid:3)(cid:36)(cid:79)(cid:69)(cid:72)(cid:85)(cid:87)(cid:3)(cid:38)(cid:85)(cid:88)(cid:93)(cid:130)(cid:15)(cid:3)(cid:37)(cid:72)(cid:79)(cid:76)(cid:81)(cid:71)(cid:68)(cid:3)(cid:47)(cid:72)(cid:13)(cid:15)(cid:3)(cid:36)(cid:86)(cid:82)(cid:81)(cid:74)(cid:88)(cid:3)(cid:55)(cid:68)(cid:80)(cid:69)(cid:82)(cid:13)(cid:3) +<br/>(cid:3) +<br/><b>Center for Research in Intelligent Systems, University of California, Riverside, CA 92521, USA</b><br/><b>cid:130) Computer Perception Lab, California State University, Bakersfield, CA 93311, USA</b><br/>(cid:36)(cid:37)(cid:54)(cid:55)(cid:53)(cid:36)(cid:38)(cid:55)(cid:3) +<br/>the background +<br/>(cid:3) +<br/>A novel feature representation of human facial expressions +<br/>for emotion recognition is developed. The representation +<br/>leveraged +<br/>texture removal ability of +<br/>Anisotropic Inhibited Gabor Filtering (AIGF) with the +<br/>compact representation of spatiotemporal +<br/>local binary +<br/>patterns. The emotion recognition system incorporated face +<br/>detection and registration followed by the proposed feature +<br/>representation: Local Anisotropic Inhibited Binary Patterns +<br/>in Three Orthogonal +<br/>and +<br/>classification. The system is evaluated on videos from Motor +<br/>(cid:55)(cid:85)(cid:72)(cid:81)(cid:71)(cid:3)(cid:48)(cid:68)(cid:74)(cid:68)(cid:93)(cid:76)(cid:81)(cid:72)(cid:182)(cid:86)(cid:3)(cid:37)(cid:72)(cid:86)(cid:87)(cid:3)(cid:39)(cid:85)(cid:76)(cid:89)(cid:72)(cid:85)(cid:3)(cid:38)(cid:68)(cid:85)(cid:3)(cid:82)(cid:73)(cid:3)(cid:87)(cid:75)(cid:72)(cid:3)(cid:60)(cid:72)(cid:68)(cid:85) 2014-2016. +<br/>The results showed improved performance compared to +<br/>other state-of-the-art feature representations.(cid:3) +<br/>(LAIBP-TOP) +<br/>Index(cid:3)Terms(cid:178)(cid:3)Facial expression, emotion recognition, +<br/>feature extraction, background texture, anisotropic Gabor +<br/>filter.(cid:3) +<br/>(cid:3) +<br/>Planes +<br/>(cid:20)(cid:17)(cid:3)(cid:44)(cid:49)(cid:55)(cid:53)(cid:50)(cid:39)(cid:56)(cid:38)(cid:55)(cid:44)(cid:50)(cid:49)(cid:3) +<br/>Facial expressions are crucial to non-verbal communication +<br/>of emotion. Automatic facial emotion recognition software +<br/>has applications in lie detection, human behavior analysis, +<br/>medical applications, and human-computer interfaces. We +<br/>develop a system to detect stress and inattention of a motor +<br/>vehicle operator from a single camera. Previous work in +<br/>observation of motor vehicle operators employed multiple +<br/>cameras for 3-D reconstruction [1], but multi-camera +<br/>systems may introduce too much complexity and too many +<br/>constraints in the design of a system. Another possible +<br/>solution is gaze, but as of yet there is no consensus on how +<br/>to detect inattention from gaze [2]. The goal of our work is a +<br/>system that can extrapolate high stress and inattention from +<br/>valence and arousal measurements on a low-cost platform so +<br/>as to prevent motor vehicle accidents. +<br/> To this end, we present a novel dynamic local appearance +<br/>feature that can compactly describe the spatiotemporal +<br/>behavior of a local neighborhood in the video. The method +<br/>is based on Local Binary Patterns in Three Orthogonal +<br/>Planes (LBP-TOP) [3] and background suppressing Gabor +<br/>Energy filtering [4], but it is significantly different. We +<br/>demonstrate that the background suppression concept can be +<br/>applied to LBP-TOP to improve performance. The system is +<br/>tested on three data sets collected from the Motor Trend +<br/>(cid:48)(cid:68)(cid:74)(cid:68)(cid:93)(cid:76)(cid:81)(cid:72)(cid:182)(cid:86)(cid:3) (cid:37)(cid:72)(cid:86)(cid:87)(cid:3) (cid:39)(cid:85)(cid:76)(cid:89)(cid:72)(cid:85)(cid:3) (cid:38)(cid:68)r of the Year 2014, 2015 and +<br/>2016. +<br/>(cid:21)(cid:17)(cid:3)(cid:53)(cid:40)(cid:47)(cid:36)(cid:55)(cid:40)(cid:39)(cid:3)(cid:58)(cid:50)(cid:53)(cid:46)(cid:3)(cid:36)(cid:49)(cid:39)(cid:3)(cid:38)(cid:50)(cid:49)(cid:55)(cid:53)(cid:44)(cid:37)(cid:56)(cid:55)(cid:44)(cid:50)(cid:49)(cid:54)(cid:3) +<br/>(cid:3) +<br/>The current challenge to dynamic facial emotion recognition +<br/>is the detection of emotion despite the various extrinsic and +<br/>intrinsic imaging conditions, and intra-personnel differences +<br/>in expression. While deep learning has been a growing trend +<br/>in image processing and computer vision, the effects of +<br/>transfer learning (cid:178) using expression data from other +<br/>datasets [5] (cid:178) are diminished possibly because of various +<br/>factors [6]. Thus, hand-crafted features, not learned from the +<br/>neural networks, are still of great interest to unconstrained +<br/>facial emotion recognition. This work focuses on the +<br/>development of a novel hand-crafted feature representation. +<br/> Local Binary Pattern (LBP) is the most commonly used +<br/>appearance-based feature extraction method [7]. LBP is a +<br/>static texture descriptor and is not suitable for dynamic +<br/>facial expressions in videos. +<br/> A variation of LBP, Volume Local Binary Patterns +<br/>(VLBP), was developed to capture dynamic textures [8]. +<br/>VLBP uses 3 parallel planes in the spatiotemporal domain +<br/>where the center pixel is on the center plane, and it records +<br/>the dynamic patterns in the neighborhood of each pixel into +<br/>a (3(cid:81)+2) dimensional histogram, where (cid:81) is the number of +<br/>neighboring pixels. +<br/> The high dimensionality of VLBP is 23(cid:81)+2, makes it +<br/>impractical to use due to the rapid increase in dimensionality +<br/>as the size of the neighborhood increases. An alternate +<br/>solution to VLBP is the Local Binary Patterns in Three +<br/>Orthogonal Planes (LBP-TOP). The dimensionality of LBP- +<br/>TOP (3*2(cid:81)) is significantly lower than VLBP. The working +<br/>of LBP-TOP is described in section 3. +<br/> The other major type of appearance feature is the Gabor +<br/>filter. Traditional Gabor filters are +<br/>in +<br/>unconstrained settings; it captures all edges within an image, +<br/>noise included. Cruz (cid:72)(cid:87)(cid:3) (cid:68)(cid:79)(cid:17)(cid:3) [4] proposed Anisotropic +<br/>Inhibited Gabor Filter (AIGF) that is robust to background +<br/>noise and computationally efficient. Almaev (cid:72)(cid:87)(cid:3) (cid:68)(cid:79)(cid:17)(cid:3) [9] +<br/>too sensitive +<br/>(cid:28)(cid:26)(cid:27)(cid:16)(cid:20)(cid:16)(cid:24)(cid:19)(cid:28)(cid:19)(cid:16)(cid:21)(cid:20)(cid:26)(cid:24)(cid:16)(cid:27)(cid:18)(cid:20)(cid:26)(cid:18)(cid:7)(cid:22)(cid:20)(cid:17)(cid:19)(cid:19)(cid:3)(cid:139)(cid:21)(cid:19)(cid:20)(cid:26)(cid:3)(cid:44)(cid:40)(cid:40)(cid:40) +<br/>(cid:27)(cid:20)(cid:19) +<br/>(cid:44)(cid:38)(cid:44)(cid:51)(cid:3)(cid:21)(cid:19)(cid:20)(cid:26) +</td><td></td><td></td></tr><tr><td>3a60678ad2b862fa7c27b11f04c93c010cc6c430</td><td>JANUARY-MARCH 2012 +<br/>A Multimodal Database for +<br/>Affect Recognition and Implicit Tagging +</td><td>('2463695', 'Mohammad Soleymani', 'mohammad soleymani')<br/>('2796371', 'Jeroen Lichtenauer', 'jeroen lichtenauer')<br/>('1809085', 'Thierry Pun', 'thierry pun')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>3a591a9b5c6d4c62963d7374d58c1ae79e3a4039</td><td>Driver Cell Phone Usage Detection From HOV/HOT NIR Images +<br/><b>Xerox Research Center Webster</b><br/>800 Phillips Rd. Webster NY 14580 +</td><td>('1762503', 'Yusuf Artan', 'yusuf artan')<br/>('2415287', 'Orhan Bulan', 'orhan bulan')<br/>('1736673', 'Robert P. Loce', 'robert p. loce')<br/>('5942563', 'Peter Paul', 'peter paul')</td><td>yusuf.artan,orhan.bulan,robert.loce,peter.paul@xerox.com +</td></tr><tr><td>3aa9c8c65ce63eb41580ba27d47babb1100df8a3</td><td>Annals of the +<br/><b>University of North Carolina Wilmington</b><br/>Master of Science in +<br/>Computer Science and Information Systems +</td><td></td><td></td></tr><tr><td>3a0a839012575ba455f2b84c2d043a35133285f9</td><td>444 +<br/>Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, pages 444–454, +<br/>Edinburgh, Scotland, UK, July 27–31, 2011. c(cid:13)2011 Association for Computational Linguistics +</td><td></td><td></td></tr><tr><td>3af1a375c7c1decbcf5c3a29774e165cafce390c</td><td>Quantifying Facial Expression Abnormality in Schizophrenia by Combining +<br/>2D and 3D Features +<br/>1 Department of Radiology +<br/><b>University of Pennsylvania</b><br/>2 Department of Psychiatry +<br/><b>University of Pennsylvania</b><br/>Philadelphia, PA 19104 +<br/>Philadelphia, PA 19104 +</td><td>('1722767', 'Peng Wang', 'peng wang')<br/>('15741672', 'Fred Barrett', 'fred barrett')<br/>('7467718', 'Ragini Verma', 'ragini verma')</td><td>{wpeng@ieee.org, ragini.verma@uphs.upenn.edu } +<br/>{kohler, fbarrett, raquel, gur}@bbl.med.upenn.edu +</td></tr><tr><td>3a9681e2e07be7b40b59c32a49a6ff4c40c962a2</td><td>Biometrics & Biostatistics International Journal +<br/>Comparing treatment means: overlapping standard +<br/>errors, overlapping confidence intervals, and tests of +<br/>hypothesis +</td><td></td><td></td></tr><tr><td>3a846704ef4792dd329a5c7a2cb8b330ab6b8b4e</td><td>in any current or +<br/>future media, +<br/>for all other uses, +<br/>© 2010 IEEE. Personal use of this material is permitted. Permission from IEEE must be +<br/>obtained +<br/>including +<br/>reprinting/republishing this material for advertising or promotional purposes, creating +<br/>new collective works, for resale or redistribution to servers or lists, or reuse of any +<br/>copyrighted component of this work in other works. +<br/>Pre-print of article that appeared at the IEEE Computer Society Workshop on Biometrics +<br/>2010. +<br/>The published article can be accessed from: +<br/>http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5544597 +</td><td></td><td></td></tr><tr><td>3a2a37ca2bdc82bba4c8e80b45d9f038fe697c7d</td><td>Handling Uncertain Tags in Visual Recognition +<br/><b>School of Computing Science, Simon Fraser University, Canada</b></td><td>('3214848', 'Arash Vahdat', 'arash vahdat')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td>{avahdat, mori}@cs.sfu.ca +</td></tr><tr><td>3a95eea0543cf05670e9ae28092a114e3dc3ab5c</td><td>Constructing the L2-Graph for Robust Subspace +<br/>Learning and Subspace Clustering +</td><td>('8249791', 'Xi Peng', 'xi peng')<br/>('1751019', 'Zhiding Yu', 'zhiding yu')<br/>('3134548', 'Huajin Tang', 'huajin tang')<br/>('9276020', 'Zhang Yi', 'zhang yi')</td><td></td></tr><tr><td>3a4f522fa9d2c37aeaed232b39fcbe1b64495134</td><td>ISSN (Online) 2321 – 2004 +<br/>ISSN (Print) 2321 – 5526 +<br/> INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN ELECTRICAL, ELECTRONICS, INSTRUMENTATION AND CONTROL ENGINEERING +<br/> Vol. 4, Issue 5, May 2016 +<br/>IJIREEICE +<br/>Face Recognition and Retrieval Using Cross +<br/>Age Reference Coding +<br/>Sricharan H S1, Srinidhi K S1, Rajath D N1, Tejas J N1, Chandrakala B M2 +<br/> BE, DSCE, Bangalore1 +<br/>Assistant Professor, DSCE, Bangalore2 +</td><td></td><td></td></tr><tr><td>54948ee407b5d32da4b2eee377cc44f20c3a7e0c</td><td>Right for the Right Reason: Training Agnostic +<br/>Networks +<br/><b>Intelligent Systems Laboratory, University of Bristol, Bristol BS8 1UB, UK</b><br/>use of classifiers in “out of domain” situations, a problem that +<br/>leads to research questions in domain adaptation [6], [18]. +<br/>Other concerns are also created around issues of bias, e.g. +<br/>classifiers incorporating biases that are present in the data +<br/>and are not intended to be used [2], which run the risk of +<br/>reinforcing or amplifying cultural (and other) biases [20]. +<br/>Therefore, both predictive accuracy and fairness are heavily +<br/>influenced by the choices made when developing black-box +<br/>machine-learning models. +</td><td>('1805367', 'Sen Jia', 'sen jia')<br/>('2031978', 'Thomas Lansdall-Welfare', 'thomas lansdall-welfare')<br/>('1685083', 'Nello Cristianini', 'nello cristianini')</td><td>{sen.jia, thomas.lansdall-welfare, nello.cristianini}@bris.ac.uk +</td></tr><tr><td>540b39ba1b8ef06293ed793f130e0483e777e278</td><td>ORIGINAL RESEARCH +<br/>published: 13 July 2018 +<br/>doi: 10.3389/fpsyg.2018.01191 +<br/>Biologically Inspired Emotional +<br/>Expressions for Artificial Agents +<br/><b>Optics and Engineering Informatics, Budapest University of Technology and Economics</b><br/><b>Budapest, Hungary, E tv s Lor nd University, Budapest, Hungary, 3 Institute for Computer Science</b><br/><b>and Control, Hungarian Academy of Sciences, Budapest, Hungary, Chuo University</b><br/>Tokyo, Japan, 5 MTA-ELTE Comparative Ethology Research Group, Budapest, Hungary, 6 Department of Telecommunications +<br/><b>and Media Informatics, Budapest University of Technology and Economics, Budapest, Hungary</b><br/>A special area of human-machine interaction, +<br/>the expression of emotions gains +<br/>importance with the continuous development of artificial agents such as social robots or +</td><td>('31575111', 'Beáta Korcsok', 'beáta korcsok')<br/>('3410664', 'Veronika Konok', 'veronika konok')<br/>('10791722', 'György Persa', 'györgy persa')<br/>('2725581', 'Tamás Faragó', 'tamás faragó')<br/>('1701851', 'Mihoko Niitsuma', 'mihoko niitsuma')<br/>('1769570', 'Péter Baranyi', 'péter baranyi')<br/>('3131165', 'Márta Gácsi', 'márta gácsi')</td><td></td></tr><tr><td>54bb25a213944b08298e4e2de54f2ddea890954a</td><td>AgeDB: the first manually collected, in-the-wild age database +<br/><b>Imperial College London</b><br/><b>Imperial College London</b><br/><b>Imperial College London, On do</b><br/><b>Imperial College London</b><br/><b>Middlesex University London</b><br/><b>Imperial College London</b></td><td>('24278037', 'Stylianos Moschoglou', 'stylianos moschoglou')<br/>('40598566', 'Athanasios Papaioannou', 'athanasios papaioannou')<br/>('3320415', 'Christos Sagonas', 'christos sagonas')<br/>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('1754270', 'Irene Kotsia', 'irene kotsia')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>s.moschoglou@imperial.ac.uk +<br/>a.papaioannou11@imperial.ac.uk +<br/>c.sagonas@imperial.ac.uk +<br/>j.deng16@imperial.ac.uk +<br/>i.kotsia@mdx.ac.uk +<br/>s.zafeiriou@imperial.ac.uk +</td></tr><tr><td>54bae57ed37ce50e859cbc4d94d70cc3a84189d5</td><td>FACE RECOGNITION COMMITTEE MACHINE +<br/>Department of Computer Science and Engineering +<br/><b>The Chinese University of Hong Kong</b><br/>Shatin, Hong Kong +</td><td>('2899702', 'Ho-Man Tang', 'ho-man tang')<br/>('1681775', 'Michael R. Lyu', 'michael r. lyu')<br/>('1706259', 'Irwin King', 'irwin king')</td><td> +</td></tr><tr><td>54f442c7fa4603f1814ebd8eba912a00dceb5cb2</td><td>The Indian Buffet Process: +<br/>Scalable Inference and Extensions +<br/>A Thesis +<br/>Presented to the Fellowship of +<br/><b>The University of Cambridge</b><br/>in Candidacy for the Degree of +<br/>Master of Science +<br/>Department of Engineering +<br/>Zoubin Ghahramani, supervisor +<br/>August 2009 +</td><td>('2292194', 'Finale Doshi-Velez', 'finale doshi-velez')</td><td></td></tr><tr><td>543f21d81bbea89f901dfcc01f4e332a9af6682d</td><td>Published as a conference paper at ICLR 2016 +<br/>UNSUPERVISED AND SEMI-SUPERVISED LEARNING +<br/>WITH CATEGORICAL GENERATIVE ADVERSARIAL +<br/>NETWORKS +<br/><b>University of Freiburg</b><br/>79110 Freiburg, Germany +</td><td>('2060551', 'Jost Tobias Springenberg', 'jost tobias springenberg')</td><td>springj@cs.uni-freiburg.de +</td></tr><tr><td>54969bcd728b0f2d3285866c86ef0b4797c2a74d</td><td>IEEE TRANSACTION SUBMISSION +<br/>Learning for Video Compression +</td><td>('31482866', 'Zhibo Chen', 'zhibo chen')<br/>('50258851', 'Tianyu He', 'tianyu he')<br/>('50562569', 'Xin Jin', 'xin jin')<br/>('1697194', 'Feng Wu', 'feng wu')</td><td></td></tr><tr><td>5456166e3bfe78a353df988897ec0bd66cee937f</td><td>Improved Boosting Performance by Exclusion +<br/>of Ambiguous Positive Examples +<br/>Computer Vision and Active Perception, KTH, Stockholm 10800, Sweden +<br/>Keywords: +<br/>Boosting, Image Classification, Algorithm Evaluation, Dataset Pruning, VOC2007. +</td><td>('1750517', 'Miroslav Kobetski', 'miroslav kobetski')<br/>('1736906', 'Josephine Sullivan', 'josephine sullivan')</td><td>{kobetski, sullivan}@kth.se +</td></tr><tr><td>54a9ed950458f4b7e348fa78a718657c8d3d0e05</td><td>Learning Neural Models for End-to-End +<br/>Clustering +<br/>1 ZHAW Datalab & School of Engineering, Winterthur, Switzerland +<br/>2 ARGUS DATA INSIGHTS Schweiz AG, Zurich, Switzerland +<br/><b>Ca Foscari University of Venice, Venice, Italy</b><br/><b>Institute of Neural Information Processing, Ulm University, Germany</b><br/><b>Institute for Optical Systems, HTWG Konstanz, Germany</b></td><td>('50415299', 'Benjamin Bruno Meier', 'benjamin bruno meier')<br/>('3469013', 'Ismail Elezi', 'ismail elezi')<br/>('1985672', 'Mohammadreza Amirian', 'mohammadreza amirian')<br/>('3238279', 'Oliver Dürr', 'oliver dürr')<br/>('2793787', 'Thilo Stadelmann', 'thilo stadelmann')</td><td></td></tr><tr><td>541f1436c8ffef1118a0121088584ddbfd3a0a8a</td><td>A Spatio-Temporal Feature based on Triangulation of Dense SURF +<br/><b>The University of Electro-Communications, Tokyo</b><br/>1-5-1 Chofu, Tokyo 182-0021 JAPAN +</td><td>('2274625', 'Do Hang Nga', 'do hang nga')<br/>('1681659', 'Keiji Yanai', 'keiji yanai')</td><td>dohang@mm.cs.uec.ac.jp, yanai@cs.uec.ac.jp +</td></tr><tr><td>54aacc196ffe49b3450059fccdf7cd3bb6f6f3c3</td><td>A Joint Learning Framework for Attribute Models and Object Descriptions +<br/>Dhruv Mahajan +<br/>Yahoo! Labs, Bangalore, India +</td><td>('1779926', 'Sundararajan Sellamanickam', 'sundararajan sellamanickam')<br/>('4989209', 'Vinod Nair', 'vinod nair')</td><td>{dkm,ssrajan,vnair}@yahoo-inc.com +</td></tr><tr><td>54ce3ff2ab6e4465c2f94eb4d636183fa7878ab7</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Local Centroids Structured Non-Negative Matrix Factorization +<br/><b>University of Texas at Arlington, Texas, USA</b><br/><b>School of Computer Science, OPTIMAL, Northwestern Polytechnical University, Xian 710072, Shaanxi, P. R. China</b></td><td>('2141896', 'Hongchang Gao', 'hongchang gao')<br/>('1688370', 'Feiping Nie', 'feiping nie')</td><td>{hongchanggao, feipingnie}@gmail.com, heng@uta.edu +</td></tr><tr><td>541bccf19086755f8b5f57fd15177dc49e77d675</td><td></td><td>('2154872', 'Lijin Aryananda', 'lijin aryananda')</td><td></td></tr><tr><td>5495e224ac7b45b9edc5cfeabbb754d8a40a879b</td><td>Feature Reconstruction Disentangling for Pose-invariant Face Recognition +<br/>Supplementary Material +<br/><b>Rutgers, The State University of New Jersey</b><br/><b>University of California, San Diego</b><br/>‡ NEC Laboratories America +<br/>1. Summary of The Supplementary +<br/>This supplementary file includes two parts: (a) Addi- +<br/>tional implementation details are presented to improve the +<br/>reproducibility; (b) More experimental results are presented +<br/>to validate our approach in different aspects, which are not +<br/>shown in the main submission due to the space limitation. +<br/>2. Additional Implementation Details +<br/>Pose-variant face generation We designed a network to +<br/>predict 3DMM parameters from a single face image. The +<br/>design is mainly based on VGG16 [4]. We use the same num- +<br/>ber of convolutional layers as VGG16 but replacing all max +<br/>pooling layers with stride-2 convolutional operations. The +<br/>fully connected (fc) layers are also different: we first use two +<br/>fc layers, each of which has 1024 neurons, to connect with +<br/>the convolutional modules; then, a fc layer of 30 neurons is +<br/>used for identity parameters, a fc layer of 29 neurons is used +<br/>for expression parameters, and a fc layer of 7 neurons is used +<br/>for pose parameters. Different from [8] uses 199 parameters +<br/>to represent the identity coefficients, we truncate the num- +<br/>ber of identity eigenvectors to 30 which preserves 90% of +<br/>variations. This truncation leads to fast convergence and less +<br/>overfitting. For texture, we only generate non-frontal faces +<br/>from frontal ones, which significantly mitigate the halluci- +<br/>nating texture issue caused by self occlusion and guarantee +<br/>high-fidelity reconstruction. We apply the Z-Buffer algo- +<br/>rithm used in [8] to prevent ambiguous pixel intensities due +<br/>to same image plane position but different depths. +<br/>Rich feature embedding The design of the rich em- +<br/>bedding network is mainly based on the architecture of +<br/>CASIA-net [6] since it is wildly used in former approach +<br/>and achieves strong performance in face recognition. During +<br/>training, CASIA+MultiPIE or CASIA+300WLP are used. +<br/>As shown in Figure 3 of the main submission, after the con- +<br/>volutional layers of CASIA-net, we use a 512-d FC for the +<br/>rich feature embedding, which is further branched into a +<br/>256-d identity feature and a 128-d non-identity feature. The +<br/>128-d non-identity feature is further connected with a 136-d +<br/>landmark prediction and a 7-d pose prediction. Notice that +<br/>in the face generation network, the number of pose parame- +<br/>ters is 7 instead of 3 because we need to uniquely depict the +<br/>projection matrix from the 3D model and the 2D face shape +<br/>in image domain, which includes scale, pitch, yaw, roll, x +<br/>translation, y translation, and z translations. +<br/>Disentanglement by feature reconstruction Once the +<br/>rich embedding network is trained, we feed genius pair that +<br/>share the same identity but different viewpoints into the +<br/>network to obtain the corresponding rich embedding, identity +<br/>and non-identity features. To disentangle the identity and +<br/>pose factors, we concatenate the identity and non-identity +<br/>features and roll though two 512-d fully connected layers +<br/>to output a reconstructed rich embedding depicted by 512 +<br/>neurons. Both self and cross reconstruction loss are designed +<br/>to eventually push the two identity features close to each +<br/>other. At the same time, a cross-entropy loss is applied on the +<br/>near-frontal identity feature to maintain the discriminative +<br/>power of the learned representation. The disentanglement +<br/>of the identity and pose is finally achieved by the proposed +<br/>feature reconstruction based metric learning. +<br/>3. Additional Experimental Results +<br/>In addition to the main submission, we present more +<br/>experimental results in this section to further validate our +<br/>approach in different aspects. +<br/>3.1. P1 and P2 protocol on MultiPIE +<br/>In the main submission, due to space considerations, we +<br/>only report the mean accuracy over 10 random training and +<br/>testing splits, on MultiPIE and 300WLP separately. In Ta- +<br/>ble 1, we report the standard deviation of our method as a +<br/>more complete comparison. From the results, the standard +<br/>deviation of our method is also very small, which suggests +<br/>that the performance is consistent across all the trials. We +</td><td>('4340744', 'Xi Peng', 'xi peng')<br/>('15644381', 'Xiang Yu', 'xiang yu')<br/>('1729571', 'Kihyuk Sohn', 'kihyuk sohn')<br/>('1711560', 'Dimitris N. Metaxas', 'dimitris n. metaxas')<br/>('2099305', 'Manmohan Chandraker', 'manmohan chandraker')</td><td>{xipeng.cs, dnm}@rutgers.edu, {xiangyu,ksohn,manu}@nec-labs.com +</td></tr><tr><td>54756f824befa3f0c2af404db0122f5b5bbf16e0</td><td>Research Statement +<br/>Computer Vision — Visual Recognition +<br/>Computational visual recognition concerns identifying what is in an image, video, or other visual data, enabling +<br/>applications such as measuring location, pose, size, activity, and identity as well as indexing for search by content. +<br/>Recent progress in making economical sensors and improvements in network, storage, and computational power +<br/>make visual recognition practical and relevant in almost all experimental sciences and commercial applications +<br/>such as image search. My work in visual recognition brings together machine learning, insights from psychology +<br/>and physiology, computer graphics, algorithms, and a great deal of computation. +<br/>While I am best known for my work on general object category detection – creating techniques and building +<br/>systems for some of the best performing approaches to categorizing and localizing objects in images, recognizing +<br/>action in video, and searching large collections of video and images – my research extends widely across visual +<br/>recognition including: +<br/>• Creating low-level image descriptors – procedures for converting pixel values to features that can be used +<br/>to model appearance for recognition. These include widely used descriptors for category recognition in +<br/>images [4, 2], object detection in images and video [11, 10, 2], and optical flow based descriptors for action +<br/>recognition in video [8]. +<br/>• Developing models for recognition – ranging from what is becoming seminal work in recognizing human +<br/>actions in video [8], to formulating object localization as approximate subgraph isomorphism [2], to models +<br/>for parsing architectural images [3], to a novel approach for face recognition based on high level describable +<br/>visual attributes [9]. +<br/>• Deriving machine learning techniques – this includes both techniques for increasing the accuracy of clas- +<br/>sification [15] and techniques that provide improvements in the trade-off between accuracy and efficiency +<br/>of classification for detection and categorization [11, 10] – making some approaches exponentially faster +<br/>and therefore useful for a new range of applications. +<br/>• Applications to web scale visual data – introducing novel techniques to automatically extract useful in- +<br/>formation from web-scale data. Successful applications include extracting models of face appearance [7] +<br/>and representative iconic images [5]. Some of my work on machine learning techniques for visual recogni- +<br/><b>tion [11, 10] is making possible very large scale visual recognition both in my own ongoing work, including</b><br/>a collaboration with the ImageNet (10 million images in 10 thousand categories) team at Princeton and +<br/>Stanford, and efforts by other researchers in industry (Google and Yahoo!) and academia. +<br/>• Applications to analyzing imagery of people – probably the most important type of content in images and +<br/>video. Several of my projects address analyzing imagery of people, from detection [10], to identification by +<br/>face recognition [9, 7, 6], to localizing limbs (pose estimation) [14], and recognizing actions [8]. +<br/>All of this work is part of an attempt to understand the structure of visual data and build better systems +<br/>for extracting information from visual signals. Such systems are useful in practice because, although for many +<br/>application areas human perceptual abilities far outstrip the ability of computational systems, automated systems +<br/>already have the upper hand in running constantly over vast amounts of data, e.g. surveillance systems and process +<br/>monitoring, and in making metric decisions about specific quantities such as size, distance, or orientation, where +<br/>humans have difficulty. Surveillance illustrates the need for recognition in order to increase performance. From +<br/>watching cells under a microscope to observing research mice in habitats to guarding national borders, surveillance +<br/>systems are limited by false detections produced due to spurious and unimportant activity. This cost can be reduced +<br/>by visual recognition algorithms that identify either activities of interest or the commonly occurring unimportant +<br/>activity. +<br/>Part of my work at Yahoo! Research emphasized another key application area for visual recognition, extracting +<br/>useful information from the vast and ever changing image and video data available on the world wide web. For +<br/>some of this data people provide partial annotation in the form of tags, captions, and freeform text on web pages. +<br/>One major challenge is to combine results from computational visual recognition with these partial annotations to +</td><td>('39668247', 'Alexander C. Berg', 'alexander c. berg')</td><td></td></tr><tr><td>54204e28af73c7aca073835a14afcc5d8f52a515</td><td>Fine-Pruning: Defending Against Backdooring Attacks +<br/>on Deep Neural Networks +<br/><b>New York University, Brooklyn, NY, USA</b></td><td>('48087922', 'Kang Liu', 'kang liu')<br/>('3337066', 'Brendan Dolan-Gavitt', 'brendan dolan-gavitt')<br/>('1696125', 'Siddharth Garg', 'siddharth garg')</td><td>{kang.liu,brendandg,siddharth.garg}@nyu.edu +</td></tr><tr><td>549c719c4429812dff4d02753d2db11dd490b2ae</td><td>YouTube-BoundingBoxes: A Large High-Precision +<br/>Human-Annotated Data Set for Object Detection in Video +<br/>Google Brain +<br/>Google Brain +<br/>Google Research +<br/>Google Brain +<br/>Google Brain +</td><td>('2892780', 'Esteban Real', 'esteban real')<br/>('1789737', 'Jonathon Shlens', 'jonathon shlens')<br/>('30554825', 'Stefano Mazzocchi', 'stefano mazzocchi')<br/>('3165011', 'Xin Pan', 'xin pan')<br/>('2657155', 'Vincent Vanhoucke', 'vincent vanhoucke')</td><td>ereal@google.com +<br/>shlens@google.com +<br/>stefanom@google.com +<br/>xpan@google.com +<br/>vanhoucke@google.com +</td></tr><tr><td>98b2f21db344b8b9f7747feaf86f92558595990c</td><td></td><td></td><td></td></tr><tr><td>98142103c311b67eeca12127aad9229d56b4a9ff</td><td>GazeDirector: Fully Articulated Eye Gaze Redirection in Video +<br/><b>University of Cambridge, UK 2Carnegie Mellon University, USA</b><br/><b>Max Planck Institute for Informatics, Germany</b><br/>4Microsoft +</td><td>('34399452', 'Erroll Wood', 'erroll wood')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td></td></tr><tr><td>9820920d4544173e97228cb4ab8b71ecf4548475</td><td>ORIGINAL RESEARCH +<br/>published: 11 September 2015 +<br/>doi: 10.3389/fpsyg.2015.01386 +<br/>Automated facial coding software +<br/>outperforms people in recognizing +<br/>neutral faces as neutral from +<br/>standardized datasets +<br/><b>The Amsterdam School of Communication Research, University of Amsterdam</b><br/>Amsterdam, Netherlands +<br/>Little is known about people’s accuracy of recognizing neutral faces as neutral. In this +<br/>paper, I demonstrate the importance of knowing how well people recognize neutral +<br/>faces. I contrasted human recognition scores of 100 typical, neutral front-up facial +<br/>images with scores of an arguably objective judge – automated facial coding (AFC) +<br/>software. I hypothesized that the software would outperform humans in recognizing +<br/>neutral faces because of the inherently objective nature of computer algorithms. Results +<br/>confirmed this hypothesis. I provided the first-ever evidence that computer software +<br/>(90%) was more accurate in recognizing neutral faces than people were (59%). I posited +<br/>two theoretical mechanisms, i.e., smile-as-a-baseline and false recognition of emotion, +<br/>as possible explanations for my findings. +<br/>Keywords: non-verbal communication, facial expression, face recognition, neutral face, automated facial coding +<br/>Introduction +<br/>lack of anger, +<br/>face should indicate lack of emotion, e.g., +<br/>Recognizing a neutral face as neutral is vital in social interactions. By virtue of “expressing” +<br/>“nothing” (for a separate discussion on faces “expressing” something, see Russell and Fernández- +<br/>Dols, 1997), a neutral +<br/>fear, or +<br/>disgust. This article’s inspiration was the interesting observation that in the literature on +<br/>facial recognition, little attention has been paid to neutral face recognition scores of human +<br/>raters. Russell (1994) and Nelson and Russell (2013), who provided the two most important +<br/>overviews on the topic, did not include or discuss recognition rates of lack of emotion +<br/>(neutral) in neutral faces. They provided overviews of matching scores (i.e., accuracy) for +<br/>six basic emotions, but they were silent on the issue of recognition accuracy of neutral +<br/>faces. +<br/>A distinct lack of articles that explicitly report accuracy scores for recognition of neutral face +<br/>could explain the silence of researchers in this field. One notable exception is the Amsterdam +<br/>Dynamic Facial Expression Set (ADFES; van der Schalk et al., 2011), where the authors provide +<br/>an average matching score of 0.67 for their neutral faces. This score is considerably low when one +<br/>considers that an average for six basic emotions is also in this range ( 0.67, see Nelson and Russell, +<br/>2013, Table A1 for datasets between pre-1994 and 2010). +<br/>Edited by: +<br/>Paola Ricciardelli, +<br/><b>University of Milano-Bicocca, Italy</b><br/>Reviewed by: +<br/>Luis J. Fuentes, +<br/>Universidad de Murcia, Spain +<br/>Francesca Gasparini, +<br/><b>University of Milano-Bicocca, Italy</b><br/>*Correspondence: +<br/>The Amsterdam School +<br/>of Communication Research, +<br/>Department of Communication +<br/><b>Science, University of Amsterdam</b><br/>Postbus 15793, +<br/>1001 NG Amsterdam, Netherlands +<br/>Specialty section: +<br/>This article was submitted to +<br/>Cognition, +<br/>a section of the journal +<br/>Frontiers in Psychology +<br/>Received: 22 April 2015 +<br/>Accepted: 31 August 2015 +<br/>Published: 11 September 2015 +<br/>Citation: +<br/>Lewinski P (2015) Automated facial +<br/>coding software outperforms people +<br/>in recognizing neutral faces as neutral +<br/>from standardized datasets. +<br/>Front. Psychol. 6:1386. +<br/>doi: 10.3389/fpsyg.2015.01386 +<br/>Frontiers in Psychology | www.frontiersin.org +<br/>September 2015 | Volume 6 | Article 1386 +</td><td>('6402753', 'Peter Lewinski', 'peter lewinski')<br/>('6402753', 'Peter Lewinski', 'peter lewinski')</td><td>p.lewinski@uva.nl +</td></tr><tr><td>9853136dbd7d5f6a9c57dc66060cab44a86cd662</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 34– No.2, November 2011 +<br/>Improving the Neural Network Training for Face +<br/>Recognition using Adaptive Learning Rate, Resilient +<br/>Back Propagation and Conjugate Gradient Algorithm +<br/>M.Sc. Student +<br/>Department of Electrical +<br/><b>Engineering, Iran University</b><br/>of Science and Technology, +<br/>Tehran, Iran +<br/>Saeid Sanei +<br/>Associate Professor +<br/>Department of Computing, +<br/>Faculty of Engineering and +<br/><b>Physical Sciences, University</b><br/>of Surrey, UK +<br/>Karim Mohammadi +<br/>Professor +<br/>Department of Electrical +<br/><b>Engineering, Iran University</b><br/>of Science and Technology, +<br/>Tehran, Iran +</td><td>('47250218', 'Hamed Azami', 'hamed azami')</td><td></td></tr><tr><td>989332c5f1b22604d6bb1f78e606cb6b1f694e1a</td><td>Recurrent Face Aging +<br/><b>University of Trento, Italy</b><br/><b>National University of Singapore</b><br/><b>Research Center for Learning Science, Southeast University, Nanjing, China</b><br/><b>Arti cial Intelligence Institute, China</b></td><td>('39792736', 'Wei Wang', 'wei wang')<br/>('10338111', 'Zhen Cui', 'zhen cui')<br/>('32059677', 'Yan Yan', 'yan yan')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('2287686', 'Xiangbo Shu', 'xiangbo shu')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')</td><td>{wei.wang,yan.yan,niculae.sebe}@unitn.it {elefjia,eleyans}@nus.edu.sg +<br/>zhen.cui@seu.edu.cn shuxb104@gmail.com +</td></tr><tr><td>982f5c625d6ad0dac25d7acbce4dabfb35dd7f23</td><td>Facial Expression Recognition by SVM-based Two-stage Classifier on +<br/>Gabor Features +<br/>School of Information Science +<br/><b>Japan Advanced Institute of Science and Technology</b><br/>Ashahi-dai 1-8, Nomi, Ishikawa 923-1292, Japan +</td><td>('1753878', 'Fan Chen', 'fan chen')<br/>('1791753', 'Kazunori Kotani', 'kazunori kotani')</td><td>chen-fan@jaist.ac.jp +<br/>ikko@jaist.ac.jp +</td></tr><tr><td>98af221afd64a23e82c40fd28d25210c352e41b7</td><td>ISCA Archive +<br/>http://www.isca-speech.org/archive +<br/>AVSP 2010 -- International Conference +<br/>on Audio-Visual Speech Processing +<br/>Hakone, Kanagawa, Japan +<br/>September 30--October 3, 2010 +<br/>Exploring Visual Features Through Gabor Representations for Facial +<br/>Expression Detection +<br/><b>Image and Video Research Laboratory, Queensland University of Technology</b><br/>GPO Box 2424, Brisbane 4001, Australia +<br/><b>Robotics Institute, Carnegie Mellon University</b><br/><b>University of Pittsburgh, Pittsburgh, USA</b></td><td>('2739248', 'Sien W. Chew', 'sien w. chew')<br/>('1713496', 'Patrick Lucey', 'patrick lucey')<br/>('1729760', 'Sridha Sridharan', 'sridha sridharan')<br/>('3140440', 'Clinton Fookes', 'clinton fookes')</td><td>s4.chew@student.qut.edu.au, patlucey@andrew.cmu.edu, {s.sridharan;c.fookes}@qut.edu.au +</td></tr><tr><td>9893865afdb1de55fdd21e5d86bbdb5daa5fa3d5</td><td>Illumination Normalization Using Logarithm Transforms +<br/>for Face Authentication +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave, Pittsburgh, USA +</td><td>('1794486', 'Marios Savvides', 'marios savvides')</td><td>msavvid@ri.cmu.edu +<br/>kumar@ece.cmu.edu +</td></tr><tr><td>988d1295ec32ce41d06e7cf928f14a3ee079a11e</td><td>Semantic Deep Learning +<br/>September 29, 2015 +</td><td>('36097730', 'Hao Wang', 'hao wang')</td><td></td></tr><tr><td>98c548a4be0d3b62971e75259d7514feab14f884</td><td>Deep generative-contrastive networks for facial expression recognition +<br/><b>Samsung Advanced Institute of Technology (SAIT), KAIST</b></td><td>('2310577', 'Youngsung Kim', 'youngsung kim')<br/>('1757573', 'ByungIn Yoo', 'byungin yoo')<br/>('9942811', 'Youngjun Kwak', 'youngjun kwak')<br/>('36995891', 'Changkyu Choi', 'changkyu choi')<br/>('1769295', 'Junmo Kim', 'junmo kim')</td><td>yo.s.ung.kim@samsung.com, byungin.yoo@kaist.ac.kr, yjk.kwak@samsung.com, changkyu choi@samsung.com, +<br/>junmo.kim@ee.kaist.ac.kr +</td></tr><tr><td>9887ab220254859ffc7354d5189083a87c9bca6e</td><td>Generic Image Classification Approaches Excel on Face Recognition +<br/><b>Nanjing University of Science and Technology, China</b><br/><b>The University of Adelaide, Australia</b></td><td>('2731972', 'Fumin Shen', 'fumin shen')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')</td><td></td></tr><tr><td>985cd420c00d2f53965faf63358e8c13d1951fa8</td><td>Pixel-Level Hand Detection with Shape-aware +<br/>Structured Forests +<br/>Department of Computer Science +<br/><b>The University of Hong Kong</b><br/>Pokfulam Road, Hong Kong +</td><td>('35130187', 'Xiaolong Zhu', 'xiaolong zhu')<br/>('34760532', 'Xuhui Jia', 'xuhui jia')</td><td>{xlzhu,xhjia,kykwong}@cs.hku.hk +</td></tr><tr><td>981449cdd5b820268c0876477419cba50d5d1316</td><td>Learning Deep Features for One-Class +<br/>Classification +</td><td>('15206897', 'Pramuditha Perera', 'pramuditha perera')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')</td><td></td></tr><tr><td>9821669a989a3df9d598c1b4332d17ae8e35e294</td><td>Minimal Correlation Classification +<br/><b>The Blavatnik School of Computer Science, Tel Aviv University, Israel</b></td><td>('21494706', 'Noga Levy', 'noga levy')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td></td></tr><tr><td>9854145f2f64d52aac23c0301f4bb6657e32e562</td><td>An Improved Face Verification Approach based on +<br/>Speedup Robust Features and Pairwise Matching +<br/>Center for Electrical Engineering and Informatics (CEEI) +<br/><b>Federal University of Campina Grande (UFCG</b><br/>Campina Grande, Para´ıba, Brazil +</td><td>('2092178', 'Herman Martins Gomes', 'herman martins gomes')</td><td>Email: {edumoura,hmg}@dsc.ufcg.edu.br, carvalho@dee.ufcg.edu.br +</td></tr><tr><td>9865fe20df8fe11717d92b5ea63469f59cf1635a</td><td>YUCEL ET AL.: WILDEST FACES +<br/>Wildest Faces: Face Detection and +<br/>Recognition in Violent Settings +<br/>Pinar Duygulu1 +<br/>1 Department of Computer Science +<br/><b>Hacettepe University</b><br/>Ankara, Turkey +<br/>2 Department of Computer Engineering +<br/><b>Middle East Technical University</b><br/>Ankara, Turkey +<br/>* indicates equal contribution. +</td><td>('46234524', 'Mehmet Kerim Yucel', 'mehmet kerim yucel')<br/>('39032755', 'Yunus Can Bilge', 'yunus can bilge')<br/>('46437368', 'Oguzhan Oguz', 'oguzhan oguz')<br/>('2011587', 'Nazli Ikizler-Cinbis', 'nazli ikizler-cinbis')<br/>('1939006', 'Ramazan Gokberk Cinbis', 'ramazan gokberk cinbis')</td><td>mkerimyucel@hacettepe.edu.tr +<br/>yunuscan.bilge@hacettepe.edu.tr +<br/>oguzhan.oguz@hacettepe.edu.tr +<br/>nazli@cs.hacettepe.edu.tr +<br/>pinar@cs.hacettepe.edu.tr +<br/>gcinbis@ceng.metu.edu.tr +</td></tr><tr><td>98c2053e0c31fab5bcb9ce5386335b647160cc09</td><td>A Distributed Framework for Spatio-temporal Analysis on Large-scale Camera +<br/>Networks +<br/><b>Georgia Institute of Technology</b><br/><b>University of Stuttgart</b><br/>†SUNY Buffalo +</td><td>('5540701', 'Kirak Hong', 'kirak hong')<br/>('1723877', 'Venu Govindaraju', 'venu govindaraju')<br/>('1752885', 'Bharat Jayaraman', 'bharat jayaraman')<br/>('1751741', 'Umakishore Ramachandran', 'umakishore ramachandran')</td><td>{khong9, rama}@cc.gatech.edu +<br/>marco.voelz@ipvs.uni-stuttgart.de +<br/>{govind, bharat}@buffalo.edu +</td></tr><tr><td>98127346920bdce9773aba6a2ffc8590b9558a4a</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Efficient Human Action Recognition using +<br/>Histograms of Motion Gradients and +<br/>VLAD with Descriptor Shape Information +<br/>Received: date / Accepted: date +</td><td>('3429470', 'Ionut C. Duta', 'ionut c. duta')<br/>('1796198', 'Bogdan Ionescu', 'bogdan ionescu')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td></td></tr><tr><td>98a660c15c821ea6d49a61c5061cd88e26c18c65</td><td>IOSR Journal of Engineering (IOSRJEN) +<br/>e-ISSN: 2250-3021, p-ISSN: 2278-8719 +<br/>Vol. 3, Issue 4 (April. 2013), ||V1 || PP 43-48 +<br/>Face Databases for 2D and 3D Facial Recognition: A Survey +<br/>R.Senthilkumar1, Dr.R.K.Gnanamurthy2 +<br/><b>Institute of Road and</b><br/><b>Odaiyappa College of</b><br/>Transport Technology,Erode-638 316. +<br/>Engineering and Technology,Theni-625 531. +</td><td></td><td></td></tr><tr><td>982fed5c11e76dfef766ad9ff081bfa25e62415a</td><td></td><td></td><td></td></tr><tr><td>98fb3890c565f1d32049a524ec425ceda1da5c24</td><td>A Robust Learning Framework Using PSM and +<br/>Ameliorated SVMs for Emotional Recognition +<br/><b>Graduate School of System Informatics, Kobe University, Kobe, 657-8501, Japan</b></td><td>('2866465', 'Jinhui Chen', 'jinhui chen')<br/>('21172382', 'Yosuke Kitano', 'yosuke kitano')<br/>('3207738', 'Yiting Li', 'yiting li')<br/>('1744026', 'Tetsuya Takiguchi', 'tetsuya takiguchi')<br/>('1678564', 'Yasuo Ariki', 'yasuo ariki')</td><td>{ianchen, kitano, liyiting }@me.cs.scitec.kobe-u.ac.jp +<br/>{takigu, ariki}@kobe-u.ac.jp +</td></tr><tr><td>98519f3f615e7900578bc064a8fb4e5f429f3689</td><td>Dictionary-based Domain Adaptation Methods +<br/>for the Re-identification of Faces +</td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('38811046', 'Jie Ni', 'jie ni')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>9825aa96f204c335ec23c2b872855ce0c98f9046</td><td>International Journal of Ethics in Engineering & Management Education +<br/>Website: www.ijeee.in (ISSN: 2348-4748, Volume 1, Issue 5, May2014) +<br/>FACE AND FACIAL EXPRESSION +<br/>RECOGNITION IN 3-D USING MASKED +<br/>PROJECTION UNDER OCCLUSION +<br/>Jyoti patil * +<br/>M.Tech (CSE) +<br/>GNDEC Bidar-585401 +<br/>BIDAR, INDIA +<br/> M.Tech (CSE) +<br/> GNDEC Bidar- 585401 +<br/> BIDAR, INDIA +<br/> M.Tech (CSE) +<br/> VKIT, Bangalore- 560040 +<br/>BANGALORE, INDIA +</td><td>('39365176', 'Gouri Patil', 'gouri patil')<br/>('4787347', 'Snehalata Patil', 'snehalata patil')</td><td>Email-jyoti.spatil35@gmail.com Email-greatgouri@gmail.com +<br/> Email-snehasharad09@gmail.com +</td></tr><tr><td>9825c4dddeb2ed7eaab668b55403aa2c38bc3320</td><td>Aerial Imagery for Roof Segmentation: A Large-Scale Dataset +<br/>towards Automatic Mapping of Buildings +<br/><b>aCenter for Spatial Information Science, University of Tokyo, Kashiwa 277-8568, Japan</b><br/><b>University of Waterloo, Waterloo, ON N2L 3G1, Canada</b><br/><b>cFaculty of Information Engineering, China University of Geosciences (Wuhan), Wuhan 430074, China</b><br/>dAtlasAI Inc., Waterloo, ON N2L 3G1, Canada +</td><td>('1783637', 'Qi Chen', 'qi chen')<br/>('48169641', 'Lei Wang', 'lei wang')<br/>('50117915', 'Yifan Wu', 'yifan wu')<br/>('3043983', 'Guangming Wu', 'guangming wu')<br/>('40477085', 'Zhiling Guo', 'zhiling guo')</td><td></td></tr><tr><td>980266ad6807531fea94252e8f2b771c20e173b3</td><td>Continuous Regression for +<br/>Non-Rigid Image Alignment +<br/>Enrique S´anchez-Lozano1 +<br/>Daniel Gonz´alez-Jim´enez1 +<br/>1Multimodal Information Area, Gradiant, Vigo, Pontevedra, 36310. Spain. +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, 15213. USA</b></td><td>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td>{esanchez,dgonzalez}@gradiant.org +<br/>ftorre@cs.cmu.edu +</td></tr><tr><td>53d78c8dbac7c9be8eb148c6a9e1d672f1dd72f9</td><td>Discriminative vs. Generative Object Recognition: +<br/>Objects, Faces, and the Web +<br/>Thesis by +<br/>In Partial Fulfillment of the Requirements +<br/>for the Degree of +<br/>Doctor of Philosophy +<br/><b>California Institute of Technology</b><br/>Pasadena, California +<br/>2007 +<br/>(Defended April 30, 2007) +</td><td>('3075121', 'Alex Holub', 'alex holub')</td><td></td></tr><tr><td>533d14e539ae5cdca0ece392487a2b19106d468a</td><td>Bidirectional Multirate Reconstruction for Temporal Modeling in Videos +<br/><b>University of Technology Sydney</b></td><td>('2948393', 'Linchao Zhu', 'linchao zhu')<br/>('2351434', 'Zhongwen Xu', 'zhongwen xu')<br/>('1698559', 'Yi Yang', 'yi yang')</td><td>{zhulinchao7, zhongwen.s.xu, yee.i.yang}@gmail.com +</td></tr><tr><td>5334ac0a6438483890d5eef64f6db93f44aacdf4</td><td></td><td></td><td></td></tr><tr><td>53dd25350d3b3aaf19beb2104f1e389e3442df61</td><td></td><td></td><td></td></tr><tr><td>53e081f5af505374c3b8491e9c4470fe77fe7934</td><td>Unconstrained Realtime Facial Performance Capture +<br/><b>University of Southern California</b><br/>† Industrial Light & Magic +<br/>Figure 1: Calibration-free realtime facial performance capture on highly occluded subjects using an RGB-D sensor. +</td><td>('2519072', 'Pei-Lun Hsieh', 'pei-lun hsieh')<br/>('1797422', 'Chongyang Ma', 'chongyang ma')<br/>('2977637', 'Jihun Yu', 'jihun yu')<br/>('1706574', 'Hao Li', 'hao li')</td><td></td></tr><tr><td>53698b91709112e5bb71eeeae94607db2aefc57c</td><td>Two-Stream Convolutional Networks +<br/>for Action Recognition in Videos +<br/><b>Visual Geometry Group, University of Oxford</b></td><td>('34838386', 'Karen Simonyan', 'karen simonyan')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>{karen,az}@robots.ox.ac.uk +</td></tr><tr><td>5394d42fd27b7e14bd875ec71f31fdd2fcc8f923</td><td>Visual Recognition Using Directional Distribution Distance +<br/>National Key Laboratory for Novel Software Technology +<br/><b>Nanjing University, China</b><br/>Minieye, Youjia Innovation LLC +</td><td>('1808816', 'Jianxin Wu', 'jianxin wu')<br/>('2226422', 'Bin-Bin Gao', 'bin-bin gao')<br/>('15527784', 'Guoqing Liu', 'guoqing liu')</td><td>guoqing@minieye.cc +<br/>wujx2001@nju.edu.cn, gaobb@lamda.nju.edu.cn +</td></tr><tr><td>530243b61fa5aea19b454b7dbcac9f463ed0460e</td><td></td><td></td><td></td></tr><tr><td>5397c34a5e396658fa57e3ca0065a2878c3cced7</td><td>Lighting Normalization with Generic Intrinsic Illumination Subspace for Face +<br/>Recognition +<br/><b>Institute of Information Science, Academia Sinica, Taipei, Taiwan</b></td><td>('1686057', 'Chia-Ping Chen', 'chia-ping chen')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')</td><td>{cpchen, song}@iis.sinica.edu.tw +</td></tr><tr><td>539ca9db570b5e43be0576bb250e1ba7a727d640</td><td></td><td></td><td></td></tr><tr><td>539287d8967cdeb3ef60d60157ee93e8724efcac</td><td>Learning Deep (cid:96)0 Encoders +<br/><b>Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA</b><br/><b>University of Science and Technology of China, Hefei, 230027, China</b></td><td>('2969311', 'Zhangyang Wang', 'zhangyang wang')<br/>('1682497', 'Qing Ling', 'qing ling')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td></td></tr><tr><td>532f7ec8e0c8f7331417dd4a45dc2e8930874066</td><td>6060 +<br/>2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +<br/>978-1-4799-2893-4/14/$31.00 ©2014 IEEE +<br/>Box 451, Thessaloniki 54124, GREECE +<br/><b>Aristotle University of Thessaloniki</b><br/>Department of Informatics +<br/>tel: +30 2310 996361 +<br/>1. INTRODUCTION +</td><td>('1905139', 'Olga Zoidi', 'olga zoidi')<br/>('1718330', 'Nikos Nikolaidis', 'nikos nikolaidis')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>{ozoidi, nikolaid, pitas}@aiia.csd.auth.gr +</td></tr><tr><td>53c8cbc4a3a3752a74f79b74370ed8aeed97db85</td><td></td><td></td><td></td></tr><tr><td>53c36186bf0ffbe2f39165a1824c965c6394fe0d</td><td>I Know How You Feel: Emotion Recognition with Facial Landmarks +<br/><b>Tooploox 2Polish-Japanese Academy of Information Technology 3Warsaw University of Technology</b></td><td>('22188614', 'Ivona Tautkute', 'ivona tautkute')<br/>('1760267', 'Tomasz Trzcinski', 'tomasz trzcinski')<br/>('48657002', 'Adam Bielski', 'adam bielski')</td><td>{firstname.lastname}@tooploox.com +</td></tr><tr><td>5366573e96a1dadfcd4fd592f83017e378a0e185</td><td>Böhlen, Chandola and Salunkhe +<br/>Server, server in the cloud. +<br/>Who is the fairest in the crowd? +</td><td></td><td></td></tr><tr><td>53a41c711b40e7fe3dc2b12e0790933d9c99a6e0</td><td>Recurrent Memory Addressing for describing videos +<br/><b>Indian Institute of Technology Kharagpur</b></td><td>('7284555', 'Arnav Kumar Jain', 'arnav kumar jain')<br/>('6565766', 'Kumar Krishna Agrawal', 'kumar krishna agrawal')<br/>('1781070', 'Pabitra Mitra', 'pabitra mitra')</td><td>{arnavkj95, abhinavagarawalla, kumarkrishna, pabitra}@iitkgp.ac.in +</td></tr><tr><td>53bfe2ab770e74d064303f3bd2867e5bf7b86379</td><td>Learning to Synthesize and Manipulate Natural Images +<br/>By +<br/>A dissertation submitted in partial satisfaction of the +<br/>requirements for the degree of +<br/>Doctor of Philosophy +<br/>in +<br/>Engineering - Electrical Engineering and Computer Science +<br/>in the +<br/>Graduate Division +<br/>of the +<br/><b>University of California, Berkeley</b><br/>Committee in charge: +<br/>Professor Alexei A. Efros, Chair +<br/>Professor Jitendra Malik +<br/>Professor Ren Ng +<br/>Professor Michael DeWeese +<br/>Fall 2017 +</td><td>('3132726', 'Junyan Zhu', 'junyan zhu')</td><td></td></tr><tr><td>533bfb82c54f261e6a2b7ed7d31a2fd679c56d18</td><td>Technical Report MSU-CSE-14-1 +<br/>Unconstrained Face Recognition: Identifying a +<br/>Person of Interest from a Media Collection +</td><td>('2180413', 'Lacey Best-Rowden', 'lacey best-rowden')<br/>('34393045', 'Hu Han', 'hu han')<br/>('40653304', 'Charles Otto', 'charles otto')<br/>('1817623', 'Brendan Klare', 'brendan klare')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>537d8c4c53604fd419918ec90d6ef28d045311d0</td><td>Active Collaborative Ensemble Tracking +<br/><b>Graduate School of Informatics, Kyoto University</b><br/>Yoshida-Honmachi, Sakyo Ward, Kyoto 606–8501, Japan +</td><td>('2146623', 'Kourosh Meshgi', 'kourosh meshgi')<br/>('31095396', 'Maryam Sadat Mirzaei', 'maryam sadat mirzaei')<br/>('38809507', 'Shigeyuki Oba', 'shigeyuki oba')<br/>('2851612', 'Shin Ishii', 'shin ishii')</td><td>meshgi-k@sys.i.kyoto-u.ac.jp +</td></tr><tr><td>530ce1097d0681a0f9d3ce877c5ba31617b1d709</td><td></td><td></td><td></td></tr><tr><td>53ce84598052308b86ba79d873082853022aa7e9</td><td>Optimized Method for Real-Time Face Recognition System Based +<br/>on PCA and Multiclass Support Vector Machine +<br/><b>IEEE Member, Shahid Rajaee Teacher training University</b><br/>Tehran, Iran +<br/><b>Institute of Computer science, Shahid Bahonar University</b><br/>Shiraz, Iran +<br/><b>Islamic Azad University, Science and Research Campus</b><br/>Hamedan, Iran +</td><td>('1763181', 'Reza Azad', 'reza azad')<br/>('39864738', 'Babak Azad', 'babak azad')<br/>('2904132', 'Iman Tavakoli Kazerooni', 'iman tavakoli kazerooni')</td><td>rezazad68@gmail.com +<br/>babak.babi72@gmail.com +<br/>iman_tavakoli2008@yahoo.com +</td></tr><tr><td>3fbd68d1268922ee50c92b28bd23ca6669ff87e5</td><td>598 +<br/>IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 10, NO. 4, APRIL 2001 +<br/>A Shape- and Texture-Based Enhanced Fisher +<br/>Classifier for Face Recognition +</td><td>('39664966', 'Chengjun Liu', 'chengjun liu')<br/>('1781577', 'Harry Wechsler', 'harry wechsler')</td><td></td></tr><tr><td>3fe4109ded039ac9d58eb9f5baa5327af30ad8b6</td><td>Spatio-Temporal GrabCut Human Segmentation for Face and Pose Recovery +<br/>Antonio Hern´andez1 +<br/><b>University of Barcelona, Gran Via de les Corts Catalanes 585, 08007 Barcelona, Spain</b><br/>1 Computer Vision Center, Campus UAB, 08193 Bellaterra, Barcelona, Spain. +</td><td>('3276130', 'Miguel Reyes', 'miguel reyes')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('1724155', 'Petia Radeva', 'petia radeva')</td><td>ahernandez@cvc.uab.es +<br/>mreyese@gmail.com +<br/>sergio@maia.ub.es +<br/>petia@cvc.uab.es +</td></tr><tr><td>3f22a4383c55ceaafe7d3cfed1b9ef910559d639</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>Robust Kronecker Component Analysis +</td><td>('11352680', 'Mehdi Bahri', 'mehdi bahri')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td></td></tr><tr><td>3fefc856a47726d19a9f1441168480cee6e9f5bb</td><td><b>Carnegie Mellon University</b><br/>Dissertations +<br/>Summer 8-2014 +<br/>Theses and Dissertations +<br/>Perceptually Valid Dynamics for Smiles and Blinks +<br/><b>Carnegie Mellon University</b><br/>Follow this and additional works at: http://repository.cmu.edu/dissertations +<br/>Recommended Citation +<br/>Trutoiu, Laura, "Perceptually Valid Dynamics for Smiles and Blinks" (2014). Dissertations. Paper 428. +</td><td>('2048839', 'Laura Trutoiu', 'laura trutoiu')</td><td>Research Showcase @ CMU +<br/>This Dissertation is brought to you for free and open access by the Theses and Dissertations at Research Showcase @ CMU. It has been accepted for +<br/>inclusion in Dissertations by an authorized administrator of Research Showcase @ CMU. For more information, please contact research- +<br/>showcase@andrew.cmu.edu. +</td></tr><tr><td>3fdcc1e2ebcf236e8bb4a6ce7baf2db817f30001</td><td>A top-down approach for a synthetic +<br/>autobiographical memory system +<br/>1Sheffield Centre for Robotics (SCentRo), Univ. of Sheffield, Sheffield, S10 2TN, UK +<br/>2Dept. of Computer Science, Univ. of Sheffield, Sheffield, S1 4DP, UK +<br/>3 CVAP Lab, KTH, Stockholm, Sweden +</td><td>('2484138', 'Carl Henrik Ek', 'carl henrik ek')<br/>('1739851', 'Neil D. Lawrence', 'neil d. lawrence')<br/>('1750570', 'Tony J. Prescott', 'tony j. prescott')</td><td>andreas.damianou@shef.ac.uk +</td></tr><tr><td>3f7cf52fb5bf7b622dce17bb9dfe747ce4a65b96</td><td>Person identity label propagation in stereo videos +<br/>Department of Informatics +<br/><b>Aristotle University of Thessaloniki</b><br/>Box 451, Thessaloniki 54124, GREECE +<br/>tel: +30 2310 996361 +</td><td>('1905139', 'Olga Zoidi', 'olga zoidi')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')<br/>('1718330', 'Nikos Nikolaidis', 'nikos nikolaidis')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>{tefas, nikolaid, pitas}@aiia.csd.auth.gr +</td></tr><tr><td>3f0c51989c516a7c5dee7dec4d7fb474ae6c28d9</td><td>Not Afraid of the Dark: NIR-VIS Face Recognition via Cross-spectral +<br/>Hallucination and Low-rank Embedding +<br/><b>IIE, Universidad de la Rep ublica, Uruguay. 2ECE, Duke University, USA</b></td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('1699339', 'Guillermo Sapiro', 'guillermo sapiro')</td><td></td></tr><tr><td>3f848d6424f3d666a1b6dd405a48a35a797dd147</td><td>GHODRATI et al.: IS 2D INFORMATION ENOUGH FOR VIEWPOINT ESTIMATION? +<br/>Is 2D Information Enough For Viewpoint +<br/>Estimation? +<br/>KU Leuven, ESAT - PSI, iMinds +<br/>Leuven, Belgium +</td><td>('3060081', 'Amir Ghodrati', 'amir ghodrati')<br/>('3048367', 'Marco Pedersoli', 'marco pedersoli')<br/>('1704728', 'Tinne Tuytelaars', 'tinne tuytelaars')</td><td>amir.ghodrati@esat.kuleuven.be +<br/>marco.pedersoli@esat.kuleuven.be +<br/>tinne.tuytelaars@esat.kuleuven.be +</td></tr><tr><td>3fa738ab3c79eacdbfafa4c9950ef74f115a3d84</td><td>DaMN – Discriminative and Mutually Nearest: +<br/>Exploiting Pairwise Category Proximity +<br/>for Video Action Recognition +<br/>1 Center for Research in Computer Vision at UCF, Orlando, USA +<br/>2 Google Research, Mountain View, USA +<br/>http://crcv.ucf.edu/projects/DaMN/ +</td><td>('2099254', 'Rui Hou', 'rui hou')<br/>('40029556', 'Amir Roshan Zamir', 'amir roshan zamir')<br/>('1694199', 'Rahul Sukthankar', 'rahul sukthankar')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td></td></tr><tr><td>3fb26f3abcf0d287243646426cd5ddeee33624d4</td><td>Joint Training of Cascaded CNN for Face Detection +<br/><b>Grad. School at Shenzhen, Tsinghua University</b><br/><b>Tsinghua University 4SenseTime</b></td><td>('2137185', 'Hongwei Qin', 'hongwei qin')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('2693308', 'Xiu Li', 'xiu li')<br/>('1705418', 'Xiaolin Hu', 'xiaolin hu')</td><td>{qhw12@mails., li.xiu@sz., xlhu@}tsinghua.edu.cn yanjunjie@outlook.com +</td></tr><tr><td>3f9ca2526013e358cd8caeb66a3d7161f5507cbc</td><td>Improving Sparse Representation-Based Classification +<br/>Using Local Principal Component Analysis +<br/>Department of Mathematics +<br/><b>University of California, Davis</b><br/>One Shields Avenue +<br/>Davis, California, 95616, United States +</td><td>('32898818', 'Chelsea Weaver', 'chelsea weaver')<br/>('3493752', 'Naoki Saito', 'naoki saito')</td><td></td></tr><tr><td>3f57c3fc2d9d4a230ccb57eed1d4f0b56062d4d5</td><td>Face Recognition Across Poses Using A Single 3D Reference Model +<br/><b>National Taiwan University of Science and Technology</b><br/>No. 43, Sec.4, Keelung Rd., Taipei, 106, Taiwan +</td><td>('38801529', 'Gee-Sern Hsu', 'gee-sern hsu')<br/>('3329222', 'Hsiao-Chia Peng', 'hsiao-chia peng')</td><td>∗jison@mail.ntust.edu.tw +</td></tr><tr><td>3feb69531653e83d0986a0643e4a6210a088e3e5</td><td>Using Group Prior to Identify People in Consumer Images +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, Pennsylvania +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, Pennsylvania +</td><td>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td>agallagh@cmu.edu +<br/>tsuhan@cmu.edu +</td></tr><tr><td>3f12701449a82a5e01845001afab3580b92da858</td><td>Joint Object Class Sequencing and Trajectory +<br/>Triangulation (JOST) +<br/><b>The University of North Carolina, Chapel Hill</b></td><td>('2873326', 'Enliang Zheng', 'enliang zheng')<br/>('1751643', 'Ke Wang', 'ke wang')<br/>('29274093', 'Enrique Dunn', 'enrique dunn')<br/>('40454588', 'Jan-Michael Frahm', 'jan-michael frahm')</td><td></td></tr><tr><td>3fb98e76ffd8ba79e1c22eda4d640da0c037e98a</td><td>Convolutional Neural Networks for Crop Yield Prediction using Satellite Images +<br/>H. Russello +</td><td></td><td></td></tr><tr><td>3fde656343d3fd4223e08e0bc835552bff4bda40</td><td>Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/>A Monthly Journal of Computer Science and Information Technology +<br/>ISSN 2320–088X +<br/> IJCSMC, Vol. 2, Issue. 4, April 2013, pg.232 – 237 +<br/>RESEARCH ARTICLE +<br/>Character Identification Using Graph +<br/>Matching Algorithm +<br/><b>Anna University Chennai, India</b><br/>5Assistant Professor, Department Of Computer Science and Engineering, +<br/><b>K.S.R. College Of Engineering, Tiruchengode, India</b></td><td>('1795761', 'S. Bharathi', 's. bharathi')<br/>('36510121', 'Ranjith Kumar', 'ranjith kumar')</td><td> 1 rathiranya@gmail.com ; 2 manirathnam60@gmail.com ; 3 ramya1736@yahoo.com ; 4 ranjith.rhl@gmail.com +</td></tr><tr><td>3f957142ef66f2921e7c8c7eadc8e548dccc1327</td><td>Merging SVMs with Linear Discriminant Analysis: A Combined Model +<br/><b>Imperial College London, United Kingdom</b><br/><b>EEMCS, University of Twente, Netherlands</b></td><td>('1793625', 'Symeon Nikitidis', 'symeon nikitidis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{s.nikitidis,s.zafeiriou,m.pantic}@imperial.ac.uk +</td></tr><tr><td>3fdfd6fa7a1cc9142de1f53e4ac7c2a7ac64c2e3</td><td>Intensity-Depth Face Alignment Using Cascade +<br/>Shape Regression +<br/>1 Center for Brain-like Computing and Machine Intelligence +<br/>Department of Computer Science and Engineering +<br/><b>Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China</b><br/>2 Key Laboratory of Shanghai Education Commission for +<br/>Intelligent Interaction and Cognitive Engineering +<br/><b>Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai 200240, China</b></td><td>('1740511', 'Yang Cao', 'yang cao')<br/>('1715839', 'Bao-Liang Lu', 'bao-liang lu')</td><td></td></tr><tr><td>3f540faf85e1f8de6ce04fb37e556700b67e4ad3</td><td>Article +<br/>Face Verification with Multi-Task and Multi-Scale +<br/>Feature Fusion +<br/><b>College of Sciences, Northeastern University, Shenyang 110819, China</b><br/><b>New York University Shanghai, 1555 Century Ave, Pudong</b><br/>Academic Editor: Maxim Raginsky +<br/>Received: 18 March 2017; Accepted: 13 May 2017; Published: 17 May 2017 +</td><td>('26337951', 'Xiaojun Lu', 'xiaojun lu')<br/>('1983143', 'Yue Yang', 'yue yang')<br/>('8030754', 'Weilin Zhang', 'weilin zhang')<br/>('40435166', 'Qi Wang', 'qi wang')<br/>('2295608', 'Yang Wang', 'yang wang')</td><td>luxiaojun@mail.neu.edu.cn (X.L.); YangY1503@163.com (Y.Y.); wangy_neu@163.com (Y.W.) +<br/>Shanghai 200122, China; wz723@nyu.edu +<br/>* Correspondence: wangqi@mail.neu.edu.cn; Tel.: +86-24-8368-7680 +</td></tr><tr><td>3f4bfa4e3655ef392eb5ad609d31c05f29826b45</td><td>ROBUST MULTI-CAMERA VIEW FACE RECOGNITION +<br/>Department of Computer Science and Engineering +<br/><b>Dr. B. C. Roy Engineering College</b><br/>Durgapur - 713206 +<br/>India +<br/>Department of Computer Science and Engineering +<br/><b>National Institute of Technology Rourkela</b><br/>Rourkela – 769008 +<br/>India +<br/>Department of Computer Science and Engineering +<br/><b>Indian Institute of Technology Kanpur</b><br/>Kanpur – 208016 +<br/>India +<br/>Department of Computer Science and Engineering +<br/><b>Jadavpur University</b><br/>Kolkata – 700032, +<br/>India +<br/>face +<br/>recognition +<br/>to face +<br/>filter banks +<br/>system uses Gabor +<br/>images produces Gabor +</td><td>('1810015', 'Dakshina Ranjan Kisku', 'dakshina ranjan kisku')<br/>('1868921', 'Hunny Mehrotra', 'hunny mehrotra')<br/>('1687389', 'Phalguni Gupta', 'phalguni gupta')<br/>('1786127', 'Jamuna Kanta Sing', 'jamuna kanta sing')</td><td>drkisku@ieee.org; hunny04@gmail.com; pg@cse.iitk.ac.in; , jksing@ieee.org +</td></tr><tr><td>3f5cf3771446da44d48f1d5ca2121c52975bb3d3</td><td></td><td></td><td></td></tr><tr><td>3fb4bf38d34f7f7e5b3df36de2413d34da3e174a</td><td>THOMAS AND KOVASHKA: PERSUASIVE FACES: GENERATING FACES IN ADS +<br/>Persuasive Faces: Generating Faces in +<br/>Advertisements +<br/>Department of Computer Science +<br/><b>University of Pittsburgh</b><br/>Pittsburgh, PA USA +</td><td>('40540691', 'Christopher Thomas', 'christopher thomas')<br/>('1770205', 'Adriana Kovashka', 'adriana kovashka')</td><td>chris@cs.pitt.edu +<br/>kovashka@cs.pitt.edu +</td></tr><tr><td>3f14b504c2b37a0e8119fbda0eff52efb2eb2461</td><td>5727 +<br/>Joint Facial Action Unit Detection and Feature +<br/>Fusion: A Multi-Conditional Learning Approach +</td><td>('2308430', 'Stefanos Eleftheriadis', 'stefanos eleftheriadis')<br/>('1729713', 'Ognjen Rudovic', 'ognjen rudovic')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>3fac7c60136a67b320fc1c132fde45205cd2ac66</td><td>Remarks on Computational Facial Expression +<br/>Recognition from HOG Features Using +<br/>Quaternion Multi-layer Neural Network +<br/><b>Information Systems Design, Doshisha University, Kyoto, Japan</b><br/><b>Graduate School of Doshisha University, Kyoto, Japan</b><br/><b>Intelligent Information Engineering and Science, Doshisha University, Kyoto, Japan</b></td><td>('39452921', 'Kazuhiko Takahashi', 'kazuhiko takahashi')<br/>('10728256', 'Sae Takahashi', 'sae takahashi')<br/>('1824476', 'Yunduan Cui', 'yunduan cui')<br/>('2565962', 'Masafumi Hashimoto', 'masafumi hashimoto')</td><td>{katakaha@mail,buj1078@mail4}.doshisha.ac.jp +<br/>dum3101@mail4.doshisha.ac.jp +<br/>mhashimo@mail.doshisha.ac.jp +</td></tr><tr><td>3f9a7d690db82cf5c3940fbb06b827ced59ec01e</td><td>VIP: Finding Important People in Images +<br/>Virginia Tech +<br/>Google Inc. +<br/>Virginia Tech +<br/>Project: https://computing.ece.vt.edu/~mclint/vip/ +<br/>Demo: http://cloudcv.org/vip/ +</td><td>('3085140', 'Clint Solomon Mathialagan', 'clint solomon mathialagan')<br/>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')<br/>('1746610', 'Dhruv Batra', 'dhruv batra')</td><td></td></tr><tr><td>3fd90098551bf88c7509521adf1c0ba9b5dfeb57</td><td>Page 1 of 21 +<br/>*****For Peer Review Only***** +<br/>10 +<br/>11 +<br/>12 +<br/>13 +<br/>14 +<br/>15 +<br/>16 +<br/>17 +<br/>18 +<br/>19 +<br/>20 +<br/>21 +<br/>22 +<br/>23 +<br/>24 +<br/>25 +<br/>26 +<br/>27 +<br/>28 +<br/>29 +<br/>30 +<br/>31 +<br/>32 +<br/>33 +<br/>34 +<br/>35 +<br/>36 +<br/>37 +<br/>38 +<br/>39 +<br/>40 +<br/>41 +<br/>42 +<br/>43 +<br/>44 +<br/>45 +<br/>46 +<br/>47 +<br/>48 +<br/>49 +<br/>50 +<br/>51 +<br/>52 +<br/>53 +<br/>54 +<br/>55 +<br/>56 +<br/>57 +<br/>58 +<br/>59 +<br/>60 +<br/>Attribute-Based Classification for Zero-Shot +<br/>Visual Object Categorization +</td><td>('1787591', 'Christoph H. Lampert', 'christoph h. lampert')<br/>('1748758', 'Hannes Nickisch', 'hannes nickisch')<br/>('1734990', 'Stefan Harmeling', 'stefan harmeling')</td><td></td></tr><tr><td>3f623bb0c9c766a5ac612df248f4a59288e4d29f</td><td>Genetic Programming for Region Detection, +<br/>Feature Extraction, Feature Construction and +<br/>Classification in Image Data +<br/>School of Engineering and Computer Science, +<br/><b>Victoria University of Wellington, PO Box 600, Wellington 6140, New Zealand</b></td><td>('39251110', 'Andrew Lensen', 'andrew lensen')<br/>('2480750', 'Harith Al-Sahaf', 'harith al-sahaf')<br/>('1679067', 'Mengjie Zhang', 'mengjie zhang')<br/>('1712740', 'Bing Xue', 'bing xue')</td><td>{Andrew.Lensen,Harith.Al-Sahaf,Mengjie.Zhang,Bing.Xue}@ecs.vuw.ac.nz +</td></tr><tr><td>3f4798c7701da044bdb7feb61ebdbd1d53df5cfe</td><td>VECTOR QUANTIZATION WITH CONSTRAINED LIKELIHOOD FOR FACE +<br/>RECOGNITION +<br/><b>University of Geneva</b><br/>Computer Science Department, Stochastic Information Processing Group +<br/>7 Route de Drize, Geneva, Switzerland +</td><td>('36133844', 'Dimche Kostadinov', 'dimche kostadinov')<br/>('8995309', 'Sviatoslav Voloshynovskiy', 'sviatoslav voloshynovskiy')<br/>('2771643', 'Maurits Diephuis', 'maurits diephuis')<br/>('1682792', 'Sohrab Ferdowsi', 'sohrab ferdowsi')</td><td></td></tr><tr><td>3f4c262d836b2867a53eefb959057350bf7219c9</td><td><b>Eastern Mediterranean University</b><br/>Gazimağusa, Mersin 10, TURKEY. +<br/> +<br/>Occlusions +<br/>Recognizing Faces under Facial Expression Variations and Partial +</td><td>('2108310', 'TIWUYA H. FAAYA', 'tiwuya h. faaya')</td><td></td></tr><tr><td>3f7723ab51417b85aa909e739fc4c43c64bf3e84</td><td>Improved Performance in Facial Expression +<br/>Recognition Using 32 Geometric Features +<br/><b>University of Bari, Bari, Italy</b><br/><b>National Institute of Optics, National Research Council, Arnesano, LE, Italy</b></td><td>('2235498', 'Giuseppe Palestra', 'giuseppe palestra')<br/>('39814343', 'Adriana Pettinicchio', 'adriana pettinicchio')<br/>('33097940', 'Marco Del Coco', 'marco del coco')<br/>('4730472', 'Marco Leo', 'marco leo')<br/>('1741861', 'Cosimo Distante', 'cosimo distante')</td><td>giuseppe.palestra@gmail.com +</td></tr><tr><td>3f5e8f884e71310d7d5571bd98e5a049b8175075</td><td>Making a Science of Model Search: Hyperparameter Optimization +<br/>in Hundreds of Dimensions for Vision Architectures +<br/>J. Bergstra +<br/><b>Rowland Institute at Harvard</b><br/>100 Edwin H. Land Boulevard +<br/>Cambridge, MA 02142, USA +<br/>D. Yamins +<br/>Department of Brain and Cognitive Sciences +<br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, MA 02139, USA +<br/>D. D. Cox +<br/><b>Rowland Institute at Harvard</b><br/>100 Edwin H. Land Boulevard +<br/>Cambridge, MA 02142, USA +</td><td></td><td></td></tr><tr><td>3f63f9aaec8ba1fa801d131e3680900680f14139</td><td>Facial Expression Recognition using Local Binary +<br/>Patterns and Kullback Leibler Divergence +<br/>AnushaVupputuri, SukadevMeher +<br/> +<br/>divergence. +<br/>role +</td><td></td><td></td></tr><tr><td>3f0e0739677eb53a9d16feafc2d9a881b9677b63</td><td>Efficient Two-Stream Motion and Appearance 3D CNNs for +<br/>Video Classification +<br/>ESAT-KU Leuven +<br/>Ali Pazandeh +<br/>Sharif UTech +<br/>ESAT-KU Leuven, ETH Zurich +</td><td>('3310120', 'Ali Diba', 'ali diba')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>ali.diba@esat.kuleuven.be +<br/>pazandeh@ee.sharif.ir +<br/>luc.vangool@esat.kuleuven.be +</td></tr><tr><td>3f5693584d7dab13ffc12122d6ddbf862783028b</td><td>Ranking CGANs: Subjective Control over Semantic Image +<br/>Attributes +<br/><b>University of Bath</b></td><td>('41020280', 'Yassir Saquil', 'yassir saquil')<br/>('1808255', 'Kwang In Kim', 'kwang in kim')</td><td></td></tr><tr><td>30b15cdb72760f20f80e04157b57be9029d8a1ab</td><td>Face Aging with Identity-Preserved +<br/>Conditional Generative Adversarial Networks +<br/><b>Shanghaitech University</b><br/>Baidu +<br/><b>Shanghaitech University</b></td><td>('50219041', 'Zongwei Wang', 'zongwei wang')<br/>('48785141', 'Xu Tang', 'xu tang')<br/>('2074878', 'Weixin Luo', 'weixin luo')<br/>('1702868', 'Shenghua Gao', 'shenghua gao')</td><td>wangzw@shanghaitech.edu.cn +<br/>tangxu02@baidu.com +<br/>{luowx, gaoshh}@shanghaitech.edu.cn +</td></tr><tr><td>3039627fa612c184228b0bed0a8c03c7f754748c</td><td>Robust Regression on Image Manifolds for Ordered Label Denoising +<br/><b>University of North Carolina at Charlotte</b></td><td>('1873911', 'Hui Wu', 'hui wu')<br/>('1690110', 'Richard Souvenir', 'richard souvenir')</td><td>{hwu13,souvenir}@uncc.edu +</td></tr><tr><td>30870ef75aa57e41f54310283c0057451c8c822b</td><td>Overcoming Catastrophic Forgetting with Hard Attention to the Task +</td><td>('50101040', 'Marius Miron', 'marius miron')</td><td></td></tr><tr><td>303065c44cf847849d04da16b8b1d9a120cef73a</td><td></td><td></td><td></td></tr><tr><td>305346d01298edeb5c6dc8b55679e8f60ba97efb</td><td>Article +<br/>Fine-Grained Face Annotation Using Deep +<br/>Multi-Task CNN +<br/><b>Systems and Communication, University of Milano-Bicocca</b><br/>Received: 3 July 2018; Accepted: 13 August 2018; Published: 14 August 2018 +</td><td>('3390122', 'Luigi Celona', 'luigi celona')<br/>('2217051', 'Simone Bianco', 'simone bianco')<br/>('1743714', 'Raimondo Schettini', 'raimondo schettini')</td><td>viale Sarca, 336 Milano, Italy; bianco@disco.unimib.it (S.B.); schettini@disco.unimib.it (R.S.) +<br/>* Correspondence: luigi.celona@disco.unimib.it +</td></tr><tr><td>303a7099c01530fa0beb197eb1305b574168b653</td><td>Occlusion-free Face Alignment: Deep Regression Networks Coupled with +<br/>De-corrupt AutoEncoders +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b><br/>3CAS Center for Excellence in Brain Science and Intelligence Technology +</td><td>('1698586', 'Jie Zhang', 'jie zhang')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{jie.zhang,meina.kan,shiguang.shan,xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>30cd39388b5c1aae7d8153c0ab9d54b61b474ffe</td><td>Deep Cascaded Regression for Face Alignment +<br/><b>School of Data and Computer Science, Sun Yat-Sen University, China</b><br/><b>National University of Singapore, Singapore</b><br/>algorithm refines the shape by estimating a shape increment +<br/>∆S. In particular, a shape increment at stage k is calculated +<br/>as: +</td><td>('3124720', 'Shengtao Xiao', 'shengtao xiao')<br/>('10338111', 'Zhen Cui', 'zhen cui')<br/>('40080379', 'Yan Pan', 'yan pan')<br/>('3029624', 'Chunyan Xu', 'chunyan xu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>303517dfc327c3004ae866a6a340f16bab2ee3e3</td><td>Inte rnational Journal of Engineering Technology, Manage ment and Applied Sciences +<br/>www.ijetmas.com August 2014, Volume 2 Issue 3, ISSN 2349-4476 +<br/> +<br/>Using Locality Preserving Projections in +<br/>Face Recognition +<br/>Galaxy Global Imperial Technical Campus +<br/>Galaxy Global Imperial Technical Campus +<br/><b>DIT UNIVERSITY, DEHRADUN</b></td><td>('34272062', 'PRACHI BANSAL', 'prachi bansal')</td><td></td></tr><tr><td>30fd1363fa14965e3ab48a7d6235e4b3516c1da1</td><td>A Deep Semi-NMF Model for Learning Hidden Representations +<br/>Stefanos Zafeiriou +<br/>Bj¨orn W. Schuller +<br/><b>Imperial College London, United Kingdom</b></td><td>('2814229', 'George Trigeorgis', 'george trigeorgis')<br/>('2732737', 'Konstantinos Bousmalis', 'konstantinos bousmalis')</td><td>GEORGE.TRIGEORGIS08@IMPERIAL.AC.UK +<br/>K.BOUSMALIS@IMPERIAL.AC.UK +<br/>S.ZAFEIRIOU@IMPERIAL.AC.UK +<br/>BJOERN.SCHULLER@IMPERIAL.AC.UK +</td></tr><tr><td>309e17e6223e13b1f76b5b0eaa123b96ef22f51b</td><td>Face Recognition based on a 3D Morphable Model +<br/><b>University of Siegen</b><br/>H¤olderlinstr. 3 +<br/>57068 Siegen, Germany +</td><td>('2880906', 'Volker Blanz', 'volker blanz')</td><td>blanz@informatik.uni-siegen.de +</td></tr><tr><td>3046baea53360a8c5653f09f0a31581da384202e</td><td>Deformable Face Alignment via Local +<br/>Measurements and Global Constraints +</td><td>('2398245', 'Jason M. Saragih', 'jason m. saragih')</td><td></td></tr><tr><td>3026722b4cbe9223eda6ff2822140172e44ed4b1</td><td>Jointly Estimating Demographics and Height with a Calibrated Camera +<br/>Eastman Kodak Company +<br/>Eastman Kodak Company +<br/><b>Cornell University</b></td><td>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')<br/>('2224373', 'Andrew C. Blose', 'andrew c. blose')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td>andrew.gallagher@kodak.com +<br/>andrew.blose@kodak.com +<br/>tsuhan@ece.cornell.edu +</td></tr><tr><td>3028690d00bd95f20842d4aec84dc96de1db6e59</td><td>Leveraging Union of Subspace Structure to Improve Constrained Clustering +</td><td>('1782134', 'John Lipor', 'john lipor')</td><td></td></tr><tr><td>30c96cc041bafa4f480b7b1eb5c45999701fe066</td><td>1090 +<br/>Discrete Cosine Transform Locality-Sensitive +<br/>Hashes for Face Retrieval +</td><td>('1784929', 'Mehran Kafai', 'mehran kafai')<br/>('1745657', 'Kave Eshghi', 'kave eshghi')<br/>('1707159', 'Bir Bhanu', 'bir bhanu')</td><td></td></tr><tr><td>306957285fea4ce11a14641c3497d01b46095989</td><td>FACE RECOGNITION UNDER VARYING LIGHTING BASED ON +<br/>DERIVATES OF LOG IMAGE +<br/>2ICT-ISVISION Joint R&D Laboratory for Face Recognition, CAS, Beijing 100080, China +<br/>1Graduate School, CAS, Beijing, 100039, China +</td><td>('2343895', 'Laiyun Qing', 'laiyun qing')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1698902', 'Wen Gao', 'wen gao')</td><td></td></tr><tr><td>304b1f14ca6a37552dbfac443f3d5b36dbe1a451</td><td>Collaborative Low-Rank Subspace Clustering +<br/><b>aSchool of Computing and Mathematics, Charles Sturt University, Bathurst, NSW</b><br/><b>bDiscipline of Business Analytics, The University of Sydney Business School</b><br/><b>The University of Sydney, NSW 2006, Australia</b><br/>cCentre for Research in Mathematics, School of Computing, Engineering and Mathematics, +<br/><b>Western Sydney University, Parramatta, NSW 2150, Australia</b><br/>Australia +</td><td>('40635684', 'Stephen Tierney', 'stephen tierney')<br/>('1767638', 'Yi Guo', 'yi guo')<br/>('1750488', 'Junbin Gao', 'junbin gao')</td><td></td></tr><tr><td>306127c3197eb5544ab1e1bf8279a01e0df26120</td><td>Sparse Coding and Dictionary Learning with Linear Dynamical Systems∗ +<br/><b>Tsinghua University, State Key Lab. of Intelligent</b><br/>Technology and Systems, Tsinghua National Lab. for Information Science and Technology (TNList); +<br/><b>Australian National University and NICTA, Australia</b></td><td>('36823190', 'Fuchun Sun', 'fuchun sun')<br/>('1678783', 'Deli Zhao', 'deli zhao')<br/>('2641547', 'Huaping Liu', 'huaping liu')<br/>('23911916', 'Mehrtash Harandi', 'mehrtash harandi')</td><td>1{huangwb12@mails, fcsun@mail, caoll12@mails, hpliu@mail}.tsinghua.edu.cn, +<br/>2zhaodeli@gmail.com, 3Mehrtash.Harandi@nicta.com.au, +</td></tr><tr><td>307a810d1bf6f747b1bd697a8a642afbd649613d</td><td>An affordable contactless security system access +<br/> for restricted area +<br/>Laboratory Le2i +<br/><b>University Bourgogne Franche-Comt , France</b><br/>2 Odalid compagny, France +<br/>Keywords – Smart Camera, Real-time Image Processing, Biometrics, Face Detection, Face Verifica- +<br/>tion, EigenFaces, Support Vector Machine, +<br/>We present in this paper a security system based on +<br/>identity verification process and a low-cost smart cam- +<br/>era, intended to avoid unauthorized access to restricted +<br/>area. The Le2i laboratory has a longstanding experi- +<br/>ence in smart cameras implementation and design [1], +<br/>for example in the case of real-time classical face de- +<br/>tection [2] or human fall detection [3]. +<br/>The principle of the system, fully thought and designed +<br/>in our laboratory, is as follows: the allowed user pre- +<br/>sents a RFID card to the reader based on Odalid system +<br/>[4]. The card ID, time and date of authorized access are +<br/>checked using connection to an online server. In the +<br/>same time, multi-modality identity verification is per- +<br/>formed using the camera. +<br/>There are many ways to perform face recognition and +<br/>face verification. As a first approach, we implemented a +<br/>standard face localization using Haar cascade [5] and +<br/>verification process based on Eigenfaces (feature ex- +<br/>traction), with the ORL face data base (or AT&T) [6], and +<br/>SVM (verification) [7]. +<br/>On the one hand, the training step has been performed +<br/>with 10-folds cross validation using the 3000 first faces +<br/>from LFW face database [8] as unauthorized class and +<br/>20 known faces were used for the authorized class. On +<br/>the other hand, the testing step has been performed us- +<br/>ing the rest of the LFW data base and 40 other faces +<br/>from the same known person. The false positive and +<br/>false negative rates are respectively 0,004% and 1,39% +<br/>with a standard deviation of respectively 0,006% and +<br/>2,08%, considering a precision of 98,9% and a recall of +<br/>98,6%. +<br/>The current PC based implementation has been de- +<br/>signed to be easily deployed on a Raspberry Pi3 or sim- +<br/>ilar based target. A combination of Eigenfaces [9], Fish- +<br/>erfaces [9] , Local Binary Patterns [9] and Generalized +<br/>Fourier Descriptors [10] will be also studied. +<br/>However, it is known that the use of single modality such +<br/>as standard face luminosity for identity control leads of- +<br/>ten to ergonomics problems due to the high intra-varia- +<br/>bility of human faces [11]. Recent work published in the +<br/>literature and developed in our laboratory showed that +<br/>it is possible to extract precise multispectral body infor- +<br/>mation from standard camera. +<br/>The next step and originality of our system will resides +<br/>in the fact that we will consider Near Infrared or multi- +<br/>spectral approach in order to improve the security level +<br/>(decrease false positive rate) as well as ergonomics +<br/>(decrease false negative rate). +<br/>The proposed platform enables security access to be +<br/>improved and original solutions based on specific illumi- +<br/>nation to be investigated. +<br/>ACKNOWLEDGMENT +<br/>This research was supported by the Conseil Regional +<br/>de Bourgogne Franche-Comte, and institut Carnot +<br/>ARTS +<br/>REFERENCES +<br/>[1] R. Mosqueron, J. Dubois, M. Mattavelli, D. Mauvilet, Smart camera +<br/>based on embedded HW/SW coprocessor, EURASIP Journal on Em- +<br/>bedded Systems, p.3:1-3:13, Hindawi Publishing Corp, 2008. +<br/>[2] K. Khattab, J. Dubois, J. Miteran, Cascade Boosting Based Object +<br/>Detection from High-Level Description to Hardware Implementation, +<br/>EURASIP Journal on Embedded System, August 2009 +<br/>[3] B. Senouci, I. Charfi, B. Barthelemy, J. Dubois, J. Miteran, Fast +<br/>prototyping of a SoC-based smart-camera: a real-time fall detection +<br/>case study, Journal of Real-Time Image Processing, p.1-14, 2014. +<br/>[4] http://odalid.com/ +<br/>[5] P. Viola, M.J. Jones, Robust Real-Time Face Detection, Interna- +<br/>tional Journal of Computer Vision, p137-154, May 2004 +<br/>[6] www.cl.cam.ac.uk/research/dtg/attarchive/facedatabase.html +<br/>[7] K. Jonsson, J. Kittler, Y.P. Li, J. Matas, Support Vector Machines +<br/>for Face Authentication, Image and Vision Computing, p543-553, +<br/>1999 +<br/>[8] G.B. Huang, M. Ramesh, T. Berg, E. Learned-Miller, Labeled +<br/>Faces in the Wild: A Database for Studying Face Recognition in Un- +<br/>constrained Environments, Tehcnical Report p07-49, October 2007 +<br/>[9] R. Jafri, H.R. Arabnia, A Survey of Face Recognition Techniques, +<br/>Journal of Information Processing Systems, p41-68, June 2009 +<br/>[10] F. Smach, C. Lemaitre, J-P. Gauthier, J. Miteran, M. Atri, Gener- +<br/>alized Fourier Descriptors With Applications to Objects Recognition in +<br/>SVM Context, Journal of Mathematical Imaging and Vision, p43-47, +<br/>2007 +<br/>[11] T. Bourlai, B. Cukic, Multi-Spectral Face Recognition: Identifica- +<br/>tion of People in Difficult Environments, p196-201, June 2012 +</td><td>('2787483', 'Johel Mitéran', 'johel mitéran')<br/>('2274333', 'Barthélémy Heyrman', 'barthélémy heyrman')<br/>('1873153', 'Dominique Ginhac', 'dominique ginhac')<br/>('33359945', 'Julien Dubois', 'julien dubois')</td><td>Contact julien.dubois@u-bourgogne.fr +</td></tr><tr><td>30180f66d5b4b7c0367e4b43e2b55367b72d6d2a</td><td>Template Adaptation for Face Verification and Identification +<br/>1 Systems and Technology Research, Woburn MA USA +<br/>2 Visionary Systems and Research, Framingham, MA USA +<br/><b>Visual Geometry Group, University of Oxford, Oxford UK</b></td><td>('3390731', 'Nate Crosswhite', 'nate crosswhite')<br/>('36067742', 'Jeffrey Byrne', 'jeffrey byrne')<br/>('34712076', 'Chris Stauffer', 'chris stauffer')<br/>('1954340', 'Qiong Cao', 'qiong cao')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>3083d2c6d4f456e01cbb72930dc2207af98a6244</td><td>16 +<br/>Perceived Age Estimation from Face Images +<br/>1NEC Soft, Ltd. +<br/><b>Tokyo Institute of Technology</b><br/>Japan +<br/>1. Introduction +<br/>In recent years, demographic analysis in public places such as shopping malls and stations +<br/>is attracting a great deal of attention. Such demographic information is useful for various +<br/>purposes, e.g., designing effective marketing strategies and targeted advertisement based +<br/>on customers’ gender and age. +<br/>For this reason, a number of approaches have been +<br/>explored for age estimation from face images (Fu et al., 2007; Geng et al., 2006; Guo et al., +<br/>2009), and several databases became publicly available recently (FG-Net Aging Database, +<br/>n.d.; Phillips et al., 2005; Ricanek & Tesafaye, 2006). +<br/>It has been reported that age can be +<br/>accurately estimated under controlled environment such as frontal faces, no expression, and +<br/>static lighting conditions. However, it is not straightforward to achieve the same accuracy +<br/>level in a real-world environment due to considerable variations in camera settings, facial +<br/>poses, and illumination conditions. The recognition performance of age prediction systems is +<br/>significantly influenced by such factors as the type of camera, camera calibration, and lighting +<br/>variations. On the other hand, the publicly available databases were mainly collected in +<br/>semi-controlled environments. For this reason, existing age prediction systems built upon +<br/>such databases tend to perform poorly in a real-world environment. +<br/>In this chapter, we address the problem of perceived age estimation from face images, and +<br/>describe our new approaches proposed in Ueki et al. (2010) and Ueki et al. (2011), which +<br/>involve three novel aspects. +<br/>The first novelty of our proposed approaches is to take the heterogeneous characteristics of +<br/>human age perception into account. +<br/>It is rare to misjudge the age of a 5-year-old child as +<br/>15 years old, but the age of a 35-year-old person is often misjudged as 45 years old. Thus, +<br/>magnitude of the error is different depending on subjects’ age. We carried out a large-scale +<br/>questionnaire survey for quantifying human age perception characteristics, and propose to +<br/>utilize the quantified characteristics in the framework of weighted regression. +<br/>The second is an efficient active learning strategy for reducing the cost of labeling face +<br/>samples. Given a large number of unlabeled face samples, we reveal the cluster structure +<br/>of the data and propose to label cluster-representative samples for covering as many +<br/>clusters as possible. This simple sampling strategy allows us to boost the performance of +<br/>a manifold-based semi-supervised learning method only with a relatively small number of +<br/>labeled samples. +<br/>The third contribution is to apply a recently proposed machine learning technique called +<br/>covariate shift adaptation (Shimodaira, 2000; Sugiyama & Kawanabe, 2011; Sugiyama et al., +</td><td>('2163491', 'Kazuya Ueki', 'kazuya ueki')<br/>('1853974', 'Yasuyuki Ihara', 'yasuyuki ihara')<br/>('1719221', 'Masashi Sugiyama', 'masashi sugiyama')</td><td></td></tr><tr><td>30cbd41e997445745b6edd31f2ebcc7533453b61</td><td>What Makes a Video a Video: Analyzing Temporal Information in Video +<br/>Understanding Models and Datasets +<br/><b>Stanford University, 2Facebook, 3Dartmouth College</b></td><td>('38485317', 'De-An Huang', 'de-an huang')<br/>('34066479', 'Vignesh Ramanathan', 'vignesh ramanathan')<br/>('49274550', 'Dhruv Mahajan', 'dhruv mahajan')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')<br/>('2210374', 'Manohar Paluri', 'manohar paluri')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')<br/>('9200530', 'Juan Carlos Niebles', 'juan carlos niebles')</td><td></td></tr><tr><td>302c9c105d49c1348b8f1d8cc47bead70e2acf08</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TCSVT.2017.2710120, IEEE +<br/>Transactions on Circuits and Systems for Video Technology +<br/>IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY +<br/>Unconstrained Face Recognition Using A Set-to-Set +<br/>Distance Measure +</td><td>('4712803', 'Jiaojiao Zhao', 'jiaojiao zhao')<br/>('1783847', 'Jungong Han', 'jungong han')</td><td></td></tr><tr><td>304a306d2a55ea41c2355bd9310e332fa76b3cb0</td><td></td><td></td><td></td></tr><tr><td>301b0da87027d6472b98361729faecf6e1d5e5f6</td><td>HEAD POSE ESTIMATION IN FACE RECOGNITION ACROSS +<br/>POSE SCENARIOS +<br/><b>Computer vision and Remote Sensing, Berlin university of Technology</b><br/>Sekr. FR-3-1, Franklinstr. 28/29, D-10587, Berlin, Germany. +<br/>Keywords: +<br/>Pose estimation, facial pose, face recognition, local energy models, shape description, local features, head +<br/>pose classification. +</td><td>('4241648', 'M. Saquib Sarfraz', 'm. saquib sarfraz')<br/>('2962236', 'Olaf Hellwich', 'olaf hellwich')</td><td>{saquib;hellwich}@fpk.tu-berlin.de +</td></tr><tr><td>30b103d59f8460d80bb9eac0aa09aaa56c98494f</td><td>Enhancing Human Action Recognition with Region Proposals +<br/>Australian Centre for Robotic Vision(ACRV), School of Electrical Engineering and Computer Science +<br/><b>Queensland University of Technology(QUT</b></td><td>('2256817', 'Fahimeh Rezazadegan', 'fahimeh rezazadegan')<br/>('34686772', 'Sareh Shirazi', 'sareh shirazi')<br/>('1771913', 'Niko Sünderhauf', 'niko sünderhauf')<br/>('1809144', 'Michael Milford', 'michael milford')<br/>('1803115', 'Ben Upcroft', 'ben upcroft')</td><td>fahimeh.rezazadegan@qut.edu.au +</td></tr><tr><td>5e59193a0fc22a0c37301fb05b198dd96df94266</td><td>Example-Based Modeling of Facial Texture from Deficient Data +<br/>1 IMB / LaBRI, Universit´e de Bordeaux, France +<br/><b>University of York, UK</b></td><td>('34895713', 'Arnaud Dessein', 'arnaud dessein')<br/>('1679753', 'Edwin R. Hancock', 'edwin r. hancock')<br/>('1687021', 'William A. P. Smith', 'william a. p. smith')<br/>('1718243', 'Richard C. Wilson', 'richard c. wilson')</td><td></td></tr><tr><td>5e6f546a50ed97658be9310d5e0a67891fe8a102</td><td>Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet? +<br/><b>National Institute of Advanced Industrial Science and Technology (AIST</b><br/>Tsukuba, Ibaraki, Japan +</td><td>('2199251', 'Kensho Hara', 'kensho hara')<br/>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')<br/>('1732705', 'Yutaka Satoh', 'yutaka satoh')</td><td>{kensho.hara, hirokatsu.kataoka, yu.satou}@aist.go.jp +</td></tr><tr><td>5e0eb34aeb2b58000726540336771053ecd335fc</td><td>Low-Quality Video Face Recognition with Deep +<br/>Networks and Polygonal Chain Distance +<br/><b>Vision and Fusion Lab, Karlsruhe Institute of Technology KIT, Karlsruhe, Germany</b><br/>†Fraunhofer IOSB, Karlsruhe, Germany +</td><td>('37646107', 'Christian Herrmann', 'christian herrmann')<br/>('1783486', 'Dieter Willersinn', 'dieter willersinn')</td><td>{christian.herrmann|dieter.willersinn|juergen.beyerer}@iosb.fraunhofer.de +</td></tr><tr><td>5e7e055ef9ba6e8566a400a8b1c6d8f827099553</td><td></td><td></td><td>Accepted manuscripts are peer-reviewed but have not been through the copyediting, formatting, or proofreadingprocess.Copyright © 2018 the authorsThis Accepted Manuscript has not been copyedited and formatted. The final version may differ from this version.Research Articles: Behavioral/CognitiveOn the role of cortex-basal ganglia interactions for category learning: Aneuro-computational approachFrancesc Villagrasa1, Javier Baladron1, Julien Vitay1, Henning Schroll1, Evan G. Antzoulatos2, Earl K.Miller3 and Fred H. Hamker11Chemnitz University of Technology, Department of Computer Science, 09107 Chemnitz, Germany2UC Davis Center for Neuroscience and Department of Neurobiology, Physiology and Behavior, Davis, CA95616, United States3The Picower Institute for Learning and Memory and Department of Brain and Cognitive Sciences,Massachusetts Institute of Technology, Cambridge, MA 02139, United StatesDOI: 10.1523/JNEUROSCI.0874-18.2018Received: 5 April 2018Revised: 7 August 2018Accepted: 28 August 2018Published: 18 September 2018Author contributions: F.V., J.V., E.G.A., and F.H.H. performed research; F.V., J.B., J.V., H.S., E.G.A., andE.K.M. analyzed data; F.V. wrote the first draft of the paper; J.B. and F.H.H. designed research; J.B., J.V., H.S.,E.G.A., E.K.M., and F.H.H. edited the paper; F.H.H. wrote the paper.Conflict of Interest: The authors declare no competing financial interests.This work has been supported by the German Research Foundation (DFG, grant agreements no. HA2630/4-2and HA2630/8-1), the European Social Fund and the Free State of Saxony (ESF, grant agreement no.ESF-100269974), the NIMH R01MH065252, and the MIT Picower Institute Innovation Fund.Corresponding author: Fred H. Hamker, fred.hamker@informatik.tu-chemnitz.de, 09107 Chemnitz, GermanyCite as: J. Neurosci ; 10.1523/JNEUROSCI.0874-18.2018Alerts: Sign up at www.jneurosci.org/cgi/alerts to receive customized email alerts when the fully formattedversion of this article is published. </td></tr><tr><td>5e28673a930131b1ee50d11f69573c17db8fff3e</td><td>Author manuscript, published in "Workshop on Faces in 'Real-Life' Images: Detection, Alignment, and Recognition, Marseille : France +<br/>(2008)" +</td><td></td><td></td></tr><tr><td>5ea9063b44b56d9c1942b8484572790dff82731e</td><td>MULTICLASS SUPPORT VECTOR MACHINES AND METRIC MULTIDIMENSIONAL +<br/>SCALING FOR FACIAL EXPRESSION RECOGNITION +<br/>Irene Kotsiay, Stefanos Zafeiriouy, Nikolaos Nikolaidisy and Ioannis Pitasy +<br/><b>yAristotle University of Thessaloniki</b><br/>Thessaloniki, Greece +</td><td></td><td>email: fekotsia, dralbert, nikolaid, pitasg@aiia.csd.auth.gr +</td></tr><tr><td>5e16f10f2d667d17c029622b9278b6b0a206d394</td><td>Learning to Rank Binary Codes +<br/><b>Columbia University</b><br/><b>IBM T. J. Watson Research Center</b><br/><b>Columbia University</b></td><td>('1710567', 'Jie Feng', 'jie feng')<br/>('1722649', 'Wei Liu', 'wei liu')<br/>('1678691', 'Yan Wang', 'yan wang')</td><td></td></tr><tr><td>5ef3e7a2c8d2876f3c77c5df2bbaea8a777051a7</td><td>Rendering or normalization? +<br/>An analysis of the 3D-aided pose-invariant face recognition +<br/>Computational Biomedicine Lab +<br/><b>University of Houston, Houston, TX, USA</b></td><td>('2461369', 'Yuhang Wu', 'yuhang wu')<br/>('2700399', 'Shishir K. Shah', 'shishir k. shah')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td>ywu36@uh.edu {sshah,ikakadia}@central.uh.edu +</td></tr><tr><td>5ea165d2bbd305dc125415487ef061bce75dac7d</td><td>Efficient Human Action Recognition by Luminance Field Trajectory and Geometry Information +<br/><b>Hong Kong Polytechnic University, Hong Kong, China</b><br/>2BBN Technologies, Cambridge, MA 02138, USA +</td><td>('3079962', 'Haomian Zheng', 'haomian zheng')<br/>('2659956', 'Zhu Li', 'zhu li')<br/>('1708679', 'Yun Fu', 'yun fu')</td><td>{cshmzheng,cszli}@comp.polyu.edu.hk, yfu@bbn.com +</td></tr><tr><td>5e6ba16cddd1797853d8898de52c1f1f44a73279</td><td>Face Identification with Second-Order Pooling +</td><td>('2731972', 'Fumin Shen', 'fumin shen')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')<br/>('1724393', 'Heng Tao Shen', 'heng tao shen')</td><td></td></tr><tr><td>5ea9cba00f74d2e113a10c484ebe4b5780493964</td><td>Automated Drowsiness Detection For Improved +<br/>Driving Safety +<br/><b>Sabanci University</b><br/>Faculty of +<br/>Engineering and Natural Sciences +<br/>Orhanli, Istanbul +<br/><b>University of California San Diego</b><br/><b>Institute of</b><br/>Neural Computation +<br/>La Jolla, San Diego +</td><td>('40322754', 'Esra Vural', 'esra vural')<br/>('21691177', 'Mujdat Cetin', 'mujdat cetin')<br/>('2724380', 'Gwen Littlewort', 'gwen littlewort')<br/>('1858421', 'Marian Bartlett', 'marian bartlett')<br/>('29794862', 'Javier Movellan', 'javier movellan')</td><td></td></tr><tr><td>5e80e2ffb264b89d1e2c468fbc1b9174f0e27f43</td><td>Naming Every Individual in News Video Monologues +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave., Pittsburgh, PA 15213, USA +<br/>1-412-268-{9747, 1448} +</td><td>('38936351', 'Jun Yang', 'jun yang')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td>{juny, alex}@cs.cmu.edu +</td></tr><tr><td>5ec94adc9e0f282597f943ea9f4502a2a34ecfc2</td><td>Leveraging the Power of Gabor Phase for Face +<br/>Identification: A Block Matching Approach +<br/><b>KTH, Royal Institute of Technology</b></td><td>('39750744', 'Yang Zhong', 'yang zhong')<br/>('40565290', 'Haibo Li', 'haibo li')</td><td></td></tr><tr><td>5e0e516226413ea1e973f1a24e2fdedde98e7ec0</td><td>The Invariance Hypothesis and the Ventral Stream +<br/>by +<br/><b>B.S./M.S. Brandeis University</b><br/>Submitted to the Department of Brain and Cognitive Sciences +<br/>in partial fulfillment of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>at the +<br/><b>MASSACHUSETTS INSTITUTE OF TECHNOLOGY</b><br/>February 2014 +<br/><b>Massachusetts Institute of Technology 2014. All rights reserved</b><br/>Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Department of Brain and Cognitive Sciences +<br/>September 5, 2013 +<br/>Certified by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Thesis Supervisor +<br/>Accepted by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Sherman Fairchild Professor of Neuroscience and Picower Scholar +<br/>Director of Graduate Education for Brain and Cognitive Sciences +</td><td>('1700356', 'Joel Zaidspiner Leibo', 'joel zaidspiner leibo')<br/>('5856191', 'Tomaso Poggio', 'tomaso poggio')<br/>('1724891', 'Eugene McDermott', 'eugene mcdermott')<br/>('3034182', 'Matthew Wilson', 'matthew wilson')</td><td></td></tr><tr><td>5e821cb036010bef259046a96fe26e681f20266e</td><td></td><td></td><td></td></tr><tr><td>5e7cb894307f36651bdd055a85fdf1e182b7db30</td><td>A Comparison of Multi-class Support Vector Machine Methods for +<br/>Face Recognition +<br/>Department of Electrical and Computer Engineering +<br/><b>The University of Maryland</b><br/>December 6, 2007 +</td><td></td><td>Naotoshi Seo, sonots@umd.edu +</td></tr><tr><td>5b693cb3bedaa2f1e84161a4261df9b3f8e77353</td><td>Proc. VIIth Digital Image Computing: Techniques and Applications, Sun C., Talbot H., Ourselin S. and Adriaansen T. (Eds.), 10-12 Dec. 2003, Sydney +<br/>Robust Face Localisation Using Motion, Colour +<br/>& Fusion +<br/>Speech, Audio, Image and Video Technologies Program +<br/>Faculty of Built Environment and Engineering +<br/><b>Queensland University of Technology</b><br/>GPO Box 2434, Brisbane QLD 4001, Australia +<br/>http://www.bee.qut.edu.au/research/prog_saivt.shtml +</td><td>('1763662', 'Chris McCool', 'chris mccool')<br/>('33258846', 'Matthew McKay', 'matthew mckay')<br/>('40453073', 'Scott Lowther', 'scott lowther')<br/>('1729760', 'Sridha Sridharan', 'sridha sridharan')</td><td></td></tr><tr><td>5b73b7b335f33cda2d0662a8e9520f357b65f3ac</td><td>Intensity Rank Estimation of Facial Expressions +<br/>Based on A Single Image +<br/><b>Institute of Information Science, Academia Sinica, Taipei, Taiwan</b><br/><b>National Taiwan University, Taipei, Taiwan</b><br/><b>Research Center for Information Technology Innovation, Academia Sinica, Taipei, Taiwan</b><br/><b>Graduate Institute of Networking and Multimedia, National Taiwan University, Taipei, Taiwan</b></td><td>('34692779', 'Kuang-Yu Chang', 'kuang-yu chang')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')<br/>('1732064', 'Yi-Ping Hung', 'yi-ping hung')</td><td>Email: song@iis.sinica.edu.tw +</td></tr><tr><td>5b6d05ce368e69485cb08dd97903075e7f517aed</td><td>Robust Active Shape Model for +<br/>Landmarking Frontal Faces +<br/>Department of Electrical and Computer Engineering +<br/><b>Carnegie Mellon University Pittsburgh, PA - 15213, USA</b><br/>June 15, 2009 +</td><td>('2363348', 'Keshav Seshadri', 'keshav seshadri')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td>kseshadr@andrew.cmu.edu, msavvid@cs.cmu.edu +</td></tr><tr><td>5b0bf1063b694e4b1575bb428edb4f3451d9bf04</td><td>Facial shape tracking via spatio-temporal cascade shape regression +<br/><b>Nanjing University of Information Science and Technology</b><br/>Nanjing, China +</td><td>('37953909', 'Jing Yang', 'jing yang')<br/>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('3198263', 'Kaihua Zhang', 'kaihua zhang')<br/>('1734954', 'Qingshan Liu', 'qingshan liu')</td><td>nuist yj@126.com +<br/>jiankangdeng@gmail.com +<br/>zhkhua@gmail.com +<br/>qsliu@nuist.edu.cn +</td></tr><tr><td>5b59e6b980d2447b2f3042bd811906694e4b0843</td><td>Two-stage Cascade Model for Unconstrained +<br/>Face Detection +<br/>Darijan Marčetić, Tomislav Hrkać, Slobodan Ribarić +<br/><b>University of Zagreb, Faculty of Electrical Engineering and Computing, Croatia</b></td><td></td><td>{darijan.marcetic, tomislav.hrkac, slobodan.ribaric}@fer.hr +</td></tr><tr><td>5bb53fb36a47b355e9a6962257dd465cd7ad6827</td><td>Mask-off: Synthesizing Face Images in the Presence of Head-mounted Displays +<br/><b>University of Kentucky</b><br/><b>North Carolina Central University</b><br/>Figure 1: Our system automatically reconstruct photo-realistic face videos for users wearing HMD. (Left) Input NIR eye images. (Middle) +<br/>Input face image with upper face blocked by HMD device. (Right) The output of our system. +</td><td>('2613340', 'Yajie Zhao', 'yajie zhao')<br/>('8285167', 'Qingguo Xu', 'qingguo xu')<br/>('2257812', 'Xinyu Huang', 'xinyu huang')<br/>('38958903', 'Ruigang Yang', 'ruigang yang')</td><td></td></tr><tr><td>5b89744d2ac9021f468b3ffd32edf9c00ed7fed7</td><td>Beyond Mahalanobis Metric: Cayley-Klein Metric Learning +<br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/>Beijing, 100190, China +</td><td>('2495602', 'Yanhong Bi', 'yanhong bi')<br/>('1684958', 'Bin Fan', 'bin fan')<br/>('3104867', 'Fuchao Wu', 'fuchao wu')</td><td>{yanhong.bi, bfan, fcwu}@nlpr.ia.ac.cn +</td></tr><tr><td>5bfc32d9457f43d2488583167af4f3175fdcdc03</td><td>International Journal of Science and Research (IJSR), India Online ISSN: 2319-7064 +<br/>Local Gray Code Pattern (LGCP): A Robust +<br/>Feature Descriptor for Facial Expression +<br/>Recognition +</td><td>('7484236', 'Mohammad Shahidul Islam', 'mohammad shahidul islam')</td><td></td></tr><tr><td>5b7cb9b97c425b52b2e6f41ba8028836029c4432</td><td>Smooth Representation Clustering +<br/>1State Key Laboratory on Intelligent Technology and Systems, TNList +<br/><b>Tsinghua University</b><br/><b>Key Lab. of Machine Perception, School of EECS, Peking University</b></td><td>('40234323', 'Han Hu', 'han hu')<br/>('33383055', 'Zhouchen Lin', 'zhouchen lin')<br/>('2632601', 'Jianjiang Feng', 'jianjiang feng')<br/>('39491387', 'Jie Zhou', 'jie zhou')</td><td>huh04@mails.thu.edu.cn, zlin@pku.edu.cn, {jfeng,jzhou}@tsinghua.edu.cn +</td></tr><tr><td>5ba7882700718e996d576b58528f1838e5559225</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TAFFC.2016.2628787, IEEE +<br/>Transactions on Affective Computing +<br/>IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, VOL. X, NO. X, OCTOBER 2016 +<br/>Predicting Personalized Image Emotion +<br/>Perceptions in Social Networks +</td><td>('1755487', 'Sicheng Zhao', 'sicheng zhao')<br/>('1720100', 'Hongxun Yao', 'hongxun yao')<br/>('33375873', 'Yue Gao', 'yue gao')<br/>('38329336', 'Guiguang Ding', 'guiguang ding')<br/>('1684968', 'Tat-Seng Chua', 'tat-seng chua')</td><td></td></tr><tr><td>5b6f0a508c1f4097dd8dced751df46230450b01a</td><td>Finding Lost Children +<br/>Ashley Michelle Eden +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2010-174 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2010/EECS-2010-174.html +<br/>December 20, 2010 +</td><td></td><td></td></tr><tr><td>5b9d41e2985fa815c0f38a2563cca4311ce82954</td><td>Exploitation of 3D Images for Face Authentication Under Pose and Illumination +<br/>Variations +<br/>1Information Processing Laboratory, Electrical and Computer Engineering Department, +<br/><b>Aristotle University of Thessaloniki, Thessaloniki 541 24, Greece</b><br/><b>Informatics and Telematics Institute, Centre for Research and Technology Hellas</b><br/>1st Km Thermi-Panorama Rd, Thessaloniki 57001, Greece +</td><td>('1807962', 'Filareti Tsalakanidou', 'filareti tsalakanidou')<br/>('1744180', 'Sotiris Malassiotis', 'sotiris malassiotis')<br/>('1721460', 'Michael G. Strintzis', 'michael g. strintzis')</td><td>Email: filareti@iti.gr, malasiot@iti.gr, strintzi@eng.auth.gr +</td></tr><tr><td>5b6593a6497868a0d19312952d2b753232414c23</td><td>Face Recognition by 3D Registration for the +<br/>Visually Impaired Using a RGB-D Sensor +<br/><b>The City College of New York, New York, NY 10031, USA</b><br/><b>Beihang University, Beijing 100191, China</b><br/>3 The CUNY Graduate Center, New York, NY 10016, USA +</td><td>('40617554', 'Wei Li', 'wei li')<br/>('3042950', 'Xudong Li', 'xudong li')<br/>('40152663', 'Martin Goldberg', 'martin goldberg')<br/>('4697712', 'Zhigang Zhu', 'zhigang zhu')</td><td>lwei000@citymail.cuny.edu, xdli@buaa.edu.cn, +<br/>mgoldberg@gc.cuny.edu, zhu@cs.ccny.cuny.edu +</td></tr><tr><td>5bb684dfe64171b77df06ba68997fd1e8daffbe1</td><td></td><td></td><td></td></tr><tr><td>5b719410e7829c98c074bc2947697fac3b505b64</td><td>ACTIVE APPEARANCE MODELS FOR AFFECT RECOGNITION USING FACIAL +<br/>EXPRESSIONS +<br/>Matthew Stephen Ratliff +<br/><b>University of North Carolina Wilmington in Partial Ful llment</b><br/>A Thesis Submitted to the +<br/>of the Requirements for the Degree of +<br/>Master of Science +<br/>Department of Computer Science +<br/>Department of Information Systems and Operations Management +<br/><b>University of North Carolina Wilmington</b><br/>2010 +<br/>Approved by +<br/>Advisory Committee +<br/>Curry Guinn +<br/>Thomas Janicki +<br/>Eric Patterson +<br/>Chair +<br/>Accepted by +<br/>Dean, Graduate School +</td><td></td><td></td></tr><tr><td>5bae9822d703c585a61575dced83fa2f4dea1c6d</td><td>MOTChallenge 2015: +<br/>Towards a Benchmark for Multi-Target Tracking +</td><td>('34761498', 'Anton Milan', 'anton milan')<br/>('34493380', 'Stefan Roth', 'stefan roth')<br/>('1803034', 'Konrad Schindler', 'konrad schindler')</td><td></td></tr><tr><td>5b0008ba87667085912ea474025d2323a14bfc90</td><td>SoS-RSC: A Sum-of-Squares Polynomial Approach to Robustifying Subspace +<br/>Clustering Algorithms∗ +<br/>Electrical and Computer Engineering +<br/><b>Northeastern University, Boston, MA</b></td><td>('1687866', 'Mario Sznaier', 'mario sznaier')</td><td>{msznaier,camps}@coe.neu.edu +</td></tr><tr><td>5b97e997b9b654373bd129b3baf5b82c2def13d1</td><td>3D Face Tracking and Texture Fusion in the Wild +<br/>Centre for Vision, Speech and Signal Processing +<br/>Image Understanding and Interactive Robotics +<br/><b>University of Surrey</b><br/>Guildford, GU2 7XH, United Kingdom +<br/>Contact: http://www.patrikhuber.ch +<br/><b>Reutlingen University</b><br/>D-72762 Reutlingen, Germany +</td><td>('39976184', 'Patrik Huber', 'patrik huber')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('49330989', 'Philipp Kopp', 'philipp kopp')</td><td></td></tr><tr><td>5bd3d08335bb4e444a86200c5e9f57fd9d719e14</td><td>3D Face Morphable Models “In-the-Wild” +<br/>,∗ +<br/>Stefanos Zafeiriou1 +<br/><b>Imperial College London, UK</b><br/>2Amazon, Berlin, Germany +<br/><b>University of Oulu, Finland</b></td><td>('47456731', 'James Booth', 'james booth')<br/>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('2015036', 'Stylianos Ploumpis', 'stylianos ploumpis')<br/>('2814229', 'George Trigeorgis', 'george trigeorgis')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')</td><td>1{james.booth,s.ploumpis,g.trigeorgis,i.panagakis,s.zafeiriou}@imperial.ac.uk +<br/>2antonak@amazon.com +</td></tr><tr><td>5babbad3daac5c26503088782fd5b62067b94fa5</td><td>Are You Sure You Want To Do That? +<br/>Classification with Verification +</td><td>('31920847', 'Harris Chan', 'harris chan')<br/>('36964031', 'Atef Chaudhury', 'atef chaudhury')<br/>('50715871', 'Kevin Shen', 'kevin shen')</td><td>hchan@cs.toronto.edu +<br/>atef@cs.toronto.edu +<br/>shenkev@cs.toronto.edu +</td></tr><tr><td>5bb87c7462c6c1ec5d60bde169c3a785ba5ea48f</td><td>Targeting Ultimate Accuracy: Face Recognition via Deep Embedding +<br/><b>Baidu Research Institute of Deep Learning</b></td><td>('2272123', 'Jingtuo Liu', 'jingtuo liu')</td><td></td></tr><tr><td>5b9d9f5a59c48bc8dd409a1bd5abf1d642463d65</td><td>Evolving Systems. manuscript No. +<br/>(will be inserted by the editor) +<br/>An evolving spatio-temporal approach for gender and age +<br/>group classification with Spiking Neural Networks +<br/>Received: date / Accepted: date +</td><td>('39323169', 'Fahad Bashir Alvi', 'fahad bashir alvi')<br/>('2662466', 'Russel Pears', 'russel pears')<br/>('1686744', 'Nikola Kasabov', 'nikola kasabov')</td><td></td></tr><tr><td>5bf70c1afdf4c16fd88687b4cf15580fd2f26102</td><td>Accepted in Pattern Recognition Letters +<br/>Pattern Recognition Letters +<br/>journal homepage: www.elsevier.com +<br/>Residual Codean Autoencoder for Facial Attribute Analysis +<br/>IIIT-Delhi, New Delhi, India +<br/>Article history: +<br/>Received 29 March 2017 +</td><td>('40639989', 'Akshay Sethi', 'akshay sethi')<br/>('2220719', 'Maneet Singh', 'maneet singh')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')</td><td></td></tr><tr><td>5b2cfee6e81ef36507ebf3c305e84e9e0473575a</td><td></td><td></td><td></td></tr><tr><td>5b01d4338734aefb16ee82c4c59763d3abc008e6</td><td>A Robust Face Recognition Algorithm Based on Kernel Regularized +<br/>Relevance-Weighted Discriminant Analysis +<br/> +<br/><b>Hunan Provincial Key Laboratory of Wind Generator and Its Control, Hunan Institute of Engineering, Xiangtan, China</b><br/><b>College of Electrical and Information Engineering</b><br/>or +<br/>In +<br/>I. INTRODUCTION +<br/>interface and security +<br/>recognition +<br/>their +<br/>from +<br/>this paper, we propose an effective +</td><td>('38296532', 'Di WU', 'di wu')<br/>('38296532', 'Di WU', 'di wu')</td><td> [e-mail: wudi6152007@163.com] +</td></tr><tr><td>5b721f86f4a394f05350641e639a9d6cb2046c45</td><td>A short version of this paper is accepted to ACM Asia Conference on Computer and Communications Security (ASIACCS) 2018 +<br/>Detection under Privileged Information (Full Paper)∗ +<br/><b>Pennsylvania State University</b><br/>Patrick McDaniel +<br/><b>Pennsylvania State University</b><br/>Vencore Labs +<br/><b>Pennsylvania State University</b><br/><b>Army Research Laboratory</b></td><td>('2950892', 'Z. Berkay Celik', 'z. berkay celik')<br/>('1804289', 'Rauf Izmailov', 'rauf izmailov')<br/>('1967156', 'Nicolas Papernot', 'nicolas papernot')<br/>('9541640', 'Ryan Sheatsley', 'ryan sheatsley')<br/>('30792942', 'Raquel Alvarez', 'raquel alvarez')<br/>('1703726', 'Ananthram Swami', 'ananthram swami')</td><td>zbc102@cse.psu.edu +<br/>mcdaniel@cse.psu.edu +<br/>rizmailov@appcomsci.com +<br/>{ngp5056,rms5643,rva5120}@cse.psu.edu +<br/>ananthram.swami.civ@mail.mil +</td></tr><tr><td>5b4b84ce3518c8a14f57f5f95a1d07fb60e58223</td><td>Diagnosing Error in Object Detectors +<br/>Department of Computer Science +<br/><b>University of Illinois at Urbana-Champaign</b></td><td>('2433269', 'Derek Hoiem', 'derek hoiem')<br/>('2918391', 'Yodsawalai Chodpathumwan', 'yodsawalai chodpathumwan')<br/>('2279233', 'Qieyun Dai', 'qieyun dai')</td><td></td></tr><tr><td>5b6ecbf5f1eecfe1a9074d31fe2fb030d75d9a79</td><td>Improving 3D Face Details based on Normal Map of Hetero-source Images +<br/><b>Tsinghua University</b><br/>Beijing, 100084, China +</td><td>('8100333', 'Chang Yang', 'chang yang')<br/>('1752427', 'Jiansheng Chen', 'jiansheng chen')<br/>('1949216', 'Nan Su', 'nan su')<br/>('7284296', 'Guangda Su', 'guangda su')</td><td>yangchang11@mails.tsinghua.edu.cn, jschenthu@tsinghua.edu.cn +<br/>v377026@sina.com, susu@tsinghua.edu.cn +</td></tr><tr><td>5b86c36e3eb59c347b81125d5dd57dd2a2c377a9</td><td>Name Identification of People in News Video +<br/>by Face Matching +<br/><b>Graduate School of Information Science, Nagoya University; Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan</b><br/>Japan Society for the Promotion of Science +<br/><b>Nagoya University</b><br/>School of Information Science, +<br/><b>Nagoya University</b></td><td>('1679187', 'Ichiro IDE', 'ichiro ide')<br/>('8027540', 'Takashi OGASAWARA', 'takashi ogasawara')<br/>('1685524', 'Tomokazu TAKAHASHI', 'tomokazu takahashi')<br/>('1725612', 'Hiroshi MURASE', 'hiroshi murase')</td><td>ide@is.nagoya-u.ac.jp, ide@nii.ac.jp +<br/>toga@murase.m.is.nagoya-u.ac.jp +<br/>ttakahashi@murase.m.is.nagoya-u.ac.jp +<br/>murase@is.nagoya-u.ac.jp Graduate +</td></tr><tr><td>5be3cc1650c918da1c38690812f74573e66b1d32</td><td>Relative Parts: Distinctive Parts for Learning Relative Attributes +<br/>Center for Visual Information Technology, IIIT Hyderabad, India - 500032 +</td><td>('32337248', 'Ramachandruni N. Sandeep', 'ramachandruni n. sandeep')<br/>('2169614', 'Yashaswi Verma', 'yashaswi verma')<br/>('1694502', 'C. V. Jawahar', 'c. v. jawahar')</td><td></td></tr><tr><td>5bc0a89f4f73523967050374ed34d7bc89e4d9e1</td><td><b></b><br/>On: 12 August 2015, At: 08:38 +<br/>Publisher: Routledge +<br/>Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: 5 +<br/>Howick Place, London, SW1P 1WG +<br/>Cognition and Emotion +<br/><b>Publication details, including instructions for authors and subscription</b><br/>information: +<br/>http://www.tandfonline.com/loi/pcem20 +<br/>The role of emotion transition for the +<br/>perception of social dominance and +<br/>affiliation +<br/><b>University of Haifa, Haifa, Israel</b><br/><b>b The Interdisciplinary Center for Research on Emotions, University of</b><br/>Haifa, Haifa, Israel +<br/><b>Humboldt-University, Berlin, Germany</b><br/>Published online: 11 Aug 2015. +<br/>Click for updates +<br/>perception of social dominance and affiliation, Cognition and Emotion, DOI: 10.1080/02699931.2015.1056107 +<br/>To link to this article: http://dx.doi.org/10.1080/02699931.2015.1056107 +<br/>PLEASE SCROLL DOWN FOR ARTICLE +<br/>Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) +<br/>contained in the publications on our platform. However, Taylor & Francis, our agents, and our +<br/>licensors make no representations or warranties whatsoever as to the accuracy, completeness, or +<br/>suitability for any purpose of the Content. Any opinions and views expressed in this publication +<br/>are the opinions and views of the authors, and are not the views of or endorsed by Taylor & +<br/>Francis. The accuracy of the Content should not be relied upon and should be independently +<br/>verified with primary sources of information. Taylor and Francis shall not be liable for any +<br/>losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities +<br/>whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or +<br/>arising out of the use of the Content. +<br/>This article may be used for research, teaching, and private study purposes. Any substantial +<br/>or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or +<br/>distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use +<br/>can be found at http://www.tandfonline.com/page/terms-and-conditions +</td><td>('3141618', 'Shlomo Hareli', 'shlomo hareli')<br/>('6885116', 'Shlomo David', 'shlomo david')<br/>('3141618', 'Shlomo Hareli', 'shlomo hareli')<br/>('6885116', 'Shlomo David', 'shlomo david')</td><td></td></tr><tr><td>5b6bed112e722c0629bcce778770d1b28e42fc96</td><td>FLOREA ET AL.:CANYOUREYESTELLMEHOWYOUTHINK? +<br/>Can Your Eyes Tell Me How You Think? A +<br/>Gaze Directed Estimation of the Mental +<br/>Activity +<br/>http://alpha.imag.pub.ro/common/staff/lflorea +<br/>http://alpha.imag.pub.ro/common/staff/cflorea +<br/>http://alpha.imag.pub.ro/common/staff/vertan +<br/>Image Processing and Analysis +<br/>Laboratory, LAPI +<br/><b>University Politehnica of Bucharest</b><br/>Bucharest, Romania +</td><td>('2143956', 'Laura Florea', 'laura florea')<br/>('2760434', 'Corneliu Florea', 'corneliu florea')<br/>('29723670', 'Ruxandra Vrânceanu', 'ruxandra vrânceanu')<br/>('2905899', 'Constantin Vertan', 'constantin vertan')</td><td>rvranceanu@alpha.imag.pub.ro +</td></tr><tr><td>5bde1718253ec28a753a892b0ba82d8e553b6bf3</td><td>JMLR: Workshop and Conference Proceedings 13: 79-94 +<br/>2nd Asian Conference on Machine Learning (ACML2010), Tokyo, Japan, Nov. 8{10, 2010. +<br/>Variational Relevance Vector Machine for Tabular Data +<br/>Dorodnicyn Computing Centre of the Russian Academy of Sciences +<br/>119333, Russia, Moscow, Vavilov str., 40 +<br/>Dmitry Vetrov +<br/><b>Lomonosov Moscow State University</b><br/>119992, Russia, Moscow, Leninskie Gory, 1, 2nd ed. bld., CMC department +<br/><b>The Blavatnik School of Computer Science, The Tel-Aviv University</b><br/><b>Schreiber Building, room 103, Tel Aviv University, P.O.B. 39040, Ramat Aviv, Tel Aviv</b><br/><b>Computer Science Division, The Open University of Israel</b><br/>108 Ravutski Str. P.O.B. 808, Raanana 43107, Israel +<br/>Editor: Masashi Sugiyama and Qiang Yang +</td><td>('3160602', 'Dmitry Kropotov', 'dmitry kropotov')<br/>('1776343', 'Lior Wolf', 'lior wolf')<br/>('1756099', 'Tal Hassner', 'tal hassner')</td><td>dmitry.kropotov@gmail.com +<br/>hassner@openu.ac.il +<br/>vetrovd@yandex.ru +<br/>wolf@cs.tau.ac.il +</td></tr><tr><td>5b0ebb8430a04d9259b321fc3c1cc1090b8e600e</td><td></td><td></td><td></td></tr><tr><td>37c8514df89337f34421dc27b86d0eb45b660a5e</td><td>Facial Landmark Tracking by Tree-based Deformable Part Model +<br/>Based Detector +<br/>Michal Uˇriˇc´aˇr, Vojtˇech Franc, and V´aclav Hlav´aˇc +<br/>Center for Machine Perception, Department of Cybernetics +<br/><b>Faculty of Electrical Engineering, Czech Technical University in Prague</b><br/>166 27 Prague 6, Technick´a 2, Czech Republic +</td><td></td><td>{uricamic, xfrancv, hlavac}@cmp.felk.cvut.cz +</td></tr><tr><td>371f40f6d32ece05cc879b6954db408b3d4edaf3</td><td>Mining Semantic Affordances of Visual Object Categories +<br/><b>Computer Science and Engineering, University of Michigan, Ann Arbor</b><br/> accelerate +<br/> intervie w +<br/> race +<br/> h urt +<br/> h u nt +<br/> fe e d +<br/> m a n ufacture +<br/> o p erate +<br/> drive +<br/> rid e +<br/> b o ard +<br/>bicycle +<br/>bird +<br/>boat +<br/>bottle +<br/>car +<br/>cat +<br/>cow +<br/>dining table +<br/>horse +<br/>person +<br/>train +<br/>tv +<br/>15 +<br/>10 +<br/>−5 +<br/>−10 +<br/>−15 +<br/>−20 +<br/>−15 +<br/>airplane +<br/>boat +<br/>car +<br/>train +<br/>bus +<br/>motorcycle +<br/>bicycle +<br/>chair +<br/>tv +<br/>couch +<br/>dining table +<br/>bottle +<br/>potted plant +<br/>person +<br/>horse +<br/>dog +<br/>cow +<br/>sheep +<br/>cat +<br/>bird +<br/>−10 +<br/>−5 +<br/>10 +<br/>15 +<br/>20 +<br/>25 +<br/>30 +<br/>(a) +<br/>(b) +<br/>Figure 1: (a) “Affordance matrix” encoding the plausibility of each action- +<br/>object pair. (b) 20 PASCAL VOC object classes in the semantic affordance +<br/>space. +<br/>Affordances are fundamental attributes of objects. Affordances reveal the +<br/>functionalities of objects and the possible actions that can be performed on +<br/>them. We can “hug” a dog, but not an ant. We can “turn on” a tv, but not a +<br/>bottle. Acquiring such knowledge is crucial for recognizing human activities +<br/>in visual data and for robots to interact with the world. The key question is: +<br/>given an object, can an action be performed on it? While this might seem +<br/>obvious to a human, there is no automated system that can readily answer +<br/>this question and there is no knowledge base that provides comprehensive +<br/>knowledge of object affordances. +<br/>In this paper, we introduce the problem of mining the knowledge of +<br/>semantic affordance: given an action and an object, determine whether the +<br/>action can be applied to the object. For example, the action of “carry” form a +<br/>valid combination with “bag”, but not with “skyscraper”. This is equivalent +<br/>to establishing connections between action concepts and object concepts, +<br/>or filling an “affordance matrix” encoding the plausibility of each action- +<br/>object pair (Fig. 1). The key scientific question is: “how can we collect +<br/>affordance knowledge”? We first introduce a new benchmark with crowd- +<br/>sourced ground truth affordances on 20 PASCAL VOC object classes and +<br/>957 action classes. We then study a variety of approaches including 1) text +<br/>mining, 2) visual mining, and 3) collaborative filtering. We quantitatively +<br/>evaluate all approaches using ground truth affordances collected through +<br/>crowdsourcing. +<br/>For our crowdsourcing study, we ask human annotators to label whether +<br/>an action-object pair is a valid combination. We use the 20 object categories +<br/>in PASCAL VOC [2]. We design experiments to obtain a list of action +<br/>categories that are both common and “visual”. Our list contains 957 ac- +<br/>tion categories extracted from the verb synsets on Wordnet [6] that has 1) a +<br/>member verb that frequently occurs in text corpora, and 2) high “visualness +<br/>score” determined by human labelers. Given the list of actions and objects, +<br/>we set up a crowdsourcing task on Amazon Mechanical Turk (AMT). We +<br/>ask crowd workers whether it is possible (for a human) to perform a given +<br/>action on a given object. For instance, +<br/>Is it possible to hunt (pursue for food or sport, as of wild animals) a car? +<br/>For every possible action-object pair formed by the 20 PASCAL VOC ob- +<br/>jects and the 957 visual verb synsets, we ask 5 workers to determine its +<br/>plausibility. This gives a total of 19K action-object questions and 96K an- +<br/>swers +<br/>What is the distribution of 20 PASCAL object classes in their affordance +<br/>space? We answer this by analyzing the human annotated affordances. Each +<br/>object has a 957 dimensional“affordance vector“, where each dimension +<br/>is the plausibility score with an action. We use PCA to project the affor- +<br/>dance vectors to a 2-dimensional space and plot the coordinates of the object +</td><td>('2820136', 'Yu-Wei Chao', 'yu-wei chao')<br/>('1718667', 'Zhan Wang', 'zhan wang')<br/>('1738516', 'Rada Mihalcea', 'rada mihalcea')<br/>('8342699', 'Jia Deng', 'jia deng')</td><td></td></tr><tr><td>374c7a2898180723f3f3980cbcb31c8e8eb5d7af</td><td>FACIAL EXPRESSION RECOGNITION IN VIDEOS USING A NOVEL MULTI-CLASS +<br/>SUPPORT VECTOR MACHINES VARIANT +<br/><b>yAristotle University of Thessaloniki</b><br/>Department of Informatics +<br/>Box 451, 54124 Thessaloniki, Greece +</td><td>('1754270', 'Irene Kotsia', 'irene kotsia')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td></td></tr><tr><td>37007af698b990a3ea8592b11d264b14d39c843f</td><td>DCMSVM: Distributed Parallel Training For Single-Machine Multiclass +<br/>Classifiers +<br/>Computer Science Department +<br/><b>Stony Brook University</b></td><td>('1682965', 'Xufeng Han', 'xufeng han')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')</td><td></td></tr><tr><td>374a0df2aa63b26737ee89b6c7df01e59b4d8531</td><td>Temporal Action Localization with Pyramid of Score Distribution Features +<br/><b>National University of Singapore, 2Shanghai Jiao Tong University</b></td><td>('1746449', 'Jun Yuan', 'jun yuan')<br/>('5796401', 'Bingbing Ni', 'bingbing ni')<br/>('1795291', 'Xiaokang Yang', 'xiaokang yang')</td><td>yuanjun@nus.edu.sg, nibingbing@sjtu.edu.cn, xkyang@sjtu.edu.cn, ashraf@nus.edu.sg +</td></tr><tr><td>378ae5ca649f023003021f5a63e393da3a4e47f0</td><td>Multi-Class Object Localization by Combining Local Contextual Interactions +<br/>Serge Belongie† +<br/>Gert Lanckriet‡ +<br/>†Computer Science and Engineering Department +<br/>‡Electrical and Computer Engineering Department +<br/><b>University of California, San Diego</b></td><td>('1954793', 'Carolina Galleguillos', 'carolina galleguillos')</td><td>{cgallegu,bmcfee,sjb}@cs.ucsd.edu, gert@ece.ucsd.edu +</td></tr><tr><td>37619564574856c6184005830deda4310d3ca580</td><td>A Deep Pyramid Deformable Part Model for Face Detection +<br/>Center for Automation Research +<br/><b>University of Maryland, College Park, MD</b></td><td>('26988560', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{rranjan1, pvishalm, rama}@umiacs.umd.edu +</td></tr><tr><td>372fb32569ced35eaf3740a29890bec2be1869fa</td><td>Running head: MU RHYTHM MODULATION BY CLASSIFICATION OF EMOTION 1 +<br/>Mu rhythm suppression is associated with the classification of emotion in faces +<br/><b>University of Otago, Dunedin, New Zealand</b><br/>Corresponding authors: +<br/>Phone: +64 (3) 479 5269; Fax: +64 (3) 479 8335 +<br/>Department of Psychology +<br/><b>University of Otago</b><br/>PO Box 56 +<br/>Dunedin, New Zealand +</td><td>('2187036', 'Elizabeth A. Franz', 'elizabeth a. franz')</td><td>Matthew Moore (matthew.moore@otago.ac.nz) & Liz Franz (lfranz@psy.otago.ac.nz) +</td></tr><tr><td>37ce1d3a6415d6fc1760964e2a04174c24208173</td><td>Pose-Invariant 3D Face Alignment +<br/>Department of Computer Science and Engineering +<br/><b>Michigan State University, East Lansing MI</b></td><td>('2357264', 'Amin Jourabloo', 'amin jourabloo')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td>{jourablo, liuxm}@msu.edu +</td></tr><tr><td>3765c26362ad1095dfe6744c6d52494ea106a42c</td><td></td><td></td><td></td></tr><tr><td>3727ac3d50e31a394b200029b2c350073c1b69e3</td><td></td><td></td><td></td></tr><tr><td>37f2e03c7cbec9ffc35eac51578e7e8fdfee3d4e</td><td>WACV +<br/>#394 +<br/>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>045 +<br/>046 +<br/>047 +<br/>048 +<br/>049 +<br/>050 +<br/>051 +<br/>052 +<br/>053 +<br/>WACV 2015 Submission #394. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +<br/>Co-operative Pedestrians Group Tracking in Crowded Scenes using an MST +<br/>Approach +<br/>Anonymous WACV submission +<br/>Paper ID 394 +</td><td></td><td></td></tr><tr><td>3795974e24296185d9b64454cde6f796ca235387</td><td>Finding your Lookalike: +<br/>Measuring Face Similarity Rather than Face Identity +<br/><b>Lafayette College</b><br/>Easton, PA +<br/>Andrew Gallagher +<br/>Google Research +<br/>Mountain View, CA +</td><td>('1803066', 'Amir Sadovnik', 'amir sadovnik')<br/>('50977255', 'Wassim Gharbi', 'wassim gharbi')<br/>('2197717', 'Thanh Vu', 'thanh vu')</td><td>{sadovnia,gharbiw,vut}@lafayette.edu +<br/>agallagher@google.com +</td></tr><tr><td>37278ffce3a0fe2c2bbf6232e805dd3f5267eba3</td><td>Can we still avoid automatic face detection? +<br/>Serge Belongie1,2 +<br/><b>Cornell University 2 Cornell Tech</b></td><td>('3035230', 'Michael J. Wilber', 'michael j. wilber')<br/>('1723945', 'Vitaly Shmatikov', 'vitaly shmatikov')</td><td></td></tr><tr><td>377a1be5113f38297716c4bb951ebef7a93f949a</td><td>Dear Faculty, IGERT Fellows, IGERT Associates and Students, +<br/>You are cordially invited to attend a Seminar presented by Albert Cruz. Please +<br/>plan to attend. +<br/> Albert Cruz +<br/>IGERT Fellow +<br/>Electrical Engineering +<br/> +<br/>Date: Friday, October 11, 2013 +<br/>Location: Bourns A265 +<br/>Time: 11:00am +<br/>Facial emotion recognition with anisotropic +<br/>inhibited gabor energy histograms +</td><td></td><td></td></tr><tr><td>377c6563f97e76a4dc836a0bd23d7673492b1aae</td><td></td><td></td><td></td></tr><tr><td>370e0d9b89518a6b317a9f54f18d5398895a7046</td><td>IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, VOL. X, NO. X, XXXXXXX 20XX +<br/>Cross-pollination of normalisation techniques +<br/>from speaker to face authentication +<br/>using Gaussian mixture models +<br/>and S´ebastien Marcel, Member, IEEE +</td><td>('1843477', 'Roy Wallace', 'roy wallace')</td><td></td></tr><tr><td>37ba12271d09d219dd1a8283bc0b4659faf3a6c6</td><td>Domain Transfer for Person Re-identification +<br/><b>Queen Mary University of London</b><br/>London, England +</td><td>('3264124', 'Ryan Layne', 'ryan layne')<br/>('1697755', 'Timothy M. Hospedales', 'timothy m. hospedales')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')</td><td>{rlayne, tmh, sgg}@eecs.qmul.ac.uk +</td></tr><tr><td>3773e5d195f796b0b7df1fca6e0d1466ad84b5e7</td><td><b>UNIVERSITY OF CALIFORNIA</b><br/>RIVERSIDE +<br/>Learning from Time Series in the Presence of Noise: Unsupervised and Semi-Supervised +<br/>Approaches +<br/>A Dissertation submitted in partial satisfaction +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>in +<br/>Computer Science +<br/>by +<br/>March 2008 +<br/>Dissertation Committee: +<br/>Dr. Eamonn Keogh, Chairperson +<br/>Dr. Vassilis Tsotras +</td><td>('40564016', 'Dragomir Dimitrov', 'dragomir dimitrov')<br/>('1736011', 'Stefano Lonardi', 'stefano lonardi')</td><td></td></tr><tr><td>37eb666b7eb225ffdafc6f318639bea7f0ba9a24</td><td>MSU Technical Report (2014): MSU-CSE-14-5 +<br/>Age, Gender and Race Estimation from +<br/>Unconstrained Face Images +</td><td>('34393045', 'Hu Han', 'hu han')<br/>('40437942', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>377f2b65e6a9300448bdccf678cde59449ecd337</td><td>Pushing the Limits of Unconstrained Face Detection: +<br/>a Challenge Dataset and Baseline Results +<br/>1Fujitsu Laboratories Ltd., Kanagawa, Japan +<br/><b>Johns Hopkins University, 3400 N. Charles St, Baltimore, MD 21218, USA</b><br/><b>Rutgers University, 94 Brett Rd, Piscataway Township, NJ 08854, USA</b></td><td>('41018586', 'Hajime Nada', 'hajime nada')<br/>('2577847', 'Vishwanath A. Sindagi', 'vishwanath a. sindagi')<br/>('46197381', 'He Zhang', 'he zhang')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')</td><td>nada.hajime@jp.fujitsu.com, vishwanath.sindagi@gmail.com, he.zhang92@rutgers.edu, +<br/>vpatel36@jhu.edu +</td></tr><tr><td>375435fb0da220a65ac9e82275a880e1b9f0a557</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI +<br/>From Pixels to Response Maps: Discriminative Image +<br/>Filtering for Face Alignment in the Wild +</td><td>('3183108', 'Akshay Asthana', 'akshay asthana')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1902288', 'Shiyang Cheng', 'shiyang cheng')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>370b6b83c7512419188f5373a962dd3175a56a9b</td><td>Face Alignment Refinement via Exploiting +<br/>Low-Rank property and Temporal Stability +<br/>Shuang LIU +<br/><b>Bournemouth University</b><br/><b>Bournemouth University</b><br/>Wenyu HU +<br/><b>Gannan Normal University</b><br/>Xiaosong YANG +<br/>Ruofeng TONG +<br/><b>Zhejiang University</b><br/>Jian J. ZHANG +<br/><b>Bournemouth University</b><br/><b>Bournemouth University</b><br/>face +<br/>and +<br/>alignment +</td><td>('48708691', 'Zhao Wang', 'zhao wang')</td><td>zwang@bournemouth.ac.uk +<br/>sliu@bournemouth.ac.uk +<br/>wenyu.huu@gmail.com +<br/>trf@zju.edu.cn +<br/>xyang@bournemouth.ac.uk +<br/>jzhang@bournemouth.ac.uk +</td></tr><tr><td>37b6d6577541ed991435eaf899a2f82fdd72c790</td><td>Vision-based Human Gender Recognition: A Survey +<br/>Universiti Tunku Abdul Rahman, Kuala Lumpur, Malaysia. +</td><td>('32877936', 'Choon Boon Ng', 'choon boon ng')<br/>('9201065', 'Yong Haur Tay', 'yong haur tay')</td><td>{ngcb,tayyh,goibm}@utar.edu.my +</td></tr><tr><td>372a8bf0ef757c08551d41e40cb7a485527b6cd7</td><td>Unsupervised Video Hashing by Exploiting +<br/>Spatio-Temporal Feature +<br/><b>Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong</b><br/><b>University, Shanghai, China</b></td><td>('46194894', 'Chao Ma', 'chao ma')<br/>('46964428', 'Yun Gu', 'yun gu')<br/>('46641573', 'Wei Liu', 'wei liu')<br/>('39264954', 'Jie Yang', 'jie yang')</td><td>{sjtu_machao,geron762,liuwei.1989,jieyang}@sjtu.edu.cn +</td></tr><tr><td>37ef18d71c1ca71c0a33fc625ef439391926bfbb</td><td>Extraction of Subject-Specific Facial Expression +<br/>Categories and Generation of Facial Expression +<br/>Feature Space using Self-Mapping +<br/>Department of Machine Intelligence and Systems Engineering, Faculty of Systems Science and Technology, +<br/><b>Akita Prefectural University, Yurihonjo, Japan</b><br/>Department of Computer Science and Engineering, Faculty of Engineering and Resource Science, +<br/><b>Akita University, Akita, Japan</b></td><td>('1932760', 'Masaki Ishii', 'masaki ishii')<br/>('2052920', 'Kazuhito Sato', 'kazuhito sato')<br/>('1738333', 'Hirokazu Madokoro', 'hirokazu madokoro')<br/>('21063785', 'Makoto Nishida', 'makoto nishida')</td><td>Email: {ishii, ksato, madokoro}@akita-pu.ac.jp +<br/>Email: nishida@ie.akita-u.ac.jp +</td></tr><tr><td>370b5757a5379b15e30d619e4d3fb9e8e13f3256</td><td>Labeled Faces in the Wild: A Database for Studying +<br/>Face Recognition in Unconstrained Environments +</td><td>('3219900', 'Gary B. Huang', 'gary b. huang')<br/>('1685538', 'Tamara Berg', 'tamara berg')<br/>('1714536', 'Erik Learned-Miller', 'erik learned-miller')</td><td></td></tr><tr><td>081189493ca339ca49b1913a12122af8bb431984</td><td>Photorealistic Facial Texture Inference Using Deep Neural Networks +<br/>Supplemental Material for +<br/>*Pinscreen +<br/><b>University of Southern California</b><br/><b>USC Institute for Creative Technologies</b><br/>Appendix I. Additional Results +<br/>Our main results in the paper demonstrate successful in- +<br/>ference of high-fidelity texture maps from unconstrained +<br/>images. The input images have mostly low resolutions, non- +<br/>frontal faces, and the subjects are often captured in chal- +<br/>lenging lighting conditions. We provide additional results +<br/>with pictures from the annotated faces-in-the-wild (AFW) +<br/>dataset [10] to further demonstrate how photorealistic pore- +<br/>level details can be synthesized using our deep learning ap- +<br/>proach. We visualize in Figure 9 the input, the intermedi- +<br/>ate low-frequency albedo map obtained using a linear PCA +<br/>model, and the synthesized high-frequency albedo texture +<br/>map. We also show several views of the final renderings us- +<br/>ing the Arnold renderer [13]. We refer to the accompanying +<br/>video for additional rotating views of the resulting textured +<br/>3D face models. +<br/>Figure 2: Even for largely downsized image resolutions, our +<br/>algorithm can produce fine-scale details while preserving +<br/>the person’s similarity. +<br/>We also evaluate the robustness of our inference frame- +<br/>work for downsized image resolutions in Figure 2. We crop +<br/>a diffuse lit face from a Light Stage capture [5]. The re- +<br/>sulting image has 435 × 652 pixels and we decrease its res- +<br/>olution to 108 × 162 pixels. In addition to complex skin +<br/>pigmentations, even the tiny mole on the lower left cheek is +<br/>properly reconstructed from the reduced input image using +<br/>our synthesis approach. +<br/>Figure 1: Comparison between different convolutional neu- +<br/>ral network architectures. +<br/>Evaluation. As Figure 1 indicates, other deep convolu- +<br/>tional neural networks can be used to extract mid-layer fea- +<br/>ture correlations to characterize multi-scale details, but it +<br/>seems that deeper architectures produce fewer artifacts and +<br/>higher quality textures. All three convolutional neural net- +<br/>works are pre-trained for classification tasks using images +<br/>from the ImageNet object recognition dataset [4]. The re- +<br/>sults of the 8 layer CaffeNet [2] show noticeable blocky ar- +<br/>tifacts in the synthesized textures and the ones from the 16 +<br/>layer VGG [12] are slightly noisy around boundaries, while +<br/>the 19 layer VGG network performs the best. +<br/>§- indicates equal contribution +<br/>Comparison. We provide in Figure 3 additional visual- +<br/>izations of our method when using the closest feature corre- +<br/>lation, unconstrained linear combinations, and convex com- +<br/>binations. We also compare against a PCA-based model +<br/>fitting [3] approach and the state-of-the-art visio-lization +<br/>framework [9]. We notice that only our proposed tech- +<br/>nique using convex combinations is effective in generating +<br/>mesoscopic-scale texture details. Both visio-lization and +<br/>the PCA-based model result in lower frequency textures and +<br/>less similar faces than the ground truth. Since our inference +<br/>also fills holes, we compare our synthesis technique with +<br/>a general inpainting solution for predicting unseen face re- +<br/>gions. We test with the widely used PatchMatch [1] tech- +<br/>nique as illustrated in Figure 4. Unsurprisingly, we observe +<br/>unwanted repeating structures and semantically wrong fill- +<br/>ings since this method is based on low-level vision cues. +<br/>CaffeNetVGG-16VGG-19albedo mapinputrendering input (magnified) </td><td>('2059597', 'Shunsuke Saito', 'shunsuke saito')<br/>('1792471', 'Lingyu Wei', 'lingyu wei')<br/>('1808579', 'Liwen Hu', 'liwen hu')<br/>('1897417', 'Koki Nagano', 'koki nagano')<br/>('40348249', 'Hao Li', 'hao li')</td><td></td></tr><tr><td>08ee541925e4f7f376538bc289503dd80399536f</td><td>Runtime Neural Pruning +<br/>Department of Automation +<br/><b>Tsinghua University</b><br/>Department of Automation +<br/><b>Tsinghua University</b><br/>Department of Automation +<br/><b>Tsinghua University</b><br/>Department of Automation +<br/><b>Tsinghua University</b></td><td>('2772283', 'Ji Lin', 'ji lin')<br/>('39358728', 'Yongming Rao', 'yongming rao')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('39491387', 'Jie Zhou', 'jie zhou')</td><td>lin-j14@mails.tsinghua.edu.cn +<br/>raoyongming95@gmail.com +<br/>lujiwen@tsinghua.edu.cn +<br/>jzhou@tsinghua.edu.cn +</td></tr><tr><td>08d2f655361335bdd6c1c901642981e650dff5ec</td><td>This is the published version: +<br/> Arandjelovic, Ognjen and Cipolla, R. 2006, Automatic cast listing in feature‐length films with +<br/>Anisotropic Manifold Space, in CVPR 2006 : Proceedings of the Computer Vision and Pattern +<br/>Recognition Conference 2006, IEEE, Piscataway, New Jersey, pp. 1513‐1520. +<br/> +<br/> http://hdl.handle.net/10536/DRO/DU:30058435 +<br/> Reproduced with the kind permission of the copyright owner. +<br/>Copyright : 2006, IEEE +<br/>Available from Deakin Research Online: +</td><td></td><td></td></tr><tr><td>08fbe3187f31b828a38811cc8dc7ca17933b91e9</td><td><b>MITSUBISHI ELECTRIC RESEARCH LABORATORIES</b><br/>http://www.merl.com +<br/>Statistical Computations on Grassmann and +<br/>Stiefel Manifolds for Image and Video-Based +<br/>Recognition +<br/>Turaga, P.; Veeraraghavan, A.; Srivastava, A.; Chellappa, R. +<br/>TR2011-084 April 2011 +</td><td></td><td></td></tr><tr><td>08ae100805d7406bf56226e9c3c218d3f9774d19</td><td>Gavrilescu and Vizireanu EURASIP Journal on Image and Video Processing (2017) 2017:59 +<br/>DOI 10.1186/s13640-017-0211-4 +<br/>EURASIP Journal on Image +<br/>and Video Processing +<br/>R ES EAR CH +<br/>Predicting the Sixteen Personality Factors +<br/>(16PF) of an individual by analyzing facial +<br/>features +<br/>Open Access +</td><td>('2132188', 'Mihai Gavrilescu', 'mihai gavrilescu')<br/>('1929703', 'Nicolae Vizireanu', 'nicolae vizireanu')</td><td></td></tr><tr><td>08c18b2f57c8e6a3bfe462e599a6e1ce03005876</td><td>A Least-Squares Framework +<br/>for Component Analysis +</td><td>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td></td></tr><tr><td>08f6ad0a3e75b715852f825d12b6f28883f5ca05</td><td>To appear in the 9th IEEE Int'l Conference on Automatic Face and Gesture Recognition, Santa Barbara, CA, March, 2011. +<br/>Face Recognition: Some Challenges in Forensics +<br/><b>Michigan State University</b><br/>East Lansing, MI, U.S.A +</td><td>('6680444', 'Anil K. Jain', 'anil k. jain')<br/>('1817623', 'Brendan Klare', 'brendan klare')<br/>('2222919', 'Unsang Park', 'unsang park')</td><td>{jain, klarebre, parkunsa}@cse.msu.edu +</td></tr><tr><td>08ff81f3f00f8f68b8abd910248b25a126a4dfa4</td><td>Papachristou, K., Tefas, A., & Pitas, I. (2014). Symmetric Subspace Learning +<br/>5697. DOI: 10.1109/TIP.2014.2367321 +<br/>Peer reviewed version +<br/>Link to published version (if available): +<br/>10.1109/TIP.2014.2367321 +<br/>Link to publication record in Explore Bristol Research +<br/>PDF-document +<br/>This is the author accepted manuscript (AAM). The final published version (version of record) is available online +<br/><b>via Institute of Electrical and Electronic Engineers at http://dx.doi.org/10.1109/TIP.2014.2367321. Please refer to</b><br/>any applicable terms of use of the publisher. +<br/><b>University of Bristol - Explore Bristol Research</b><br/>General rights +<br/>This document is made available in accordance with publisher policies. Please cite only the published +<br/>version using the reference above. Full terms of use are available: +<br/>http://www.bristol.ac.uk/pure/about/ebr-terms +<br/> </td><td></td><td></td></tr><tr><td>081a431107eb38812b74a8cd036ca5e97235b499</td><td></td><td></td><td></td></tr><tr><td>084bd02d171e36458f108f07265386f22b34a1ae</td><td>Face Alignment at 3000 FPS via Regressing Local Binary Features +<br/><b>University of Science and Technology of China</b><br/>Microsoft Research +</td><td>('2032273', 'Xudong Cao', 'xudong cao')<br/>('3080683', 'Shaoqing Ren', 'shaoqing ren')<br/>('1732264', 'Yichen Wei', 'yichen wei')<br/>('40055995', 'Jian Sun', 'jian sun')</td><td>sqren@mail.ustc.edu.cn +<br/>{xudongca,yichenw,jiansun}@microsoft.com +</td></tr><tr><td>081cb09791e7ff33c5d86fd39db00b2f29653fa8</td><td>Square Loss based Regularized LDA for Face Recognition Using Image Sets +<br/><b>Center for Information Science, Peking University, Beijing 100871, China</b><br/>2Philips Research, High Tech Campus 36, 5656 AE Eindhoven, The Netherlands +<br/><b>Queen Mary, University of London, London E1 4NS, UK</b></td><td>('37536447', 'Yanlin Geng', 'yanlin geng')<br/>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('1685266', 'Pengwei Hao', 'pengwei hao')</td><td>gengyanlin@cis.pku.edu.cn, caifeng.shan@philips.com, phao@dcs.qmul.ac.uk +</td></tr><tr><td>086131159999d79adf6b31c1e604b18809e70ba8</td><td>Deep Action Unit Classification using a Binned +<br/>Intensity Loss and Semantic Context Model +<br/>Department of Computing Sciences +<br/><b>Villanova University</b><br/>Villanova, Pennsylvania 19085 +<br/>Department of Computing Sciences +<br/><b>Villanova University</b><br/>Villanova, Pennsylvania 19085 +</td><td>('1904114', 'Edward Kim', 'edward kim')<br/>('35266734', 'Shruthika Vangala', 'shruthika vangala')</td><td>Email: edward.kim@villanova.edu +<br/>Email: svagal1@villanova.edu +</td></tr><tr><td>0831a511435fd7d21e0cceddb4a532c35700a622</td><td></td><td></td><td></td></tr><tr><td>0861f86fb65aa915fbfbe918b28aabf31ffba364</td><td>International Journal of Computer Trends and Technology (IJCTT) – volume 22 Number 3–April 2015 +<br/> An Efficient Facial Annotation with Machine Learning Approach +<br/>1A.Anusha,2R.Srinivas +<br/>1Final M.Tech Student, 2Associate Professor +<br/><b>Aditya Institute of Technology And Management, Tekkali, Srikakulam, Andhra Pradesh</b></td><td></td><td></td></tr><tr><td>089513ca240c6d672c79a46fa94a92cde28bd567</td><td>RNN Fisher Vectors for Action Recognition and Image Annotation +<br/><b>The Blavatnik School of Computer Science, Tel Aviv University, Israel</b><br/>2IBM Research, Haifa, Israel +</td><td>('3004979', 'Guy Lev', 'guy lev')<br/>('2251827', 'Gil Sadeh', 'gil sadeh')<br/>('2205955', 'Benjamin Klein', 'benjamin klein')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td></td></tr><tr><td>089b5e8eb549723020b908e8eb19479ba39812f5</td><td>A Cross Benchmark Assessment of A Deep Convolutional Neural +<br/>Network for Face Recognition +<br/><b>National Institute of Standards and Technology</b><br/>Gaithersburg, MD 20899 USA +</td><td>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')</td><td></td></tr><tr><td>080c204edff49bf85b335d3d416c5e734a861151</td><td>CLAD: A Complex and Long Activities +<br/>Dataset with Rich Crowdsourced +<br/>Annotations +<br/>Journal Title +<br/>XX(X):1–6 +<br/>c(cid:13)The Author(s) 2016 +<br/>Reprints and permission: +<br/>sagepub.co.uk/journalsPermissions.nav +<br/>DOI: 10.1177/ToBeAssigned +<br/>www.sagepub.com/ +</td><td>('3280554', 'Jawad Tayyub', 'jawad tayyub')<br/>('2762811', 'Majd Hawasly', 'majd hawasly')<br/>('1967104', 'David C. Hogg', 'david c. hogg')<br/>('1703235', 'Anthony G. Cohn', 'anthony g. cohn')</td><td></td></tr><tr><td>08f4832507259ded9700de81f5fd462caf0d5be8</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 118 – No.14, May 2015 +<br/>Geometric Approach for Human Emotion +<br/>Recognition using Facial Expression +<br/>S. S. Bavkar +<br/>Assistant Professor +<br/>J. S. Rangole +<br/>Assistant Professor +<br/>V. U. Deshmukh +<br/>Assistant Professor +</td><td></td><td></td></tr><tr><td>08a1fc55d03e4a73cad447e5c9ec79a6630f3e2d</td><td>BERG, BELHUMEUR: TOM-VS-PETE CLASSIFIERS AND IDENTITY-PRESERVING ALIGNMENT +<br/>Tom-vs-Pete Classifiers and Identity-Preserving +<br/>Alignment for Face Verification +<br/><b>Columbia University</b><br/>New York, NY +</td><td>('1778562', 'Thomas Berg', 'thomas berg')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')</td><td>tberg@cs.columbia.edu +<br/>belhumeur@cs.columbia.edu +</td></tr><tr><td>08d40ee6e1c0060d3b706b6b627e03d4b123377a</td><td>Human Action Localization +<br/>with Sparse Spatial Supervision +</td><td>('2492127', 'Philippe Weinzaepfel', 'philippe weinzaepfel')<br/>('3269403', 'Xavier Martin', 'xavier martin')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>08c1f8f0e69c0e2692a2d51040ef6364fb263a40</td><td></td><td></td><td></td></tr><tr><td>088aabe3da627432fdccf5077969e3f6402f0a80</td><td>Under review as a conference paper at ICLR 2018 +<br/>CLASSIFIER-TO-GENERATOR ATTACK: ESTIMATION +<br/>OF TRAINING DATA DISTRIBUTION FROM CLASSIFIER +<br/>Anonymous authors +<br/>Paper under double-blind review +</td><td></td><td></td></tr><tr><td>087002ab569e35432cdeb8e63b2c94f1abc53ea9</td><td>Looking at People +<br/>CVPRW 2015 +<br/>Spatio-temporal Analysis of RGB-D-T Facial +<br/>Images for Multimodal Pain Level +<br/>Recognition +<br/><b>Visual Analysis of People Lab, Aalborg University, Denmark</b><br/>Computer Vision Center, UAB, Barcelona, Spain +<br/><b>Aalborg University, Denmark</b></td><td>('37541412', 'Ramin Irani', 'ramin irani')<br/>('1803459', 'Kamal Nasrollahi', 'kamal nasrollahi')<br/>('3321700', 'Ciprian A. Corneanu', 'ciprian a. corneanu')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('40526933', 'Tanja L. Pedersen', 'tanja l. pedersen')<br/>('31627926', 'Maria-Louise Klitgaard', 'maria-louise klitgaard')<br/>('35675498', 'Laura Petrini', 'laura petrini')</td><td></td></tr><tr><td>08903bf161a1e8dec29250a752ce9e2a508a711c</td><td>Joint Dimensionality Reduction and Metric Learning: A Geometric Take +</td><td>('2862871', 'Mathieu Salzmann', 'mathieu salzmann')</td><td></td></tr><tr><td>08cb294a08365e36dd7ed4167b1fd04f847651a9</td><td>EXAMINING VISIBLE ARTICULATORY FEATURES IN CLEAR AND +<br/>CONVERSATIONAL SPEECH +<br/><b>Medical Image Analysis Lab, School of Computing Science, Simon Fraser University, Canada</b><br/><b>Language and Brain Lab, Simon Fraser University, Canada</b><br/><b>KU Phonetics and Psycholinguistics Lab, University of Kansas</b></td><td>('2664514', 'Lisa Tang', 'lisa tang')<br/>('26839551', 'Beverly Hannah', 'beverly hannah')<br/>('3200950', 'Allard Jongman', 'allard jongman')<br/>('1723309', 'Yue Wang', 'yue wang')<br/>('3049056', 'Ghassan Hamarneh', 'ghassan hamarneh')</td><td> lisat@sfu.ca, beverlyw@sfu.ca, jongman@ku.edu, sereno@ku.edu, yuew@sfu.ca, hamarneh@sfu.ca +</td></tr><tr><td>081286ede247c5789081502a700b378b6223f94b</td><td>ORIGINAL RESEARCH +<br/>published: 06 February 2018 +<br/>doi: 10.3389/fpsyg.2018.00052 +<br/>Neural Correlates of Facial Mimicry: +<br/>Simultaneous Measurements of EMG +<br/>and BOLD Responses during +<br/>Perception of Dynamic Compared to +<br/>Static Facial Expressions +<br/><b>Institute of Cognitive and Behavioural Neuroscience, SWPS University of Social</b><br/>Sciences and Humanities, Warsaw, Poland, 2 Laboratory of Psychophysiology, Department of Neurophysiology, Nencki +<br/><b>Institute of Experimental Biology of Polish Academy of Sciences, Warsaw, Poland</b><br/>Facial mimicry (FM) is an automatic response to imitate the facial expressions of others. +<br/>However, neural correlates of the phenomenon are as yet not well established. We +<br/>investigated this issue using simultaneously recorded EMG and BOLD signals during +<br/>perception of dynamic and static emotional facial expressions of happiness and anger. +<br/>During display presentations, BOLD signals and zygomaticus major (ZM), corrugator +<br/>supercilii (CS) and orbicularis oculi (OO) EMG responses were recorded simultaneously +<br/>from 46 healthy individuals. Subjects reacted spontaneously to happy facial expressions +<br/>with increased EMG activity in ZM and OO muscles and decreased CS activity, which +<br/>was interpreted as FM. Facial muscle responses correlated with BOLD activity in regions +<br/>associated with motor simulation of facial expressions [i.e., inferior frontal gyrus, a +<br/>classical Mirror Neuron System (MNS)]. Further, we also found correlations for regions +<br/>associated with emotional processing (i.e., insula, part of the extended MNS). It is +<br/>concluded that FM involves both motor and emotional brain structures, especially during +<br/>perception of natural emotional expressions. +<br/>Keywords: facial mimicry, EMG, fMRI, mirror neuron system, emotional expressions, dynamic, happiness, anger +<br/>INTRODUCTION +<br/>Facial mimicry (FM) is an unconscious and unintentional automatic response to the facial +<br/>expressions of others. Numerous studies have shown that observing the emotional states of others +<br/>leads to congruent facial muscle activity. For example, observing angry facial expressions can result +<br/>in enhanced activity in the viewer’s muscle responsible for frowning (CS), while viewing happy +<br/>images leads to Increased activity in the facial muscle involved in smiling (ZM), and decreased +<br/>activity of the CS (Hess et al., 1998; Dimberg and Petterson, 2000). However, it has recently been +<br/>suggested that FM may not be an exclusive automatic reaction but rather a multifactorial response +<br/>dependent on properties such as stimulus modality (e.g., static or dynamic) or interpersonal +<br/>characteristics (e.g., emotional contagion susceptibility) (for review see Seibt et al., 2015). +<br/>There are two main psychological approaches trying to explain the mechanisms of +<br/>FM. One of +<br/>these is the perception-behavior link model which assumes perception and +<br/>execution of a specific action show a certain overlap (Chartrand and Bargh, 1999). +<br/>Edited by: +<br/>Alessio Avenanti, +<br/>Università di Bologna, Italy +<br/>Reviewed by: +<br/>Sebastian Korb, +<br/><b>University of Vienna, Austria</b><br/>Frank A. Russo, +<br/><b>Ryerson University, Canada</b><br/>*Correspondence: +<br/>Łukasz ˙Zurawski +<br/>Specialty section: +<br/>This article was submitted to +<br/>Emotion Science, +<br/>a section of the journal +<br/>Frontiers in Psychology +<br/>Received: 20 July 2017 +<br/>Accepted: 12 January 2018 +<br/>Published: 06 February 2018 +<br/>Citation: +<br/>Rymarczyk K, ˙Zurawski Ł, +<br/>Jankowiak-Siuda K and Szatkowska I +<br/>(2018) Neural Correlates of Facial +<br/>Mimicry: Simultaneous Measurements +<br/>of EMG and BOLD Responses during +<br/>Perception of Dynamic Compared to +<br/>Static Facial Expressions. +<br/>Front. Psychol. 9:52. +<br/>doi: 10.3389/fpsyg.2018.00052 +<br/>Frontiers in Psychology | www.frontiersin.org +<br/>February 2018 | Volume 9 | Article 52 +</td><td>('4079953', 'Krystyna Rymarczyk', 'krystyna rymarczyk')<br/>('4022705', 'Kamila Jankowiak-Siuda', 'kamila jankowiak-siuda')<br/>('4970569', 'Iwona Szatkowska', 'iwona szatkowska')<br/>('4079953', 'Krystyna Rymarczyk', 'krystyna rymarczyk')</td><td>krymarczyk@swps.edu.pl +<br/>l.zurawski@nencki.gov.pl +</td></tr><tr><td>08e995c080a566fe59884a527b72e13844b6f176</td><td>A New KSVM + KFD Model for Improved +<br/>Classification and Face Recognition +<br/><b>School of Computer Science, University of Windsor, Windsor, ON, Canada N9B 3P</b></td><td>('1687000', 'Riadh Ksantini', 'riadh ksantini')</td><td>Email: {ksantini, boufama, imran}@uwindsor.ca +</td></tr><tr><td>08e24f9df3d55364290d626b23f3d42b4772efb6</td><td>ENHANCING FACIAL EXPRESSION CLASSIFICATION BY INFORMATION +<br/>FUSION +<br/>I. Buciu1, Z. Hammal 2, A. Caplier2, N. Nikolaidis 1, and I. Pitas 1 +<br/><b></b><br/>GR-54124, Thessaloniki, Box 451, Greece +<br/>2 Laboratoire des Images et des Signaux / Institut National Polytechnique de Grenoble +<br/>web: http://www.aiia.csd.auth.gr +<br/>38031 Grenoble, France +<br/>web: http://www.lis.inpg.fr +</td><td></td><td>phone: + 30(2310)99.6361, fax: + 30(2310)99.8453, email: {nelu,nikolaid,pitas}@aiia.csd.auth.gr +<br/>phone: + 33(0476)574363, fax: + 33(0476)57 47 90, email: alice.caplier@inpg.fr +</td></tr><tr><td>085ceda1c65caf11762b3452f87660703f914782</td><td>Large-pose Face Alignment via CNN-based Dense 3D Model Fitting +<br/>Department of Computer Science and Engineering +<br/><b>Michigan State University, East Lansing MI</b></td><td>('2357264', 'Amin Jourabloo', 'amin jourabloo')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td>{jourablo, liuxm}@msu.edu +</td></tr><tr><td>0830c9b9f207007d5e07f5269ffba003235e4eff</td><td></td><td></td><td></td></tr><tr><td>08d55271589f989d90a7edce3345f78f2468a7e0</td><td>Quality Aware Network for Set to Set Recognition +<br/>SenseTime Group Limited +<br/>SenseTime Group Limited +<br/><b>University of Sydney</b></td><td>('1715752', 'Yu Liu', 'yu liu')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('3001348', 'Wanli Ouyang', 'wanli ouyang')</td><td>liuyuisanai@gmail.com +<br/>yanjunjie@sensetime.com +<br/>wanli.ouyang@gmail.com +</td></tr><tr><td>081fb4e97d6bb357506d1b125153111b673cc128</td><td></td><td></td><td></td></tr><tr><td>08a98822739bb8e6b1388c266938e10eaa01d903</td><td>SensorSift: Balancing Sensor Data Privacy and Utility in +<br/>Automated Face Understanding +<br/><b>University of Washington</b><br/>**Microsoft Research, Redmond WA +</td><td>('3299424', 'Miro Enev', 'miro enev')<br/>('33481800', 'Jaeyeon Jung', 'jaeyeon jung')<br/>('1766509', 'Liefeng Bo', 'liefeng bo')<br/>('1728501', 'Xiaofeng Ren', 'xiaofeng ren')<br/>('1769675', 'Tadayoshi Kohno', 'tadayoshi kohno')</td><td></td></tr><tr><td>084bebc5c98872e9307cd8e7f571d39ef9c1b81e</td><td>A Discriminative Feature Learning Approach +<br/>for Deep Face Recognition +<br/>1 Shenzhen Key Lab of Computer Vision and Pattern Recognition, +<br/><b>Shenzhen Institutes of Advanced Technology, CAS, Shenzhen, China</b><br/><b>The Chinese University of Hong Kong, Sha Tin, Hong Kong</b></td><td>('2512949', 'Yandong Wen', 'yandong wen')<br/>('3393556', 'Kaipeng Zhang', 'kaipeng zhang')<br/>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td>yandongw@andrew.cmu.edu, {kp.zhang,zhifeng.li,yu.qiao}@siat.ac.cn +</td></tr><tr><td>0857281a3b6a5faba1405e2c11f4e17191d3824d</td><td>Chude-Olisah et al. EURASIP Journal on Advances in Signal Processing 2014, 2014:102 +<br/>http://asp.eurasipjournals.com/content/2014/1/102 +<br/>R ES EAR CH +<br/>Face recognition via edge-based Gabor feature +<br/>representation for plastic surgery-altered images +<br/>Open Access +</td><td>('2529988', 'Ghazali Sulong', 'ghazali sulong')</td><td></td></tr><tr><td>08f1e9e14775757298afd9039f46ec56e80677f9</td><td>Attentional Push: Augmenting Salience with +<br/>Shared Attention Modeling +<br/>Centre for Intelligent Machines, Department of Electrical and Computer Engineering, +<br/><b>McGill University</b><br/>Montreal, Quebec, Canada +</td><td>('38111179', 'Siavash Gorji', 'siavash gorji')<br/>('1713608', 'James J. Clark', 'james j. clark')</td><td>siagorji@cim.mcgill.ca clark@cim.mcgill.ca +</td></tr><tr><td>08d41d2f68a2bf0091dc373573ca379de9b16385</td><td>Recursive Chaining of Reversible Image-to-Image +<br/>Translators for Face Aging +<br/><b>Aalto University, Espoo, Finland</b><br/>1 GenMind Ltd, Finland +<br/>{ari.heljakka,arno.solin,juho.kannala}aalto.fi +</td><td>('2622083', 'Ari Heljakka', 'ari heljakka')<br/>('1768402', 'Arno Solin', 'arno solin')<br/>('1776374', 'Juho Kannala', 'juho kannala')</td><td></td></tr><tr><td>08f6745bc6c1b0fb68953ea61054bdcdde6d2fc7</td><td>Understanding Kin Relationships in a Photo +</td><td>('2025056', 'Ming Shao', 'ming shao')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')<br/>('1708679', 'Yun Fu', 'yun fu')</td><td></td></tr><tr><td>082ad50ac59fc694ba4369d0f9b87430553b11db</td><td></td><td></td><td></td></tr><tr><td>6d0fe30444c6f4e4db3ad8b02fb2c87e2b33c58d</td><td>Robust Deep Appearance Models +<br/><b>Concordia University, Montreal, Quebec, Canada</b><br/>2 CyLab Biometrics Center and the Department of Electrical and Computer Engineering, +<br/><b>Carnegie Mellon University, Pittsburgh, PA, USA</b><br/>face images. In this approach, +</td><td>('2687827', 'Kha Gia Quach', 'kha gia quach')<br/>('1876581', 'Chi Nhan Duong', 'chi nhan duong')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('1699922', 'Tien D. Bui', 'tien d. bui')</td><td>Email: {k q, c duon, bui}@encs.concordia.ca +<br/>Email: kluu@andrew.cmu.edu +</td></tr><tr><td>6dbdb07ce2991db0f64c785ad31196dfd4dae721</td><td>Seeing Small Faces from Robust Anchor’s Perspective +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Avenue, Pittsburgh, PA 15213, USA +</td><td>('47894545', 'Chenchen Zhu', 'chenchen zhu')<br/>('1794486', 'Marios Savvides', 'marios savvides')<br/>('47599820', 'Ran Tao', 'ran tao')<br/>('1769788', 'Khoa Luu', 'khoa luu')</td><td>{chenchez, rant, kluu, marioss}@andrew.cmu.edu +</td></tr><tr><td>6dd052df6b0e89d394192f7f2af4a3e3b8f89875</td><td>International Journal of Engineering and Advanced Technology (IJEAT) +<br/>ISSN: 2249 – 8958, Volume-2, Issue-4, April 2013 +<br/>A literature survey on Facial Expression +<br/>Recognition using Global Features +<br/> +</td><td>('9318822', 'Mahesh M. Goyani', 'mahesh m. goyani')</td><td></td></tr><tr><td>6d7a32f594d46f4087b71e2a2bb66a4b25da5e30</td><td>Towards Person Authentication by Fusing Visual and Thermal Face +<br/>Biometrics +<br/>1 Department of Engineering +<br/><b>University of Cambridge</b><br/>Cambridge, CB2 1TQ +<br/>UK +<br/>2 Delphi Corporation +<br/>Delphi Electronics and Safety +<br/>Kokomo, IN 46901-9005 +<br/>USA +</td><td>('2214319', 'Riad Hammoud', 'riad hammoud')<br/>('1745672', 'Roberto Cipolla', 'roberto cipolla')</td><td>{oa214,cipolla}@eng.cam.ac.uk +<br/>riad.hammoud@delphi.com +</td></tr><tr><td>6dd5dbb6735846b214be72983e323726ef77c7a9</td><td>Josai Mathematical Monographs +<br/>vol. 7 (2014), pp. 25-40 +<br/>A Survey on Newer Prospective +<br/>Biometric Authentication Modalities +</td><td>('3322335', 'Narishige Abe', 'narishige abe')<br/>('2395689', 'Takashi Shinzaki', 'takashi shinzaki')</td><td></td></tr><tr><td>6d10beb027fd7213dd4bccf2427e223662e20b7d</td><td></td><td></td><td>ResearchArticleUserAdaptiveandContext-AwareSmartHomeUsingPervasiveandSemanticTechnologiesAggelikiVlachostergiou,1GeorgiosStratogiannis,1GeorgeCaridakis,1,2GeorgeSiolas,1andPhivosMylonas1,31IntelligentSystemsContentandInteractionLaboratory,NationalTechnicalUniversityofAthens,IroonPolytexneiou9,15780Zografou,Greece2DepartmentofCulturalTechnologyandCommunication,UniversityoftheAegean,Mytilene,Lesvos,Greece3DepartmentofInformatics,IonianUniversity,Corfu,GreeceCorrespondenceshouldbeaddressedtoAggelikiVlachostergiou;aggelikivl@image.ntua.grReceived17January2016;Revised6July2016;Accepted17July2016AcademicEditor:JohnN.SahalosCopyright©2016AggelikiVlachostergiouetal.ThisisanopenaccessarticledistributedundertheCreativeCommonsAttributionLicense,whichpermitsunrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalworkisproperlycited.UbiquitousComputingismovingtheinteractionawayfromthehuman-computerparadigmandtowardsthecreationofsmartenvironmentsthatusersandthings,fromtheIoTperspective,interactwith.Usermodelingandadaptationisconsistentlypresenthavingthehumanuserasaconstantbutpervasiveinteractionintroducestheneedforcontextincorporationtowardscontext-awaresmartenvironments.Thecurrentarticlediscussesbothaspectsoftheusermodelingandadaptationaswellascontextawarenessandincorporationintothesmarthomedomain.Usersaremodeledasfuzzypersonasandthesemodelsaresemanticallyrelated.Contextinformationiscollectedviasensorsandcorrespondstovariousaspectsofthepervasiveinteractionsuchastemperatureandhumidity,butalsosmartcitysensorsandservices.Thiscontextinformationenhancesthesmarthomeenvironmentviatheincorporationofuserdefinedhomerules.SemanticWebtechnologiessupporttheknowledgerepresentationofthisecosystemwhiletheoverallarchitecturehasbeenexperimentallyverifiedusinginputfromtheSmartSantandersmartcityandapplyingittotheSandSsmarthomewithinFIREandFIWAREframeworks.1.IntroductionAlthoughintheirinitialdefinitionanddevelopmentstagespervasivecomputingpracticesdidnotnecessarilyrelyontheuseoftheInternet,currenttrendsshowtheemergenceofmanyconvergencepointswiththeInternetofThings(IoT)paradigm,whereobjectsareidentifiedasInternetresourcesandcanbeaccessedandutilizedassuch.Inthesametime,theHuman-ComputerInteraction(HCI)paradigminthedomainofdomoticshaswideneditsscopeconsiderably,placingthehumaninhabitantinapervasiveenvironmentandinacontinuousinteractionwithsmartobjectsandappliances.SmarthomesthatadditionallyadheretotheIoTapproachconsiderthatthisdatacontinuouslyproducedbyappliances,sensors,andhumanscanbeprocessedandassessedcollaboratively,remotely,andevensocially.Inthepresentpaper,wetrytobuildanewknowledgerepresentationframeworkwherewefirstplacethehumanuserinthecenterofthisinteraction.Wethenproposetobreakdownthemultitudeofpossibleuserbehaviorstoafewprototypicalusermodelsandthentoresynthesizethemusingfuzzyreasoning.Then,wediscusstheubiquityofcontextinformationinrelationtotheuserandthedifficultyofproposingauniversalformalizationframeworkfortheopenworld.Weshowthat,byrestrictinguser-relatedcontexttothesmarthomeenvironment,wecanreliablydefinesimplerulestructuresthatcorrelatespecificsensorinputdataanduseractionsthatcanbeusedtotriggerarbitrarysmarthomeevents.ThisrationaleisthenevolvedtoahigherlevelsemanticrepresentationofthedomoticecosysteminwhichcomplexhomerulescanbedefinedusingSemanticWebtechnologies.Itisthusobservedthatasmarthomeusingpervasiveandsemantictechnologiesinwhichthehumanuserisinthecenteroftheinteractionhastobeadaptive(itsbehaviorcanchangeinresponsetoaperson’sactionsandenvironment)andpersonalized(itsbehaviorcanbetailoredtotheuser’sHindawi Publishing CorporationJournal of Electrical and Computer EngineeringVolume 2016, Article ID 4789803, 20 pageshttp://dx.doi.org/10.1155/2016/4789803 </td></tr><tr><td>6d2ca1ddacccc8c865112bd1fbf8b931c2ee8e75</td><td>ROC Speak: Semi-Automated Personalized Feedback on +<br/>Nonverbal Behavior from Recorded Videos +<br/><b>Rochester Human-Computer Interaction (ROC HCI), University of Rochester, NY</b><br/>Figure 1. An overview of our system. Once the user finishes recording, the video is analyzed on the server for objective feedback +<br/>and sent to Mechanical Turk for subjective feedback. The objective feedback is then combined with subjective feedback that is +<br/>scored based on helpfulness, under which the sentiment is then classified. +</td><td>('1825866', 'Michelle Fung', 'michelle fung')<br/>('2961433', 'Yina Jin', 'yina jin')<br/>('2171034', 'RuJie Zhao', 'rujie zhao')</td><td>{mfung, yjin18, rzhao2, mehoque}@cs.rochester.edu +</td></tr><tr><td>6dddf1440617bf7acda40d4d75c7fb4bf9517dbb</td><td>JOURNAL OF LATEX CLASS FILES, VOL. XX, NO. X, MM YY +<br/>Beyond Counting: Comparisons of Density Maps for Crowd +<br/>Analysis Tasks - Counting, Detection, and Tracking +</td><td>('41201301', 'Di Kang', 'di kang')<br/>('1730232', 'Zheng Ma', 'zheng ma')<br/>('3651407', 'Antoni B. Chan', 'antoni b. chan')</td><td></td></tr><tr><td>6de18708218988b0558f6c2f27050bb4659155e4</td><td></td><td></td><td></td></tr><tr><td>6d97e69bbba5d1f5c353f9a514d62aff63bc0fb1</td><td>Semi-Supervised Learning for Facial Expression +<br/>Recognition +<br/>1HP Labs, Palo Alto, CA, USA +<br/><b>Faculty of Science, University of Amsterdam, The Netherlands</b><br/>3Escola Polit´ecnica, Universidade de S˜ao Paulo, Brazil +<br/><b>Beckman Institute, University of Illinois at Urbana-Champaign, IL, USA</b></td><td>('1774778', 'Ira Cohen', 'ira cohen')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>Ira.cohen@hp.com +<br/>nicu@science.uva.nl +<br/>fgcozman@usp.br +<br/>huang@ifp.uiuc.edu +</td></tr><tr><td>6d91da37627c05150cb40cac323ca12a91965759</td><td></td><td></td><td></td></tr><tr><td>6d07e176c754ac42773690d4b4919a39df85d7ec</td><td>Face Attribute Prediction Using Off-The-Shelf Deep +<br/>Learning Networks +<br/>Computer Science and Communication +<br/><b>KTH Royal Institute of Technology</b><br/>100 44 Stockholm, Sweden +</td><td>('50262049', 'Yang Zhong', 'yang zhong')<br/>('1736906', 'Josephine Sullivan', 'josephine sullivan')<br/>('40565290', 'Haibo Li', 'haibo li')</td><td>{yzhong, sullivan, haiboli}@kth.se +</td></tr><tr><td>6dd2a0f9ca8a5fee12edec1485c0699770b4cfdf</td><td>Webly-supervised Video Recognition by Mutually +<br/>Voting for Relevant Web Images and Web Video Frames +<br/><b>IIIS, Tsinghua University</b><br/>2Google Research +<br/>3Amazon +<br/><b>CRCV, University of Central Florida</b></td><td>('2551285', 'Chuang Gan', 'chuang gan')<br/>('1726241', 'Chen Sun', 'chen sun')<br/>('2055900', 'Lixin Duan', 'lixin duan')<br/>('40206014', 'Boqing Gong', 'boqing gong')</td><td></td></tr><tr><td>6d4b5444c45880517213a2fdcdb6f17064b3fa91</td><td>Journal of Information Engineering and Applications +<br/>ISSN 2224-5782 (print) ISSN 2225-0506 (online) +<br/>Vol 2, No.3, 2012 +<br/>www.iiste.org +<br/>Harvesting Image Databases from The Web +<br/><b>G.H.Raisoni College of Engg. and Mgmt., Pune, India</b><br/><b>G.H.Raisoni College of Engg. and Mgmt., Pune, India</b><br/><b>G.H.Raisoni College of Engg. and Mgmt., Pune, India</b></td><td>('2671016', 'Snehal M. Gaikwad', 'snehal m. gaikwad')<br/>('40050646', 'Snehal S. Pathare', 'snehal s. pathare')</td><td>*gaikwad.snehal99@gmail.com +<br/>*snehalpathare4@gmail.com +<br/>*truptijachak311991@gmail.com +</td></tr><tr><td>6d8c9a1759e7204eacb4eeb06567ad0ef4229f93</td><td>Face Alignment Robust to Pose, Expressions and +<br/>Occlusions +</td><td>('2232940', 'Vishnu Naresh Boddeti', 'vishnu naresh boddeti')<br/>('1767616', 'Myung-Cheol Roh', 'myung-cheol roh')<br/>('2526145', 'Jongju Shin', 'jongju shin')<br/>('3149566', 'Takaharu Oguri', 'takaharu oguri')<br/>('1733113', 'Takeo Kanade', 'takeo kanade')</td><td></td></tr><tr><td>6dc1f94b852538d572e4919238ddb10e2ee449a4</td><td>Objects as context for detecting their semantic parts +<br/><b>University of Edinburgh</b></td><td>('20758701', 'Abel Gonzalez-Garcia', 'abel gonzalez-garcia')<br/>('1996209', 'Davide Modolo', 'davide modolo')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')</td><td>a.gonzalez-garcia@sms.ed.ac.uk +<br/>davide.modolo@gmail.com +<br/>vferrari@staffmail.ed.ac.uk +</td></tr><tr><td>6d4e3616d0b27957c4107ae877dc0dd4504b69ab</td><td>Shuffle and Learn: Unsupervised Learning using +<br/>Temporal Order Verification +<br/><b>The Robotics Institute, Carnegie Mellon University</b><br/>2 Facebook AI Research +</td><td>('1806773', 'Ishan Misra', 'ishan misra')<br/>('1699161', 'C. Lawrence Zitnick', 'c. lawrence zitnick')<br/>('1709305', 'Martial Hebert', 'martial hebert')</td><td>{imisra, hebert}@cs.cmu.edu, zitnick@fb.com +</td></tr><tr><td>6d5125c9407c7762620eeea7570af1a8ee7d76f3</td><td>Video Frame Interpolation by Plug-and-Play +<br/>Deep Locally Linear Embedding +<br/><b>Yonsei University</b></td><td>('1886286', 'Anh-Duc Nguyen', 'anh-duc nguyen')<br/>('47902684', 'Woojae Kim', 'woojae kim')<br/>('2078790', 'Jongyoo Kim', 'jongyoo kim')<br/>('39200200', 'Sanghoon Lee', 'sanghoon lee')</td><td></td></tr><tr><td>6d8e3f3a83514381f890ab7cd2a1f1c5be597b69</td><td><b>University of Massachusetts - Amherst</b><br/>Doctoral Dissertations 2014-current +<br/>Dissertations and Theses +<br/>2014 +<br/>Improving Text Recognition in Images of Natural +<br/>Scenes +<br/>Jacqueline Feild +<br/>Follow this and additional works at: http://scholarworks.umass.edu/dissertations_2 +<br/>Recommended Citation +<br/>Feild, Jacqueline, "Improving Text Recognition in Images of Natural Scenes" (2014). Doctoral Dissertations 2014-current. Paper 37. +</td><td></td><td>ScholarWorks@UMass Amherst +<br/>University of Massachusetts - Amherst, jacqueline.feild@gmail.com +<br/>This Open Access Dissertation is brought to you for free and open access by the Dissertations and Theses at ScholarWorks@UMass Amherst. It has +<br/>been accepted for inclusion in Doctoral Dissertations 2014-current by an authorized administrator of ScholarWorks@UMass Amherst. For more +<br/>information, please contact scholarworks@library.umass.edu. +</td></tr><tr><td>6d8eef8f8d6cd8436c55018e6ca5c5907b31ac19</td><td>Understanding Representations and Reducing +<br/>their Redundancy in Deep Networks +<br/>Thesis submitted to the Faculty of +<br/><b>Virginia Polytechnic Institute and State University</b><br/>in partial fulfillment of the requirements for the degree of +<br/>Master of Science +<br/>in +<br/>Computer Science and Applications +<br/>Chair +<br/>Co-chair +<br/>February 18, 2016 +<br/>Blacksburg, Virginia +<br/>Keywords: Computer Vision, Machine Learning, Object Recognition, Overfitting +</td><td>('3358085', 'Micheal Cogswell', 'micheal cogswell')<br/>('40486307', 'Bert Huang', 'bert huang')<br/>('1746610', 'Dhruv Batra', 'dhruv batra')<br/>('38013066', 'B. Aditya Prakash', 'b. aditya prakash')</td><td>Copyright @ 2016 Michael Cogswell +</td></tr><tr><td>6d618657fa5a584d805b562302fe1090957194ba</td><td>Full Paper +<br/>NNGT Int. J. of Artificial Intelligence , Vol. 1, July 2014 +<br/>Human Facial Expression Recognition based +<br/>on Principal Component Analysis and +<br/>Artificial Neural Network +<br/>Laboratory of Automatic and Signals Annaba (LASA) , Department of electronics, Faculty of Engineering, +<br/>Zermi.Narima, Ramdani.M, Saaidia.M +<br/><b>Badji-Mokhtar University, P.O.Box 12, Annaba-23000, Algeria</b></td><td></td><td>E-Mail : naili.narima@gmail.com, messaoud.ramdani@univ-annaba.org +</td></tr><tr><td>6d66c98009018ac1512047e6bdfb525c35683b16</td><td>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 25, NO. 9, SEPTEMBER 2003 +<br/>1063 +<br/>Face Recognition Based on +<br/>Fitting a 3D Morphable Model +</td><td>('2880906', 'Volker Blanz', 'volker blanz')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td></td></tr><tr><td>016cbf0878db5c40566c1fbc237686fbad666a33</td><td></td><td></td><td></td></tr><tr><td>016800413ebd1a87730a5cf828e197f43a08f4b3</td><td>Learning Attributes Equals +<br/>Multi-Source Domain Generalization +<br/><b>IIIS, Tsinghua University</b><br/><b>University of Iowa</b><br/>CRCV, U. of Central Florida +</td><td>('2551285', 'Chuang Gan', 'chuang gan')<br/>('40381920', 'Tianbao Yang', 'tianbao yang')<br/>('40206014', 'Boqing Gong', 'boqing gong')</td><td>ganchuang1990@gmail.com +<br/>tianbao-yang@uiowa.edu +<br/>bgong@crcv.ucf.edu +</td></tr><tr><td>0172867f4c712b33168d9da79c6d3859b198ed4c</td><td>Technique for Face Recognition +<br/><b>Faculty of Engineering, Ain Shams University, Cairo, Egypt</b><br/>Expression and Illumination Invariant Preprocessing +</td><td>('1726416', 'A. Abbas', 'a. abbas')<br/>('9159923', 'S. Abdel-Hay', 's. abdel-hay')</td><td></td></tr><tr><td>0145dc4505041bf39efa70ea6d95cf392cfe7f19</td><td>Human Action Segmentation with Hierarchical Supervoxel Consistency +<br/><b>University of Michigan</b><br/>Detailed analysis of human action, such as classification, detection and lo- +<br/>calization has received increasing attention from the community; datasets +<br/>like J-HMDB [1] have made it plausible to conduct studies analyzing the +<br/>impact that such deeper information has on the greater action understanding +<br/>problem. However, detailed automatic segmentation of human action has +<br/>comparatively been unexplored. In this paper, we introduce a hierarchical +<br/>MRF model to automatically segment human action boundaries in videos +<br/>“in-the-wild” (see Fig. 1). +<br/>We first propose a human motion saliency representation which incor- +<br/>porates two parts: foreground motion and human appearance information. +<br/>For foreground motion estimation, we propose a new motion saliency fea- +<br/>ture by using long-term trajectories to build a camera motion model, and +<br/>then measure the motion saliency via the deviation from the camera model. +<br/>For human appearance information, we use a DPM person detector trained +<br/>on PASCAL VOC 2007 and construct a saliency map by averaging the nor- +<br/>malized detection score of all the scale and all components. +<br/>Then, to segment the human action, we start by applying hierarchical +<br/>graph-based video segmentation [2] to form a hierarchy of supervoxels. On +<br/>this hierarchy, we define an MRF model, using our novel human motion +<br/>saliency as the unary term. We consider the joint information of temporal +<br/>connections in the direction of optical flow and human appearance-aware +<br/>spatial neighbors as pairwise potential. We design an innovative high-order +<br/>potential between different supervoxels on different levels of the hierar- +<br/>chy to alleviate leaks and sustain better semantic information. Given the +<br/>graph structure G = (X ,E) induced by the supervoxel hierarchy (E is the +<br/>set of edges in the graph hiearchy). We introduce an energy function over +<br/>G = (X ,E) that enforces hierarchical supervoxel consistency through higher +<br/>order potentials derived from supervoxel V. +<br/>E(Y ) = ∑ +<br/>i∈X +<br/>Φi(yi) + ∑ +<br/>(i, j)∈E +<br/>Φi, j(yi,y j) + ∑ +<br/>v∈V +<br/>Φv(yv) +<br/>(1) +<br/>where Φi(yi) denotes unary potential for a supervoxel with index i, Φi, j(yi,y j) +<br/>denotes pairwise potential between two supervoxels with edge, and Φv(yv) +<br/>denotes high order potential of supervoxels between two layers. Unary po- +<br/>tential: We encode the motion saliency and human saliency feature into +<br/>supervoxels to get the unary potential components: +<br/>Φi(yi) = γMMi(yi) + γPPi(yi) + γSSi(yi) +<br/>(2) +<br/>where γM, γP and γS are weights for the unary terms. Mi(yi) reflects the +<br/>motion evidence, Pi(yi) and Si(yi) reflect the human evidence. Pairwise +<br/>potential: we constrain the edge space with only two types of neighbors: +<br/>temporal supervoxel neighbors and human-aware spatial neighbors, so we +<br/>define the pairwise potential as: +<br/>Φi, j(yi,y j) = γIIi, j(yi,y j) + γKKi, j(yi,y j)) +<br/>(3) +<br/>where γI and γK are pairwise potential weights. Ii, j(yi,y j) is the cost be- +<br/>tween supervoxel i and supervoxel j with human detection constraints, which +<br/>ensures the smoothness spatially. Note that i and j could be determined as +<br/>neighbors without pixel-level connection. Ki, j(yi,y j) is the virtual dissim- +<br/>ilarity which ensures the smoothness temporally. Higher order potential: +<br/>We define the hierarchical supervoxel label consistency potential. We utilize +<br/>the connection between different supervoxel hierarchical levels. In practice, +<br/>we adopt the Robust Pn model [3] to define the potentials, +<br/>if N(yv) (cid:54) Q +<br/>otherwise +<br/>(cid:26) N(yv) 1 +<br/>Φv(yv) = +<br/>Q γmax(v) +<br/>γmax(v) +</td><td>('8553015', 'Jiasen Lu', 'jiasen lu')<br/>('1856629', 'Ran Xu', 'ran xu')</td><td></td></tr><tr><td>01bef320b83ac4405b3fc5b1cff788c124109fb9</td><td>de Lausanne +<br/>RLC D1 740, CH-1015 +<br/>Lausanne +<br/>de Lausanne +<br/>RLC D1 740, CH-1015 +<br/>Lausanne +<br/>de Lausanne +<br/>RLC D1 740, CH-1015 +<br/>Lausanne +<br/>Translating Head Motion into Attention - Towards +<br/>Processing of Student’s Body-Language +<br/>CHILI Laboratory +<br/>Łukasz Kidzi´nski +<br/>CHILI Laboratory +<br/>CHILI Laboratory +<br/>École polytechnique fédérale +<br/>École polytechnique fédérale +<br/>École polytechnique fédérale +</td><td>('1850245', 'Mirko Raca', 'mirko raca')<br/>('1799133', 'Pierre Dillenbourg', 'pierre dillenbourg')</td><td>mirko.raca@epfl.ch +<br/>lukasz.kidzinski@epfl.ch +<br/>pierre.dillenbourg@epfl.ch +</td></tr><tr><td>01c9dc5c677aaa980f92c4680229db482d5860db</td><td>Temporal Action Detection using a Statistical Language Model +<br/><b>University of Bonn, Germany</b></td><td>('32774629', 'Alexander Richard', 'alexander richard')<br/>('2946643', 'Juergen Gall', 'juergen gall')</td><td>{richard,gall}@iai.uni-bonn.de +</td></tr><tr><td>013909077ad843eb6df7a3e8e290cfd5575999d2</td><td>A semi-automatic methodology for facial landmark annotation +<br/><b>Imperial College London, UK</b><br/><b>School of Computer Science, University of Lincoln, U.K</b><br/><b>EEMCS, University of Twente, The Netherlands</b></td><td>('3320415', 'Christos Sagonas', 'christos sagonas')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{c.sagonas, gt204, s.zafeiriou, m.pantic}@imperial.ac.uk +</td></tr><tr><td>01c7a778cde86ad1b89909ea809d55230e569390</td><td>A Supervised Low-rank Method for Learning Invariant Subspaces +<br/><b>West Virginia University</b><br/>Morgantown, WV 26508 +</td><td>('1803400', 'Farzad Siyahjani', 'farzad siyahjani')<br/>('3360490', 'Ranya Almohsen', 'ranya almohsen')<br/>('36911226', 'Sinan Sabri', 'sinan sabri')<br/>('1736352', 'Gianfranco Doretto', 'gianfranco doretto')</td><td>{fsiyahja, ralmohse, sisabri, gidoretto}@mix.wvu.edu +</td></tr><tr><td>01c8d7a3460422412fba04e7ee14c4f6cdff9ad7</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 4, No. 7, 2013 +<br/>Rule Based System for Recognizing Emotions Using +<br/>Multimodal Approach +<br/>Information System +<br/>SBM, SVKM’s NMIMS +<br/>Mumbai, India +<br/> +</td><td>('9575671', 'Preeti Khanna', 'preeti khanna')</td><td></td></tr><tr><td>0115f260069e2e501850a14845feb400142e2443</td><td>An On-Line Handwriting Recognizer +<br/>with Fisher Matching, Hypotheses +<br/>Propagation Network and Context +<br/>Constraint Models +<br/>By +<br/>A dissertation submitted in partial fulfillment of +<br/>the requirements for the degree of +<br/>Doctor of Philosophy +<br/>Department of Computer Science +<br/><b>New York University</b><br/>May 2001 +<br/>_____________________ +<br/>Davi Geiger +</td><td>('2034318', 'Jong Oh', 'jong oh')</td><td></td></tr><tr><td>01cc8a712e67384f9ef9f30580b7415bfd71e980</td><td>14750 • The Journal of Neuroscience, November 3, 2010 • 30(44):14750 –14758 +<br/>Behavioral/Systems/Cognitive +<br/>Failing to Ignore: Paradoxical Neural Effects of Perceptual +<br/>Load on Early Attentional Selection in Normal Aging +<br/><b>2Program in Neuroscience, and 3Rotman Research Institute, University of Toronto, Toronto, Ontario M5S 3G3, Canada</b><br/>We examined visual selective attention under perceptual load—simultaneous presentation of task-relevant and -irrelevant informa- +<br/>tion—in healthy young and older adult human participants to determine whether age differences are observable at early stages of +<br/>selection in the visual cortices. Participants viewed 50/50 superimposed face/place images and judged whether the faces were male or +<br/>female, rendering places perceptible but task-irrelevant. Each stimulus was repeated, allowing us to index dynamic stimulus-driven +<br/>competition from places. Consistent with intact early selection in young adults, we observed no adaptation to unattended places in +<br/>parahippocampal place area (PPA) and significant adaptation to attended faces in fusiform face area (FFA). Older adults, however, +<br/>exhibited both PPA adaptation to places and weak FFA adaptation to faces. We also probed participants’ associative recognition for +<br/>face-place pairs post-task. Older adults with better place recognition memory scores were found to exhibit both the largest magnitudes of +<br/>PPA adaptation and the smallest magnitudes of FFA adaptation on the attention task. In a control study, we removed the competing +<br/>perceptual information to decrease perceptual load. These data revealed that the initial age-related impairments in selective attention +<br/>were not due to a general decline in visual cortical selectivity; both young and older adults exhibited robust FFA adaptation and neither +<br/>group exhibited PPA adaptation to repeated faces. Accordingly, distracting information does not merely interfere with attended input in +<br/>older adults, but is co-encoded along with the contents of attended input, to the extent that this information can subsequently be +<br/>recovered from recognition memory. +<br/>Introduction +<br/>Age-related changes in selective attention have traditionally been +<br/>examined using manipulations of executive attention, e.g., the +<br/>capacity to selectively maintain targets and suppress distractors +<br/>in working memory (WM) (Hasher and Zacks, 1988; Gazzaley et +<br/>al., 2005, 2008; Healey et al., 2008). Under cognitive load from +<br/>WM, older adults appear more susceptible to interference from +<br/>distracting stimuli compared with young controls. +<br/>At the neural level, executive attention appears to reconcile +<br/>interference from unattended distractors at stages of processing +<br/>after encoding in the perceptual cortices, i.e., late selection, and +<br/>relies on prefrontal control mechanisms (de Fockert et al., 2001; +<br/>Gehring and Knight, 2002). Experimental tasks that manipulate +<br/>executive attention, such as distractor exclusion (de Fockert et al., +<br/>2001; Yi et al., 2004) and attentional blink (Luck et al., 1996; +<br/>Marois et al., 2000) have routinely demonstrated late selection of +<br/>unattended information. +<br/>However, the focus of aging research on executive attention +<br/>and distractor interference has left several questions unexplored. +<br/>Executive attention appears to be dissociable from the type of +<br/>perceptual attention used for reconciling distractor competition +<br/>Received May 26, 2010; revised Aug. 28, 2010; accepted Sept. 11, 2010. +<br/><b>This work was supported by Grant MOP102637 from the Canadian Institutes of Health Research to E.D.R. and the</b><br/>Vanier National Science and Engineering Research Council Scholarship to T.W.S. We also thank Adam K. Anderson +<br/>and Daniel H. Lee for helpful editorial input on the manuscript. +<br/>DOI:10.1523/JNEUROSCI.2687-10.2010 +<br/>Copyright © 2010 the authors +<br/>0270-6474/10/3014750-09$15.00/0 +<br/>within the visual field, which is thought to be embedded in pos- +<br/>terior cortical subsystems (Treisman, 1969; Desimone and Dun- +<br/>can, 1995; Lavie et al., 2004). For instance, Yi et al. (2004) +<br/>observed that under perceptual load but not WM load, unat- +<br/>tended distractors were suppressed at stages of visual processing +<br/>before extrastriate encoding. These finding indicate that percep- +<br/>tual attention relies on a distinct early selection mechanism. +<br/>In the present study, we therefore explored with functional +<br/>magnetic resonance imaging (fMRI) whether perceptual at- +<br/>tention is also susceptible to an age-related impairment. We +<br/>hypothesized that under perceptually demanding conditions, +<br/>when task-relevant and -irrelevant stimuli were simultaneously +<br/>presented in the visual field, early competitive perceptual inter- +<br/>actions from task-irrelevant sensory information would be suc- +<br/>cessfully filtered in younger adults before encoding (Lavie, 1995; +<br/>Yi et al., 2004). By contrast, if older adults do exhibit impaired +<br/>perceptual attention, then age-differences in distractor encoding +<br/>should be observable in extrastriate cortex sensitive to the +<br/>unattended stream of input. We were also interested in eluci- +<br/>dating the precise neural fate of this unattended information +<br/>in older adults. Specifically, do distractors merely interfere +<br/>with attended input, or are distractors co-encoded along with +<br/>the content of attended input to the extent that they can sub- +<br/>sequently be recognized? +<br/>To interrogate these hypotheses, we acquired fMRI while a +<br/>group of healthy young (mean age ⫽ 22.2 years) and older (mean +<br/>age ⫽ 77.4 years) adults viewed 50/50 threshold superimposed +<br/>face and place images (O’Craven et al., 1999; Yi et al., 2006) (Fig. +<br/>1a). Participants decided whether faces were male or female, ren- +</td><td>('4258285', 'Eve De Rosa', 'eve de rosa')<br/>('4258285', 'Eve De Rosa', 'eve de rosa')</td><td>George Street, Toronto, ON M5S 3G3, Canada. E-mail: taylor@aclab.ca or derosa@psych.utoronto.ca. +</td></tr><tr><td>01e12be4097fa8c94cabeef0ad61498c8e7762f2</td><td></td><td></td><td></td></tr><tr><td>0163d847307fae508d8f40ad193ee542c1e051b4</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 6, NO. 1, JANUARY 2007 +<br/>Classemes and Other Classifier-based +<br/>Features for Efficient Object Categorization +<br/>- Supplementary material - +<br/>1 LOW-LEVEL FEATURES +<br/>We extract the SIFT [1] features for our descriptor +<br/>according to the following pipeline. We first convert +<br/>each image to gray-scale, then we normalize the con- +<br/>trast by forcing the 0.01% of lightest and darkest pixels +<br/>to be mapped to white and black respectively, and +<br/>linearly rescaling the values in between. All images +<br/>exceeding 786,432 pixels of resolution are downsized +<br/>to this maximum value while keeping the aspect ratio. +<br/>The 128-dimensional SIFT descriptors are computed +<br/>from the interest points returned by a DoG detec- +<br/>tor [2]. We finally compute a Bag-Of-Word histogram +<br/>of these descriptors, using a K-means vocabulary of +<br/>500 words. +<br/>2 CLASSEMES +<br/>The LSCOM categories were developed specifically +<br/>for multimedia annotation and retrieval, and have +<br/>been used in the TRECVID video retrieval series. +<br/>We took the LSCOM CYC ontology dated 2006-06-30, +<br/>which contains 2832 unique categories. We removed +</td><td>('34338883', 'Alessandro Bergamo', 'alessandro bergamo')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')</td><td></td></tr><tr><td>01dc1e03f39901e212bdf291209b7686266aeb13</td><td>Actionness Estimation Using Hybrid Fully Convolutional Networks +<br/><b>Shenzhen key lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b><br/><b>The Chinese University of Hong Kong, Hong Kong</b><br/>3Computer Vision Lab, ETH Zurich, Switzerland +</td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('33427555', 'Yu Qiao', 'yu qiao')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>016f49a54b79ec787e701cc8c7d0280273f9b1ef</td><td>SELF ORGANIZING MAPS FOR REDUCING THE NUMBER OF CLUSTERS BY ONE ON +<br/>SIMPLEX SUBSPACES +<br/><b>Aristotle University of Thessaloniki</b><br/>Box 451, Thessaloniki 541 24, Greece +</td><td>('1736143', 'Constantine Kotropoulos', 'constantine kotropoulos')<br/>('1762248', 'Vassiliki Moschou', 'vassiliki moschou')</td><td>E-mail: {costas, vmoshou}@aiia.csd.auth.gr +</td></tr><tr><td>01c4cf9c7c08f0ad3f386d88725da564f3c54679</td><td>Interpretability Beyond Feature Attribution: +<br/>Quantitative Testing with Concept Activation Vectors (TCAV) +</td><td>('3351164', 'Been Kim', 'been kim')<br/>('2217654', 'Rory Sayres', 'rory sayres')</td><td></td></tr><tr><td>017ce398e1eb9f2eed82d0b22fb1c21d3bcf9637</td><td>FACE RECOGNITION WITH HARMONIC DE-LIGHTING +<br/>2ICT-ISVISION Joint R&D Laboratory for Face Recognition, CAS, Beijing, China, 100080 +<br/>1Graduate School, CAS, Beijing, China, 100080 +<br/>Emails: {lyqing, sgshan, wgao}jdl.ac.cn +</td><td>('2343895', 'Laiyun Qing', 'laiyun qing')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('40049005', 'Wen Gao', 'wen gao')</td><td></td></tr><tr><td>014e3d0fa5248e6f4634dc237e2398160294edce</td><td>Int J Comput Vis manuscript No. +<br/>(will be inserted by the editor) +<br/>What does 2D geometric information really tell us about +<br/>3D face shape? +<br/>Received: date / Accepted: date +</td><td>('39180407', 'Anil Bas', 'anil bas')</td><td></td></tr><tr><td>01125e3c68edb420b8d884ff53fb38d9fbe4f2b8</td><td>Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric +<br/>CNN Regression +<br/><b>The University of Nottingham, UK</b><br/><b>Kingston University, UK</b><br/><b>Figure 1: A few results from our VRN - Guided method, on a full range of pose, including large expressions</b></td><td>('34596685', 'Aaron S. Jackson', 'aaron s. jackson')<br/>('3458121', 'Adrian Bulat', 'adrian bulat')<br/>('1689047', 'Vasileios Argyriou', 'vasileios argyriou')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>1{aaron.jackson, adrian.bulat, yorgos.tzimiropoulos}@nottingham.ac.uk +<br/>2 vasileios.argyriou@kingston.ac.uk +</td></tr><tr><td>01c09acf0c046296643de4c8b55a9330e9c8a419</td><td>MANIFOLD LEARNING USING EUCLIDEAN +<br/>-NEAREST NEIGHBOR GRAPHS +<br/>Department of Electrical Engineering and Computer Science +<br/><b>University of Michigan, Ann Arbor, MI</b></td><td>('1759109', 'Jose A. Costa', 'jose a. costa')<br/>('35806564', 'Alfred O. Hero', 'alfred o. hero')</td><td>Email: jcosta@umich.edu, hero@eecs.umich.edu +</td></tr><tr><td>01d23cbac762b0e46251f5dbde08f49f2d13b9f8</td><td>Combining Face Verification Experts +<br/>+Telecommunication laboratory, Universit´e catholique de Louvain, B-1348 Belgium +<br/>⁄Center for Vision, Speech and Signal Processing, +<br/><b>University of Surrey, Guildford, Surrey GU2 7XH, UK</b></td><td>('34964585', 'Jacek Czyz', 'jacek czyz')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('1698047', 'Luc Vandendorpe', 'luc vandendorpe')</td><td>czyz@tele.ucl.ac.be +</td></tr><tr><td>014143aa16604ec3f334c1407ceaa496d2ed726e</td><td>Large-Scale Manifold Learning +<br/><b>Courant Institute</b><br/>New York, NY +<br/>Google Research +<br/>New York, NY +<br/>Henry Rowley +<br/>Google Research +<br/>Mountain View, CA +</td><td>('8395559', 'Ameet Talwalkar', 'ameet talwalkar')<br/>('2794322', 'Sanjiv Kumar', 'sanjiv kumar')</td><td>ameet@cs.nyu.edu +<br/>sanjivk@google.com +<br/>har@google.com +</td></tr><tr><td>011e6146995d5d63c852bd776f782cc6f6e11b7b</td><td>Fast Training of Triplet-based Deep Binary Embedding Networks +<br/><b>The University of Adelaide; and Australian Centre for Robotic Vision</b></td><td>('3194022', 'Bohan Zhuang', 'bohan zhuang')<br/>('2604251', 'Guosheng Lin', 'guosheng lin')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')</td><td></td></tr><tr><td>0182d090478be67241392df90212d6cd0fb659e6</td><td>Discovering Localized Attributes for Fine-grained Recognition +<br/><b>Indiana University</b><br/>Bloomington, IN +<br/>TTI-Chicago +<br/>Chicago, IL +<br/>David Crandall +<br/><b>Indiana University</b><br/>Bloomington, IN +<br/><b>University of Texas</b><br/>Austin, TX +</td><td>('2481141', 'Kun Duan', 'kun duan')<br/>('1713589', 'Devi Parikh', 'devi parikh')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>kduan@indiana.edu +<br/>dparikh@ttic.edu +<br/>djcran@indiana.edu +<br/>grauman@cs.utexas.edu +</td></tr><tr><td>016a8ed8f6ba49bc669dbd44de4ff31a79963078</td><td>1Graduate School, CAS, Beijing, 100039, China, +<br/>2ICT-ISVISION Joint R&D Laboratory for Face Recognition, CAS, Beijing, China, 100080 +<br/><b>Harbin Institute of Technology, Harbin, China</b><br/>FACE RELIGHTING FOR FACE RECOGNTION UNDER GENERIC ILLUMINATION +</td><td>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>01beab8f8293a30cf48f52caea6ca0fb721c8489</td><td></td><td></td><td></td></tr><tr><td>0178929595f505ef7655272cc2c339d7ed0b9507</td><td></td><td></td><td></td></tr><tr><td>0181fec8e42d82bfb03dc8b82381bb329de00631</td><td>Discriminative Subspace Clustering +<br/><b>CVL, Link oping University, Link oping, Sweden</b><br/><b>VSI Lab, Goethe University, Frankfurt, Germany</b></td><td>('1797883', 'Vasileios Zografos', 'vasileios zografos')<br/>('34824636', 'Rudolf Mester', 'rudolf mester')</td><td></td></tr><tr><td>01b4b32c5ef945426b0396d32d2a12c69c282e29</td><td></td><td></td><td></td></tr><tr><td>0113b302a49de15a1d41ca4750191979ad756d2f</td><td>1424403677/06/$20.00 ©2006 IEEE +<br/>537 +<br/>ICME 2006 +</td><td></td><td></td></tr><tr><td>019e471667c72b5b3728b4a9ba9fe301a7426fb2</td><td>Cross-Age Face Verification by Coordinating with Cross-Face Age Verification +<br/><b>Temple University, Philadelphia, USA</b></td><td>('38909760', 'Liang Du', 'liang du')<br/>('1805398', 'Haibin Ling', 'haibin ling')</td><td>{liang.du, hbling}@temple.edu +</td></tr><tr><td>0601416ade6707c689b44a5bb67dab58d5c27814</td><td>Feature Selection in Face Recognition: A Sparse +<br/>Representation Perspective +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2007-99 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2007/EECS-2007-99.html +<br/>August 14, 2007 +</td><td>('2223304', 'Allan Y. Yang', 'allan y. yang')<br/>('1738310', 'John Wright', 'john wright')<br/>('7777470', 'Yi Ma', 'yi ma')<br/>('1717598', 'S. Shankar Sastry', 's. shankar sastry')</td><td></td></tr><tr><td>064b797aa1da2000640e437cacb97256444dee82</td><td>Coarse-to-fine Face Alignment with Multi-Scale Local Patch Regression +<br/>Megvii Inc. +<br/>Megvii Inc. +<br/>Megvii Inc. +</td><td>('18036051', 'Zhiao Huang', 'zhiao huang')<br/>('1848243', 'Erjin Zhou', 'erjin zhou')<br/>('2695115', 'Zhimin Cao', 'zhimin cao')</td><td>hza@megvii.com +<br/>zej@megvii.com +<br/>czm@megvii.com +</td></tr><tr><td>06f146dfcde10915d6284981b6b84b85da75acd4</td><td>Scalable Face Image Retrieval using +<br/>Attribute-Enhanced Sparse Codewords +</td><td>('33970300', 'Bor-Chun Chen', 'bor-chun chen')<br/>('35081710', 'Yan-Ying Chen', 'yan-ying chen')<br/>('1692811', 'Yin-Hsi Kuo', 'yin-hsi kuo')<br/>('1716836', 'Winston H. Hsu', 'winston h. hsu')</td><td></td></tr><tr><td>067126ce1f1a205f98e33db7a3b77b7aec7fb45a</td><td>On Improving Dissimilarity-Based Classifications Using +<br/>a Statistical Similarity Measure(cid:2) +<br/><b>Myongji University</b><br/>Yongin, 449-728 South Korea +<br/>2 Faculty of Electrical Engineering, Mathematics and Computer Science, +<br/><b>Delft University of Technology, The Netherlands</b></td><td>('34959719', 'Sang-Woon Kim', 'sang-woon kim')</td><td>kimsw@mju.ac.kr +<br/>r.p.w.duin@tudelft.nl +</td></tr><tr><td>06466276c4955257b15eff78ebc576662100f740</td><td>Where Is Who: Large-Scale Photo Retrieval by Facial +<br/>Attributes and Canvas Layout +<br/><b>National Taiwan University, Taipei, Taiwan</b></td><td>('2476032', 'Yu-Heng Lei', 'yu-heng lei')<br/>('35081710', 'Yan-Ying Chen', 'yan-ying chen')<br/>('33970300', 'Bor-Chun Chen', 'bor-chun chen')<br/>('2817570', 'Lime Iida', 'lime iida')<br/>('1716836', 'Winston H. Hsu', 'winston h. hsu')</td><td>{siriushpa, limeiida}@gmail.com, winston@csie.ntu.edu.tw +<br/>{ryanlei, yanying}@cmlab.csie.ntu.edu.tw, +</td></tr><tr><td>0697bd81844d54064d992d3229162fe8afcd82cb</td><td>User-driven mobile robot storyboarding: Learning image interest and +<br/>saliency from pairwise image comparisons +</td><td>('1699287', 'Michael Burke', 'michael burke')</td><td></td></tr><tr><td>06262d6beeccf2784e4e36a995d5ee2ff73c8d11</td><td>Recognize Actions by Disentangling Components of Dynamics +<br/><b>CUHK - SenseTime Joint Lab, The Chinese University of Hong Kong 2Amazon Rekognition</b></td><td>('47827548', 'Yue Zhao', 'yue zhao')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('1807606', 'Dahua Lin', 'dahua lin')</td><td>{zy317,dhlin}@ie.cuhk.edu.hk {yuanjx}@amazon.com +</td></tr><tr><td>06f585a3a05dd3371cd600a40dc35500e2f82f9b</td><td>Better and Faster: Knowledge Transfer from Multiple Self-supervised Learning +<br/>Tasks via Graph Distillation for Video Classification +<br/><b>Institute of Computer Science and Technology, Peking University</b><br/>Beijing 100871, China +</td><td>('2439211', 'Chenrui Zhang', 'chenrui zhang')<br/>('1704081', 'Yuxin Peng', 'yuxin peng')</td><td>pengyuxin@pku.edu.cn +</td></tr><tr><td>06f8aa1f436a33014e9883153b93581eea8c5c70</td><td>Leaving Some Stones Unturned: +<br/>Dynamic Feature Prioritization +<br/>for Activity Detection in Streaming Video +<br/><b>The University of Texas at Austin</b><br/>Current approaches for activity recognition often ignore con- +<br/>straints on computational resources: 1) they rely on extensive +<br/>feature computation to obtain rich descriptors on all frames, +<br/>and 2) they assume batch-mode access to the entire test video at +<br/>once. We propose a new active approach to activity recognition +<br/>that prioritizes “what to compute when” in order to make timely +<br/>predictions. The main idea is to learn a policy that dynamically +<br/>schedules the sequence of features to compute on selected frames +<br/>of a given test video. In contrast to traditional static feature +<br/>selection, our approach continually re-prioritizes computation +<br/>based on the accumulated history of observations and accounts +<br/>for the transience of those observations in ongoing video. We +<br/>develop variants to handle both the batch and streaming settings. +<br/>On two challenging datasets, our method provides significantly +<br/>better accuracy than alternative techniques for a wide range of +<br/>computational budgets. +<br/>I. INTRODUCTION +<br/>Activity recognition in video is a core vision challenge. It +<br/>has applications in surveillance, autonomous driving, human- +<br/>robot interaction, and automatic tagging for large-scale video +<br/>retrieval. In any such setting, a system that can both categorize +<br/>and temporally localize activities would be of great value. +<br/>Activity recognition has attracted a steady stream of in- +<br/>teresting research [1]. Recent methods are largely learning- +<br/>based, and tackle realistic everyday activities (e.g., making +<br/>tea, riding a bike). Due to the complexity of the problem, +<br/>as well as the density of raw data comprising even short +<br/>videos, useful video representations are often computationally +<br/>intensive—whether dense trajectories, interest points, object +<br/>detectors, or convolutional neural network (CNN) features run +<br/>on each frame [2]–[8]. In fact, the expectation is that the more +<br/>features one extracts from the video, the better for accuracy. +<br/>For a practitioner wanting reliable activity recognition, then, +<br/>the message is to “leave no stone unturned”, ideally extracting +<br/>complementary descriptors from all video frames. +<br/>However, the “no stone unturned” strategy is problematic. +<br/>Not only does it assume virtually unbounded computational +<br/>resources, it also assumes that an entire video is available +<br/>at once for batch processing. In reality, a recognition system +<br/>will have some computational budget. Further, it may need +<br/>to perform in a streaming manner, with access to only a short +<br/>buffer of recent frames. Together, these considerations suggest +<br/>some form of feature triage is needed. +<br/>Yet prioritizing features for activity in video is challenging, +<br/>for two key reasons. First, +<br/>informative features +<br/>may depend critically on what has been observed so far in +<br/>the most +<br/>the specific test video, making traditional fixed/static feature +<br/>selection methods inadequate. In other words, the recognition +<br/>system’s belief state must evolve over time, and its priorities of +<br/>which features to extract next must evolve too. Second, when +<br/>processing streaming video, the entire video is never available +<br/>to the algorithm at once. This puts limits on what features can +<br/>even be considered each time step, and requires accounting +<br/>for the feature extractors’ framerates when allocating compu- +<br/>tation. +<br/>In light of these challenges, we propose a dynamic approach +<br/>to prioritize which features to compute when for activity +<br/>recognition. We formulate the problem as policy learning in a +<br/>Markov decision process. In particular, we learn a non-myopic +<br/>policy that maps the accumulated feature history (state) to the +<br/>subsequent feature and space-time location (action) that, once +<br/>extracted, is most expected to improve recognition accuracy +<br/>(reward) over a sequence of such actions. We develop two +<br/>variants of our approach: one for batch processing, where +<br/>we are free to “jump” around the video to get +<br/>the next +<br/>desired feature, and one for streaming video, where we are +<br/>confined to a buffer of newly received frames. By dynamically +<br/>allocating feature extraction effort, our method wisely leaves +<br/>some stones unturned—that is, some features unextracted—in +<br/>order to meet real computational budget constraints. +<br/>To our knowledge, our work is the first to actively triage +<br/>feature computation for streaming activity recognition.1 While +<br/>recent work explores ways to intelligently order feature com- +<br/>putation in a static image for the sake of object or scene +<br/>recognition [10]–[17] or offline batch activity detection [18], +<br/>streaming video presents unique challenges, as we explain in +<br/>detail below. While methods for “early” detection can fire on +<br/>an action prior to its completion [19]–[21], they nonetheless +<br/>passively extract all features in each incoming frame. +<br/>We validate our approach on two public datasets consist- +<br/>ing of third- and first-person video from over 120 activity +<br/>categories. We show its impact in both the streaming and +<br/>batch settings, and we further consider scenarios where the test +<br/>video is “untrimmed”. Comparisons with status quo passive +<br/>feature extraction, traditional feature selection approaches, and +<br/>a state-of-the-art early event detector demonstrate the clear +<br/>advantages of our approach. +<br/>1This paper extends our earlier technical report [9]. +</td><td>('39523296', 'Yu-Chuan Su', 'yu-chuan su')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>061c84a4143e859a7caf6e6d283dfb30c23ee56e</td><td>DEEP-CARVING : Discovering Visual Attributes by Carving Deep Neural Nets +<br/><b>Machine Intelligence Lab (MIL), Cambridge University</b><br/>∗Computer Science & Artificial Intelligence Lab (CSAIL), MIT +<br/>Most of the approaches for discovering visual attributes in images de- +<br/>mand significant supervision, which is cumbersome to obtain. In this paper, +<br/>we aim to discover visual attributes in a weakly supervised setting that is +<br/>commonly encountered with contemporary image search engines. +<br/>For instance, given a noun (say forest) and its associated attributes (say +<br/>dense, sunlit, autumn), search engines can now generate many valid im- +<br/>ages for any attribute-noun pair (dense forests, autumn forests, etc). How- +<br/>ever, images for an attribute-noun pair do not contain any information about +<br/>other attributes (like which forests in the autumn are dense too). Thus, a +<br/>weakly supervised scenario occurs. Let A = {a1, . . . ,aM} be the set of +<br/>M attributes under consideration. We have a weakly supervised training +<br/>set, S = {(x1,y1), . . . , (xN,yN )} of N images x1, . . . ,xN ∈ X having labels +<br/>y1, . . . ,yN ∈ A respectively. Equivalently, segregating the training images +<br/>based on their label, we obtain M sets Sm = Xm × am, where Xm = {x ∈ +<br/>X|(x,am) ∈ S} denotes the set of Nm = |Xm| images each having the (sin- +<br/>gle) positive training label am,m ∈ {1, . . . ,M}. For a test image xt, the task +<br/>is to predict yt ⊆ A, i.e. all the attributes present in xt. The aforemen- +<br/>tioned weakly supervised problem setting is more challenging for attributes +<br/>as compared to object and scene detection, because attributes can highly co- +</td><td>('1808862', 'Sukrit Shankar', 'sukrit shankar')<br/>('3307138', 'Vikas K. Garg', 'vikas k. garg')<br/>('1745672', 'Roberto Cipolla', 'roberto cipolla')</td><td></td></tr><tr><td>06d93a40365da90f30a624f15bf22a90d9cfe6bb</td><td>Learning from Candidate Labeling Sets +<br/><b>Idiap Research Institute and EPF Lausanne</b><br/>Luo Jie +<br/>DSI, Universit`a degli Studi di Milano +</td><td>('1721068', 'Francesco Orabona', 'francesco orabona')</td><td>jluo@idiap.ch +<br/>orabona@dsi.unimi.it +</td></tr><tr><td>061e29eae705f318eee703b9e17dc0989547ba0c</td><td>Enhancing Expression Recognition in the Wild +<br/>with Unlabeled Reference Data +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>CAS), Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>Graduate University of Chinese Academy of Sciences, Beijing 100049, China</b></td><td>('1730228', 'Mengyi Liu', 'mengyi liu')<br/>('1688086', 'Shaoxin Li', 'shaoxin li')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{mengyi.liu, shaoxin.li, shiguang.shan, xilin.chen}@vipl.ict.ac.cn; +</td></tr><tr><td>06850b60e33baa4ea9473811d58c0d5015da079e</td><td>A SURVEY OF THE TRENDS IN FACIAL AND +<br/>EXPRESSION RECOGNITION DATABASES AND +<br/>METHODS +<br/><b>University of Washington, Bothell</b><br/><b>University of Washington, Bothell</b></td><td>('2971095', 'Sohini Roychowdhury', 'sohini roychowdhury')<br/>('31448697', 'Michelle Emmons', 'michelle emmons')</td><td>roych@uw.edu +<br/>memmons1442@gmail.com +</td></tr><tr><td>06e7e99c1fdb1da60bc3ec0e2a5563d05b63fe32</td><td>WhittleSearch: Image Search with Relative Attribute Feedback +<br/>(Supplementary Material) +<br/>1 Comparative Qualitative Search Results +<br/>We present three qualitative search results for human-generated feedback, in addition to those +<br/>shown in the paper. Each example shows one search iteration, where the 20 reference images are +<br/>randomly selected (rather than ones that match a keyword search, as the image examples in the +<br/>main paper illustrate). For each result, the first figure shows our method and the second figure +<br/>shows the binary feedback result for the corresponding target image. Note that for our method, +<br/>“more/less X” (where X is an attribute) means that the target image is more/less X than the +<br/>reference image which is shown. +<br/>Figures 1 and 2 show results for human-generated relative attribute and binary feedback, re- +<br/>spectively, when both methods are used to target the same “mental image” of a shoe shown in the +<br/>top left bubble. The top right grid of 20 images are the reference images displayed to the user, and +<br/>those outlined and annotated with constraints are the ones chosen by the user to give feedback. +<br/>The bottom row of images in either figure shows the top-ranked images after integrating the user’s +<br/>feedback into the scoring function, revealing the two methods’ respective performance. We see that +<br/>while both methods retrieve high-heeled shoes, only our method retrieves images that are as “open” +<br/>as the target image. This is because using the proposed approach, the user was able to comment +<br/>explicitly on the desired openness property. +</td><td>('1770205', 'Adriana Kovashka', 'adriana kovashka')<br/>('1713589', 'Devi Parikh', 'devi parikh')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td></td></tr><tr><td>06a6347ac14fd0c6bb3ad8190cbe9cdfa5d59efc</td><td>Active Image Clustering: Seeking Constraints from Humans to Complement +<br/>Algorithms +<br/>Computer Science Department +<br/><b>University of Maryland, College Park</b></td><td>('2221075', 'Arijit Biswas', 'arijit biswas')</td><td>arijitbiswas87@gmail.com, djacobs@umiacs.umd.edu +</td></tr><tr><td>066d71fcd997033dce4ca58df924397dfe0b5fd1</td><td>(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:3)(cid:4)(cid:6)(cid:7)(cid:3)(cid:8)(cid:9)(cid:6)(cid:10)(cid:3)(cid:11)(cid:3)(cid:12)(cid:3)(cid:13)(cid:9) +<br/>(cid:3)(cid:4)(cid:14)(cid:6)(cid:15)(cid:16)(cid:3)(cid:17)(cid:18)(cid:3)(cid:11)(cid:5)(cid:19)(cid:4) (cid:20)(cid:5)(cid:11)(cid:21)(cid:6)(cid:3)(cid:6)(cid:22)(cid:9)(cid:20)(cid:6)(cid:10)(cid:9)(cid:11)(cid:9)(cid:8)(cid:11)(cid:5)(cid:19)(cid:4)(cid:6)(cid:23)(cid:17)(cid:24)(cid:19)(cid:2)(cid:5)(cid:11)(cid:21)(cid:25) +<br/>(cid:26)(cid:11)(cid:5)(cid:8)(cid:17)(cid:6)(cid:27)(cid:1)(cid:9)(cid:22)(cid:8)(cid:18)(cid:1)(cid:28)(cid:12)(cid:6)(cid:29)(cid:4)(cid:20)(cid:11)(cid:6)(cid:24)(cid:30)(cid:1)(cid:15)(cid:25)(cid:1)(cid:31)(cid:8)(cid:20)(cid:8) (cid:14)(cid:1)!(cid:8) (cid:8)(cid:6)(cid:4)(cid:1)"(cid:16)(cid:8)(cid:16)(cid:20)(cid:14)(cid:1)(cid:3)(cid:15)(cid:8)(cid:22)(cid:4)(cid:12)(cid:1)(cid:23)(cid:5)(cid:29)(cid:18)(cid:14)(cid:1)(cid:31)(cid:8)(cid:20)(cid:8) (cid:14)(cid:1)(cid:26)!(cid:9)(cid:13)(cid:14)(cid:1)#(cid:17)(cid:8)(cid:6)(cid:5)$(cid:1)(cid:17)(cid:4)(cid:5)%(cid:8)(cid:10)(cid:8)(cid:11)(cid:6)(cid:8)(cid:12)&(cid:30)(cid:8)(cid:16)(cid:15)(cid:15)(cid:21)(cid:27)(cid:15)(cid:17) +<br/>(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:1)(cid:9)(cid:10)(cid:10)(cid:8)(cid:11)(cid:6)(cid:8)(cid:12)(cid:1)(cid:13)(cid:6)(cid:7)(cid:14) (cid:3)(cid:15)(cid:16)(cid:8)(cid:17)(cid:17)(cid:8)(cid:18)(cid:1)(cid:3)(cid:8)(cid:16)(cid:18)(cid:6)(cid:1)(cid:19)(cid:4)(cid:16)(cid:11)(cid:16)(cid:6)(cid:10)(cid:6)(cid:14)(cid:1)(cid:19)(cid:20)(cid:21)(cid:1)(cid:9)(cid:22)(cid:8)(cid:17)(cid:1)(cid:23)(cid:8)(cid:11)(cid:24)(cid:8)(cid:12)(cid:25)(cid:8)(cid:20)(cid:18) +<br/>(cid:23)(cid:12)(cid:13)(cid:11)(cid:2)(cid:3)(cid:8)(cid:11)$(cid:1)’(cid:16)(cid:6)(cid:11) ((cid:8)((cid:4)(cid:20)(cid:1)(cid:6)(cid:12)(cid:24)(cid:20)(cid:15)(cid:18))(cid:27)(cid:4)(cid:11)(cid:1)(cid:8)(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)(cid:15)(cid:25)(cid:1)(cid:15)(cid:29)(cid:4)(cid:20)(cid:1)*(cid:14)+,,(cid:1)(cid:27)(cid:15)(cid:5)(cid:15)(cid:20)(cid:1)(cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1).(cid:4)(cid:1)(cid:27)(cid:15)(cid:5)(cid:5)(cid:4)(cid:27)(cid:24)(cid:4)(cid:18)(cid:1)(cid:25)(cid:20)(cid:15)(cid:17)(cid:1)+(cid:2)+(cid:1)(cid:18)(cid:6)(cid:25)(cid:25)(cid:4)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1)(cid:16))(cid:17)(cid:8)(cid:12)(cid:1)(cid:25)(cid:8)(cid:27)(cid:4)(cid:11) (cid:6)(cid:12)(cid:1)(cid:8)-(cid:4)(cid:11)(cid:1)(cid:10)(cid:4)(cid:24).(cid:4)(cid:4)(cid:12)(cid:1)/ +<br/>(cid:8)(cid:12)(cid:18) 01(cid:21)(cid:1)2(cid:4)(cid:1)(cid:12)(cid:8)(cid:17)(cid:4)(cid:18)(cid:1)(cid:24)(cid:16)(cid:6)(cid:11)(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)(cid:24)(cid:16)(cid:4)(cid:1)(cid:26)(cid:20)(cid:8)(cid:12)(cid:6)(cid:8)(cid:12)(cid:1)3(cid:8)(cid:27)(cid:4)(cid:1)(cid:19)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)4(cid:26)3(cid:19)(cid:23)5(cid:21)(cid:1)’(cid:15)(cid:1)(cid:4)(cid:29)(cid:8)(cid:5))(cid:8)(cid:24)(cid:4)(cid:1)(cid:24)(cid:16)(cid:4)(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)(cid:24)(cid:16)(cid:4)(cid:1)(cid:4)6((cid:4)(cid:20)(cid:6)(cid:17)(cid:4)(cid:12)(cid:24)(cid:8)(cid:5)(cid:1)(cid:20)(cid:4)(cid:11))(cid:5)(cid:24)(cid:1)(cid:15)(cid:25)(cid:1)(cid:8)(cid:1)(cid:12)(cid:4).(cid:1)(cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) +<br/>(cid:25)(cid:4)(cid:8)(cid:24))(cid:20)(cid:4)(cid:1)(cid:18)(cid:4)(cid:24)(cid:4)(cid:27)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1)(cid:8)(cid:5)-(cid:15)(cid:20)(cid:6)(cid:24)(cid:16)(cid:17)(cid:1)(cid:6)(cid:11)(cid:1)(cid:20)(cid:4)((cid:15)(cid:20)(cid:24)(cid:4)(cid:18)(cid:21) +<br/>(cid:26)(cid:9)(cid:27) (cid:28)(cid:19)(cid:2)(cid:14)(cid:13)$(cid:1)3(cid:8)(cid:27)(cid:4)(cid:1)(cid:26)(cid:17)(cid:8)-(cid:4)(cid:1)(cid:19)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:14)(cid:1)3(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1)3(cid:4)(cid:8)(cid:24))(cid:20)(cid:4)(cid:1)(cid:19)(cid:4)(cid:24)(cid:4)(cid:27)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1)(cid:9)(cid:5)-(cid:15)(cid:20)(cid:6)(cid:24)(cid:16)(cid:17)(cid:11)(cid:14)(cid:1)(cid:9)-(cid:4)(cid:1)7(cid:5)(cid:8)(cid:11)(cid:11)(cid:6)(cid:25)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:21) +<br/>(cid:29) (cid:1)(cid:4)(cid:11)(cid:2)(cid:19)(cid:14)(cid:18)(cid:8)(cid:11)(cid:5)(cid:19)(cid:4) +<br/>8)(cid:17)(cid:8)(cid:12)(cid:1) (cid:25)(cid:8)(cid:27)(cid:4)(cid:1) (cid:6)(cid:11)(cid:1) (cid:24)(cid:16)(cid:4)(cid:1) (cid:17)(cid:15)(cid:11)(cid:24)(cid:1) (cid:27)(cid:15)(cid:17)(cid:17)(cid:15)(cid:12)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) )(cid:11)(cid:4)(cid:25))(cid:5)(cid:1) (cid:7)(cid:4)(cid:30)(cid:1) (cid:24)(cid:15)(cid:1) (cid:8)(cid:1) +<br/>((cid:4)(cid:20)(cid:11)(cid:15)(cid:12)9(cid:11)(cid:1) (cid:6)(cid:18)(cid:4)(cid:12)(cid:24)(cid:6)(cid:24)(cid:30)(cid:21)(cid:1) (cid:9)(cid:11)(cid:1) (cid:16))(cid:17)(cid:8)(cid:12)(cid:11)(cid:14)(cid:1) .(cid:4)(cid:1) (cid:8)(cid:20)(cid:4)(cid:1) (cid:8)(cid:10)(cid:5)(cid:4)(cid:1) (cid:24)(cid:15)(cid:1) (cid:27)(cid:8)(cid:24)(cid:4)-(cid:15)(cid:20)(cid:6)(cid:22)(cid:4)(cid:1) (cid:8)(cid:1) +<br/>((cid:4)(cid:20)(cid:11)(cid:15)(cid:12):(cid:11)(cid:1)(cid:8)-(cid:4)(cid:1)-(cid:20)(cid:15))((cid:1)(cid:25)(cid:20)(cid:15)(cid:17)(cid:1)(cid:8)(cid:1)((cid:4)(cid:20)(cid:11)(cid:15)(cid:12):(cid:11)(cid:1)(cid:25)(cid:8)(cid:27)(cid:4)(cid:1)(cid:6)(cid:17)(cid:8)-(cid:4)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:8)(cid:20)(cid:4)(cid:1)(cid:15)(cid:25)(cid:24)(cid:4)(cid:12)(cid:1) +<br/>(cid:8)(cid:10)(cid:5)(cid:4)(cid:1)(cid:24)(cid:15)(cid:1)(cid:10)(cid:4)(cid:1);)(cid:6)(cid:24)(cid:4)(cid:1)((cid:20)(cid:4)(cid:27)(cid:6)(cid:11)(cid:4)(cid:1)(cid:6)(cid:12)(cid:1)(cid:24)(cid:16)(cid:6)(cid:11)(cid:1)(cid:4)(cid:11)(cid:24)(cid:6)(cid:17)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1)<(cid:2)=(cid:21)(cid:1)(cid:26)(cid:12)(cid:1)(cid:20)(cid:4)(cid:27)(cid:4)(cid:12)(cid:24)(cid:1)(cid:30)(cid:4)(cid:8)(cid:20)(cid:11)(cid:14)(cid:1) +<br/>(cid:25)(cid:8)(cid:27)(cid:4)(cid:1) (cid:20)(cid:4)(cid:27)(cid:15)-(cid:12)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:20)(cid:4)(cid:5)(cid:8)(cid:24)(cid:4)(cid:18)(cid:1) .(cid:15)(cid:20)(cid:7)(cid:11)(cid:1) (cid:16)(cid:8)(cid:29)(cid:4)(cid:1) (cid:20)(cid:4)(cid:27)(cid:4)(cid:6)(cid:29)(cid:4)(cid:18)(cid:1) (cid:11))(cid:10)(cid:11)(cid:24)(cid:8)(cid:12)(cid:24)(cid:6)(cid:8)(cid:5)(cid:1) +<br/>(cid:8)(cid:24)(cid:24)(cid:4)(cid:12)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) (cid:25)(cid:20)(cid:15)(cid:17)(cid:1) (cid:20)(cid:4)(cid:11)(cid:4)(cid:8)(cid:20)(cid:27)(cid:16)(cid:4)(cid:20)(cid:11)(cid:1) (cid:6)(cid:12)(cid:1) (cid:10)(cid:6)(cid:15)(cid:17)(cid:4)(cid:24)(cid:20)(cid:6)(cid:27)(cid:11)(cid:14)(cid:1) ((cid:8)(cid:24)(cid:24)(cid:4)(cid:20)(cid:12)(cid:1) (cid:20)(cid:4)(cid:27)(cid:15)-(cid:12)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:14)(cid:1) +<br/>(cid:8)(cid:12)(cid:18)(cid:1) (cid:27)(cid:15)(cid:17)()(cid:24)(cid:4)(cid:20) (cid:29)(cid:6)(cid:11)(cid:6)(cid:15)(cid:12)(cid:1) (cid:27)(cid:15)(cid:17)(cid:17))(cid:12)(cid:6)(cid:24)(cid:6)(cid:4)(cid:11)(cid:1) </(cid:14)(cid:1) *(cid:14)(cid:1) > (cid:8)(cid:12)(cid:18) 1=(cid:21)(cid:1) ’(cid:16)(cid:4)(cid:11)(cid:4)(cid:1) +<br/>(cid:27)(cid:15)(cid:17)(cid:17)(cid:15)(cid:12)(cid:1)(cid:6)(cid:12)(cid:24)(cid:4)(cid:20)(cid:4)(cid:11)(cid:24)(cid:11)(cid:1)(cid:8)(cid:17)(cid:15)(cid:12)-(cid:1)(cid:20)(cid:4)(cid:11)(cid:4)(cid:8)(cid:20)(cid:27)(cid:16)(cid:4)(cid:20)(cid:11)(cid:1)(cid:17)(cid:15)(cid:24)(cid:6)(cid:29)(cid:8)(cid:24)(cid:4)(cid:18)(cid:1))(cid:11)(cid:1)(cid:24)(cid:15)(cid:1)(cid:27)(cid:15)(cid:5)(cid:5)(cid:4)(cid:27)(cid:24)(cid:1)(cid:8)(cid:1) +<br/>(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1) (cid:15)(cid:25)(cid:1) (cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) (cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1) (cid:25)(cid:20)(cid:15)(cid:17)(cid:1) ((cid:4)(cid:15)((cid:5)(cid:4)(cid:1) (cid:6)(cid:12)(cid:1) (cid:18)(cid:6)(cid:25)(cid:25)(cid:4)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1) (cid:8)-(cid:4)(cid:11)(cid:21) ’(cid:16)(cid:4)(cid:1) +<br/>(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)(cid:6)(cid:11)(cid:1)(cid:6)(cid:12)(cid:24)(cid:4)(cid:12)(cid:18)(cid:4)(cid:18)(cid:1)(cid:25)(cid:15)(cid:20)(cid:1)(cid:18)(cid:6)(cid:11)(cid:24)(cid:20)(cid:6)(cid:10))(cid:24)(cid:6)(cid:15)(cid:12)(cid:1)(cid:24)(cid:15)(cid:1)(cid:20)(cid:4)(cid:11)(cid:4)(cid:8)(cid:20)(cid:27)(cid:16)(cid:4)(cid:20)(cid:11)(cid:21) +<br/>’(cid:16)(cid:4)(cid:20)(cid:4)(cid:1) (cid:8)(cid:20)(cid:4)(cid:1) (cid:17)(cid:8)(cid:12)(cid:30)(cid:1) ()(cid:10)(cid:5)(cid:6)(cid:27)(cid:8)(cid:5)(cid:5)(cid:30)(cid:1) (cid:8)(cid:29)(cid:8)(cid:6)(cid:5)(cid:8)(cid:10)(cid:5)(cid:4)(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)(cid:1) (cid:25)(cid:15)(cid:20)(cid:1) (cid:25)(cid:8)(cid:27)(cid:4)(cid:1) +<br/>(cid:20)(cid:4)(cid:27)(cid:15)-(cid:12)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:14)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) (cid:4)6((cid:20)(cid:4)(cid:11)(cid:11)(cid:6)(cid:15)(cid:12)(cid:1) (cid:8)(cid:12)(cid:8)(cid:5)(cid:30)(cid:11)(cid:6)(cid:11)(cid:21)(cid:1) (cid:23)(cid:4)(cid:11)(cid:6)(cid:18)(cid:4)(cid:1) (cid:8)(cid:10)(cid:15)(cid:29)(cid:4)(cid:1) +<br/>(cid:8)(((cid:5)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:14)(cid:1)(cid:26)(cid:20)(cid:8)(cid:12)(cid:6)(cid:8)(cid:12)(cid:1)3(cid:8)(cid:27)(cid:4)(cid:1)(cid:19)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)4(cid:26)3(cid:19)(cid:23)5(cid:1)(cid:27)(cid:8)(cid:12)(cid:1)(cid:10)(cid:4)(cid:1))(cid:11)(cid:4)(cid:18)(cid:1)(cid:25)(cid:15)(cid:20)(cid:1)(cid:8)-(cid:4)(cid:1) +<br/>(cid:27)(cid:5)(cid:8)(cid:11)(cid:11)(cid:6)(cid:25)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:14)(cid:1) (cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) (cid:11))(cid:20)-(cid:4)(cid:20)(cid:30)(cid:14)(cid:1) (cid:20)(cid:8)(cid:27)(cid:4)(cid:1) (cid:18)(cid:4)(cid:24)(cid:4)(cid:27)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) 4(cid:10)(cid:4)(cid:11)(cid:6)(cid:18)(cid:4)(cid:1) (cid:15)(cid:24)(cid:16)(cid:4)(cid:20)(cid:1) +<br/>(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)5(cid:14)(cid:1) (cid:11)(cid:24))(cid:18)(cid:30)(cid:6)(cid:12)-(cid:1) (cid:6)(cid:12)(cid:25)(cid:5))(cid:4)(cid:12)(cid:27)(cid:4)(cid:1) (cid:15)(cid:25)(cid:1) (cid:27)(cid:8)(cid:20)(cid:4)(cid:4)(cid:20)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:7)(cid:6)(cid:12)(cid:18)(cid:1) (cid:15)(cid:25)(cid:1) (cid:11)(cid:7)(cid:6)(cid:12)(cid:1) (cid:15)(cid:12)(cid:1) +<br/>(cid:8)-(cid:6)(cid:12)-(cid:14)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:15)(cid:24)(cid:16)(cid:4)(cid:20)(cid:1)(cid:11)(cid:6)(cid:17)(cid:6)(cid:5)(cid:8)(cid:20)(cid:1)(cid:20)(cid:4)(cid:11)(cid:4)(cid:8)(cid:20)(cid:27)(cid:16)(cid:4)(cid:11)(cid:21) +<br/>(cid:26)(cid:12)(cid:1) (cid:24)(cid:16)(cid:4)(cid:1) (cid:20)(cid:4)(cid:17)(cid:8)(cid:6)(cid:12)(cid:6)(cid:12)-(cid:1) ((cid:8)(cid:20)(cid:24)(cid:11) (cid:18)(cid:4)(cid:24)(cid:8)(cid:6)(cid:5)(cid:11)(cid:1) (cid:15)(cid:25)(cid:1) (cid:24)(cid:16)(cid:4)(cid:1) (cid:4)6(cid:6)(cid:11)(cid:24)(cid:6)(cid:12)-(cid:1) (cid:25)(cid:8)(cid:27)(cid:4)(cid:1) (cid:6)(cid:17)(cid:8)-(cid:4)(cid:1) +<br/>(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11) (cid:8)(cid:12)(cid:18) (cid:24)(cid:16)(cid:4)(cid:1)(cid:26)(cid:20)(cid:8)(cid:12)(cid:6)(cid:8)(cid:12)(cid:1)3(cid:8)(cid:27)(cid:4)(cid:1)(cid:26)(cid:17)(cid:8)-(cid:4)(cid:1)(cid:19)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)(cid:6)(cid:11)(cid:1)-(cid:6)(cid:29)(cid:4)(cid:12)(cid:21) (cid:9)(cid:5)(cid:11)(cid:15)(cid:1) +<br/>(cid:24)(cid:16)(cid:4)(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1) (cid:6)(cid:11)(cid:1) (cid:4)(cid:29)(cid:8)(cid:5))(cid:8)(cid:24)(cid:4)(cid:18)(cid:1) (cid:10)(cid:30)(cid:1) (cid:8)(((cid:5)(cid:30)(cid:6)(cid:12)- (cid:8)(cid:1) (cid:12)(cid:4).(cid:1) (cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) (cid:25)(cid:4)(cid:8)(cid:24))(cid:20)(cid:4)(cid:1) +<br/>(cid:18)(cid:4)(cid:24)(cid:4)(cid:27)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1)(cid:8)(cid:5)-(cid:15)(cid:20)(cid:6)(cid:24)(cid:16)(cid:17)(cid:21)(cid:1) +<br/>(cid:30) (cid:15)(cid:31)(cid:5)(cid:13)(cid:11)(cid:5)(cid:4)(cid:24)(cid:6)(cid:7)(cid:3)(cid:8)(cid:9)(cid:6)(cid:1)(cid:25)(cid:3)(cid:24)(cid:9)(cid:6)(cid:10)(cid:3)(cid:11)(cid:3)(cid:12)(cid:3)(cid:13)(cid:9)(cid:13) +<br/>(cid:3)(cid:8)(cid:12)(cid:30)(cid:1) (cid:25)(cid:8)(cid:27)(cid:4)(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)(cid:1) (cid:8)(cid:20)(cid:4)(cid:1) (cid:20)(cid:4)(cid:27)(cid:15)(cid:20)(cid:18)(cid:4)(cid:18)(cid:1) )(cid:12)(cid:18)(cid:4)(cid:20)(cid:1) (cid:8)(cid:1) (cid:29)(cid:8)(cid:20)(cid:6)(cid:4)(cid:24)(cid:30)(cid:1) (cid:15)(cid:25)(cid:1) +<br/>(cid:27)(cid:15)(cid:12)(cid:18)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1).(cid:6)(cid:24)(cid:16)(cid:1)(cid:29)(cid:8)(cid:20)(cid:6)(cid:15))(cid:11)(cid:1)(cid:8)(((cid:5)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:1)(cid:6)(cid:12)(cid:1)(cid:17)(cid:6)(cid:12)(cid:18)(cid:21)(cid:1)(cid:9)(cid:5)(cid:15)(cid:12)-(cid:1).(cid:6)(cid:24)(cid:16)(cid:1) +<br/>(cid:24)(cid:16)(cid:4)(cid:1) (cid:18)(cid:4)(cid:29)(cid:4)(cid:5)(cid:15)((cid:17)(cid:4)(cid:12)(cid:24)(cid:1) (cid:15)(cid:25)(cid:1) (cid:25)(cid:8)(cid:27)(cid:4)(cid:1) (cid:20)(cid:4)(cid:27)(cid:15)-(cid:12)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) (cid:4)6((cid:20)(cid:4)(cid:11)(cid:11)(cid:6)(cid:15)(cid:12)(cid:1) +<br/>(cid:8)(cid:12)(cid:8)(cid:5)(cid:30)(cid:11)(cid:6)(cid:11)(cid:1) (cid:8)(cid:5)-(cid:15)(cid:20)(cid:6)(cid:24)(cid:16)(cid:17)(cid:11)(cid:14)(cid:1) (cid:8)(cid:1) (cid:27)(cid:15)(cid:17)((cid:8)(cid:20)(cid:8)(cid:24)(cid:6)(cid:29)(cid:4)(cid:5)(cid:30)(cid:1) (cid:5)(cid:8)(cid:20)-(cid:4)(cid:1) (cid:12))(cid:17)(cid:10)(cid:4)(cid:20)(cid:1) (cid:15)(cid:25)(cid:1) (cid:25)(cid:8)(cid:27)(cid:4)(cid:1) +<br/>A(cid:8)(cid:5)(cid:4)(cid:1)<0=(cid:14)(cid:1)(cid:3)(cid:26)’(cid:1)<B=(cid:14)(cid:1)C(cid:9)33#(cid:1)<(cid:2),=(cid:14)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:17)(cid:8)(cid:12)(cid:30)(cid:1)(cid:15)(cid:24)(cid:16)(cid:4)(cid:20)(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)(cid:1)<(cid:2)(cid:2)(cid:14)(cid:1) +<br/>(cid:2)/=(cid:21)(cid:1)8(cid:4)(cid:20)(cid:4)(cid:1)3#!#’(cid:1)<(cid:2)*= (cid:8)(cid:12)(cid:18)(cid:1)3DE(cid:13)#’(cid:1)<(cid:2)>=(cid:1)(cid:8)(cid:20)(cid:4)(cid:1)(cid:20)(cid:4)(cid:29)(cid:6)(cid:4).(cid:4)(cid:18)(cid:21) +<br/>(cid:30) (cid:29) (cid:7)(cid:15)!(cid:15)"(cid:6)(cid:10)(cid:3)(cid:11)(cid:3)(cid:12)(cid:3)(cid:13)(cid:9) +<br/>’(cid:16)(cid:4)(cid:1) 3(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) !(cid:4)(cid:27)(cid:15)-(cid:12)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) ’(cid:4)(cid:27)(cid:16)(cid:12)(cid:15)(cid:5)(cid:15)-(cid:30)(cid:1) 43#!#’5(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1) +<br/>.(cid:8)(cid:11)(cid:1)(cid:27)(cid:15)(cid:5)(cid:5)(cid:4)(cid:27)(cid:24)(cid:4)(cid:18)(cid:1) (cid:8)(cid:24)(cid:1) D(cid:4)(cid:15)(cid:20)-(cid:4)(cid:1)(cid:3)(cid:8)(cid:11)(cid:15)(cid:12)(cid:1) (cid:28)(cid:12)(cid:6)(cid:29)(cid:4)(cid:20)(cid:11)(cid:6)(cid:24)(cid:30)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1)(cid:24)(cid:16)(cid:4)(cid:1) (cid:28)"(cid:1) (cid:9)(cid:20)(cid:17)(cid:30)(cid:1) +<br/>!(cid:4)(cid:11)(cid:4)(cid:8)(cid:20)(cid:27)(cid:16)(cid:1)F(cid:8)(cid:10)(cid:15)(cid:20)(cid:8)(cid:24)(cid:15)(cid:20)(cid:30)(cid:1)(cid:25)(cid:8)(cid:27)(cid:6)(cid:5)(cid:6)(cid:24)(cid:6)(cid:4)(cid:11)(cid:1) (cid:8)(cid:11)(cid:1)((cid:8)(cid:20)(cid:24)(cid:1)(cid:15)(cid:25)(cid:1) (cid:24)(cid:16)(cid:4)(cid:1)3#!#’(cid:1) ((cid:20)(cid:15)-(cid:20)(cid:8)(cid:17)(cid:1) +<br/><(cid:2)*=(cid:21)(cid:1)(cid:26)(cid:12)(cid:1)3#!#’(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)(cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1)(cid:15)(cid:25)(cid:1)(cid:2)(cid:2)BB(cid:1)(cid:11))(cid:10) (cid:4)(cid:27)(cid:24)(cid:1)(cid:4)6(cid:6)(cid:11)(cid:24)(cid:1)(cid:6)(cid:12)(cid:1)BE/, +<br/>(cid:18)(cid:6)(cid:25)(cid:25)(cid:4)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1) ((cid:15)(cid:11)(cid:4)(cid:11)(cid:14)(cid:1) /(cid:1) +<br/>(cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) (cid:4)6((cid:20)(cid:4)(cid:11)(cid:11)(cid:6)(cid:15)(cid:12)(cid:11)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) /(cid:1) (cid:18)(cid:6)(cid:25)(cid:25)(cid:4)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1) +<br/>(cid:6)(cid:5)(cid:5))(cid:17)(cid:6)(cid:12)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:1)(cid:6)(cid:12)(cid:1)/(cid:1)(cid:18)(cid:6)(cid:25)(cid:25)(cid:4)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1)(cid:24)(cid:6)(cid:17)(cid:4)(cid:11)(cid:21)(cid:1)(cid:1)’(cid:16)(cid:4)(cid:20)(cid:4)(cid:1)(cid:8)(cid:20)(cid:4) (cid:2)>(cid:14),1(cid:2)(cid:1)(cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1)(cid:6)(cid:12)(cid:1) +<br/>/1+G*0>(cid:1)((cid:6)6(cid:4)(cid:5)(cid:11)(cid:1)(cid:6)(cid:12)(cid:1)(cid:11)(cid:6)(cid:22)(cid:4)(cid:21)(cid:1)(cid:26)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1).(cid:4)(cid:20)(cid:4)(cid:1)(cid:27)(cid:15)(cid:5)(cid:5)(cid:4)(cid:27)(cid:24)(cid:4)(cid:18)(cid:1)(cid:8)(cid:24)(cid:1)(cid:24)(cid:16)(cid:4)(cid:1)(cid:25)(cid:15)(cid:5)(cid:5)(cid:15).(cid:6)(cid:12)-(cid:1) +<br/>((cid:15)(cid:11)(cid:4)(cid:11)$(cid:1)(cid:20)(cid:6)-(cid:16)(cid:24)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:5)(cid:4)(cid:25)(cid:24)(cid:1)((cid:20)(cid:15)(cid:25)(cid:6)(cid:5)(cid:4)(cid:14)(cid:1)(cid:20)(cid:6)-(cid:16)(cid:24)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:5)(cid:4)(cid:25)(cid:24)(cid:1);)(cid:8)(cid:20)(cid:24)(cid:4)(cid:20)(cid:1)((cid:20)(cid:15)(cid:25)(cid:6)(cid:5)(cid:4)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1) +<br/>(cid:20)(cid:6)-(cid:16)(cid:24)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:5)(cid:4)(cid:25)(cid:24)(cid:1) (cid:16)(cid:8)(cid:5)(cid:25)(cid:1) ((cid:20)(cid:15)(cid:25)(cid:6)(cid:5)(cid:4)(cid:21)(cid:1) (cid:26)(cid:12)(cid:1) (cid:24)(cid:16)(cid:4)(cid:11)(cid:4)(cid:1) (cid:27)(cid:8)(cid:24)(cid:4)-(cid:15)(cid:20)(cid:6)(cid:4)(cid:11)(cid:1) (cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1) .(cid:4)(cid:20)(cid:4)(cid:1) +<br/>(cid:20)(cid:4)(cid:27)(cid:15)(cid:20)(cid:18)(cid:4)(cid:18)(cid:1)(cid:25)(cid:15)(cid:20)(cid:1)1,0(cid:1)(cid:24)(cid:15)(cid:1)B0,(cid:1)(cid:11))(cid:10) (cid:4)(cid:27)(cid:24)(cid:11)(cid:21) +<br/>(cid:30) (cid:30)(cid:6)(cid:7)#$(cid:22)(cid:15)"(cid:6)(cid:23)(cid:24)(cid:5)(cid:4)(cid:24)(cid:6)(cid:10)(cid:3)(cid:11)(cid:3)(cid:12)(cid:3)(cid:13)(cid:9) +<br/>(cid:2) +<br/>’(cid:16)(cid:4)(cid:1)3DE(cid:13)#’(cid:1)(cid:9)-(cid:6)(cid:12)-(cid:1)(cid:19)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1).(cid:8)(cid:11)(cid:1)-(cid:4)(cid:12)(cid:4)(cid:20)(cid:8)(cid:24)(cid:4)(cid:18)(cid:1)(cid:8)(cid:11)(cid:1)((cid:8)(cid:20)(cid:24)(cid:1)(cid:15)(cid:25)(cid:1)(cid:24)(cid:16)(cid:4)(cid:1) +<br/>#)(cid:20)(cid:15)((cid:4)(cid:8)(cid:12)(cid:1) (cid:28)(cid:12)(cid:6)(cid:15)(cid:12)(cid:1) ((cid:20)(cid:15) (cid:4)(cid:27)(cid:24)(cid:1) 3DE(cid:13)#’(cid:1) +<br/>43(cid:8)(cid:27)(cid:4)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) D(cid:4)(cid:11)(cid:24))(cid:20)(cid:4)(cid:1) +<br/>!(cid:4)(cid:27)(cid:15)-(cid:12)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) !(cid:4)(cid:11)(cid:4)(cid:8)(cid:20)(cid:27)(cid:16)(cid:1) (cid:13)(cid:4)(cid:24).(cid:15)(cid:20)(cid:7)5(cid:21)’(cid:16)(cid:6)(cid:11)(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1) (cid:6)(cid:11)(cid:1) (cid:27)(cid:15)(cid:12)(cid:24)(cid:8)(cid:6)(cid:12)(cid:6)(cid:12)-(cid:1) +<br/>(cid:2),,/(cid:1) (cid:11)(cid:27)(cid:8)(cid:12)(cid:12)(cid:4)(cid:18)(cid:1) (cid:25)(cid:8)(cid:27)(cid:4)(cid:1) (cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1) (cid:11)(cid:16)(cid:15).(cid:6)(cid:12)-(cid:1) 0/(cid:1) (cid:11))(cid:10) (cid:4)(cid:27)(cid:24)(cid:11)(cid:1) (cid:8)(cid:24)(cid:1) (cid:18)(cid:6)(cid:25)(cid:25)(cid:4)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1) +<br/>(cid:8)-(cid:4)(cid:11)(cid:21)(cid:1)(cid:26)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1)(cid:16)(cid:8)(cid:29)(cid:4)(cid:1)(cid:29)(cid:8)(cid:20)(cid:30)(cid:6)(cid:12)-(cid:1)(cid:20)(cid:4)(cid:11)(cid:15)(cid:5))(cid:24)(cid:6)(cid:15)(cid:12)?(cid:1)(cid:8)(((cid:20)(cid:15)6(cid:6)(cid:17)(cid:8)(cid:24)(cid:4)(cid:5)(cid:30)(cid:1)>,,G1,, +<br/>((cid:6)6(cid:4)(cid:5)(cid:11)(cid:21)(cid:1) ’(cid:16)(cid:4)(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1) .(cid:8)(cid:11)(cid:1) (cid:18)(cid:4)(cid:29)(cid:4)(cid:5)(cid:15)((cid:4)(cid:18)(cid:1) (cid:6)(cid:12)(cid:1) (cid:8)(cid:12)(cid:1) (cid:8)(cid:24)(cid:24)(cid:4)(cid:17)((cid:24)(cid:1) (cid:24)(cid:15)(cid:1) (cid:8)(cid:11)(cid:11)(cid:6)(cid:11)(cid:24)(cid:1) +<br/>(cid:20)(cid:4)(cid:11)(cid:4)(cid:8)(cid:20)(cid:27)(cid:16)(cid:4)(cid:20)(cid:11)(cid:1) .(cid:16)(cid:15)(cid:1) (cid:6)(cid:12)(cid:29)(cid:4)(cid:11)(cid:24)(cid:6)-(cid:8)(cid:24)(cid:4)(cid:1) (cid:24)(cid:16)(cid:4)(cid:1) (cid:4)(cid:25)(cid:25)(cid:4)(cid:27)(cid:24)(cid:11)(cid:1) (cid:15)(cid:25)(cid:1) (cid:8)-(cid:6)(cid:12)-(cid:1) (cid:15)(cid:12)(cid:1) (cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) +<br/>(cid:8)(((cid:4)(cid:8)(cid:20)(cid:8)(cid:12)(cid:27)(cid:4)(cid:1)<(cid:2)> =(cid:21) +<br/>(cid:30) % (cid:22)(cid:9)(cid:9)(cid:14)(cid:6)(cid:7)(cid:19)(cid:2)(cid:6)(cid:23)(cid:6)(cid:22)(cid:9)(cid:20)(cid:6)(cid:10)(cid:3)(cid:11)(cid:3)(cid:12)(cid:3)(cid:13)(cid:9) +<br/>(cid:26)(cid:12)(cid:1)(cid:15)(cid:20)(cid:18)(cid:4)(cid:20)(cid:1)(cid:24)(cid:15)(cid:1) (cid:10))(cid:6)(cid:5)(cid:18)(cid:14)(cid:1) (cid:24)(cid:20)(cid:8)(cid:6)(cid:12)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:20)(cid:4)(cid:5)(cid:6)(cid:8)(cid:10)(cid:5)(cid:30)(cid:1) (cid:24)(cid:4)(cid:11)(cid:24)(cid:1) (cid:8)-(cid:4)(cid:1) (cid:27)(cid:5)(cid:8)(cid:11)(cid:11)(cid:6)(cid:25)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) +<br/>(cid:8)(cid:5)-(cid:15)(cid:20)(cid:6)(cid:24)(cid:16)(cid:17)(cid:11)(cid:14)(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)(cid:1).(cid:6)(cid:24)(cid:16)(cid:1)(cid:27)(cid:15)(cid:12)(cid:24)(cid:20)(cid:15)(cid:5)(cid:5)(cid:4)(cid:18)(cid:1)(cid:29)(cid:8)(cid:20)(cid:6)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:1)(cid:15)(cid:25)(cid:1)(cid:25)(cid:8)(cid:27)(cid:24)(cid:15)(cid:20)(cid:11)(cid:1)(cid:11))(cid:27)(cid:16)(cid:1) +<br/>(cid:8)(cid:11)(cid:1)(cid:8)-(cid:4)(cid:14)(cid:1)(cid:25)(cid:8)(cid:27)(cid:4)(cid:1)((cid:15)(cid:11)(cid:4)(cid:14)(cid:1)(cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1)(cid:4)6((cid:20)(cid:4)(cid:11)(cid:11)(cid:6)(cid:15)(cid:12)(cid:14)(cid:1)(cid:15)(cid:27)(cid:27)(cid:5))(cid:11)(cid:6)(cid:15)(cid:12)(cid:1)(cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1)(cid:16)(cid:8)(cid:6)(cid:20)(cid:14)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1) +<br/>(cid:6)(cid:5)(cid:5))(cid:17)(cid:6)(cid:12)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) (cid:6)(cid:11)(cid:1) (cid:12)(cid:4)(cid:4)(cid:18)(cid:4)(cid:18)(cid:21)(cid:1) (cid:26)(cid:12)(cid:1) (cid:11)((cid:6)(cid:24)(cid:4)(cid:1) (cid:15)(cid:25)(cid:1) (cid:29)(cid:8)(cid:20)(cid:6)(cid:15))(cid:11)(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)(cid:14)(cid:1) (cid:24)(cid:16)(cid:4)(cid:20)(cid:4)(cid:1) (cid:6)(cid:11)(cid:1) +<br/>(cid:12)(cid:15)(cid:24)(cid:1)(cid:8)(cid:12)(cid:1)(cid:8)(((cid:20)(cid:15)((cid:20)(cid:6)(cid:8)(cid:24)(cid:4)(cid:1)(cid:15)(cid:12)(cid:4)(cid:1)(cid:25)(cid:15)(cid:20)(cid:1)(cid:8)-(cid:4)(cid:1)(cid:27)(cid:5)(cid:8)(cid:11)(cid:11)(cid:6)(cid:25)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:21)(cid:1)(cid:3)(cid:15)(cid:11)(cid:24)(cid:1)(cid:27))(cid:20)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:1) +<br/>(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)(cid:1)(cid:18)(cid:15)(cid:12):(cid:24)(cid:1)(cid:16)(cid:8)(cid:29)(cid:4)(cid:1)(cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1)(cid:15)(cid:25)(cid:1)((cid:4)(cid:15)((cid:5)(cid:4)(cid:1)(cid:6)(cid:12)(cid:1)(cid:18)(cid:6)(cid:25)(cid:25)(cid:4)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1)(cid:8)-(cid:4)(cid:11)(cid:14)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:6)(cid:25)(cid:1)(cid:24)(cid:16)(cid:4)(cid:30)(cid:1) +<br/>(cid:16)(cid:8)(cid:29)(cid:4)(cid:14)(cid:1) (cid:24)(cid:16)(cid:4)(cid:30)(cid:1) (cid:18)(cid:15)(cid:1) (cid:12)(cid:15)(cid:24)(cid:1) (cid:17)(cid:4)(cid:12)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) (cid:24)(cid:16)(cid:4)(cid:6)(cid:20)(cid:1) (cid:8)-(cid:4)(cid:11)(cid:21)(cid:1) 3DE(cid:13)#’(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1) +<br/>(cid:27)(cid:15)(cid:12)(cid:24)(cid:8)(cid:6)(cid:12)(cid:11)(cid:1) (cid:11)(cid:27)(cid:8)(cid:12)(cid:12)(cid:4)(cid:18)(cid:1) (cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1) (cid:15)(cid:25)(cid:1) ((cid:4)(cid:20)(cid:11)(cid:15)(cid:12)(cid:11)(cid:1) .(cid:6)(cid:24)(cid:16)(cid:1) (cid:17)(cid:4)(cid:12)(cid:24)(cid:6)(cid:15)(cid:12)(cid:6)(cid:12)-(cid:1) (cid:24)(cid:16)(cid:4)(cid:6)(cid:20)(cid:1) +<br/>(cid:8)-(cid:4)(cid:11)?(cid:1)(cid:10))(cid:24)(cid:1)(cid:18)(cid:6)(cid:25)(cid:25)(cid:4)(cid:20)(cid:4)(cid:12)(cid:24)(cid:1)(cid:5)(cid:6)-(cid:16)(cid:24)(cid:6)(cid:12)-(cid:1)(cid:27)(cid:15)(cid:12)(cid:18)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:14)(cid:1)(cid:10)(cid:8)(cid:27)(cid:7)-(cid:20)(cid:15))(cid:12)(cid:18)(cid:14)(cid:1)((cid:15)(cid:11)(cid:4)(cid:11)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1) +<br/>(cid:4)6((cid:20)(cid:4)(cid:11)(cid:11)(cid:6)(cid:15)(cid:12)(cid:11)(cid:21)(cid:1)(cid:23)(cid:30)(cid:1)(cid:11)(cid:24))(cid:18)(cid:30)(cid:6)(cid:12)-(cid:1)(cid:15)(cid:24)(cid:16)(cid:4)(cid:20)(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)(cid:1)(cid:6)(cid:24)(cid:1) .(cid:8)(cid:11)(cid:1) (cid:27)(cid:15)(cid:12)(cid:27)(cid:5))(cid:18)(cid:4)(cid:18)(cid:1)(cid:24)(cid:15)(cid:1) +<br/>((cid:20)(cid:15)(cid:29)(cid:6)(cid:18)(cid:4)(cid:1) (cid:8)(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1) .(cid:6)(cid:24)(cid:16)(cid:1) (cid:27)(cid:15)(cid:12)(cid:18)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:1) (cid:15)(cid:25)(cid:1) (cid:8)(cid:12)(cid:1) (cid:8)-(cid:4)(cid:1) (cid:27)(cid:5)(cid:8)(cid:11)(cid:11)(cid:6)(cid:25)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:1) +<br/>((cid:20)(cid:15) (cid:4)(cid:27)(cid:24)(cid:21)(cid:1) (cid:9)-(cid:4)(cid:14)(cid:1) (cid:4)(cid:12)(cid:15))-(cid:16)(cid:1) (cid:20)(cid:4)(cid:11)(cid:15)(cid:5))(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) (cid:25)(cid:15)(cid:20)(cid:1) .(cid:20)(cid:6)(cid:12)(cid:7)(cid:5)(cid:4)(cid:1) (cid:8)(cid:12)(cid:8)(cid:5)(cid:30)(cid:11)(cid:6)(cid:11)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) +<br/>(cid:25)(cid:20)(cid:15)(cid:12)(cid:24)(cid:8)(cid:5)(cid:1)((cid:15)(cid:11)(cid:4)(cid:11)(cid:1)(cid:8)(cid:20)(cid:4)(cid:1)(cid:10)(cid:8)(cid:11)(cid:6)(cid:27)(cid:1)(cid:12)(cid:4)(cid:4)(cid:18)(cid:11)(cid:1)(cid:25)(cid:15)(cid:20)(cid:1)(cid:24)(cid:16)(cid:6)(cid:11)(cid:1)(cid:25)(cid:6)(cid:4)(cid:5)(cid:18)(cid:21)(cid:1) +<br/>% (cid:10)(cid:9)(cid:13)(cid:8)(cid:2)(cid:5)&(cid:11)(cid:5)(cid:19)(cid:4)(cid:6) ’((cid:6) +<br/>(cid:10)(cid:3)(cid:11)(cid:3)(cid:12)(cid:3)(cid:13)(cid:9) +<br/>(cid:1)(cid:2)(cid:3)(cid:4)(cid:5)(cid:3)(cid:4)(cid:6) (cid:7)(cid:3)(cid:8)(cid:9)(cid:6) +<br/>’(cid:16)(cid:4)(cid:1) (cid:26)(cid:20)(cid:8)(cid:12)(cid:6)(cid:8)(cid:12)(cid:1) 3(cid:8)(cid:27)(cid:4)(cid:1) (cid:19)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:14)(cid:1) (cid:24)(cid:16)(cid:4)(cid:1) (cid:25)(cid:6)(cid:20)(cid:11)(cid:24)(cid:1) (cid:6)(cid:17)(cid:8)-(cid:4)(cid:1) (cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1) (cid:6)(cid:12)(cid:1) +<br/>(cid:17)(cid:6)(cid:18)(cid:18)(cid:5)(cid:4)E(cid:4)(cid:8)(cid:11)(cid:24)(cid:14)(cid:1)(cid:27)(cid:15)(cid:12)(cid:24)(cid:8)(cid:6)(cid:12)(cid:11)(cid:1)(cid:27)(cid:15)(cid:5)(cid:15)(cid:20)(cid:1)(cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1)(cid:6)(cid:17)(cid:8)-(cid:4)(cid:20)(cid:30)(cid:1)(cid:15)(cid:25)(cid:1)(cid:8)(cid:1)(cid:5)(cid:8)(cid:20)-(cid:4)(cid:1)(cid:12))(cid:17)(cid:10)(cid:4)(cid:20)(cid:1)(cid:15)(cid:25)(cid:1) +<br/>(cid:26)(cid:20)(cid:8)(cid:12)(cid:6)(cid:8)(cid:12)(cid:1)(cid:11))(cid:10) (cid:4)(cid:27)(cid:24)(cid:11) (cid:10)(cid:4)(cid:24).(cid:4)(cid:4)(cid:12)(cid:1)/(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)01(cid:1)(cid:30)(cid:4)(cid:8)(cid:20)(cid:11)(cid:1)(cid:15)(cid:5)(cid:18)(cid:21) +<br/>(cid:26)3(cid:19)(cid:23)(cid:1)(cid:6)(cid:11)(cid:1)(cid:8)(cid:1)(cid:5)(cid:8)(cid:20)-(cid:4)(cid:1)(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:1)(cid:24)(cid:16)(cid:8)(cid:24)(cid:1)(cid:27)(cid:8)(cid:12)(cid:1)(cid:11))(((cid:15)(cid:20)(cid:24)(cid:1)(cid:11)(cid:24))(cid:18)(cid:6)(cid:4)(cid:11)(cid:1)(cid:15)(cid:25)(cid:1)(cid:24)(cid:16)(cid:4)(cid:1)(cid:8)-(cid:4)(cid:1) +<br/>(cid:27)(cid:5)(cid:8)(cid:11)(cid:11)(cid:6)(cid:25)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) (cid:11)(cid:30)(cid:11)(cid:24)(cid:4)(cid:17)(cid:11)(cid:21)(cid:1) (cid:26)(cid:24)(cid:1) (cid:27)(cid:15)(cid:12)(cid:24)(cid:8)(cid:6)(cid:12)(cid:11)(cid:1) (cid:15)(cid:29)(cid:4)(cid:20)(cid:1) *(cid:14)+,,(cid:1) (cid:27)(cid:15)(cid:5)(cid:15)(cid:20)(cid:1) (cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1) +<br/>(cid:13)(cid:15)(cid:1)(cid:20)(cid:4)(cid:11)(cid:24)(cid:20)(cid:6)(cid:27)(cid:24)(cid:6)(cid:15)(cid:12)(cid:11)(cid:1)(cid:15)(cid:12)(cid:1).(cid:4)(cid:8)(cid:20)(cid:1)4(cid:27)(cid:5)(cid:15)(cid:24)(cid:16)(cid:4)(cid:11)(cid:14)(cid:1)-(cid:5)(cid:8)(cid:11)(cid:11)(cid:4)(cid:11)(cid:14)(cid:1)(cid:4)(cid:24)(cid:27)(cid:21)5(cid:14)(cid:1) (cid:17)(cid:8)(cid:7)(cid:4)E)((cid:14)(cid:1)(cid:16)(cid:8)(cid:6)(cid:20)(cid:1) +<br/>(cid:11)(cid:24)(cid:30)(cid:5)(cid:4)(cid:14)(cid:1) (cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) (cid:16)(cid:8)(cid:6)(cid:20)(cid:1) .(cid:4)(cid:20)(cid:4)(cid:1) (cid:6)(cid:17)((cid:15)(cid:11)(cid:4)(cid:18)(cid:1) (cid:24)(cid:15)(cid:1) ((cid:8)(cid:20)(cid:24)(cid:6)(cid:27)(cid:6)((cid:8)(cid:12)(cid:24)(cid:11)(cid:21)(cid:1) D(cid:20)(cid:15))(cid:12)(cid:18)E(cid:24)(cid:20))(cid:24)(cid:16)(cid:1) +<br/>(cid:6)(cid:12)(cid:25)(cid:15)(cid:20)(cid:17)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:14)(cid:1)(cid:6)(cid:12)(cid:27)(cid:5))(cid:18)(cid:6)(cid:12)-(cid:1)(cid:26)(cid:19)(cid:14)(cid:1)(cid:8)-(cid:4)(cid:14)(cid:1)(cid:7)(cid:6)(cid:12)(cid:18)(cid:1)(cid:15)(cid:25) ((cid:15)(cid:11)(cid:4)(cid:1)(cid:15)(cid:20)(cid:1)(cid:4)6((cid:20)(cid:4)(cid:11)(cid:11)(cid:6)(cid:15)(cid:12)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1) +<br/>(cid:6)(cid:25)(cid:1) (cid:24)(cid:16)(cid:4)(cid:1) (cid:11))(cid:10) (cid:4)(cid:27)(cid:24)(cid:1) (cid:16)(cid:8)(cid:11)(cid:1) -(cid:5)(cid:8)(cid:11)(cid:11)(cid:4)(cid:11)(cid:1) (cid:6)(cid:11)(cid:1) ((cid:20)(cid:15)(cid:29)(cid:6)(cid:18)(cid:4)(cid:18)(cid:21)(cid:1) #6((cid:4)(cid:20)(cid:6)(cid:17)(cid:4)(cid:12)(cid:24)(cid:8)(cid:5)(cid:1) (cid:11))(cid:10) (cid:4)(cid:27)(cid:24)(cid:11)(cid:1) +<br/>.(cid:4)(cid:20)(cid:4)(cid:1)((cid:16)(cid:15)(cid:24)(cid:15)-(cid:20)(cid:8)((cid:16)(cid:4)(cid:18)(cid:1).(cid:6)(cid:24)(cid:16)(cid:1)(cid:8)(cid:1)(cid:25)(cid:6)(cid:12)(cid:4)E(cid:20)(cid:4)(cid:11)(cid:15)(cid:5))(cid:24)(cid:6)(cid:15)(cid:12)(cid:1)(cid:27)(cid:15)(cid:5)(cid:15)(cid:20)(cid:1)(cid:18)(cid:6)-(cid:6)(cid:24)(cid:8)(cid:5)(cid:1)(cid:27)(cid:8)(cid:17)(cid:4)(cid:20)(cid:8)(cid:1) +<br/>(cid:6)(cid:12)(cid:1)(cid:18)(cid:8)(cid:30)(cid:5)(cid:6)-(cid:16)(cid:24)(cid:21)(cid:1)’(cid:16)(cid:4)(cid:1)(cid:11))(cid:10) (cid:4)(cid:27)(cid:24)(cid:11)(cid:1).(cid:4)(cid:20)(cid:4)(cid:1)(cid:11)(cid:4)(cid:8)(cid:24)(cid:4)(cid:18)(cid:1)(cid:15)(cid:12)(cid:1)(cid:8)(cid:1)(cid:11)(cid:24)(cid:15)(cid:15)(cid:5)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:6)(cid:12)(cid:11)(cid:24)(cid:20))(cid:27)(cid:24)(cid:4)(cid:18)(cid:1) +<br/>(cid:24)(cid:15)(cid:1) (cid:17)(cid:8)(cid:6)(cid:12)(cid:24)(cid:8)(cid:6)(cid:12)(cid:1) (cid:8)(cid:1) (cid:27)(cid:15)(cid:12)(cid:11)(cid:24)(cid:8)(cid:12)(cid:24)(cid:1) (cid:16)(cid:4)(cid:8)(cid:18)(cid:1) ((cid:15)(cid:11)(cid:6)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) 4(cid:8)(cid:5)(cid:24)(cid:16)(cid:15))-(cid:16)(cid:1) (cid:11)(cid:5)(cid:6)-(cid:16)(cid:24)(cid:1) +<br/>(cid:17)(cid:15)(cid:29)(cid:4)(cid:17)(cid:4)(cid:12)(cid:24)(cid:11)(cid:1).(cid:4)(cid:20)(cid:4)(cid:1))(cid:12)(cid:8)(cid:29)(cid:15)(cid:6)(cid:18)(cid:8)(cid:10)(cid:5)(cid:4)5(cid:21) +<br/>’(cid:16)(cid:4)(cid:1)(cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1)(cid:8)(cid:20)(cid:4)(cid:1)(cid:6)(cid:12)(cid:1)>0,G+>,(cid:1)((cid:6)6(cid:4)(cid:5)(cid:11)(cid:1)(cid:20)(cid:4)(cid:11)(cid:15)(cid:5))(cid:24)(cid:6)(cid:15)(cid:12)(cid:14)(cid:1)/>(cid:1)(cid:10)(cid:6)(cid:24)(cid:1)(cid:18)(cid:4)((cid:24)(cid:16) (cid:14)(cid:1) +<br/>(cid:8)(cid:10)(cid:15))(cid:24)(cid:1)>,(cid:1)(cid:31)(cid:10)(cid:30)(cid:24)(cid:4)(cid:11)(cid:1)(cid:11)(cid:6)(cid:22)(cid:4)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)CHD(cid:1)(cid:25)(cid:15)(cid:20)(cid:17)(cid:8)(cid:24) (cid:21)(cid:1) +<br/>#(cid:12)(cid:15))-(cid:16)(cid:1) (cid:5))(cid:17)(cid:6)(cid:12)(cid:15)(cid:11)(cid:6)(cid:24)(cid:30)(cid:1) (cid:25)(cid:15)(cid:20)(cid:1) .(cid:20)(cid:6)(cid:12)(cid:7)(cid:5)(cid:4)(cid:1) ((cid:20)(cid:15)(cid:27)(cid:4)(cid:11)(cid:11)(cid:6)(cid:12)-(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:25)(cid:8)(cid:27)(cid:6)(cid:8)(cid:5)(cid:1) +<br/>(cid:25)(cid:4)(cid:8)(cid:24))(cid:20)(cid:4)(cid:11)(cid:1) .(cid:6)(cid:24)(cid:16)(cid:15))(cid:24)(cid:1) (cid:11)(cid:16)(cid:8)(cid:18)(cid:15).(cid:11)(cid:1) (cid:6)(cid:11)(cid:1) (cid:12)(cid:4)(cid:4)(cid:18)(cid:4)(cid:18)(cid:1) 4(cid:6)(cid:12)(cid:1) (cid:8)-(cid:4)(cid:1) (cid:27)(cid:5)(cid:8)(cid:11)(cid:11)(cid:6)(cid:25)(cid:6)(cid:27)(cid:8)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) +<br/>.(cid:20)(cid:6)(cid:12)(cid:7)(cid:5)(cid:4)(cid:1) (cid:18)(cid:4)(cid:24)(cid:4)(cid:27)(cid:24)(cid:6)(cid:15)(cid:12)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:8)(cid:12)(cid:8)(cid:5)(cid:30)(cid:11)(cid:6)(cid:11)(cid:1) +<br/>(cid:24)(cid:16)(cid:4)(cid:1) +<br/>(cid:18)(cid:6)(cid:11)(cid:24)(cid:6)(cid:12)-)(cid:6)(cid:11)(cid:16)(cid:6)(cid:12)-(cid:1)(cid:15)(cid:25)(cid:1)(cid:11)(cid:4)(cid:12)(cid:6)(cid:15)(cid:20)(cid:11)(cid:1)(cid:25)(cid:20)(cid:15)(cid:17)(cid:1)(cid:24)(cid:16)(cid:15)(cid:11)(cid:4)(cid:1)(cid:6)(cid:12)(cid:1)(cid:24)(cid:16)(cid:4)(cid:1)(cid:30)(cid:15))(cid:12)-(cid:4)(cid:20)(cid:1)(cid:27)(cid:8)(cid:24)(cid:4)-(cid:15)(cid:20)(cid:6)(cid:4)(cid:11)(cid:1) +<br/><(cid:2)=5(cid:21)(cid:1) ")(cid:10) (cid:4)(cid:27)(cid:24)(cid:11)(cid:1) .(cid:4)(cid:20)(cid:4)(cid:1) ((cid:16)(cid:15)(cid:24)(cid:15)-(cid:20)(cid:8)((cid:16)(cid:4)(cid:18)(cid:1) .(cid:6)(cid:24)(cid:16)(cid:15))(cid:24)(cid:1) (cid:8)(cid:12)(cid:30)(cid:1) ((cid:20)(cid:15) (cid:4)(cid:27)(cid:24)(cid:15)(cid:20)(cid:11)(cid:1) (cid:15)(cid:20)(cid:1) +<br/>(cid:6)(cid:17)((cid:15)(cid:20)(cid:24)(cid:8)(cid:12)(cid:24)(cid:1) +<br/>(cid:25)(cid:15)(cid:20)(cid:1) +<br/>(cid:6)(cid:11)(cid:1) +</td><td></td><td>(cid:18)(cid:8)(cid:24)(cid:8)(cid:10)(cid:8)(cid:11)(cid:4)(cid:11)(cid:1) (cid:16)(cid:8)(cid:29)(cid:4)(cid:1) (cid:10)(cid:4)(cid:4)(cid:12)(cid:1) (cid:27)(cid:15)(cid:5)(cid:5)(cid:4)(cid:27)(cid:24)(cid:4)(cid:18)?(cid:1) (cid:11))(cid:27)(cid:16)(cid:1) (cid:8)(cid:11)(cid:1) (cid:9)!(cid:1) <+=(cid:14)(cid:1) (cid:23)(cid:9)(cid:13)7(cid:9)(cid:1) <@=(cid:14)(cid:1) +<br/>(cid:27)(cid:15)(cid:20)(cid:20)(cid:4)(cid:11)((cid:15)(cid:12)(cid:18)(cid:6)(cid:12)-(cid:1) (cid:24)(cid:15)(cid:1) +(cid:2)+(cid:1) ((cid:4)(cid:15)((cid:5)(cid:4)9(cid:11)(cid:1) (cid:25)(cid:8)(cid:27)(cid:4)(cid:11)(cid:1) 4>0@(cid:1) (cid:17)(cid:4)(cid:12)(cid:14)(cid:1) (cid:2)/B(cid:1) .(cid:15)(cid:17)(cid:4)(cid:12)5(cid:21)(cid:1) +</td></tr><tr><td>06560d5721ecc487a4d70905a485e22c9542a522</td><td>SUN, YU: DEEP FACIAL ATTRIBUTE DETECTION IN THE WILD +<br/>Deep Facial Attribute Detection in the Wild: +<br/>From General to Specific +<br/>Department of Automation +<br/><b>University of Science and Technology</b><br/>of China +<br/>Hefei, China +</td><td>('4364455', 'Yuechuan Sun', 'yuechuan sun')<br/>('1720236', 'Jun Yu', 'jun yu')</td><td>ycsun@mail.ustc.edu.cn +<br/>harryjun@ustc.edu.cn +</td></tr><tr><td>06526c52a999fdb0a9fd76e84f9795a69480cecf</td><td></td><td></td><td></td></tr><tr><td>06bad0cdda63e3fd054e7b334a5d8a46d8542817</td><td>Sharing Features Between Objects and Their Attributes +<br/>1Department of Computer Science +<br/><b>University of Texas at Austin</b><br/>2Computer Science Department +<br/><b>University of Southern California</b></td><td>('35788904', 'Sung Ju Hwang', 'sung ju hwang')<br/>('1693054', 'Fei Sha', 'fei sha')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>{sjhwang,grauman}@cs.utexas.edu +<br/>feisha@usc.edu +</td></tr><tr><td>06fe63b34fcc8ff68b72b5835c4245d3f9b8a016</td><td>Mach Learn +<br/>DOI 10.1007/s10994-013-5336-9 +<br/>Learning semantic representations of objects +<br/>and their parts +<br/>Received: 24 May 2012 / Accepted: 26 February 2013 +<br/>© The Author(s) 2013 +</td><td>('1935910', 'Grégoire Mesnil', 'grégoire mesnil')<br/>('1732280', 'Gal Chechik', 'gal chechik')</td><td></td></tr><tr><td>06aab105d55c88bd2baa058dc51fa54580746424</td><td>Image Set based Collaborative Representation for +<br/>Face Recognition +</td><td>('2873638', 'Pengfei Zhu', 'pengfei zhu')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('36685537', 'Lei Zhang', 'lei zhang')<br/>('1698371', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>0641dbee7202d07b6c78a39eecd312c17607412e</td><td>283 +<br/>978-1-4799-5751-4/14/$31.00 ©2014 IEEE +<br/>ICIP 2014 +<br/>WITH APPLICATIONS TO MOTION SEGMENTATION AND FACE CLUSTERING +<br/>NULL SPACE CLUSTERING +<br/><b>Australian National University, Canberra</b><br/>2NICTA, Canberra +</td><td>('2744345', 'Pan Ji', 'pan ji')<br/>('2015152', 'Yiran Zhong', 'yiran zhong')<br/>('40124570', 'Hongdong Li', 'hongdong li')<br/>('2862871', 'Mathieu Salzmann', 'mathieu salzmann')</td><td>fpan.ji,hongdong.lig@anu.edu.au,mathieu.salzmann@nicta.com.au +</td></tr><tr><td>06262d14323f9e499b7c6e2a3dec76ad9877ba04</td><td>Real-Time Pose Estimation Piggybacked on Object Detection +<br/>Brno, Czech Republic +</td><td>('1785162', 'Adam Herout', 'adam herout')</td><td>Graph@FIT, Brno University of Technology +<br/>ijuranek,herout,idubska,zemcik@fit.vutbr.cz +</td></tr><tr><td>062c41dad67bb68fefd9ff0c5c4d296e796004dc</td><td>Temporal Generative Adversarial Nets with Singular Value Clipping +<br/>Preferred Networks inc., Japan +</td><td>('49160719', 'Masaki Saito', 'masaki saito')<br/>('8252749', 'Eiichi Matsumoto', 'eiichi matsumoto')<br/>('3083107', 'Shunta Saito', 'shunta saito')</td><td>{msaito, matsumoto, shunta}@preferred.jp +</td></tr><tr><td>06400a24526dd9d131dfc1459fce5e5189b7baec</td><td>Event Recognition in Photo Collections with a Stopwatch HMM +<br/>1Computer Vision Lab +<br/>ETH Z¨urich, Switzerland +<br/>2ESAT, PSI-VISICS +<br/>K.U. Leuven, Belgium +</td><td>('1696393', 'Lukas Bossard', 'lukas bossard')<br/>('2737253', 'Matthieu Guillaumin', 'matthieu guillaumin')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>lastname@vision.ee.ethz.ch +<br/>vangool@esat.kuleuven.be +</td></tr><tr><td>062d67af7677db086ef35186dc936b4511f155d7</td><td>They Are Not Equally Reliable: Semantic Event Search +<br/>using Differentiated Concept Classifiers +<br/><b>Centre for Quantum Computation and Intelligent Systems, University of Technology Sydney</b><br/><b>Carnegie Mellon University</b></td><td>('1729163', 'Xiaojun Chang', 'xiaojun chang')<br/>('1698559', 'Yi Yang', 'yi yang')<br/>('1752601', 'Eric P. Xing', 'eric p. xing')</td><td>cxj273@gmail.com, yaoliang@cs.cmu.edu, yi.yang@uts.edu.au, epxing@cs.cmu.edu +</td></tr><tr><td>06c2086f7f72536bf970ca629151b16927104df3</td><td>PALMERO ET AL.: MULTI-MODAL RECURRENT CNN FOR 3D GAZE ESTIMATION +<br/>Recurrent CNN for 3D Gaze Estimation +<br/>using Appearance and Shape Cues +<br/>1 Dept. Mathematics and Informatics +<br/>Universitat de Barcelona, Spain +<br/>2 Computer Vision Center +<br/>Campus UAB, Bellaterra, Spain +<br/>3 Dept. Electrical and Computer Eng. +<br/><b>University of Calgary, Canada</b><br/>4 Dept. Engineering +<br/><b>University of Larestan, Iran</b></td><td>('3413560', 'Cristina Palmero', 'cristina palmero')<br/>('38081877', 'Javier Selva', 'javier selva')<br/>('1921285', 'Mohammad Ali Bagheri', 'mohammad ali bagheri')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')</td><td>crpalmec7@alumnes.ub.edu +<br/>javier.selva.castello@est.fib.upc.edu +<br/>mohammadali.bagheri@ucalgary.ca +<br/>sergio@maia.ub.es +</td></tr><tr><td>0694b05cbc3ef5d1c5069a4bfb932a5a7b4d5ff0</td><td>Iosifidis, A., Tefas, A., & Pitas, I. (2014). Exploiting Local Class Information +<br/>in Extreme Learning Machine. Paper presented at International Joint +<br/>Conference on Computational Intelligence (IJCCI), Rome, Italy. +<br/>Peer reviewed version +<br/>Link to publication record in Explore Bristol Research +<br/>PDF-document +<br/><b>University of Bristol - Explore Bristol Research</b><br/>General rights +<br/>This document is made available in accordance with publisher policies. Please cite only the published +<br/>version using the reference above. Full terms of use are available: +<br/>http://www.bristol.ac.uk/pure/about/ebr-terms +<br/> </td><td></td><td></td></tr><tr><td>060034b59275c13746413ca9c67d6304cba50da6</td><td>Ordered Trajectories for Large Scale Human Action Recognition +<br/>1Vision & Sensing, HCC Lab, +<br/><b>ESTeM, University of Canberra</b><br/>2IHCC, RSCS, CECS, +<br/><b>Australian National University</b></td><td>('1793720', 'O. V. Ramana Murthy', 'o. v. ramana murthy')<br/>('1717204', 'Roland Goecke', 'roland goecke')</td><td>O.V.RamanaMurthy@ieee.org +<br/>roland.goecke@ieee.org +</td></tr><tr><td>060820f110a72cbf02c14a6d1085bd6e1d994f6a</td><td>Fine-Grained Classification of Pedestrians in Video: Benchmark and State of the Art +<br/><b>California Institute of Technology</b><br/>The dataset was labelled with bounding boxes, tracks, pose and fine- +<br/>grained labels. To achieve this, crowdsourcing, using workers from Ama- +<br/>zon’s Mechanical Turk (MTURK) was used. A summary of the dataset’s +<br/>statistics can be found in Table 1. +<br/>Number of Frames Sent to MTURK +<br/>Number of Frames with at least 1 Pedestrian +<br/>Number of Bounding Box Labels +<br/>Number of Pose Labels +<br/>Number of Tracks +<br/>38,708 +<br/>20,994 +<br/>32,457 +<br/>27,454 +<br/>4,222 +<br/>Table 1: Dataset Statistics +<br/>A state-of-the-art algorithm for fine-grained classification was tested us- +<br/>ing the dataset. The results are reported as a useful performance baseline. +<br/>The dataset is split into a training/validation set containing 4 videos, with +<br/>the remaining 3 videos forming the test set. Since each video was collected +<br/>on a unique day, different images of the same person do not appear in both +<br/>the training and testing sets. +<br/>The fine-grained categorisation benchmark uses ’pose normalised deep +<br/>convolutional nets’ as proposed by Branson et al. [1]. In this framework, +<br/>features are extracted by applying deep convolutional nets to image re- +<br/>gions that are normalised by pose. It has state-of the-art performance on +<br/>bird species categorisation and we believe that it will generalise to the CRP +<br/>dataset. Results can be found in Figure 2 +<br/>Figure 2: Fine-grained classification results. We report the mean average +<br/>accuracy across 10 different train/test splits, for each of the subcategories +<br/>in CRP, using the method of [1]. Average accuracy is computed assuming +<br/>that there is a uniform prior across the classes. The reference value for +<br/>each subcategory corresponds to chance. The results suggest that CRP is a +<br/>challenging dataset. +<br/>A novel feature of our dataset is the occlusion labelling of the keypoints. +<br/>Exploiting this information may be the first step towards improving perfor- +<br/>mance for fine-grained classification. Using temporal information is another +<br/>alternative. Most pedestrians in CRP appear multiple times over large inter- +<br/>vals of time. We are planning on adding an identity label for each individ- +<br/>ual, to make our dataset useful for studying individual re-identification from +<br/>a moving camera. +<br/>Improved Bird Species Recognition Using Pose Normalized Deep Con- +<br/>volutional Nets. In BMVC, 2014. +<br/>Figure 1: Three examples from the CRP dataset. Annotations include a +<br/>bounding box, tracks, parts, occlusion, sex, age, weight and clothing style. +<br/>People are an important component of a machine’s environment. De- +<br/>tecting, tracking, and recognising people, interpreting their behaviour and +<br/>interacting with them is a valuable capability for machines. Using vision to +<br/>estimate human attributes such as: age, sex, activity, social status, health, +<br/>pose and motion patterns is useful for interpreting and predicting behaviour. +<br/>This motivates our interest in fine-grained categorisation of people. +<br/>In this work, we introduce a public video dataset—Caltech Roadside +<br/>Pedestrians (CRP)—to further advance the state-of-the-art in fine-grained +<br/>categorisation of people using the entire human body. This dataset is also +<br/>useful for benchmarking tracking, detection and pose estimation of pedes- +<br/>trians. +<br/>Its novel and distinctive features are: +<br/>1. Size (27,454 bounding box and pose labels) – making it suitable for +<br/>training deep-networks. +<br/>2. Natural behaviour – subjects are recorded “in-the-wild” so are un- +<br/>aware, and behave naturally. +<br/>3. Viewpoint – Pedestrians are viewed from front, profile, back and ev- +<br/>erything in between. +<br/>4. Moving camera – More general and challenging than surveillance +<br/>video with static background. +<br/>5. Realism – There is a variety of outdoor background and lighting con- +<br/>ditions +<br/>6. Multi-class subcategories – age, clothing style and body shape. +<br/>7. Detailed annotation – bounding boxes, tracks and 14 keypoints with +<br/>occlusion information; examples can be found in Figure 1. Each +<br/>bounding box is also labelled with the fine-grained categories of age +<br/>(5 classes), sex (2 classes), weight (3 classes) and clothing type (4 +<br/>classes). +<br/>8. Availability – All videos and annotations are publicly available +<br/>CRP contains seven, twenty-one minute videos. Each video is captured +<br/>by mounting a rightwards-pointing, GoPro Hero3 camera to the roof of a +<br/>car. The car then completes three laps of a ring road within a park where +<br/>there are many walkers and joggers. Each video was recorded on a different +<br/>day. +</td><td>('1990633', 'David Hall', 'david hall')<br/>('1690922', 'Pietro Perona', 'pietro perona')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td></td></tr><tr><td>0653dcdff992ad980cd5ea5bc557efb6e2a53ba1</td><td></td><td></td><td></td></tr><tr><td>063a3be18cc27ba825bdfb821772f9f59038c207</td><td>This is a repository copy of The development of spontaneous facial responses to others’ +<br/>emotions in infancy. An EMG study. +<br/>White Rose Research Online URL for this paper: +<br/>http://eprints.whiterose.ac.uk/125231/ +<br/>Version: Published Version +<br/>Article: +<br/>Kaiser, Jakob, Crespo-Llado, Maria Magdalena, Turati, Chiara et al. (1 more author) +<br/>(2017) The development of spontaneous facial responses to others’ emotions in infancy. +<br/>An EMG study. Scientific Reports. ISSN 2045-2322 +<br/>https://doi.org/10.1038/s41598-017-17556-y +<br/>Reuse +<br/>This article is distributed under the terms of the Creative Commons Attribution (CC BY) licence. This licence +<br/>allows you to distribute, remix, tweak, and build upon the work, even commercially, as long as you credit the +<br/>authors for the original work. More information and the full terms of the licence here: +<br/>https://creativecommons.org/licenses/ +<br/>Takedown +<br/>If you consider content in White Rose Research Online to be in breach of UK law, please notify us by +<br/>https://eprints.whiterose.ac.uk/ +</td><td></td><td>emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request. +<br/>eprints@whiterose.ac.uk +</td></tr><tr><td>064cd41d323441209ce1484a9bba02a22b625088</td><td>Selective Transfer Machine for Personalized Facial Action Unit Detection +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA</b><br/><b>University of Pittsburgh, Pittsburgh, PA</b></td><td>('39336289', 'Wen-Sheng Chu', 'wen-sheng chu')</td><td></td></tr><tr><td>06c2dfe1568266ad99368fc75edf79585e29095f</td><td>Bayesian Active Appearance Models +<br/><b>Imperial College London, United Kingdom</b></td><td>('2575567', 'Joan Alabort-i-Medina', 'joan alabort-i-medina')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>{ja310,s.zafeiriou}@imperial.ac.uk +</td></tr><tr><td>06f39834e870278243dda826658319be2d5d8ded</td><td>RECOGNIZING UNSEEN ACTIONS IN A DOMAIN-ADAPTED EMBEDDING SPACE +<br/><b>Arizona State University</b></td><td>('2180892', 'Yikang Li', 'yikang li')<br/>('8060096', 'Sheng-hung Hu', 'sheng-hung hu')<br/>('2913552', 'Baoxin Li', 'baoxin li')</td><td></td></tr><tr><td>06d7ef72fae1be206070b9119fb6b61ce4699587</td><td>On One-Shot Similarity Kernels: explicit feature maps and properties +<br/>†Department of Computing +<br/><b>Imperial College London</b><br/>,†,(cid:2) +<br/>∗Electronics Laboratory, Department of Physics, +<br/><b>University of Patras, Greece</b><br/>(cid:2)School of Science and Technology, +<br/><b>Middlesex University, London</b></td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1754270', 'Irene Kotsia', 'irene kotsia')</td><td>s.zafeiriou@imperial.ac.uk +</td></tr><tr><td>062d0813815c2b9864cd9bb4f5a1dc2c580e0d90</td><td>Encouraging LSTMs to Anticipate Actions Very Early +<br/><b>Australian National University, 2CVLab, EPFL, Switzerland, 3Smart Vision Systems, CSIRO</b></td><td>('2862871', 'Mathieu Salzmann', 'mathieu salzmann')<br/>('1688071', 'Basura Fernando', 'basura fernando')<br/>('2370776', 'Lars Petersson', 'lars petersson')<br/>('34234277', 'Lars Andersson', 'lars andersson')</td><td>firstname.lastname@data61.csiro.au, mathieu.salzmann@epfl.ch, basura.fernando@anu.edu.au +</td></tr><tr><td>06a9ed612c8da85cb0ebb17fbe87f5a137541603</td><td>Deep Learning of Player Trajectory Representations for Team +<br/>Activity Analysis +</td><td>('10386960', 'Nazanin Mehrasa', 'nazanin mehrasa')<br/>('19198359', 'Yatao Zhong', 'yatao zhong')<br/>('2123865', 'Frederick Tung', 'frederick tung')<br/>('3004771', 'Luke Bornn', 'luke bornn')<br/>('10771328', 'Greg Mori', 'greg mori')<br/>('2190580', 'Simon Fraser', 'simon fraser')</td><td>{nmehrasa, yataoz, ftung, lbornn}@sfu.ca, mori@cs.sfu.ca +</td></tr><tr><td>06ad99f19cf9cb4a40741a789e4acbf4433c19ae</td><td>SenTion: A framework for Sensing Facial +<br/>Expressions +</td><td>('31623038', 'Rahul Islam', 'rahul islam')<br/>('3451315', 'Karan Ahuja', 'karan ahuja')<br/>('1784438', 'Sandip Karmakar', 'sandip karmakar')</td><td>{rahul.islam, karan.ahuja, sandip, ferdous}@iiitg.ac.in +</td></tr><tr><td>6c27eccf8c4b22510395baf9f0d0acc3ee547862</td><td>Using CMU PIE Human Face Database to a +<br/>Convolutional Neural Network - Neocognitron +<br/><b></b><br/>Rodovia Washington Luis, Km 235, São Carlos – SP - Brazil +<br/><b>Systems and Telematics - Neurolab</b><br/>Via Opera Pia, 13 – I-16145 – Genoa - Italy +</td><td>('2231336', 'José Hiroki Saito', 'josé hiroki saito')<br/>('3261775', 'Marcelo Hirakuri', 'marcelo hirakuri')<br/>('2558289', 'André Saunite', 'andré saunite')<br/>('36243877', 'Alessandro Noriaki Ide', 'alessandro noriaki ide')<br/>('40209065', 'Sandra Abib', 'sandra abib')</td><td>{saito,hirakuri,sabib}@dc.ufscar.br, tiagocarvalho@uol.com.br, saunite@fai.com.br +<br/>noriaki@dist.unige.it +</td></tr><tr><td>6c66ae815e7e508e852ecb122fb796abbcda16a8</td><td>International Journal of Computer Science & Engineering Survey (IJCSES) Vol.6, No.5, October 2015 +<br/>A SURVEY OF THE TRENDS IN FACIAL AND +<br/>EXPRESSION RECOGNITION DATABASES AND +<br/>METHODS +<br/><b>University of Washington, Bothell, USA</b></td><td>('2971095', 'Sohini Roychowdhury', 'sohini roychowdhury')<br/>('33073434', 'Michelle Emmons', 'michelle emmons')</td><td></td></tr><tr><td>6ca2c5ff41e91c34696f84291a458d1312d15bf2</td><td>LIPNET: SENTENCE-LEVEL LIPREADING +<br/><b>University of Oxford, Oxford, UK</b><br/>Google DeepMind, London, UK 2 +<br/>CIFAR, Canada 3 +<br/>{yannis.assael,brendan.shillingford, +</td><td>('3365565', 'Yannis M. Assael', 'yannis m. assael')<br/>('3144580', 'Brendan Shillingford', 'brendan shillingford')<br/>('1766767', 'Shimon Whiteson', 'shimon whiteson')</td><td>shimon.whiteson,nando.de.freitas}@cs.ox.ac.uk +</td></tr><tr><td>6cefb70f4668ee6c0bf0c18ea36fd49dd60e8365</td><td>Privacy-Preserving Deep Inference for Rich User +<br/>Data on The Cloud +<br/><b>Sharif University of Technology</b><br/><b>Queen Mary University of London</b><br/><b>Nokia Bell Labs and University of Oxford</b></td><td>('9920557', 'Ali Shahin Shamsabadi', 'ali shahin shamsabadi')<br/>('2251846', 'Ali Taheri', 'ali taheri')<br/>('2226725', 'Kleomenis Katevas', 'kleomenis katevas')<br/>('1688652', 'Hamid R. Rabiee', 'hamid r. rabiee')<br/>('2772904', 'Nicholas D. Lane', 'nicholas d. lane')<br/>('1763096', 'Hamed Haddadi', 'hamed haddadi')</td><td></td></tr><tr><td>6c690af9701f35cd3c2f6c8d160b8891ad85822a</td><td>Multi-Task Learning with Low Rank Attribute Embedding for Person +<br/>Re-identification +<br/><b>Peking University</b><br/><b>University of Maryland College Park</b><br/><b>University of Texas at San Antonio</b></td><td>('20798990', 'Chi Su', 'chi su')<br/>('1752128', 'Fan Yang', 'fan yang')<br/>('1776581', 'Shiliang Zhang', 'shiliang zhang')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td></td></tr><tr><td>6c5fbf156ef9fc782be0089309074cc52617b868</td><td>Controllable Video Generation with Sparse Trajectories +<br/><b>Cornell University</b></td><td>('19235216', 'Zekun Hao', 'zekun hao')<br/>('47932904', 'Xun Huang', 'xun huang')<br/>('50172592', 'Serge Belongie', 'serge belongie')</td><td>{hz472,xh258,sjb344}@cornell.edu +</td></tr><tr><td>6c304f3b9c3a711a0cca5c62ce221fb098dccff0</td><td>Attentive Semantic Video Generation using Captions +<br/>IIT Hyderabad +<br/>IIT Hyderabad +</td><td>('8268761', 'Tanya Marwah', 'tanya marwah')<br/>('47351893', 'Gaurav Mittal', 'gaurav mittal')<br/>('1699429', 'Vineeth N. Balasubramanian', 'vineeth n. balasubramanian')</td><td>ee13b1044@iith.ac.in +<br/>gaurav.mittal.191013@gmail.com +<br/>vineethnb@iith.ac.in +</td></tr><tr><td>6ce23cf4f440021b7b05aa3c1c2700cc7560b557</td><td>Learning Local Convolutional Features for Face +<br/>Recognition with 2D-Warping +<br/>Human Language Technology and Pattern Recognition Group, +<br/><b>RWTH Aachen University</b></td><td>('1804963', 'Harald Hanselmann', 'harald hanselmann')<br/>('1685956', 'Hermann Ney', 'hermann ney')</td><td>surname@cs.rwth-aachen.de +</td></tr><tr><td>6c80c834d426f0bc4acd6355b1946b71b50cbc0b</td><td>Pose-Based Two-Stream Relational Networks for +<br/>Action Recognition in Videos +<br/>1Center for Research on Intelligent Perception and Computing (CRIPAC), +<br/>National Laboratory of Pattern Recognition (NLPR) +<br/>2Center for Excellence in Brain Science and Intelligence Technology (CEBSIT), +<br/><b>Institute of Automation, Chinese Academy of Sciences (CASIA</b><br/><b>University of Chinese Academy of Sciences (UCAS</b></td><td>('47824598', 'Wei Wang', 'wei wang')<br/>('47539600', 'Jinjin Zhang', 'jinjin zhang')<br/>('39927579', 'Chenyang Si', 'chenyang si')<br/>('1693997', 'Liang Wang', 'liang wang')</td><td>{wangwei, wangliang}@nlpr.ia.ac.cn, {jinjin.zhang, +<br/>chenyang.si}@cripac.ia.ac.cn +</td></tr><tr><td>6cb7648465ba7757ecc9c222ac1ab6402933d983</td><td>Visual Forecasting by Imitating Dynamics in Natural Sequences +<br/><b>Stanford University National Tsing Hua University</b></td><td>('32970572', 'Kuo-Hao Zeng', 'kuo-hao zeng')</td><td>{khzeng, bshen88, dahuang, jniebles}@cs.stanford.edu sunmin@ee.nthu.edu.tw +</td></tr><tr><td>6c2b392b32b2fd0fe364b20c496fcf869eac0a98</td><td>DOI 10.1007/s00138-012-0423-7 +<br/>ORIGINAL PAPER +<br/>Fully automatic face recognition framework based +<br/>on local and global features +<br/>Received: 30 May 2011 / Revised: 21 February 2012 / Accepted: 29 February 2012 / Published online: 22 March 2012 +<br/>© Springer-Verlag 2012 +</td><td>('36048866', 'Cong Geng', 'cong geng')</td><td></td></tr><tr><td>6c6bb85a08b0bdc50cf8f98408d790ccdb418798</td><td>Recognition of facial expressions in presence of +<br/>partial occlusion +<br/>AIIA Laboratory +<br/>Computer Vision and Image Processing Group +<br/>Department of Informatics +<br/><b>Aristotle University of Thessaloniki, GR-54124 Thessaloniki, Greece</b><br/>Phone: +30 2310 996361 +<br/>Fax: +30 2310 998453 +<br/>Web: http://poseidon.csd.auth.gr +</td><td>('2336758', 'Ioan Buciu', 'ioan buciu')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>E-mail: {nelu,ekotsia,pitas}@zeus.csd.auth.gr +</td></tr><tr><td>6c705285c554985ecfe1117e854e1fe1323f8c21</td><td>DIY Human Action Data Set Generation +<br/>Illya Zharkov +<br/><b>Simon Fraser University</b><br/>Microsoft +<br/>Microsoft +<br/>Microsoft +</td><td>('1916516', 'Mehran Khodabandeh', 'mehran khodabandeh')<br/>('3227254', 'Hamid Reza Vaezi Joze', 'hamid reza vaezi joze')<br/>('3811436', 'Vivek Pradeep', 'vivek pradeep')</td><td>mkhodaba@sfu.ca +<br/>hava@microsoft.com +<br/>zharkov@microsoft.com +<br/>vpradeep@microsoft.com +</td></tr><tr><td>6cddc7e24c0581c50adef92d01bb3c73d8b80b41</td><td>Face Verification Using the LARK +<br/>Representation +</td><td>('3326805', 'Hae Jong Seo', 'hae jong seo')<br/>('1718280', 'Peyman Milanfar', 'peyman milanfar')</td><td></td></tr><tr><td>6cfc337069868568148f65732c52cbcef963f79d</td><td>Audio-Visual Speaker Localization via Weighted +<br/>Clustering +<br/>To cite this version: +<br/>Localization via Weighted Clustering. IEEE Workshop on Machine Learning for Signal Processing, +<br/>Sep 2014, Reims, France. pp.1-6, 2014, <10.1109/MLSP.2014.6958874>. <hal-01053732> +<br/>HAL Id: hal-01053732 +<br/>https://hal.archives-ouvertes.fr/hal-01053732 +<br/>Submitted on 11 Aug 2014 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('1780201', 'Xavier Alameda-Pineda', 'xavier alameda-pineda')<br/>('1794229', 'Radu Horaud', 'radu horaud')<br/>('1785817', 'Florence Forbes', 'florence forbes')<br/>('1780201', 'Xavier Alameda-Pineda', 'xavier alameda-pineda')<br/>('1794229', 'Radu Horaud', 'radu horaud')<br/>('1785817', 'Florence Forbes', 'florence forbes')</td><td></td></tr><tr><td>6cd96f2b63c6b6f33f15c0ea366e6003f512a951</td><td>A New Approach in Solving Illumination and Facial Expression Problems +<br/>for Face Recognition +<br/><b>a The University of Nottingham Malaysia Campus</b><br/>Tel : 03-89248358, Fax : 03-89248017 +<br/>Jalan Broga +<br/>43500 Semenyih, Selangor +</td><td>('1968167', 'Yee Wan Wong', 'yee wan wong')<br/>('9273662', 'Kah Phooi Seng', 'kah phooi seng')<br/>('2808528', 'Li-Minn Ang', 'li-minn ang')</td><td>E-mail : yeewan.wong@nottingham.edu.my +</td></tr><tr><td>6c8c7065d1041146a3604cbe15c6207f486021ba</td><td>Attention Modeling for Face Recognition via Deep Learning +<br/>Department of Computing, Hung Hom, Kowloon +<br/>Hong Kong, 999077 CHINA +<br/>Department of Computing, Hung Hom, Kowloon +<br/>Hong Kong, 99907 CHINA +<br/>Department of Computing, Hung Hom, Kowloon +<br/>Hong Kong, 99907 CHINA +<br/>Department of Computing, Hung Hom, Kowloon +<br/>Hong Kong, 99907 CHINA +</td><td></td><td>Sheng-hua Zhong (csshzhong@comp.polyu.edu.hk) +<br/>Yan Liu (csyliu@comp.polyu.edu.hk) +<br/>Yao Zhang (csyaozhang@comp.polyu.edu.hk) +<br/>Fu-lai Chung (cskchung@comp.polyu.edu.hk) +</td></tr><tr><td>390f3d7cdf1ce127ecca65afa2e24c563e9db93b</td><td>Learning Deep Representation for Face +<br/>Alignment with Auxiliary Attributes +</td><td>('3152448', 'Zhanpeng Zhang', 'zhanpeng zhang')<br/>('1693209', 'Ping Luo', 'ping luo')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>39ed31ced75e6151dde41944a47b4bdf324f922b</td><td>Pose-Guided Photorealistic Face Rotation +<br/><b>CRIPAC and NLPR and CEBSIT, CASIA 2University of Chinese Academy of Sciences</b><br/>3Noah’s Ark Laboratory, Huawei Technologies Co., Ltd. +</td><td>('49995036', 'Yibo Hu', 'yibo hu')<br/>('47150161', 'Xiang Wu', 'xiang wu')<br/>('46806278', 'Bing Yu', 'bing yu')<br/>('50361927', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>{yibo.hu, xiang.wu}@cripac.ia.ac.cn, yubing5@huawei.com, {rhe, znsun}@nlpr.ia.ac.cn +</td></tr><tr><td>3918b425bb9259ddff9eca33e5d47bde46bd40aa</td><td>Copyright +<br/>by +<br/>David Lieh-Chiang Chen +<br/>2012 +</td><td></td><td></td></tr><tr><td>39ce143238ea1066edf0389d284208431b53b802</td><td></td><td></td><td></td></tr><tr><td>39ce2232452c0cd459e32a19c1abe2a2648d0c3f</td><td></td><td></td><td></td></tr><tr><td>3998c5aa6be58cce8cb65a64cb168864093a9a3e</td><td></td><td></td><td></td></tr><tr><td>39dc2ce4cce737e78010642048b6ed1b71e8ac2f</td><td>Recognition of Six Basic Facial Expressions by Feature-Points Tracking using +<br/>RBF Neural Network and Fuzzy Inference System +<br/><b>Islamic Azad University of AHAR</b><br/><b>Elect. Eng. Faculty, Tabriz University, Tabriz, Iran</b><br/> +</td><td>('3210269', 'Hadi Seyedarabi', 'hadi seyedarabi')<br/>('2488201', 'Ali Aghagolzadeh', 'ali aghagolzadeh')<br/>('1766050', 'Sohrab Khanmohammadi', 'sohrab khanmohammadi')</td><td>seyedarabi@tabrizu.ac.ir , aghagol@tabrizu.ac.ir , khan@tabrizu.ac.ir +</td></tr><tr><td>397aeaea61ecdaa005b09198942381a7a11cd129</td><td></td><td></td><td></td></tr><tr><td>3991223b1dc3b87883cec7af97cf56534178f74a</td><td>A Unified Framework for Context Assisted Face Clustering +<br/>Department of Computer Science +<br/><b>University of California, Irvine</b></td><td>('3338094', 'Liyan Zhang', 'liyan zhang')<br/>('1818681', 'Dmitri V. Kalashnikov', 'dmitri v. kalashnikov')<br/>('1686199', 'Sharad Mehrotra', 'sharad mehrotra')</td><td></td></tr><tr><td>39b22bcbd452d5fea02a9ee63a56c16400af2b83</td><td></td><td></td><td></td></tr><tr><td>399a2c23bd2592ebe20aa35a8ea37d07c14199da</td><td></td><td></td><td></td></tr><tr><td>396a19e29853f31736ca171a3f40c506ef418a9f</td><td>Real World Real-time Automatic Recognition of Facial Expressions +<br/><b>Exploratory Computer Vision Group, IBM T. J. Watson Research Center</b><br/>PO Box 704, Yorktown Heights, NY 10598 +</td><td>('8193125', 'Ying-li Tian', 'ying-li tian')<br/>('1773140', 'Ruud Bolle', 'ruud bolle')</td><td>{yltian,lisabr,arunh,sharat,aws,bolle}@us.ibm.com +</td></tr><tr><td>392d35bb359a3b61cca1360272a65690a97a2b3f</td><td>YAN, YAP, MORI: ONE-SHOT MULTI-TASK LEARNING FOR VIDEO EVENT DETECTION 1 +<br/>Multi-Task Transfer Methods to Improve +<br/>One-Shot Learning for Multimedia Event +<br/>Detection +<br/>School of Computing Science +<br/><b>Simon Fraser University</b><br/>Burnaby, BC, CANADA +</td><td>('34289418', 'Wang Yan', 'wang yan')<br/>('32874186', 'Jordan Yap', 'jordan yap')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td>wyan@sfu.ca +<br/>jjyap@sfu.ca +<br/>mori@cs.sfu.ca +</td></tr><tr><td>397085122a5cade71ef6c19f657c609f0a4f7473</td><td>GHIASI, FOWLKES: USING SEGMENTATION TO DETECT OCCLUSION +<br/>Using Segmentation to Predict the Absence +<br/>of Occluded Parts +<br/>Dept. of Computer Science +<br/><b>University of California</b><br/>Irvine, CA +</td><td>('1898210', 'Golnaz Ghiasi', 'golnaz ghiasi')<br/>('3157443', 'Charless C. Fowlkes', 'charless c. fowlkes')</td><td>gghiasi@ics.uci.edu +<br/>fowlkes@ics.uci.edu +</td></tr><tr><td>39c48309b930396a5a8903fdfe781d3e40d415d0</td><td>Learning Spatial and Temporal Cues for Multi-label Facial Action Unit Detection +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh PA</b><br/><b>University of Pittsburgh, Pittsburgh PA</b></td><td>('39336289', 'Wen-Sheng Chu', 'wen-sheng chu')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td></td></tr><tr><td>39c8b34c1b678235b60b648d0b11d241a34c8e32</td><td>Learning to Deblur Images with Exemplars +</td><td>('9416825', 'Jinshan Pan', 'jinshan pan')<br/>('2776845', 'Wenqi Ren', 'wenqi ren')<br/>('1786024', 'Zhe Hu', 'zhe hu')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>3986161c20c08fb4b9b791b57198b012519ea58b</td><td>International Journal of Soft Computing and Engineering (IJSCE) +<br/>ISSN: 2231-2307, Volume-4 Issue-4, September 2014 +<br/>An Efficient Method for Face Recognition based on +<br/>Fusion of Global and Local Feature Extraction +</td><td>('9218646', 'E. Gomathi', 'e. gomathi')<br/>('1873007', 'K. Baskaran', 'k. baskaran')</td><td></td></tr><tr><td>392425be1c9d9c2ee6da45de9df7bef0d278e85f</td><td></td><td></td><td></td></tr><tr><td>392c3cabe516c0108b478152902a9eee94f4c81e</td><td>Computer Science and Artificial Intelligence Laboratory +<br/>Technical Report +<br/>MIT-CSAIL-TR-2007-024 +<br/>April 23, 2007 +<br/>Tiny images +<br/>m a s s a c h u s e t t s i n s t i t u t e o f t e c h n o l o g y, c a m b r i d g e , m a 0 213 9 u s a — w w w. c s a i l . m i t . e d u +</td><td>('34943293', 'Antonio Torralba', 'antonio torralba')<br/>('2276554', 'Rob Fergus', 'rob fergus')<br/>('1768236', 'William T. Freeman', 'william t. freeman')</td><td></td></tr><tr><td>39f525f3a0475e6bbfbe781ae3a74aca5b401125</td><td>Deep Joint Face Hallucination and Recognition +<br/><b>Sun Yat-sen University</b><br/><b>Sun Yat-sen University</b><br/><b>Sun Yat-sen University</b><br/><b>Sun Yat-sen University</b><br/>November 28, 2016 +</td><td>('4080607', 'Junyu Wu', 'junyu wu')<br/>('2442939', 'Shengyong Ding', 'shengyong ding')<br/>('1723992', 'Wei Xu', 'wei xu')<br/>('38255852', 'Hongyang Chao', 'hongyang chao')</td><td>wujunyu2@mail2.sysu.edu.cn +<br/>1633615231@qq.com +<br/>xuwei1993@qq.com +<br/>isschhy@mail.sysu.edu.cn +</td></tr><tr><td>3946b8f862ecae64582ef0912ca2aa6d3f6f84dc</td><td>Who and Where: People and Location Co-Clustering +<br/>Electrical Engineering +<br/><b>Stanford University</b></td><td>('8491578', 'Zixuan Wang', 'zixuan wang')</td><td>zxwang@stanford.edu +</td></tr><tr><td>3933416f88c36023a0cba63940eb92f5cef8001a</td><td>Learning Robust Subspace Clustering +<br/>Department of Electrical and Computer Engineering +<br/><b>Duke University</b><br/>Durham, NC, 27708 +<br/>May 11, 2014 +</td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('1699339', 'Guillermo Sapiro', 'guillermo sapiro')</td><td>{qiang.qiu, guillermo.sapiro}@duke.edu +</td></tr><tr><td>39150acac6ce7fba56d54248f9c0badbfaeef0ea</td><td>Proceedings, Digital Signal Processing for in-Vehicle and mobile systems, Istanbul, Turkey, June 2007. +<br/><b>Sabanci University</b><br/>Faculty of +<br/>Engineering and Natural Sciences +<br/>Orhanli, Istanbul +</td><td>('40322754', 'Esra Vural', 'esra vural')<br/>('21691177', 'Mujdat Cetin', 'mujdat cetin')<br/>('31849282', 'Aytul Ercil', 'aytul ercil')<br/>('2724380', 'Gwen Littlewort', 'gwen littlewort')<br/>('1858421', 'Marian Bartlett', 'marian bartlett')<br/>('29794862', 'Javier Movellan', 'javier movellan')</td><td></td></tr><tr><td>3947b64dcac5bcc1d3c8e9dcb50558efbb8770f1</td><td></td><td></td><td></td></tr><tr><td>3965d61c4f3b72044f43609c808f8760af8781a2</td><td></td><td></td><td></td></tr><tr><td>39f03d1dfd94e6f06c1565d7d1bb14ab0eee03bc</td><td>Simultaneous Local Binary Feature Learning and Encoding for Face Recognition +<br/><b>Tsinghua University, Beijing, China</b><br/>2Rapid-Rich Object Search (ROSE) Lab, Interdisciplinary Graduate School, +<br/><b>Nanyang Technological University, Singapore</b></td><td>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('1754854', 'Venice Erin Liong', 'venice erin liong')<br/>('39491387', 'Jie Zhou', 'jie zhou')</td><td>elujiwen@gmail.com; veniceer001@e.ntu.edu.sg; jzhou@tsinghua.edu.cn +</td></tr><tr><td>395bf182983e0917f33b9701e385290b64e22f9a</td><td></td><td></td><td></td></tr><tr><td>3983637022992a329f1d721bed246ae76bc934f7</td><td>Wide-Baseline Stereo for Face Recognition with Large Pose Variation +<br/>Computer Science Department +<br/><b>University of Maryland, College Park</b></td><td>('38171682', 'Carlos D. Castillo', 'carlos d. castillo')<br/>('34734622', 'David W. Jacobs', 'david w. jacobs')</td><td>{carlos,djacobs}@cs.umd.edu +</td></tr><tr><td>3933e323653ff27e68c3458d245b47e3e37f52fd</td><td>Evaluation of a 3D-aided Pose Invariant 2D Face Recognition System +<br/>Computational Biomedicine Lab +<br/>4800 Calhoun Rd. Houston, TX, USA +</td><td>('26401746', 'Ha A. Le', 'ha a. le')<br/>('39634395', 'Pengfei Dou', 'pengfei dou')<br/>('2461369', 'Yuhang Wu', 'yuhang wu')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td>{xxu18, hale4, pdou, ywu35, ikakadia}@central.uh.edu +</td></tr><tr><td>39b452453bea9ce398613d8dd627984fd3a0d53c</td><td></td><td></td><td></td></tr><tr><td>3958db5769c927cfc2a9e4d1ee33ecfba86fe054</td><td>Describable Visual Attributes for +<br/>Face Verification and Image Search +</td><td>('40631426', 'Neeraj Kumar', 'neeraj kumar')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')<br/>('1750470', 'Shree K. Nayar', 'shree k. nayar')</td><td></td></tr><tr><td>39ecdbad173e45964ffe589b9ced9f1ebfe2d44e</td><td>Automatic Recognition of Lower Facial Action Units +<br/>Joint Research Group on Audio Visual Signal Processing (AVSP), +<br/><b>Vrije Universiteit Brussel</b><br/>Pleinlaan 2, 1050 Brussels +<br/>lower +<br/>recognize +</td><td>('1802474', 'Werner Verhelst', 'werner verhelst')<br/>('34068333', 'Isabel Gonzalez', 'isabel gonzalez')<br/>('1970907', 'Hichem Sahli', 'hichem sahli')</td><td>igonzale@etro.vub.ac.be +<br/>hichem.sahli@etro.vub.ac.be +<br/>wverhels@etro.vub.ac.be +</td></tr><tr><td>39b5f6d6f8d8127b2b97ea1a4987732c0db6f9df</td><td></td><td></td><td></td></tr><tr><td>99ced8f36d66dce20d121f3a29f52d8b27a1da6c</td><td>Organizing Multimedia Data in Video +<br/>Surveillance Systems Based on Face Verification +<br/>with Convolutional Neural Networks +<br/><b>National Research University Higher School of Economics, Nizhny Novgorod, Russian</b><br/>Federation +</td><td>('26376584', 'Anastasiia D. Sokolova', 'anastasiia d. sokolova')<br/>('26427828', 'Angelina S. Kharchevnikova', 'angelina s. kharchevnikova')<br/>('35153729', 'Andrey V. Savchenko', 'andrey v. savchenko')</td><td>adsokolova96@mail.ru +</td></tr><tr><td>994f7c469219ccce59c89badf93c0661aae34264</td><td>1 +<br/>Model Based Face Recognition Across Facial +<br/>Expressions +<br/> +<br/>screens, embedded into mobiles and installed into everyday +<br/>living and working environments they become valuable tools +<br/>for human system interaction. A particular important aspect of +<br/>this interaction is detection and recognition of faces and +<br/>interpretation of facial expressions. These capabilities are +<br/>deeply rooted in the human visual system and a crucial +<br/>building block for social interaction. Consequently, these +<br/>capabilities are an important step towards the acceptance of +<br/>many technical systems. +<br/>trees as a classifier +<br/>lies not only +</td><td>('1725709', 'Zahid Riaz', 'zahid riaz')<br/>('50565622', 'Christoph Mayer', 'christoph mayer')<br/>('32131501', 'Matthias Wimmer', 'matthias wimmer')<br/>('1699132', 'Bernd Radig', 'bernd radig')<br/>('31311898', 'Senior Member', 'senior member')</td><td></td></tr><tr><td>9949ac42f39aeb7534b3478a21a31bc37fe2ffe3</td><td>Parametric Stereo for Multi-Pose Face Recognition and +<br/>3D-Face Modeling +<br/>PSI ESAT-KUL +<br/>Leuven, Belgium +</td><td>('2733505', 'Rik Fransens', 'rik fransens')<br/>('2404667', 'Christoph Strecha', 'christoph strecha')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>999289b0ef76c4c6daa16a4f42df056bf3d68377</td><td>The Role of Color and Contrast in Facial Age Estimation +<br/><b>Intelligent Systems Lab Amsterdam, University of Amsterdam, The Netherlands</b><br/><b>Pattern Recognition and Bioinformatics Group, Delft University of Technology, The Netherlands</b><br/><b>Bo gazic i University, Istanbul, Turkey</b></td><td>('1695527', 'Theo Gevers', 'theo gevers')<br/>('1764521', 'Albert Ali Salah', 'albert ali salah')</td><td>{h.dibeklioglu,th.gevers,m.p.Lucassen}@uva.nl +<br/>salah@boun.edu.tr +</td></tr><tr><td>9958942a0b7832e0774708a832d8b7d1a5d287ae</td><td>The Sparse Matrix Transform for Covariance +<br/>Estimation and Analysis of High Dimensional +<br/>Signals +</td><td>('1696925', 'Guangzhi Cao', 'guangzhi cao')<br/>('1709256', 'Leonardo R. Bachega', 'leonardo r. bachega')<br/>('1745655', 'Charles A. Bouman', 'charles a. bouman')</td><td></td></tr><tr><td>995d55fdf5b6fe7fb630c93a424700d4bc566104</td><td>The One Triangle Three Parallelograms Sampling Strategy and Its Application +<br/>in Shape Regression +<br/>Centre of Mathematical Sciences +<br/><b>Lund University, Lund, Sweden</b></td><td>('38481779', 'Mikael Nilsson', 'mikael nilsson')</td><td>mikael.nilsson@math.lth.se +</td></tr><tr><td>99726ad232cef837f37914b63de70d8c5101f4e2</td><td>International Journal of Scientific & Engineering Research, Volume 5, Issue 5, May-2014 570 +<br/>ISSN 2229-5518 +<br/>Facial Expression Recognition Using PCA & Distance Classifier +<br/>Dept. of Electronics & Telecomm. Engg. +<br/> Ph.D Scholar,VSSUT +<br/>BURLA, ODISHA, INDIA +<br/>Nilamani Bhoi +<br/>Reader in Dept. of Electronics & Telecomm. Engg. +<br/><b>VEER SURENDRA SAI UNIVERSITY OF</b><br/>TECHNOLOGY +<br/>BURLA, ODISHA, INDIA +</td><td></td><td>alpesh.d123@gmail.com +<br/>nilamanib@gmail.com +</td></tr><tr><td>993d189548e8702b1cb0b02603ef02656802c92b</td><td>Highly-Economized Multi-View Binary +<br/>Compression for Scalable Image Clustering +<br/><b>Harbin Institute of Technology (Shenzhen), China</b><br/><b>The University of Queensland, Australia</b><br/><b>Inception Institute of Arti cial Intelligence, UAE</b><br/>4 Computer Vision Laboratory, ETH Zurich, Switzerland +<br/><b>University of Electronic Science and Technology of China, China</b></td><td>('38448016', 'Zheng Zhang', 'zheng zhang')<br/>('40241836', 'Li Liu', 'li liu')<br/>('1747229', 'Jie Qin', 'jie qin')<br/>('39986542', 'Fan Zhu', 'fan zhu')<br/>('2731972', 'Fumin Shen', 'fumin shen')<br/>('1725160', 'Yong Xu', 'yong xu')<br/>('40799321', 'Ling Shao', 'ling shao')<br/>('1724393', 'Heng Tao Shen', 'heng tao shen')</td><td></td></tr><tr><td>9931c6b050e723f5b2a189dd38c81322ac0511de</td><td></td><td></td><td></td></tr><tr><td>994b52bf884c71a28b4f5be4eda6baaacad1beee</td><td>Categorizing Big Video Data on the Web: +<br/>Challenges and Opportunities +<br/>School of Computer Science +<br/><b>Fudan University</b><br/>Shanghai, China +<br/>http://www.yugangjiang.info +</td><td>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')</td><td></td></tr><tr><td>99001ac9fdaf7649c0d0bd8d2078719bafd216d9</td><td>> TPAMI-0571-1005< +<br/>General Tensor Discriminant Analysis and +<br/>Gabor Features for Gait Recognition +<br/><b>School of Computer Science and Information Systems, Birkbeck College, University of London</b><br/><b>University of Vermont, 33 Colchester Avenue, Burlington</b><br/>Malet Street, London WC1E 7HX, United Kingdom. +<br/>Vermont 05405, United States of America. +</td><td>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('1720243', 'Xuelong Li', 'xuelong li')<br/>('1748808', 'Xindong Wu', 'xindong wu')<br/>('1740503', 'Stephen J. Maybank', 'stephen j. maybank')</td><td>{dacheng, xuelong, sjmaybank}@dcs.bbk.ac.uk; xwu@cs.uvm.edu. +</td></tr><tr><td>9993f1a7cfb5b0078f339b9a6bfa341da76a3168</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>A Simple, Fast and Highly-Accurate Algorithm to +<br/>Recover 3D Shape from 2D Landmarks on a Single +<br/>Image +</td><td>('39071836', 'Ruiqi Zhao', 'ruiqi zhao')<br/>('1678691', 'Yan Wang', 'yan wang')</td><td></td></tr><tr><td>9901f473aeea177a55e58bac8fd4f1b086e575a4</td><td>Human and Sheep Facial Landmarks Localisation +<br/>by Triplet Interpolated Features +<br/><b>University of Cambridge</b></td><td>('2966679', 'Heng Yang', 'heng yang')<br/>('2271111', 'Renqiao Zhang', 'renqiao zhang')<br/>('39626495', 'Peter Robinson', 'peter robinson')</td><td>hy306, rz264, pr10@cam.ac.uk +</td></tr><tr><td>992ebd81eb448d1eef846bfc416fc929beb7d28b</td><td>Exemplar-Based Face Parsing +<br/>Supplementary Material +<br/><b>University of Wisconsin Madison</b><br/>Adobe Research +<br/>http://www.cs.wisc.edu/~lizhang/projects/face-parsing/ +<br/>1. Additional Selected Results +<br/>Figures 1 and 2 supplement Figure 4 in our paper. In all cases, the input images come from our Helen [1] test set. We note +<br/>that our algorithm generally produces accurate results, as shown in Figures 1. However, our algorithm is not perfect and makes +<br/>mistakes on especially challenging input images, as shown in Figure 2. +<br/>In our view, the mouth is the most challenging region of the face to segment: the shape and appearance of the lips vary +<br/>widely from subject to subject, mouths deform significantly, and the overall appearance of the mouth region changes depending +<br/>on whether the inside of the mouth is visible or not. Unusual mouth expressions, like those shown in Figure 2, are not repre- +<br/>sented well in the exemplar images, which results in poor label transfer from the top exemplars to the test image. Despite these +<br/>challenges, our algorithm generally performs well on the mouth, with large segmentation errors occurring infrequently. +<br/>2. Comparisons with Liu et al. [2] +<br/>The scene parsing approach by Liu et al. [2] shares sevaral similarities with our work. Like our approach, they propose a +<br/>nonparametric system that transfers labels from exemplars in a database to annotate a test image. This begs the question, Why +<br/>not simply apply the approach from Liu et al. to face images? +<br/>To help answer this question, we used the code provided by Liu et al. on our Helen [1] images; our exemplar set is used for +<br/>training their system, and our test set is used for testing. Please see Section 4.3 in our paper for more details. Figure 3 shows +<br/>several selected results for qualitative comparison. In general, our algorithm performs much better than Liu et al.’s algorithm. +<br/>References +<br/>[1] V. Le, J. Brandt, Z. Lin, L. Bourdev, and T. S. Huang. Interactive facial feature localization. In ECCV, 2012. +<br/>[2] C. Liu, J. Yuen, and A. Torralba. Nonparametric scene parsing via label transfer. In PAMI, December 2011. +</td><td>('2721523', 'Brandon M. Smith', 'brandon m. smith')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')</td><td></td></tr><tr><td>99c20eb5433ed27e70881d026d1dbe378a12b342</td><td>ISCA Archive +<br/>http://www.isca-speech.org/archive +<br/>First Workshop on Speech, Language +<br/>and Audio in Multimedia +<br/>Marseille, France +<br/>August 22-23, 2013 +<br/>Proceedings of the First Workshop on Speech, Language and Audio in Multimedia (SLAM), Marseille, France, August 22-23, 2013. +<br/>78 +</td><td></td><td></td></tr><tr><td>99facca6fc50cc30f13b7b6dd49ace24bc94f702</td><td>Front.Comput.Sci. +<br/>DOI +<br/>RESEARCH ARTICLE +<br/>VIPLFaceNet: An Open Source Deep Face Recognition SDK +<br/>1 Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b><br/>c(cid:13) Higher Education Press and Springer-Verlag Berlin Heidelberg 2016 +</td><td>('46522348', 'Xin Liu', 'xin liu')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('3468240', 'Wanglong Wu', 'wanglong wu')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>9990e0b05f34b586ffccdc89de2f8b0e5d427067</td><td>International Journal of Modeling and Optimization, Vol. 3, No. 2, April 2013 +<br/>Auto-Optimized Multimodal Expression Recognition +<br/>Framework Using 3D Kinect Data for ASD Therapeutic +<br/>Aid +<br/> +<br/>regarding +<br/>emotion +<br/>and +<br/>to +<br/>recognize +</td><td>('25833279', 'Amira E. Youssef', 'amira e. youssef')<br/>('1720250', 'Ahmed S. Ibrahim', 'ahmed s. ibrahim')<br/>('1731164', 'A. Lynn Abbott', 'a. lynn abbott')</td><td></td></tr><tr><td>99d7678039ad96ee29ab520ff114bb8021222a91</td><td>Political image analysis with deep neural +<br/>networks +<br/>November 28, 2017 +</td><td>('41096358', 'L. Jason Anastasopoulos', 'l. jason anastasopoulos')<br/>('2361255', 'Shiry Ginosar', 'shiry ginosar')<br/>('2007721', 'Dhruvil Badani', 'dhruvil badani')<br/>('2459453', 'Jake Ryland Williams', 'jake ryland williams')<br/>('50521070', 'Crystal Lee', 'crystal lee')</td><td></td></tr><tr><td>52012b4ecb78f6b4b9ea496be98bcfe0944353cd</td><td> +<br/> JOURNAL OF COMPUTATION IN BIOSCIENCES AND ENGINEERING +<br/> +<br/> Journal homepage: http://scienceq.org/Journals/JCLS.php +<br/> +<br/>Research Article +<br/>Using Support Vector Machine and Local Binary Pattern for Facial Expression +<br/>Recognition +<br/>Open Access +<br/><b>Federal University Technology Akure, PMB 704, Akure, Nigeria</b><br/>2. Department of computer science, Kwara state polytechnic Ilorin, Kwara-State, Nigeria. +<br/> Received: September 22, 2015, Accepted: December 14, 2015, Published: December 14, 2015. +</td><td>('10698338', 'Alese Boniface Kayode', 'alese boniface kayode')</td><td>. *Corresponding author: Ayeni Olaniyi Abiodun Mail Id: oaayeni@futa.edu.ng +</td></tr><tr><td>523854a7d8755e944bd50217c14481fe1329a969</td><td>A Differentially Private Kernel Two-Sample Test +<br/>MPI-IS +<br/><b>University Of Oxford</b><br/><b>University Of Oxford</b><br/>MPI-IS +<br/>April 17, 2018 +</td><td>('39565862', 'Anant Raj', 'anant raj')<br/>('35142231', 'Ho Chung Leon Law', 'ho chung leon law')<br/>('1698032', 'Dino Sejdinovic', 'dino sejdinovic')<br/>('37292171', 'Mijung Park', 'mijung park')</td><td>anant.raj@tuebingen.mpg.de +<br/>ho.law@stats.ox.ac.uk +<br/>dino.sejdinovic@stats.ox.ac.uk +<br/>mijung.park@tuebingen.mpg.de +</td></tr><tr><td>521cfbc1949289a7ffc3ff90af7c55adeb43db2a</td><td>Action Recognition with Coarse-to-Fine Deep Feature Integration and +<br/>Asynchronous Fusion +<br/><b>Shanghai Jiao Tong University, China</b><br/><b>National Key Laboratory for Novel Software Technology, Nanjing University, China</b><br/><b>University of Chinese Academy of Sciences, China</b></td><td>('8131625', 'Weiyao Lin', 'weiyao lin')<br/>('1926641', 'Yang Mi', 'yang mi')<br/>('1808816', 'Jianxin Wu', 'jianxin wu')<br/>('1875882', 'Ke Lu', 'ke lu')<br/>('37028145', 'Hongkai Xiong', 'hongkai xiong')</td><td>{wylin, deyangmiyang, xionghongkai}@sjtu.edu.cn, wujx2001@nju.edu.cn, luk@ucas.ac.cn +</td></tr><tr><td>529e2ce6fb362bfce02d6d9a9e5de635bde81191</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. +<br/>> TIP-05732-2009< +<br/>1 +<br/>Normalization of Face Illumination Based +<br/>on Large- and Small- Scale Features +</td><td>('2002129', 'Xiaohua Xie', 'xiaohua xie')<br/>('3333315', 'Wei-Shi Zheng', 'wei-shi zheng')<br/>('1768574', 'Pong C. Yuen', 'pong c. yuen')<br/>('1713795', 'Ching Y. Suen', 'ching y. suen')</td><td></td></tr><tr><td>52887969107956d59e1218abb84a1f834a314578</td><td>1283 +<br/>Travel Recommendation by Mining People +<br/>Attributes and Travel Group Types From +<br/>Community-Contributed Photos +</td><td>('35081710', 'Yan-Ying Chen', 'yan-ying chen')<br/>('2363522', 'An-Jung Cheng', 'an-jung cheng')<br/>('1716836', 'Winston H. Hsu', 'winston h. hsu')</td><td></td></tr><tr><td>521482c2089c62a59996425603d8264832998403</td><td></td><td></td><td></td></tr><tr><td>521b625eebea73b5deb171a350e3709a4910eebf</td><td></td><td></td><td></td></tr><tr><td>52258ec5ec73ce30ca8bc215539c017d279517cf</td><td>Recognizing Faces with Expressions: Within-class Space and Between-class Space +<br/><b>Zhejang University, Hangzhou 310027, P.R.China</b><br/>Yu Bing Chen Ping Jin Lianfu +</td><td></td><td>Email: BingbingYu@21cn.com Pchen@mail.hz.zj.cn Lfjin@mail.hz.zj.cn +</td></tr><tr><td>5253c94f955146ba7d3566196e49fe2edea1c8f4</td><td>Internet-based Morphable Model +<br/><b>University of Washington</b><br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/>! +<br/>! +<br/> +<br/> +<br/>Figure 1. Overview of the method. We construct a morphable +<br/>model directly from Internet photos, the model is then used for +<br/>single view reconstruction from any new input image (Face An- +<br/>alyzer) and further for shape modification (Face Modifier), e.g., +<br/>from neutral to smile in 3D. +</td><td>('2419955', 'Ira Kemelmacher-Shlizerman', 'ira kemelmacher-shlizerman')</td><td>kemelmi@cs.washington.edu +</td></tr><tr><td>527dda77a3864d88b35e017d542cb612f275a4ec</td><td></td><td></td><td></td></tr><tr><td>529b1f33aed49dbe025a99ac1d211c777ad881ec</td><td>FAST AND EXACT BI-DIRECTIONAL FITTING OF ACTIVE APPEARANCE MODELS +<br/>Jean Kossaifi(cid:63) +<br/><b>cid:63) Imperial College London, UK</b><br/><b>University of Nottingham, UK, School of Computer Science</b><br/><b>University of Twente, The Netherlands</b></td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>523b2cbc48decfabffb66ecaeced4fe6a6f2ac78</td><td>Photorealistic Facial Expression Synthesis by the Conditional Difference Adversarial +<br/>Autoencoder +<br/>Department of Electronic and Computer Engineering +<br/><b>The Hong Kong University of Science and Technology</b><br/>HKSAR, China +</td><td>('1698743', 'Yuqian Zhou', 'yuqian zhou')</td><td>yzhouas@ust.hk, eebert@ust.hk +</td></tr><tr><td>52472ec859131844f38fc7d57944778f01d109ac</td><td>Improving speaker turn embedding by +<br/>crossmodal transfer learning from face embedding +<br/><b>Idiap Research Institute, Martigny, Switzerland</b><br/>2 ´Ecole Polytechnique F´ed´eral de Lausanne, Switzerland +</td><td>('39560344', 'Nam Le', 'nam le')<br/>('1719610', 'Jean-Marc Odobez', 'jean-marc odobez')</td><td>{nle, odobez}@idiap.ch +</td></tr><tr><td>5287d8fef49b80b8d500583c07e935c7f9798933</td><td>Generative Adversarial Text to Image Synthesis +<br/><b>University of Michigan, Ann Arbor, MI, USA (UMICH.EDU</b><br/><b>Max Planck Institute for Informatics, Saarbr ucken, Germany (MPI-INF.MPG.DE</b><br/>REEDSCOT1, AKATA2, XCYAN1, LLAJAN1 +<br/>HONGLAK1, SCHIELE2 +</td><td>('2893664', 'Zeynep Akata', 'zeynep akata')<br/>('3084614', 'Xinchen Yan', 'xinchen yan')<br/>('2876316', 'Lajanugen Logeswaran', 'lajanugen logeswaran')<br/>('1697141', 'Honglak Lee', 'honglak lee')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td></td></tr><tr><td>52c59f9f4993c8248dd3d2d28a4946f1068bcbbe</td><td>Structural Similarity and Distance in Learning +<br/>Dept. of Electrical and +<br/>Computer Engineering +<br/><b>Boston University</b><br/>Boston, MA 02215 +<br/>Dept. of Electrical and +<br/>Computer Engineering +<br/><b>Boston University</b><br/>Boston, MA 02215 +<br/>David A. Casta˜n´on +<br/>Dept. of Electrical and +<br/>Computer Engineering +<br/><b>Boston University</b><br/>Boston, MA 02215 +<br/>information, +</td><td>('1928419', 'Joseph Wang', 'joseph wang')<br/>('1699322', 'Venkatesh Saligrama', 'venkatesh saligrama')</td><td>Email: joewang@bu.edu +<br/>Email: srv@bu.edu +<br/>Email: dac@bu.edu +</td></tr><tr><td>52bf00df3b970e017e4e2f8079202460f1c0e1bd</td><td>Learning High-level Prior with Convolutional Neural Networks +<br/>for Semantic Segmentation +<br/><b>University of Science and Technology of China</b><br/>Hefei, China +<br/><b>Tsinghua University</b><br/>Beijing, China +<br/><b>The Hong Kong University of Science and Technology</b><br/>HongKong, China +</td><td>('2743695', 'Haitian Zheng', 'haitian zheng')<br/>('1697194', 'Feng Wu', 'feng wu')<br/>('39987643', 'Lu Fang', 'lu fang')<br/>('1680777', 'Yebin Liu', 'yebin liu')<br/>('1916870', 'Mengqi Ji', 'mengqi ji')</td><td>{zhenght,fengwu,fanglu}@mail.ustc.edu.cn +<br/>liuyebin@mail.tsinghua.edu.cn +<br/>mji@ust.hk +</td></tr><tr><td>52c91fcf996af72d191520d659af44e310f86ef9</td><td>Interactive Image Search with Attribute-based Guidance and Personalization +<br/><b>The University of Texas at Austin</b></td><td>('1770205', 'Adriana Kovashka', 'adriana kovashka')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>{adriana, grauman}@cs.utexas.edu +</td></tr><tr><td>52885fa403efbab5ef21274282edd98b9ca70cbf</td><td>Discriminant Graph Structures for Facial +<br/>Expression Recognition +<br/><b>Aristotle University of Thessaloniki</b><br/>Department of Informatics +<br/>Box 451 +<br/>54124 Thessaloniki, Greece +<br/>Address for correspondence : +<br/><b>Aristotle University of Thessaloniki</b><br/>54124 Thessaloniki +<br/>GREECE +<br/>Tel. ++ 30 231 099 63 04 +<br/>Fax ++ 30 231 099 63 04 +<br/>April 2, 2008 +<br/>DRAFT +</td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>email: pitas@zeus.csd.auth.gr +</td></tr><tr><td>52f23e1a386c87b0dab8bfdf9694c781cd0a3984</td><td></td><td></td><td></td></tr><tr><td>52d7eb0fbc3522434c13cc247549f74bb9609c5d</td><td>WIDER FACE: A Face Detection Benchmark +<br/><b>The Chinese University of Hong Kong</b></td><td>('1692609', 'Shuo Yang', 'shuo yang')<br/>('47571885', 'Ping Luo', 'ping luo')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{ys014, pluo, ccloy, xtang}@ie.cuhk,edu.hk +</td></tr><tr><td>528069963f0bd0861f380f53270c96c269a3ea1c</td><td><b>Cardi University</b><br/>School of Computer Science and Informatics +<br/>Visual Computing Group +<br/>4D (3D Dynamic) Statistical Models of +<br/>Conversational Expressions and the +<br/>Synthesis of Highly-Realistic 4D Facial +<br/>Expression Sequences +<br/>Submitted in part fulfilment of the requirements for the degree of +<br/><b>Doctor of Philosophy in Computer Science at Cardi University, July 24th</b></td><td>('1812779', 'Jason Vandeventer', 'jason vandeventer')</td><td></td></tr><tr><td>529baf1a79cca813f8c9966ceaa9b3e42748c058</td><td>Triangle Wise Mapping Technique to Transform one Face Image into Another Face Image +<br/> +<br/>{tag} {/tag} +<br/> +<br/> International Journal of Computer Applications +<br/> +<br/> © 2014 by IJCA Journal +<br/> Volume 87 - Number 6 +<br/> +<br/> Year of Publication: 2014 +<br/> +<br/> +<br/> +<br/> Authors: +<br/> +<br/>Bhogeswar Borah +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> 10.5120/15209-3714 +<br/> {bibtex}pxc3893714.bib{/bibtex} +</td><td></td><td></td></tr><tr><td>5239001571bc64de3e61be0be8985860f08d7e7e</td><td>SUBMITTED TO IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, JUNE 2016 +<br/>Deep Appearance Models: A Deep Boltzmann +<br/>Machine Approach for Face Modeling +</td><td>('1876581', 'Chi Nhan Duong', 'chi nhan duong')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('2687827', 'Kha Gia Quach', 'kha gia quach')<br/>('1699922', 'Tien D. Bui', 'tien d. bui')</td><td></td></tr><tr><td>556b9aaf1bc15c928718bc46322d70c691111158</td><td>Exploiting Qualitative Domain Knowledge for Learning Bayesian +<br/>Network Parameters with Incomplete Data +<br/>Thomson-Reuters Corporation +<br/><b>Rensselaer Polytechnic Institute</b></td><td>('2460793', 'Wenhui Liao', 'wenhui liao')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>wenhui.liao@thomsonreuters.com +<br/>qji@ecse.rpi.edu +</td></tr><tr><td>55ea0c775b25d9d04b5886e322db852e86a556cd</td><td>DOCK: Detecting Objects +<br/>by transferring Common-sense Knowledge +<br/><b>University of California, Davis 2University of Washington 3Allen Institute for AI</b><br/>https://dock-project.github.io +</td><td>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('19553871', 'Krishna Kumar Singh', 'krishna kumar singh')<br/>('1883898', 'Yong Jae Lee', 'yong jae lee')</td><td></td></tr><tr><td>550858b7f5efaca2ebed8f3969cb89017bdb739f</td><td></td><td></td><td></td></tr><tr><td>554b9478fd285f2317214396e0ccd81309963efd</td><td>Spatio-Temporal Action Localization For Human Action +<br/>Recognition in Large Dataset +<br/>1L2TI, Institut Galil´ee, Universit´e Paris 13, France; +<br/>2SERCOM, Ecole Polytechnique de Tunisie +</td><td>('3240115', 'Sameh MEGRHI', 'sameh megrhi')<br/>('2504338', 'Marwa JMAL', 'marwa jmal')<br/>('1731553', 'Azeddine BEGHDADI', 'azeddine beghdadi')<br/>('14521102', 'Wided Mseddi', 'wided mseddi')</td><td></td></tr><tr><td>55c68c1237166679d2cb65f266f496d1ecd4bec6</td><td>Learning to Score Figure Skating Sport Videos +</td><td>('2708397', 'Chengming Xu', 'chengming xu')<br/>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('10110775', 'Zitian Chen', 'zitian chen')<br/>('40379722', 'Bing Zhang', 'bing zhang')<br/>('1717861', 'Yu-Gang Jiang', 'yu-gang jiang')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')</td><td></td></tr><tr><td>558fc9a2bce3d3993a9c1f41b6c7f290cefcf92f</td><td>DEPARTMENT OF INFORMATION ENGINEERING AND COMPUTER SCIENCE +<br/>ICT International Doctoral School +<br/>Efficient and Effective Solutions +<br/>for Video Classification +<br/>Advisor: +<br/>Prof. Nicu Sebe +<br/><b>University of Trento</b><br/>Co-Advisor: +<br/>Prof. Bogdan Ionescu +<br/><b>University Politehnica of Bucharest</b><br/>November 2017 +</td><td>('28957796', 'Ionut Cosmin Duta', 'ionut cosmin duta')</td><td></td></tr><tr><td>55138c2b127ebdcc508503112bf1d1eeb5395604</td><td>Ensemble Nystr¨om Method +<br/>Google Research +<br/>New York, NY +<br/><b>Courant Institute and Google Research</b><br/>New York, NY +<br/><b>Courant Institute of Mathematical Sciences</b><br/>New York, NY +</td><td>('2794322', 'Sanjiv Kumar', 'sanjiv kumar')<br/>('1709415', 'Mehryar Mohri', 'mehryar mohri')<br/>('8395559', 'Ameet Talwalkar', 'ameet talwalkar')</td><td>sanjivk@google.com +<br/>mohri@cs.nyu.edu +<br/>ameet@cs.nyu.edu +</td></tr><tr><td>5502dfe47ac26e60e0fb25fc0f810cae6f5173c0</td><td>Affordance Prediction via Learned Object Attributes +</td><td>('2749326', 'Tucker Hermans', 'tucker hermans')<br/>('1692956', 'James M. Rehg', 'james m. rehg')<br/>('1688328', 'Aaron Bobick', 'aaron bobick')</td><td></td></tr><tr><td>55e18e0dde592258882134d2dceeb86122b366ab</td><td>Journal of Artificial Intelligence Research 37 (2010) 397-435 +<br/>Submitted 11/09; published 03/10 +<br/>Training a Multilingual Sportscaster: +<br/>Using Perceptual Context to Learn Language +<br/>Department of Computer Science +<br/><b>The University of Texas at Austin</b><br/><b>University Station C0500, Austin TX 78712, USA</b></td><td>('39230960', 'David L. Chen', 'david l. chen')<br/>('1765656', 'Joohyun Kim', 'joohyun kim')<br/>('1797655', 'Raymond J. Mooney', 'raymond j. mooney')</td><td>DLCC@CS.UTEXAS.EDU +<br/>SCIMITAR@CS.UTEXAS.EDU +<br/>MOONEY@CS.UTEXAS.EDU +</td></tr><tr><td>55a158f4e7c38fe281d06ae45eb456e05516af50</td><td>The 22nd International Conference on Computer Graphics and Vision +<br/>108 +<br/>GraphiCon’2012 +</td><td></td><td></td></tr><tr><td>5506a1a1e1255353fde05d9188cb2adc20553af5</td><td></td><td></td><td></td></tr><tr><td>55966926e7c28b1eee1c7eb7a0b11b10605a1af0</td><td>Surpassing Human-Level Face Verification Performance on LFW with +<br/>GaussianFace +<br/><b>The Chinese University of Hong Kong</b></td><td>('2312486', 'Chaochao Lu', 'chaochao lu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{lc013, xtang}@ie.cuhk.edu.hk +</td></tr><tr><td>552c55c71bccfc6de7ce1343a1cd12208e9a63b3</td><td>Accurate Eye Center Location and Tracking Using Isophote Curvature +<br/>Intelligent Systems Lab Amsterdam +<br/><b>University of Amsterdam, The Netherlands</b></td><td>('9301018', 'Roberto Valenti', 'roberto valenti')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td>{rvalenti,gevers}@science.uva.nl +</td></tr><tr><td>5517b28795d7a68777c9f3b2b46845dcdb425b2c</td><td>Deep video gesture recognition using illumination invariants +<br/><b>Massachusetts Institute of Technology</b><br/>Figure 1: Automated facial gesture recognition is a fundamental problem in human computer interaction. While tackling real world tasks of +<br/>expression recognition sudden changes in illumination from multiple sources can be expected. We show how to build a robust system to detect +<br/>human emotions while showing invariance to illumination. +</td><td>('37381309', 'Otkrist Gupta', 'otkrist gupta')<br/>('2283049', 'Dan Raviv', 'dan raviv')<br/>('1717566', 'Ramesh Raskar', 'ramesh raskar')</td><td></td></tr><tr><td>55c81f15c89dc8f6eedab124ba4ccab18cf38327</td><td></td><td></td><td></td></tr><tr><td>5550a6df1b118a80c00a2459bae216a7e8e3966c</td><td>ISSN: 0974-2115 +<br/>www.jchps.com Journal of Chemical and Pharmaceutical Sciences +<br/>A perusal on Facial Emotion Recognition System (FERS) +<br/><b>School of Information Technology and Engineering, VIT University, Vellore, 632014, India</b></td><td></td><td>*Corresponding author: E-Mail: krithika.lb@vit.ac.in +</td></tr><tr><td>55e87050b998eb0a8f0b16163ef5a28f984b01fa</td><td>CAN YOU FIND A FACE IN A HEVC BITSTREAM? +<br/><b>School of Engineering Science, Simon Fraser University, Burnaby, BC, Canada</b></td><td>('3393216', 'Saeed Ranjbar Alvar', 'saeed ranjbar alvar')<br/>('3320198', 'Hyomin Choi', 'hyomin choi')</td><td></td></tr><tr><td>55bc7abcef8266d76667896bbc652d081d00f797</td><td>Impact of Facial Cosmetics on Automatic Gender and Age Estimation +<br/>Algorithms +<br/><b>Computer Science and Electrical Engineering, West Virginia University, Morgantown, USA</b><br/><b>Computer Science and Engineering, Michigan State University, East Lansing, USA</b><br/>Keywords: +<br/>Biometrics, Face Recognition, Facial Cosmetics, Makeup, Gender Spoofing, Age Alteration, Automatic +<br/>Gender Estimation, Automatic Age Estimation +</td><td>('1751335', 'Cunjian Chen', 'cunjian chen')<br/>('3299530', 'Antitza Dantcheva', 'antitza dantcheva')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td>cchen10@csee.wvu.edu, {antitza, rossarun}@msu.edu +</td></tr><tr><td>55b4b1168c734eeb42882082bd131206dbfedd5b</td><td>Learning to Align from Scratch +<br/><b>University of Massachusetts, Amherst, MA</b><br/><b>University of Michigan, Ann Arbor, MI</b></td><td>('3219900', 'Gary B. Huang', 'gary b. huang')</td><td>{gbhuang,mmattar,elm}@cs.umass.edu +<br/>honglak@eecs.umich.edu +</td></tr><tr><td>55079a93b7d1eb789193d7fcdcf614e6829fad0f</td><td>Efficient and Robust Inverse Lighting of a Single Face Image using Compressive +<br/>Sensing +<br/><b>Center for Sensor Systems (ZESS) and Institute for Vision and Graphics#, University of Siegen</b><br/>57076 Siegen, Germany +</td><td>('1747804', 'Miguel Heredia Conde', 'miguel heredia conde')<br/>('1967283', 'Davoud Shahlaei', 'davoud shahlaei')<br/>('2880906', 'Volker Blanz', 'volker blanz')<br/>('1698728', 'Otmar Loffeld', 'otmar loffeld')</td><td>heredia@zess.uni-siegen.de +</td></tr><tr><td>55804f85613b8584d5002a5b0ddfe86b0d0e3325</td><td>Data Complexity in Machine Learning +<br/><b>Learning Systems Group, California Institute of Technology</b></td><td>('37715538', 'Ling Li', 'ling li')<br/>('1817975', 'Yaser S. Abu-Mostafa', 'yaser s. abu-mostafa')</td><td></td></tr><tr><td>551fa37e8d6d03b89d195a5c00c74cc52ff1c67a</td><td>GeThR-Net: A Generalized Temporally Hybrid +<br/>Recurrent Neural Network for Multimodal +<br/>Information Fusion +<br/>1 Xerox Research Centre India; 2 Amazon Development Center India +</td><td>('2757149', 'Ankit Gandhi', 'ankit gandhi')<br/>('34751361', 'Arjun Sharma', 'arjun sharma')<br/>('2221075', 'Arijit Biswas', 'arijit biswas')<br/>('2116262', 'Om Deshmukh', 'om deshmukh')</td><td>{ankit.g1290,arjunsharma.iitg,arijitbiswas87}@gmail.com; +<br/>om.deshmukh@xerox.com (*-equal contribution) +</td></tr><tr><td>55eb7ec9b9740f6c69d6e62062a24bfa091bbb0c</td><td>CAS(ME)2: A Database of Spontaneous +<br/>Macro-expressions and Micro-expressions +<br/><b>State Key Laboratory of Brain and Cognitive Science, Institute of Psychology</b><br/>Chinese Academy of Sciences, Beijing, China +<br/><b>University of Chinese Academy of Sciences, Beijing, China</b><br/><b>Key Laboratory of Behavior Sciences, Institute of Psychology</b><br/>Chinese Academy of Sciences, Beijing, China +<br/><b>Institute of Psychology and Behavioral Sciences</b><br/><b>Wenzhou University, Wenzhou, China</b></td><td>('34495371', 'Fangbing Qu', 'fangbing qu')<br/>('9185305', 'Wen-Jing Yan', 'wen-jing yan')<br/>('1684007', 'Xiaolan Fu', 'xiaolan fu')</td><td>{qufb,fuxl}@psych.ac.cn +<br/>wangsujing@psych.ac.cn +<br/>yanwj@wzu.edu.cn +</td></tr><tr><td>55b9b1c1c5487f5f62b44340104a9c4cc2ed7c96</td><td>1 Million Full-Sentences Visual Question Answering (FSVQA) +<br/>The Color of the Cat is Gray: +<br/><b>The University of Tokyo</b><br/>7 Chome-3-1 Hongo, Bunkyo +<br/>Tokyo 113-8654, Japan +</td><td>('2518695', 'Andrew Shin', 'andrew shin')<br/>('3250559', 'Yoshitaka Ushiku', 'yoshitaka ushiku')<br/>('1790553', 'Tatsuya Harada', 'tatsuya harada')</td><td></td></tr><tr><td>55c40cbcf49a0225e72d911d762c27bb1c2d14aa</td><td>Indian Face Age Database: A Database for Face Recognition with Age Variation +<br/>{tag} {/tag} +<br/> International Journal of Computer Applications +<br/> +<br/> Foundation of Computer Science (FCS), NY, USA +<br/> +<br/> +<br/>Volume 126 +<br/>- +<br/>Number 5 +<br/> +<br/> +<br/> Year of Publication: 2015 +<br/> +<br/> +<br/> +<br/> +<br/> Authors: +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> 10.5120/ijca2015906055 +<br/> {bibtex}2015906055.bib{/bibtex} +</td><td>('2029759', 'Reecha Sharma', 'reecha sharma')</td><td></td></tr><tr><td>9788b491ddc188941dadf441fc143a4075bff764</td><td>LOGAN: Membership Inference Attacks Against Generative Models∗ +<br/><b>University College London</b></td><td>('9200194', 'Jamie Hayes', 'jamie hayes')<br/>('2008164', 'Luca Melis', 'luca melis')<br/>('1722262', 'George Danezis', 'george danezis')<br/>('1728207', 'Emiliano De Cristofaro', 'emiliano de cristofaro')</td><td>{j.hayes, l.melis, g.danezis, e.decristofaro}@cs.ucl.ac.uk +</td></tr><tr><td>973e3d9bc0879210c9fad145a902afca07370b86</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 7, No. 7, 2016 +<br/>From Emotion Recognition to Website +<br/>Customizations +<br/>O.B. Efremides +<br/>School of Web Media +<br/>Bahrain Polytechnic +<br/>Isa Town, Kingdom of Bahrain +</td><td></td><td></td></tr><tr><td>970c0d6c0fd2ebe7c5921a45bc70f6345c844ff3</td><td>Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-16) +<br/>Discriminative Log-Euclidean Feature Learning for Sparse +<br/>Representation-Based Recognition of Faces from Videos +<br/><b>Center for Automation Research, University of Maryland</b><br/><b>College Park, MD</b><br/>{mefathy, azadeh, rama} (at) umiacs.umd.edu +</td><td>('4570075', 'Mohammed E. Fathy', 'mohammed e. fathy')<br/>('2943431', 'Azadeh Alavi', 'azadeh alavi')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>97b8249914e6b4f8757d22da51e8347995a40637</td><td>28 +<br/>Large-Scale Vehicle Detection, Indexing, +<br/>and Search in Urban Surveillance Videos +</td><td>('1832513', 'Behjat Siddiquie', 'behjat siddiquie')<br/>('3151405', 'James Petterson', 'james petterson')<br/>('2029646', 'Yun Zhai', 'yun zhai')<br/>('3233207', 'Ankur Datta', 'ankur datta')<br/>('34609371', 'Lisa M. Brown', 'lisa m. brown')<br/>('1767897', 'Sharath Pankanti', 'sharath pankanti')</td><td></td></tr><tr><td>972ef9ddd9059079bdec17abc8b33039ed25c99c</td><td>International Journal of Innovations in Engineering and Technology (IJIET) +<br/>A Novel on understanding How IRIS +<br/>Recognition works +<br/>Dept. of Comp. Science +<br/><b>M.P.M. College, Bhopal, India</b><br/>Asst. Professor CSE +<br/><b>M.P.M. College, Bhopal, India</b></td><td>('37930830', 'Vijay Shinde', 'vijay shinde')<br/>('9345591', 'Prakash Tanwar', 'prakash tanwar')</td><td></td></tr><tr><td>97032b13f1371c8a813802ade7558e816d25c73f</td><td>Total Recall Final Report +<br/>Supervisor: Professor Duncan Gillies +<br/>January 11, 2006 +</td><td>('2561350', 'Peter Collingbourne', 'peter collingbourne')<br/>('3036326', 'Khilan Gudka', 'khilan gudka')<br/>('15490561', 'Steve Lovegrove', 'steve lovegrove')<br/>('35260800', 'Jiefei Ma', 'jiefei ma')</td><td></td></tr><tr><td>97137d5154a9f22a5d9ecc32e8e2b95d07a5a571</td><td>The final publication is available at Springer via http://dx.doi.org/10.1007/s11042-016-3418-y +<br/>Facial Expression Recognition based on Local Region +<br/>Specific Features and Support Vector Machines +<br/>Park1 +<br/><b>Korea Electronics Technology Institute, Jeonju-si, Jeollabuk-do 561-844, Rep. of Korea; E</b><br/><b>Division of Computer Engineering, Jeonbuk National University, Jeonju-si, Jeollabuk-do</b><br/>Tel.: +82-63-270-2406; Fax: +82-63-270-2394. +</td><td>('32322842', 'Deepak Ghimire', 'deepak ghimire')<br/>('31984909', 'SungHwan Jeong', 'sunghwan jeong')<br/>('2034182', 'Joonwhoan Lee', 'joonwhoan lee')</td><td>Mails: (deepak, shjeong, shpark)@keti.re.kr +<br/>756, Rep. of Korea; E-Mail: chlee@jbnu.ac.kr +<br/>♣ Corresponding Author; E-Mail: chlee@jbnu.ac.kr; +</td></tr><tr><td>9730b9cd998c0a549601c554221a596deda8af5b</td><td>Spatio-temporal Person Retrieval via Natural Language Queries +<br/><b>Graduate School of Information Science and Technology, The University of Tokyo</b></td><td>('3369734', 'Masataka Yamaguchi', 'masataka yamaguchi')<br/>('8915348', 'Kuniaki Saito', 'kuniaki saito')<br/>('3250559', 'Yoshitaka Ushiku', 'yoshitaka ushiku')<br/>('1790553', 'Tatsuya Harada', 'tatsuya harada')</td><td>{yamaguchi, ksaito, ushiku, harada}@mi.t.u-tokyo.ac.jp +</td></tr><tr><td>978a219e07daa046244821b341631c41f91daccd</td><td>Emotional Intelligence: Giving Computers +<br/>Effective Emotional Skills to Aid Interaction +<br/><b>School of Computer Science, University of Birmingham, UK</b><br/>1 Introduction +<br/>Why do computers need emotional intelligence? Science fiction often por- +<br/>trays emotional computers as dangerous and frightening, and as a serious +<br/>threat to human life. One of the most famous examples is HAL, the supercom- +<br/>puter onboard the spaceship Discovery, in the movie 2001: A Space Odyssey. +<br/>HAL could express, recognize and respond to human emotion, and generally +<br/>had strong emotional skills – the consequences of which were catastrophic. +<br/>However, since the movie’s release almost 40 years ago, the traditional view +<br/>of emotions as contributing to irrational and unpredictable behaviour has +<br/>changed. Recent research has suggested that emotions play an essential role +<br/>in important areas such as learning, memory, motivation, attention, creativ- +<br/>ity, and decision making. These findings have prompted a large number of +<br/>research groups around the world to start examining the role of emotions and +<br/>emotional intelligence in human-computer interaction (HCI). +<br/>For almost half a century, computer scientists have been attempting to build +<br/>machines that can interact intelligently with us, and despite initial optimism, +<br/>they are still struggling to do so. For much of this time, the role of emotion in +<br/>developing intelligent computers was largely overlooked, and it is only recently +<br/>that interest in this area has risen dramatically. This increased interest can +<br/>largely be attributed to the work of [6] and [85] who were amongst the first to +<br/>bring emotion to the attention of computer scientists. The former highlighted +<br/>emotion as a fundamental component required in building believable agents, +<br/>while the latter further raised the awareness of emotion and its potential +<br/>importance in HCI. Since these publications, the literature on emotions and +<br/>computing has grown considerably with progress being made on a number of +<br/>different fronts. +<br/>The concept of designing computers to have emotional intelligence may seem +<br/>strange, but equipping computers with this type of intelligence may provide +<br/>a number of important advantages. For example, in spite of a computer’s +</td><td>('3134697', 'Chris Creed', 'chris creed')<br/>('2282865', 'Russell Beale', 'russell beale')</td><td>cpc@cs.bham.ac.uk +<br/>r.beale@cs.bham.ac.uk +</td></tr><tr><td>976e0264bb57786952a987d4456850e274714fb8</td><td>Improving Semantic Concept Detection through the +<br/>Dictionary of Visually-distinct Elements +<br/><b>Center for Research in Computer Vision, University of Central Florida</b></td><td>('1707795', 'Afshin Dehghan', 'afshin dehghan')<br/>('1803711', 'Haroon Idrees', 'haroon idrees')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>{adehghan, haroon, shah}@cs.ucf.edu +</td></tr><tr><td>9758f3fd94239a8d974217fe12599f88fb413f3d</td><td>UC-HCC Submission to Thumos 2014 +<br/><b>Vision and Sensing, HCC, ESTeM, University of Canberra</b></td><td>('1793720', 'O. V. Ramana Murthy', 'o. v. ramana murthy')<br/>('1717204', 'Roland Goecke', 'roland goecke')</td><td></td></tr><tr><td>97f9c3bdb4668f3e140ded2da33fe704fc81f3ea</td><td>AnExperimentalComparisonofAppearance +<br/>andGeometricModelBasedRecognition +<br/>J.Mundy,A.Liu,N.Pillow,A.Zisserman,S.Abdallah,S.Utcke, +<br/>S.NayarandC.Rothwell +<br/>GeneralElectricCorporateResearchandDevelopment,Schenectady,NY,USA +<br/><b>RoboticsResearchGroup, UniversityofOxford, Oxford, UK</b><br/><b>ColumbiaUniversity, NY, USA</b><br/>INRIA,SophiaAntipolis,France +</td><td></td><td></td></tr><tr><td>97e569159d5658760eb00ca9cb662e6882d2ab0e</td><td>Correlation Filters for Object Alignment +<br/><b>Carnegie Mellon University</b><br/><b>Carnegie Mellon University</b><br/>B.V.K. Vijaya Kumar +<br/><b>Carnegie Mellon University</b></td><td>('2232940', 'Vishnu Naresh Boddeti', 'vishnu naresh boddeti')<br/>('1733113', 'Takeo Kanade', 'takeo kanade')</td><td>naresh@cmu.edu +<br/>tk@cs.cmu.edu +<br/>kumar@ece.cmu.edu +</td></tr><tr><td>97cf04eaf1fc0ac4de0f5ad4a510d57ce12544f5</td><td>manuscript No. +<br/>(will be inserted by the editor) +<br/>Deep Affect Prediction in-the-wild: Aff-Wild Database and Challenge, +<br/>Deep Architectures, and Beyond +<br/>Zafeiriou4 +</td><td>('1811396', 'Dimitrios Kollias', 'dimitrios kollias')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')</td><td></td></tr><tr><td>97d1d561362a8b6beb0fdbee28f3862fb48f1380</td><td>1955 +<br/>Age Synthesis and Estimation via Faces: +<br/>A Survey +</td><td>('1708679', 'Yun Fu', 'yun fu')<br/>('1822413', 'Guodong Guo', 'guodong guo')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td></td></tr><tr><td>97540905e4a9fdf425989a794f024776f28a3fa9</td><td></td><td></td><td></td></tr><tr><td>97865d31b5e771cf4162bc9eae7de6991ceb8bbf</td><td>Face and Gender Classification in Crowd Video +<br/>IIIT-D-MTech-CS-GEN-13-100 +<br/>July 16, 2015 +<br/><b>Indraprastha Institute of Information Technology</b><br/>New Delhi +<br/>Thesis Advisors +<br/>Dr. Richa Singh +<br/>Submitted in partial fulfillment of the requirements +<br/>for the Degree of M.Tech. in Computer Science +<br/>c(cid:13) Verma, 2015 +<br/>Keywords : Face Recognition, Gender Classification, Crowd database +</td><td>('2578160', 'Priyanka Verma', 'priyanka verma')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')</td><td></td></tr><tr><td>975978ee6a32383d6f4f026b944099e7739e5890</td><td>Privacy-Preserving Age Estimation +<br/>for Content Rating +<br/>Binglin Li∗ +<br/><b>University of Manitoba</b><br/><b>Simon Fraser University</b><br/>Winnipeg, Canada +<br/>Burnaby, Canada +<br/>Noman Mohammed +<br/><b>University of Manitoba</b><br/>Winnipeg, Canada +<br/>Yang Wang +<br/>Jie Liang +<br/><b>University of Manitoba</b><br/><b>Simon Fraser University</b><br/>Winnipeg, Canada +<br/>Burnaby, Canada +</td><td>('2373631', 'Linwei Ye', 'linwei ye')</td><td>yel3@cs.umanitoba.ca +<br/>binglinl@sfu.ca +<br/>noman@cs.umanitoba.ca +<br/>ywang@cs.umanitoba.ca +<br/>jiel@sfu.ca +</td></tr><tr><td>9755554b13103df634f9b1ef50a147dd02eab02f</td><td>How Transferable are CNN-based Features for +<br/>Age and Gender Classification? +<br/> 1 +</td><td>('2850086', 'Gökhan Özbulak', 'gökhan özbulak')<br/>('3152281', 'Yusuf Aytar', 'yusuf aytar')</td><td></td></tr><tr><td>635158d2da146e9de559d2742a2fa234e06b52db</td><td></td><td></td><td></td></tr><tr><td>63d8110ac76f57b3ba8a5947bc6bdbb86f25a342</td><td>On Modeling Variations for Face Authentication +<br/><b>Carnegie Mellon University, Pittsburgh, PA</b></td><td>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td>xiaoming@andrew.cmu.edu tsuhan@cmu.edu kumar@ece.cmu.edu +</td></tr><tr><td>63cf5fc2ee05eb9c6613043f585dba48c5561192</td><td>Prototype Selection for +<br/>Classification in Standard +<br/>and Generalized +<br/>Dissimilarity Spaces +</td><td></td><td></td></tr><tr><td>632b24ddd42fda4aebc5a8af3ec44f7fd3ecdc6c</td><td>Real-Time Facial Segmentation +<br/>and Performance Capture from RGB Input +<br/>Pinscreen +<br/><b>University of Southern California</b></td><td>('2059597', 'Shunsuke Saito', 'shunsuke saito')<br/>('50290121', 'Tianye Li', 'tianye li')<br/>('1706574', 'Hao Li', 'hao li')</td><td></td></tr><tr><td>6324fada2fb00bd55e7ff594cf1c41c918813030</td><td>Uncertainty Reduction For Active Image Clustering +<br/>via a Hybrid Global-Local Uncertainty Model +<br/><b>State University of New York at Buffalo</b><br/>Department of Computer Science and Engineering +<br/>338 Davis Hall, Buffalo, NY, 14260-2500 +</td><td>('2228109', 'Caiming Xiong', 'caiming xiong')<br/>('34187462', 'David M. Johnson', 'david m. johnson')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')</td><td>{cxiong,davidjoh,jcorso}@buffalo.edu +</td></tr><tr><td>6308e9c991125ee6734baa3ec93c697211237df8</td><td>LEARNING THE SPARSE REPRESENTATION FOR CLASSIFICATION +<br/><b>Beckman Institute, University of Illinois at Urbana-Champaign, USA</b></td><td>('1706007', 'Jianchao Yang', 'jianchao yang')<br/>('7898154', 'Jiangping Wang', 'jiangping wang')</td><td>{jyang29, jwang63, huang}@ifp.illinois.edu +</td></tr><tr><td>6342a4c54835c1e14159495373ab18b4233d2d9b</td><td>TOWARDS POSE-ROBUST +<br/>FACE RECOGNITION ON VIDEO +<br/>Submitted as a requirement of the degree +<br/>of doctor of philosophy +<br/>at the +<br/>Science and Engineering Faculty +<br/><b>Queensland University of Technology</b><br/>September, 2014 +</td><td>('23168868', 'Moh Edi Wibowo', 'moh edi wibowo')</td><td></td></tr><tr><td>63d8d69e90e79806a062cb8654ad78327c8957bb</td><td></td><td></td><td></td></tr><tr><td>63c109946ffd401ee1195ed28f2fb87c2159e63d</td><td>14-1 +<br/>MVA2011 IAPR Conference on Machine Vision Applications, June 13-15, 2011, Nara, JAPAN +<br/>Robust Facial Feature Localization using Improved Active Shape +<br/>Model and Gabor Filter +<br/><b>Engineering, National Formosa University</b><br/>Taiwan +</td><td>('1711364', 'Hui-Yu Huang', 'hui-yu huang')</td><td>E-mail: hyhuang@nfu.edu.tw +</td></tr><tr><td>63b29886577a37032c7e32d8899a6f69b11a90de</td><td>Image-set based Face Recognition Using Boosted Global +<br/>and Local Principal Angles +<br/><b>Xi an Jiaotong University, China</b><br/><b>University of Tsukuba, Japan</b></td><td>('6916241', 'Xi Li', 'xi li')<br/>('1770128', 'Kazuhiro Fukui', 'kazuhiro fukui')<br/>('1715389', 'Nanning Zheng', 'nanning zheng')</td><td>lxaccv09@yahoo.com, +<br/>znn@xjtu.edu.cn +<br/>kf@cs.tsukuba.ac.jp +</td></tr><tr><td>631483c15641c3652377f66c8380ff684f3e365c</td><td>Sync-DRAW: Automatic Video Generation using Deep Recurrent +<br/>A(cid:130)entive Architectures +<br/>Gaurav Mi(cid:138)al∗ +<br/>IIT Hyderabad +<br/>Vineeth N Balasubramanian +<br/>IIT Hyderabad +</td><td>('8268761', 'Tanya Marwah', 'tanya marwah')</td><td>gaurav.mi(cid:138)al.191013@gmail.com +<br/>ee13b1044@iith.ac.in +<br/>vineethnb@iith.ac.in +</td></tr><tr><td>63a6c256ec2cf2e0e0c9a43a085f5bc94af84265</td><td>Complexity of Multiverse Networks and +<br/>their Multilayer Generalization +<br/>The Blavatnik School of Computer Science +<br/><b>Tel Aviv University</b></td><td>('1762320', 'Etai Littwin', 'etai littwin')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td></td></tr><tr><td>63213d080a43660ac59ea12e3c35e6953f6d7ce8</td><td>ActionVLAD: Learning spatio-temporal aggregation for action classification +<br/><b>Robotics Institute, Carnegie Mellon University</b><br/>2Adobe Research +<br/>3INRIA +<br/>http://rohitgirdhar.github.io/ActionVLAD +</td><td>('3102850', 'Rohit Girdhar', 'rohit girdhar')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')<br/>('1782755', 'Josef Sivic', 'josef sivic')<br/>('2015670', 'Bryan Russell', 'bryan russell')</td><td></td></tr><tr><td>630d1728435a529d0b0bfecb0e7e335f8ea2596d</td><td>Facial Action Unit Detection by Cascade of Tasks +<br/><b>School of Information Science and Engineering, Southeast University, Nanjing, China</b><br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA</b><br/><b>University of Pittsburgh, Pittsburgh, PA</b></td><td>('2499751', 'Xiaoyu Ding', 'xiaoyu ding')<br/>('18870591', 'Qiao Wang', 'qiao wang')</td><td></td></tr><tr><td>63eefc775bcd8ccad343433fc7a1dd8e1e5ee796</td><td></td><td></td><td></td></tr><tr><td>632fa986bed53862d83918c2b71ab953fd70d6cc</td><td>GÜNEL ET AL.: WHAT FACE AND BODY SHAPES CAN TELL ABOUT HEIGHT +<br/>What Face and Body Shapes Can Tell +<br/>About Height +<br/>CVLab +<br/>EPFL, +<br/>Lausanne, Switzerland +</td><td>('46211822', 'Semih Günel', 'semih günel')<br/>('2933543', 'Helge Rhodin', 'helge rhodin')<br/>('1717736', 'Pascal Fua', 'pascal fua')</td><td>semih.gunel@epfl.ch +<br/>helge.rhodin@epfl.ch +<br/>pascal.fua@epfl.ch +</td></tr><tr><td>63340c00896d76f4b728dbef85674d7ea8d5ab26</td><td>1732 +<br/>Discriminant Subspace Analysis: +<br/>A Fukunaga-Koontz Approach +</td><td>('40404906', 'Sheng Zhang', 'sheng zhang')<br/>('1715286', 'Terence Sim', 'terence sim')</td><td></td></tr><tr><td>633101e794d7b80f55f466fd2941ea24595e10e6</td><td>In submission to IEEE conference +<br/>Face Attribute Prediction with classification CNN +<br/>FACE ATTRIBUTE PREDICTION WITH +<br/>CLASSIFICATION CNN +<br/>Computer Science and Communication +<br/><b>KTH Royal Institute of Technology</b><br/>100 44 Stockholm, Sweden +</td><td>('50262049', 'Yang Zhong', 'yang zhong')<br/>('1736906', 'Josephine Sullivan', 'josephine sullivan')<br/>('40565290', 'Haibo Li', 'haibo li')</td><td>{yzhong, sullivan, haiboli}@kth.se +</td></tr><tr><td>63a2e2155193dc2da9764ae7380cdbd044ff2b94</td><td>A Dense SURF and Triangulation based +<br/>Spatio-Temporal Feature for Action Recognition +<br/><b>The University of Electro-Communications</b><br/>Chofu, Tokyo 182-8585 JAPAN +</td><td>('2274625', 'Do Hang Nga', 'do hang nga')<br/>('1681659', 'Keiji Yanai', 'keiji yanai')</td><td>fdohang,yanaig@mm.cs.uec.ac.jp +</td></tr><tr><td>63d865c66faaba68018defee0daf201db8ca79ed</td><td>Deep Regression for Face Alignment +<br/>1Dept. of Electronics and Information Engineering, Huazhong Univ. of Science and Technology, China +<br/>2Microsoft Research, Beijing, China +</td><td>('2276155', 'Baoguang Shi', 'baoguang shi')<br/>('1688516', 'Jingdong Wang', 'jingdong wang')</td><td>shibaoguang@gmail.com,{xbai,liuwy}@hust.edu.cn,jingdw@microsoft.com +</td></tr><tr><td>63cff99eff0c38b633c8a3a2fec8269869f81850</td><td>Feature Correlation Filter for Face Recognition +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern +<br/>Recognition, +<br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/>95 Zhongguancun East Road, 100080 Beijing, China +<br/>http://www.cbsr.ia.ac.cn +</td><td>('32015491', 'XiangXin Zhu', 'xiangxin zhu')<br/>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('3168566', 'Rong Liu', 'rong liu')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{xxzhu,scliao,zlei,rliu,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>634541661d976c4b82d590ef6d1f3457d2857b19</td><td>AAllmmaa MMaatteerr SSttuuddiioorruumm –– UUnniivveerrssiittàà ddii BBoollooggnnaa +<br/>in cotutela con Università di Sassari +<br/>DOTTORATO DI RICERCA IN +<br/>INGEGNERIA ELETTRONICA, INFORMATICA E DELLE +<br/>TELECOMUNICAZIONI +<br/>Ciclo XXVI +<br/>Settore Concorsuale di afferenza: 09/H1 +<br/>Settore Scientifico disciplinare: ING-INF/05 +<br/>ADVANCED TECHNIQUES FOR FACE RECOGNITION +<br/>UNDER CHALLENGING ENVIRONMENTS +<br/>TITOLO TESI +<br/>Presentata da: +<br/>Coordinatore Dottorato +<br/>ALESSANDRO VANELLI-CORALLI +<br/> +<br/>Relatore +<br/> DAVIDE MALTONI +<br/>Relatore +<br/> MASSIMO TISTARELLI +<br/>Esame finale anno 2014 +</td><td>('2384894', 'Yunlian Sun', 'yunlian sun')</td><td></td></tr><tr><td>6332a99e1680db72ae1145d65fa0cccb37256828</td><td>MASTER IN COMPUTER VISION AND ARTIFICIAL INTELLIGENCE +<br/>REPORT OF THE RESEARCH PROJECT +<br/>OPTION: COMPUTER VISION +<br/>Pose and Face Recovery via +<br/>Spatio-temporal GrabCut Human +<br/>Segmentation +<br/>Date: 13/07/2010 +</td><td>('4765407', 'Antonio Hernández Vela', 'antonio hernández vela')<br/>('10722928', 'Sergio Escalera Guerrero', 'sergio escalera guerrero')</td><td></td></tr><tr><td>63488398f397b55552f484409b86d812dacde99a</td><td>Learning Universal Multi-view Age Estimator by Video Contexts +<br/><b>2 School of Computing, National University of Singapore</b><br/>3 Advanced Digital Sciences Center, Singapore; 4 Facebook +</td><td>('1964516', 'Zheng Song', 'zheng song')<br/>('5796401', 'Bingbing Ni', 'bingbing ni')<br/>('39034731', 'Dong Guo', 'dong guo')<br/>('1715286', 'Terence Sim', 'terence sim')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>{zheng.s, eleyans}@nus.edu.sg, bingbing.ni@adsc.com.sg, dnguo@fb.com, tsim@comp.nus.edu.sg +</td></tr><tr><td>6341274aca0c2977c3e1575378f4f2126aa9b050</td><td>A Multi-Scale Cascade Fully Convolutional +<br/>Network Face Detector +<br/><b>Institute for Robotics and Intelligent Systems</b><br/><b>University of Southern California</b><br/>Los Angeles, California 90089 +</td><td>('3469030', 'Zhenheng Yang', 'zhenheng yang')<br/>('1694832', 'Ramakant Nevatia', 'ramakant nevatia')</td><td>Email:(cid:8)zhenheny,nevatia(cid:9)@usc.edu +</td></tr><tr><td>63c022198cf9f084fe4a94aa6b240687f21d8b41</td><td>425 +</td><td></td><td></td></tr><tr><td>632441c9324cd29489cee3da773a9064a46ae26b</td><td>Video-based Cardiac Physiological Measurements Using +<br/>Joint Blind Source Separation Approaches +<br/>by +<br/><b>B. Eng., Zhejiang University</b><br/>A THESIS SUBMITTED IN PARTIAL FULFILLMENT +<br/>OF THE REQUIREMENTS FOR THE DEGREE OF +<br/>Master of Applied Science +<br/>in +<br/>THE FACULTY OF GRADUATE AND POSTDOCTORAL +<br/>STUDIES +<br/>(Electrical and Computer Engineering) +<br/><b>The University of British Columbia</b><br/>(Vancouver) +<br/>July 2015 +</td><td>('33064881', 'Huan Qi', 'huan qi')<br/>('33064881', 'Huan Qi', 'huan qi')</td><td></td></tr><tr><td>0f65c91d0ed218eaa7137a0f6ad2f2d731cf8dab</td><td>Multi-Directional Multi-Level Dual-Cross +<br/>Patterns for Robust Face Recognition +</td><td>('37990555', 'Changxing Ding', 'changxing ding')<br/>('3826759', 'Jonghyun Choi', 'jonghyun choi')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td></td></tr><tr><td>0f112e49240f67a2bd5aaf46f74a924129f03912</td><td>947 +<br/>Age-Invariant Face Recognition +</td><td>('2222919', 'Unsang Park', 'unsang park')<br/>('3225345', 'Yiying Tong', 'yiying tong')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>0fc254272db096a9305c760164520ad9914f4c9e</td><td>UNSUPERVISED CONVOLUTIONAL NEURAL NETWORKS FOR MOTION ESTIMATION +<br/>School of Electronic Engineering and Computer Science +<br/><b>Queen Mary University of London</b><br/>Mile End road, E1 4NS, London, UK +</td><td>('29946980', 'Aria Ahmadi', 'aria ahmadi')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')</td><td></td></tr><tr><td>0fae5d9d2764a8d6ea691b9835d497dd680bbccd</td><td>Face Recognition using Canonical Correlation Analysis +<br/>Department of Electrical Engineering +<br/><b>Indian Institute of Technology, Madras</b><br/>Department of Electrical Engineering +<br/><b>Indian Institute of Technology, Madras</b></td><td>('37274547', 'Amit C. Kale', 'amit c. kale')<br/>('4436239', 'R. Aravind', 'r. aravind')</td><td>ee04s043@ee.iitm.ac.in +<br/>aravind@tenet.res.in +</td></tr><tr><td>0f4cfcaca8d61b1f895aa8c508d34ad89456948e</td><td>LOCAL APPEARANCE BASED FACE RECOGNITION USING +<br/>DISCRETE COSINE TRANSFORM (WedPmPO4) +<br/>Author(s) : +</td><td></td><td></td></tr><tr><td>0fdcfb4197136ced766d538b9f505729a15f0daf</td><td>Multiple Pattern Classification by Sparse Subspace Decomposition +<br/><b>Institute of Media and Information Technology, Chiba University</b><br/>1-33 Yayoi, Inage, Chiba, Japan +</td><td>('1688743', 'Tomoya Sakai', 'tomoya sakai')</td><td>tsakai@faculty.chiba-u.jp +</td></tr><tr><td>0fad544edfc2cd2a127436a2126bab7ad31ec333</td><td>Decorrelating Semantic Visual Attributes by Resisting the Urge to Share +<br/>UT Austin +<br/>USC +<br/>UT Austin +</td><td>('2228235', 'Dinesh Jayaraman', 'dinesh jayaraman')<br/>('1693054', 'Fei Sha', 'fei sha')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>dineshj@cs.utexas.edu +<br/>feisha@usc.edu +<br/>grauman@cs.utexas.edu +</td></tr><tr><td>0f32df6ae76402b98b0823339bd115d33d3ec0a0</td><td>Emotion recognition from embedded bodily +<br/>expressions and speech during dyadic interactions +</td><td>('40404576', 'Sikandar Amin', 'sikandar amin')<br/>('2766593', 'Prateek Verma', 'prateek verma')<br/>('1906895', 'Mykhaylo Andriluka', 'mykhaylo andriluka')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')</td><td>∗Max Planck Institute for Informatics, Germany, {pmueller,andriluk,bulling}@mpi-inf.mpg.de +<br/>†Stanford University, USA, prateekv@stanford.edu +<br/>‡Technical University of Munich, Germany, sikandar.amin@in.tum.de +</td></tr><tr><td>0fd1715da386d454b3d6571cf6d06477479f54fc</td><td>J Intell Robot Syst (2016) 82:101–133 +<br/>DOI 10.1007/s10846-015-0259-2 +<br/>A Survey of Autonomous Human Affect Detection Methods +<br/>for Social Robots Engaged in Natural HRI +<br/>Received: 10 December 2014 / Accepted: 11 August 2015 / Published online: 23 August 2015 +<br/>© Springer Science+Business Media Dordrecht 2015 +</td><td>('2929516', 'Derek McColl', 'derek mccoll')<br/>('31839336', 'Naoaki Hatakeyama', 'naoaki hatakeyama')<br/>('1719617', 'Beno Benhabib', 'beno benhabib')</td><td></td></tr><tr><td>0f9bf5d8f9087fcba419379600b86ae9e9940013</td><td></td><td></td><td></td></tr><tr><td>0f829fee12e86f980a581480a9e0cefccb59e2c5</td><td>Bird Part Localization Using Exemplar-Based Models with Enforced +<br/>Pose and Subcategory Consistency +<br/><b>Columbia University</b><br/>Problem +<br/>The goal of our work is to localize the parts au- +<br/>tomatically and accurately for fine-grained cate- +<br/>gories. We evaluate our method on bird images in +<br/>the CUB-200-2011 [1] dataset. +<br/>Pipeline +<br/>Approach +<br/>Subcategory Detectors +<br/>Localization Examples +<br/>(1) Sliding-window detection. (2) Matching and ranking exemplars. (3) Predicting the final part configuration. +<br/>Does Xk,t match the image I? ⇐⇒ P (Xk,t|I) =? +<br/>k, si +<br/>k,t|di +<br/>k,t])} +<br/>P (Xk,t|I) = P (Xk,t|Dp)αP (Xk,t|Ds)1−α +<br/>P (Xk,t|Dp) = Gavg{P (xi +<br/>P (Xk,t|Ds) = max +<br/>P (Xk,t|l, Ds) = Gavg{P (xi +<br/>(1) +<br/>(2) +<br/>(3) +<br/>k,t])} (4) +<br/>We use the most likely models M to predict the +<br/>part locations of the testing sample: +<br/>k,t)P (xi|di +<br/>p[ci +<br/>P (Xk,t|l, Ds) +<br/>s[l, si +<br/>ˆxi = arg max +<br/>(cid:88) +<br/>P ((cid:52)xi +<br/>k,t]) (5) +<br/>k,t|di +<br/>p[ci +<br/>k, si +<br/>k,t, θi +<br/>xi +<br/>k,t∈M +<br/>Species 1 +<br/>Species 2 +<br/>Species 3 +<br/>Subcategory clusters of Back +<br/>For each species l of part i, we build a detector after +<br/>aligning the samples. Assuming the detector scans +<br/>the image over scales and orientations, then the re- +<br/>sponse map of this detector at a particular scale si +<br/>and orientation θi is denoted as di +<br/>Enforcing Consistency +<br/>P (xi +<br/>P (xi +<br/>s[l, si, θi]. +<br/>s[l, si +<br/>k,t, θi +<br/>k,t|di +<br/>k,t|di +<br/>p[ci +<br/>k, si +<br/>k,t]) +<br/>k,t]) +<br/>Pose Detectors +<br/>Pose 1 +<br/>Pose 2 +<br/>Pose 3 +<br/>Poses clusters of Back +<br/>For each pose cluster ci of part i, we build a de- +<br/>tector. The detector scans the image over scales, +<br/>and the response map of this detector at a particu- +<br/>lar scale si is denoted as di +<br/>p[ci, si]. +<br/>References +<br/>[1] C. Wah, S. Branson, P. Welinder, P. Perona, S. Belongie. The +<br/>Caltech-UCSD Birds-200-2011 Dataset. Computation & Neu- +<br/>ral Systems Technical Report, CNS-TR-2011-001, 2011 +<br/>[2] P. N. Belhumeur, D. W. Jacobs, D. J. Kriegman, N. Kumar. +<br/>Localizing Parts of Faces Using a Consensus of Exemplars. +<br/>In CVPR ’11 +<br/>Comparisons +<br/>PCP +<br/>Back +<br/>Beak +<br/>Belly +<br/>Breast +<br/>Crown +<br/>Forehead +<br/>Left Eye +<br/>Left Leg +<br/>Left Wing +<br/>Nape +<br/>Right Eye +<br/>Right Leg +<br/>Right Wing +<br/>Tail +<br/>Throat +<br/>Average +<br/>CoE [2] Ours +<br/>62.08 +<br/>46.29 +<br/>49.02 +<br/>43.08 +<br/>69.02 +<br/>54.44 +<br/>66.98 +<br/>54.19 +<br/>72.85 +<br/>64.69 +<br/>58.46 +<br/>51.48 +<br/>55.78 +<br/>47.53 +<br/>40.94 +<br/>29.67 +<br/>71.57 +<br/>59.58 +<br/>70.78 +<br/>58.91 +<br/>55.51 +<br/>46.50 +<br/>40.52 +<br/>29.03 +<br/>71.56 +<br/>58.47 +<br/>40.16 +<br/>27.77 +<br/>70.83 +<br/>58.89 +<br/>59.74 +<br/>48.70 +<br/>mAP +<br/>Birdlets +<br/>Template bagging +<br/>Pose pooling +<br/>Ours +<br/>200 species +<br/>14 species +<br/>28.18 +<br/>44.13 +<br/>40.25 +<br/>44.73 +<br/>57.44 +<br/>62.42 +</td><td>('2454675', 'Jiongxin Liu', 'jiongxin liu')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')</td><td>{liujx09, belhumeur}@cs.columbia.edu +</td></tr><tr><td>0faee699eccb2da6cf4307ded67ba8434368257b</td><td>TAIGMAN: MULTIPLE ONE-SHOTS FOR UTILIZING CLASS LABEL INFORMATION +<br/>Multiple One-Shots for Utilizing Class Label +<br/>Information +<br/>1 The Blavatnik School of Computer +<br/>Science, +<br/><b>Tel-Aviv University, Israel</b><br/>2 Computer Science Division, +<br/><b>The Open University of Israel</b><br/>3 face.com +<br/>Tel-Aviv, Israel +</td><td>('2188620', 'Yaniv Taigman', 'yaniv taigman')<br/>('1776343', 'Lior Wolf', 'lior wolf')<br/>('1756099', 'Tal Hassner', 'tal hassner')</td><td>yaniv@face.com +<br/>wolf@cs.tau.ac.il +<br/>hassner@openu.ac.il +</td></tr><tr><td>0fabb4a40f2e3a2502cd935e54e090a304006c1c</td><td>Regularized Robust Coding for Face Recognition +<br/><b>The Hong Kong Polytechnic University, Hong Kong, China</b><br/>bSchool of Computer Science and Technology, Nanjing Univ. of Science and Technology, Nanjing, China +</td><td>('5828998', 'Meng Yang', 'meng yang')<br/>('36685537', 'Lei Zhang', 'lei zhang')<br/>('37081450', 'Jian Yang', 'jian yang')<br/>('1698371', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>0f92e9121e9c0addc35eedbbd25d0a1faf3ab529</td><td>MORPH-II: A Proposed Subsetting Scheme +<br/>NSF-REU Site at UNC Wilmington, Summer 2017 +</td><td>('1940145', 'K. Park', 'k. park')<br/>('11134292', 'Y. Wang', 'y. wang')<br/>('1693283', 'C. Chen', 'c. chen')<br/>('3369885', 'T. Kling', 't. kling')</td><td></td></tr><tr><td>0f0366070b46972fcb2976775b45681e62a94a26</td><td>Reliable Posterior Probability Estimation for Streaming Face Recognition +<br/><b>University of Colorado at Colorado Springs</b><br/>Terrance Boult +<br/><b>University of Colorado at Colorado Springs</b></td><td>('3274223', 'Abhijit Bendale', 'abhijit bendale')</td><td>abendale@vast.uccs.edu +<br/>tboult@vast.uccs.edu +</td></tr><tr><td>0ff23392e1cb62a600d10bb462d7a1f171f579d0</td><td>Toward Sparse Coding on Cosine +<br/>Distance +<br/>Jonghyun Choi, Hyunjong Cho, Jungsuk Kwak#, +<br/>Larry S. Davis +<br/><b>UMIACS | University of Maryland, College Park</b><br/><b>Stanford University</b></td><td></td><td></td></tr><tr><td>0fd3a7ee228bbc3dd4a111dae04952a1ee58a8cd</td><td>Hair Style Retrieval by Semantic Mapping on +<br/>Informative Patches +<br/><b>Tsinghua University, Beijing, China</b></td><td>('38081719', 'Nan Wang', 'nan wang')<br/>('1679380', 'Haizhou Ai', 'haizhou ai')</td><td>wang-n04@mails.tsinghua.edu.cn, ahz@mail.tsinghua.edu.cn +</td></tr><tr><td>0f533bc9fdfb75a3680d71c84f906bbd59ee48f1</td><td>Illumination Invariant Feature Extraction Based on Natural Images Statistics – +<br/>Taking Face Images as An Example +<br/><b>Research Center for Information Technology Innovation, Academia Sinica, Taipei, Taiwan</b><br/><b>National Taiwan University, Taipei, Taiwan</b></td><td>('2314709', 'Lu-Hung Chen', 'lu-hung chen')<br/>('1934873', 'Yao-Hsiang Yang', 'yao-hsiang yang')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')<br/>('2809590', 'Ming-Yen Cheng', 'ming-yen cheng')</td><td>luhung.chen,yhyang@statistics.twbbs.org song@iis.sincia.edu.tw +<br/>cheng@math.ntu.edu.tw +</td></tr><tr><td>0f4eb63402a4f3bae8f396e12133684fb760def1</td><td>LONG, LIU, SHAO: ATTRIBUTE EMBEDDING WITH VSAR FOR ZERO-SHOT LEARNING 1 +<br/>Attribute Embedding with Visual-Semantic +<br/>Ambiguity Removal for Zero-shot Learning +<br/>1 Department of Electronic and Electrical +<br/>Engineering +<br/><b>The University of Shef eld</b><br/>Sheffield , UK +<br/>2 Department of Computer and +<br/>Information Sciences +<br/><b>Northumbria University</b><br/>Newcastle upon Tyne, UK +</td><td>('39650869', 'Yang Long', 'yang long')<br/>('40017778', 'Li Liu', 'li liu')<br/>('40799321', 'Ling Shao', 'ling shao')</td><td>ylong2@sheffield.ac.uk +<br/>li2.liu@northumbria.ac.uk +<br/>ling.shao@ieee.org +</td></tr><tr><td>0fba39bf12486c7684fd3d51322e3f0577d3e4e8</td><td>Task Specific Local Region Matching +<br/>Department of Computer Science and Engineering +<br/><b>University of California, San Diego</b></td><td>('2490700', 'Boris Babenko', 'boris babenko')</td><td>{bbabenko,pdollar,sjb}@cs.ucsd.edu +</td></tr><tr><td>0f395a49ff6cbc7e796656040dbf446a40e300aa</td><td>ORIGINAL RESEARCH +<br/>published: 22 December 2015 +<br/>doi: 10.3389/fpsyg.2015.01937 +<br/>The Change of Expression +<br/>Configuration Affects +<br/>Identity-Dependent Expression +<br/>Aftereffect but Not +<br/>Identity-Independent Expression +<br/>Aftereffect +<br/><b>College of Information Engineering, Shanghai Maritime University, Shanghai, China, 2 School of Information, Kochi University</b><br/><b>of Technology, Kochi, Japan, 3 Yunnan Key Laboratory of Computer Technology Applications, Kunming University of Science</b><br/>and Technology, Kunming, China +<br/>The present study examined the influence of expression configuration on cross-identity +<br/>expression aftereffect. The expression configuration refers to the spatial arrangement +<br/>of facial features in a face for conveying an emotion, e.g., an open-mouth smile vs. +<br/>a closed-mouth smile. In the first of two experiments, the expression aftereffect is +<br/>measured using a cross-identity/cross-expression configuration factorial design. The +<br/>facial +<br/>identities of test faces were the same or different from the adaptor, while +<br/>orthogonally, the expression configurations of those facial identities were also the same +<br/>or different. The results show that the change of expression configuration impaired +<br/>the expression aftereffect when the facial +<br/>identities of adaptor and tests were the +<br/>same; however, the impairment effect disappears when facial identities were different, +<br/>indicating the identity-independent expression representation is more robust to the +<br/>change of the expression configuration in comparison with the identity-dependent +<br/>expression representation. In the second experiment, we used schematic line faces +<br/>as adaptors and real faces as tests to minimize the similarity between the adaptor +<br/>and tests, which is expected to exclude the contribution from the identity-dependent +<br/>expression representation to expression aftereffect. The second experiment yields a +<br/>similar result as the identity-independent expression aftereffect observed in Experiment 1. +<br/>The findings indicate the different neural sensitivities to expression configuration for +<br/>identity-dependent and identity-independent expression systems. +<br/>Keywords: facial expression, adaptation, aftereffect, visual representation, vision +<br/>INTRODUCTION +<br/>One key issue in face study is to understand how emotional expression is represented in the +<br/>human visual system. According to the classical cognitive model (Bruce and Young, 1986) and +<br/>neural model (Haxby et al., 2000), emotional expression is consider to be represented and +<br/>processed independent of facial identity. This view is supported by several lines of evidence. +<br/>Edited by: +<br/>Wenfeng Chen, +<br/><b>Institute of Psychology, Chinese</b><br/>Academy of Sciences, China +<br/>Reviewed by: +<br/>Marianne Latinus, +<br/>Aix Marseille Université, France +<br/>Jan Van den Stock, +<br/>KU Leuven, Belgium +<br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Emotion Science, +<br/>a section of the journal +<br/>Frontiers in Psychology +<br/>Received: 03 January 2015 +<br/>Accepted: 02 December 2015 +<br/>Published: 22 December 2015 +<br/>Citation: +<br/>Song M, Shinomori K, Qian Q, Yin J +<br/>and Zeng W (2015) The Change of +<br/>Expression Configuration Affects +<br/>Identity-Dependent Expression +<br/>Aftereffect but Not +<br/>Identity-Independent Expression +<br/>Aftereffect. Front. Psychol. 6:1937. +<br/>doi: 10.3389/fpsyg.2015.01937 +<br/>Frontiers in Psychology | www.frontiersin.org +<br/>December 2015 | Volume 6 | Article 1937 +</td><td>('1692572', 'Miao Song', 'miao song')<br/>('1970678', 'Keizo Shinomori', 'keizo shinomori')<br/>('2431558', 'Qian Qian', 'qian qian')<br/>('40596849', 'Jun Yin', 'jun yin')<br/>('2161630', 'Weiming Zeng', 'weiming zeng')<br/>('1692572', 'Miao Song', 'miao song')</td><td>songmiaolm@gmail.com +</td></tr><tr><td>0fb8317a8bf5feaf297af8e9b94c50c5ed0e8277</td><td>Detecting Hands in Egocentric Videos: Towards +<br/>Action Recognition +<br/>Gran Via de les Corts Catalanes, 585, 08007 Barcelona, Spain +<br/><b>University of Barcelona</b><br/>2 Computer Vision Centre, +<br/>Campus UAB, 08193 Cerdanyola del Valls, Barcelona, Spain +</td><td>('1901010', 'Alejandro Cartas', 'alejandro cartas')<br/>('2837527', 'Mariella Dimiccoli', 'mariella dimiccoli')<br/>('1724155', 'Petia Radeva', 'petia radeva')</td><td>alejandro.cartas@ub.edu +</td></tr><tr><td>0fe96806c009e8d095205e8f954d41b2b9fd5dcf</td><td>On-the-Job Learning with Bayesian Decision Theory +<br/>Department of Computer Science +<br/><b>Stanford University</b><br/>Arun Chaganty +<br/>Department of Computer Science +<br/><b>Stanford University</b><br/>Department of Computer Science +<br/><b>Stanford University</b><br/>Department of Computer Science +<br/><b>Stanford University</b></td><td>('2795219', 'Keenon Werling', 'keenon werling')<br/>('40085065', 'Percy Liang', 'percy liang')<br/>('1812612', 'Christopher D. Manning', 'christopher d. manning')</td><td>keenon@cs.stanford.edu +<br/>chaganty@cs.stanford.edu +<br/>pliang@cs.stanford.edu +<br/>manning@cs.stanford.edu +</td></tr><tr><td>0f940d2cdfefc78c92ec6e533a6098985f47a377</td><td>A Hierarchical Framework for Simultaneous Facial Activity Tracking +<br/>Department of Electrical,Computer and System Engineering +<br/><b>Rensselaer Polytechnic Institute</b><br/>Troy, NY 12180 +</td><td>('1713712', 'Jixu Chen', 'jixu chen')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>chenj4@rpi.edu +<br/>qji@ecse.rpi.edu +</td></tr><tr><td>0f21a39fa4c0a19c4a5b4733579e393cb1d04f71</td><td>Evaluation of optimization +<br/>components of a 3D to 2D +<br/>landmark fitting algorithm for +<br/>head pose estimation +<br/>11029668 +<br/>Bachelor thesis +<br/>Credits: 18 EC +<br/>Bachelor Opleiding Kunstmatige Intelligentie +<br/><b>University of Amsterdam</b><br/>Faculty of Science +<br/>Science Park 904 +<br/>1098 XH Amsterdam +<br/>Supervisors +<br/>dr. Sezer Karaoglu +<br/>MSc. Minh Ngo +<br/><b>Informatics Institute</b><br/>Faculty of Science +<br/><b>University of Amsterdam</b><br/>Science Park 904 +<br/>1090 GH Amsterdam +<br/>June 29th, 2018 +</td><td></td><td></td></tr><tr><td>0fd1bffb171699a968c700f206665b2f8837d953</td><td>Weakly Supervised Object Localization with +<br/>Multi-fold Multiple Instance Learning +</td><td>('1939006', 'Ramazan Gokberk Cinbis', 'ramazan gokberk cinbis')<br/>('34602236', 'Jakob Verbeek', 'jakob verbeek')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>0faeec0d1c51623a511adb779dabb1e721a6309b</td><td>Seeing is Worse than Believing: Reading +<br/>People’s Minds Better than Computer-Vision +<br/>Methods Recognize Actions +<br/>1 MIT, Cambridge, MA, USA +<br/><b>Purdue University, West Lafayette, IN, USA</b><br/>3 SUNY Buffalo, Buffalo, NY, USA +<br/><b>Stanford University, Stanford, CA, USA</b><br/><b>University of California at Los Angeles, Los Angeles, CA, USA</b><br/><b>University of Michigan, Ann Arbor, MI, USA</b><br/><b>Princeton University, Princeton, NJ, USA</b><br/><b>Rutgers University, Newark, NJ, USA</b><br/><b>University of Texas at Arlington, Arlington, TX, USA</b><br/><b>National University of Ireland Maynooth, Co. Kildare, Ireland</b></td><td>('21570451', 'Andrei Barbu', 'andrei barbu')<br/>('1728624', 'Wei Chen', 'wei chen')<br/>('2228109', 'Caiming Xiong', 'caiming xiong')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')<br/>('2663295', 'Christiane D. Fellbaum', 'christiane d. fellbaum')<br/>('32218165', 'Catherine Hanson', 'catherine hanson')<br/>('20009336', 'Evguenia Malaia', 'evguenia malaia')<br/>('1700974', 'Barak A. Pearlmutter', 'barak a. pearlmutter')<br/>('2465833', 'Ronnie B. Wilbur', 'ronnie b. wilbur')</td><td>andrei@0xab.com +<br/>{dpbarret,shelie,qobi,tmt,wilbur}@purdue.edu +<br/>wchen23@buffalo.edu +<br/>nsid@stanford.edu +<br/>caimingxiong@ucla.edu +<br/>jjcorso@eecs.umich.edu +<br/>fellbaum@princeton.edu +<br/>{cat,jose}@psychology.rutgers.edu +<br/>malaia@uta.edu +<br/>barak@cs.nuim.ie +</td></tr><tr><td>0f81b0fa8df5bf3fcfa10f20120540342a0c92e5</td><td>Mirror, mirror on the wall, tell me, is the error small? +<br/><b>Queen Mary University of London</b><br/><b>Queen Mary University of London</b></td><td>('2966679', 'Heng Yang', 'heng yang')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')</td><td>heng.yang@qmul.ac.uk +<br/>i.patras@qmul.ac.uk +</td></tr><tr><td>0f0241124d6092a0bb56259ac091467c2c6938ca</td><td>Associating Faces and Names in Japanese Photo News Articles on the Web +<br/><b>The University of Electro-Communications, JAPAN</b></td><td>('32572703', 'Akio Kitahara', 'akio kitahara')<br/>('2558848', 'Taichi Joutou', 'taichi joutou')<br/>('1681659', 'Keiji Yanai', 'keiji yanai')</td><td></td></tr><tr><td>0a6d344112b5af7d1abbd712f83c0d70105211d0</td><td>Constrained Local Neural Fields for robust facial landmark detection in the wild +<br/>Tadas Baltruˇsaitis +<br/><b>University of Cambridge Computer Laboratory</b><br/><b>USC Institute for Creative Technologies</b><br/>15 JJ Thomson Avenue +<br/>12015 Waterfront Drive +</td><td>('40609287', 'Peter Robinson', 'peter robinson')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td>tb346@cl.cam.ac.uk +<br/>pr10@cl.cam.ac.uk +<br/>morency@ict.usc.edu +</td></tr><tr><td>0a64f4fec592662316764283575d05913eb2135b</td><td>Joint Pixel and Feature-level Domain Adaptation in the Wild +<br/><b>Michigan State University</b><br/>2NEC Labs America +<br/>3UC San Diego +</td><td>('1849929', 'Luan Tran', 'luan tran')</td><td></td></tr><tr><td>0a0321785c8beac1cbaaec4d8ad0cfd4a0d6d457</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Learning Invariant Deep Representation +<br/>for NIR-VIS Face Recognition +<br/>National Laboratory of Pattern Recognition, CASIA +<br/>Center for Research on Intelligent Perception and Computing, CASIA +<br/>Center for Excellence in Brain Science and Intelligence Technology, CAS +<br/><b>University of Chinese Academy of Sciences, Beijing 100190, China</b></td><td>('1705643', 'Ran He', 'ran he')<br/>('2225749', 'Xiang Wu', 'xiang wu')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')<br/>('1688870', 'Tieniu Tan', 'tieniu tan')</td><td>{rhe,znsun,tnt}@nlpr.ia.ac.cn, alfredxiangwu@gmail.com +</td></tr><tr><td>0a2ddf88bd1a6c093aad87a8c7f4150bfcf27112</td><td>Patch-based Models For Visual Object Classes +<br/>A dissertation submitted in partial fulfilment +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>at +<br/><b>University College London</b><br/>Department of Computer Science +<br/><b>University College London</b><br/>February 24, 2011 +</td><td>('1904148', 'Jania Aghajanian', 'jania aghajanian')</td><td></td></tr><tr><td>0a5ffc55b584da7918c2650f9d8602675d256023</td><td>Efficient Face Alignment via Locality-constrained Representation for Robust +<br/>Recognition +<br/><b>School of Electronic and Information Engineering, South China University of Technology</b><br/><b>School of Electronic and Computer Engineering, Peking University</b><br/><b>School of Computer Science and Software Engineering, Shenzhen University</b><br/>4SIAT, Chinese Academy of Sciences +</td><td>('36326884', 'Weiyang Liu', 'weiyang liu')</td><td></td></tr><tr><td>0aeb5020003e0c89219031b51bd30ff1bceea363</td><td>Sparsifying Neural Network Connections for Face Recognition +<br/>1SenseTime Group +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('1681656', 'Yi Sun', 'yi sun')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>sunyi@sensetime.com +<br/>xgwang@ee.cuhk.edu.hk +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>0a511058edae582e8327e8b9d469588c25152dc6</td><td></td><td></td><td></td></tr><tr><td>0a4f3a423a37588fde9a2db71f114b293fc09c50</td><td></td><td></td><td></td></tr><tr><td>0aa74ad36064906e165ac4b79dec298911a7a4db</td><td>Variational Inference for the Indian Buffet Process +<br/>Engineering Department +<br/><b>Cambridge University</b><br/>Cambridge, UK +<br/>Engineering Department +<br/><b>Cambridge University</b><br/>Cambridge, UK +<br/>Gatsby Unit +<br/><b>University College London</b><br/>London, UK +<br/>Kurt T. Miller∗ +<br/>Computer Science Division +<br/><b>University of California, Berkeley</b><br/>Berkeley, CA +</td><td>('2292194', 'Finale Doshi-Velez', 'finale doshi-velez')<br/>('1689857', 'Jurgen Van Gael', 'jurgen van gael')<br/>('1725303', 'Yee Whye Teh', 'yee whye teh')</td><td></td></tr><tr><td>0abf67e7bd470d9eb656ea2508beae13ca173198</td><td>Going Deeper into First-Person Activity Recognition +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213, USA +</td><td>('2238622', 'Minghuang Ma', 'minghuang ma')<br/>('2681569', 'Haoqi Fan', 'haoqi fan')<br/>('37991449', 'Kris M. Kitani', 'kris m. kitani')</td><td>minghuam@andrew.cmu.edu haoqif@andrew.cmu.edu kkitani@cs.cmu.edu +</td></tr><tr><td>0af33f6b5fcbc5e718f24591b030250c6eec027a</td><td>Text Analysis for Automatic Image Annotation +<br/>Interdisciplinary Centre for Law & IT +<br/>Department of Computer Science +<br/><b>Katholieke Universiteit Leuven</b><br/>Tiensestraat 41, 3000 Leuven, Belgium +</td><td>('1797588', 'Koen Deschacht', 'koen deschacht')<br/>('1802161', 'Marie-Francine Moens', 'marie-francine moens')</td><td>{koen.deschacht,marie-france.moens}@law.kuleuven.ac.be +</td></tr><tr><td>0a3863a0915256082aee613ba6dab6ede962cdcd</td><td>Early and Reliable Event Detection Using Proximity Space Representation +<br/>LTCI, CNRS, T´el´ecom ParisTech, Universit´e Paris-Saclay, 75013, Paris, France +<br/>J´erˆome Gauthier +<br/>LADIS, CEA, LIST, 91191, Gif-sur-Yvette, France +<br/>Normandie Universit´e, UR, LITIS EA 4108, Avenue de l’universit´e, 76801, Saint-Etienne-du-Rouvray, France +</td><td>('2527457', 'Maxime Sangnier', 'maxime sangnier')<br/>('1792962', 'Alain Rakotomamonjy', 'alain rakotomamonjy')</td><td>MAXIME.SANGNIER@TELECOM-PARISTECH.FR +<br/>JEROME.GAUTHIER@CEA.FR +<br/>ALAIN.RAKOTO@INSA-ROUEN.FR +</td></tr><tr><td>0a60d9d62620e4f9bb3596ab7bb37afef0a90a4f</td><td>Chimpanzee Faces in the Wild: Log-Euclidean CNNs for Predicting Identities and Attributes of Primates. GCPR 2016 +<br/>c(cid:13) Copyright by Springer. The final publication will be available at link.springer.com +<br/>A. Freytag, E. Rodner, M. Simon, A. Loos, H. K¨uhl and J. Denzler +<br/>Chimpanzee Faces in the Wild: +<br/>Log-Euclidean CNNs for Predicting Identities +<br/>and Attributes of Primates +<br/><b>Computer Vision Group, Friedrich Schiller University Jena, Germany</b><br/>2Michael Stifel Center Jena, Germany +<br/><b>Fraunhofer Institute for Digital Media Technology, Germany</b><br/><b>Max Planck Institute for Evolutionary Anthropology, Germany</b><br/>5German Centre for Integrative Biodiversity Research (iDiv), Germany +</td><td>('1720839', 'Alexander Freytag', 'alexander freytag')<br/>('1679449', 'Erik Rodner', 'erik rodner')<br/>('49675890', 'Marcel Simon', 'marcel simon')<br/>('4572597', 'Alexander Loos', 'alexander loos')<br/>('1728382', 'Joachim Denzler', 'joachim denzler')</td><td></td></tr><tr><td>0a34fe39e9938ae8c813a81ae6d2d3a325600e5c</td><td>FacePoseNet: Making a Case for Landmark-Free Face Alignment +<br/><b>Institute for Robotics and Intelligent Systems, USC, CA, USA</b><br/><b>Information Sciences Institute, USC, CA, USA</b><br/><b>The Open University of Israel, Israel</b></td><td>('1752756', 'Feng-Ju Chang', 'feng-ju chang')<br/>('46634688', 'Anh Tuan Tran', 'anh tuan tran')<br/>('1756099', 'Tal Hassner', 'tal hassner')<br/>('11269472', 'Iacopo Masi', 'iacopo masi')</td><td>{fengjuch,anhttran,iacopoma,nevatia,medioni}@usc.edu, hassner@isi.edu +</td></tr><tr><td>0ad8149318912b5449085187eb3521786a37bc78</td><td>CP-mtML: Coupled Projection multi-task Metric Learning +<br/>for Large Scale Face Retrieval +<br/>Frederic Jurie1,∗ +<br/><b>University of Caen, France</b><br/>2MPI for Informatics, Germany +<br/>3IIT Kanpur, India +</td><td>('2078892', 'Binod Bhattarai', 'binod bhattarai')<br/>('2515597', 'Gaurav Sharma', 'gaurav sharma')</td><td></td></tr><tr><td>0a9d204db13d395f024067cf70ac19c2eeb5f942</td><td>Viewpoint-aware Video Summarization +<br/><b>The University of Tokyo, 2RIKEN, 3ETH Z urich, 4KU Leuven</b></td><td>('2551640', 'Atsushi Kanehira', 'atsushi kanehira')<br/>('1681236', 'Luc Van Gool', 'luc van gool')<br/>('3250559', 'Yoshitaka Ushiku', 'yoshitaka ushiku')<br/>('1790553', 'Tatsuya Harada', 'tatsuya harada')</td><td></td></tr><tr><td>0aa9872daf2876db8d8e5d6197c1ce0f8efee4b7</td><td><b>Imperial College of Science, Technology and Medicine</b><br/>Department of Computing +<br/>Timing is everything +<br/>A spatio-temporal approach to the analysis of facial +<br/>actions +<br/>Michel Fran¸cois Valstar +<br/>Submitted in part fulfilment of the requirements for the degree of +<br/><b>Doctor of Philosophy in Computing of Imperial College, February</b></td><td></td><td></td></tr><tr><td>0aae88cf63090ea5b2c80cd014ef4837bcbaadd8</td><td>3D Face Structure Extraction from Images at Arbitrary Poses and under +<br/>Arbitrary Illumination Conditions +<br/>A Thesis +<br/>Submitted to the Faculty +<br/>Of +<br/><b>Drexel University</b><br/>By +<br/>In partial fulfillment of the +<br/>Requirements for the degree +<br/>Of +<br/>Doctor of Philosophy +<br/>October 2006 +</td><td>('40531119', 'Cuiping Zhang', 'cuiping zhang')</td><td></td></tr><tr><td>0a87d781fe2ae2e700237ddd00314dbc10b1429c</td><td>Distribution Statement A: Approved for public release; distribution unlimited. +<br/>Multi-scale HOG Prescreening Algorithm for Detection of Buried +<br/>Explosive Hazards in FL-IR and FL-GPR Data +<br/><b>University of Missouri, Columbia, MO</b></td><td>('2741325', 'K. Stone', 'k. stone')<br/>('9187168', 'J. M. Keller', 'j. m. keller')</td><td></td></tr><tr><td>0ad90118b4c91637ee165f53d557da7141c3fde0</td><td></td><td></td><td></td></tr><tr><td>0a82860d11fcbf12628724333f1e7ada8f3cd255</td><td>Action Temporal Localization in Untrimmed Videos via Multi-stage CNNs +<br/><b>Columbia University</b><br/>New York, NY, USA +</td><td>('2195345', 'Zheng Shou', 'zheng shou')<br/>('2704179', 'Dongang Wang', 'dongang wang')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>{zs2262,dw2648,sc250}@columbia.edu +</td></tr><tr><td>0a4fc9016aacae9cdf40663a75045b71e64a70c9</td><td>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING XX, XXX-XXX (201X) +<br/> Illumination Normalization Based on +<br/>Homomorphic Wavelet Filtering for Face Recognition +<br/>1School of Electronic and Information Engineering +<br/><b>Beijing Jiaotong University</b><br/>No.3 Shang Yuan Cun,Hai Dian District +<br/>Beijing 100044,China +<br/>2School of Physics Electrical Information Engineering +<br/><b>Ningxia University</b><br/>Yinchuan Ningxia 750021,China +<br/>Phone number: 086-010-51688165 +<br/>The performance of face recognition techniques is greatly challenged by the pose, +<br/>expression and illumination of the image. For most existing systems, the recognition rate +<br/>will decrease due to changes in environmental illumination. In this paper, a +<br/>Homomorphic Wavelet-based Illumination Normalization (HWIN) method is proposed. +<br/>The purpose of this method is to normalize the uneven illumination of the facial image. +<br/>The image is analyzed in the logarithm domain with wavelet transform, the +<br/>approximation coefficients of the image are mapped according to the reference +<br/>illumination map in order to normalize the distribution of illumination energy resulting +<br/>from different lighting effects, and the detail components are enhanced to achieve detail +<br/>information emphasis. Then, a Difference of Gaussian (DoG) filter is also applied to +<br/>reduce the noise resulting from different lighting effects, which exists on detail +<br/>components. The proposed methods are implemented on Yale B and Extended Yale B +<br/>facial databases. The experimental results show that the methods described in this study +<br/>are capable of effectively eliminating the effects of uneven illumination and of greatly +<br/>improving the recognition rate, and are therefore more effective than other popular +<br/>methods. +<br/>Keywords: face recognition; homomorphic filtering; wavelet transfer; illumination +<br/>mapping +<br/>1. INTRODUCTION +<br/>Automatic face recognition has received significant attention over the past several +<br/>decades due to its numerous potential applications, such as human-computer interfaces, +<br/>access control, security and surveillance, e-commerce, entertainment, and so on. Related +<br/>research performed in recent years has made great progress, and a number of face +<br/>recognition systems have achieved strong results, as shown in the latest report of Face +<br/>Recognition Vendor Test (FRVT, 2006). Despite this remarkable progress, face +<br/>recognition still faces a challenging problem, which is its sensitivity to the dramatic +<br/>variations among images of the same face. For example, facial expression, pose, ageing, +<br/>make-up, background and illumination variations are all factors which may result in +<br/>significant variations [1-26]. +<br/>Illumination variation is one of the most significant factors limiting the performance +<br/>of face recognition. Since several images of the same person appear to be dramatically +<br/>1 +</td><td>('2613621', 'Xue Yuan', 'xue yuan')<br/>('47884608', 'Yifei Meng', 'yifei meng')</td><td>E-mail: 10111045@bjtu.edu.cn +</td></tr><tr><td>0a85afebaa19c80fddb660110a4352fd22eb2801</td><td>Neural Animation and Reenactment of Human Actor Videos +<br/>Fig. 1. We propose a novel learning-based approach for the animation and reenactment of human actor videos. The top row shows some frames of the video +<br/>We propose a method for generating (near) video-realistic animations of +<br/>real humans under user control. In contrast to conventional human char- +<br/>acter rendering, we do not require the availability of a production-quality +<br/>photo-realistic 3D model of the human, but instead rely on a video sequence +<br/>in conjunction with a (medium-quality) controllable 3D template model +<br/>of the person. With that, our approach significantly reduces production +<br/>cost compared to conventional rendering approaches based on production- +<br/>quality 3D models, and can also be used to realistically edit existing videos. +<br/>Technically, this is achieved by training a neural network that translates +<br/>simple synthetic images of a human character into realistic imagery. For +<br/>training our networks, we first track the 3D motion of the person in the +<br/>video using the template model, and subsequently generate a synthetically +<br/><b>mpg.de, Max Planck Institute for Informatics</b><br/>Permission to make digital or hard copies of part or all of this work for personal or +<br/>classroom use is granted without fee provided that copies are not made or distributed +<br/>for profit or commercial advantage and that copies bear this notice and the full citation +<br/>on the first page. Copyrights for third-party components of this work must be honored. +<br/>For all other uses, contact the owner/author(s). +<br/>© 2018 Copyright held by the owner/author(s). +<br/>XXXX-XXXX/2018/9-ART282 +<br/>https://doi.org/10.475/123_4 +<br/>rendered version of the video. These images are then used to train a con- +<br/>ditional generative adversarial network that translates synthetic images of +<br/>the 3D model into realistic imagery of the human. We evaluate our method +<br/>for the reenactment of another person that is tracked in order to obtain the +<br/>motion data, and show video results generated from artist-designed skeleton +<br/>motion. Our results outperform the state-of-the-art in learning-based human +<br/>image synthesis. +<br/>CCS Concepts: • Computing methodologies → Computer graphics; +<br/>Neural networks; Appearance and texture representations; Animation; Ren- +<br/>dering; +<br/>Additional Key Words and Phrases: Video-based Characters, Deep Learning, +<br/>Conditional GAN, Rendering-to-Video Translation +<br/>ACM Reference Format: +<br/>Animation and Reenactment of Human Actor Videos. 1, 1, Article 282 +<br/>(September 2018), 13 pages. https://doi.org/10.475/123_4 +<br/>INTRODUCTION +<br/>The creation of realistically rendered and controllable animations +<br/>of human characters is a crucial task in many computer graphics +<br/>applications. Virtual actors play a key role in games and visual ef- +<br/>fects, in telepresence, or in virtual and augmented reality. Today, the +<br/>plausible rendition of video-realistic characters—i.e., animations in- +<br/>distinguishable from a video of a human—under user control is also +<br/>Submission ID: 282. 2018-09-12 00:32. Page 1 of 1–13. +<br/>, Vol. 1, No. 1, Article 282. Publication date: September 2018. +</td><td>('46458089', 'Lingjie Liu', 'lingjie liu')<br/>('9765909', 'Weipeng Xu', 'weipeng xu')<br/>('1699058', 'Michael Zollhöfer', 'michael zollhöfer')<br/>('3022958', 'Hyeongwoo Kim', 'hyeongwoo kim')<br/>('39600032', 'Florian Bernard', 'florian bernard')<br/>('14210288', 'Marc Habermann', 'marc habermann')<br/>('1698520', 'Wenping Wang', 'wenping wang')<br/>('1680185', 'Christian Theobalt', 'christian theobalt')<br/>('3022958', 'Hyeongwoo Kim', 'hyeongwoo kim')<br/>('46458089', 'Lingjie Liu', 'lingjie liu')<br/>('9765909', 'Weipeng Xu', 'weipeng xu')<br/>('1699058', 'Michael Zollhöfer', 'michael zollhöfer')<br/>('3022958', 'Hyeongwoo Kim', 'hyeongwoo kim')<br/>('39600032', 'Florian Bernard', 'florian bernard')<br/>('14210288', 'Marc Habermann', 'marc habermann')<br/>('1698520', 'Wenping Wang', 'wenping wang')<br/>('1680185', 'Christian Theobalt', 'christian theobalt')</td><td>Authors’ addresses: Lingjie Liu, liulingjie0206@gmail.com, University of Hong Kong, +<br/>Max Planck Institute for Informatics; Weipeng Xu, wxu@mpi-inf.mpg.de, Max Planck +<br/>Institute for Informatics; Michael Zollhöfer, zollhoefer@cs.stanford.edu, Stanford +<br/>kim@mpi-inf.mpg.de; Florian Bernard, fbernard@mpi-inf.mpg.de; Marc Habermann, +<br/>mhaberma@mpi-inf.mpg.de, Max Planck Institute for Informatics; Wenping Wang, +<br/>wenping@cs.hku.hk, University of Hong Kong; Christian Theobalt, theobalt@mpi-inf. +</td></tr><tr><td>0ac442bb570b086d04c4d51a8410fcbfd0b1779d</td><td>WarpNet: Weakly Supervised Matching for Single-view Reconstruction +<br/><b>University of Maryland, College Park</b><br/>Manmohan Chandraker +<br/>NEC Labs America +</td><td>('20615377', 'Angjoo Kanazawa', 'angjoo kanazawa')<br/>('34734622', 'David W. Jacobs', 'david w. jacobs')</td><td></td></tr><tr><td>0af48a45e723f99b712a8ce97d7826002fe4d5a5</td><td>2982 +<br/>Toward Wide-Angle Microvision Sensors +<br/>Todd Zickler, Member, IEEE +</td><td>('2724462', 'Sanjeev J. Koppal', 'sanjeev j. koppal')<br/>('2407724', 'Ioannis Gkioulekas', 'ioannis gkioulekas')<br/>('3091204', 'Travis Young', 'travis young')<br/>('2070262', 'Hyunsung Park', 'hyunsung park')<br/>('2140759', 'Kenneth B. Crozier', 'kenneth b. crozier')<br/>('40431923', 'Geoffrey L. Barrows', 'geoffrey l. barrows')</td><td></td></tr><tr><td>0aa8a0203e5f406feb1815f9b3dd49907f5fd05b</td><td>Mixture subclass discriminant analysis +</td><td>('1827419', 'Nikolaos Gkalelis', 'nikolaos gkalelis')<br/>('1737436', 'Vasileios Mezaris', 'vasileios mezaris')</td><td></td></tr><tr><td>0ac664519b2b8abfb8966dafe60d093037275573</td><td>Facial Action Unit Detection Using Kernel Partial Least Squares - +<br/>Supplemental Material +<br/><b>Facial Image Processing and Analysis Group, Institute for Anthropomatics</b><br/><b>Karlsruhe Institute of Technology</b><br/>D-76131 Karlsruhe, P.O. Box 6980 Germany +<br/>1. Introduction +<br/>In this document we present additional results corre- +<br/>sponding to the experiments shown in [1]. +<br/>A. ROC Curves +<br/>The ROC curves for the AU estimates are shown in this +<br/>section. +<br/>A.1. Evaluation on a Single Dataset +<br/>A.1.1 Experiment on the CK+ Dataset with Eye Labels +<br/>See Figure 1. +<br/>A.1.2 Experiment on the CK+ Dataset with Automatic +<br/>Eye Detection +<br/>See Figure 2. +<br/>A.1.3 Experiment on the GEMEP-FERA Dataset +<br/>See Figure 3. +<br/>A.2. Evaluation across Datasets +<br/>A.2.1 Generalization from Constrained to less Con- +<br/>strained Condition +<br/>See Figure 4. +<br/>A.2.2 Generalization from less Constrained to Con- +<br/>strained Condition +<br/>See Figure 5. +<br/>B. F1-Score +<br/>The F1-Scores for the AU estimates are shown in this +<br/>section. If no threshold optimization is performed then the +<br/>thresholds are set to 0.5 for the PLS-based approaches and +<br/>Table 1. F1 scores in % on CK+ using eye labels. AVG is the +<br/>weighted average over the individual results, depending on the +<br/>number of positive samples given by in the column N. +<br/>linear PLS +<br/>RBF PLS +<br/>linear SVM RBF SVM +<br/>176 +<br/>117 +<br/>193 +<br/>102 +<br/>123 +<br/>120 +<br/>75 +<br/>34 +<br/>131 +<br/>94 +<br/>201 +<br/>79 +<br/>60 +<br/>58 +<br/>324 +<br/>50 +<br/>81 +<br/>AU +<br/>11 +<br/>12 +<br/>15 +<br/>17 +<br/>20 +<br/>23 +<br/>24 +<br/>25 +<br/>26 +<br/>27 +<br/>AVG +<br/>78.1 +<br/>80.4 +<br/>74.2 +<br/>77.5 +<br/>72.8 +<br/>64.0 +<br/>84.3 +<br/>15.0 +<br/>84.7 +<br/>60.3 +<br/>77.4 +<br/>64.8 +<br/>35.2 +<br/>38.2 +<br/>85.4 +<br/>15.6 +<br/>85.9 +<br/>72.3 +<br/>77.5 +<br/>76.2 +<br/>75.9 +<br/>76.2 +<br/>68.2 +<br/>51.0 +<br/>84.2 +<br/>5.7 +<br/>81.9 +<br/>51.5 +<br/>78.3 +<br/>57.1 +<br/>28.6 +<br/>26.7 +<br/>86.5 +<br/>7.4 +<br/>83.0 +<br/>69.5 +<br/>69.6 +<br/>78.9 +<br/>72.8 +<br/>74.3 +<br/>67.0 +<br/>51.9 +<br/>84.5 +<br/>14.6 +<br/>78.3 +<br/>52.6 +<br/>73.6 +<br/>49.6 +<br/>28.9 +<br/>14.1 +<br/>86.5 +<br/>5.9 +<br/>84.6 +<br/>67.4 +<br/>71.5 +<br/>76.7 +<br/>68.0 +<br/>73.8 +<br/>65.7 +<br/>42.3 +<br/>83.0 +<br/>0.0 +<br/>80.0 +<br/>49.6 +<br/>76.8 +<br/>28.0 +<br/>14.3 +<br/>9.0 +<br/>86.1 +<br/>0.0 +<br/>77.7 +<br/>64.4 +<br/>0.0 for the SVM-based approaches. Otherwise thresholds +<br/>are optimized using equal error rate (EER) or F1 score as +<br/>metrics [2] on either the training folds of the LOSO scheme +<br/>or the whole training data in case of the cross-dataset tests. +<br/>B.1. Evaluation on a Single Dataset +<br/>B.1.1 Experiment on the CK+ Dataset with Eye Labels +<br/>See Table 1 for F1 scores without threshold optimization, +<br/>Table 2 for F1 scores using threshold optimization based on +<br/>EER and Table 3 for F1 scores using threshold optimization +<br/>based on F1 score. +<br/>B.1.2 Experiment on the CK+ Dataset with Automatic +<br/>Eye Detection +<br/>See Table 4 for F1 scores without threshold optimization, +<br/>Table 5 for F1 scores using threshold optimization based on +<br/>EER and Table 6 for F1 scores using threshold optimization +<br/>based on F1 score. +</td><td>('40303076', 'Tobias Gehrig', 'tobias gehrig')</td><td>{tobias.gehrig, ekenel}@kit.edu +</td></tr><tr><td>0a9345ea6e488fb936e26a9ba70b0640d3730ba7</td><td>Deep Bi-directional Cross-triplet Embedding for +<br/>Cross-Domain Clothing Retrieval +<br/><b>Northeastern University, Boston, USA</b><br/><b>College of Computer and Information Science, Northeastern University, Boston, USA</b></td><td>('3343578', 'Shuhui Jiang', 'shuhui jiang')<br/>('1746738', 'Yue Wu', 'yue wu')<br/>('37771688', 'Yun Fu', 'yun fu')</td><td>{shjiang, yuewu, yunfu}@ece.neu.edu +</td></tr><tr><td>0a79d0ba1a4876086e64fc0041ece5f0de90fbea</td><td>FACE ILLUMINATION NORMALIZATION +<br/>WITH SHADOW CONSIDERATION +<br/>By +<br/>SUBMITTED IN PARTIAL FULFILLMENT OF THE +<br/>REQUIREMENTS FOR THE DEGREE OF +<br/>MASTER OF SCIENCE +<br/>AT +<br/><b>CARNEGIE MELLON UNIVERSITY</b><br/>5000 FORBES AVENUE PITTSBURGH PA 15213-3890 +<br/>MAY 2004 +</td><td>('3039721', 'Avinash B. Baliga', 'avinash b. baliga')<br/>('3039721', 'Avinash B. Baliga', 'avinash b. baliga')</td><td></td></tr><tr><td>0a7309147d777c2f20f780a696efe743520aa2db</td><td>Stories for Images-in-Sequence by using Visual +<br/>and Narrative Components (cid:63) +<br/><b>Ss. Cyril and Methodius University, Skopje, Macedonia</b><br/>2 Pendulibrium, Skopje, Macedonia +<br/>3 Elevate Global, Skopje, Macedonia +</td><td>('46205557', 'Marko Smilevski', 'marko smilevski')<br/>('46242132', 'Ilija Lalkovski', 'ilija lalkovski')</td><td>{marko.smilevski,ilija}@webfactory.mk, gjorgji.madjarov@finki.ukim.mk +</td></tr><tr><td>0a11b82aa207d43d1b4c0452007e9388a786be12</td><td>Feature Level Multiple Model Fusion Using Multilinear +<br/>Subspace Analysis with Incomplete Training Set +<br/>and Its Application to Face Image Analysis +<br/><b>School of IoT Engineering, Jiangnan University, Wuxi, 214122, China</b><br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, Guildford, GU2 7XH</b><br/>United Kingdom +</td><td>('2976854', 'Zhen-Hua Feng', 'zhen-hua feng')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td>xiaojun wu jnu@163.com +<br/>{Z.Feng,J.Kittler,W.Christmas}@surrey.ac.uk +</td></tr><tr><td>0a1138276c52c734b67b30de0bf3f76b0351f097</td><td>This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. +<br/>The final version of record is available at +<br/> http://dx.doi.org/10.1109/TIP.2016.2539502 +<br/>Discriminant Incoherent Component Analysis +</td><td>('2812961', 'Christos Georgakis', 'christos georgakis')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>0a6a25ee84fc0bf7284f41eaa6fefaa58b5b329a</td><td></td><td>('1802883', 'Soufiane Belharbi', 'soufiane belharbi')</td><td></td></tr><tr><td>0ae9cc6a06cfd03d95eee4eca9ed77b818b59cb7</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Multi-task, multi-label and multi-domain learning with +<br/>residual convolutional networks for emotion recognition +<br/>Received: date / Accepted: date +</td><td>('10157512', 'Gerard Pons', 'gerard pons')</td><td></td></tr><tr><td>0acf23485ded5cb9cd249d1e4972119239227ddb</td><td>Dual coordinate solvers for large-scale structural SVMs +<br/>UC Irvine +<br/>This manuscript describes a method for training linear SVMs (including binary SVMs, SVM regression, +<br/>and structural SVMs) from large, out-of-core training datasets. Current strategies for large-scale learning fall +<br/>into one of two camps; batch algorithms which solve the learning problem given a finite datasets, and online +<br/>algorithms which can process out-of-core datasets. The former typically requires datasets small enough to fit +<br/>in memory. The latter is often phrased as a stochastic optimization problem [4, 15]; such algorithms enjoy +<br/>strong theoretical properties but often require manual tuned annealing schedules, and may converge slowly +<br/>for problems with large output spaces (e.g., structural SVMs). We discuss an algorithm for an “intermediate” +<br/>regime in which the data is too large to fit in memory, but the active constraints (support vectors) are small +<br/>enough to remain in memory. +<br/>In this case, one can design rather efficient learning algorithms that are +<br/>as stable as batch algorithms, but capable of processing out-of-core datasets. We have developed such a +<br/>MATLAB-based solver and used it to train a series of recognition systems [19, 7, 21, 12] for articulated pose +<br/>estimation, facial analysis, 3D object recognition, and action classification, all with publicly-available code. +<br/>This writeup describes the solver in detail. +<br/>Approach: Our approach is closely based on data-subsampling algorithms for collecting hard exam- +<br/>ples [9, 10, 6], combined with the dual coordinate quadratic programming (QP) solver described in liblinear +<br/>[8]. The latter appears to be current fastest method for learning linear SVMs. We make two extensions (1) +<br/>We show how to generalize the solver to other types of SVM problems such as (latent) structural SVMs (2) +<br/>We show how to modify it to behave as a partially-online algorithm, which only requires access to small +<br/>amounts of data at a time. +<br/>Overview: Sec. 1 describes a general formulation of an SVM problem that encompasses many standard +<br/>tasks such as multi-class classification and (latent) structural prediction. Sec. 2 derives its dual QP, and Sec. 3 +<br/>describes a dual coordinate descent optimization algorithm. Sec. 4 describes modifications for optimizing +<br/>in an online fashion, allowing one to learn near-optimal models with a single pass over large, out-of-core +<br/>datasets. Sec. 5 briefly touches on some theoretical issues that are necessary to ensure convergence. Finally, +<br/>Sec. 6 and Sec. 7 describe modifications to our basic formulation to accommodate non-negativity constraints +<br/>and flexible regularization schemes during learning. +<br/>1 Generalized SVMs +<br/>We first describe a general formulation of a SVM which encompasses various common problems such as +<br/>binary classification, regression, and structured prediction. Assume we are given training data where the ith +<br/>example is described by a set of Ni vectors {xij} and a set of Ni scalars {lij}, where j varies from 1 to Ni. +<br/>We wish to solve the following optimization problem: +<br/>(0, lij − wT xij) +<br/>max +<br/>j∈Ni +<br/>(1) +<br/>(cid:88) +<br/>argmin +<br/>L(w) = +<br/>||w||2 + +</td><td>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td></td></tr><tr><td>0ad4a814b30e096ad0e027e458981f812c835aa0</td><td></td><td></td><td></td></tr><tr><td>6448d23f317babb8d5a327f92e199aaa45f0efdc</td><td></td><td></td><td></td></tr><tr><td>6412d8bbcc01f595a2982d6141e4b93e7e982d0f</td><td>Deep Convolutional Neural Network using Triplets of Faces, Deep Ensemble, and +<br/>Score-level Fusion for Face Recognition +<br/>1Department of Creative IT Engineering, POSTECH, Korea +<br/>2Department of Computer Science and Engineering, POSTECH, Korea +</td><td>('2794366', 'Bong-Nam Kang', 'bong-nam kang')<br/>('1804861', 'Yonghyun Kim', 'yonghyun kim')<br/>('1695669', 'Daijin Kim', 'daijin kim')</td><td>{bnkang, gkyh0805, dkim}@postech.ac.kr +</td></tr><tr><td>641f0989b87bf7db67a64900dcc9568767b7b50f</td><td>Reconstructing Faces from their Signatures using RBF +<br/>Regression +<br/>To cite this version: +<br/>sion. British Machine Vision Conference 2013, Sep 2013, Bristol, United Kingdom. pp.103.1– +<br/>103.12, 2013, <10.5244/C.27.103>. <hal-00943426> +<br/>HAL Id: hal-00943426 +<br/>https://hal.archives-ouvertes.fr/hal-00943426 +<br/>Submitted on 13 Feb 2014 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('34723309', 'Alexis Mignon', 'alexis mignon')<br/>('34723309', 'Alexis Mignon', 'alexis mignon')</td><td></td></tr><tr><td>6409b8879c7e61acf3ca17bcc62f49edca627d4c</td><td>Learning Finite Beta-Liouville Mixture Models via +<br/>Variational Bayes for Proportional Data Clustering +<br/>Electrical and Computer Engineering +<br/><b>Institute for Information Systems Engineering</b><br/><b>Concordia University, Canada</b><br/><b>Concordia University, Canada</b></td><td>('2038786', 'Wentao Fan', 'wentao fan')<br/>('1729109', 'Nizar Bouguila', 'nizar bouguila')</td><td>wenta fa@encs.concordia.ca +<br/>nizar.bouguila@concordia.ca +</td></tr><tr><td>64153df77fe137b7c6f820a58f0bdb4b3b1a879b</td><td>Shape Invariant Recognition of Segmented Human +<br/>Faces using Eigenfaces +<br/>Department of Informatics +<br/><b>Technical University of Munich, Germany</b></td><td>('1725709', 'Zahid Riaz', 'zahid riaz')<br/>('1746229', 'Michael Beetz', 'michael beetz')<br/>('1699132', 'Bernd Radig', 'bernd radig')</td><td>{riaz,beetz,radig}@in.tum.de +</td></tr><tr><td>649eb674fc963ce25e4e8ce53ac7ee20500fb0e3</td><td></td><td></td><td></td></tr><tr><td>64ec0c53dd1aa51eb15e8c2a577701e165b8517b</td><td>Online Regression with Feature Selection in +<br/>Stochastic Data Streams +<br/><b>Florida State University</b><br/><b>Florida State University</b></td><td>('5517409', 'Lizhe Sun', 'lizhe sun')<br/>('2455529', 'Adrian Barbu', 'adrian barbu')</td><td>lizhe.sun@stat.fsu.edu +<br/>abarbu@stat.fsu.edu +</td></tr><tr><td>642c66df8d0085d97dc5179f735eed82abf110d0</td><td></td><td></td><td></td></tr><tr><td>6459f1e67e1ea701b8f96177214583b0349ed964</td><td>GENERALIZED SUBSPACE BASED HIGH DIMENSIONAL DENSITY ESTIMATION +<br/><b>University of California Santa Barbara</b><br/><b>University of California Santa Barbara</b></td><td>('3231876', 'Karthikeyan Shanmuga Vadivel', 'karthikeyan shanmuga vadivel')</td><td>(cid:63){karthikeyan,msargin,sjoshi,manj}@ece.ucsb.edu +<br/>†grafton@psych.ucsb.edu +</td></tr><tr><td>64cf86ba3b23d3074961b485c16ecb99584401de</td><td>Single Image 3D Interpreter Network +<br/><b>Massachusetts Institute of Technology</b><br/><b>Stanford University</b><br/>3Facebook AI Research +<br/>4Google Research +</td><td>('3045089', 'Jiajun Wu', 'jiajun wu')<br/>('3222730', 'Tianfan Xue', 'tianfan xue')<br/>('35198686', 'Joseph J. Lim', 'joseph j. lim')<br/>('39402399', 'Yuandong Tian', 'yuandong tian')<br/>('1763295', 'Joshua B. Tenenbaum', 'joshua b. tenenbaum')<br/>('1690178', 'Antonio Torralba', 'antonio torralba')<br/>('1768236', 'William T. Freeman', 'william t. freeman')</td><td></td></tr><tr><td>6424b69f3ff4d35249c0bb7ef912fbc2c86f4ff4</td><td>Deep Learning Face Attributes in the Wild∗ +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b></td><td>('3243969', 'Ziwei Liu', 'ziwei liu')<br/>('1693209', 'Ping Luo', 'ping luo')</td><td>{lz013,pluo,xtang}@ie.cuhk.edu.hk, xgwang@ee.cuhk.edu.hk +</td></tr><tr><td>6479b61ea89e9d474ffdefa71f068fbcde22cc44</td><td><b>University of Exeter</b><br/>Department of Computer Science +<br/>Some Topics on Similarity Metric Learning +<br/>June 2015 +<br/>Supervised by Dr. Yiming Ying +<br/>Philosophy in Computer Science , June 2015. +<br/>This thesis is available for Library use on the understanding that it is copyright material +<br/>and that no quotation from the thesis may be published without proper acknowledgement. +<br/>I certify that all material in this thesis which is not my own work has been identified and +<br/>that no material has previously been submitted and approved for the award of a degree by this or +<br/><b>any other University</b><br/>(signature) ................................................................................................. +</td><td>('1954340', 'Qiong Cao', 'qiong cao')<br/>('1954340', 'Qiong Cao', 'qiong cao')</td><td></td></tr><tr><td>64e75f53ff3991099c3fb72ceca55b76544374e5</td><td>Simultaneous Feature Selection and Classifier Training via Linear +<br/>Programming: A Case Study for Face Expression Recognition +<br/>Computer Sciences Department +<br/><b>University of Wisconsin-Madison</b><br/>Madison, WI 53706 +</td><td>('1822413', 'Guodong Guo', 'guodong guo')<br/>('1724754', 'Charles R. Dyer', 'charles r. dyer')</td><td>fgdguo, dyerg@cs.wisc.edu +</td></tr><tr><td>64f9519f20acdf703984f02e05fd23f5e2451977</td><td>Learning Temporal Alignment Uncertainty for +<br/>Efficient Event Detection +<br/><b>Image and Video Laboratory, Queensland University of Technology (QUT), Brisbane, QLD, Australia</b><br/><b>The Robotics Institute, Carnegie Mellon University, 5000 Forbes Ave, PA, USA</b></td><td>('2838646', 'Iman Abbasnejad', 'iman abbasnejad')<br/>('1729760', 'Sridha Sridharan', 'sridha sridharan')<br/>('1980700', 'Simon Denman', 'simon denman')<br/>('3140440', 'Clinton Fookes', 'clinton fookes')<br/>('1820249', 'Simon Lucey', 'simon lucey')</td><td>Email:{i.abbasnejad, s.sridharan, s.denman, c.fookes}@qut.edu.au, slucey@cs.cmu.edu +</td></tr><tr><td>641f34deb3bdd123c6b6e7b917519c3e56010cb7</td><td></td><td></td><td></td></tr><tr><td>64782a2bc5da11b1b18ca20cecf7bdc26a538d68</td><td>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING XX, XXX-XXX (2011) +<br/>Facial Expression Recognition using +<br/>Spectral Supervised Canonical Correlation Analysis* +<br/><b>Institute of Information Science</b><br/><b>Beijing Jiaotong University</b><br/>Beijing, 100044 P.R. China +<br/>Feature extraction plays an important role in facial expression recognition. Canoni- +<br/>cal correlation analysis (CCA), which studies the correlation between two random vec- +<br/>tors, is a major linear feature extraction method based on feature fusion. Recent studies +<br/>have shown that facial expression images often reside on a latent nonlinear manifold. +<br/>However, either CCA or its kernel version KCCA, which is globally linear or nonlinear, +<br/>cannot effectively utilize the local structure information to discover the low-dimensional +<br/>manifold embedded in the original data. Inspired by the successful application of spectral +<br/>graph theory in classification, we proposed spectral supervised canonical correlation +<br/>analysis (SSCCA) to overcome the shortcomings of CCA and KCCA. In SSCCA, we +<br/>construct an affinity matrix, which incorporates both the class information and local +<br/>structure information of the data points, as the supervised matrix. The spectral feature of +<br/>covariance matrices is used to extract a new combined feature with more discriminative +<br/>information, and it can reveal the nonlinear manifold structure of the data. Furthermore, +<br/>we proposed a unified framework for CCA to offer an effective methodology for +<br/>non-empirical structural comparison of different forms of CCA as well as providing a +<br/>way to extend the CCA algorithm. The correlation feature extraction power is then pro- +<br/>posed to evaluate the effectiveness of our method. Experimental results on two facial ex- +<br/>pression databases validate the effectiveness of our method. +<br/>Keywords: spectral supervised canonical correlation analysis, spectral classification, fea- +<br/>ture fusion, feature extraction, facial expression recognition +<br/>1. INTRODUCTION +<br/>Facial expression conveys visual human emotions, which makes the facial expres- +<br/>sion recognition (FER) plays an important role in human–computer interaction, image +<br/>retrieval, synthetic face animation, video conferencing, human emotion analysis [1, 2]. +<br/>Due to its wide range of applications, FER has attracted much attention in recent years. +<br/>Generally speaking, a FER system consists of three major components: face detection, +<br/>facial expression feature extraction and facial expression classification [1, 2]. Since ap- +<br/>propriate facial expression representation can effectively alleviate the complexity of the +<br/>design of classification and improve the performance of the FER system, most researches +<br/>currently concentrate on how to extract effective facial expression features. +<br/>A variety of methods have been proposed for facial expression feature extraction +<br/>[3-7], and there are generally two common approaches: single feature extraction and +<br/>feature fusion. Single feature extraction is based on a particular method, i.e. principal +<br/>component analysis (PCA) [3], fisher’s linear discriminant (FLD) [4], locality preserving +<br/>*This paper was supported by the National Natural Science Foundation of China (Grant No.60973060), Spe- +<br/>cialized Research Fund for the Doctoral Program of Higher Education (Grant No. 200800040008), Beijing +<br/>Program (Grant No. YB20081000401) and the Fundamental Research Funds for the Central Universities +<br/>(Grant No. 2011JBM022). +<br/>1 +</td><td>('1701978', 'Song Guo', 'song guo')<br/>('1738408', 'Qiuqi Ruan', 'qiuqi ruan')<br/>('1718667', 'Zhan Wang', 'zhan wang')<br/>('1702894', 'Shuai Liu', 'shuai liu')</td><td></td></tr><tr><td>645de797f936cb19c1b8dba3b862543645510544</td><td>Deep Temporal Linear Encoding Networks +<br/>1ESAT-PSI, KU Leuven, 2CVL, ETH Z¨urich +</td><td>('3310120', 'Ali Diba', 'ali diba')<br/>('50633941', 'Vivek Sharma', 'vivek sharma')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{firstname.lastname}@esat.kuleuven.be +</td></tr><tr><td>6462ef39ca88f538405616239471a8ea17d76259</td><td></td><td></td><td></td></tr><tr><td>64d5772f44efe32eb24c9968a3085bc0786bfca7</td><td>Morphable Displacement Field Based Image +<br/>Matching for Face Recognition across Pose +<br/>1 Key Lab of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>CAS), Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>Graduate University of Chinese Academy of Sciences, Beijing 100049, China</b><br/>3 Omron Social Solutions Co., LTD., Kyoto, Japan +</td><td>('1688086', 'Shaoxin Li', 'shaoxin li')<br/>('1731144', 'Xin Liu', 'xin liu')<br/>('1695600', 'Xiujuan Chai', 'xiujuan chai')<br/>('1705483', 'Haihong Zhang', 'haihong zhang')<br/>('1710195', 'Shihong Lao', 'shihong lao')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')</td><td>{shaoxin.li,xiujuan.chai,xin.liu,shiguang.shan}@vipl.ict.ac.cn, +<br/>lao@ari.ncl.omron.co.jp, angelazhang@ssb.kusatsu.omron.co.jp +</td></tr><tr><td>64d7e62f46813b5ad08289aed5dc4825d7ec5cff</td><td>YAMAGUCHI et al.: MIX AND MATCH +<br/>Mix and Match: Joint Model for Clothing and +<br/>Attribute Recognition +<br/>http://vision.is.tohoku.ac.jp/~kyamagu +<br/><b>Tohoku University</b><br/>Sendai, Japan +<br/>2 NTT +<br/>Yokosuka, Japan +<br/><b>Tokyo University of Science</b><br/>Tokyo, Japan +</td><td>('1721910', 'Kota Yamaguchi', 'kota yamaguchi')<br/>('1718872', 'Takayuki Okatani', 'takayuki okatani')<br/>('1745497', 'Kyoko Sudo', 'kyoko sudo')<br/>('2023568', 'Kazuhiko Murasaki', 'kazuhiko murasaki')<br/>('2113938', 'Yukinobu Taniguchi', 'yukinobu taniguchi')</td><td>okatani@vision.is.tohoku.ac.jp +<br/>sudo.kyoko@lab.ntt.co.jp +<br/>murasaki.kazuhiko@lab.ntt.co.jp +<br/>ytaniguti@ms.kagu.tus.ac.jp +</td></tr><tr><td>90ac0f32c0c29aa4545ed3d5070af17f195d015f</td><td></td><td></td><td></td></tr><tr><td>90d735cffd84e8f2ae4d0c9493590f3a7d99daf1</td><td>Original Research Paper +<br/>American Journal of Engineering and Applied Sciences +<br/>Recognition of Faces using Efficient Multiscale Local Binary +<br/>Pattern and Kernel Discriminant Analysis in Varying +<br/>Environment +<br/>V.H. Mankar +<br/><b>Priyadarshini College of Engg, Nagpur, India</b><br/>2Department of Electronics Engg, Government Polytechnic, Nagpur, India +<br/>Article history +<br/>Received: 20-06-2017 +<br/>Revised: 18-07-2017 +<br/>Accepted: 21-08-2017 +<br/>Corresponding Author: +<br/>Department of Electronics +<br/><b>Engg, Priyadarshini College of</b><br/>Engg, Nagpur, India +<br/>face +</td><td>('9128944', 'Sujata G. Bhele', 'sujata g. bhele')<br/>('9128944', 'Sujata G. Bhele', 'sujata g. bhele')</td><td>Email: sujata_bhele@yahoo.co.in +</td></tr><tr><td>90298f9f80ebe03cb8b158fd724551ad711d4e71</td><td>A Pursuit of Temporal Accuracy in General Activity Detection +<br/><b>The Chinese University of Hong Kong</b><br/>2Computer Vision Laboratory, ETH Zurich, Switzerland +</td><td>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('1695765', 'Yue Zhao', 'yue zhao')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('1807606', 'Dahua Lin', 'dahua lin')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>900207b3bc3a4e5244cae9838643a9685a84fee0</td><td>Reconstructing Geometry from Its Latent Structures +<br/>A Thesis +<br/>Submitted to the Faculty +<br/>of +<br/><b>Drexel University</b><br/>by +<br/>Geoffrey Oxholm +<br/>in partial fulfillment of the +<br/>requirements for the degree +<br/>of +<br/>Doctor of Philosophy +<br/>June 2014 +</td><td></td><td></td></tr><tr><td>90498b95fe8b299ce65d5cafaef942aa58bd68b7</td><td>Face Recognition: Primates in the Wild∗ +<br/><b>Michigan State University, East Lansing, MI, USA</b><br/><b>University of Chester, UK, 3Conservation Biologist</b></td><td>('32623642', 'Debayan Deb', 'debayan deb')<br/>('46516859', 'Susan Wiper', 'susan wiper')<br/>('9658130', 'Sixue Gong', 'sixue gong')<br/>('9644181', 'Yichun Shi', 'yichun shi')<br/>('41022894', 'Cori Tymoszek', 'cori tymoszek')</td><td>E-mail: 1{debdebay, gongsixu, shiyichu, tymoszek, jain}@cse.msu.edu, +<br/>2s.wiper@chester.ac.uk, 3alexandra.h.russo@gmail.com +</td></tr><tr><td>90cc2f08a6c2f0c41a9dd1786bae097f9292105e</td><td>Top-down Attention Recurrent VLAD Encoding +<br/>for Action Recognition in Videos +<br/>1 Fondazione Bruno Kessler, Trento, Italy +<br/><b>University of Trento, Trento, Italy</b></td><td>('1756362', 'Swathikiran Sudhakaran', 'swathikiran sudhakaran')<br/>('1717522', 'Oswald Lanz', 'oswald lanz')</td><td>{sudhakaran,lanz}@fbk.eu +</td></tr><tr><td>90fb58eeb32f15f795030c112f5a9b1655ba3624</td><td>INTERNATIONAL JOURNAL OF RESEARCH IN COMPUTER APPLICATIONS AND ROBOTICS +<br/> www.ijrcar.com +<br/>Vol.4 Issue 6, Pg.: 12-27 +<br/>June 2016 +<br/>INTERNATIONAL JOURNAL OF +<br/>RESEARCH IN COMPUTER +<br/>APPLICATIONS AND ROBOTICS +<br/>ISSN 2320-7345 +<br/>FACE AND IRIS RECOGNITION IN A +<br/>VIDEO SEQUENCE USING DBPNN AND +<br/>ADAPTIVE HAMMING DISTANCE +<br/><b>PG Scholar, Hindusthan College of Engineering and Technology, Coimbatore, India</b><br/><b>Hindusthan College of Engineering and Technology, Coimbatore, India</b></td><td>('3406423', 'S. Revathy', 's. revathy')</td><td>Email id: revathysreeni14@gmail.com +</td></tr><tr><td>90c4f15f1203a3a8a5bf307f8641ba54172ead30</td><td>A 2D Morphable Model of Craniofacial Profile +<br/>and Its Application to Craniosynostosis +<br/><b>University of York, York, UK</b><br/>2 Alder Hey Craniofacial Unit, Liverpool, UK +<br/>https://www-users.cs.york.ac.uk/~nep/research/LYHM/ +</td><td>('1694260', 'Hang Dai', 'hang dai')<br/>('1737428', 'Nick Pears', 'nick pears')<br/>('14154312', 'Christian Duncan', 'christian duncan')</td><td>{hd816,nick.pears}@york.ac.uk +<br/>Christian.Duncan@alderhey.nhs.uk +</td></tr><tr><td>902114feaf33deac209225c210bbdecbd9ef33b1</td><td>KAN et al.: SIDE-INFORMATION BASED LDA FOR FACE RECOGNITION +<br/>Side-Information based Linear +<br/>Discriminant Analysis for Face +<br/>Recognition +<br/><b>Digital Media Research Center</b><br/><b>Institute of Computing</b><br/>Technology, CAS, Beijing, China +<br/>2 Key Laboratory of Intelligent +<br/>Information Processing, Chinese +<br/>Academy of Sciences, Beijing, +<br/>China +<br/>3 School of Computer Engineering, +<br/>Nanyang Technological +<br/><b>University, Singapore</b></td><td>('1693589', 'Meina Kan', 'meina kan')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1714390', 'Dong Xu', 'dong xu')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>mnkan@jdl.ac.cn +<br/>sgshan@jdl.ac.cn +<br/>dongxu@ntu.edu.sg +<br/>xlchen@jdl.ac.cn +</td></tr><tr><td>90ad0daa279c3e30b360f9fe9371293d68f4cebf</td><td>SPATIO-TEMPORAL FRAMEWORK AND +<br/>ALGORITHMS FOR VIDEO-BASED FACE +<br/>RECOGNITION +<br/>DOCTOR OF PHILOSOPHY +<br/><b>MULTIMEDIA UNIVERSITY</b><br/>MAY 2014 +</td><td>('2339975', 'JOHN SEE', 'john see')</td><td></td></tr><tr><td>90a754f597958a2717862fbaa313f67b25083bf9</td><td>REVIEW +<br/>published: 16 November 2015 +<br/>doi: 10.3389/frobt.2015.00028 +<br/>A Review of Human Activity +<br/>Recognition Methods +<br/><b>University of Ioannina, Ioannina, Greece, 2 Computational Biomedicine</b><br/><b>Laboratory, University of Houston, Houston, TX, USA</b><br/>Recognizing human activities from video sequences or still images is a challenging task +<br/>due to problems, such as background clutter, partial occlusion, changes in scale, view- +<br/><b>point, lighting, and appearance. Many applications, including video surveillance systems</b><br/>human-computer interaction, and robotics for human behavior characterization, require +<br/>a multiple activity recognition system. In this work, we provide a detailed review of recent +<br/>and state-of-the-art research advances in the field of human activity classification. We +<br/>propose a categorization of human activity methodologies and discuss their advantages +<br/>and limitations. In particular, we divide human activity classification methods into two large +<br/>categories according to whether they use data from different modalities or not. Then, each +<br/>of these categories is further analyzed into sub-categories, which reflect how they model +<br/>human activities and what type of activities they are interested in. Moreover, we provide +<br/>a comprehensive analysis of the existing, publicly available human activity classification +<br/>datasets and examine the requirements for an ideal human activity recognition dataset. +<br/>Finally, we report the characteristics of future research directions and present some open +<br/>issues on human activity recognition. +<br/>Keywords: human activity recognition, activity categorization, activity datasets, action representation, +<br/>review, survey +<br/>1. INTRODUCTION +<br/>Human activity recognition plays a significant role in human-to-human interaction and interper- +<br/>sonal relations. Because it provides information about the identity of a person, their personality, +<br/>and psychological state, it is difficult to extract. The human ability to recognize another person’s +<br/>activities is one of the main subjects of study of the scientific areas of computer vision and machine +<br/><b>learning. As a result of this research, many applications, including video surveillance systems</b><br/>human-computer interaction, and robotics for human behavior characterization, require a multiple +<br/>activity recognition system. +<br/>Among various classification techniques two main questions arise: “What action?” (i.e., the +<br/>recognition problem) and “Where in the video?” (i.e., the localization problem). When attempting to +<br/>recognize human activities, one must determine the kinetic states of a person, so that the computer +<br/>can efficiently recognize this activity. Human activities, such as “walking” and “running,” arise very +<br/>naturally in daily life and are relatively easy to recognize. On the other hand, more complex activities, +<br/>such as “peeling an apple,” are more difficult to identify. Complex activities may be decomposed into +<br/>other simpler activities, which are generally easier to recognize. Usually, the detection of objects in +<br/>a scene may help to better understand human activities as it may provide useful information about +<br/>the ongoing event (Gupta and Davis, 2007). +<br/>Edited by: +<br/>Venkatesh Babu Radhakrishnan, +<br/><b>Indian Institute of Science, India</b><br/>Reviewed by: +<br/>Stefano Berretti, +<br/><b>University of Florence, Italy</b><br/>Xinlei Chen, +<br/><b>Carnegie Mellon University, USA</b><br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to Vision +<br/>Systems Theory, Tools and +<br/>Applications, a section of the +<br/>journal Frontiers in Robotics and AI +<br/>Received: 09 July 2015 +<br/>Accepted: 29 October 2015 +<br/>Published: 16 November 2015 +<br/>Citation: +<br/>Vrigkas M, Nikou C and Kakadiaris IA +<br/>(2015) A Review of Human Activity +<br/>Recognition Methods. +<br/>Front. Robot. AI 2:28. +<br/>doi: 10.3389/frobt.2015.00028 +<br/>Frontiers in Robotics and AI | www.frontiersin.org +<br/>November 2015 | Volume 2 | Article 28 +</td><td>('2045915', 'Michalis Vrigkas', 'michalis vrigkas')<br/>('1727495', 'Christophoros Nikou', 'christophoros nikou')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')<br/>('1727495', 'Christophoros Nikou', 'christophoros nikou')</td><td>cnikou@cs.uoi.gr +</td></tr><tr><td>90d9209d5dd679b159051a8315423a7f796d704d</td><td>Temporal Sequence Distillation: Towards Few-Frame Action +<br/>Recognition in Videos +<br/><b>Wuhan University</b><br/>SenseTime Research +<br/>SenseTime Research +<br/><b>The Chinese University of Hong Kong</b><br/>SenseTime Research +<br/>SenseTime Research +</td><td>('40192003', 'Zhaoyang Zhang', 'zhaoyang zhang')<br/>('1874900', 'Zhanghui Kuang', 'zhanghui kuang')<br/>('47571885', 'Ping Luo', 'ping luo')<br/>('1739512', 'Litong Feng', 'litong feng')<br/>('1726357', 'Wei Zhang', 'wei zhang')</td><td>zhangzhaoyang@whu.edu.cn +<br/>kuangzhanghui@sensetime.com +<br/>pluo@ie.cuhk.edu.hk +<br/>fenglitong@sensetime.com +<br/>wayne.zhang@sensetime.com +</td></tr><tr><td>90dd2a53236b058c79763459b9d8a7ba5e58c4f1</td><td>Capturing Correlations Among Facial Parts for +<br/>Facial Expression Analysis +<br/>Department of Computer Science +<br/><b>Queen Mary, University of London</b><br/>Mile End Road, London E1 4NS, UK +</td><td>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('2803283', 'Peter W. McOwan', 'peter w. mcowan')</td><td>{cfshan, sgg, pmco}@dcs.qmul.ac.uk +</td></tr><tr><td>90cb074a19c5e7d92a1c0d328a1ade1295f4f311</td><td>MIT. Media Laboratory Affective Computing Technical Report #571 +<br/>Appears in IEEE International Workshop on Analysis and Modeling of Faces and Gestures , Oct 2003 +<br/>Fully Automatic Upper Facial Action Recognition +<br/>MIT Media Laboratory +<br/>Cambridge, MA 02139 +</td><td>('2189118', 'Ashish Kapoor', 'ashish kapoor')</td><td></td></tr><tr><td>90b11e095c807a23f517d94523a4da6ae6b12c76</td><td></td><td></td><td></td></tr><tr><td>90c2d4d9569866a0b930e91713ad1da01c2a6846</td><td>528 +<br/>The Open Automation and Control Systems Journal, 2014, 6, 528-534 +<br/>Dimensionality Reduction Based on Low Rank Representation +<br/>Open Access +<br/><b>School of Electronic and Information Engineering, Tongji University, Shanghai, China</b></td><td>('40328872', 'Cheng Luo', 'cheng luo')<br/>('40174994', 'Yang Xiang', 'yang xiang')</td><td>Send Orders for Reprints to reprints@benthamscience.ae +</td></tr><tr><td>907475a4febf3f1d4089a3e775ea018fbec895fe</td><td>STATISTICAL MODELING FOR FACIAL EXPRESSION ANALYSIS AND SYNTHESIS +<br/><b>Heudiasyc Laboratory, CNRS, University of Technology of Compi`egne</b><br/>BP 20529, 60205 COMPIEGNE Cedex, FRANCE. +</td><td>('2371236', 'Bouchra Abboud', 'bouchra abboud')<br/>('1742818', 'Franck Davoine', 'franck davoine')</td><td>E-mail: Franck.Davoine@hds.utc.fr +</td></tr><tr><td>9028fbbd1727215010a5e09bc5758492211dec19</td><td>Solving the Uncalibrated Photometric Stereo +<br/>Problem using Total Variation +<br/>1 IRIT, UMR CNRS 5505, Toulouse, France +<br/>2 Dept. of Computer Science, Univ. of Copenhagen, Denmark +</td><td>('2233590', 'Jean-Denis Durou', 'jean-denis durou')</td><td>yvain.queau@enseeiht.fr +<br/>durou@irit.fr +<br/>francois@diku.dk +</td></tr><tr><td>bff77a3b80f40cefe79550bf9e220fb82a74c084</td><td>Facial Expression Recognition Based on Local Binary Patterns and +<br/>Local Fisher Discriminant Analysis +<br/>1School of Physics and Electronic Engineering +<br/><b>Taizhou University</b><br/>Taizhou 318000 +<br/>CHINA +<br/> 2Department of Computer Science +<br/><b>Taizhou University</b><br/>Taizhou 318000 +<br/>CHINA +</td><td>('1695589', 'SHIQING ZHANG', 'shiqing zhang')<br/>('1730594', 'XIAOMING ZHAO', 'xiaoming zhao')<br/>('38909691', 'BICHENG LEI', 'bicheng lei')</td><td>tzczsq@163.com, leibicheng@163.com +<br/>tzxyzxm@163.com +</td></tr><tr><td>bf03f0fe8f3ba5b118bdcbb935bacb62989ecb11</td><td>EFFECT OF FACIAL EXPRESSIONS ON FEATURE-BASED +<br/>LANDMARK LOCALIZATION IN STATIC GREY SCALE +<br/>IMAGES +<br/>Research Group for Emotions, Sociality, and Computing, Tampere Unit for Computer-Human Interaction (TAUCHI) +<br/><b>University of Tampere, Kanslerinnrinne 1, 33014, Tampere, Finland</b><br/>Keywords: +<br/>Image processing and computer vision, segmentation, edge detection, facial landmark localization, facial +<br/>expressions, action units. +</td><td>('2935367', 'Yulia Gizatdinova', 'yulia gizatdinova')<br/>('1718377', 'Veikko Surakka', 'veikko surakka')</td><td>{yulia.gizatdinova, veikko.surakka}@cs.uta.fi +</td></tr><tr><td>bf961e4a57a8f7e9d792e6c2513ee1fb293658e9</td><td>EURASIP Journal on Applied Signal Processing 2004:16, 2533–2543 +<br/>c(cid:1) 2004 Hindawi Publishing Corporation +<br/>Robust Face Image Matching under +<br/>Illumination Variations +<br/><b>National Tsing Hua University, 101 Kuang Fu Road, Section 2, Hsinchu 300, Taiwan</b><br/><b>National Tsing Hua University, 101 Kuang Fu Road, Section 2, Hsinchu 300, Taiwan</b><br/><b>National Tsing Hua University, 101 Kuang Fu Road, Section 2, Hsinchu 300, Taiwan</b><br/>Received 1 September 2003; Revised 21 September 2004 +<br/>Face image matching is an essential step for face recognition and face verification. It is difficult to achieve robust face matching +<br/>under various image acquisition conditions. In this paper, a novel face image matching algorithm robust against illumination +<br/>variations is proposed. The proposed image matching algorithm is motivated by the characteristics of high image gradient along +<br/>the face contours. We define a new consistency measure as the inner product between two normalized gradient vectors at the +<br/>corresponding locations in two images. The normalized gradient is obtained by dividing the computed gradient vector by the +<br/>corresponding locally maximal gradient magnitude. Then we compute the average consistency measures for all pairs of the corre- +<br/>sponding face contour pixels to be the robust matching measure between two face images. To alleviate the problem due to shadow +<br/>and intensity saturation, we introduce an intensity weighting function for each individual consistency measure to form a weighted +<br/>average of the consistency measure. This robust consistency measure is further extended to integrate multiple face images of the +<br/>same person captured under different illumination conditions, thus making our robust face matching algorithm. Experimental +<br/>results of applying the proposed face image matching algorithm on some well-known face datasets are given in comparison with +<br/>some existing face recognition methods. The results show that the proposed algorithm consistently outperforms other methods +<br/>and achieves higher than 93% recognition rate with three reference images for different datasets under different lighting condi- +<br/>tions. +<br/>Keywords and phrases: robust image matching, face recognition, illumination variations, normalized gradient. +<br/>INTRODUCTION +<br/>1. +<br/>Face recognition has attracted the attention of a number +<br/>of researchers from academia and industry because of its +<br/>challenges and related applications, such as security access +<br/>control, personal ID verification, e-commerce, video surveil- +<br/>lance, and so forth. The details of these applications are re- +<br/>ferred to in the surveys [1, 2, 3]. Face matching is the most +<br/>important and crucial component in face recognition. Al- +<br/>though there have been many efforts in previous works to +<br/>achieve robust face matching under a wide variety of dif- +<br/>ferent image capturing conditions, such as lighting changes, +<br/>head pose or view angle variations, expression variations, +<br/>and so forth, these problems are still difficult to overcome. +<br/>It is a great challenge to achieve robust face matching under +<br/>all kinds of different face imaging variations. A practical face +<br/>recognition system needs to work under different imaging +<br/>conditions, such as different face poses, or different illumi- +<br/>nation conditions. Therefore, a robust face matching method +<br/>is essential to the development of an illumination-insensitive +<br/>face recognition system. In this paper, we particularly focus +<br/>on robust face matching under different illumination condi- +<br/>tions. +<br/>Many researchers have proposed face recognition meth- +<br/>ods or face verification systems under different illumination +<br/>conditions. Some of these methods extracted representative +<br/>features from face images to compute the distance between +<br/>these features. In general, these methods can be categorized +<br/>into the feature-based approach [4, 5, 6, 7, 8, 9, 10, 11], the +<br/>appearance-based approach [12, 13, 14, 15, 16, 17, 18, 19, 20, +<br/>21, 22, 23], and the hybrid approach [22, 24]. +</td><td>('2393568', 'Chyuan-Huei Thomas Yang', 'chyuan-huei thomas yang')<br/>('1696527', 'Shang-Hong Lai', 'shang-hong lai')<br/>('39505245', 'Long-Wen Chang', 'long-wen chang')</td><td>Email: chyang@cs.nthu.edu.tw +<br/>Email: lai@cs.nthu.edu.tw +<br/>Email: lchang@cs.nthu.edu.tw +</td></tr><tr><td>bf54b5586cdb0b32f6eed35798ff91592b03fbc4</td><td>Journal of Signal and Information Processing, 2017, 8, 78-98 +<br/>http://www.scirp.org/journal/jsip +<br/>ISSN Online: 2159-4481 +<br/>ISSN Print: 2159-4465 +<br/>Methodical Analysis of Western-Caucasian and +<br/>East-Asian Basic Facial Expressions of Emotions +<br/>Based on Specific Facial Regions +<br/><b>The University of Electro-Communications, Tokyo, Japan</b><br/>How to cite this paper: Benitez-Garcia, G., +<br/>Nakamura, T. and Kaneko, M. (2017) Me- +<br/>thodical Analysis of Western-Caucasian and +<br/>East-Asian Basic Facial Expressions of Emo- +<br/>tions Based on Specific Facial Regions. Jour- +<br/>nal of Signal and Information Processing, 8, +<br/>78-98. +<br/>https://doi.org/10.4236/jsip.2017.82006 +<br/>Received: March 30, 2017 +<br/>Accepted: May 15, 2017 +<br/>Published: May 18, 2017 +<br/>Copyright © 2017 by authors and +<br/>Scientific Research Publishing Inc. +<br/>This work is licensed under the Creative +<br/>Commons Attribution International +<br/>License (CC BY 4.0). +<br/>http://creativecommons.org/licenses/by/4.0/ +<br/> +<br/>Open Access +</td><td>('2567776', 'Gibran Benitez-Garcia', 'gibran benitez-garcia')<br/>('1693821', 'Tomoaki Nakamura', 'tomoaki nakamura')<br/>('49061848', 'Masahide Kaneko', 'masahide kaneko')</td><td></td></tr><tr><td>bf1e0279a13903e1d43f8562aaf41444afca4fdc</td><td> International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 +<br/> Volume: 04 Issue: 10 | Oct -2017 www.irjet.net p-ISSN: 2395-0072 +<br/>Different Viewpoints of Recognizing Fleeting Facial Expressions with +<br/>DWT +<br/>information +<br/>to get desired +<br/>information +<br/>Introduction +<br/>---------------------------------------------------------------------***--------------------------------------------------------------------- +</td><td>('1848141', 'SANJEEV SHRIVASTAVA', 'sanjeev shrivastava')<br/>('34417227', 'MOHIT GANGWAR', 'mohit gangwar')</td><td></td></tr><tr><td>bf0f0eb0fb31ee498da4ae2ca9b467f730ea9103</td><td>Brain Sci. 2015, 5, 369-386; doi:10.3390/brainsci5030369 +<br/>OPEN ACCESS +<br/>brain sciences +<br/>ISSN 2076-3425 +<br/>www.mdpi.com/journal/brainsci/ +<br/>Article +<br/>Emotion Regulation in Adolescent Males with Attention-Deficit +<br/>Hyperactivity Disorder: Testing the Effects of Comorbid +<br/>Conduct Disorder +<br/><b>School of Psychology, Cardiff University, Cardiff, CF10 3AT, UK</b><br/><b>MRC Centre for Neuropsychiatric Genetics and Genomics, Cardiff University, Cardiff</b><br/>Tel.: +44-2920-874630; Fax: +44-2920-874545. +<br/>Received: 17 July 2015 / Accepted: 25 August 2015 / Published: 7 September 2015 +</td><td>('5383377', 'Clare Northover', 'clare northover')<br/>('4094135', 'Anita Thapar', 'anita thapar')<br/>('39373878', 'Kate Langley', 'kate langley')<br/>('4552820', 'Stephanie van Goozen', 'stephanie van goozen')<br/>('2928107', 'Derek G.V. Mitchell', 'derek g.v. mitchell')</td><td>E-Mails: NorthoverC@cardiff.ac.uk (C.N.); LangleyK@cardiff.ac.uk (K.L.) +<br/>CF24 4HQ, UK; E-Mail: Thapar@cardiff.ac.uk +<br/>* Author to whom correspondence should be addressed; E-Mail: vangoozens@cardiff.ac.uk; +</td></tr><tr><td>bf3f8726f2121f58b99b9e7287f7fbbb7ab6b5f5</td><td>Visual face scanning and emotion +<br/>perception analysis between Autistic +<br/>and Typically Developing children +<br/><b>University of Dhaka</b><br/><b>University of Dhaka</b><br/>Dhaka, Bangladesh +<br/>Dhaka, Bangladesh +</td><td>('24613724', 'Uzma Haque Syeda', 'uzma haque syeda')<br/>('24572640', 'Syed Mahir Tazwar', 'syed mahir tazwar')</td><td></td></tr><tr><td>bf4825474673246ae855979034c8ffdb12c80a98</td><td><b>UNIVERSITY OF CALIFORNIA</b><br/>RIVERSIDE +<br/>Active Learning in Multi-Camera Networks, With Applications in Person +<br/>Re-Identification +<br/>A Dissertation submitted in partial satisfaction +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>in +<br/>Electrical Engineering +<br/>by +<br/>December 2015 +<br/>Dissertation Committee: +</td><td>('40521893', 'Abir Das', 'abir das')<br/>('1688416', 'Amit K. Roy-Chowdhury', 'amit k. roy-chowdhury')<br/>('1751869', 'Anastasios Mourikis', 'anastasios mourikis')<br/>('1778860', 'Walid Najjar', 'walid najjar')</td><td></td></tr><tr><td>bf8a520533f401347e2f55da17383a3e567ef6d8</td><td>Bounded-Distortion Metric Learning +<br/><b>The Chinese University of Hong Kong</b><br/><b>University of Chinese Academy of Sciences</b><br/><b>Tsinghua University</b><br/><b>The Chinese University of Hong Kong</b></td><td>('2246396', 'Renjie Liao', 'renjie liao')<br/>('1788070', 'Jianping Shi', 'jianping shi')<br/>('2376789', 'Ziyang Ma', 'ziyang ma')<br/>('37670465', 'Jun Zhu', 'jun zhu')<br/>('1729056', 'Jiaya Jia', 'jiaya jia')</td><td>rjliao,jpshi@cse.cuhk.edu.hk +<br/>maziyang08@gmail.com +<br/>dcszj@mail.tsinghua.edu.cn +<br/>leojia@cse.cuhk.edu.hk +</td></tr><tr><td>bf5940d57f97ed20c50278a81e901ae4656f0f2c</td><td>Query-free Clothing Retrieval via Implicit +<br/>Relevance Feedback +</td><td>('26331884', 'Zhuoxiang Chen', 'zhuoxiang chen')<br/>('1691461', 'Zhe Xu', 'zhe xu')<br/>('48380192', 'Ya Zhang', 'ya zhang')<br/>('48531192', 'Xiao Gu', 'xiao gu')</td><td></td></tr><tr><td>bff567c58db554858c7f39870cff7c306523dfee</td><td>Neural Task Graphs: Generalizing to Unseen +<br/>Tasks from a Single Video Demonstration +<br/><b>Stanford University</b></td><td>('38485317', 'De-An Huang', 'de-an huang')<br/>('4734949', 'Suraj Nair', 'suraj nair')<br/>('2068265', 'Danfei Xu', 'danfei xu')<br/>('2117748', 'Yuke Zhu', 'yuke zhu')<br/>('1873736', 'Animesh Garg', 'animesh garg')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')<br/>('1702137', 'Silvio Savarese', 'silvio savarese')<br/>('9200530', 'Juan Carlos Niebles', 'juan carlos niebles')</td><td></td></tr><tr><td>bfb98423941e51e3cd067cb085ebfa3087f3bfbe</td><td>Sparseness helps: Sparsity Augmented +<br/>Collaborative Representation for Classification +</td><td>('2941543', 'Naveed Akhtar', 'naveed akhtar')<br/>('1688013', 'Faisal Shafait', 'faisal shafait')</td><td></td></tr><tr><td>bffbd04ee5c837cd919b946fecf01897b2d2d432</td><td><b>Boston University Computer Science Technical Report No</b><br/>Facial Feature Tracking and Occlusion +<br/>Recovery in American Sign Language +<br/>1 Department of Computer Science, 2 Department of Modern Foreign Languages +<br/><b>Boston University</b><br/>Facial features play an important role in expressing grammatical information +<br/><b>in signed languages, including American Sign Language (ASL). Gestures such</b><br/>as raising or furrowing the eyebrows are key indicators of constructions such +<br/>as yes-no questions. Periodic head movements (nods and shakes) are also an +<br/>essential part of the expression of syntactic information, such as negation +<br/>(associated with a side-to-side headshake). Therefore, identification of these +<br/>facial gestures is essential to sign language recognition. One problem with +<br/>detection of such grammatical indicators is occlusion recovery. If the signer’s +<br/>hand blocks his/her eyebrows during production of a sign, it becomes difficult +<br/>to track the eyebrows. We have developed a system to detect such grammatical +<br/>markers in ASL that recovers promptly from occlusion. +<br/>Our system detects and tracks evolving templates of facial features, which +<br/>are based on an anthropometric face model, and interprets the geometric +<br/>relationships of these templates to identify grammatical markers. It was tested +<br/>on a variety of ASL sentences signed by various Deaf 1native signers and +<br/>detected facial gestures used to express grammatical information, such as +<br/>raised and furrowed eyebrows as well as headshakes. +<br/>1 Introduction +<br/>A computer-based translator of American Sign Language (ASL) would be +<br/>useful in enabling people who do not know ASL to communicate with Deaf1 +<br/>individuals. Facial gesture interpretation would be an essential part of an in- +<br/>terface that eliminates the language barrier between Deaf and hearing people. +<br/>Our work focuses on facial feature detection and tracking in ASL, specifically +<br/>in occlusion processing and recovery. +<br/>1 The word “Deaf” is capitalized to designate those individuals who are linguisti- +<br/>cally and culturally deaf and who use ASL as their primary language, whereas +<br/>“deaf” refers to the status of those who cannot hear [25]. +</td><td>('2313369', 'Thomas J. Castelli', 'thomas j. castelli')<br/>('1723703', 'Margrit Betke', 'margrit betke')<br/>('1732359', 'Carol Neidle', 'carol neidle')</td><td></td></tr><tr><td>d35534f3f59631951011539da2fe83f2844ca245</td><td>Published as a conference paper at ICLR 2018 +<br/>SEMANTICALLY DECOMPOSING THE LATENT SPACES +<br/>OF GENERATIVE ADVERSARIAL NETWORKS +<br/>Department of Music +<br/><b>University of California, San Diego</b><br/>Department of Genetics +<br/><b>Stanford University</b><br/>Zachary C. Lipton +<br/><b>Carnegie Mellon University</b><br/>Amazon AI +<br/>Department of Computer Science +<br/><b>University of California, San Diego</b></td><td>('1872307', 'Chris Donahue', 'chris donahue')<br/>('1693411', 'Akshay Balsubramani', 'akshay balsubramani')<br/>('1814008', 'Julian McAuley', 'julian mcauley')</td><td>cdonahue@ucsd.edu +<br/>abalsubr@stanford.edu +<br/>zlipton@cmu.edu +<br/>jmcauley@eng.ucsd.edu +</td></tr><tr><td>d3edbfe18610ce63f83db83f7fbc7634dde1eb40</td><td>Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-17) +<br/>Large Graph Hashing with Spectral Rotation +<br/>School of Computer Science and Center for OPTical IMagery Analysis and Learning (OPTIMAL), +<br/><b>Northwestern Polytechnical University</b><br/>Xi’an 710072, Shaanxi, P. R. China +</td><td>('1720243', 'Xuelong Li', 'xuelong li')<br/>('48080389', 'Di Hu', 'di hu')<br/>('1688370', 'Feiping Nie', 'feiping nie')</td><td>xuelong li@opt.ac.cn, hdui831@mail.nwpu.edu.cn, feipingnie@gmail.com +</td></tr><tr><td>d3424761e06a8f5f3c1f042f1f1163a469872129</td><td>Pose-invariant, model-based object +<br/>recognition, using linear combination of views +<br/>and Bayesian statistics. +<br/>A dissertation submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>of the +<br/><b>University of London</b><br/>Department of Computer Science +<br/><b>University College London</b><br/>2009 +</td><td>('1797883', 'Vasileios Zografos', 'vasileios zografos')</td><td></td></tr><tr><td>d33b26794ea6d744bba7110d2d4365b752d7246f</td><td>Transfer Feature Representation via Multiple Kernel Learning +<br/>1. Science and Technology on Integrated Information System Laboratory +<br/>2. State Key Laboratory of Computer Science +<br/><b>Institute of Software, Chinese Academy of Sciences, Beijing 100190, China</b></td><td>('40451597', 'Wei Wang', 'wei wang')<br/>('39483391', 'Hao Wang', 'hao wang')<br/>('1783918', 'Chen Zhang', 'chen zhang')<br/>('34532334', 'Fanjiang Xu', 'fanjiang xu')</td><td>weiwangpenny@gmail.com +</td></tr><tr><td>d3b73e06d19da6b457924269bb208878160059da</td><td>Proceedings of the 5th International Conference on Computing and Informatics, ICOCI 2015 +<br/>11-13 August, 2015 Istanbul, Turkey. Universiti Utara Malaysia (http://www.uum.edu.my ) +<br/>Paper No. +<br/>065 +<br/>IMPLEMENTATION OF AN AUTOMATED SMART HOME +<br/>CONTROL FOR DETECTING HUMAN EMOTIONS VIA FACIAL +<br/>DETECTION +<br/>Osman4 +</td><td>('9164797', 'Lim Teck Boon', 'lim teck boon')<br/>('2229534', 'Mohd Heikal Husin', 'mohd heikal husin')<br/>('1881455', 'Zarul Fitri Zaaba', 'zarul fitri zaaba')</td><td>1Universiti Sains Malaysia, Malaysia, ltboon.ucom10@student.usm.my +<br/>2Universiti Sains Malaysia, Malaysia, heikal@usm.my +<br/>3Universiti Sains Malaysia, Malaysia, zarulfitri@usm.my +<br/>4Universiti Sains Malaysia, Malaysia, azam@usm.my +</td></tr><tr><td>d3d5d86afec84c0713ec868cf5ed41661fc96edc</td><td>A Comprehensive Analysis of Deep Learning Based Representation +<br/>for Face Recognition +<br/>Mostafa Mehdipour Ghazi +<br/>Faculty of Engineering and Natural Sciences +<br/><b>Sabanci University, Istanbul, Turkey</b><br/>Hazım Kemal Ekenel +<br/>Department of Computer Engineering +<br/><b>Istanbul Technical University, Istanbul, Turkey</b></td><td></td><td>mehdipour@sabanciuniv.edu +<br/>ekenel@itu.edu.tr +</td></tr><tr><td>d3e04963ff42284c721f2bc6a90b7a9e20f0242f</td><td>On Forensic Use of Biometrics +<br/><b>University of Southampton, UK, 2University of Warwick, UK</b><br/>This chapter discusses the use of biometrics techniques within forensic science. It outlines the +<br/>historic connections between the subjects and then examines face and ear biometrics as two +<br/>case studies to demonstrate the application, the challenges and the acceptability of biometric +<br/>features and techniques in forensics. The detailed examination starts with one of the most +<br/>common and familiar biometric features, face, and then examines an emerging biometric +<br/>feature, ear. +<br/>1.1 Introduction +<br/>Forensic science largely concerns the analysis of crime: its existence, the perpetrator(s) and +<br/>the modus operandi. The science of biometrics has been developing approaches that can +<br/>be used to automatically identify individuals by personal characteristics. The relationship +<br/>of biometrics and forensics centers primarily on identifying people: the central question is +<br/>whether a perpetrator can reliably be identified from scene-of-crime data or can reliably +<br/>be excluded, wherein the reliability concerns reasonable doubt. The personal characteristics +<br/>which can be used as biometrics include face, finger, iris, gait, ear, electroencephalogram +<br/>(EEG), handwriting, voice and palm. Those which are suited to forensic use concern traces +<br/>left at a scene-of-crime, such as latent fingerprints, palmprints or earprints, or traces which +<br/>have been recorded, such as face, gait or ear in surveillance video. +<br/>Biometrics is generally concerned with the recognition of individuals based on their +<br/>physical or behavioral attributes. So far, biometric techniques have primarily been used to +<br/>assure identity (in immigration and commerce etc.). These techniques are largely automatic +<br/>or semi-automatic approaches steeped in pattern recognition and computer vision. The main +<br/>steps of a biometric recognition approach are: (1) acquisition of the biometric data; (2) +<br/>localization and alignment of the data; (3) feature extraction; and (4) matching. Feature +<br/>This is a Book Title Name of the Author/Editor +<br/>c(cid:13) XXXX John Wiley & Sons, Ltd +</td><td>('2804800', 'Banafshe Arbab-Zavar', 'banafshe arbab-zavar')<br/>('40655450', 'Xingjie Wei', 'xingjie wei')<br/>('2365596', 'John D. Bustard', 'john d. bustard')<br/>('1727698', 'Mark S. Nixon', 'mark s. nixon')<br/>('1799504', 'Chang-Tsun Li', 'chang-tsun li')</td><td>1{baz10v,jdb,msn}@ecs.soton.ac.uk, 2{x.wei, c-t.li}@warwick.ac.uk +</td></tr><tr><td>d3d71a110f26872c69cf25df70043f7615edcf92</td><td>2736 +<br/>Learning Compact Feature Descriptor and Adaptive +<br/>Matching Framework for Face Recognition +<br/>improvements +</td><td>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('2856494', 'Dihong Gong', 'dihong gong')<br/>('1720243', 'Xuelong Li', 'xuelong li')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')</td><td></td></tr><tr><td>d35c82588645b94ce3f629a0b98f6a531e4022a3</td><td>Scalable Online Annotation & +<br/>Object Localisation +<br/>For Broadcast Media Production +<br/>Submitted for the Degree of +<br/>Master of Philosophy +<br/>from the +<br/><b>University of Surrey</b><br/>Centre for Vision, Speech and Signal Processing +<br/>Faculty of Engineering and Physical Sciences +<br/><b>University of Surrey</b><br/>Guildford, Surrey GU2 7XH, U.K. +<br/>August 2016 +</td><td>('39222045', 'Charles Gray', 'charles gray')<br/>('39222045', 'Charles Gray', 'charles gray')</td><td></td></tr><tr><td>d3b18ba0d9b247bfa2fb95543d172ef888dfff95</td><td>Learning and Using the Arrow of Time +<br/><b>Harvard University 2University of Southern California</b><br/><b>University of Oxford 4Massachusetts Institute of Technology 5Google Research</b><br/>(a) +<br/>(c) +<br/>(b) +<br/>(d) +<br/>Figure 1: Seeing these ordered frames from videos, can you tell whether each video is playing forward or backward? (answer +<br/>below1). Depending on the video, solving the task may require (a) low-level understanding (e.g. physics), (b) high-level +<br/>reasoning (e.g. semantics), or (c) familiarity with very subtle effects or with (d) camera conventions. In this work, we learn +<br/>and exploit several types of knowledge to predict the arrow of time automatically with neural network models trained on +<br/>large-scale video datasets. +</td><td>('1766333', 'Donglai Wei', 'donglai wei')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')<br/>('1768236', 'William T. Freeman', 'william t. freeman')</td><td>donglai@seas.harvard.edu, limjj@usc.edu, az@robots.ox.ac.uk, billf@mit.edu +</td></tr><tr><td>d309e414f0d6e56e7ba45736d28ee58ae2bad478</td><td>Efficient Two-Stream Motion and Appearance 3D CNNs for +<br/>Video Classification +<br/>Ali Diba +<br/>ESAT-KU Leuven +<br/>Ali Pazandeh +<br/>Sharif UTech +<br/>Luc Van Gool +<br/>ESAT-KU Leuven, ETH Zurich +</td><td></td><td>ali.diba@esat.kuleuven.be +<br/>pazandeh@ee.sharif.ir +<br/>luc.vangool@esat.kuleuven.be +</td></tr><tr><td>d394bd9fbaad1f421df8a49347d4b3fca307db83</td><td>Recognizing Facial Expressions at Low Resolution +<br/><b>Deparment of Computer Science, Queen Mary, University of London, London, E1 4NS, UK</b></td><td>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('2803283', 'Peter W. McOwan', 'peter w. mcowan')</td><td>{cfshan, sgg, pmco}@dcs.qmul.ac.uk +</td></tr><tr><td>d3f5a1848b0028d8ab51d0b0673732cad2e3c8c9</td><td></td><td></td><td></td></tr><tr><td>d3b550e587379c481392fb07f2cbbe11728cf7a6</td><td>Small Sample Size Face Recognition using Random Quad-Tree based +<br/>Ensemble Algorithm +<br/><b>Graduate School of Informatics, Kyoto University, Kyoto 606-8501, Japan</b></td><td>('7923772', 'Cuicui Zhang', 'cuicui zhang')<br/>('2735528', 'Xuefeng Liang', 'xuefeng liang')<br/>('1731351', 'Takashi Matsuyama', 'takashi matsuyama')</td><td>zhang@vision.kuee.kyoto-u.ac.jp, fxliang, tmg@i.kyoto-u.ac.jp +</td></tr><tr><td>d307a766cc9c728a24422313d4c3dcfdb0d16dd5</td><td>Deep Keyframe Detection in Human Action Videos +<br/><b>School of Physics and Optoelectronic Engineering, Xidian University, China</b><br/><b>School of Computer Science and Software Engineering, University of Western Australia</b><br/><b>College of Electrical and Information Engineering, Hunan University, China</b><br/><b>School of Software, Xidian University, China</b></td><td>('46580760', 'Xiang Yan', 'xiang yan')<br/>('1746166', 'Syed Zulqarnain Gilani', 'syed zulqarnain gilani')<br/>('2404621', 'Hanlin Qin', 'hanlin qin')<br/>('3446916', 'Mingtao Feng', 'mingtao feng')<br/>('48570713', 'Liang Zhang', 'liang zhang')<br/>('46332747', 'Ajmal Mian', 'ajmal mian')</td><td>xyan@stu.xidian.edu.cn, hlqin@mail.xidian.edu.cn +<br/>{zulqarnain.gilani, ajmal.mian}@uwa.edu.au +<br/>mintfeng@hnu.edu.cn +<br/>liangzhang@xidian.edu.cn +</td></tr><tr><td>d31af74425719a3840b496b7932e0887b35e9e0d</td><td>Article +<br/>A Multimodal Deep Log-Based User Experience (UX) +<br/>Platform for UX Evaluation +<br/><b>Ubiquitous Computing Lab, Kyung Hee University</b><br/><b>College of Electronics and Information Engineering, Sejong University</b><br/>Received: 16 March 2018; Accepted: 15 May 2018; Published: 18 May 2018 +</td><td>('33081617', 'Jamil Hussain', 'jamil hussain')<br/>('2794241', 'Wajahat Ali Khan', 'wajahat ali khan')<br/>('27531310', 'Anees Ul Hassan', 'anees ul hassan')<br/>('1765947', 'Muhammad Afzal', 'muhammad afzal')<br/>('1700806', 'Sungyoung Lee', 'sungyoung lee')</td><td>Giheung-gu, Yongin-si, Gyeonggi-do, Seoul 446-701, Korea; jamil@oslab.khu.ac.kr (J.H.); +<br/>wajahat.alikhan@oslab.khu.ac.kr (W.A.K.); hth@oslab.khu.ac.kr (T.H.); bilalrizvi@oslab.khu.ac.kr (H.S.M.B.); +<br/>jhb@oslab.khu.ac.kr (J.B.); anees@oslab.khu.ac.kr (A.U.H.) +<br/>Seoul 05006, Korea; mafzal@sejong.ac.kr +<br/>* Correspondence: sylee@oslab.khu.ac.kr; Tel.: +82-31-201-2514 +</td></tr><tr><td>d3b0839324d0091e70ce34f44c979b9366547327</td><td>Precise Box Score: Extract More Information from Datasets to Improve the +<br/>Performance of Face Detection +<br/>1School of Information and Communication Engineering +<br/>2Beijing Key Laboratory of Network System and Network Culture +<br/><b>Beijing University of Posts and Telecommunications, Beijing, China</b></td><td>('49712251', 'Ce Qi', 'ce qi')<br/>('1684263', 'Fei Su', 'fei su')<br/>('8120542', 'Pingyu Wang', 'pingyu wang')</td><td></td></tr><tr><td>d30050cfd16b29e43ed2024ae74787ac0bbcf2f7</td><td>Facial Expression Classification Using +<br/>Convolutional Neural Network and Support Vector +<br/>Machine +<br/>Graduate Program in Electrical and Computer Engineering +<br/><b>Federal University of Technology - Paran a</b><br/>Department of Electrical and Computer Engineering +<br/><b>Opus College of Engineering</b><br/><b>Marquette University</b></td><td>('11857183', 'Cristian Bortolini', 'cristian bortolini')<br/>('2357308', 'Humberto R. Gamba', 'humberto r. gamba')<br/>('2432946', 'Gustavo Benvenutti Borba', 'gustavo benvenutti borba')<br/>('2767912', 'Henry Medeiros', 'henry medeiros')</td><td>Email: vpillajr@mail.com +</td></tr><tr><td>d3faed04712b4634b47e1de0340070653546deb2</td><td>Neural Best-Buddies: Sparse Cross-Domain Correspondence +<br/>Fig. 1. Top 5 Neural Best-Buddies for two cross-domain image pairs. Using deep features of a pre-trained neural network, our coarse-to-fine sparse +<br/>correspondence algorithm first finds high-level, low resolution, semantically matching areas (indicated by the large blue circles), then narrows down the search +<br/>area to intermediate levels (middle green circles), until precise localization on well-defined edges in the pixel space (colored in corresponding unique colors). +<br/>Correspondence between images is a fundamental problem in computer +<br/>vision, with a variety of graphics applications. This paper presents a novel +<br/>method for sparse cross-domain correspondence. Our method is designed for +<br/>pairs of images where the main objects of interest may belong to different +<br/>semantic categories and differ drastically in shape and appearance, yet still +<br/>contain semantically related or geometrically similar parts. Our approach +<br/>operates on hierarchies of deep features, extracted from the input images +<br/>by a pre-trained CNN. Specifically, starting from the coarsest layer in both +<br/>hierarchies, we search for Neural Best Buddies (NBB): pairs of neurons +<br/>that are mutual nearest neighbors. The key idea is then to percolate NBBs +<br/>through the hierarchy, while narrowing down the search regions at each +<br/>level and retaining only NBBs with significant activations. Furthermore, in +<br/>order to overcome differences in appearance, each pair of search regions is +<br/>transformed into a common appearance. +<br/>We evaluate our method via a user study, in addition to comparisons +<br/>with alternative correspondence approaches. The usefulness of our method +<br/><b>is demonstrated using a variety of graphics applications, including cross</b><br/>domain image alignment, creation of hybrid images, automatic image mor- +<br/>phing, and more. +<br/>CCS Concepts: • Computing methodologies → Interest point and salient +<br/>region detections; Matching; Image manipulation; +<br/><b>University</b><br/>© 2018 Association for Computing Machinery. +<br/>This is the author’s version of the work. It is posted here for your personal use. Not for +<br/>redistribution. The definitive Version of Record was published in ACM Transactions on +<br/>Graphics, https://doi.org/10.1145/3197517.3201332. +<br/>Additional Key Words and Phrases: cross-domain correspondence, image +<br/>hybrids, image morphing +<br/>ACM Reference Format: +<br/>Cohen-Or. 2018. Neural Best-Buddies: Sparse Cross-Domain Correspon- +<br/>//doi.org/10.1145/3197517.3201332 +<br/>INTRODUCTION +<br/>Finding correspondences between a pair of images has been a long +<br/>standing problem, with a multitude of applications in computer +<br/>vision and graphics. In particular, sparse sets of corresponding point +<br/>pairs may be used for tasks such as template matching, image align- +<br/>ment, and image morphing, to name a few. Over the years, a variety +<br/>of dense and sparse correspondence methods have been developed, +<br/>most of which assume that the input images depict the same scene +<br/>or object (with differences in viewpoint, lighting, object pose, etc.), +<br/>or a pair of objects from the same class. +<br/>In this work, we are concerned with sparse cross-domain corre- +<br/>spondence: a more general and challenging version of the sparse +<br/>correspondence problem, where the object of interest in the two +<br/>input images can differ more drastically in their shape and appear- +<br/>ance, such as objects belonging to different semantic categories +<br/>(domains). It is, however, assumed that the objects contain at least +<br/>some semantically related parts or geometrically similar regions, oth- +<br/>erwise the correspondence task cannot be considered well-defined. +<br/>Two examples of cross-domain scenarios and the results of our ap- +<br/>proach are shown in Figure 1. We focus on sparse correspondence, +<br/>since in many cross-domain image pairs, dense correspondence +<br/>ACM Transactions on Graphics, Vol. 37, No. 4, Article 69. Publication date: August 2018. +</td><td>('3451442', 'Kfir Aberman', 'kfir aberman')<br/>('39768043', 'Jing Liao', 'jing liao')<br/>('5807605', 'Mingyi Shi', 'mingyi shi')<br/>('1684384', 'Dani Lischinski', 'dani lischinski')<br/>('1748939', 'Baoquan Chen', 'baoquan chen')<br/>('1701009', 'Daniel Cohen-Or', 'daniel cohen-or')<br/>('3451442', 'Kfir Aberman', 'kfir aberman')<br/>('39768043', 'Jing Liao', 'jing liao')<br/>('5807605', 'Mingyi Shi', 'mingyi shi')<br/>('1684384', 'Dani Lischinski', 'dani lischinski')<br/>('1748939', 'Baoquan Chen', 'baoquan chen')<br/>('1701009', 'Daniel Cohen-Or', 'daniel cohen-or')<br/>('3451442', 'Kfir Aberman', 'kfir aberman')<br/>('39768043', 'Jing Liao', 'jing liao')<br/>('5807605', 'Mingyi Shi', 'mingyi shi')<br/>('1684384', 'Dani Lischinski', 'dani lischinski')<br/>('1748939', 'Baoquan Chen', 'baoquan chen')</td><td></td></tr><tr><td>d3c004125c71942846a9b32ae565c5216c068d1e</td><td>RESEARCH ARTICLE +<br/>Recognizing Age-Separated Face Images: +<br/>Humans and Machines +<br/><b>West Virginia University, Morgantown, West Virginia, United States of America, 2. IIIT Delhi, New Delhi</b><br/>Delhi, India +</td><td>('3017294', 'Daksha Yadav', 'daksha yadav')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('2487227', 'Afzel Noore', 'afzel noore')</td><td>*mayank@iiitd.ac.in +</td></tr><tr><td>d350a9390f0818703f886138da27bf8967fe8f51</td><td>LIGHTING DESIGN FOR PORTRAITS WITH A VIRTUAL LIGHT STAGE +<br/><b>Institute for Vision and Graphics, University of Siegen, Germany</b></td><td>('1967283', 'Davoud Shahlaei', 'davoud shahlaei')<br/>('2712313', 'Marcel Piotraschke', 'marcel piotraschke')<br/>('2880906', 'Volker Blanz', 'volker blanz')</td><td></td></tr><tr><td>d33fcdaf2c0bd0100ec94b2c437dccdacec66476</td><td>Neurons with Paraboloid Decision Boundaries for +<br/>Improved Neural Network Classification +<br/>Performance +</td><td>('2320550', 'Nikolaos Tsapanos', 'nikolaos tsapanos')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td></td></tr><tr><td>d4a5eaf2e9f2fd3e264940039e2cbbf08880a090</td><td>An Occluded Stacked Hourglass Approach to Facial +<br/>Landmark Localization and Occlusion Estimation +<br/><b>University of California San Diego</b></td><td>('2812409', 'Kevan Yuen', 'kevan yuen')</td><td>kcyuen@eng.ucsd.edu, mtrivedi@eng.ucsd.edu +</td></tr><tr><td>d46b790d22cb59df87f9486da28386b0f99339d3</td><td>Learning Face Deblurring Fast and Wide +<br/><b>University of Bern</b><br/>Switzerland +<br/>Amazon Research +<br/>Germany +<br/><b>University of Bern</b><br/>Switzerland +</td><td>('39866194', 'Meiguang Jin', 'meiguang jin')<br/>('36266446', 'Michael Hirsch', 'michael hirsch')<br/>('1739080', 'Paolo Favaro', 'paolo favaro')</td><td>jin@inf.unibe.ch +<br/>hirsch@amazon.com +<br/>favaro@inf.unibe.ch +</td></tr><tr><td>d41c11ebcb06c82b7055e2964914b9af417abfb2</td><td>CDI-Type I: Unsupervised and Weakly-Supervised +<br/>1 Introduction +<br/>Discovery of Facial Events +<br/>The face is one of the most powerful channels of nonverbal communication. Facial expression has been a +<br/>focus of emotion research for over a hundred years [12]. It is central to several leading theories of emotion +<br/>[18, 31, 54] and has been the focus of at times heated debate about issues in emotion science [19, 24, 50]. +<br/><b>Facial expression gures prominently in research on almost every aspect of emotion, including psychophys</b><br/>iology [40], neural correlates [20], development [11], perception [4], addiction [26], social processes [30], +<br/>depression [49] and other emotion disorders [55], to name a few. In general, facial expression provides cues +<br/>about emotional response, regulates interpersonal behavior, and communicates aspects of psychopathology. +<br/>Because of its importance to behavioral science and the emerging fields of computational behavior +<br/>science, perceptual computing, and human-robot interaction, significant efforts have been applied toward +<br/>developing algorithms that automatically detect facial expression. With few exceptions, previous work on +<br/>facial expression relies on supervised approaches to learning (i.e. event categories are defined in advance +<br/>in labeled training data). While supervised learning has important advantages, two critical limitations may +<br/>be noted. One, because labeling facial expression is highly labor intensive, progress in automated facial +<br/>expression recognition and analysis is slowed. For the most detailed and comprehensive labeling or coding +<br/>systems, such as Facial Action Coding System (FACS), three to four months is typically required to train +<br/>a coder (’coding’ refers to the labeling of video using behavioral descriptors). Once trained, each minute +<br/>of video may require 1 hour or more to code [9]. No wonder relatively few databases are yet available, +<br/>especially those of real-world rather than posed behavior [61]. Second, research has been limited to the +<br/>perceptual categories used by human observers. Those categories were operationalized in large part based +<br/>on technology available in the past [36]. While a worthy goal of computer vision and machine learning +<br/>is to efficiently replicate human-based measurement, should that be our only goal? New measurement +<br/>approaches make possible new scientific discoveries. Two in particular, unsupervised and weakly-supervised +<br/>learning have the potential to inform new ways of perceiving and modeling human behavior, to impact the +<br/>infrastructure of science, and contribute to the design of perceptual computing applications. +<br/>We propose that unsupervised and weakly-supervised approaches to automatic facial expression analysis +<br/>can increase the efficiency of current measurement approaches in behavioral science, demonstrate conver- +<br/>gent validity with supervised approaches, and lead to new knowledge in clinical and developmental science. +<br/>Specifically, we will: +<br/>• Develop two novel non-parametric algorithms for unsupervised and weakly-supervised time-series +<br/>analysis. The proposed approaches are general and can be applied to a myriad of problems in behav- +<br/>ioral science and computer vision (e.g., gesture or activity recognition). +<br/>• Exploit the potential of these algorithms in four applications: +<br/>1) New tools to improve the reliability and utility of human FACS coding. Using unsupervised learn- +<br/>ing, we will develop and validate a computer-assisted approach to FACS coding that doubles the +<br/>efficiency of human FACS coding. +<br/>2) At present, taxonomies of facial expression are based on FACS or other observer-based schemes. +<br/>Consequently, approaches to automatic facial expression recognition are dependent on access to cor- +<br/>puses of FACS or similarly labeled video. In the proposed work we raise the question of whether +</td><td></td><td></td></tr><tr><td>d444e010049944c1b3438c9a25ae09b292b17371</td><td>Structure Preserving Video Prediction +<br/><b>Shanghai Institute for Advanced Communication and Data Science</b><br/>Shanghai Key Laboratory of Digital Media Processing and Transmission +<br/><b>Shanghai Jiao Tong University, Shanghai 200240, China</b></td><td>('47882735', 'Jingwei Xu', 'jingwei xu')<br/>('47889348', 'Shuo Cheng', 'shuo cheng')</td><td>{xjwxjw,nibingbing,Leezf,xkyang}@sjtu.edu.cn, acccheng94@gmail.com +</td></tr><tr><td>d46fda4b49bbc219e37ef6191053d4327e66c74b</td><td>Facial Expression Recognition Based on Complexity Perception Classification +<br/>Algorithm +<br/><b>School of Computer Science and Engineering, South China University of Technology, Guangzhou, China</b></td><td>('36047279', 'Tianyuan Chang', 'tianyuan chang')<br/>('9725901', 'Guihua Wen', 'guihua wen')<br/>('39946628', 'Yang Hu', 'yang hu')<br/>('35847383', 'JiaJiong Ma', 'jiajiong ma')</td><td>tianyuan_chang@163.com, crghwen@scut.edu.cn +</td></tr><tr><td>d448d67c6371f9abf533ea0f894ef2f022b12503</td><td>Weakly Supervised Collective Feature Learning from Curated Media +<br/>1. NTT Communication Science Laboratories, Japan. +<br/><b>University of Cambridge, United Kingdom</b><br/><b>The University of Tokyo, Japan</b><br/><b>Technical University of Munich, Germany</b><br/>5. Uber AI Labs, USA. +</td><td>('2374364', 'Yusuke Mukuta', 'yusuke mukuta')<br/>('34454585', 'Akisato Kimura', 'akisato kimura')<br/>('2584289', 'David B. Adrian', 'david b. adrian')<br/>('1983575', 'Zoubin Ghahramani', 'zoubin ghahramani')</td><td>mukuta@mi.t.u-tokyo.ac.jp, akisato@ieee.org, david.adrian@tum.de, zoubin@eng.cam.ac.uk +</td></tr><tr><td>d492dbfaa42b4f8b8a74786d7343b3be6a3e9a1d</td><td>Deep Cost-Sensitive and Order-Preserving Feature Learning for +<br/>Cross-Population Age Estimation +<br/><b>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences</b><br/><b>University of Chinese Academy of Sciences</b><br/>3 KingSoft Ltd. +<br/>4 CAS Center for Excellence in Brain Science and Intelligence Technology +<br/>5 Vimicro AI Chip Technology Corporation +<br/><b>Birkbeck University of London</b></td><td>('2168945', 'Kai Li', 'kai li')<br/>('1757173', 'Junliang Xing', 'junliang xing')<br/>('49734675', 'Chi Su', 'chi su')<br/>('40506509', 'Weiming Hu', 'weiming hu')<br/>('2373307', 'Yundong Zhang', 'yundong zhang')</td><td>{kai.li,jlxing,wmhu}@nlpr.ia.ac.cn suchi@kingsoft.com raymond@vimicro.com sjmaybank@dcs.bbk.ac.uk +</td></tr><tr><td>d444368421f456baf8c3cb089244e017f8d32c41</td><td>CNN for IMU Assisted Odometry Estimation using Velodyne LiDAR +</td><td>('3414588', 'Martin Velas', 'martin velas')<br/>('2131298', 'Michal Spanel', 'michal spanel')<br/>('1700956', 'Michal Hradis', 'michal hradis')<br/>('1785162', 'Adam Herout', 'adam herout')</td><td></td></tr><tr><td>d4885ca24189b4414031ca048a8b7eb2c9ac646c</td><td>Efficient Facial Representations for Age, Gender +<br/>and Identity Recognition in Organizing Photo +<br/>Albums using Multi-output CNN +<br/><b>Samsung-PDMI Joint AI Center</b><br/>Mathematics +<br/><b>National Research University Higher School of Economics</b><br/>Nizhny Novgorod, Russia +</td><td>('35153729', 'Andrey V. Savchenko', 'andrey v. savchenko')</td><td></td></tr><tr><td>d4c7d1a7a03adb2338704d2be7467495f2eb6c7b</td><td></td><td></td><td></td></tr><tr><td>d4001826cc6171c821281e2771af3a36dd01ffc0</td><td>Modélisation de contextes pour l’annotation sémantique +<br/>de vidéos +<br/>To cite this version: +<br/>Ecole Nationale Supérieure des Mines de Paris, 2013. Français. <NNT : 2013ENMP0051>. <pastel- +<br/>00958135> +<br/>HAL Id: pastel-00958135 +<br/>https://pastel.archives-ouvertes.fr/pastel-00958135 +<br/>Submitted on 11 Mar 2014 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('2482072', 'Nicolas Ballas', 'nicolas ballas')<br/>('2482072', 'Nicolas Ballas', 'nicolas ballas')</td><td></td></tr><tr><td>d46b4e6871fc9974542215f001e92e3035aa08d9</td><td>A Gabor Quotient Image for Face Recognition +<br/>under Varying Illumination +<br/><b>Mahanakorn University of Technology</b><br/>51 Cheum-Sampan Rd., Nong Chok, Bangkok, THAILAND 10530 +</td><td>('1805935', 'Sanun Srisuk', 'sanun srisuk')<br/>('2337544', 'Amnart Petpon', 'amnart petpon')</td><td>sanun@mut.ac.th, amnartpe@dtac.co.th +</td></tr><tr><td>d458c49a5e34263c95b3393386b5d76ba770e497</td><td>Middle-East Journal of Scientific Research 20 (1): 01-13, 2014 +<br/>ISSN 1990-9233 +<br/>© IDOSI Publications, 2014 +<br/>DOI: 10.5829/idosi.mejsr.2014.20.01.11434 +<br/>A Comparative Analysis of Gender Classification Techniques +<br/><b>Shaheed Zulfikar Ali Bhutto Institute of Science and Technology, Islamabad, Pakistan</b></td><td>('46883468', 'Sajid Ali Khan', 'sajid ali khan')<br/>('48767110', 'Maqsood Ahmad', 'maqsood ahmad')<br/>('2521631', 'Naveed Riaz', 'naveed riaz')</td><td></td></tr><tr><td>d454ad60b061c1a1450810a0f335fafbfeceeccc</td><td>Deep Regression Forests for Age Estimation +<br/>1 Key Laboratory of Specialty Fiber Optics and Optical Access Networks, +<br/><b>Shanghai Institute for Advanced Communication and Data Science</b><br/><b>School of Communication and Information Engineering, Shanghai University</b><br/><b>Johns Hopkins University</b><br/><b>College of Computer and Control Engineering, Nankai University 4 Hikvision Research</b></td><td>('41187410', 'Wei Shen', 'wei shen')<br/>('9544564', 'Yilu Guo', 'yilu guo')<br/>('47906413', 'Yan Wang', 'yan wang')<br/>('1681247', 'Kai Zhao', 'kai zhao')<br/>('49292319', 'Bo Wang', 'bo wang')</td><td>{shenwei1231,gyl.luan0,wyanny.9,zhaok1206,wangbo.yunze,alan.l.yuille}@gmail.com +</td></tr><tr><td>d40cd10f0f3e64fd9b0c2728089e10e72bea9616</td><td>Article +<br/>Enhancing Face Identification Using Local Binary +<br/>Patterns and K-Nearest Neighbors +<br/><b>School of Communication Engineering, Hangzhou Dianzi University, Xiasha Higher Education Zone</b><br/>Received: 21 March 2017; Accepted: 29 August 2017; Published: 5 September 2017 +</td><td>('11249315', 'Idelette Laure Kambi Beli', 'idelette laure kambi beli')<br/>('2826297', 'Chunsheng Guo', 'chunsheng guo')</td><td>Hangzhou 310018, China; guo.chsh@gmail.com +<br/>* Correspondence: kblaure@yahoo.fr +</td></tr><tr><td>d4ebf0a4f48275ecd8dbc2840b2a31cc07bd676d</td><td></td><td></td><td></td></tr><tr><td>d4e669d5d35fa0ca9f8d9a193c82d4153f5ffc4e</td><td>A Lightened CNN for Deep Face Representation +<br/>School of Computer and Communication Engineering +<br/><b>University of Science and Technology Beijing, Beijing, China</b><br/>National Laboratory of Pattern Recognition +<br/><b>Institute of Automation Chinese Academy of Sciences, Beijing, China</b></td><td>('2225749', 'Xiang Wu', 'xiang wu')<br/>('1705643', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>aflredxiangwu@gmail.com +<br/>{rhe, znsun}@nlpr.ia.ac.cn +</td></tr><tr><td>d46e793b945c4f391031656357625e902c4405e8</td><td>Face-off: Automatic Alteration of Facial Features +<br/>Department of Information Management +<br/><b>National Taiwan University of Science and Technology</b><br/>No. 43, Sec. 4, Keelung Road +<br/>Taipei, 106, Taiwan, ROC +</td><td>('40119465', 'Jia-Kai Chou', 'jia-kai chou')<br/>('2241272', 'Chuan-Kai Yang', 'chuan-kai yang')<br/>('2553196', 'Sing-Dong Gong', 'sing-dong gong')</td><td>A9409004@mail.ntust.edu.tw,ckyang@cs.ntust.edu.tw,hgznrn@uj.com.tw +</td></tr><tr><td>d44a93027208816b9e871101693b05adab576d89</td><td></td><td></td><td></td></tr><tr><td>d4c2d26523f577e2d72fc80109e2540c887255c8</td><td>Face-space Action Recognition by Face-Object Interactions +<br/><b>Weizmann Institute of Science</b><br/>Rehovot, 7610001, Israel +</td><td>('32928116', 'Amir Rosenfeld', 'amir rosenfeld')<br/>('1743045', 'Shimon Ullman', 'shimon ullman')</td><td>{amir.rosenfeld,shimon.ullman}@weizmann.ac.il +</td></tr><tr><td>d4b88be6ce77164f5eea1ed2b16b985c0670463a</td><td>TECHNICAL REPORT JAN.15.2016 +<br/>A Survey of Different 3D Face Reconstruction +<br/>Methods +<br/>Department of Computer Science and Engineering +</td><td>('2357264', 'Amin Jourabloo', 'amin jourabloo')</td><td>jourablo@msu.edu +</td></tr><tr><td>d44ca9e7690b88e813021e67b855d871cdb5022f</td><td>QUT Digital Repository: +<br/>http://eprints.qut.edu.au/ +<br/>Zhang, Ligang and Tjondronegoro, Dian W. (2009) Selecting, optimizing and +<br/>fusing ‘salient’ Gabor features for facial expression recognition. In: Neural +<br/>Information Processing (Lecture Notes in Computer Science), 1-5 December +<br/>2009, Hotel Windsor Suites Bangkok, Bangkok. +<br/> +<br/> © Copyright 2009 Springer-Verlag GmbH Berlin Heidelberg +<br/> +</td><td></td><td></td></tr><tr><td>baaaf73ec28226d60d923bc639f3c7d507345635</td><td><b>Stanford University</b><br/>CS229 : Machine Learning techniques +<br/>Project report +<br/>Emotion Classification on face images +<br/>Authors: +<br/>Instructor +<br/>December 12, 2015 +</td><td>('40503018', 'Mikael Jorda', 'mikael jorda')<br/>('2765850', 'Nina Miolane', 'nina miolane')<br/>('34699434', 'Andrew Ng', 'andrew ng')</td><td></td></tr><tr><td>ba2bbef34f05551291410103e3de9e82fdf9dddd</td><td>A Study on Cross-Population Age Estimation +<br/><b>LCSEE, West Virginia University</b><br/><b>LCSEE, West Virginia University</b></td><td>('1822413', 'Guodong Guo', 'guodong guo')<br/>('1720735', 'Chao Zhang', 'chao zhang')</td><td>guodong.guo@mai1.wvu.edu +<br/>cazhang@mix.wvu.edu +</td></tr><tr><td>bafb8812817db7445fe0e1362410a372578ec1fc</td><td>805 +<br/>Image-Quality-Based Adaptive Face Recognition +</td><td>('2284264', 'Harin Sellahewa', 'harin sellahewa')</td><td></td></tr><tr><td>baa0fe4d0ac0c7b664d4c4dd00b318b6d4e09143</td><td>International Journal of Signal Processing, Image Processing and Pattern Recognition +<br/>Vol. 8, No. 1 (2015), pp. 9-22 +<br/>http://dx.doi.org/10.14257/ijsip.2015.8.1.02 +<br/>Facial Expression Analysis using Active Shape Model +<br/><b>School of Engineering, University of Portsmouth, United Kingdom</b></td><td>('2226048', 'Reda Shbib', 'reda shbib')<br/>('32991189', 'Shikun Zhou', 'shikun zhou')</td><td>reda.shbib@port.ac.uk, Shikun.zhou@port.ac.uk +</td></tr><tr><td>ba99c37a9220e08e1186f21cab11956d3f4fccc2</td><td>A Fast Factorization-based Approach to Robust PCA +<br/><b>Southern Illinois University, Carbondale, IL 62901 USA</b></td><td>('33048613', 'Chong Peng', 'chong peng')<br/>('1686710', 'Zhao Kang', 'zhao kang')<br/>('39951979', 'Qiang Cheng', 'qiang cheng')</td><td>Email: {pchong,zhao.kang,qcheng}@siu.edu +</td></tr><tr><td>ba816806adad2030e1939450226c8647105e101c</td><td>MindLAB at the THUMOS Challenge +<br/>Fabi´an P´aez +<br/>Fabio A. Gonz´alez +<br/>MindLAB Research Group +<br/>MindLAB Research Group +<br/>MindLAB Research Group +<br/>Bogot´a, Colombia +<br/>Bogot´a, Colombia +<br/>Bogot´a, Colombia +</td><td>('1939861', 'Jorge A. Vanegas', 'jorge a. vanegas')</td><td>fmpaezri@unal.edu.co +<br/>javanegasr@unal.edu.co +<br/>fagonzalezo@unal.edu.co +</td></tr><tr><td>badcd992266c6813063c153c41b87babc0ba36a3</td><td>Recent Advances in Object Detection in the Age +<br/>of Deep Convolutional Neural Networks +<br/>,1,2), Fr´ed´eric Jurie(1) +<br/>(∗) equal contribution +<br/>(1)Normandie Univ, UNICAEN, ENSICAEN, CNRS +<br/>(2)Safran Electronics and Defense +<br/>September 11, 2018 +</td><td>('51443250', 'Shivang Agarwal', 'shivang agarwal')<br/>('35527701', 'Jean Ogier Du Terrail', 'jean ogier du terrail')</td><td></td></tr><tr><td>ba788365d70fa6c907b71a01d846532ba3110e31</td><td></td><td></td><td></td></tr><tr><td>badcfb7d4e2ef0d3e332a19a3f93d59b4f85668e</td><td>The Application of Extended Geodesic Distance +<br/>in Head Poses Estimation +<br/><b>Institute of Computing Technology</b><br/>Chinese Academy of Sciences, Beijing 100080, China +<br/>2 Department of Computer Science and Engineering, +<br/><b>Harbin Institute of Technology, Harbin, China</b><br/>3 Graduate School of the Chinese Academy of Sciences, Beijing 100039, China +</td><td>('1798982', 'Bingpeng Ma', 'bingpeng ma')<br/>('1684164', 'Fei Yang', 'fei yang')<br/>('1698902', 'Wen Gao', 'wen gao')<br/>('1740430', 'Baochang Zhang', 'baochang zhang')</td><td></td></tr><tr><td>ba8a99d35aee2c4e5e8a40abfdd37813bfdd0906</td><td>ELEKTROTEHNI ˇSKI VESTNIK 78(1-2): 12–17, 2011 +<br/>EXISTING SEPARATE ENGLISH EDITION +<br/>Uporaba emotivno pogojenega raˇcunalniˇstva v +<br/>priporoˇcilnih sistemih +<br/>Marko Tkalˇciˇc, Andrej Koˇsir, Jurij Tasiˇc +<br/>1Univerza v Ljubljani, Fakulteta za elektrotehniko, Trˇzaˇska 25, 1000 Ljubljana, Slovenija +<br/>2Univerza v Ljubljani, Fakulteta za raˇcunalniˇstvo in informatiko, Trˇzaˇska 25, 1000 Ljubljana, Slovenija +<br/>Povzetek. V ˇclanku predstavljamo rezultate treh raziskav, vezanih na izboljˇsanje delovanja multimedijskih +<br/>priporoˇcilnih sistemov s pomoˇcjo metod emotivno pogojenega raˇcunalniˇstva (ang. affective computing). +<br/>Vsebinski priporoˇcilni sistem smo izboljˇsali s pomoˇcjo metapodatkov, ki opisujejo emotivne odzive uporabnikov. +<br/>Pri skupinskem priporoˇcilnem sistemu smo dosegli znaˇcilno izboljˇsanje v obmoˇcju hladnega zagona z uvedbo +<br/>nove mere podobnosti, ki temelji na osebnostnem modelu velikih pet (ang. five factor model). Razvili smo tudi +<br/>sistem za neinvazivno oznaˇcevanje vsebin z emotivnimi parametri, ki pa ˇse ni zrel za uporabo v priporoˇcilnih +<br/>sistemih. +<br/>Kljuˇcne besede: priporoˇcilni sistemi, emotivno pogojeno raˇcunalniˇstvo, strojno uˇcenje, uporabniˇski profil, +<br/>emocije +<br/>Uporaba emotivnega raˇcunalniˇstva v priporoˇcilnih +<br/>sistemih +<br/>In this paper we present the results of three investigations of +<br/>our broad research on the usage of affect and personality in +<br/>recommender systems. We improved the accuracy of content- +<br/>based recommender system with the inclusion of affective +<br/>parameters of user and item modeling. We improved the +<br/>accuracy of a content filtering recommender system under the +<br/>cold start conditions with the introduction of a personality +<br/>based user similarity measure. Furthermore we developed a +<br/>system for implicit tagging of content with affective metadata. +<br/>1 UVOD +<br/>Uporabniki (porabniki) multimedijskih (MM) vsebin so +<br/>v ˇcedalje teˇzjem poloˇzaju, saj v veliki koliˇcini vse- +<br/>bin teˇzko najdejo zanje primerne. Pomagajo si s pri- +<br/>poroˇcilnimi sistemi, ki na podlagi osebnih preferenc +<br/>uporabnikov izberejo manjˇso koliˇcino relevantnih MM +<br/>vsebin, med katerimi uporabnik laˇze izbira. Noben danes +<br/>znan priporoˇcilni sistem ne zadoˇsˇca v celoti potrebam +<br/>uporabnikov, saj je izbor priporoˇcenih vsebin obiˇcajno +<br/>nezadovoljive kakovosti [10]. Cilj tega ˇclanka je pred- +<br/>staviti metode emotivno pogojenega raˇcunalniˇstva (ang. +<br/>affective computing - glej [12]) za izboljˇsanje kakovosti +<br/>priporoˇcilnih sistemov in utrditi za slovenski prostor +<br/>novo terminologijo. +<br/>1.1 Opis problema +<br/>Za izboljˇsanje kakovosti priporoˇcilnih sistemov sta +<br/>na voljo dve poti: (i) optimizacija algoritmov ali (ii) +<br/>uporaba boljˇsih znaˇcilk, ki bolje razloˇzijo neznano +<br/>Prejet 13. oktober, 2010 +<br/>Odobren 1. februar, 2011 +<br/>varianco [8]. V tem ˇclanku predstavljamo izboljˇsanje +<br/>priporoˇcilnih sistemov z uporabo novih znaˇcilk, ki te- +<br/>meljijo na emotivnih odzivih uporabnikov in na njiho- +<br/>vih osebnostnih lastnostih. Te znaˇcilke razloˇzijo velik +<br/>del uporabnikovih preferenc, ki se izraˇzajo v obliki +<br/>ocen posameznih vsebin (npr. Likertova lestvica, binarne +<br/>ocene itd.). Ocene vsebin se pri priporoˇcilnih sistemih +<br/>zajemajo eksplicitno (ocena) ali implicitno, pri ˇcemer o +<br/>oceni sklepamo na podlagi opazovanj (npr. ˇcas gledanja +<br/>kot indikator vˇseˇcnosti [7]. +<br/>Izboljˇsanja uˇcinkovitosti priporoˇcilnih sistemov smo +<br/>se lotili na treh podroˇcjih: (i) uporaba emotivnega +<br/>modeliranja uporabnikov v vsebinskem priporoˇcilnem +<br/>sistemu, (ii) neinvazivna (implicitna) detekcija emocij za +<br/>emotivno modeliranje in (iii) uporaba osebnostne mere +<br/>podobnosti v skupinskem priporoˇcilnem sistemu. Slika 1 +<br/>prikazuje arhitekturo emotivnega priporoˇcilnega sistema +<br/>in mesta, kjer smo vnesli opisane izboljˇsave. +<br/>Preostanek ˇclanka je strukturiran tako: v razdelku +<br/>2 je predstavljen zajem podatkov. V razdelku 3 je +<br/>predstavljen vsebinski priporoˇcilni sistem z emotivnimi +<br/>metapodatki. V razdelku 4 je predstavljen skupinski +<br/>priporoˇcilni sistem, ki uporablja mero podobnosti na +<br/>podlagi osebnosti, v razdelku 5 pa algoritem za razpo- +<br/>znavo emocij. Vsak od teh razdelov je sestavljen iz opisa +<br/>eksperimenta in predstavitve rezultatov. V razdelku 6 so +<br/>predstavljeni sklepi. +<br/>1.2 Sorodno delo +<br/>Najbolj groba delitev priporoˇcilnih sistemov je na vse- +<br/>binske, skupinske ter hibridne sisteme [1]. Z izjemo vse- +<br/>binskih priporoˇcilnih sistemov, ki sta ga razvila Arapakis +<br/>[2] in Tkalˇciˇc [14], sorodnega dela na podroˇcju emotivno +<br/>pogojenih priporoˇcilnih sistemov takorekoˇc ni. Panti´c in +</td><td></td><td>E-poˇsta: avtor@naslov.com +</td></tr><tr><td>bac11ce0fb3e12c466f7ebfb6d036a9fe62628ea</td><td>Weakly Supervised Learning of Heterogeneous +<br/>Concepts in Videos +<br/>Larry Davis1 +<br/><b>University of Maryland, College Park; 2Arizona State University; 3Xerox Research Centre</b><br/>India +</td><td>('36861219', 'Sohil Shah', 'sohil shah')<br/>('40222634', 'Kuldeep Kulkarni', 'kuldeep kulkarni')<br/>('2221075', 'Arijit Biswas', 'arijit biswas')<br/>('2757149', 'Ankit Gandhi', 'ankit gandhi')<br/>('2116262', 'Om Deshmukh', 'om deshmukh')</td><td></td></tr><tr><td>ba29ba8ec180690fca702ad5d516c3e43a7f0bb8</td><td></td><td></td><td></td></tr><tr><td>ba7b12c8e2ff3c5e4e0f70b58215b41b18ff8feb</td><td>Natural and Effective Obfuscation by Head Inpainting +<br/><b>Max Planck Institute for Informatics, Saarland Informatics Campus</b><br/>2KU-Leuven/PSI, Toyota Motor Europe (TRACE) +<br/>3ETH Zurich +</td><td>('32222907', 'Qianru Sun', 'qianru sun')<br/>('1681236', 'Luc Van Gool', 'luc van gool')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td>{qsun, joon, schiele, mfritz}@mpi-inf.mpg.de +<br/>{liqian.ma, luc.vangool}@esat.kuleuven.be +<br/>vangool@vision.ee.ethz.ch +</td></tr><tr><td>bab88235a30e179a6804f506004468aa8c28ce4f</td><td></td><td></td><td></td></tr><tr><td>badd371a49d2c4126df95120902a34f4bee01b00</td><td>GONDA, WEI, PARAG, PFISTER: PARALLEL SEPARABLE 3D CONVOLUTION +<br/>Parallel Separable 3D Convolution for Video +<br/>and Volumetric Data Understanding +<br/>Harvard John A. Paulson School of +<br/>Engineering and Applied Sciences +<br/>Camabridge MA, USA +<br/>Toufiq Parag +<br/>Hanspeter Pfister +</td><td>('49147616', 'Felix Gonda', 'felix gonda')<br/>('1766333', 'Donglai Wei', 'donglai wei')</td><td>fgonda@g.harvard.edu +<br/>donglai@seas.harvard.edu +<br/>paragt@seas.harvard.edu +<br/>pfister@g.harvard.edu +</td></tr><tr><td>a065080353d18809b2597246bb0b48316234c29a</td><td>FHEDN: A based on context modeling Feature Hierarchy +<br/>Encoder-Decoder Network for face detection +<br/><b>College of Computer Science, Chongqing University, Chongqing, China</b><br/><b>College of Medical Informatics, Chongqing Medical University, Chongqing, China</b><br/><b>Sichuan Fine Arts Institute, Chongqing, China</b></td><td>('6030130', 'Zexun Zhou', 'zexun zhou')<br/>('7686690', 'Zhongshi He', 'zhongshi he')<br/>('2685579', 'Ziyu Chen', 'ziyu chen')<br/>('33458882', 'Yuanyuan Jia', 'yuanyuan jia')<br/>('1768826', 'Haiyan Wang', 'haiyan wang')<br/>('8784203', 'Jinglong Du', 'jinglong du')<br/>('2961485', 'Dingding Chen', 'dingding chen')</td><td>{zexunzhou,zshe,chenziyu,yyjia,jldu,dingding}@cqu.edu.cn;{why}@scfai.edu.cn +</td></tr><tr><td>a0f94e9400938cbd05c4b60b06d9ed58c3458303</td><td>1118 +<br/>Value-Directed Human Behavior Analysis +<br/>from Video Using Partially Observable +<br/>Markov Decision Processes +</td><td>('1773895', 'Jesse Hoey', 'jesse hoey')<br/>('1710980', 'James J. Little', 'james j. little')</td><td></td></tr><tr><td>a022eff5470c3446aca683eae9c18319fd2406d5</td><td>2017-ENST-0071 +<br/>EDITE - ED 130 +<br/>Doctorat ParisTech +<br/>T H È S E +<br/>pour obtenir le grade de docteur délivré par +<br/>TÉLÉCOM ParisTech +<br/>Spécialité « SIGNAL et IMAGES » +<br/>présentée et soutenue publiquement par +<br/>le 15 décembre 2017 +<br/>Apprentissage Profond pour la Description Sémantique des Traits +<br/>Visuels Humains +<br/>Directeur de thèse : Jean-Luc DUGELAY +<br/>Co-encadrement de la thèse : Moez BACCOUCHE +<br/>Jury +<br/>Mme Bernadette DORIZZI, PRU, Télécom SudParis +<br/>Mme Jenny BENOIS-PINEAU, PRU, Université de Bordeaux +<br/>M. Christian WOLF, MC/HDR, INSA de Lyon +<br/>M. Patrick PEREZ, Chercheur/HDR, Technicolor Rennes +<br/>M. Moez BACCOUCHE, Chercheur/Docteur, Orange Labs Rennes +<br/>M. Jean-Luc DUGELAY, PRU, Eurecom Sophia Antipolis +<br/>M. Sid-Ahmed BERRANI, Directeur de l’Innovation/HDR, Algérie Télécom +<br/>Présidente +<br/>Rapporteur +<br/>Rapporteur +<br/>Examinateur +<br/>Encadrant +<br/>Directeur de Thèse +<br/>Invité +<br/>TÉLÉCOM ParisTech +<br/>école de l’Institut Télécom - membre de ParisTech +<br/>N°: 2009 ENAM XXXX T H È S E </td><td>('3116433', 'Grigory Antipov', 'grigory antipov')</td><td></td></tr><tr><td>a0f193c86e3dd7e0020c0de3ec1e24eaff343ce4</td><td>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 21, 819-828 (2005) +<br/>Short Paper_________________________________________________ +<br/>A New Classification Approach using +<br/>Discriminant Functions +<br/>Department of Computer Engineering +<br/>+Department of Electrical and Electronic Engineering +<br/><b>Sakarya University</b><br/>54187 Sakarya, Turkey +<br/>In this study, an approach involving new types of cost functions is given for the +<br/>construction of discriminant functions. Centers of mass, not specified a priori, around +<br/>feature vectors are clustered using cost function. Thus, the algorithms yield both the +<br/>centers of mass and the distinct classes. +<br/>Keywords: classification, feature vectors, linear discriminant function, Fisher’s LDF, +<br/>dimension reduction +<br/>1. INTRODUCTION +<br/>There are many algorithms for, and many applications of classification and dis- +<br/>crimination (grouping of a set of objects into subsets of similar objects where the objects +<br/>in different subsets are different) in several diverse fields [2-15, 23, 24], ranging from +<br/>engineering to medicine, to econometrics, etc. Some examples are automatic target rec- +<br/>ognition (ATR), fault and maintenance-time recognition, optical character recognition +<br/>(OCR), speech and speaker recognition, etc. +<br/>In this study, a new approach and algorithm to the classification problem are de- +<br/>scribed with the goal of finding a single (possibly vector-valued) linear discriminant +<br/>function. This approach is in terms of some optimal centers of mass for the transformed +<br/>feature vectors of each class, the transforms being performed via the discriminant func- +<br/>tions. As such, it follows the same philosophy which is behind the approaches such as +<br/>principal component analysis (PCA), Fisher’s linear discriminant functions (LDF), and +<br/>minimum total covariance (MTC) [1-16, 22, 25-28], providing alternatives which extend +<br/>this work. +<br/>Linear discriminant functions (LDF) are often used in pattern recognition to classify +<br/>a given object or pattern, based on its features, into one of several given classes. For sim- +<br/>plicity, consider the discrimination problem for two classes. Let x = [x1, x2, …, xm] be the +<br/>Received April 28, 2003; revised March 1 and March 29, 2004; accepted May 3, 2004. +<br/>Communicated by H. Y. Mark Liao. +<br/>819 +</td><td>('7605725', 'Zafer Demir', 'zafer demir')<br/>('2279264', 'Erol Emre', 'erol emre')</td><td>E-mail: {askind, zdemir, eemre}@sakarya.edu.tr +</td></tr><tr><td>a0c37f07710184597befaa7e6cf2f0893ff440e9</td><td></td><td></td><td></td></tr><tr><td>a0dc68c546e0fc72eb0d9ca822cf0c9ccb4b4c4f</td><td>Fusing with Context: a Bayesian Approach to Combining Descriptive Attributes +<br/><b>University of Colorado at Colorado Springs and Securics, Inc., Colorado Springs, CO, USA</b><br/><b>Columbia University, New York, NY, USA</b><br/><b>University of North Carolina Wilmington, Wilmington, NC, USA</b></td><td>('2613438', 'Walter J. Scheirer', 'walter j. scheirer')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')</td><td></td></tr><tr><td>a0021e3bbf942a88e13b67d83db7cf52e013abfd</td><td>Human concerned object detecting in video +<br/><b>School of Computer Science and Technology, Shandong Institute of Business and Technology</b><br/>Yantai, Shandong, 264005, China +<br/><b>School of Computer Science and Technology, Shandong University</b><br/>Jinan, Shandong, 250101, China +<br/>Received 11 December 2014 +</td><td>('2525711', 'Jinjiang LI', 'jinjiang li')<br/>('1733582', 'Jie GUO', 'jie guo')<br/>('9242942', 'Hui FAN', 'hui fan')</td><td>E-mail: lijinjiang@gmail.com +</td></tr><tr><td>a0d6390dd28d802152f207940c7716fe5fae8760</td><td>Bayesian Face Revisited: A Joint Formulation +<br/><b>University of Science and Technology of China</b><br/><b>The Chinese University of Hong Kong</b><br/>3 Microsoft Research Asia, Beijing, China +</td><td>('39447786', 'Dong Chen', 'dong chen')<br/>('2032273', 'Xudong Cao', 'xudong cao')<br/>('34508239', 'Liwei Wang', 'liwei wang')<br/>('1716835', 'Fang Wen', 'fang wen')<br/>('40055995', 'Jian Sun', 'jian sun')</td><td>chendong@mail.ustc.edu.cn +<br/>lwwang@cse.cuhk.edu.hk +<br/>{xudongca,fangwen,jiansun}@microsoft.com +</td></tr><tr><td>a0fb5b079dd1ee5ac6ac575fe29f4418fdb0e670</td><td></td><td></td><td></td></tr><tr><td>a0fd85b3400c7b3e11122f44dc5870ae2de9009a</td><td>Learning Deep Representation for Face +<br/>Alignment with Auxiliary Attributes +</td><td>('3152448', 'Zhanpeng Zhang', 'zhanpeng zhang')<br/>('47571885', 'Ping Luo', 'ping luo')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>a0dfb8aae58bd757b801e2dcb717a094013bc178</td><td>Reconocimiento de expresiones faciales con base +<br/>en la din´amica de puntos de referencia faciales +<br/>Instituto Nacional de Astrof´ısica ´Optica y Electr´onica, +<br/>Divisi´on de Ciencias Computacionales, Tonantzintla, Puebla, +<br/>M´exico +<br/>Resumen. Las expresiones faciales permiten a las personas comunicar +<br/>emociones, y es pr´acticamente lo primero que observamos al interactuar +<br/>con alguien. En el ´area de computaci´on, el reconocimiento de expresiones +<br/>faciales es importante debido a que su an´alisis tiene aplicaci´on directa en +<br/>´areas como psicolog´ıa, medicina, educaci´on, entre otras. En este articulo +<br/>se presenta el proceso de dise˜no de un sistema para el reconocimiento de +<br/>expresiones faciales utilizando la din´amica de puntos de referencia ubi- +<br/>cados en el rostro, su implementaci´on, experimentos realizados y algunos +<br/>de los resultados obtenidos hasta el momento. +<br/>Palabras clave: Expresiones faciales, clasificaci´on, m´aquinas de soporte +<br/>vectorial,modelos activos de apariencia. +<br/>Facial Expressions Recognition Based on Facial +<br/>Landmarks Dynamics +</td><td>('40452660', 'E. Morales-Vargas', 'e. morales-vargas')<br/>('2737777', 'Hayde Peregrina-Barreto', 'hayde peregrina-barreto')</td><td>emoralesv@inaoep.mx, kargaxxi@inaoep.mx, hperegrina@inaoep.mx +</td></tr><tr><td>a0aa32bb7f406693217fba6dcd4aeb6c4d5a479b</td><td>Cascaded Regressor based 3D Face Reconstruction +<br/>from a Single Arbitrary View Image +<br/><b>College of Computer Science, Sichuan University, Chengdu, China</b></td><td>('50207647', 'Feng Liu', 'feng liu')<br/>('39422721', 'Dan Zeng', 'dan zeng')<br/>('1723081', 'Jing Li', 'jing li')<br/>('7345195', 'Qijun Zhao', 'qijun zhao')</td><td>qjzhao@scu.edu.cn +</td></tr><tr><td>a03cfd5c0059825c87d51f5dbf12f8a76fe9ff60</td><td>Simultaneous Learning and Alignment: +<br/>Multi-Instance and Multi-Pose Learning? +<br/>1 Comp. Science & Eng. +<br/>Univ. of CA, San Diego +<br/>2 Electrical Engineering +<br/>California Inst. of Tech. +<br/>3 Lab of Neuro Imaging +<br/>Univ. of CA, Los Angeles +</td><td>('2490700', 'Boris Babenko', 'boris babenko')<br/>('1736745', 'Zhuowen Tu', 'zhuowen tu')<br/>('1769406', 'Serge Belongie', 'serge belongie')</td><td>{bbabenko,sjb}@cs.ucsd.edu +<br/>pdollar@caltech.edu +<br/>zhuowen.tu@loni.ucla.edu +</td></tr><tr><td>a06b6d30e2b31dc600f622ab15afe5e2929581a7</td><td>Robust Joint and Individual Variance Explained +<br/><b>Imperial College London, UK</b><br/>2Onfido, UK +<br/><b>Middlesex University London, UK</b></td><td>('3320415', 'Christos Sagonas', 'christos sagonas')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('28943361', 'Alina Leidinger', 'alina leidinger')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>christos.sagonas@onfido.com, {i.panagakis, s.zafeiriou}@imperial.ac.uk +</td></tr><tr><td>a0b1990dd2b4cd87e4fd60912cc1552c34792770</td><td>Deep Constrained Local Models for Facial Landmark Detection +<br/><b>Carnegie Mellon University</b><br/>Tadas Baltruaitis +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +<br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +</td><td>('1783029', 'Amir Zadeh', 'amir zadeh')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td>abagherz@cs.cmu.edu +<br/>tbaltrus@cs.cmu.edu +<br/>morency@cs.cmu.edu +</td></tr><tr><td>a090d61bfb2c3f380c01c0774ea17929998e0c96</td><td>On the Dimensionality of Video Bricks under Varying Illumination +<br/>Beijing Lab of Intelligent Information Technology, School of Computer Science, +<br/><b>Beijing Institute of Technology, Beijing 100081, PR China</b></td><td>('2852150', 'Youdong Zhao', 'youdong zhao')<br/>('38150687', 'Xi Song', 'xi song')<br/>('7415267', 'Yunde Jia', 'yunde jia')</td><td>{zyd458, songxi, jiayunde}@bit.edu.cn +</td></tr><tr><td>a0e7f8771c7d83e502d52c276748a33bae3d5f81</td><td>Ensemble Nystr¨om +<br/>A common problem in many areas of large-scale machine learning involves ma- +<br/>nipulation of a large matrix. This matrix may be a kernel matrix arising in Support +<br/>Vector Machines [9, 15], Kernel Principal Component Analysis [47] or manifold +<br/>learning [43,51]. Large matrices also naturally arise in other applications, e.g., clus- +<br/>tering, collaborative filtering, matrix completion, and robust PCA. For these large- +<br/>scale problems, the number of matrix entries can easily be in the order of billions +<br/>or more, making them hard to process or even store. An attractive solution to this +<br/>problem involves the Nystr¨om method, in which one samples a small number of +<br/>columns from the original matrix and generates its low-rank approximation using +<br/>the sampled columns [53]. The accuracy of the Nystr¨om method depends on the +<br/>number columns sampled from the original matrix. Larger the number of samples, +<br/>higher the accuracy but slower the method. +<br/>In the Nystr¨om method, one needs to perform SVD on a l × l matrix where l is +<br/>the number of columns sampled from the original matrix. This SVD operation is +<br/>typically carried out on a single machine. Thus, the maximum value of l used for an +<br/>application is limited by the capacity of the machine. That is why in practice, one +<br/>restricts l to be less than 20K or 30K, even when the size of matrix is in millions. +<br/>This restricts the accuracy of the Nystr¨om method in very large-scale settings. +<br/>This chapter describes a family of algorithms based on mixtures of Nystr¨om +<br/>approximations called, Ensemble Nystr¨om algorithms, which yields more accurate +<br/>low-rank approximations than the standard Nystr¨om method. The core idea of En- +<br/>semble Nystr¨om is to sample many subsets of columns from the original matrix, +<br/>each containing a relatively small number of columns. Then, Nystr¨om method is +<br/><b>Division of Computer Science, University of California, Berkeley, CA, USA e-mail</b></td><td>('2794322', 'Sanjiv Kumar', 'sanjiv kumar')<br/>('1709415', 'Mehryar Mohri', 'mehryar mohri')<br/>('8395559', 'Ameet Talwalkar', 'ameet talwalkar')<br/>('2794322', 'Sanjiv Kumar', 'sanjiv kumar')<br/>('1709415', 'Mehryar Mohri', 'mehryar mohri')<br/>('8395559', 'Ameet Talwalkar', 'ameet talwalkar')</td><td>Google Research, New York, NY, USA e-mail: sanjivk@google.com +<br/>Courant Institute, New York, NY, USA e-mail: mohri@cs.nyu.edu +<br/>ameet@eecs.berkeley.edu +</td></tr><tr><td>a0061dae94d916f60a5a5373088f665a1b54f673</td><td>Research Article +<br/>Lensless computational imaging through deep +<br/>learning +<br/><b>Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA</b><br/><b>Institute for Medical Engineering Science, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA</b><br/>3Singapore-MIT Alliance for Research and Technology (SMART) Centre, One Create Way, Singapore 117543, Singapore +<br/>†These authors contributed equally +<br/>Compiled March 1, 2017 +<br/>Deep learning has been proven to yield reliably generalizable answers to numerous classification and +<br/>decision tasks. Here, we demonstrate for the first time, to our knowledge, that deep neural networks +<br/>(DNNs) can be trained to solve inverse problems in computational imaging. We experimentally demon- +<br/>strate a lens-less imaging system where a DNN was trained to recover a phase object given a raw +<br/>intensity image recorded some distance away. © +<br/>OCIS codes: +<br/>(110.1758) Computational imaging. +<br/>(100.3190) Inverse problems; (100.4996) Pattern recognition, neural networks; (100.5070) Phase retrieval; +<br/>http://dx.doi.org/10.1364/optica.XX.XXXXXX +<br/>1. INTRODUCTION +<br/>Neural network training can be thought of as generic function approxi- +<br/>mation, as follows: given a training set (i.e., examples of matched input +<br/>and output data obtained from a hitherto-unknown model), generate +<br/>the computational architecture that most accurately maps all inputs in +<br/>In this paper, we propose that deep neural networks may “learn” to +<br/>approximate solutions to inverse problems in computational imaging. +<br/>A general computational imaging system consists of a physical part +<br/>where light propagates through one or more objects of interest as well +<br/>as optical elements such as lenses, prisms, etc. finally producing a +<br/>raw intensity image on a digital camera. The raw intensity image is +<br/>then computationally processed to yield object attributes, e.g. a spatial +<br/>map of light attenuation and/or phase delay through the object—what +<br/>we call traditionally “intensity image” and “quantitative phase image,” +<br/>respectively. The computational part of the system is then said to solve +<br/>the inverse problem. +<br/>The study of inverse problems is traced back at least a century ago +<br/>to Tikhonov [1] and Wiener [2]. A good introductory book with rigor- +<br/>ous but not overwhelming discussion of the underlying mathematical +<br/>concepts, especially regularization, is [3]. During the past decade, the +<br/>field experienced a renaissance due to the almost simultaneous matura- +<br/>tion of two related mathematics disciplines: convex optimization and +<br/>harmonic analysis, especially sparse representations. A light technical +<br/>introduction to these fascinating developments is in [4]. +<br/>Neural networks have their own history of legendary ups-and-downs +<br/>[5] culminating with an even more recent renaissance. This was driven +<br/>by Hinton’s insight that multi-layer architectures with numerous layers, +<br/>dubbed as “deep networks,” DNNs, can generalize better than had been +<br/>previously thought after some simple but ingenious changes in the +<br/>nonlinearity and training algorithms [6]. Even more recently developed +<br/>architectures [7–9] have enabled neural networks to “learn deeper;” +<br/>and modern DNNs have shown spectacular success at solving “hard” +<br/>computational problems, such as: playing complex games like Atari +<br/>[17] and Go [18], object detection [19], and image restoration (e.g., +<br/>colorization [20], deblurring [21–23], in-painting [24]). +<br/>The idea of using neural networks to clean up images isn’t exactly +<br/>new: for example, Hopfield’s associative memory network [25] was +<br/>capable of retrieving entire faces from partially obscured inputs, and +<br/>was implemented in an all-optical architecture [26] when computers +<br/>weren’t nearly as powerful as they are now. Recently, Horisaki et al. +<br/>[27] used support-vector machines, a form of bi-layer neural network +<br/>with nonlinear discriminant functions, also to recover face images +<br/>when the obscuration is caused by scattering media. +<br/>The hypothesis that we set out to test in this paper is whether +<br/>a neural network can be trained by being presented pairs of known +<br/>objects and their raw intensity image representations on the digital +<br/>camera of a computational imaging system; and then be used to produce +<br/>object estimates given raw intensity images from hitherto unknown +<br/>test objects, thus solving the inverse problem. This is a rather general +<br/>question and may take several flavors, depending on the nature of the +<br/>object, the physical design of the imaging system, etc. We chose to +</td><td>('3365480', 'Ayan Sinha', 'ayan sinha')<br/>('2371140', 'Justin Lee', 'justin lee')<br/>('1804684', 'Shuai Li', 'shuai li')<br/>('2455899', 'George Barbastathis', 'george barbastathis')</td><td></td></tr><tr><td>a0848d7b1bb43f4b4f1b4016e58c830f40944817</td><td>Face matching for post-disaster family reunification +<br/><b>Lister Hill National Center for Biomedical Communications, National Library of Medicine, National Institutes of Health</b><br/>8600 Rockville Pike, Bethesda, MD USA +</td><td>('1744255', 'Eugene Borovikov', 'eugene borovikov')<br/>('2075836', 'Girish Lingappa', 'girish lingappa')</td><td>FaceMatch@NIH.gov +</td></tr><tr><td>a000149e83b09d17e18ed9184155be140ae1266e</td><td>Chapter 9 +<br/>Action Recognition in Realistic +<br/>Sports Videos +</td><td>('1799979', 'Khurram Soomro', 'khurram soomro')<br/>('40029556', 'Amir R. Zamir', 'amir r. zamir')</td><td></td></tr><tr><td>a01f9461bc8cf8fe40c26d223ab1abea5d8e2812</td><td>Facial Age Estimation Through the Fusion of Texture +<br/>and local appearance Descriptors +<br/><b>DPDCE, University IUAV, Santa Croce 1957, 30135 Venice, Italy</b><br/>2 Herta Security, Pau Claris 165 4-B, 08037 Barcelona, Spain +</td><td>('1733945', 'Andrea Prati', 'andrea prati')</td><td>huertacasado@iuav.it, aprati@iuav.it +<br/>carles.fernandez@hertasecurity.com +</td></tr><tr><td>a702fc36f0644a958c08de169b763b9927c175eb</td><td>FACIAL EXPRESSION RECOGNITION USING HOUGH FOREST +<br/><b>National Tsing-Hua University, Hsin-Chu, Taiwan</b><br/><b>Asia University, Taichung, Taiwan</b></td><td>('2867389', 'Chi-Ting Hsu', 'chi-ting hsu')<br/>('2790846', 'Shih-Chung Hsu', 'shih-chung hsu')<br/>('1793389', 'Chung-Lin Huang', 'chung-lin huang')</td><td>Email: s9961601@m99.nthu.edu.tw, d9761817@oz.nthu.edu.tw, clhuang@ee.nthu.edu.tw +</td></tr><tr><td>a7267bc781a4e3e79213bb9c4925dd551ea1f5c4</td><td>Proceedings of eNTERFACE’15 +<br/>The 11th Summer Workshop +<br/>on Multimodal Interfaces +<br/>August 10th - September 4th, 2015 +<br/><b>Numediart Institute, University of Mons</b><br/>Mons, Belgium +<br/> +</td><td></td><td></td></tr><tr><td>a784a0d1cea26f18626682ab108ce2c9221d1e53</td><td>Anchored Regression Networks applied to Age Estimation and Super Resolution +<br/>D-ITET, ETH Zurich +<br/>Switzerland +<br/>D-ITET, ETH Zurich +<br/>Merantix GmbH +<br/>D-ITET, ETH Zurich +<br/>ESAT, KU Leuven +</td><td>('2794259', 'Eirikur Agustsson', 'eirikur agustsson')<br/>('1732855', 'Radu Timofte', 'radu timofte')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>aeirikur@vision.ee.ethz.ch +<br/>timofter@vision.ee.ethz.ch +<br/>vangool@vision.ee.ethz.ch +</td></tr><tr><td>a77e9f0bd205a7733431a6d1028f09f57f9f73b0</td><td>Multimodal feature fusion for CNN-based gait recognition: an +<br/>empirical comparison +<br/>F.M. Castroa,, M.J. Mar´ın-Jim´enezb, N. Guila, N. P´erez de la Blancac +<br/><b>University of Malaga, Spain</b><br/><b>University of Cordoba, Spain</b><br/><b>University of Granada, Spain</b></td><td></td><td></td></tr><tr><td>a74251efa970b92925b89eeef50a5e37d9281ad0</td><td></td><td></td><td></td></tr><tr><td>a7d23c699a5ae4ad9b8a5cbb8c38e5c3b5f5fb51</td><td>Postgraduate Annual Research Seminar 2007 (3-4 July 2007) +<br/>A Summary of literature review : Face Recognition +<br/>Faculty of Computer Science & Information System, +<br/><b>University Technology of Malaysia, 81310 Skudai, Johor, Malaysia</b></td><td></td><td>kittmee@yahoo.com; dzulkifli@fsksm.utm.my +</td></tr><tr><td>a70e36daf934092f40a338d61e0fe27be633f577</td><td>Enhanced Facial Feature Tracking of Spontaneous and Continuous Expressions +<br/>A.Goneid and R. El Kaliouby +<br/><b>The American University in Cairo, Egypt</b></td><td></td><td>goneid@aucegypt.edu, ranak@aucegypt.edu +</td></tr><tr><td>a7664247a37a89c74d0e1a1606a99119cffc41d4</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>3287 +</td><td></td><td></td></tr><tr><td>a7191958e806fce2505a057196ccb01ea763b6ea</td><td>Convolutional Neural Network based +<br/>Age Estimation from Facial Image and +<br/>Depth Prediction from Single Image +<br/>B. Eng. (Honours) +<br/><b>Australian National University</b><br/>January 2016 +<br/>A thesis submitted for the degree of Master of Philosophy +<br/><b>at The Australian National University</b><br/>Computer Vision Group +<br/>Research School of Engineering +<br/><b>College of Engineering and Computer Science</b><br/><b>The Australian National University</b></td><td>('2124180', 'Jiayan Qiu', 'jiayan qiu')</td><td></td></tr><tr><td>a7e1327bd76945a315f2869bfae1ce55bb94d165</td><td>Kernel Fisher Discriminant Analysis with Locality Preserving for Feature Extraction and +<br/>Recognition +<br/><b>School of Information Engineering, Guangdong Medical College, Song Shan Hu</b><br/>Dongguan, Guangdong, China +<br/><b>Shaoguan University, Da Tang Lu</b><br/>Shaoguan, Guangdong, China +<br/><b>School of Information Engineering, Guangdong Medical College, Song Shan Hu</b><br/>Dongguan, Guangdong, China +</td><td>('2588058', 'Di Zhang', 'di zhang')<br/>('2007270', 'Jiazhong He', 'jiazhong he')<br/>('20374749', 'Yun Zhao', 'yun zhao')</td><td>E-mail: changnuode@163.com +<br/>E-mail: hejiazhong@126.com +<br/>E-mail: zyun@gdmc.edu.cn +</td></tr><tr><td>a7a6eb53bee5e2224f2ecd56a14e3a5a717e55b9</td><td>11th International Symposium of Robotics Research (ISRR2003), pp.192-201, 2003 +<br/>Face Recognition Using Multi-viewpoint Patterns for +<br/>Robot Vision +<br/>Corporate Research and Development Center, TOSHIBA Corporation +<br/>1, KomukaiToshiba-cho, Saiwai-ku, Kawasaki 212-8582 Japan +</td><td>('1770128', 'Kazuhiro Fukui', 'kazuhiro fukui')<br/>('1708862', 'Osamu Yamaguchi', 'osamu yamaguchi')</td><td>kazuhiro.fukui@toshiba.co.jp / osamu1.yamaguchi@toshiba.co.jp +</td></tr><tr><td>a758b744a6d6962f1ddce6f0d04292a0b5cf8e07</td><td> +<br/>ISSN XXXX XXXX © 2017 IJESC +<br/> +<br/> +<br/>Research Article Volume 7 Issue No.4 +<br/>Study on Human Face Recognition under Invariant Pose, Illumination +<br/>and Expression using LBP, LoG and SVM +<br/>Amrutha +<br/>Depart ment of Co mputer Science & Engineering +<br/><b>Mangalore Institute of Technology and Engineering, Moodabidri, Mangalore, India</b><br/>INTRODUCTION +<br/>RELATED WORK +<br/>Abstrac t: +<br/>Face recognition system uses human face for the identification of the user. Face recognition is a difficu lt task there is no unique +<br/>method that provide accurate an accurate and effic ient solution in all the situations like the face image with differen t pose , +<br/>illu mination and exp ression. Local Binary Pattern (LBP) and Laplac ian of Gaussian (Lo G) operators. Support Vector Machine +<br/>classifier is used to recognize the human face. The Lo G algorith m is used to preprocess the image to detect the edges of the face +<br/>image to get the image information. The LBP operator divides the face image into several blocks to generate the features informat ion +<br/>on pixe l level by creating LBP labels for all the blocks of image is obtained by concatenating all the individual local histo grams. +<br/>Support Vector Machine classifier (SVM ) is used to classify t he image. The a lgorith m performances is verified under the constraints +<br/>like illu mination, e xp ression and pose variation +<br/>Ke ywor ds: Face Recognition, Local Binary Pattern, Laplac ian of Gaussian, histogram, illu mination, pose angle, exp ression +<br/>variations, SVM . +<br/>1. +<br/>The Technology used for recognizing the face under security +<br/>systems works on the bio metric principles. There are many +<br/>human characteristics which can be used +<br/>for biometric +<br/>identification such that palm, finger print, face, and iris etc. one +<br/>of these biometrics methods face recognition is advantageous +<br/>because of it can be detected fro m much more d istance without +<br/>need of scanning devices this provides easy observation to +<br/>identify indiv iduals in group of persons. Most of the military +<br/>application security systems, attendance systems, authentication, +<br/>criminal identity etc. are performed using this technology. The +<br/>computer uses this recognition technology to identify or to +<br/>compare the person with same person or with some other person. +<br/>The human faces are very important factor to identify who the +<br/>person is and how the people will ma ke out his/her face. The +<br/>images of faces are taken fro m the distance without having +<br/>contact with a person, capturing the face images. Verification +<br/>and Identification s teps are used for comparison. The first +<br/>method is verification wh ich co mpares the face image with +<br/>his/her image wh ich is a lready stored in database. It is one to one +<br/>matching because it tries to match individual against same +<br/>person's image stored in database. The second method is +<br/>called one to n matching because it matches individual person's +<br/>face image with every person's face images. If the face images +<br/>are effected by lightning condition, different posing angle or +<br/>diffe rent expression then it is difficult to identify the human +<br/>face. Many algorithms are used to extract features of face and to +<br/>match the face images such as Principal Co mponent Analysis +<br/>(PCA) and Independent Component Analysis (ICA) [1], Elastic +<br/>Bunch Graph Matching (EBGM) [2], K -nearest neighbor +<br/>algorith m classifier and Linear Discriminant Analysis (LDA) +<br/>[3]. Th is paper is organized as fo llo ws: Section II revie ws the +<br/>related works done on data security in cloud. Section III +<br/>describes the proposed system and assumptions. Section IV +<br/>provides the conclusion of the paper +<br/>2. +<br/>the most biometrics +<br/>Face Recognition becomes one of +<br/>authentication +<br/>the past few years. Face +<br/>recognition is an interesting and successful application of Pattern +<br/>recognition and Image analysis. It co mpares a query face image +<br/>against all image te mplates in a face database. Face recognition +<br/>is very important due to its wide range of commercia l and law +<br/>enforcement applicat ions, which include forensic identificat ion, +<br/>access control, border surveillance and human interactions and +<br/>availability of low cost recording devices. Principa l Co mponent +<br/>Analysis and Independent Component Analysis [1], Elastic +<br/>Bunch Graph Matching [2], K-nearest neighbor algorithm +<br/>classifier and Linear Discriminant Analysis [3], Loca l Derivative +<br/>pattern and Local Binary Pattern [4]. These algorithms are still +<br/>having some proble ms +<br/>the +<br/>constraints like variations in pose, expression and illu mination. +<br/>This variation in the image degrades the performance of +<br/>recognition rate. Local Binary Pattern (LBP) and Laplac ian of +<br/>Gaussian (Lo G) is used to reduce the illu mination effects by +<br/>increasing the contrast of the image which does not effect to the +<br/>original +<br/>image and diffe rential e xc itation pixe l used for +<br/>preprocessing which is to make the algorithm invariant to the +<br/>illu mination changes +<br/>[4]. The Local Direct ional Pattern +<br/>descriptor (LDP) uses the edge values of surrounding pixe l of +<br/>the center pixe l and Two Dimensional Principal Analysis (2D- +<br/>PCA) is used for feature extraction which uses Euclidean +<br/>distance to measure the simila rity between tra ining database +<br/>images and test image features. The nearest neighbor classifier is +<br/>used to classify the images [5]. To reduce the influence of +<br/>illu mination fro m an input image an adaptive homo morphic +<br/>filtering is used in adaptive homo morphic eight local d irectional +<br/>to recognize +<br/>the face under +<br/>techniques from +<br/>International Journal of Engineering Science and Computing, April 2017 10081 http://ije sc.org/ +</td><td></td><td></td></tr><tr><td>a7c39a4e9977a85673892b714fc9441c959bf078</td><td>Automated Individualization of Deformable Eye Region Model and Its +<br/>Application to Eye Motion Analysis +<br/>Dept. of Media and Image Technology, +<br/><b>Tokyo Polytechnic University</b><br/>1583 Iiyama, Atsugi, +<br/>Kanagawa 243-0297, Japan +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>5000 Forbes Avenue Pittsburgh, +<br/>PA 15213-3891, USA +</td><td>('1683262', 'Tsuyoshi Moriyama', 'tsuyoshi moriyama')<br/>('1733113', 'Takeo Kanade', 'takeo kanade')</td><td>moriyama@mega.t-kougei.ac.jp +<br/>tk@cs.cmu.edu +</td></tr><tr><td>a75edf8124f5b52690c08ff35b0c7eb8355fe950</td><td>Authentic Emotion Detection in Real-Time Video +<br/><b>School of Computer Science and Engineering, Sichuan University, China</b><br/><b>Faculty of Science, University of Amsterdam, The Netherlands</b><br/><b>LIACS Media Lab, Leiden University, The Netherlands</b></td><td>('1840164', 'Yafei Sun', 'yafei sun')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1731570', 'Michael S. Lew', 'michael s. lew')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td></td></tr><tr><td>a75ee7f4c4130ef36d21582d5758f953dba03a01</td><td>DD2427 Final Project Report +<br/>DD2427 Final Project Report +<br/>Human face attributes prediction with Deep +<br/>Learning +</td><td></td><td>moaah@kth.se +</td></tr><tr><td>a775da3e6e6ea64bffab7f9baf665528644c7ed3</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 142 – No.9, May 2016 +<br/>Human Face Pose Estimation based on Feature +<br/>Extraction Points +<br/>Research scholar, +<br/> Department of ECE +<br/>SBSSTC, Moga Road, +<br/> Ferozepur, Punjab, India +</td><td></td><td></td></tr><tr><td>a703d51c200724517f099ee10885286ddbd8b587</td><td>Fuzzy Neural Networks(FNN)-based Approach for +<br/>Personalized Facial Expression Recognition with +<br/>Novel Feature Selection Method +<br/>Div. of EE, Dept. of EECS, KAIST +<br/>373-1 Guseong-dong, Yuseong-gu, Daejeon 305-701, Korea +<br/><b>Human-friendly Welfare Robotic System Engineering Research Center, KAIST</b><br/>373-1 Guseong-dong, Yuseong-gu, Daejeon 305-701, Korea +</td><td>('1793114', 'Dae-Jin Kim', 'dae-jin kim')<br/>('5960489', 'Kwang-Hyun Park', 'kwang-hyun park')</td><td>djkim@mail.kaist.ac.kr, zbien@ee.kaist.ac.kr +<br/>akaii@robotian.net +</td></tr><tr><td>a75dfb5a839f0eb4b613d150f54a418b7812aa90</td><td>MULTIBIOMETRIC SECURE SYSTEM BASED ON DEEP LEARNING +<br/><b>West Virginia University, Morgantown, USA</b></td><td>('23980155', 'Veeru Talreja', 'veeru talreja')<br/>('1709360', 'Matthew C. Valenti', 'matthew c. valenti')<br/>('8147588', 'Nasser M. Nasrabadi', 'nasser m. nasrabadi')</td><td></td></tr><tr><td>b88ceded6467e9b286f048bb1b17be5998a077bd</td><td>Sparse Subspace Clustering via Diffusion Process +<br/><b>Curtin University, Perth, Australia</b></td><td>('2191968', 'Qilin Li', 'qilin li')<br/>('1919769', 'Ling Li', 'ling li')<br/>('1713220', 'Wanquan Liu', 'wanquan liu')</td><td>kylinlovesummer@gmail.com +</td></tr><tr><td>b871d1b8495025ff8a6255514ed39f7765415935</td><td>Application of Completed Local Binary Pattern for Facial Expression +<br/>Recognition on Gabor Filtered Facial Images +<br/><b>University of Ulsan, Ulsan, Republic of Korea</b></td><td>('2288674', 'Tanveer Ahsan', 'tanveer ahsan')</td><td>1tanveerahsan@gmail.com, 2rsbdce@yahoo.com, *3upchong@ulsan.ac.kr +</td></tr><tr><td>b8375ff50b8a6f1a10dd809129a18df96888ac8b</td><td>Published as a conference paper at ICLR 2017 +<br/>DECOMPOSING MOTION AND CONTENT FOR +<br/>NATURAL VIDEO SEQUENCE PREDICTION +<br/><b>University of Michigan, Ann Arbor, USA</b><br/>2Adobe Research, San Jose, CA 95110 +<br/>3POSTECH, Pohang, Korea +<br/><b>Beihang University, Beijing, China</b><br/>5Google Brain, Mountain View, CA 94043 +</td><td>('2241528', 'Seunghoon Hong', 'seunghoon hong')<br/>('10668384', 'Xunyu Lin', 'xunyu lin')<br/>('1697141', 'Honglak Lee', 'honglak lee')<br/>('1768964', 'Jimei Yang', 'jimei yang')<br/>('1711926', 'Ruben Villegas', 'ruben villegas')</td><td></td></tr><tr><td>b88d5e12089f6f598b8c72ebeffefc102cad1fc0</td><td>Robust 2DPCA and Its Application +<br/><b>Xidian University</b><br/>Xi’an China +<br/><b>Xidian University</b><br/>Xi’an China +</td><td>('40326660', 'Qianqian Wang', 'qianqian wang')<br/>('38469552', 'Quanxue Gao', 'quanxue gao')</td><td>610887187@qq.com +<br/>xd ste pr@163.com +</td></tr><tr><td>b84b7b035c574727e4c30889e973423fe15560d7</td><td>Human Age Estimation Using Ranking SVM +<br/><b>HoHai University</b><br/>2Center for Biometrics and Security Research & National Laboratory of Pattern +<br/><b>Recognition, Institute of Automation, Chinese Academy of Sciences</b><br/>3China Research and Development Center for Internet of Thing +</td><td>('40478348', 'Dong Cao', 'dong cao')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1959072', 'Zhiwei Zhang', 'zhiwei zhang')<br/>('39189280', 'Jun Feng', 'jun feng')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>fdcao,zlei,zwzhang,szlig@cbsr.ia.ac.cn, fengjun@hhu.edu.cn +</td></tr><tr><td>b8dba0504d6b4b557d51a6cf4de5507141db60cf</td><td>Comparing Performances of Big Data Stream +<br/>Processing Platforms with RAM3S +</td><td></td><td></td></tr><tr><td>b89862f38fff416d2fcda389f5c59daba56241db</td><td>A Web Survey for Facial Expressions Evaluation +<br/>Ecole Polytechnique Federale de Lausanne +<br/><b>Signal Processing Institute</b><br/>Ecublens, 1015 Lausanne, Switzerland +<br/>Ecole Polytechnique Federale de Lausanne, Operation Research Group +<br/>Ecublens, 1015 Lausanne, Switzerland +<br/>June 9, 2008 +</td><td>('2916630', 'Matteo Sorci', 'matteo sorci')<br/>('1794461', 'Gianluca Antonini', 'gianluca antonini')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')<br/>('1690395', 'Michel Bierlaire', 'michel bierlaire')</td><td>{Matteo.Sorci,Gianluca.Antonini,JP.Thiran}@epfl.ch +<br/>Michel.Bierlaire@epfl.ch +</td></tr><tr><td>b8caf1b1bc3d7a26a91574b493c502d2128791f6</td><td>RESEARCH ARTICLE +<br/>As Far as the Eye Can See: Relationship +<br/>between Psychopathic Traits and Pupil +<br/>Response to Affective Stimuli +<br/>Daniel T. Burley1*, Nicola S. Gray2,3, Robert J. Snowden1* +<br/><b>School of Psychology, Cardiff University, Cardiff, United Kingdom, College of</b><br/><b>Human and Health Sciences, Swansea University, Swansea, United Kingdom, 3 Abertawe Bro-Morgannwg</b><br/><b>University Health Board, Swansea, United Kingdom</b></td><td></td><td>* BurleyD2@Cardiff.ac.uk (DTB); Snowden@Cardiff.ac.uk (RJS) +</td></tr><tr><td>b8084d5e193633462e56f897f3d81b2832b72dff</td><td>DeepID3: Face Recognition with Very Deep Neural Networks +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b><br/>2SenseTime Group +</td><td>('1681656', 'Yi Sun', 'yi sun')<br/>('1865674', 'Ding Liang', 'ding liang')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>sy011@ie.cuhk.edu.hk +<br/>xgwang@ee.cuhk.edu.hk +<br/>liangding@sensetime.com +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>b8378ab83bc165bc0e3692f2ce593dcc713df34a</td><td></td><td></td><td></td></tr><tr><td>b8f3f6d8f188f65ca8ea2725b248397c7d1e662d</td><td>Selfie Detection by Synergy-Constriant Based +<br/>Convolutional Neural Network +<br/>Electrical and Electronics Engineering, NITK-Surathkal, India. +</td><td>('7245071', 'Yashas Annadani', 'yashas annadani')<br/>('8341302', 'Akshay Kumar Jagadish', 'akshay kumar jagadish')<br/>('2139966', 'Krishnan Chemmangat', 'krishnan chemmangat')</td><td></td></tr><tr><td>b8ebda42e272d3617375118542d4675a0c0e501d</td><td>Deep Hashing Network for Unsupervised Domain Adaptation +<br/><b>Center for Cognitive Ubiquitous Computing, Arizona State University, Tempe, AZ, USA</b></td><td>('3151995', 'Hemanth Venkateswara', 'hemanth venkateswara')<br/>('30443430', 'Jose Eusebio', 'jose eusebio')<br/>('2471253', 'Shayok Chakraborty', 'shayok chakraborty')<br/>('1743991', 'Sethuraman Panchanathan', 'sethuraman panchanathan')</td><td>{hemanthv, jeusebio, shayok.chakraborty, panch}@asu.edu +</td></tr><tr><td>b85580ff2d8d8be0a2c40863f04269df4cd766d9</td><td>HCMUS team at the Multimodal Person Discovery in +<br/>Broadcast TV Task of MediaEval 2016 +<br/>Faculty of Information Technology +<br/><b>University of Science, Vietnam National University-Ho Chi Minh city</b></td><td>('34453615', 'Vinh-Tiep Nguyen', 'vinh-tiep nguyen')<br/>('30097677', 'Manh-Tien H. Nguyen', 'manh-tien h. nguyen')<br/>('8176737', 'Quoc-Huu Che', 'quoc-huu che')<br/>('7736164', 'Van-Tu Ninh', 'van-tu ninh')<br/>('38994364', 'Tu-Khiem Le', 'tu-khiem le')<br/>('7213584', 'Thanh-An Nguyen', 'thanh-an nguyen')<br/>('1780348', 'Minh-Triet Tran', 'minh-triet tran')</td><td>nvtiep@fit.hcmus.edu.vn, {nhmtien, cqhuu, nvtu, ltkhiem}@apcs.vn, +<br/>1312016@student.hcmus.edu.vn, tmtriet@fit.hcmus.edu.vn +</td></tr><tr><td>b87b0fa1ac0aad0ca563844daecaeecb2df8debf</td><td>Computational Aesthetics in Graphics, Visualization, and Imaging +<br/>EXPRESSIVE 2015 +<br/>Non-Photorealistic Rendering of Portraits +<br/><b>Cardiff University, UK</b><br/>Figure 1: President Obama re-rendered in “puppet” style and in the style of Julian Opie. +</td><td>('1734823', 'Paul L. Rosin', 'paul l. rosin')<br/>('1734823', 'Paul L. Rosin', 'paul l. rosin')<br/>('7827503', 'Yu-Kun Lai', 'yu-kun lai')</td><td></td></tr><tr><td>b87db5ac17312db60e26394f9e3e1a51647cca66</td><td>Semi-definite Manifold Alignment +<br/><b>Tsinghua University</b><br/>Beijing, China +</td><td>('2066355', 'Liang Xiong', 'liang xiong')<br/>('34410258', 'Fei Wang', 'fei wang')<br/>('1700883', 'Changshui Zhang', 'changshui zhang')</td><td>{xiongl,feiwang03}@mails.tsinghua.edu.cn, zcs@mail.tsinghua.edu.cn +</td></tr><tr><td>b81cae2927598253da37954fb36a2549c5405cdb</td><td>Experiments on Visual Information Extraction with the Faces of Wikipedia +<br/>D´epartement de g´enie informatique et g´enie logiciel, Polytechnique Montr´eal +<br/>2500, Chemin de Polytechnique, Universit´e de Montr´eal, Montr`eal, Qu´ebec, Canada +</td><td>('2811524', 'Md. Kamrul Hasan', 'md. kamrul hasan')</td><td></td></tr><tr><td>b8a829b30381106b806066d40dd372045d49178d</td><td>1872 +<br/>A Probabilistic Framework for Joint Pedestrian Head +<br/>and Body Orientation Estimation +</td><td>('2869660', 'Fabian Flohr', 'fabian flohr')<br/>('1898318', 'Madalin Dumitru-Guzu', 'madalin dumitru-guzu')<br/>('34846285', 'Julian F. P. Kooij', 'julian f. p. kooij')</td><td></td></tr><tr><td>b1d89015f9b16515735d4140c84b0bacbbef19ac</td><td>Too Far to See? Not Really! +<br/>— Pedestrian Detection with Scale-aware +<br/>Localization Policy +</td><td>('47957574', 'Xiaowei Zhang', 'xiaowei zhang')<br/>('50791064', 'Li Cheng', 'li cheng')<br/>('49729740', 'Bo Li', 'bo li')<br/>('2938403', 'Hai-Miao Hu', 'hai-miao hu')</td><td></td></tr><tr><td>b191aa2c5b8ece06c221c3a4a0914e8157a16129</td><td>: DEEP SPATIO-TEMPORAL MANIFOLD NETWORK FOR ACTION RECOGNITION +<br/>Deep Spatio-temporal Manifold Network for +<br/>Action Recognition +<br/>Department of Computer Science +<br/><b>China University of Mining and Technol</b><br/>ogy, Beijing, China +<br/>Center for Research in Computer +<br/>Vision (CRCV) +<br/><b>University of Central Florida, Orlando</b><br/>FL, USA +<br/>School of Automation Science and +<br/>electrical engineering +<br/><b>Beihang University, Beijing, China</b><br/><b>University of Chinese Academy of</b><br/>Sciences +<br/>Beijing, China +<br/>Nortumbria Univesity +<br/>Newcastle, UK +<br/>Xiamen Univesity +<br/>Xiamen, China +</td><td>('2606761', 'Ce Li', 'ce li')<br/>('9497155', 'Chen Chen', 'chen chen')<br/>('1740430', 'Baochang Zhang', 'baochang zhang')<br/>('1694936', 'Qixiang Ye', 'qixiang ye')<br/>('1783847', 'Jungong Han', 'jungong han')<br/>('1725599', 'Rongrong Ji', 'rongrong ji')</td><td>celi@cumtb.edu.cn +<br/>chenchen870713@gmail.com +<br/>bczhang@139.com +<br/>qxye@ucas.ac.cn +<br/>jungonghan77@gmail.com +<br/>rrji@xmu.edu.cn +</td></tr><tr><td>b13bf657ca6d34d0df90e7ae739c94a7efc30dc3</td><td>Attribute and Simile Classifiers for Face Verification (In submission please do +<br/>not distribute.) +<br/><b>Columbia University</b><br/>New York, NY +<br/><b>Columbia University</b><br/>New York, NY +<br/><b>Columbia University</b><br/><b>Columbia University</b><br/>New York, NY +</td><td>('3586464', 'Neeraj Kumar', 'neeraj kumar')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')<br/>('1750470', 'Shree K. Nayar', 'shree k. nayar')</td><td>belhumeur@cs.columbia.edu +<br/>neeraj@cs.columbia.edu +<br/>aberg@cs.columbia.edu +<br/>nayar@cs.columbia.edu +</td></tr><tr><td>b13a882e6168afc4058fe14cc075c7e41434f43e</td><td>Recognition of Humans and Their Activities Using Video +<br/>Center for Automation Research +<br/><b>University of Maryland</b><br/><b>College Park, MD</b><br/>Dept. of Electrical Engineering +<br/><b>University of California</b><br/>Riverside, CA 92521 +<br/>Shaohua K. Zhou +<br/>Siemens Research +<br/>Princeton, NJ 08540 +</td><td>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('1688416', 'Amit K. Roy-Chowdhury', 'amit k. roy-chowdhury')</td><td></td></tr><tr><td>b14b672e09b5b2d984295dfafb05604492bfaec5</td><td>LearningImageClassificationandRetrievalModelsThomasMensink </td><td></td><td></td></tr><tr><td>b1665e1ddf9253dcaebecb48ac09a7ab4095a83e</td><td>EMOTION RECOGNITION USING FACIAL EXPRESSIONS WITH ACTIVE +<br/>APPEARANCE MODELS +<br/>Department of Computer Science +<br/><b>University of North Carolina Wilmington</b><br/><b>South College Road</b><br/>Wilmington, NC, USA +<br/>Department of Computer Science +<br/><b>University of North Carolina Wilmington</b><br/><b>South College Road</b><br/>Wilmington, NC, USA +</td><td>('12675740', 'Matthew S. Ratliff', 'matthew s. ratliff')<br/>('37804931', 'Eric Patterson', 'eric patterson')</td><td>msr3520@uncw.edu +<br/>pattersone@uncw.edu +</td></tr><tr><td>b16580d27bbf4e17053f2f91bc1d0be12045e00b</td><td>Pose-invariant Face Recognition with a +<br/>Two-Level Dynamic Programming Algorithm +<br/>1 Human Language Technology and Pattern Recognition Group +<br/><b>RWTH Aachen University, Aachen, Germany</b><br/>2 Robert Bosch GmbH, Hildesheim, Germany +</td><td>('1804963', 'Harald Hanselmann', 'harald hanselmann')<br/>('1685956', 'Hermann Ney', 'hermann ney')<br/>('1967060', 'Philippe Dreuw', 'philippe dreuw')</td><td><surname>@cs.rwth-aachen.de +<br/>philippe.dreuw@de.bosch.com +</td></tr><tr><td>b1b993a1fbcc827bcb99c4cc1ba64ae2c5dcc000</td><td>Deep Variation-structured Reinforcement Learning for Visual Relationship and +<br/>Attribute Detection +<br/><b>School of Computer Science, Carnegie Mellon University</b></td><td>('40250403', 'Xiaodan Liang', 'xiaodan liang')<br/>('1752601', 'Eric P. Xing', 'eric p. xing')<br/>('49441821', 'Lisa Lee', 'lisa lee')</td><td>{xiaodan1,lslee,epxing}@cs.cmu.edu +</td></tr><tr><td>b11bb6bd63ee6f246d278dd4edccfbe470263803</td><td>Joint Voxel and Coordinate Regression for Accurate +<br/>3D Facial Landmark Localization +<br/>†Center for Research on Intelligent Perception and Computing (CRIPAC) +<br/><b>Institute of Automation, Chinese Academy of Sciences (CASIA</b><br/>†National Laboratory of Pattern Recognition (NLPR) +<br/><b>University of Chinese Academy of Sciences (UCAS</b><br/>§Center for Excellence in Brain Science and Intelligence Technology (CEBSIT), CAS +</td><td>('37536613', 'Hongwen Zhang', 'hongwen zhang')<br/>('39763795', 'Qi Li', 'qi li')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>Email: hongwen.zhang@cripac.ia.ac.cn, {qli, znsun}@nlpr.ia.ac.cn +</td></tr><tr><td>b171f9e4245b52ff96790cf4f8d23e822c260780</td><td></td><td></td><td></td></tr><tr><td>b1a3b19700b8738b4510eecf78a35ff38406df22</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TAFFC.2017.2731763, IEEE +<br/>Transactions on Affective Computing +<br/>JOURNAL OF LATEX CLASS FILES, VOL. 13, NO. 9, SEPTEMBER 2014 +<br/>Automatic Analysis of Facial Actions: A Survey +<br/>and Maja Pantic, Fellow, IEEE +</td><td>('1680608', 'Brais Martinez', 'brais martinez')<br/>('1795528', 'Michel F. Valstar', 'michel f. valstar')<br/>('39532631', 'Bihan Jiang', 'bihan jiang')</td><td></td></tr><tr><td>b166ce267ddb705e6ed855c6b679ec699d62e9cb</td><td>Turk J Elec Eng & Comp Sci +<br/>(2017) 25: 4421 { 4430 +<br/>c⃝ T (cid:127)UB_ITAK +<br/>doi:10.3906/elk-1702-49 +<br/>Sample group and misplaced atom dictionary learning for face recognition +<br/><b>Faculty of Electronics and Communication, Yanshan University</b><br/><b>Faculty of Electronics and Communication, Taishan University</b><br/>Qinhuangdao, P.R. China +<br/>Tai’an, P.R. China +<br/>Received: 04.02.2017 +<br/>(cid:15) +<br/>Accepted/Published Online: 01.06.2017 +<br/>(cid:15) +<br/>Final Version: 05.10.2017 +</td><td>('39980529', 'Meng Wang', 'meng wang')<br/>('49576759', 'Zhe Sun', 'zhe sun')<br/>('6410069', 'Mei Zhu', 'mei zhu')<br/>('49632877', 'Mei Sun', 'mei sun')</td><td></td></tr><tr><td>b13e2e43672e66ba45d1b852a34737e4ce04226b</td><td>CROWLEY, PARKHI, ZISSERMAN: FACE PAINTING +<br/>Face Painting: querying art with photos +<br/>Elliot J. Crowley +<br/>Visual Geometry Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b></td><td>('3188342', 'Omkar M. Parkhi', 'omkar m. parkhi')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>elliot@robots.ox.ac.uk +<br/>omkar@robots.ox.ac.uk +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>b1e4f8c15ff30cc7d35ab25ff3eddaf854e0a87c</td><td>RESEARCH ARTICLE +<br/>Conveying facial expressions to blind and +<br/>visually impaired persons through a wearable +<br/>vibrotactile device +<br/><b>MIRA Institute, University of Twente, Enschede, The</b><br/><b>Netherlands, Donders Institute, Radboud University, Nijmegen, The</b><br/>Netherlands, 3 VicarVision, Amsterdam, The Netherlands, 4 Department of Media, Communication, & +<br/><b>Organization, University of Twente, Enschede, The Netherlands, HAN</b><br/><b>University of Applied Sciences, Arnhem, The Netherlands</b></td><td>('1950480', 'Hendrik P. Buimer', 'hendrik p. buimer')<br/>('25188062', 'Marian Bittner', 'marian bittner')<br/>('3427220', 'Tjerk Kostelijk', 'tjerk kostelijk')<br/>('49432294', 'Abdellatif Nemri', 'abdellatif nemri')<br/>('2968885', 'Richard J. A. van Wezel', 'richard j. a. van wezel')</td><td>* h.buimer@donders.ru.nl +</td></tr><tr><td>b1301c722886b6028d11e4c2084ee96466218be4</td><td></td><td></td><td></td></tr><tr><td>b15a06d701f0a7f508e3355a09d0016de3d92a6d</td><td>Running head: FACIAL CONTRAST LOOKS HEALTHY +<br/>1 +<br/>Facial contrast is a cue for perceiving health from the face +<br/>Mauger2, Frederique Morizot2 +<br/><b>Gettysburg College, Gettysburg, PA, USA</b><br/>2 CHANEL Recherche et Technologie, Chanel PB +<br/>3 Université Grenoble Alpes +<br/>Author Note +<br/>Psychologie et NeuroCognition, Université Grenoble Alpes. +<br/>This is a prepublication copy. This article may not exactly replicate the authoritative document +<br/>published in the APA journal. It is not the copy of record. The authoritative document can be +<br/>found through this DOI: http://psycnet.apa.org/doi/10.1037/xhp0000219 +</td><td>('40482411', 'Richard Russell', 'richard russell')<br/>('4556101', 'Aurélie Porcheron', 'aurélie porcheron')<br/>('40482411', 'Richard Russell', 'richard russell')<br/>('4556101', 'Aurélie Porcheron', 'aurélie porcheron')<br/>('6258499', 'Emmanuelle Mauger', 'emmanuelle mauger')<br/>('4556101', 'Aurélie Porcheron', 'aurélie porcheron')<br/>('40482411', 'Richard Russell', 'richard russell')</td><td>College, Gettysburg, PA 17325, USA. Email: rrussell@gettysburg.edu +</td></tr><tr><td>b1c5581f631dba78927aae4f86a839f43646220c</td><td></td><td></td><td></td></tr><tr><td>b18858ad6ec88d8b443dffd3e944e653178bc28b</td><td><b>Purdue University</b><br/>Purdue e-Pubs +<br/>Department of Computer Science Technical +<br/>Reports +<br/>Department of Computer Science +<br/>2017 +<br/>Trojaning Attack on Neural Networks +<br/>See next page for additional authors +<br/>Report Number: +<br/>17-002 +<br/>Liu, Yingqi; Ma, Shiqing; Aafer, Yousra; Lee, Wen-Chuan; Zhai, Juan; Wang, Weihang; and Zhang, Xiangyu, "Trojaning Attack on +<br/>Neural Networks" (2017). Department of Computer Science Technical Reports. Paper 1781. +<br/>https://docs.lib.purdue.edu/cstech/1781 +<br/>additional information. +</td><td>('3347155', 'Yingqi Liu', 'yingqi liu')<br/>('40306181', 'Shiqing Ma', 'shiqing ma')<br/>('3216258', 'Yousra Aafer', 'yousra aafer')<br/>('2547748', 'Wen-Chuan Lee', 'wen-chuan lee')<br/>('3293342', 'Juan Zhai', 'juan zhai')</td><td>Purdue University, liu1751@purdue.edu +<br/>Purdue University, ma229@purdue.edu +<br/>Purdue University, yaafer@purdue.edu +<br/>Purdue University, lee1938@purdue.edu +<br/>Nanjing University, China, zhaijuan@nju.edu.cn +<br/>This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact epubs@purdue.edu for +</td></tr><tr><td>b1444b3bf15eec84f6d9a2ade7989bb980ea7bd1</td><td>LOCAL DIRECTIONAL RELATION PATTERN +<br/>Local Directional Relation Pattern for +<br/>Unconstrained and Robust Face Retrieval +</td><td>('34992579', 'Shiv Ram Dubey', 'shiv ram dubey')</td><td></td></tr><tr><td>b133b2d7df9b848253b9d75e2ca5c68e21eba008</td><td><b>Kobe University, NICT and University of Siegen</b><br/>at TRECVID 2017 AVS Task +<br/><b>Graduate School of System Informatics, Kobe University</b><br/><b>Center for Information and Neural Networks, National Institute of Information and Communications Technology (NICT</b><br/><b>Pattern Recognition Group, University of Siegen</b></td><td>('2240008', 'Zhenying He', 'zhenying he')<br/>('8183718', 'Takashi Shinozaki', 'takashi shinozaki')<br/>('1707938', 'Kimiaki Shirahama', 'kimiaki shirahama')<br/>('1727057', 'Marcin Grzegorzek', 'marcin grzegorzek')<br/>('1711781', 'Kuniaki Uehara', 'kuniaki uehara')</td><td>jennyhe@ai.cs.kobe-u.ac.jp, uehara@kobe-u.ac.jp +<br/>tshino@nict.go.jp +<br/>kimiaki.shirahama@uni-siegen.de, marcin.grzegorzek@uni-siegen.de +</td></tr><tr><td>b1451721864e836069fa299a64595d1655793757</td><td>Criteria Sliders: Learning Continuous +<br/>Database Criteria via Interactive Ranking +<br/><b>Brown University 2University of Bath</b><br/><b>Harvard University 4Max Planck Institute for Informatics</b></td><td>('1854493', 'James Tompkin', 'james tompkin')<br/>('1808255', 'Kwang In Kim', 'kwang in kim')<br/>('1680185', 'Christian Theobalt', 'christian theobalt')</td><td></td></tr><tr><td>b1df214e0f1c5065f53054195cd15012e660490a</td><td>Supplementary Material to Sparse Coding and Dictionary Learning with Linear +<br/>Dynamical Systems∗ +<br/><b>Tsinghua University, State Key Lab. of Intelligent</b><br/>Technology and Systems, Tsinghua National Lab. for Information Science and Technology (TNList); +<br/><b>Australian National University and NICTA, Australia</b><br/>In this supplementary material, we present the proofs of Theorems (1-3), the algorithm for learning the transition matrix +<br/>of LDSST, and the reconstruction error approach for classification in LDS-SC, LDSST-SC and covLDSST-SC. In addition, +<br/>we describe the details of the benchmark datasets that are applied in our experiments. Our dictionary learning algorithm for +<br/>anormaly detection is also explored in this supplementary material. +<br/>1. Proofs +<br/>Theorem 1. Suppose V1, V2, · · · , VM ∈ S(n, ∞), and y1, y2, · · · , yM ∈ R, we have +<br/>Xi=1 +<br/>yiΠ(Vi) k2 +<br/>F = +<br/>Xi,j=1 +<br/>yiyj k VT +<br/>i Vj k2 +<br/>F , +<br/>i Oj can be computed with the Lyapunov equation defined in Equation (2), Li and Lj +<br/>i Vj = L−1 +<br/>where VT +<br/>are Cholesky decomposition matrices for OT +<br/>i Oj L−T +<br/>i OT +<br/>. OT +<br/>i Oi and OT +<br/>j Oj , respectively. +<br/>i)]T by +<br/>Proof. We denote the sub-matrix of the extended observability matrix Oi as Oi(t) = [CT +<br/>taking the first t rows. We suppose that the Cholesky decomposition matrix for Oi is Li and denote that Vi(t) = Oi(t)L−T +<br/>Then, we derive +<br/>i , (CiAi)T, · · · , (CiAt +<br/>Xi=1 +<br/>yiΠ(Vi) k2 +<br/>F = lim +<br/>t→∞ +<br/>= lim +<br/>t→∞ +<br/>= lim +<br/>t→∞ +<br/>yiVi(t)Vi(t)T +<br/>yiVi(t)Vi(t)T k2 +<br/>Xi=1 +<br/>Tr +<br/>Xi=1 +<br/>Xi,j=1 +<br/>yiyjTr(cid:0)Vi(t)TVj(t)Vj(t)TVi(t)(cid:1) +<br/>yj Vj(t)Vj(t)T +<br/>Xj=1 +<br/>yiyj lim +<br/>t→∞ +<br/>k Vi(t)TVj(t) k2 +<br/>yiyj lim +<br/>t→∞ +<br/>k L−1 +<br/>(Oi(t)TOj(t))L−T +<br/>k2 +<br/>yiyj k L−1 +<br/>i Oij L−T +<br/>k2 +<br/>F , +<br/>(13) +<br/>Xi,j=1 +<br/>Xi,j=1 +<br/>Xi,j=1 +<br/>∗This work is jointly supported by National Natural Science Foundation of China under Grant No. 61327809, 61210013, 91420302 and 91520201. +</td><td>('8984539', 'Wenbing Huang', 'wenbing huang')<br/>('40203750', 'Fuchun Sun', 'fuchun sun')<br/>('2507718', 'Lele Cao', 'lele cao')<br/>('1678783', 'Deli Zhao', 'deli zhao')<br/>('31833173', 'Huaping Liu', 'huaping liu')<br/>('23911916', 'Mehrtash Harandi', 'mehrtash harandi')</td><td>1{huangwb12@mails, fcsun@mail, caoll12@mails, hpliu@mail}.tsinghua.edu.cn, +<br/>zhaodeli@gmail.com, +<br/>Mehrtash.Harandi@nicta.com.au, +</td></tr><tr><td>b185f0a39384ceb3c4923196aeed6d68830a069f</td><td>Describing Clothing by Semantic Attributes +<br/><b>Stanford University, Stanford, California</b><br/><b>Kodak Research Laboratories, Rochester, New York</b><br/><b>Cornell University, Ithaca, New York</b></td><td>('2896700', 'Huizhong Chen', 'huizhong chen')<br/>('1739786', 'Bernd Girod', 'bernd girod')</td><td></td></tr><tr><td>b19e83eda4a602abc5a8ef57467c5f47f493848d</td><td>JOURNAL OF LATEX CLASS FILES +<br/>Heat Kernel Based Local Binary Pattern for +<br/>Face Representation +</td><td>('38979129', 'Xi Li', 'xi li')<br/>('40506509', 'Weiming Hu', 'weiming hu')<br/>('1720488', 'Zhongfei Zhang', 'zhongfei zhang')<br/>('37414077', 'Hanzi Wang', 'hanzi wang')</td><td></td></tr><tr><td>b1429e4d3dd3412e92a37d2f9e0721ea719a9b9e</td><td>Person Re-identification Using Multiple First-Person-Views on Wearable Devices +<br/><b>Nanyang Technological University, Singapore</b><br/><b>Institute for Infocomm Research (I2R), A*STAR, Singapore</b><br/>Istituto Italiano di Tecnologia (IIT), Genova, 16163, Italy +</td><td>('37287044', 'Anirban Chakraborty', 'anirban chakraborty')<br/>('1709001', 'Bappaditya Mandal', 'bappaditya mandal')<br/>('2860592', 'Hamed Kiani Galoogahi', 'hamed kiani galoogahi')</td><td>a.chakraborty@ntu.edu.sg +<br/>bmandal@i2r.a-star.edu.sg +<br/>kiani.galoogahi@iit.it +</td></tr><tr><td>b1fdd4ae17d82612cefd4e78b690847b071379d3</td><td>Supervised Descent Method +<br/>CMU-RI-TR-15-28 +<br/>September 2015 +<br/><b>The Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>Thesis Committee: +<br/>Fernando De la Torre, Chair +<br/>Srinivasa Narasimhan +<br/>Kris Kitani +<br/>Aleix Martinez +<br/>Submitted in partial fulfillment of the requirements +<br/>for the degree of Doctor of Philosophy in Robotics. +</td><td>('3182065', 'Xuehan Xiong', 'xuehan xiong')<br/>('3182065', 'Xuehan Xiong', 'xuehan xiong')</td><td></td></tr><tr><td>dde5125baefa1141f1ed50479a3fd67c528a965f</td><td>Synthesizing Normalized Faces from Facial Identity Features +<br/><b>Google, Inc. 2University of Massachusetts Amherst 3MIT CSAIL</b></td><td>('39578349', 'Forrester Cole', 'forrester cole')<br/>('1707347', 'Dilip Krishnan', 'dilip krishnan')</td><td>{fcole, dbelanger, dilipkay, sarna, inbarm, wfreeman}@google.com +</td></tr><tr><td>dd8084b2878ca95d8f14bae73e1072922f0cc5da</td><td>Model Distillation with Knowledge Transfer from +<br/>Face Classification to Alignment and Verification +<br/>Beijing Orion Star Technology Co., Ltd. Beijing, China +</td><td>('1747751', 'Chong Wang', 'chong wang')<br/>('26403761', 'Xipeng Lan', 'xipeng lan')</td><td>{chongwang.nlpr, xipeng.lan, caveman1984}@gmail.com +</td></tr><tr><td>ddf55fc9cf57dabf4eccbf9daab52108df5b69aa</td><td>International Journal of Grid and Distributed Computing +<br/>Vol. 4, No. 3, September, 2011 +<br/>Methodology and Performance Analysis of 3-D Facial Expression +<br/>Recognition Using Statistical Shape Representation +<br/><b>ADSIP Research Centre, University of Central Lancashire</b><br/><b>School of Psychology, University of Central Lancashire</b></td><td>('2343120', 'Wei Quan', 'wei quan')<br/>('2647218', 'Bogdan J. Matuszewski', 'bogdan j. matuszewski')<br/>('2550166', 'Lik-Kwan Shark', 'lik-kwan shark')<br/>('2942330', 'Charlie Frowd', 'charlie frowd')</td><td>{WQuan, BMatuszewski1, LShark}@uclan.ac.uk +<br/>CFrowd@uclan.ac.uk +</td></tr><tr><td>dd85b6fdc45bf61f2b3d3d92ce5056c47bd8d335</td><td>Unsupervised Learning and Segmentation of Complex Activities from Video +<br/><b>University of Bonn, Germany</b></td><td>('34678431', 'Fadime Sener', 'fadime sener')<br/>('2569989', 'Angela Yao', 'angela yao')</td><td>{sener,yao}@cs.uni-bonn.de +</td></tr><tr><td>dda35768681f74dafd02a667dac2e6101926a279</td><td>MULTI-LAYER TEMPORAL GRAPHICAL MODEL +<br/>FOR HEAD POSE ESTIMATION IN REAL-WORLD VIDEOS +<br/><b>McGill University</b><br/>Centre for Intelligent Machines, +</td><td>('2515930', 'Meltem Demirkus', 'meltem demirkus')<br/>('1724729', 'Doina Precup', 'doina precup')<br/>('1713608', 'James J. Clark', 'james j. clark')<br/>('1699104', 'Tal Arbel', 'tal arbel')</td><td></td></tr><tr><td>dd0760bda44d4e222c0a54d41681f97b3270122b</td><td></td><td></td><td></td></tr><tr><td>ddea3c352f5041fb34433b635399711a90fde0e8</td><td>Facial Expression Classification using Visual Cues and Language +<br/>Department of Computer Science and Engineering, IIT Kanpur +</td><td>('2094658', 'Abhishek Kar', 'abhishek kar')<br/>('1803835', 'Amitabha Mukerjee', 'amitabha mukerjee')</td><td>{akar,amit}@iitk.ac.in +</td></tr><tr><td>dd033d4886f2e687b82d893a2c14dae02962ea70</td><td>Electronic Letters on Computer Vision and Image Analysis 11(1):41-54; 2012 +<br/>Facial Expression Recognition Using New Feature Extraction +<br/>Algorithm +<br/><b>National Cheng Kung University, Tainan, Taiwan</b><br/> Received 10th Oct. 2011; accepted 5th Sep. 2012 +</td><td>('2499819', 'Hung-Fu Huang', 'hung-fu huang')<br/>('1751725', 'Shen-Chuan Tai', 'shen-chuan tai')</td><td></td></tr><tr><td>ddbd24a73ba3d74028596f393bb07a6b87a469c0</td><td>Multi-region two-stream R-CNN +<br/>for action detection +<br/>Inria(cid:63) +</td><td>('1766837', 'Xiaojiang Peng', 'xiaojiang peng')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td>{xiaojiang.peng,cordelia.schmid}@inria.fr +</td></tr><tr><td>ddf099f0e0631da4a6396a17829160301796151c</td><td>IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY +<br/>Learning Face Image Quality from +<br/>Human Assessments +</td><td>('2180413', 'Lacey Best-Rowden', 'lacey best-rowden')<br/>('40217643', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>dd0a334b767e0065c730873a95312a89ef7d1c03</td><td>Eigenexpressions: Emotion Recognition using Multiple +<br/>Eigenspaces +<br/>Luis Marco-Gim´enez1, Miguel Arevalillo-Herr´aez1, and Cristina Cuhna-P´erez2 +<br/><b></b><br/>Burjassot. Valencia 46100, Spain, +<br/>2 Universidad Cat´olica San Vicente M´artir de Valencia (UCV), +<br/>Burjassot. Valencia. Spain +</td><td></td><td>margi4@alumni.uv.es +</td></tr><tr><td>dd2f6a1ba3650075245a422319d86002e1e87808</td><td></td><td></td><td></td></tr><tr><td>ddaa8add8528857712424fd57179e5db6885df7c</td><td>METTES, SNOEK, CHANG: ACTION LOCALIZATION WITH PSEUDO-ANNOTATIONS +<br/>Localizing Actions from Video Labels +<br/>and Pseudo-Annotations +<br/>Cees G.M. Snoek1 +<br/><b>University of Amsterdam</b><br/>Amsterdam, NL +<br/><b>Columbia University</b><br/>New York, USA +</td><td>('2606260', 'Pascal Mettes', 'pascal mettes')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td></td></tr><tr><td>dd8d53e67668067fd290eb500d7dfab5b6f730dd</td><td>69 +<br/>A Parameter-Free Framework for General +<br/>Supervised Subspace Learning +</td><td>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('7137861', 'Jianzhuang Liu', 'jianzhuang liu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td></td></tr><tr><td>ddbb6e0913ac127004be73e2d4097513a8f02d37</td><td>264 +<br/>IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 1, NO. 3, SEPTEMBER 1999 +<br/>Face Detection Using Quantized Skin Color +<br/>Regions Merging and Wavelet Packet Analysis +</td><td>('34798028', 'Christophe Garcia', 'christophe garcia')<br/>('2441655', 'Georgios Tziritas', 'georgios tziritas')</td><td></td></tr><tr><td>dd600e7d6e4443ebe87ab864d62e2f4316431293</td><td></td><td></td><td></td></tr><tr><td>dc550f361ae82ec6e1a0cf67edf6a0138163382e</td><td> +<br/>ISSN XXXX XXXX © 2018 IJESC +<br/> +<br/> +<br/>Research Article Volume 8 Issue No.3 +<br/>Emotion Based Music Player +<br/>Professor1, UG Student2, 3, 4, 5, 6 +<br/>Department of Electronics Engineering +<br/><b>K.D.K. College of Engineering Nagpur, India</b></td><td>('9217928', 'Vijay Chakole', 'vijay chakole')<br/>('48228560', 'Kalyani Trivedi', 'kalyani trivedi')</td><td></td></tr><tr><td>dcf71245addaf66a868221041aabe23c0a074312</td><td>S3FD: Single Shot Scale-invariant Face Detector +<br/><b>CBSR and NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, China</b></td><td>('3220556', 'Shifeng Zhang', 'shifeng zhang')</td><td>{shifeng.zhang,xiangyu.zhu,zlei,hailin.shi,xiaobo.wang,szli}@nlpr.ia.ac.cn +</td></tr><tr><td>dcb44fc19c1949b1eda9abe998935d567498467d</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>1916 +</td><td></td><td></td></tr><tr><td>dcc38db6c885444694f515d683bbb50521ff3990</td><td>Learning to hallucinate face images via Component Generation and Enhancement +<br/><b>City University of Hong Kong</b><br/><b>South China University of Technology</b><br/>3Tencent AI Lab +<br/><b>University of Science and Technology of China</b></td><td>('2255687', 'Yibing Song', 'yibing song')<br/>('1718428', 'Jiawei Zhang', 'jiawei zhang')<br/>('2548483', 'Shengfeng He', 'shengfeng he')<br/>('2780029', 'Linchao Bao', 'linchao bao')<br/>('1777434', 'Qingxiong Yang', 'qingxiong yang')</td><td></td></tr><tr><td>dc5cde7e4554db012d39fc41ac8580f4f6774045</td><td>FAKTOR, IRANI: VIDEO SEGMENTATION BY NON-LOCAL CONSENSUS VOTING +<br/>Video Segmentation by Non-Local +<br/>Consensus Voting +<br/>http://www.wisdom.weizmann.ac.il/~alonf/ +<br/>http://www.wisdom.weizmann.ac.il/~irani/ +<br/>Dept. of Computer Science and +<br/>Applied Math +<br/><b>The Weizmann Institute of Science</b><br/>ISRAEL +</td><td>('2859022', 'Alon Faktor', 'alon faktor')<br/>('1696887', 'Michal Irani', 'michal irani')</td><td></td></tr><tr><td>dc7df544d7c186723d754e2e7b7217d38a12fcf7</td><td>Facial expression recognition using salient facial patches +<br/>MIRACL-ENET’COM +<br/><b>University of Sfax</b><br/>Tunisia (3018), Sfax +<br/>MIRACL-FSS +<br/><b>University of Sfax</b><br/>Tunisia (3018), Sfax +</td><td>('2049116', 'Hazar Mliki', 'hazar mliki')<br/>('1749733', 'Mohamed Hammami', 'mohamed hammami')</td><td>mliki.hazar@gmail.com +<br/>mohamed.hammami@fss.rnu.tn +</td></tr><tr><td>dc77287bb1fcf64358767dc5b5a8a79ed9abaa53</td><td>Fashion Conversation Data on Instagram +<br/>∗Graduate School of Culture Technology, KAIST, South Korea +<br/>†Department of Communication Studies, UCLA, USA +</td><td>('3459091', 'Yu-i Ha', 'yu-i ha')<br/>('2399803', 'Sejeong Kwon', 'sejeong kwon')<br/>('1775511', 'Meeyoung Cha', 'meeyoung cha')<br/>('1834047', 'Jungseock Joo', 'jungseock joo')</td><td></td></tr><tr><td>dc2e805d0038f9d1b3d1bc79192f1d90f6091ecb</td><td></td><td></td><td></td></tr><tr><td>dced05d28f353be971ea2c14517e85bc457405f3</td><td>Multimodal Priority Verification of Face and Speech +<br/>Using Momentum Back-Propagation Neural Network +<br/>1 Image Processing and Intelligent Systems Laboratory, Department of Image Engineering, +<br/><b>Graduate School of Advanced Imaging Science, Multimedia, and Film, Chung-Ang University</b><br/>221 Huksuk-dong, Tongjak-Ku, Seoul 156-756, Korea, +<br/>2 Broadcasting Media Research Group, Digital Broadcasting Research Division, ETRI, 161 +<br/>Gajeong-dong, Yuseong-Gu, Daejeon 305-700, Korea, +<br/>3 Intelligent Image Communication Laboratory, Department of Computer Engineering, +<br/><b>Kwangwoon University, 447-1 Wolge-dong, Nowon-Gu, Seoul 139-701, Korea</b></td><td>('1727735', 'Changhan Park', 'changhan park')<br/>('1722181', 'Myungseok Ki', 'myungseok ki')<br/>('1723542', 'Jaechan Namkung', 'jaechan namkung')<br/>('1684329', 'Joonki Paik', 'joonki paik')</td><td>initialchp@wm.cau.ac.kr, http://ipis.cau.ac.kr, +<br/>kkim@etri.re.kr, http://www.etri.re.kr, +<br/>namjc@daisy.kw.ac.kr, http://vision.kw.ac.kr. +</td></tr><tr><td>dce5e0a1f2cdc3d4e0e7ca0507592860599b0454</td><td>Facelet-Bank for Fast Portrait Manipulation +<br/><b>The Chinese University of Hong Kong</b><br/>2Tencent Youtu Lab +<br/><b>Johns Hopkins University</b></td><td>('2070527', 'Ying-Cong Chen', 'ying-cong chen')<br/>('40898180', 'Yangang Ye', 'yangang ye')<br/>('1729056', 'Jiaya Jia', 'jiaya jia')</td><td>{ycchen, linhj, ryli, xtao}@cse.cuhk.edu.hk +<br/>goodshenxy@gmail.com +<br/>Mshu1@jhu.edu +<br/>yangangye@tecent.com +<br/>leojia9@gmail.com +</td></tr><tr><td>dc9d62087ff93a821e6bb8a15a8ae2da3e39dcdd</td><td>Learning with Confident Examples: +<br/>Rank Pruning for Robust Classification with Noisy Labels +<br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, MA 02139 +</td><td>('39972987', 'Curtis G. Northcutt', 'curtis g. northcutt')<br/>('3716141', 'Tailin Wu', 'tailin wu')<br/>('1706040', 'Isaac L. Chuang', 'isaac l. chuang')</td><td>{cgn, tailin, ichuang}@mit.edu +</td></tr><tr><td>dcce3d7e8d59041e84fcdf4418702fb0f8e35043</td><td>Probabilistic Identity Characterization for Face Recognition∗ +<br/>Center for Automation Research (CfAR) and +<br/>Department of Electrical and Computer Engineering +<br/><b>University of Maryland, College Park, MD</b></td><td>('1682187', 'Shaohua Kevin Zhou', 'shaohua kevin zhou')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{shaohua, rama}@cfar.umd.edu +</td></tr><tr><td>dce3dff9216d63c4a77a2fcb0ec1adf6d2489394</td><td>Manifold Learning for Gender Classification +<br/>from Face Sequences +<br/><b>Machine Vision Group, P.O. Box 4500, FI-90014, University of Oulu, Finland</b></td><td>('1751372', 'Abdenour Hadid', 'abdenour hadid')</td><td></td></tr><tr><td>dc974c31201b6da32f48ef81ae5a9042512705fe</td><td>Am I done? Predicting Action Progress in Video +<br/>1 Media Integration and Communication Center, Univ. of Florence, Italy +<br/>2 Department of Mathematics “Tullio Levi-Civita”, Univ. of Padova, Italy +</td><td>('41172759', 'Federico Becattini', 'federico becattini')<br/>('1789269', 'Tiberio Uricchio', 'tiberio uricchio')<br/>('2831602', 'Lorenzo Seidenari', 'lorenzo seidenari')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')<br/>('1795847', 'Lamberto Ballan', 'lamberto ballan')</td><td></td></tr><tr><td>b6f758be954d34817d4ebaa22b30c63a4b8ddb35</td><td>A Proximity-Aware Hierarchical Clustering of Faces +<br/><b>University of Maryland, College Park</b></td><td>('3329881', 'Wei-An Lin', 'wei-an lin')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>walin@terpmail.umd.edu, pullpull@cs.umd.edu, rama@umiacs.umd.edu +</td></tr><tr><td>b62571691a23836b35719fc457e093b0db187956</td><td> Volume 3, Issue 5, May 2013 ISSN: 2277 128X +<br/>International Journal of Advanced Research in +<br/> Computer Science and Software Engineering +<br/> Research Paper +<br/> Available online at: www.ijarcsse.com +<br/>A Novel approach for securing biometric template +<br/>Dr.Chander Kant +<br/> Department of computer Science & applications Department of computer Science & applications +<br/><b>Kurukshetra University, Kurukshetra, India</b><br/><b>Kurukshetra University, Kurukshetra, India</b><br/> +</td><td>('3384880', 'Shweta Malhotra', 'shweta malhotra')</td><td></td></tr><tr><td>b69b239217d4e9a20fe4fe1417bf26c94ded9af9</td><td>A Temporally-Aware Interpolation Network for +<br/>Video Frame Inpainting +<br/><b>University of Michigan, Ann Arbor, USA</b></td><td>('2582303', 'Ximeng Sun', 'ximeng sun')<br/>('34246012', 'Ryan Szeto', 'ryan szeto')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')</td><td>{sunxm,szetor,jjcorso}@umich.edu +</td></tr><tr><td>b6c047ab10dd86b1443b088029ffe05d79bbe257</td><td></td><td></td><td></td></tr><tr><td>b6052dc718c72f2506cfd9d29422642ecf3992ef</td><td>A Survey on Human Motion Analysis from +<br/>Depth Data +<br/><b>University of Kentucky, 329 Rose St., Lexington, KY, 40508, U.S.A</b><br/>2 Microsoft, One Microsoft Way, Redmond, WA, 98052, U.S.A +<br/>3 SRI International Sarnoff, 201 Washington Rd, Princeton, NJ, 08540, U.S.A +<br/><b>University of Bonn, Roemerstrasse 164, 53117 Bonn, Germany</b></td><td>('3876303', 'Mao Ye', 'mao ye')<br/>('1681771', 'Qing Zhang', 'qing zhang')<br/>('40476140', 'Liang Wang', 'liang wang')<br/>('2446676', 'Jiejie Zhu', 'jiejie zhu')<br/>('38958903', 'Ruigang Yang', 'ruigang yang')<br/>('2946643', 'Juergen Gall', 'juergen gall')</td><td>mao.ye@uky.edu, qing.zhang@uky.edu, ryang@cs.uky.edu +<br/>liangwan@microsoft.com +<br/>jiejie.zhu@sri.com +<br/>gall@iai.uni-bonn.de +</td></tr><tr><td>b6145d3268032da70edc9cfececa1f9ffa4e3f11</td><td>c(cid:2) 2001 Kluwer Academic Publishers. Manufactured in The Netherlands. +<br/>Face Recognition Using the Discrete Cosine Transform +<br/><b>Center for Intelligent Machines, McGill University, 3480 University Street, Montreal, Canada H3A 2A</b></td><td>('1693521', 'Ziad M. Hafed', 'ziad m. hafed')<br/>('3631473', 'Martin D. Levine', 'martin d. levine')</td><td>zhafed@cim.mcgill.ca +<br/>levine@cim.mcgill.ca +</td></tr><tr><td>b6c53891dff24caa1f2e690552a1a5921554f994</td><td></td><td></td><td></td></tr><tr><td>b6ef158d95042f39765df04373c01546524c9ccd</td><td>Im2vid: Future Video Prediction for Static Image Action +<br/>Recognition +<br/>Badour Ahmad AlBahar +<br/>Thesis submitted to the Faculty of the +<br/><b>Virginia Polytechnic Institute and State University</b><br/>in partial fulfillment of the requirements for the degree of +<br/>Master of Science +<br/>in +<br/>Computer Engineering +<br/>Jia-Bin Huang, Chair +<br/>A. Lynn Abbott +<br/>Pratap Tokekar +<br/>May 9, 2018 +<br/>Blacksburg, Virginia +<br/>Keywords: Human Action Recognition, Static Image Action Recognition, Video Action +<br/>Recognition, Future Video Prediction. +<br/>Copyright 2018, Badour Ahmad AlBahar +</td><td></td><td></td></tr><tr><td>b68150bfdec373ed8e025f448b7a3485c16e3201</td><td>Adversarial Image Perturbation for Privacy Protection +<br/>A Game Theory Perspective +<br/><b>Max Planck Institute for Informatics, Saarland Informatics Campus, Saarbr cken, Germany</b></td><td>('2390510', 'Seong Joon Oh', 'seong joon oh')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td>{joon,mfritz,schiele}@mpi-inf.mpg.de +</td></tr><tr><td>b613b30a7cbe76700855479a8d25164fa7b6b9f1</td><td>1 +<br/>Identifying User-Specific Facial Affects from +<br/>Spontaneous Expressions with Minimal Annotation +</td><td>('23417737', 'Michael Xuelin Huang', 'michael xuelin huang')<br/>('1706729', 'Grace Ngai', 'grace ngai')<br/>('1730455', 'Kien A. Hua', 'kien a. hua')<br/>('1714454', 'Hong Va Leong', 'hong va leong')</td><td></td></tr><tr><td>b64cfb39840969b1c769e336a05a30e7f9efcd61</td><td>ORIGINAL RESEARCH +<br/>published: 15 June 2016 +<br/>doi: 10.3389/fict.2016.00009 +<br/>CRF-Based Context Modeling for +<br/>Person Identification in Broadcast +<br/>Videos +<br/><b>LIUM Laboratory, Le Mans, France, 2 Idiap Research Institute, Martigny, Switzerland</b><br/>We are investigating the problem of speaker and face identification in broadcast videos. +<br/>Identification is performed by associating automatically extracted names from overlaid +<br/>texts with speaker and face clusters. We aimed at exploiting the structure of news +<br/>videos to solve name/cluster association ambiguities and clustering errors. The proposed +<br/>approach combines iteratively two conditional random fields (CRF). The first CRF performs +<br/>the person diarization (joint temporal segmentation, clustering, and association of voices +<br/>jointly over the speech segments and the face tracks. It benefits from +<br/>and faces) +<br/>contextual +<br/>information being extracted from the image backgrounds and the overlaid +<br/>texts. The second CRF associates names with person clusters, thanks to co-occurrence +<br/>statistics. Experiments conducted on a recent and substantial public dataset containing +<br/>reports and debates demonstrate the interest and complementarity of the different +<br/>modeling steps and information sources: the use of these elements enables us to obtain +<br/>better performances in clustering and identification, especially in studio scenes. +<br/>Keywords: face identification, speaker identification, broadcast videos, conditional random field, face clustering, +<br/>speaker diarization +<br/>1. INTRODUCTION +<br/>For the last two decades, researchers have been trying to create indexing and fast search and +<br/>browsing tools capable of handling the growing amount of available video collections. Among the +<br/>associated possibilities, person identification is an important one. Indeed, video contents can often +<br/>be browsed through the appearances of their different actors. Moreover, the availability of each +<br/>person intervention allows easier access to video structure elements, such as the scene segmentation. +<br/>Both motivations are especially verified in the case of news collections. The focus of this paper is, +<br/>therefore, to develop a program able to identify persons in broadcast videos. That is, the program +<br/>must be able to provide all temporal segments corresponding to each face and speaker. +<br/>Person identification can be supervised. A face and/or a speaker model of the queried person is +<br/>then learned over manually labeled training data. However, this raises the problem of annotation +<br/>cost. An unsupervised and complementary approach consists of using the naming information +<br/>already present in the documents. Such resources include overlaid texts, speech transcripts, and +<br/>metadata. Motivated by this opportunity, unsupervised identification has been investigated for +<br/>15 years from the early work of Satoh et al. (1999) to the development of more complex news- +<br/>browsing systems exploiting this paradigm (Jou et al., 2013), or thanks to sponsored competitions +<br/>(Giraudel et al., 2012). Whatever the source of naming information, it must tackle two main +<br/>obstacles: associate the names to co-occurring speech and face segments and propagate this naming +<br/>information from the co-occurring segments to the other segments of this person. +<br/>Edited by: +<br/>Shin’Ichi Satoh, +<br/><b>National Institute of Informatics, Japan</b><br/>Reviewed by: +<br/>Thanh Duc Ngo, +<br/><b>Vietnam National University Ho Chi</b><br/>Minh City, Vietnam +<br/>Ichiro Ide, +<br/><b>Nagoya University, Japan</b><br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Computer Image Analysis, a section +<br/>of the journal Frontiers in ICT +<br/>Received: 16 October 2015 +<br/>Accepted: 12 May 2016 +<br/>Published: 15 June 2016 +<br/>Citation: +<br/>Gay P, Meignier S, Deléglise P and +<br/>Odobez J-M (2016) CRF-Based +<br/>Context Modeling for Person +<br/>Identification in Broadcast Videos. +<br/>doi: 10.3389/fict.2016.00009 +<br/>Frontiers in ICT | www.frontiersin.org +<br/>June 2016 | Volume 3 | Article 9 +</td><td>('14556501', 'Paul Gay', 'paul gay')<br/>('2446815', 'Sylvain Meignier', 'sylvain meignier')<br/>('1682046', 'Paul Deléglise', 'paul deléglise')<br/>('1719610', 'Jean-Marc Odobez', 'jean-marc odobez')<br/>('1719610', 'Jean-Marc Odobez', 'jean-marc odobez')</td><td>odobez@idiap.ch +</td></tr><tr><td>b6f682648418422e992e3ef78a6965773550d36b</td><td>February 8, 2017 +</td><td></td><td></td></tr><tr><td>b689d344502419f656d482bd186a5ee6b0140891</td><td>2009, Vol. 9, No. 2, 260 –264 +<br/>© 2009 American Psychological Association +<br/>1528-3542/09/$12.00 DOI: 10.1037/a0014681 +<br/>CORRECTED JULY 1, 2009; SEE LAST PAGE +<br/>BRIEF REPORTS +<br/>Christopher P. Said +<br/><b>Princeton University</b><br/><b>University of Amsterdam, University of Trento, Italy</b><br/><b>Princeton University</b><br/>People make trait inferences based on facial appearance despite little evidence that these inferences +<br/>accurately reflect personality. The authors tested the hypothesis that these inferences are driven in part +<br/>neutral faces on a set of trait dimensions. The authors then submitted the face images to a Bayesian +<br/>expression. In general, neutral faces that are perceived to have positive valence resemble happiness, faces +<br/>that are perceived to have negative valence resemble disgust and fear, and faces that are perceived to be +<br/>threatening resemble anger. These results support the idea that trait inferences are in part the result of an +<br/>then be misattributed as traits. +<br/>People evaluate neutral faces on multiple trait dimensions and +<br/>these evaluations have social consequences (Hassin & Trope, +<br/>2000). For instance, political candidates whose faces are perceived +<br/>as more competent are more likely to win elections (Ballew & +<br/>Todorov, 2007; Todorov, Mandisodza, Goren, & Hall, 2005), and +<br/>cadets whose faces are perceived as more dominant are more likely +<br/>to be promoted to higher military ranks (Mazur, Mazur, & Keating, +<br/>1984). +<br/>Although inferences about traits based on facial appearance are +<br/>made reliably across observers, there is little evidence that these +<br/>inferences accurately reflect the personality of the observed face. +<br/>Most correlations between perceived traits and actual traits are +<br/>weak though positive (Bond, Berry, & Omar, 1994), some are +<br/>inconsistent for men and women (Zebrowitz, Voinescu, & Collins, +<br/>1996), and some are negative (Zebrowitz, Andreoletti, Collins, +<br/>ogy and the Center for the Study of Brain, Mind and Behavior at +<br/><b>versity of Amsterdam, Amsterdam and University of Trento</b><br/>We thank Valerie Loehr for her assistance with the acquisition of trait +<br/>ratings, and Nick Oosterhof for helpful discussions. This research was +<br/>supported by National Science Foundation Grant BCS-0446846. +<br/>Correspondence should be addressed to Christopher P. Said, Department +<br/><b>of Psychology, Princeton University, Princeton, NJ 08540. E-mail</b><br/>260 +<br/>Lee, & Blumenthal, 1998). It is therefore puzzling that people +<br/>make reliable and rapid trait inferences from faces (Willis & +<br/>Todorov, 2006) when only little accurate information, at best, is +<br/>provided about personality. One intriguing explanation is that +<br/>neutral faces may contain structural properties that cause them to +<br/>resemble faces with more accurate and ecologically relevant in- +<br/>son, 1996; Montepare & Dobish, 2003). +<br/>Under this hypothesis, the adaptive ability to recognize emo- +<br/>tions overgeneralizes to neutral faces that merely bear a subtle +<br/>faces vary on trait dimensions such as trustworthiness (Engell, +<br/>Haxby, & Todorov, 2007). One possibility is that the source of +<br/>consensus in judging faces on social dimensions is the similarity of +<br/>the face to expressions corresponding to the dimension of trait +<br/>judgment (e.g., aggressiveness and anger). When given the task of +<br/>could base their judgments on this similarity. Evidence for this +<br/>hypothesis comes from research showing that the more a neutral +<br/>face is rated as happy by one group of participants the higher it is +<br/>rated on dominance and affiliation by another group of partici- +<br/>pants, and the more a face is rated as angry the higher it is rated on +<br/>dominance and the lower on affiliation (Montepare & Dobish, +<br/>2003). One interpretation of these findings is that people misat- +</td><td>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('2913698', 'Alexander Todorov', 'alexander todorov')<br/>('2913698', 'Alexander Todorov', 'alexander todorov')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')</td><td>csaid@princeton.edu +</td></tr><tr><td>b6d3caccdcb3fbce45ce1a68bb5643f7e68dadb3</td><td>Learning Spatio-Temporal Representation with Pseudo-3D Residual Networks ∗ +<br/><b>University of Science and Technology of China, Hefei, China</b><br/>‡ Microsoft Research, Beijing, China +</td><td>('3430743', 'Zhaofan Qiu', 'zhaofan qiu')<br/>('2053452', 'Ting Yao', 'ting yao')<br/>('1724211', 'Tao Mei', 'tao mei')</td><td>zhaofanqiu@gmail.com, {tiyao, tmei}@microsoft.com +</td></tr><tr><td>b6d0e461535116a675a0354e7da65b2c1d2958d4</td><td>Deep Directional Statistics: +<br/>Pose Estimation with +<br/>Uncertainty Quantification +<br/><b>Max Planck Institute for Intelligent Systems, T ubingen, Germany</b><br/>2 Amazon, T¨ubingen, Germany +<br/>3 Microsoft Research, Cambridge, UK +</td><td>('15968671', 'Sergey Prokudin', 'sergey prokudin')<br/>('2388416', 'Sebastian Nowozin', 'sebastian nowozin')</td><td>sergey.prokudin@tuebingen.mpg.de +</td></tr><tr><td>b656abc4d1e9c8dc699906b70d6fcd609fae8182</td><td></td><td></td><td></td></tr><tr><td>b6a01cd4572b5f2f3a82732ef07d7296ab0161d3</td><td>Kernel-Based Supervised Discrete Hashing for +<br/>Image Retrieval +<br/><b>University of Florida, Gainesville, FL, 32611, USA</b></td><td>('2766473', 'Xiaoshuang Shi', 'xiaoshuang shi')<br/>('2082604', 'Fuyong Xing', 'fuyong xing')<br/>('3457945', 'Jinzheng Cai', 'jinzheng cai')<br/>('2476328', 'Zizhao Zhang', 'zizhao zhang')<br/>('1877955', 'Yuanpu Xie', 'yuanpu xie')<br/>('1705066', 'Lin Yang', 'lin yang')</td><td>xsshi2015@ufl.edu +</td></tr><tr><td>a9791544baa14520379d47afd02e2e7353df87e5</td><td>Technical Note +<br/>The Need for Careful Data Collection for Pattern Recognition in +<br/>Digital Pathology +<br/><b>Montefiore Institute, University of Li ge, 4000 Li ge, Belgium</b><br/>Received: 08 December 2016 +<br/>Accepted: 15 March 2017 +<br/> Published: 10 April 2017 +</td><td>('1689882', 'Raphaël Marée', 'raphaël marée')</td><td></td></tr><tr><td>a9eb6e436cfcbded5a9f4b82f6b914c7f390adbd</td><td>(IJARAI) International Journal of Advanced Research in Artificial Intelligence, +<br/>Vol. 5, No.6, 2016 +<br/>A Model for Facial Emotion Inference Based on +<br/>Planar Dynamic Emotional Surfaces +<br/>Ruivo, J. P. P. +<br/>Escola Polit´ecnica +<br/>Negreiros, T. +<br/>Escola Polit´ecnica +<br/>Barretto, M. R. P. +<br/>Escola Polit´ecnica +<br/>Tinen, B. +<br/>Escola Polit´ecnica +<br/>Universidade de S˜ao Paulo +<br/>Universidade de S˜ao Paulo +<br/>Universidade de S˜ao Paulo +<br/>Universidade de S˜ao Paulo +<br/>S˜ao Paulo, Brazil +<br/>S˜ao Paulo, Brazil +<br/>S˜ao Paulo, Brazil +<br/>S˜ao Paulo, Brazil +</td><td></td><td></td></tr><tr><td>a955033ca6716bf9957b362b77092592461664b4</td><td> ISSN(Online): 2320-9801 +<br/> ISSN (Print): 2320-9798 +<br/>International Journal of Innovative Research in Computer +<br/>and Communication Engineering +<br/>(An ISO 3297: 2007 Certified Organization) +<br/>Video Based Face Recognition Using Artificial +<br/>Vol. 3, Issue 6, June 2015 +<br/>Neural Network +<br/><b>Pursuing M.Tech, Caarmel Engineering College, MG University, Kerala, India</b><br/><b>Caarmel Engineering College, MG University, Kerala, India</b></td><td></td><td></td></tr><tr><td>a956ff50ca958a3619b476d16525c6c3d17ca264</td><td>A Novel Bidirectional Neural Network for Face Recognition +<br/>JalilMazloum, Ali Jalali and Javad Amiryan +<br/>Electrical and Computer Engineering Department +<br/><b>ShahidBeheshti University</b><br/>Tehran, Iran +</td><td></td><td>J_Mazloum@sbu.ac.ir, A_Jalali@sbu.ac.ir, Amiryan.j@robocyrus.ir +</td></tr><tr><td>a92adfdd8996ab2bd7cdc910ea1d3db03c66d34f</td><td></td><td></td><td></td></tr><tr><td>a98316980b126f90514f33214dde51813693fe0d</td><td>Collaborations on YouTube: From Unsupervised Detection to the +<br/>Impact on Video and Channel Popularity +<br/>Multimedia Communications Lab (KOM), Technische Universität Darmstadt, Germany +</td><td>('49495293', 'Christian Koch', 'christian koch')<br/>('46203604', 'Moritz Lode', 'moritz lode')<br/>('2214486', 'Denny Stohr', 'denny stohr')<br/>('2869441', 'Amr Rizk', 'amr rizk')<br/>('1725298', 'Ralf Steinmetz', 'ralf steinmetz')</td><td>E-Mail: {Christian.Koch | Denny.Stohr | Amr.Rizk | Ralf.Steinmetz}@kom.tu-darmstadt.de +</td></tr><tr><td>a93781e6db8c03668f277676d901905ef44ae49f</td><td>Recent Datasets on Object Manipulation: A Survey +</td><td>('3112203', 'Yongqiang Huang', 'yongqiang huang')<br/>('39545911', 'Matteo Bianchi', 'matteo bianchi')<br/>('2646612', 'Minas Liarokapis', 'minas liarokapis')<br/>('1681376', 'Yu Sun', 'yu sun')</td><td></td></tr><tr><td>a9fc23d612e848250d5b675e064dba98f05ad0d9</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 9, No. 2, 2018 +<br/>Face Age Estimation Approach based on Deep +<br/>Learning and Principle Component Analysis +<br/>Faculty of Computers and +<br/>Informatics, +<br/><b>Benha University, Egypt</b><br/> Faculty of Computers and +<br/>Information, +<br/><b>Minia University, Egypt</b><br/>Faculty of Computers and +<br/>Informatics, +<br/><b>Benha University, Egypt</b></td><td>('3488856', 'Essam H. Houssein', 'essam h. houssein')<br/>('33680569', 'Hala H. Zayed', 'hala h. zayed')</td><td></td></tr><tr><td>a9adb6dcccab2d45828e11a6f152530ba8066de6</td><td>Aydınlanma Alt-uzaylarına dayalı Gürbüz Yüz Tanıma +<br/>Illumination Subspaces based Robust Face Recognition +<br/>Interactive Systems Labs, Universität Karlsruhe (TH) +<br/>76131 Karlsruhe, Almanya +<br/>web: http://isl.ira.uka.de/face_recognition +<br/>Özetçe +<br/>yönlerine +<br/>aydınlanma +<br/>kaynaklanan +<br/>sonra, yüz uzayı +<br/>Bu çalışmada aydınlanma alt-uzaylarına dayalı bir yüz tanıma +<br/>sistemi sunulmuştur. Bu sistemde, +<br/>ilk olarak, baskın +<br/>aydınlanma yönleri, bir topaklandırma algoritması kullanılarak +<br/>öğrenilmiştir. Topaklandırma algoritması sonucu önden, sağ +<br/>ve sol yanlardan olmak üzere üç baskın aydınlanma yönü +<br/>gözlemlenmiştir. Baskın +<br/>karar +<br/>-yüzün görünümündeki +<br/>kılındıktan +<br/>aydınlanmadan +<br/>kişi +<br/>kimliklerinden kaynaklanan değişimlerden ayırmak için- bu üç +<br/>aydınlanma uzayına bölünmüştür. Daha sonra, ek aydınlanma +<br/>yönü bilgisinden faydalanmak için aydınlanma alt-uzaylarına +<br/>dayalı yüz +<br/>tanıma algoritması kullanılmıştır. Önerilen +<br/>yaklaşım, CMU PIE veritabanında, “illumination” ve +<br/>“lighting” kümelerinde yer alan yüz +<br/>imgeleri üzerinde +<br/>sınanmıştır. Elde edilen deneysel sonuçlar, aydınlanma +<br/>yönünden yararlanmanın ve aydınlanma alt-uzaylarına dayalı +<br/>yüz tanıma algoritmasının yüz tanıma başarımını önemli +<br/>ölçüde arttırdığını göstermiştir. +<br/>değişimleri, +<br/>farklı +</td><td>('1770336', 'D. Kern', 'd. kern')<br/>('1742325', 'R. Stiefelhagen', 'r. stiefelhagen')</td><td>ekenel@ira.uka.de +</td></tr><tr><td>a967426ec9b761a989997d6a213d890fc34c5fe3</td><td>Relative Ranking of Facial Attractiveness +<br/>Department of Computer Science and Engineering +<br/><b>University of California, San Diego</b></td><td>('3079766', 'Hani Altwaijry', 'hani altwaijry')</td><td>{haltwaij,sjb}@cs.ucsd.edu +</td></tr><tr><td>a95dc0c4a9d882a903ce8c70e80399f38d2dcc89</td><td> TR-IIS-14-003 +<br/>Review and Implementation of +<br/>High-Dimensional Local Binary +<br/>Patterns and Its Application to +<br/>Face Recognition +<br/>July. 24, 2014 || Technical Report No. TR-IIS-14-003 +<br/>http://www.iis.sinica.edu.tw/page/library/TechReport/tr2014/tr14.html +</td><td>('33970300', 'Bor-Chun Chen', 'bor-chun chen')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')</td><td></td></tr><tr><td>a9286519e12675302b1d7d2fe0ca3cc4dc7d17f6</td><td>Learning to Succeed while Teaching to Fail: +<br/>Privacy in Closed Machine Learning Systems +</td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('4838771', 'Miguel R. D. Rodrigues', 'miguel r. d. rodrigues')<br/>('1699339', 'Guillermo Sapiro', 'guillermo sapiro')</td><td></td></tr><tr><td>a949b8700ca6ba96ee40f75dfee1410c5bbdb3db</td><td>Instance-weighted Transfer Learning of Active Appearance Models +<br/><b>Computer Vision Group, Friedrich Schiller University of Jena, Germany</b><br/>Ernst-Abbe-Platz 2-4, 07743 Jena, Germany +</td><td>('1708249', 'Daniel Haase', 'daniel haase')<br/>('1679449', 'Erik Rodner', 'erik rodner')<br/>('1728382', 'Joachim Denzler', 'joachim denzler')</td><td>{daniel.haase,erik.rodner,joachim.denzler}@uni-jena.de +</td></tr><tr><td>a92b5234b8b73e06709dd48ec5f0ec357c1aabed</td><td></td><td></td><td></td></tr><tr><td>a9be20954e9177d8b2bc39747acdea4f5496f394</td><td>Event-specific Image Importance +<br/><b>University of California, San Diego</b><br/>2Adobe Research +</td><td>('35259685', 'Yufei Wang', 'yufei wang')</td><td>{yuw176, gary}@ucsd.edu +<br/>{zlin, xshen, rmech, gmiller}@adobe.com +</td></tr><tr><td>d5afd7b76f1391321a1340a19ba63eec9e0f9833</td><td>Journal of Information Hiding and Multimedia Signal Processing +<br/>Ubiquitous International +<br/>c⃝2010 ISSN 2073-4212 +<br/>Volume 1, Number 3, July 2010 +<br/>Statistical Analysis of Human Facial Expressions +<br/>Department of Informatics +<br/><b>Aristotle University of Thessaloniki</b><br/>Box 451, 54124 Thessaloniki, Greece +<br/>Department of Informatics +<br/><b>Aristotle University of Thessaloniki</b><br/>Box 451, 54124 Thessaloniki, Greece +<br/><b>Informatics and Telematics Institute</b><br/>CERTH, Greece +<br/>Received March 2010; revised June 2010 +</td><td>('2764130', 'Stelios Krinidis', 'stelios krinidis')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>stelios.krinidis@mycosmos.gr +<br/>pitas@aiia.csd.auth.gr +</td></tr><tr><td>d5375f51eeb0c6eff71d6c6ad73e11e9353c1f12</td><td>Manifold Ranking-Based Locality Preserving Projections +<br/><b>School of Computer Science and Engineering, South China University of Technology</b><br/>Guangzhou 510006, Guangdong, China +</td><td>('2132230', 'Jia Wei', 'jia wei')<br/>('3231018', 'Zewei Chen', 'zewei chen')<br/>('1837988', 'Pingyang Niu', 'pingyang niu')<br/>('2524825', 'Yishun Chen', 'yishun chen')<br/>('7307608', 'Wenhui Chen', 'wenhui chen')</td><td>csjwei@scut.edu.cn +</td></tr><tr><td>d5d7e89e6210fcbaa52dc277c1e307632cd91dab</td><td>DOTA: A Large-scale Dataset for Object Detection in Aerial Images∗ +<br/><b>State Key Lab. LIESMARS, Wuhan University, China</b><br/>2EIS, Huazhong Univ. Sci. and Tech., China +<br/><b>Computer Science Depart., Cornell University, USA</b><br/><b>Computer Science Depart., Rochester University, USA</b><br/>5German Aerospace Center (DLR), Germany +<br/><b>DAIS, University of Venice, Italy</b><br/>January 30, 2018 +</td><td>('39943835', 'Gui-Song Xia', 'gui-song xia')<br/>('1686737', 'Xiang Bai', 'xiang bai')<br/>('1749386', 'Jian Ding', 'jian ding')<br/>('48148046', 'Zhen Zhu', 'zhen zhu')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')<br/>('1777167', 'Mihai Datcu', 'mihai datcu')<br/>('8111020', 'Marcello Pelillo', 'marcello pelillo')<br/>('1733213', 'Liangpei Zhang', 'liangpei zhang')</td><td>{guisong.xia, jding, zlp62}@whu.edu.cn +<br/>{xbai, zzhu}@hust.edu.cn +<br/>sjb344@cornell.edu +<br/>jiebo.luo@gmail.com +<br/>mihai.datcu@dlr.de +<br/>pelillo@dsi.unive.it +</td></tr><tr><td>d50c6d22449cc9170ab868b42f8c72f8d31f9b6c</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>1668 +</td><td></td><td></td></tr><tr><td>d522c162bd03e935b1417f2e564d1357e98826d2</td><td>He et al. EURASIP Journal on Advances in Signal Processing 2013, 2013:19 +<br/>http://asp.eurasipjournals.com/content/2013/1/19 +<br/>RESEARCH +<br/>Open Access +<br/>Weakly supervised object extraction with +<br/>iterative contour prior for remote sensing +<br/>images +</td><td>('2456383', 'Chu He', 'chu he')<br/>('40382947', 'Yu Zhang', 'yu zhang')<br/>('1813780', 'Bo Shi', 'bo shi')<br/>('1727252', 'Xin Su', 'xin su')<br/>('32514309', 'Xin Xu', 'xin xu')<br/>('2048631', 'Mingsheng Liao', 'mingsheng liao')</td><td></td></tr><tr><td>d59f18fcb07648381aa5232842eabba1db52383e</td><td>International Conference on Systemics, Cybernetics and Informatics, February 12–15, 2004 +<br/>ROBUST FACIAL EXPRESSION RECOGNITION USING SPATIALLY +<br/>LOCALIZED GEOMETRIC MODEL +<br/>Department of Electrical Engineering +<br/>Dept. of Computer Sc. and Engg. +<br/>IIT Kanpur +<br/> Kanpur 208016, India +<br/>Kanpur 208016, India +<br/> IIT Kanpur +<br/>Dept. of Computer Sc. and Engg. +<br/> IIT Kanpur +<br/>Kanpur 208016, India +<br/>While approaches based on 3D deformable facial model have +<br/>achieved expression recognition rates of as high as 98% [2], they +<br/>are computationally inefficient and require considerable apriori +<br/>training based on 3D information, which is often unavailable. +<br/>Recognition from 2D images remains a difficult yet important +<br/>problem for areas such as +<br/>image database querying and +<br/>classification. The accuracy rates achieved for 2D images are +<br/>around 90% [3,4,5,11]. In a recent review of expression +<br/>recognition, Fasel [1] considers the problem along several +<br/>dimensions: whether features such as lips or eyebrows are first +<br/>identified in the face (local [4] vs holistic [11]), or whether the +<br/>image model used is 2D or 3D. Methods proposed for expression +<br/>recognition from 2D images include the Gabor-Wavelet [5] or +<br/>Holistic Optical flow [11] approach. +<br/>This paper describes a more robust system for facial expression +<br/>recognition from image sequences using 2D appearance-based +<br/>local approach for the extraction of intransient facial features, i.e. +<br/>features such as eyebrows, lips, or mouth, which are always +<br/>present in the image, but may be deformed [1] (in contrast, +<br/>transient features are wrinkles or bulges that disappear at other +<br/>times). The main advantages of such an approach is low +<br/>computational requirements, ability to work with both colored and +<br/>grayscale images and robustness in handling partial occlusions +<br/>[3]. +<br/>Edge projection analysis which is used here for feature extraction +<br/>(eyebrows and lips) is well known [6]. Unlike [6] which describes +<br/>a template based matching as an essential starting point, we use +<br/>contours analysis. Our system computes a feature vector based on +<br/>geometrical model of the face and then classifies it into four +<br/>expression classes using a feed-forward basis function net. The +<br/>system detects open and closed state of the mouth as well. The +<br/>algorithm presented here works on both color and grayscale image +<br/>sequences. An important aspect of our work is the use of color +<br/>information for robust and more accurate segmentation of lip +<br/>region in case of color images. The novel lip-enhancement +<br/>transform is based on Gaussian modeling of skin and lip color. +<br/>To place the work in a larger context of face analysis and +<br/>recognition, the overall task requires that the part of the image +<br/>involving the face be detected and segmented. We assume that a +<br/>near-frontal view of the face is available. Tests on a grayscale +<br/>and two color face image databases ([8] and [9,10]) demonstrate a +<br/>superior recognition rate for four facial expressions (smile, +<br/>surprise, disgust and sad against neutral). +<br/>image sequences +</td><td>('1681995', 'Ashutosh Saxena', 'ashutosh saxena')<br/>('40101676', 'Ankit Anand', 'ankit anand')<br/>('1803835', 'Amitabha Mukerjee', 'amitabha mukerjee')</td><td>ashutosh.saxena@ieee.org +<br/>ankanand@cse.iitk.ac.in +<br/>amit@cse.iitk.ac.in +</td></tr><tr><td>d5fa9d98c8da54a57abf353767a927d662b7f026</td><td> VOL. 1, NO. 2, Oct 2010 E-ISSN 2218-6301 +<br/>Journal of Emerging Trends in Computing and Information Sciences +<br/>©2009-2010 CIS Journal. All rights reserved. +<br/>http://www.cisjournal.org +<br/>Age Estimation based on Neural Networks using Face Features +<br/> +<br/>Corresponding Author: Faculty of Information Technology +<br/><b>Islamic University of Gaza - Palestine</b><br/>Email +<br/>: +<br/>edu. +<br/> ps. +</td><td>('1714298', 'Nabil Hewahi', 'nabil hewahi')</td><td>nhewahi@iugaza +</td></tr><tr><td>d588dd4f305cdea37add2e9bb3d769df98efe880</td><td> +<br/>Audio-Visual Authentication System over the +<br/>Internet Protocol +<br/>abandoned. +<br/>in +<br/>illumination based +<br/>is developed with the objective to +</td><td>('1968167', 'Yee Wan Wong', 'yee wan wong')</td><td></td></tr><tr><td>d5444f9475253bbcfef85c351ea9dab56793b9ea</td><td>IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS +<br/>BoxCars: Improving Fine-Grained Recognition +<br/>of Vehicles using 3D Bounding Boxes +<br/>in Traffic Surveillance +<br/>in contrast +</td><td>('34891870', 'Jakub Sochor', 'jakub sochor')<br/>('1785162', 'Adam Herout', 'adam herout')</td><td></td></tr><tr><td>d5ab6aa15dad26a6ace5ab83ce62b7467a18a88e</td><td>World Journal of Computer Application and Technology 2(7): 133-138, 2014 +<br/>DOI: 10.13189/wjcat.2014.020701 +<br/>http://www.hrpub.org +<br/>Optimized Structure for Facial Action Unit Relationship +<br/>Using Bayesian Network +<br/>Intelligent Biometric Group, School of Electrical and Electronic Engineering, Engineering Campus, Universiti Sains Malaysia, Pulau +<br/>Pinang, Malaysia +<br/>Copyright © 2014 Horizon Research Publishing All rights reserved. +</td><td>('9115930', 'Yee Koon Loh', 'yee koon loh')<br/>('3120408', 'Shahrel A. Suandi', 'shahrel a. suandi')</td><td>*Corresponding Author: lyk10_eee045@student.usm.my +</td></tr><tr><td>d5b0e73b584be507198b6665bcddeba92b62e1e5</td><td>CHEN ET AL.: MULTI-REGION ENSEMBLE CNNS FOR AGE ESTIMATION +<br/>Multi-Region Ensemble Convolutional Neural +<br/>Networks for High-Accuracy Age Estimation +<br/>1 Faculty of Information Technology +<br/><b>Macau University of Science and</b><br/>Technology, Macau SAR +<br/>2 National Laboratory of Pattern +<br/><b>Recognition, Institute of Automation</b><br/>Chinese Academy of Sciences +<br/><b>University of Chinese Academy of</b><br/>Sciences +<br/>4 Computing, School of Science and +<br/><b>Engineering, University of Dundee</b></td><td>('38141486', 'Yiliang Chen', 'yiliang chen')<br/>('9645431', 'Zichang Tan', 'zichang tan')<br/>('1916793', 'Alex Po Leung', 'alex po leung')<br/>('1756538', 'Jun Wan', 'jun wan')<br/>('40539612', 'Jianguo Zhang', 'jianguo zhang')</td><td>elichan5168@gmail.com +<br/>tanzichang2016@ia.ac.cn +<br/>pleung@must.edu.mo +<br/>jun.wan@ia.ac.cn +<br/>jnzhang@dundee.ac.uk +</td></tr><tr><td>d56fe69cbfd08525f20679ffc50707b738b88031</td><td>Training of multiple classifier systems utilizing +<br/>partially labelled sequences +<br/><b></b><br/>89069 Ulm - Germany +</td><td>('3037635', 'Martin Schels', 'martin schels')<br/>('2307794', 'Patrick Schillinger', 'patrick schillinger')<br/>('1685857', 'Friedhelm Schwenker', 'friedhelm schwenker')</td><td></td></tr><tr><td>d5de42d37ee84c86b8f9a054f90ddb4566990ec0</td><td>Asynchronous Temporal Fields for Action Recognition +<br/><b>Carnegie Mellon University 2University of Washington 3Allen Institute for Arti cial Intelligence</b><br/>github.com/gsig/temporal-fields/ +</td><td>('34280810', 'Gunnar A. Sigurdsson', 'gunnar a. sigurdsson')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')</td><td></td></tr><tr><td>d50751da2997e7ebc89244c88a4d0d18405e8507</td><td></td><td></td><td></td></tr><tr><td>d511e903a882658c9f6f930d6dd183007f508eda</td><td></td><td></td><td></td></tr><tr><td>d50a40f2d24363809a9ac57cf7fbb630644af0e5</td><td>END-TO-END TRAINED CNN ENCODER-DECODER NETWORKS FOR IMAGE +<br/>STEGANOGRAPHY +<br/><b>National University of Computer and Emerging Sciences (NUCES-FAST), Islamabad, Pakistan</b><br/>Reveal.ai (Recognition, Vision & Learning) Lab +</td><td>('9205693', 'Atique ur Rehman', 'atique ur rehman')<br/>('2695106', 'Sibt ul Hussain', 'sibt ul hussain')</td><td></td></tr><tr><td>d5b5c63c5611d7b911bc1f7e161a0863a34d44ea</td><td>Extracting Scene-dependent Discriminant +<br/>Features for Enhancing Face Recognition +<br/>under Severe Conditions +<br/><b>Information and Media Processing Research Laboratories, NEC Corporation</b><br/>1753, Shimonumabe, Nakahara-Ku, Kawasaki 211-8666 Japan +</td><td>('1709089', 'Rui Ishiyama', 'rui ishiyama')<br/>('35577655', 'Nobuyuki Yasukawa', 'nobuyuki yasukawa')</td><td></td></tr><tr><td>d59404354f84ad98fa809fd1295608bf3d658bdc</td><td>International Journal of Computer Vision manuscript No. +<br/>(will be inserted by the editor) +<br/>Face Synthesis from Visual Attributes via Sketch using +<br/>Conditional VAEs and GANs +<br/>Received: date / Accepted: date +</td><td>('29673017', 'Xing Di', 'xing di')</td><td></td></tr><tr><td>d5e1173dcb2a51b483f86694889b015d55094634</td><td></td><td></td><td></td></tr><tr><td>d28d32af7ef9889ef9cb877345a90ea85e70f7f1</td><td>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>Local-Global Landmark Confidences for Face Recognition +<br/><b>Institute for Robotics and Intelligent Systems, University of Southern California, CA, USA</b><br/><b>Language Technologies Institute, Carnegie Mellon University, PA, USA</b></td><td>('2792633', 'KangGeon Kim', 'kanggeon kim')<br/>('1752756', 'Feng-Ju Chang', 'feng-ju chang')<br/>('1689391', 'Jongmoo Choi', 'jongmoo choi')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')<br/>('1694832', 'Ramakant Nevatia', 'ramakant nevatia')</td><td></td></tr><tr><td>d28d697b578867500632b35b1b19d3d76698f4a9</td><td>Appears in the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR’99, Fort Collins, Colorado, USA, June 23-25, 1999. +<br/>Face Recognition Using Shape and Texture +<br/>Department of Computer Science +<br/><b>George Mason University</b><br/>Fairfax, VA 22030-4444 +<br/> +</td><td>('39664966', 'Chengjun Liu', 'chengjun liu')<br/>('1781577', 'Harry Wechsler', 'harry wechsler')</td><td>@cs.gmu.edu +</td></tr><tr><td>d231a81b38fde73bdbf13cfec57d6652f8546c3c</td><td>SUPERRESOLUTION TECHNIQUES +<br/> FOR FACE RECOGNITION FROM VIDEO +<br/>by +<br/><b>B.S., E.E., Bo azi i University</b><br/>Submitted to the Graduate School of Engineering +<br/> and Natural Sciences in partially fulfillment of +<br/>the requirement for the degree of +<br/>Master of Science +<br/>Graduate Program in Electronics Engineering and Computer Science +<br/><b>Sabanc University</b><br/>Spring 2005 +</td><td>('2258053', 'Osman Gökhan Sezer', 'osman gökhan sezer')</td><td></td></tr><tr><td>d22785eae6b7503cb16402514fd5bd9571511654</td><td>Evaluating Facial Expressions with Different +<br/>Occlusion around Image Sequence +<br/> +<br/>Department of Computer Science +<br/><b>Sanghvi Institute of Management and Science</b><br/>Indore (MP), India +<br/>I. +<br/>local +<br/>INTRODUCTION +<br/> +</td><td>('2890210', 'Ramchand Hablani', 'ramchand hablani')</td><td></td></tr><tr><td>d2eb1079552fb736e3ba5e494543e67620832c52</td><td>ANNUNZIATA, SAGONAS, CALÌ: DENSELY FUSED SPATIAL TRANSFORMER NETWORKS1 +<br/>DeSTNet: Densely Fused Spatial +<br/>Transformer Networks1 +<br/>Onfido Research +<br/>3 Finsbury Avenue +<br/>London, UK +</td><td>('31336510', 'Roberto Annunziata', 'roberto annunziata')<br/>('3320415', 'Christos Sagonas', 'christos sagonas')<br/>('1997807', 'Jacques Calì', 'jacques calì')</td><td>roberto.annunziata@onfido.com +<br/>christos.sagonas@onfido.com +<br/>jacques.cali@onfido.com +</td></tr><tr><td>d24dafe10ec43ac8fb98715b0e0bd8e479985260</td><td>J Nonverbal Behav (2018) 42:81–99 +<br/>https://doi.org/10.1007/s10919-017-0266-z +<br/>O R I G I N A L P A P E R +<br/>Effects of Social Anxiety on Emotional Mimicry +<br/>and Contagion: Feeling Negative, but Smiling Politely +<br/>• Gerben A. van Kleef2 +<br/>• Agneta H. Fischer2 +<br/>Published online: 25 September 2017 +<br/>Ó The Author(s) 2017. This article is an open access publication +</td><td>('4041392', 'Corine Dijk', 'corine dijk')<br/>('35427440', 'Charlotte van Eeuwijk', 'charlotte van eeuwijk')<br/>('1878851', 'Nexhmedin Morina', 'nexhmedin morina')</td><td></td></tr><tr><td>d29eec5e047560627c16803029d2eb8a4e61da75</td><td>Feature Transfer Learning for Deep Face +<br/>Recognition with Long-Tail Data +<br/><b>Michigan State University, NEC Laboratories America</b></td><td>('39708770', 'Xi Yin', 'xi yin')<br/>('15644381', 'Xiang Yu', 'xiang yu')<br/>('1729571', 'Kihyuk Sohn', 'kihyuk sohn')<br/>('40022363', 'Xiaoming Liu', 'xiaoming liu')<br/>('2099305', 'Manmohan Chandraker', 'manmohan chandraker')</td><td>{yinxi1,liuxm}@cse.msu.edu,{xiangyu,ksohn,manu}@nec-labs.com +</td></tr><tr><td>d280bcbb387b1d548173917ae82cb6944e3ceca6</td><td>FACIAL GRID TRANSFORMATION: A NOVEL FACE REGISTRATION APPROACH FOR +<br/>IMPROVING FACIAL ACTION UNIT RECOGNITION +<br/><b>University of South Carolina, Columbia, USA</b></td><td>('3225915', 'Shizhong Han', 'shizhong han')<br/>('3091647', 'Zibo Meng', 'zibo meng')<br/>('40205868', 'Ping Liu', 'ping liu')<br/>('1686235', 'Yan Tong', 'yan tong')</td><td></td></tr><tr><td>d278e020be85a1ccd90aa366b70c43884dd3f798</td><td>Learning From Less Data: Diversified Subset Selection and +<br/>Active Learning in Image Classification Tasks +<br/>IIT Bombay +<br/>Mumbai, Maharashtra, India +<br/>AITOE Labs +<br/>Mumbai, Maharashtra, India +<br/>AITOE Labs +<br/>Mumbai, Maharashtra, India +<br/>Rishabh Iyer +<br/>AITOE Labs +<br/>Seattle, Washington, USA +<br/>AITOE Labs +<br/>Seattle, Washington, USA +<br/>Narsimha Raju +<br/>IIT Bombay +<br/>Mumbai, Maharashtra, India +<br/>IIT Bombay +<br/>Mumbai, Maharashtra, India +<br/>IIT Bombay +<br/>Mumbai, Maharashtra, India +<br/>May 30, 2018 +</td><td>('3333118', 'Vishal Kaushal', 'vishal kaushal')<br/>('40224337', 'Khoshrav Doctor', 'khoshrav doctor')<br/>('33911191', 'Suyash Shetty', 'suyash shetty')<br/>('10710354', 'Anurag Sahoo', 'anurag sahoo')<br/>('49613683', 'Pankaj Singh', 'pankaj singh')<br/>('1697088', 'Ganesh Ramakrishnan', 'ganesh ramakrishnan')</td><td>vkaushal@cse.iitb.ac.in +<br/>khoshrav@gmail.com +<br/>suyashshetty29@gmail.com +<br/>rishabh@aitoelabs.com +<br/>anurag@aitoelabs.com +<br/>uavnraju@cse.iitb.ac.in +<br/>pr.pankajsingh@gmail.com +<br/>ganesh@cse.iitb.ac.in +</td></tr><tr><td>d26b443f87df76034ff0fa9c5de9779152753f0c</td><td>A GPU-Oriented Algorithm Design for +<br/>Secant-Based Dimensionality Reduction +<br/>Department of Mathematics +<br/><b>Colorado State University</b><br/>Fort Collins, CO 80523-1874 +<br/>tool +<br/>for extracting useful +</td><td>('51042250', 'Henry Kvinge', 'henry kvinge')<br/>('51121534', 'Elin Farnell', 'elin farnell')<br/>('41211081', 'Michael Kirby', 'michael kirby')<br/>('30383278', 'Chris Peterson', 'chris peterson')</td><td></td></tr><tr><td>d2cd9a7f19600370bce3ea29aba97d949fe0ceb9</td><td>Separability Oriented Preprocessing for +<br/>Illumination-Insensitive Face Recognition +<br/>1 Key Lab of Intelligent Information Processing +<br/>of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing 100190, China</b><br/>2 Department of Computer Science and Engineering, +<br/><b>Michigan State University, East Lansing, MI 48824, U.S.A</b><br/>3 Omron Social Solutions Co., LTD., Kyoto, Japan +<br/><b>Institute of Digital Media, Peking University, Beijing 100871, China</b><br/>some +<br/>last decade, +</td><td>('34393045', 'Hu Han', 'hu han')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')<br/>('1710195', 'Shihong Lao', 'shihong lao')<br/>('1698902', 'Wen Gao', 'wen gao')</td><td>{hhan,sgshan,xlchen}@jdl.ac.cn, lao@ari.ncl.omron.co.jp, wgao@pku.edu.cn +</td></tr><tr><td>d22b378fb4ef241d8d210202893518d08e0bb213</td><td>Random Faces Guided Sparse Many-to-One Encoder +<br/>for Pose-Invariant Face Recognition +<br/><b>Polytechnic Institute of NYU, NY, USA</b><br/><b>College of Computer and Information Science, Northeastern University, MA, USA</b><br/><b>Northeastern University, MA, USA</b></td><td>('3272356', 'Yizhe Zhang', 'yizhe zhang')</td><td>zhangyizhe1987@gmail.com, mingshao@ccs.neu.edu, wong@poly.edu, yunfu@ece.neu.edu +</td></tr><tr><td>aac39ca161dfc52aade063901f02f56d01a1693c</td><td>The Analysis of Parameters t and k of LPP on +<br/>Several Famous Face Databases +<br/><b>College of Computer Science and Technology</b><br/><b>Jilin University, Changchun 130012, China</b></td><td>('7489436', 'Sujing Wang', 'sujing wang')<br/>('1758249', 'Na Zhang', 'na zhang')<br/>('3028807', 'Mingfang Sun', 'mingfang sun')<br/>('8239114', 'Chunguang Zhou', 'chunguang zhou')</td><td>{wangsj08, nazhang08}@mails.jlu.edu.cn; cgzhou@jlu.edu.cn +</td></tr><tr><td>aadf4b077880ae5eee5dd298ab9e79a1b0114555</td><td>Dynamics-based Facial Emotion Recognition and Pain Detection +<br/>Using Hankel Matrices for +<br/><b>DICGIM - University of Palermo</b><br/>V.le delle Scienze, Ed. 6, 90128 Palermo (Italy) +</td><td>('1711610', 'Liliana Lo Presti', 'liliana lo presti')<br/>('9127836', 'Marco La Cascia', 'marco la cascia')</td><td>liliana.lopresti@unipa.it +</td></tr><tr><td>aa127e6b2dc0aaccfb85e93e8b557f83ebee816b</td><td>Advancing Human Pose and +<br/>Gesture Recognition +<br/>DPhil Thesis +<br/>Supervisor: Professor Andrew Zisserman +<br/>Tomas Pfister +<br/>Visual Geometry Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b><br/><b>Wolfson College</b><br/>April 2015 +</td><td></td><td></td></tr><tr><td>aafb271684a52a0b23debb3a5793eb618940c5dd</td><td></td><td></td><td></td></tr><tr><td>aae742779e8b754da7973949992d258d6ca26216</td><td>Robust Facial Expression Classification Using Shape +<br/>and Appearance Features +<br/>Department of Electrical Engineering, +<br/><b>Indian Institute of Technology Kharagpur, India</b></td><td>('2680543', 'Aurobinda Routray', 'aurobinda routray')</td><td></td></tr><tr><td>aa8ef6ba6587c8a771ec4f91a0dd9099e96f6d52</td><td>Improved Face Tracking Thanks to Local Features +<br/>Correspondence +<br/>Department of Information Engineering +<br/><b>University of Brescia</b></td><td>('3134795', 'Alberto Piacenza', 'alberto piacenza')<br/>('1806359', 'Fabrizio Guerrini', 'fabrizio guerrini')<br/>('1741369', 'Riccardo Leonardi', 'riccardo leonardi')</td><td></td></tr><tr><td>aab3561acbd19f7397cbae39dd34b3be33220309</td><td>Quantization Mimic: Towards Very Tiny CNN +<br/>for Object Detection +<br/><b>Tsinghua University, Beijing, China</b><br/><b>The Chinese University of Hong Kong, Hong Kong, China</b><br/>3SenseTime, Beijing, China +<br/><b>The University of Sydney, SenseTime Computer Vision Research Group, Sydney</b><br/>New South Wales, Australia +</td><td>('49019561', 'Yi Wei', 'yi wei')<br/>('7418754', 'Xinyu Pan', 'xinyu pan')<br/>('46636770', 'Hongwei Qin', 'hongwei qin')<br/>('1721677', 'Junjie Yan', 'junjie yan')</td><td>wei-y15@mails.tsinghua.edu.cn,THUSEpxy@gmail.com +<br/>qinhongwei@sensetime.com,wanli.ouyang@sydney.edu.au +<br/>yanjunjie@sensetime.com +</td></tr><tr><td>aa912375eaf50439bec23de615aa8a31a3395ad3</td><td>International Journal on Cryptography and Information Security(IJCIS),Vol.2, No.2, June 2012 +<br/>Implementation of a New Methodology to Reduce +<br/>the Effects of Changes of Illumination in Face +<br/>Recognition-based Authentication +<br/><b>Howard University, Washington DC</b><br/><b>Howard University, Washington DC</b></td><td>('3437323', 'Andres Alarcon-Ramirez', 'andres alarcon-ramirez')<br/>('2522254', 'Mohamed F. Chouikha', 'mohamed f. chouikha')</td><td>alarconramirezandr@bison.howard.edu +<br/>mchouikha@howard.edu +</td></tr><tr><td>aa52910c8f95e91e9fc96a1aefd406ffa66d797d</td><td>FACE RECOGNITION SYSTEM BASED +<br/>ON 2DFLD AND PCA +<br/>E&TC Department +<br/>Sinhgad Academy of Engineering +<br/>Pune, India +<br/>Mr. Hulle Rohit Rajiv +<br/>ME E&TC [Digital System] +<br/>Sinhgad Academy of Engineering +<br/>Pune, India +</td><td>('2985198', 'Sachin D. Ruikar', 'sachin d. ruikar')</td><td>ruikarsachin@gmail.com +<br/>rohithulle@gmail.com +</td></tr><tr><td>aaeb8b634bb96a372b972f63ec1dc4db62e7b62a</td><td>ISSN (e): 2250 – 3005 || Vol, 04 || Issue, 12 || December – 2014 || +<br/>International Journal of Computational Engineering Research (IJCER) +<br/>Facial Expression Recognition System: A Digital Printing +<br/>Application +<br/><b>Jadavpur University, India</b><br/><b>Jadavpur University, India</b></td><td>('2226316', 'Somnath Banerjee', 'somnath banerjee')</td><td></td></tr><tr><td>aafb8dc8fda3b13a64ec3f1ca7911df01707c453</td><td>Excitation Backprop for RNNs +<br/><b>Boston University 2Pattern Analysis and Computer Vision (PAVIS</b><br/>Istituto Italiano di Tecnologia 3Adobe Research 4Computer Science Department, Universit`a di Verona +<br/>Figure 1: Our proposed framework spatiotemporally highlights/grounds the evidence that an RNN model used in producing a class label +<br/>or caption for a given input video. In this example, by using our proposed back-propagation method, the evidence for the activity class +<br/>CliffDiving is highlighted in a video that contains CliffDiving and HorseRiding. Our model employs a single backward pass to produce +<br/>saliency maps that highlight the evidence that a given RNN used in generating its outputs. +</td><td>('3298267', 'Sarah Adel Bargal', 'sarah adel bargal')<br/>('40063519', 'Andrea Zunino', 'andrea zunino')<br/>('40622560', 'Donghyun Kim', 'donghyun kim')<br/>('1701293', 'Jianming Zhang', 'jianming zhang')<br/>('1727204', 'Vittorio Murino', 'vittorio murino')<br/>('1749590', 'Stan Sclaroff', 'stan sclaroff')</td><td>{sbargal,donhk,sclaroff}@bu.edu, {andrea.zunino,vittorio.murino}@iit.it, jianmzha@adobe.com +</td></tr><tr><td>aa0c30bd923774add6e2f27ac74acd197b9110f2</td><td>DYNAMIC PROBABILISTIC LINEAR DISCRIMINANT ANALYSIS FOR VIDEO +<br/>CLASSIFICATION +<br/><b>Deparment of Computing, Imperial College London, UK</b><br/><b>Deparment of Computing, Goldsmiths, University of London, UK</b><br/><b>Middlesex University London, 4International Hellenic University</b><br/><b>Center for Machine Vision and Signal Analysis, University of Oulu, Finland</b></td><td>('35340264', 'Alessandro Fabris', 'alessandro fabris')<br/>('1752913', 'Mihalis A. Nicolaou', 'mihalis a. nicolaou')<br/>('1754270', 'Irene Kotsia', 'irene kotsia')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td></td></tr><tr><td>aadfcaf601630bdc2af11c00eb34220da59b7559</td><td>Multi-view Hybrid Embedding: +<br/>A Divide-and-Conquer Approach +</td><td>('30443690', 'Jiamiao Xu', 'jiamiao xu')<br/>('2462771', 'Shujian Yu', 'shujian yu')<br/>('1744228', 'Xinge You', 'xinge you')<br/>('3381421', 'Mengjun Leng', 'mengjun leng')<br/>('15132338', 'Xiao-Yuan Jing', 'xiao-yuan jing')<br/>('1697202', 'C. L. Philip Chen', 'c. l. philip chen')</td><td></td></tr><tr><td>aaa4c625f5f9b65c7f3df5c7bfe8a6595d0195a5</td><td>Biometrics in Ambient Intelligence +</td><td>('1725688', 'Massimo Tistarelli', 'massimo tistarelli')</td><td></td></tr><tr><td>aac934f2eed758d4a27562dae4e9c5415ff4cdb7</td><td>TS-LSTM and Temporal-Inception: +<br/>Exploiting Spatiotemporal Dynamics for Activity Recognition +<br/><b>Georgia Institute of Technology</b><br/>2Georgia Tech Research Institution +</td><td>('7437104', 'Chih-Yao Ma', 'chih-yao ma')<br/>('1960668', 'Min-Hung Chen', 'min-hung chen')<br/>('1746245', 'Zsolt Kira', 'zsolt kira')</td><td>{cyma, cmhungsteve, zkira, alregib}@gatech.edu +</td></tr><tr><td>aa331fe378056b6d6031bb8fe6676e035ed60d6d</td><td></td><td></td><td></td></tr><tr><td>aae0e417bbfba701a1183d3d92cc7ad550ee59c3</td><td>844 +<br/>A Statistical Method for 2-D Facial Landmarking +</td><td>('1764521', 'Albert Ali Salah', 'albert ali salah')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td></td></tr><tr><td>aa577652ce4dad3ca3dde44f881972ae6e1acce7</td><td>Deep Attribute Networks +<br/>Department of EE, KAIST +<br/>Daejeon, South Korea +<br/>Department of EE, KAIST +<br/>Daejeon, South Korea +<br/>Department of EE, KAIST +<br/>Daejeon, South Korea +<br/>Department of EE, KAIST +<br/>Daejeon, South Korea +</td><td>('8270717', 'Junyoung Chung', 'junyoung chung')<br/>('2350325', 'Donghoon Lee', 'donghoon lee')<br/>('2397884', 'Youngjoo Seo', 'youngjoo seo')<br/>('5578091', 'Chang D. Yoo', 'chang d. yoo')</td><td>jych@kaist.ac.kr +<br/>iamdh@kaist.ac.kr +<br/>minerrba@kaist.ac.kr +<br/>cdyoo@ee.kaist.ac.kr +</td></tr><tr><td>aa3c9de34ef140ec812be85bb8844922c35eba47</td><td>Reducing Gender Bias Amplification using Corpus-level Constraints +<br/>Men Also Like Shopping: +<br/><b>University of Virginia</b><br/><b>University of Washington</b></td><td>('3456473', 'Tianlu Wang', 'tianlu wang')<br/>('2064210', 'Mark Yatskar', 'mark yatskar')<br/>('33524946', 'Jieyu Zhao', 'jieyu zhao')<br/>('2782886', 'Kai-Wei Chang', 'kai-wei chang')<br/>('2004053', 'Vicente Ordonez', 'vicente ordonez')</td><td>{jz4fu, tw8cb, vicente, kc2wc}@virginia.edu +<br/>my89@cs.washington.edu +</td></tr><tr><td>aa94f214bb3e14842e4056fdef834a51aecef39c</td><td>Reconhecimento de padrões faciais: Um estudo +<br/>Universidade Federal +<br/>Rural do Semi-Árido +<br/>Departamento de Ciências Naturais +<br/>Mossoró, RN - 59625-900 +<br/>Resumo—O reconhecimento facial tem sido utilizado em di- +<br/>versas áreas para identificação e autenticação de usuários. Um +<br/>dos principais mercados está relacionado a segurança, porém há +<br/>uma grande variedade de aplicações relacionadas ao uso pessoal, +<br/>conveniência, aumento de produtividade, etc. O rosto humano +<br/>possui um conjunto de padrões complexos e mutáveis. Para +<br/>reconhecer esses padrões, são necessárias técnicas avançadas de +<br/>reconhecimento de padrões capazes, não apenas de reconhecer, +<br/>mas de se adaptar às mudanças constantes das faces das pessoas. +<br/>Este documento apresenta um método de reconhecimento facial +<br/>proposto a partir da análise comparativa de trabalhos encontra- +<br/>dos na literatura. +<br/>biométrica é o uso da biometria para reconhecimento, identi- +<br/>ficação ou verificação, de um ou mais traços biométricos de +<br/>um indivíduo com o objetivo de autenticar sua identidade. Os +<br/>traços biométricos são os atributos analisados pelas técnicas +<br/>de reconhecimento biométrico. +<br/>A tarefa de reconhecimento facial é composta por três +<br/>processos distintos: Registro, verificação e identificação bio- +<br/>métrica. Os processos se diferenciam pela forma de determinar +<br/>a identidade de um indivíduo. Na Figura 1 são descritos os +<br/>processos de registro, verificação e identificação biométrica. +<br/>I. INTRODUÇÃO +<br/>Biometria é a ciência que estabelece a identidade de um +<br/>indivíduo baseada em seus atributos físicos, químicos ou +<br/>comportamentais [1]. Possui inúmeras aplicações em diver- +<br/>sas áreas, se destacando mais na área de segurança, como +<br/>por exemplo sistemas de gerenciamento de identidade, cuja +<br/>funcionalidade é autenticar a identidade de um indivíduo no +<br/>contexto de uma aplicação. +<br/>O reconhecimento facial é uma técnica biométrica que +<br/>consiste em identificar padrões em características faciais como +<br/>formato da boca, do rosto, distância dos olhos, entre outros. +<br/>Um humano é capaz de reconhecer uma pessoa familiar +<br/>mesmo com muitos obstáculos com distância, sombras ou +<br/>apenas a visão parcial do rosto. Uma máquina, no entanto, +<br/>precisa realizar inúmeros processos para detectar e reconhecer +<br/>um conjunto de padrões específicos para rotular uma face +<br/>como conhecida ou desconhecida. Para isso, exitem métodos +<br/>capazes de detectar, extrair e classificar as características +<br/>faciais, fornecendo um reconhecimento automático de pessoas. +<br/>II. RECONHECIMENTO FACIAL +<br/>A tecnologia biométrica oferece vantagens em relação a +<br/>outros métodos tradicionais de identificação como senhas, +<br/>documentos e tokens. Entre elas estão o fato de que os +<br/>traços biométricos não podem ser perdidos ou esquecidos, são +<br/>difíceis de serem copiados, compartilhados ou distribuídos. Os +<br/>métodos requerem que a pessoa autenticada esteja presente +<br/>na hora e lugar da autenticação, evitando que pessoas má +<br/>intencionadas tenham acesso sem autorização. +<br/>A autenticação é o ato de estabelecer ou confirmar alguém, +<br/>ou alguma coisa, como autêntico, isto é, que as alegações +<br/>feitas por ou sobre a coisa é verdadeira [2]. Autenticação +<br/>(a) +<br/>(b) +<br/>(c) +<br/>Figura 1: Registro biométrico (a), identificação biométrica (b) +<br/>e verificação biométrica (c) +<br/>A Figura 1a descreve o processo de registro de dados +</td><td>('2545499', 'Marcos Evandro Cintra', 'marcos evandro cintra')</td><td>Email: alexdemise@gmail.com, mecintra@gmail.com +</td></tr><tr><td>aac101dd321e6d2199d8c0b48c543b541c181b66</td><td>USING CONTEXT TO ENHANCE THE +<br/>UNDERSTANDING OF FACE IMAGES +<br/>A Dissertation Presented +<br/>by +<br/>VIDIT JAIN +<br/>Submitted to the Graduate School of the +<br/><b>University of Massachusetts Amherst in partial ful llment</b><br/>of the requirements for the degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>September 2010 +<br/>Department of Computer Science +</td><td></td><td></td></tr><tr><td>af8fe1b602452cf7fc9ecea0fd4508ed4149834e</td><td></td><td></td><td></td></tr><tr><td>af6e351d58dba0962d6eb1baf4c9a776eb73533f</td><td>How to Train Your Deep Neural Network with +<br/>Dictionary Learning +<br/>*IIIT Delhi +<br/>Okhla Phase 3 +<br/>Delhi, 110020, India +<br/>+IIIT Delhi +<br/>Okhla Phase 3 +<br/>#IIIT Delhi +<br/>Okhla Phase 3 +<br/>Delhi, 110020, India +<br/>Delhi, 110020, India +</td><td>('30255052', 'Vanika Singhal', 'vanika singhal')<br/>('38608015', 'Shikha Singh', 'shikha singh')<br/>('2641605', 'Angshul Majumdar', 'angshul majumdar')</td><td>vanikas@iiitd.ac.in +<br/>shikhas@iiitd.ac.in +<br/>angshul@iiitd.ac.in +</td></tr><tr><td>aff92784567095ee526a705e21be4f42226bbaab</td><td>Face Recognition in Uncontrolled +<br/>Environments +<br/>A dissertation submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>at +<br/><b>University College London</b><br/>Department of Computer Science +<br/><b>University College London</b><br/>May 26, 2015 +</td><td>('38098063', 'Yun Fu', 'yun fu')</td><td></td></tr><tr><td>aff8705fb2f2ae460cb3980b47f2e85c2e6dd41a</td><td>Attributes in Multiple Facial Images +<br/><b>West Virginia University, Morgantown</b><br/>WV 26506, USA +</td><td>('1767347', 'Xudong Liu', 'xudong liu')<br/>('1822413', 'Guodong Guo', 'guodong guo')</td><td>xdliu@mix.wvu.edu, guodong.guo@mail.wvu.edu +</td></tr><tr><td>af13c355a2a14bb74847aedeafe990db3fc9cbd4</td><td>Happy and Agreeable? Multi-Label Classification of +<br/>Impressions in Social Video +<br/><b>Idiap Research Institute</b><br/>Switzerland +<br/>Instituto Potosino de +<br/>Investigación Científica y +<br/>Tecnológica +<br/>Mexico +<br/><b>Idiap Research Institute</b><br/>École Polytechnique Fédérale +<br/>de Lausanne +<br/>Switzerland +</td><td>('2389354', 'Gilberto Chávez-Martínez', 'gilberto chávez-martínez')<br/>('1934619', 'Salvador Ruiz-Correa', 'salvador ruiz-correa')<br/>('1698682', 'Daniel Gatica-Perez', 'daniel gatica-perez')</td><td>gchavez@idiap.ch +<br/>src@cmls.pw +<br/>gatica@idiap.ch +</td></tr><tr><td>af6cae71f24ea8f457e581bfe1240d5fa63faaf7</td><td></td><td></td><td></td></tr><tr><td>af62621816fbbe7582a7d237ebae1a4d68fcf97d</td><td>International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 +<br/>International Conference on Humming Bird ( 01st March 2014) +<br/>RESEARCH ARTICLE +<br/> OPEN ACCESS +<br/>Active Shape Model Based Recognition Of Facial Expression +<br/>AncyRija V , Gayathri. S2 +<br/><b>AncyRijaV, Author is currently pursuing M.E (Software Engineering) in Vins Christian College of</b><br/>Engineering, +<br/><b>Gayathri.S, M.E., Vins Christian college of Engineering</b></td><td></td><td>e-mail: ancyrija@gmail.com. +</td></tr><tr><td>afdf9a3464c3b015f040982750f6b41c048706f5</td><td>A Recurrent Encoder-Decoder Network for Sequential Face Alignment +<br/><b>Rutgers University</b><br/>Rogerio Feris +<br/>IBM T. J. Watson +<br/>Snapchat Research +<br/>Dimitris Metaxas +<br/><b>Rutgers University</b></td><td>('4340744', 'Xi Peng', 'xi peng')<br/>('48631738', 'Xiaoyu Wang', 'xiaoyu wang')</td><td>xipeng.cs@rutgers.edu +<br/>rsferis@us.ibm.com +<br/>fanghuaxue@gmail.com +<br/>dnm@cs.rutgers.edu +</td></tr><tr><td>af54dd5da722e104740f9b6f261df9d4688a9712</td><td></td><td></td><td></td></tr><tr><td>afa57e50570a6599508ee2d50a7b8ca6be04834a</td><td>Motion in action : optical flow estimation and action +<br/>localization in videos +<br/>To cite this version: +<br/>Computer Vision and Pattern Recognition [cs.CV]. Université Grenoble Alpes, 2016. English. <NNT : +<br/>2016GREAM013>. <tel-01407258> +<br/>HAL Id: tel-01407258 +<br/>https://tel.archives-ouvertes.fr/tel-01407258 +<br/>Submitted on 1 Dec 2016 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('2492127', 'Philippe Weinzaepfel', 'philippe weinzaepfel')<br/>('2492127', 'Philippe Weinzaepfel', 'philippe weinzaepfel')</td><td></td></tr><tr><td>afe9cfba90d4b1dbd7db1cf60faf91f24d12b286</td><td>Principal Directions of Synthetic Exact Filters +<br/>for Robust Real-Time Eye Localization +<br/>Vitomir ˇStruc1;2, Jerneja ˇZganec Gros1, and Nikola Paveˇsi´c2 +<br/>1 Alpineon Ltd, Ulica Iga Grudna 15, SI-1000 Ljubljana, Slovenia, +<br/><b>Faculty of Electrical Engineering, University of Ljubljana, Tr za ska cesta</b><br/>SI-1000 Ljubljana, Slovenia, +</td><td></td><td>fvitomir.struc, jerneja.grosg@alpineon.com, +<br/>fvitomir.struc, nikola.pavesicg@fe.uni-lj.si +</td></tr><tr><td>afa84ff62c9f5b5c280de2996b69ad9fa48b7bc3</td><td>Two-stream Flow-guided Convolutional Attention Networks for Action +<br/>Recognition +<br/><b>National University of Singapore</b><br/>Loong-Fah Cheong +</td><td>('25205026', 'An Tran', 'an tran')</td><td>an.tran@u.nus.edu +<br/>eleclf@nus.edu.sg +</td></tr><tr><td>af278274e4bda66f38fd296cfa5c07804fbc26ee</td><td>RESEARCH ARTICLE +<br/>A Novel Maximum Entropy Markov Model for +<br/>Human Facial Expression Recognition +<br/><b>College of Information and Communication Engineering, Sungkyunkwan University, Suwon-si, Gyeonggi</b><br/><b>do, Rep. of Korea, Kyung Hee University, Suwon, Rep. of Korea</b><br/><b>Innopolis University, Kazan, Russia</b><br/>a11111 +<br/>☯ These authors contributed equally to this work. +</td><td>('1711083', 'Muhammad Hameed Siddiqi', 'muhammad hameed siddiqi')<br/>('2401685', 'Md. Golam Rabiul Alam', 'md. golam rabiul alam')<br/>('1683244', 'Choong Seon Hong', 'choong seon hong')<br/>('1734679', 'Hyunseung Choo', 'hyunseung choo')</td><td>* choo@skku.edu +</td></tr><tr><td>af654a7ec15168b16382bd604889ea07a967dac6</td><td>FACE RECOGNITION COMMITTEE MACHINE +<br/>Department of Computer Science and Engineering +<br/><b>The Chinese University of Hong Kong</b><br/>Shatin, Hong Kong +</td><td>('2899702', 'Ho-Man Tang', 'ho-man tang')<br/>('1681775', 'Michael R. Lyu', 'michael r. lyu')<br/>('1706259', 'Irwin King', 'irwin king')</td><td> +</td></tr><tr><td>afc7092987f0d05f5685e9332d83c4b27612f964</td><td>Person-Independent Facial Expression Detection using Constrained +<br/>Local Models +</td><td>('1713496', 'Patrick Lucey', 'patrick lucey')<br/>('1820249', 'Simon Lucey', 'simon lucey')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')<br/>('1729760', 'Sridha Sridharan', 'sridha sridharan')</td><td></td></tr><tr><td>b730908bc1f80b711c031f3ea459e4de09a3d324</td><td>2024 +<br/>Active Orientation Models for Face +<br/>Alignment In-the-Wild +</td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('2575567', 'Joan Alabort-i-Medina', 'joan alabort-i-medina')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>b7426836ca364603ccab0e533891d8ac54cf2429</td><td>Hindawi +<br/>Journal of Healthcare Engineering +<br/>Volume 2017, Article ID 3090343, 31 pages +<br/>https://doi.org/10.1155/2017/3090343 +<br/>Review Article +<br/>A Review on Human Activity Recognition Using +<br/>Vision-Based Method +<br/><b>College of Information Science and Engineering, Ocean University of China, Qingdao, China</b><br/><b>Tsinghua University, Beijing, China</b><br/>Received 22 February 2017; Accepted 11 June 2017; Published 20 July 2017 +<br/>Academic Editor: Dong S. Park +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Human activity recognition (HAR) aims to recognize activities from a series of observations on the actions of subjects and the +<br/>environmental conditions. The vision-based HAR research is the basis of many applications including video surveillance, health +<br/>care, and human-computer interaction (HCI). This review highlights the advances of state-of-the-art activity recognition +<br/>approaches, especially for the activity representation and classification methods. For the representation methods, we sort out a +<br/>chronological research trajectory from global representations to local representations, and recent depth-based representations. +<br/>For the classification methods, we conform to the categorization of template-based methods, discriminative models, and +<br/>generative models and review several prevalent methods. Next, representative and available datasets are introduced. Aiming to +<br/>provide an overview of those methods and a convenient way of comparing them, we classify existing literatures with a detailed +<br/>taxonomy including representation and classification methods, as well as the datasets they used. Finally, we investigate the +<br/>directions for future research. +<br/>1. Introduction +<br/>Human activity recognition (HAR) is a widely studied com- +<br/>puter vision problem. Applications of HAR include video +<br/>surveillance, health care, and human-computer interaction. +<br/>As the imaging technique advances and the camera device +<br/>upgrades, novel approaches for HAR constantly emerge. This +<br/>review aims to provide a comprehensive introduction to the +<br/>video-based human activity recognition, giving an overview +<br/>of various approaches as well as their evolutions by covering +<br/>both the representative classical literatures and the state-of- +<br/>the-art approaches. +<br/>Human activities have an inherent hierarchical structure +<br/>that indicates the different levels of it, which can be consid- +<br/>ered as a three-level categorization. First, for the bottom level, +<br/>there is an atomic element and these action primitives consti- +<br/>tute more complex human activities. After the action primi- +<br/>tive level, the action/activity comes as the second level. +<br/>Finally, the complex interactions form the top level, which +<br/>refers to the human activities that involve more than two +<br/>persons and objects. In this paper, we follow this three-level +<br/>categorization namely action primitives, actions/activities, +<br/>and interactions. This three-level categorization varies a little +<br/>from previous surveys [1–4] and maintains a consistent +<br/>theme. Action primitives are those atomic actions at the limb +<br/>level, such as “stretching the left arm,” and “raising the right +<br/>leg.” Atomic actions are performed by a specific part of the +<br/>human body, such as the hands, arms, or upper body part +<br/>[4]. Actions and activities are used interchangeably in this +<br/>review, referring to the whole-body movements composed +<br/>of several action primitives in temporal sequential order +<br/>and performed by a single person with no more person or +<br/>additional objects. Specifically, we refer the terminology +<br/>human activities as all movements of the three layers and +<br/>the activities/actions as the middle level of human activities. +<br/>Human activities like walking, running, and waving hands +<br/>are categorized in the actions/activities level. Finally, similar +<br/>to Aggarwal et al.’s review [2], interactions are human activ- +<br/>ities that involve two or more persons and objects. The +<br/>additional person or object is an important characteristic of +</td><td>('7671146', 'Shugang Zhang', 'shugang zhang')<br/>('39868595', 'Zhiqiang Wei', 'zhiqiang wei')<br/>('2896895', 'Jie Nie', 'jie nie')<br/>('40284611', 'Lei Huang', 'lei huang')<br/>('40658604', 'Shuang Wang', 'shuang wang')<br/>('40166799', 'Zhen Li', 'zhen li')<br/>('7671146', 'Shugang Zhang', 'shugang zhang')</td><td>Correspondence should be addressed to Zhen Li; lizhen0130@gmail.com +</td></tr><tr><td>b73795963dc623a634d218d29e4a5b74dfbc79f1</td><td>ZHAO, YANG: IDENTITY PRESERVING FACE COMPLETION FOR LARGE OCULAR RO +<br/>Identity Preserving Face Completion for +<br/>Large Ocular Region Occlusion +<br/>1 Computer Science Department +<br/><b>University of Kentucky</b><br/>Lexington, KY, USA +<br/><b>Institute for Creative Technologies</b><br/><b>University of Southern California</b><br/>Playa Vista, California, USA +<br/>3 School of Computer Science and +<br/>Technology +<br/><b>Harbin Institute of Technology</b><br/>Harbin, China +<br/><b>Hangzhou Institute of Service</b><br/>Engineering +<br/><b>Hangzhou Normal University</b><br/>Hangzhou, China +</td><td>('2613340', 'Yajie Zhao', 'yajie zhao')<br/>('47483055', 'Weikai Chen', 'weikai chen')<br/>('1780032', 'Jun Xing', 'jun xing')<br/>('21515518', 'Xiaoming Li', 'xiaoming li')<br/>('3408065', 'Zach Bessinger', 'zach bessinger')<br/>('1752129', 'Fuchang Liu', 'fuchang liu')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('38958903', 'Ruigang Yang', 'ruigang yang')</td><td>yajie.zhao@uky.edu +<br/>wechen@ict.usc.edu +<br/>junxnui@gmail.com +<br/>hit.xmshr@gmail.com +<br/>zach.bessinger@gmail.com +<br/>20140022@hznu.edu.cn +<br/>cswmzuo@gmail.com +<br/>ryang@cs.uky.edu +</td></tr><tr><td>b7cf7bb574b2369f4d7ebc3866b461634147041a</td><td>Neural Comput & Applic (2012) 21:1575–1583 +<br/>DOI 10.1007/s00521-011-0728-x +<br/>O R I G I N A L A R T I C L E +<br/>From NLDA to LDA/GSVD: a modified NLDA algorithm +<br/>Received: 2 August 2010 / Accepted: 3 August 2011 / Published online: 19 August 2011 +<br/>Ó Springer-Verlag London Limited 2011 +</td><td>('1692984', 'Jun Yin', 'jun yin')</td><td></td></tr><tr><td>b750b3d8c34d4e57ecdafcd5ae8a15d7fa50bc24</td><td>Unified Solution to Nonnegative Data Factorization Problems +<br/><b>Huazhong University of Science and Technology, Wuhan, China</b><br/><b>National University of Singapore, Singapore</b></td><td>('1817910', 'Xiaobai Liu', 'xiaobai liu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('2156156', 'Hai Jin', 'hai jin')</td><td></td></tr><tr><td>b7894c1f805ffd90ab4ab06002c70de68d6982ab</td><td>Biomedical Research 2017; Special Issue: S610-S618 +<br/>ISSN 0970-938X +<br/>www.biomedres.info +<br/>A comprehensive age estimation on face images using hybrid filter based +<br/>feature extraction. +<br/>Karthikeyan D1*, Balakrishnan G2 +<br/><b>Srinivasan Engineering College, Perambalur, India</b><br/><b>Indra Ganesan College of Engineering, Trichy, India</b></td><td></td><td></td></tr><tr><td>b7eead8586ffe069edd190956bd338d82c69f880</td><td>A VIDEO DATABASE FOR FACIAL +<br/>BEHAVIOR UNDERSTANDING +<br/>D. Freire-Obreg´on and M. Castrill´on-Santana. +<br/>SIANI, Universidad de Las Palmas de Gran Canaria, Spain +</td><td></td><td>dfreire@iusiani.ulpgc.es, mcastrillon@iusiani.ulpgc.es +</td></tr><tr><td>b75cee96293c11fe77ab733fc1147950abbe16f9</td><td></td><td></td><td></td></tr><tr><td>b7774c096dc18bb0be2acef07ff5887a22c2a848</td><td>Distance metric learning for image and webpage +<br/>comparison +<br/>To cite this version: +<br/>versité Pierre et Marie Curie - Paris VI, 2015. English. <NNT : 2015PA066019>. <tel-01135698v2> +<br/>HAL Id: tel-01135698 +<br/>https://tel.archives-ouvertes.fr/tel-01135698v2 +<br/>Submitted on 18 Mar 2015 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('32868306', 'Marc Teva Law', 'marc teva law')<br/>('32868306', 'Marc Teva Law', 'marc teva law')</td><td></td></tr><tr><td>b7f05d0771da64192f73bdb2535925b0e238d233</td><td> MVA2005 IAPR Conference on Machine VIsion Applications, May 16-18, 2005 Tsukuba Science City, Japan +<br/>4-3 +<br/>Robust Active Shape Model using AdaBoosted Histogram Classifiers +<br/>W ataru Ito +<br/>Imaging Software Technology Center +<br/>Imaging Software Technology Center +<br/>FUJI PHOTO FILM CO., LTD. +<br/>FUJI PHOTO FILM CO., LTD. +</td><td>('1724928', 'Yuanzhong Li', 'yuanzhong li')</td><td>li_yuanzhong@ fujifilm.co.jp +<br/>wataru_ito@ fujifilm.co.jp +</td></tr><tr><td>b755505bdd5af078e06427d34b6ac2530ba69b12</td><td>To appear in the International Joint Conf. Biometrics, Washington D.C., October, 2011 +<br/>NFRAD: Near-Infrared Face Recognition at a Distance +<br/>aDept. of Brain and Cognitive Eng. Korea Univ., Seoul, Korea +<br/>bDept. of Comp. Sci. & Eng. Michigan State Univ., E. Lansing, MI, USA 48824 +</td><td>('2429013', 'Hyunju Maeng', 'hyunju maeng')<br/>('2131755', 'Hyun-Cheol Choi', 'hyun-cheol choi')<br/>('2222919', 'Unsang Park', 'unsang park')<br/>('1703007', 'Seong-Whan Lee', 'seong-whan lee')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>{hjmaeng, hcchoi}@korea.ac.kr, parkunsa@cse.msu.edu, swlee@image.korea.ac.kr , jain@cse.msu.edu +</td></tr><tr><td>b7820f3d0f43c2ce613ebb6c3d16eb893c84cf89</td><td>Visual Data Synthesis via GAN for Zero-Shot Video Classification +<br/><b>Institute of Computer Science and Technology, Peking University</b><br/>Beijing 100871, China +</td><td>('2439211', 'Chenrui Zhang', 'chenrui zhang')<br/>('1704081', 'Yuxin Peng', 'yuxin peng')</td><td>pengyuxin@pku.edu.cn +</td></tr><tr><td>b7b461f82c911f2596b310e2b18dd0da1d5d4491</td><td>2961 +<br/>2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +<br/>978-1-4799-2893-4/14/$31.00 ©2014 IEEE +<br/>K-MAPPINGS AND REGRESSION TREES +<br/><b>SAMSI and Duke University</b><br/>1. INTRODUCTION +<br/>argminM1,...,MK +<br/>P1,...PK +<br/>2.1. Partitioning Y +<br/>K(cid:2) +<br/>(cid:2) +<br/>(cid:3) +<br/>(cid:4) +</td><td>('3149531', 'Arthur Szlam', 'arthur szlam')</td><td></td></tr><tr><td>b73fdae232270404f96754329a1a18768974d3f6</td><td></td><td></td><td></td></tr><tr><td>b76af8fcf9a3ebc421b075b689defb6dc4282670</td><td>Face Mask Extraction in Video Sequence +</td><td>('2563750', 'Yujiang Wang', 'yujiang wang')</td><td></td></tr><tr><td>b7c5f885114186284c51e863b58292583047a8b4</td><td>GAdaBoost: Accelerating Adaboost Feature Selection with Genetic +<br/>Algorithms +<br/><b>The American University In Cairo, Road 90, New Cairo, Cairo, Egypt</b><br/>Keywords: +<br/>Object Detection, Genetic Algorithms, Haar Features, Adaboost, Face Detection. +</td><td>('3468033', 'Mai F. Tolba', 'mai f. tolba')<br/>('27045559', 'Mohamed Moustafa', 'mohamed moustafa')</td><td>maitolba@aucegypt.edu, m.moustafa@aucegypt.edu +</td></tr><tr><td>b73d9e1af36aabb81353f29c40ecdcbdf731dbed</td><td>Sensors 2015, 15, 20945-20966; doi:10.3390/s150920945 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>Head Pose Estimation on Top of Haar-Like Face Detection: +<br/>A Study Using the Kinect Sensor +<br/><b>Institute for Information Technology and Communications (IIKT), Otto-von-Guericke-University</b><br/><b>College of Computer Science and Information Sciences</b><br/><b>College of Science, Menou a University, Menou a 32721, Egypt</b><br/>Tel.: +49-391-67-11033; Fax: +49-391-67-11231. +<br/>Academic Editor: Vittorio M. N. Passaro +<br/>Received: 3 July 2015 / Accepted: 6 August 2015 / Published: 26 August 2015 +</td><td>('2712124', 'Anwar Saeed', 'anwar saeed')<br/>('1741165', 'Ayoub Al-Hamadi', 'ayoub al-hamadi')<br/>('1889194', 'Ahmed Ghoneim', 'ahmed ghoneim')</td><td>Magdeburg, Magdeburg D-39016, Germany; E-Mail: Ayoub.Al-Hamadi@ovgu.de +<br/>King Saud University, Riyadh 11451, Saudi Arabia; E-Mail: ghoneim@KSU.EDU.SA +<br/>* Author to whom correspondence should be addressed; E-Mail: anwar.saeed@ovgu.de; +</td></tr><tr><td>b747fcad32484dfbe29530a15776d0df5688a7db</td><td></td><td></td><td></td></tr><tr><td>b7f7a4df251ff26aca83d66d6b479f1dc6cd1085</td><td>Bouges et al. EURASIP Journal on Image and Video Processing 2013, 2013:55 +<br/>http://jivp.eurasipjournals.com/content/2013/1/55 +<br/>RESEARCH +<br/>Open Access +<br/>Handling missing weak classifiers in boosted +<br/>cascade: application to multiview and +<br/>occluded face detection +</td><td>('3212236', 'Pierre Bouges', 'pierre bouges')<br/>('1865978', 'Thierry Chateau', 'thierry chateau')<br/>('32323470', 'Christophe Blanc', 'christophe blanc')<br/>('1685767', 'Gaëlle Loosli', 'gaëlle loosli')</td><td></td></tr><tr><td>db848c3c32464d12da33b2f4c3a29fe293fc35d1</td><td>Pose Guided Human Video Generation +<br/>1 CUHK-SenseTime Joint Lab, CUHK, Hong Kong S.A.R. +<br/>2 SenseTime Research, Beijing, China +<br/><b>Carnegie Mellon University</b></td><td>('49984891', 'Ceyuan Yang', 'ceyuan yang')<br/>('1915826', 'Zhe Wang', 'zhe wang')<br/>('22689408', 'Xinge Zhu', 'xinge zhu')<br/>('2000034', 'Chen Huang', 'chen huang')<br/>('1788070', 'Jianping Shi', 'jianping shi')<br/>('1807606', 'Dahua Lin', 'dahua lin')</td><td>yangceyuan@sensetime.com +</td></tr><tr><td>db1f48a7e11174d4a724a4edb3a0f1571d649670</td><td>Joint Constrained Clustering and Feature +<br/>Learning based on Deep Neural Networks +<br/>by +<br/><b>B.Sc., University of Science and Technology of China</b><br/>Thesis Submitted in Partial Fulfillment of the +<br/>Requirements for the Degree of +<br/>Master of Science +<br/>in the +<br/>School of Computing Science +<br/>Faculty of Applied Sciences +<br/><b>SIMON FRASER UNIVERSITY</b><br/>Summer 2017 +<br/>However, in accordance with the Copyright Act of Canada, this work may be +<br/>reproduced without authorization under the conditions for “Fair Dealing.” +<br/>Therefore, limited reproduction of this work for the purposes of private study, +<br/>research, education, satire, parody, criticism, review and news reporting is likely +<br/>All rights reserved. +<br/>to be in accordance with the law, particularly if cited appropriately. +</td><td>('1707706', 'Xiaoyu Liu', 'xiaoyu liu')<br/>('1707706', 'Xiaoyu Liu', 'xiaoyu liu')</td><td></td></tr><tr><td>db227f72bb13a5acca549fab0dc76bce1fb3b948</td><td>International Refereed Journal of Engineering and Science (IRJES) +<br/>ISSN (Online) 2319-183X, (Print) 2319-1821 +<br/>Volume 4, Issue 6 (June 2015), PP.169-169-174 +<br/>Characteristic Based Image Search using Re-Ranking method +<br/>1Chitti Babu, 2Yasmeen Jaweed, 3G.Vijay Kumar +<br/><b></b></td><td></td><td></td></tr><tr><td>dbb16032dd8f19bdfd045a1fc0fc51f29c70f70a</td><td>PARKHI et al.: DEEP FACE RECOGNITION +<br/>Deep Face Recognition +<br/>Visual Geometry Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b></td><td>('3188342', 'Omkar M. Parkhi', 'omkar m. parkhi')<br/>('1687524', 'Andrea Vedaldi', 'andrea vedaldi')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>omkar@robots.ox.ac.uk +<br/>vedaldi@robots.ox.ac.uk +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>dbaf89ca98dda2c99157c46abd136ace5bdc33b3</td><td>Nonlinear Cross-View Sample Enrichment for +<br/>Action Recognition +<br/>Institut Mines-T´el´ecom; T´el´ecom ParisTech; CNRS LTCI +</td><td>('1695223', 'Ling Wang', 'ling wang')<br/>('1692389', 'Hichem Sahbi', 'hichem sahbi')</td><td></td></tr><tr><td>dbab6ac1a9516c360cdbfd5f3239a351a64adde7</td><td></td><td></td><td></td></tr><tr><td>dbe255d3d2a5d960daaaba71cb0da292e0af36a7</td><td>Evolutionary Cost-sensitive Extreme Learning +<br/>Machine +<br/>1 +</td><td>('36904370', 'Lei Zhang', 'lei zhang')</td><td></td></tr><tr><td>dbb0a527612c828d43bcb9a9c41f1bf7110b1dc8</td><td>Chapter 7 +<br/>Machine Learning Techniques +<br/>for Face Analysis +</td><td>('9301018', 'Roberto Valenti', 'roberto valenti')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1695527', 'Theo Gevers', 'theo gevers')<br/>('1774778', 'Ira Cohen', 'ira cohen')</td><td></td></tr><tr><td>db5a00984fa54b9d2a1caad0067a9ff0d0489517</td><td>Multi-Task Adversarial Network for Disentangled Feature Learning +<br/>Ian Wassell1 +<br/><b>University of Cambridge</b><br/>2Adobe Research +</td><td>('49421489', 'Yang Liu', 'yang liu')<br/>('48707577', 'Zhaowen Wang', 'zhaowen wang')</td><td>1{yl504,ijw24}@cam.ac.uk +<br/>2{zhawang,hljin}@adobe.com +</td></tr><tr><td>dbd958ffedc3eae8032be67599ec281310c05630</td><td>Automated Restyling of Human Portrait Based on Facial Expression Recognition +<br/>and 3D Reconstruction +<br/><b>Stanford University</b><br/>350 Serra Mall, Stanford, CA 94305, USA +</td><td>('46740443', 'Cheng-Han Wu', 'cheng-han wu')</td><td>1chw0208@stanford.edu +<br/>2hsinc@stanford.edu +</td></tr><tr><td>dbed26cc6d818b3679e46677abc9fa8e04e8c6a6</td><td>A Hierarchical Generative Model for Eye Image Synthesis and Eye Gaze +<br/>Estimation +<br/><b>ECSE, Rensselaer Polytechnic Institute, Troy, NY, USA</b></td><td>('1771700', 'Kang Wang', 'kang wang')<br/>('49832825', 'Rui Zhao', 'rui zhao')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>{wangk10, zhaor, jiq}@rpi.edu +</td></tr><tr><td>db3545a983ffd24c97c18bf7f068783102548ad7</td><td>Enriching the Student Model in an +<br/>Intelligent Tutoring System +<br/>Submitted in partial fulfillment of the requirements for the degree +<br/>of Doctor of Philosophy +<br/>of the +<br/><b>Indian Institute of Technology, Bombay, India</b><br/>and +<br/><b>Monash University, Australia</b><br/>by +<br/>Supervisors: +<br/>The course of study for this award was developed jointly by +<br/><b>the Indian Institute of Technology, Bombay and Monash University, Australia</b><br/>and given academic recognition by each of them. +<br/>The programme was administered by The IITB-Monash Research Academy. +<br/>2014 +</td><td>('2844237', 'Ramkumar Rajendran', 'ramkumar rajendran')<br/>('1946438', 'Sridhar Iyer', 'sridhar iyer')<br/>('1791910', 'Sahana Murthy', 'sahana murthy')<br/>('38751653', 'Campbell Wilson', 'campbell wilson')<br/>('1727078', 'Judithe Sheard', 'judithe sheard')</td><td></td></tr><tr><td>dba493caf6647214c8c58967a8251641c2bda4c2</td><td>Automatic 3D Facial Expression Editing in Videos +<br/><b>University of California, Santa Barbara</b><br/>2IMPA – Instituto de Matematica Pura e Aplicada +</td><td>('13303219', 'Ya Chang', 'ya chang')<br/>('2428542', 'Marcelo Vieira', 'marcelo vieira')<br/>('1752714', 'Matthew Turk', 'matthew turk')<br/>('1705620', 'Luiz Velho', 'luiz velho')</td><td></td></tr><tr><td>dbb7f37fb9b41d1aa862aaf2d2e721a470fd2c57</td><td>Face Image Analysis With +<br/>Convolutional Neural Networks +<br/>Dissertation +<br/>Zur Erlangung des Doktorgrades +<br/>der Fakult¨at f¨ur Angewandte Wissenschaften +<br/>an der Albert-Ludwigs-Universit¨at Freiburg im Breisgau +<br/>von +<br/>Stefan Duffner +<br/>2007 +</td><td></td><td></td></tr><tr><td>db36e682501582d1c7b903422993cf8d70bb0b42</td><td>Deep Trans-layer Unsupervised Networks for +<br/>Representation Learning +<br/>aKey Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>bSchool of Computer and Control Engineering, University of Chinese Academy of Sciences</b><br/>Beijing 100049, China +</td><td>('1778018', 'Wentao Zhu', 'wentao zhu')<br/>('35048816', 'Jun Miao', 'jun miao')<br/>('2343895', 'Laiyun Qing', 'laiyun qing')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>dbe0e533d715f8543bcf197f3b8e5cffa969dfc0</td><td>International Journal of Advanced Research in Electrical, +<br/>Electronics and Instrumentation Engineering +<br/> ISSN (Print) : 2320 – 3765 +<br/> ISSN (Online): 2278 – 8875 +<br/>(An ISO 3297: 2007 Certified Organization) +<br/>Vol. 3, Issue 5, May 2014 +<br/>A Comprehensive Comparative Performance +<br/>Analysis of Eigenfaces, Laplacianfaces and +<br/>Orthogonal Laplacianfaces for Face Recognition +<br/><b>UG student, Amity school of Engineering and Technology, Amity University, Haryana, India</b><br/><b>Lecturer, Amity school of Engineering and Technology, Amity University, Haryana, India</b></td><td></td><td></td></tr><tr><td>dbd5e9691cab2c515b50dda3d0832bea6eef79f2</td><td>Image-basedFaceRecognition:IssuesandMethods +<br/>WenYiZhao +<br/>RamaChellappa +<br/>Sarno(cid:11)Corporation +<br/>CenterforAutomationResearch +<br/> +<br/><b>UniversityofMaryland</b><br/>Princeton,NJ +<br/><b>CollegePark, MD</b></td><td></td><td>Email:wzhao@sarno(cid:11).com +<br/>Email:rama@cfar.umd.edu +</td></tr><tr><td>db67edbaeb78e1dd734784cfaaa720ba86ceb6d2</td><td>SPECFACE - A Dataset of Human Faces Wearing Spectacles +<br/><b>Indian Institute of Technology Kharagpur</b><br/>India +</td><td>('30654921', 'Anirban Dasgupta', 'anirban dasgupta')<br/>('30572870', 'Shubhobrata Bhattacharya', 'shubhobrata bhattacharya')<br/>('2680543', 'Aurobinda Routray', 'aurobinda routray')</td><td></td></tr><tr><td>db82f9101f64d396a86fc2bd05b352e433d88d02</td><td>A Spatio-Temporal Probabilistic Framework for +<br/>Dividing and Predicting Facial Action Units +<br/><b>Electrical and Computer Engineering, The University of Memphis</b></td><td>('2497319', 'Md. Iftekhar Tanveer', 'md. iftekhar tanveer')<br/>('1828610', 'Mohammed Yeasin', 'mohammed yeasin')</td><td></td></tr><tr><td>db428d03e3dfd98624c23e0462817ad17ef14493</td><td>Oxford TRECVID 2006 – Notebook paper +<br/>Department of Engineering Science +<br/><b>University of Oxford</b><br/>United Kingdom +</td><td>('2276542', 'James Philbin', 'james philbin')<br/>('8873555', 'Anna Bosch', 'anna bosch')<br/>('1720149', 'Jan-Mark Geusebroek', 'jan-mark geusebroek')<br/>('1782755', 'Josef Sivic', 'josef sivic')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>a83fc450c124b7e640adc762e95e3bb6b423b310</td><td>Deep Face Feature for Face Alignment +</td><td>('15679675', 'Boyi Jiang', 'boyi jiang')<br/>('2938279', 'Juyong Zhang', 'juyong zhang')<br/>('2964129', 'Bailin Deng', 'bailin deng')<br/>('8280113', 'Yudong Guo', 'yudong guo')<br/>('1724542', 'Ligang Liu', 'ligang liu')</td><td></td></tr><tr><td>a85e9e11db5665c89b057a124547377d3e1c27ef</td><td>Dynamics of Driver’s Gaze: Explorations in +<br/>Behavior Modeling & Maneuver Prediction +</td><td>('1841835', 'Sujitha Martin', 'sujitha martin')<br/>('22254044', 'Sourabh Vora', 'sourabh vora')<br/>('2812409', 'Kevan Yuen', 'kevan yuen')</td><td></td></tr><tr><td>a8117a4733cce9148c35fb6888962f665ae65b1e</td><td>IEEE TRANSACTIONS ON XXXX, VOL. XX, NO. XX, XX 201X +<br/>A Good Practice Towards Top Performance of Face +<br/>Recognition: Transferred Deep Feature Fusion +</td><td>('33419682', 'Lin Xiong', 'lin xiong')<br/>('1785111', 'Jayashree Karlekar', 'jayashree karlekar')<br/>('2052311', 'Jian Zhao', 'jian zhao')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')<br/>('2668358', 'Sugiri Pranata', 'sugiri pranata')<br/>('3493398', 'Shengmei Shen', 'shengmei shen')</td><td></td></tr><tr><td>a87ab836771164adb95d6744027e62e05f47fd96</td><td>Understanding human-human interactions: a survey +<br/><b>Utrecht University, Buys Ballotgebouw, Princetonplein 5, Utrecht, 3584CC, Netherlands</b><br/><b>Utrecht University, Buys Ballotgebouw, Princetonplein 5, Utrecht, 3584CC, Netherlands</b></td><td>('26936326', 'Alexandros Stergiou', 'alexandros stergiou')<br/>('1754666', 'Ronald Poppe', 'ronald poppe')</td><td></td></tr><tr><td>a896ddeb0d253739c9aaef7fc1f170a2ba8407d3</td><td>SSH: Single Stage Headless Face Detector +<br/><b>University of Maryland</b></td><td>('40465379', 'Mahyar Najibi', 'mahyar najibi')<br/>('3383048', 'Pouya Samangouei', 'pouya samangouei')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td>{pouya,rama,lsd}@umiacs.umd.edu +<br/>najibi@cs.umd.edu +</td></tr><tr><td>a820941eaf03077d68536732a4d5f28d94b5864a</td><td>Leveraging Datasets with Varying Annotations for Face Alignment +<br/>via Deep Regression Network +<br/>1Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b><br/>3CAS Center for Excellence in Brain Science and Intelligence Technology +</td><td>('1698586', 'Jie Zhang', 'jie zhang')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{jie.zhang,meina.kan,shiguang.shan,xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>a8035ca71af8cc68b3e0ac9190a89fed50c92332</td><td>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>IIIT-CFW: A Benchmark Database of +<br/>Cartoon Faces in the Wild +<br/>1 IIIT Chittoor, Sri City, India +<br/>2 CVIT, KCIS, IIIT Hyderabad, India +</td><td>('2154430', 'Ashutosh Mishra', 'ashutosh mishra')<br/>('31821293', 'Shyam Nandan Rai', 'shyam nandan rai')<br/>('39719398', 'Anand Mishra', 'anand mishra')<br/>('1694502', 'C. V. Jawahar', 'c. v. jawahar')</td><td></td></tr><tr><td>a88640045d13fc0207ac816b0bb532e42bcccf36</td><td>ARXIV VERSION +<br/>Simultaneously Learning Neighborship and +<br/>Projection Matrix for Supervised +<br/>Dimensionality Reduction +</td><td>('34116743', 'Yanwei Pang', 'yanwei pang')<br/>('2521321', 'Bo Zhou', 'bo zhou')<br/>('1688370', 'Feiping Nie', 'feiping nie')</td><td></td></tr><tr><td>a803453edd2b4a85b29da74dcc551b3c53ff17f9</td><td>Pose Invariant Face Recognition Under Arbitrary +<br/>Illumination Based on 3D Face Reconstruction +<br/><b>School of Computer Science and Technology, Harbin Institute of Technology</b><br/>150001 Harbin, China +<br/>2 ICT-ISVISION Joint R&D Lab for Face Recognition, ICT, CAS, 100080 Beijing, China +</td><td>('1695600', 'Xiujuan Chai', 'xiujuan chai')<br/>('2343895', 'Laiyun Qing', 'laiyun qing')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')<br/>('1698902', 'Wen Gao', 'wen gao')</td><td>{xjchai,xlchen,wgao}@jdl.ac.cn +<br/>{lyqing,sgshan}@jdl.ac.cn +</td></tr><tr><td>a8a30a8c50d9c4bb8e6d2dd84bc5b8b7f2c84dd8</td><td>This is a repository copy of Modelling of Orthogonal Craniofacial Profiles. +<br/>White Rose Research Online URL for this paper: +<br/>http://eprints.whiterose.ac.uk/131767/ +<br/>Version: Published Version +<br/>Article: +<br/>Dai, Hang, Pears, Nicholas Edwin orcid.org/0000-0001-9513-5634 and Duncan, Christian +<br/>(2017) Modelling of Orthogonal Craniofacial Profiles. Journal of Imaging. ISSN 2313-433X +<br/>https://doi.org/10.3390/jimaging3040055 +<br/>Reuse +<br/>This article is distributed under the terms of the Creative Commons Attribution (CC BY) licence. This licence +<br/>allows you to distribute, remix, tweak, and build upon the work, even commercially, as long as you credit the +<br/>authors for the original work. More information and the full terms of the licence here: +<br/>https://creativecommons.org/licenses/ +<br/>Takedown +<br/>If you consider content in White Rose Research Online to be in breach of UK law, please notify us by +<br/>https://eprints.whiterose.ac.uk/ +</td><td></td><td>emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request. +<br/>eprints@whiterose.ac.uk +</td></tr><tr><td>a8638a07465fe388ae5da0e8a68e62a4ee322d68</td><td>How to predict the global instantaneous feeling induced +<br/>by a facial picture? +<br/>To cite this version: +<br/>feeling induced by a facial picture?. Signal Processing: Image Communication, Elsevier, 2015, +<br/>pp.1-30. . +<br/>HAL Id: hal-01198718 +<br/>https://hal.archives-ouvertes.fr/hal-01198718 +<br/>Submitted on 14 Sep 2015 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('25030249', 'Arnaud Lienhard', 'arnaud lienhard')<br/>('2216412', 'Patricia Ladret', 'patricia ladret')<br/>('1788869', 'Alice Caplier', 'alice caplier')<br/>('25030249', 'Arnaud Lienhard', 'arnaud lienhard')<br/>('2216412', 'Patricia Ladret', 'patricia ladret')<br/>('1788869', 'Alice Caplier', 'alice caplier')</td><td></td></tr><tr><td>a8e75978a5335fd3deb04572bb6ca43dbfad4738</td><td>Sparse Graphical Representation based Discriminant +<br/>Analysis for Heterogeneous Face Recognition +</td><td>('2299758', 'Chunlei Peng', 'chunlei peng')<br/>('10699750', 'Xinbo Gao', 'xinbo gao')<br/>('2870173', 'Nannan Wang', 'nannan wang')<br/>('38158055', 'Jie Li', 'jie li')</td><td></td></tr><tr><td>a8d52265649c16f95af71d6f548c15afc85ac905</td><td>Situation Recognition with Graph Neural Networks +<br/><b>The Chinese University of Hong Kong, 2University of Toronto, 3Youtu Lab, Tencent</b><br/><b>Uber Advanced Technologies Group, 5Vector Institute</b></td><td>('8139953', 'Ruiyu Li', 'ruiyu li')<br/>('2103464', 'Makarand Tapaswi', 'makarand tapaswi')<br/>('2246396', 'Renjie Liao', 'renjie liao')<br/>('1729056', 'Jiaya Jia', 'jiaya jia')<br/>('2422559', 'Raquel Urtasun', 'raquel urtasun')<br/>('37895334', 'Sanja Fidler', 'sanja fidler')</td><td>ryli@cse.cuhk.edu.hk, {makarand,rjliao,urtasun,fidler}@cs.toronto.edu, leojia9@gmail.com +</td></tr><tr><td>a8583e80a455507a0f146143abeb35e769d25e4e</td><td>A DISTANCE-ACCURACY HYBRID WEIGHTED VOTING SCHEME +<br/>FOR PARTIAL FACE RECOGNITION +<br/>1Dept. of Information Engineering and Computer Science, +<br/><b>Feng Chia University, Taichung, Taiwan</b><br/>2Department of Photonics, +<br/><b>National Chiao Tung University, Taiwan</b></td><td>('40609876', 'Yung-Hui Li', 'yung-hui li')<br/>('3072232', 'Bo-Ren Zheng', 'bo-ren zheng')<br/>('2532474', 'Wei-Cheng Huang', 'wei-cheng huang')</td><td>ayunghui@gmail.com, bzawdcx@gmail.com, cs75757775@gmail.com, dchtien@mail.nctu.edu.tw +</td></tr><tr><td>a87e37d43d4c47bef8992ace408de0f872739efc</td><td>Review +<br/>A Comprehensive Review on Handcrafted and +<br/>Learning-Based Action Representation Approaches +<br/>for Human Activity Recognition +<br/><b>School of Computing and Communications Infolab21, Lancaster University, Lancaster LA1 4WA, UK</b><br/><b>COMSATS Institute of Information Technology, Lahore 54000, Pakistan</b><br/>Academic Editor: José Santamaria +<br/>Received: 5 September 2016; Accepted: 13 January 2017; Published: 23 January 2017 +</td><td>('2145942', 'Allah Bux Sargano', 'allah bux sargano')<br/>('5736243', 'Plamen Angelov', 'plamen angelov')</td><td>p.angelov@lancaster.ac.uk +<br/>drzhabib@ciitlahore.edu.pk +<br/>* Correspondence: a.bux@lancaster.ac.uk; Tel.: +44-152-451-0525 +</td></tr><tr><td>a8c8a96b78e7b8e0d4a4a422fcb083e53ad06531</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 8, No. 4, 2017 +<br/>3D Human Action Recognition using Hu Moment +<br/>Invariants and Euclidean Distance Classifier +<br/>System Engineering Department +<br/>System Engineering Department +<br/>Computer Science Department +<br/><b>University of Arkansas at Little Rock</b><br/><b>University of Arkansas at Little Rock</b><br/><b>University of Arkansas at Little Rock</b><br/>Arkansas, USA +<br/>Arkansas, USA +<br/>Arkansas, USA +</td><td>('19305764', 'Fadwa Al-Azzo', 'fadwa al-azzo')<br/>('22768683', 'Arwa Mohammed Taqi', 'arwa mohammed taqi')<br/>('1795699', 'Mariofanna Milanova', 'mariofanna milanova')</td><td></td></tr><tr><td>a8748a79e8d37e395354ba7a8b3038468cb37e1f</td><td>Seeing the Forest from the Trees: A Holistic Approach to Near-infrared +<br/>Heterogeneous Face Recognition +<br/><b>U.S. Army Research Laboratory</b><br/><b>University of Maryland, College Park</b><br/><b>West Virginia University</b></td><td>('39412489', 'Christopher Reale', 'christopher reale')<br/>('8147588', 'Nasser M. Nasrabadi', 'nasser m. nasrabadi')<br/>('1688527', 'Heesung Kwon', 'heesung kwon')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>reale@umiacs.umd.edu +<br/>heesung.kwon.civ@mail.mil +<br/>nasser.nasrabadi@mail.wvu.edu +<br/>rama@umiacs.umd.edu +</td></tr><tr><td>a8a61badec9b8bc01f002a06e1426a623456d121</td><td>JOINT SPATIO-TEMPORAL ACTION LOCALIZATION +<br/>IN UNTRIMMED VIDEOS WITH PER-FRAME SEGMENTATION +<br/><b>Xi an Jiaotong University</b><br/>2HERE Technologies +<br/>3Alibaba Group +<br/>4Microsoft Research +</td><td>('46809347', 'Xuhuan Duan', 'xuhuan duan')<br/>('40367806', 'Le Wang', 'le wang')<br/>('51262903', 'Changbo Zhai', 'changbo zhai')<br/>('46324995', 'Qilin Zhang', 'qilin zhang')<br/>('1786361', 'Zhenxing Niu', 'zhenxing niu')<br/>('1715389', 'Nanning Zheng', 'nanning zheng')<br/>('1745420', 'Gang Hua', 'gang hua')</td><td></td></tr><tr><td>a8154d043f187c6640cb6aedeaa8385a323e46cf</td><td>MURRUGARRA, KOVASHKA: IMAGE RETRIEVAL WITH MIXED INITIATIVE +<br/>Image Retrieval with Mixed Initiative and +<br/>Multimodal Feedback +<br/>Department of Computer Science +<br/><b>University of Pittsburgh</b><br/>Pittsburgh, PA, USA +</td><td>('1916866', 'Nils Murrugarra-Llerena', 'nils murrugarra-llerena')<br/>('1770205', 'Adriana Kovashka', 'adriana kovashka')</td><td>nineil@cs.pitt.edu +<br/>kovashka@cs.pitt.edu +</td></tr><tr><td>a812368fe1d4a186322bf72a6d07e1cf60067234</td><td><b>Imperial College London</b><br/>Department of Computing +<br/>Gaussian Processes +<br/>for Modeling of Facial Expressions +<br/>September, 2016 +<br/>Supervised by Prof. Maja Pantic +<br/>Submitted in part fulfilment of the requirements for the degree of PhD in Computing and +<br/><b>the Diploma of Imperial College London. This thesis is entirely my own work, and, except</b><br/>where otherwise indicated, describes my own research. +</td><td>('2308430', 'Stefanos Eleftheriadis', 'stefanos eleftheriadis')</td><td></td></tr><tr><td>de7f5e4ccc2f38e0c8f3f72a930ae1c43e0fdcf0</td><td>Merge or Not? Learning to Group Faces via Imitation Learning +<br/>SenseTime +<br/>SenseTime +<br/>SenseTime +<br/>Chen Chang Loy +<br/><b>The Chinese University of Hong Kong</b></td><td>('49990550', 'Yue He', 'yue he')<br/>('9963152', 'Kaidi Cao', 'kaidi cao')<br/>('46651787', 'Cheng Li', 'cheng li')</td><td>heyue@sensetime.com +<br/>caokaidi@sensetime.com +<br/>chengli@sensetime.com +<br/>ccloy@ie.cuhk.edu.hk +</td></tr><tr><td>de8381903c579a4fed609dff3e52a1dc51154951</td><td><b>Graz University of Technology</b><br/><b>Institute for Computer Graphics and Vision</b><br/>Dissertation +<br/>Shape and Appearance Based Analysis +<br/>of Facial Images for Assessing ICAO +<br/>Compliance +<br/>Graz, Austria, December 2010 +<br/>Thesis supervisors +<br/>Prof. Dr. Horst Bischof +<br/>Prof. Dr. Fernando De la Torre +</td><td>('3464430', 'Markus Storer', 'markus storer')</td><td></td></tr><tr><td>ded968b97bd59465d5ccda4f1e441f24bac7ede5</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Large scale 3D Morphable Models +<br/>Zafeiriou +<br/>Received: date / Accepted: date +</td><td>('47456731', 'James Booth', 'james booth')</td><td></td></tr><tr><td>de0eb358b890d92e8f67592c6e23f0e3b2ba3f66</td><td>ACCEPTED BY IEEE TRANS. PATTERN ANAL. AND MACH. INTELL. +<br/>Inference-Based Similarity Search in +<br/>Randomized Montgomery Domains for +<br/>Privacy-Preserving Biometric Identification +</td><td>('46393453', 'Yi Wang', 'yi wang')<br/>('2087574', 'Jianwu Wan', 'jianwu wan')<br/>('39954962', 'Jun Guo', 'jun guo')<br/>('32840387', 'Yiu-ming Cheung', 'yiu-ming cheung')</td><td></td></tr><tr><td>def569db592ed1715ae509644444c3feda06a536</td><td>Discovery and usage of joint attention in images +<br/><b>Weizmann Institute of Science, Rehovot, Israel</b><br/><b>The Center for Brains, Minds and Machines, Massachusetts Institute of Technology, Cambridge, MA USA</b><br/><b>Massachusetts Institute of Technology, Cambridge, MA USA</b><br/><b>Weizmann Institute of Science, Rehovot, Israel</b></td><td></td><td>Daniel Harari (hararid@weizmann.ac.il) +<br/>Joshua B. Tenenbaum (jbt@mit.edu) +<br/>Shimon Ullman (shimon.ullman@weizmann.ac.il) +</td></tr><tr><td>dee406a7aaa0f4c9d64b7550e633d81bc66ff451</td><td>Content-Adaptive Sketch Portrait Generation by +<br/>Decompositional Representation Learning +</td><td>('8335563', 'Dongyu Zhang', 'dongyu zhang')<br/>('1737218', 'Liang Lin', 'liang lin')<br/>('1765674', 'Tianshui Chen', 'tianshui chen')<br/>('1738906', 'Xian Wu', 'xian wu')<br/>('1989769', 'Wenwei Tan', 'wenwei tan')<br/>('1732655', 'Ebroul Izquierdo', 'ebroul izquierdo')</td><td></td></tr><tr><td>de15af84b1257211a11889b6c2adf0a2bcf59b42</td><td>Anomaly Detection in Non-Stationary and +<br/>Distributed Environments +<br/>Colin O’Reilly +<br/>Submitted for the Degree of +<br/>Doctor of Philosophy +<br/>from the +<br/><b>University of Surrey</b><br/><b>Institute for Communication Systems</b><br/>Faculty of Engineering and Physical Sciences +<br/><b>University of Surrey</b><br/>Guildford, Surrey GU2 7XH, U.K. +<br/>November 2014 +<br/>© Colin O’Reilly 2014 +</td><td></td><td></td></tr><tr><td>de3285da34df0262a4548574c2383c51387a24bf</td><td>Two-Stream Convolutional Networks for Dynamic Texture Synthesis +<br/>Department of Electrical Engineering and Computer Science +<br/><b>York University, Toronto</b></td><td>('19251410', 'Matthew Tesfaldet', 'matthew tesfaldet')</td><td>{mtesfald,mab}@eecs.yorku.ca +</td></tr><tr><td>dedabf9afe2ae4a1ace1279150e5f1d495e565da</td><td>3294 +<br/>Robust Face Recognition With Structurally +<br/>Incoherent Low-Rank Matrix Decomposition +</td><td>('2017922', 'Chia-Po Wei', 'chia-po wei')<br/>('2624492', 'Chih-Fan Chen', 'chih-fan chen')<br/>('2733735', 'Yu-Chiang Frank Wang', 'yu-chiang frank wang')</td><td></td></tr><tr><td>dec0c26855da90876c405e9fd42830c3051c2f5f</td><td>Supplementary Material: Learning Compositional Visual Concepts with Mutual +<br/>Consistency +<br/><b>School of Electrical and Computer Engineering, Cornell University, Ithaca NY</b><br/>3Siemens Corporate Technology, Princeton NJ +<br/>Contents +<br/>1. Objective functions +<br/>1.1. Adversarial loss +<br/>1.2. Extended cycle-consistency loss . +<br/>1.3. Commutative loss +<br/>. . . +<br/>. . . +<br/>. . . +<br/>2. Additional implementation details +<br/>3. Additional results +<br/>4. Discussion +<br/>5. Generalizing ConceptGAN +<br/>5.1. Assumption: Concepts have distinct states . . +<br/>5.2. Assumption: Concepts are mutually compatible +<br/>5.3. Generalization . +<br/>. . . +<br/>1. Objective functions +<br/>In this section, we provide complete mathematical +<br/>expressions for each of the three terms in our loss func- +<br/>tion, following the notation defined in Section 3 of the main +<br/>paper and the assumption that no training data is available +<br/>in subdomain Σ11. +<br/>1.1. Adversarial loss +<br/>For generator G1 and discriminator D10, for example, +<br/>the adversarial loss is expressed as: +<br/>Ladv(G1, D10, Σ00, Σ10) = Eσ10∼P10 [log D10(σ10)] +<br/>+Eσ00∼P00[log(1 − D10(G1(σ00)))] +<br/>(1) +<br/>where the generator G1 and discriminator D10 are +<br/>learned to optimize a minimax objective such that +<br/>G∗ +<br/>1 = arg min +<br/>G1 +<br/>max +<br/>D10 +<br/>Ladv(G1, D10, Σ00, Σ10) +<br/>(2) +<br/>For generator G2 and discriminator D01, the adversarial +<br/>loss is expressed as: +<br/>Ladv(G2, D01, Σ00, Σ01) = Eσ01∼P01 [log D01(σ01)] +<br/>+Eσ00∼P00[log(1 − D01(G2(σ00)))] +<br/>For generator F1 and discriminator D00, the adversarial +<br/>loss is expressed as: +<br/>Ladv(F1, D00, Σ10, Σ00) = Eσ00∼P00 [log D00(σ00)] +<br/>+Eσ10∼P10 [log(1 − D00(F1(σ10)))] +<br/>For generator F2 and discriminator D00, the adversarial +<br/>loss is expressed as: +<br/>Ladv(F2, D00, Σ01, Σ00) = Eσ00∼P00 [log D00(σ00)] +<br/>+Eσ01∼P01 [log(1 − D00(F2(σ01)))] +<br/>(5) +<br/>The overall adversarial loss LADV is the sum of these four +<br/>terms. +<br/>(3) +<br/>(4) +<br/>(6) +<br/>LADV =Ladv(G1, D10, Σ00, Σ10) +<br/>+ Ladv(G2, D01, Σ00, Σ01) +<br/>+ Ladv(F1, D00, Σ10, Σ00) +<br/>+ Ladv(F2, D00, Σ01, Σ00) +<br/>1.2. Extended cycle-consistency loss +<br/>Following our discussion in Section 3.2 of the main +<br/>paper, for any data sample σ00 in subdomain Σ00, a +<br/>distance-4 cycle consistency constraint is defined in the +<br/>clockwise direction (F2 ◦ F1 ◦ G2 ◦ G1)(σ00) ≈ σ00 and in +<br/>the counterclockwise direction (F1 ◦ F2 ◦ G1 ◦ G2)(σ00) ≈ +<br/>σ00. Such constraints are implemented by the penalty func- +<br/>tion: +<br/>Lcyc4(G, F, Σ00) +<br/>= Eσ00∼P00[(cid:107)(F2 ◦ F1 ◦ G2 ◦ G1)(σ00) − σ00(cid:107)1] +<br/>+ Eσ00∼P00[(cid:107)(F1 ◦ F2 ◦ G1 ◦ G2)(σ00) − σ00(cid:107)1]. +<br/>(7) +</td><td>('3303727', 'Yunye Gong', 'yunye gong')<br/>('1976152', 'Srikrishna Karanam', 'srikrishna karanam')<br/>('3311781', 'Ziyan Wu', 'ziyan wu')<br/>('2692770', 'Kuan-Chuan Peng', 'kuan-chuan peng')<br/>('39497207', 'Jan Ernst', 'jan ernst')<br/>('1767099', 'Peter C. Doerschuk', 'peter c. doerschuk')</td><td>{yg326,pd83}@cornell.edu,{first.last}@siemens.com +</td></tr><tr><td>de398bd8b7b57a3362c0c677ba8bf9f1d8ade583</td><td>Hierarchical Bayesian Theme Models for +<br/>Multi-pose Facial Expression Recognition +</td><td>('3069077', 'Qirong Mao', 'qirong mao')<br/>('1851510', 'Qiyu Rao', 'qiyu rao')<br/>('1770550', 'Yongbin Yu', 'yongbin yu')<br/>('1710341', 'Ming Dong', 'ming dong')</td><td></td></tr><tr><td>ded41c9b027c8a7f4800e61b7cfb793edaeb2817</td><td></td><td></td><td></td></tr><tr><td>defa8774d3c6ad46d4db4959d8510b44751361d8</td><td>FEBEI - Face Expression Based Emoticon Identification +<br/>CS - B657 Computer Vision +<br/>Robert J Henderson - rojahend +</td><td>('1854614', 'Nethra Chandrasekaran', 'nethra chandrasekaran')<br/>('1830695', 'Prashanth Kumar Murali', 'prashanth kumar murali')</td><td></td></tr><tr><td>b0c512fcfb7bd6c500429cbda963e28850f2e948</td><td></td><td></td><td></td></tr><tr><td>b08203fca1af7b95fda8aa3d29dcacd182375385</td><td>OBJECT AND TEXT-GUIDED SEMANTICS FOR CNN-BASED ACTIVITY RECOGNITION +<br/><b>U.S. Army Research Laboratory, Adelphi, MD, USA</b><br/>§Booz Allen Hamilton Inc., McLean, VA, USA +</td><td>('3090299', 'Sungmin Eum', 'sungmin eum')<br/>('39412489', 'Christopher Reale', 'christopher reale')<br/>('1688527', 'Heesung Kwon', 'heesung kwon')<br/>('3202888', 'Claire Bonial', 'claire bonial')</td><td></td></tr><tr><td>b03b4d8b4190361ed2de66fcbb6fda0c9a0a7d89</td><td>Deep Alternative Neural Network: Exploring +<br/>Contexts as Early as Possible for Action Recognition +<br/><b>School of Electronics Engineering and Computer Science, Peking University</b><br/><b>School of Electronics and Computer Engineering, Peking University</b></td><td>('3258842', 'Jinzhuo Wang', 'jinzhuo wang')<br/>('1788029', 'Wenmin Wang', 'wenmin wang')<br/>('8082703', 'Xiongtao Chen', 'xiongtao chen')<br/>('1702330', 'Ronggang Wang', 'ronggang wang')<br/>('1698902', 'Wen Gao', 'wen gao')</td><td>jzwang@pku.edu.cn, wangwm@ece.pku.edu.cn +<br/>cxt@pku.edu.cn, rgwang@ece.pku.edu.cn, wgao@pku.edu.cn +</td></tr><tr><td>b09b693708f412823053508578df289b8403100a</td><td>WANG et al.: TWO-STREAM SR-CNNS FOR ACTION RECOGNITION IN VIDEOS +<br/>Two-Stream SR-CNNs for Action +<br/>Recognition in Videos +<br/>1 Advanced Interactive Technologies Lab +<br/>ETH Zurich +<br/>Zurich, Switzerland +<br/>2 Computer Vision Lab +<br/>ETH Zurich +<br/>Zurich, Switzerland +</td><td>('46394691', 'Yifan Wang', 'yifan wang')<br/>('40403685', 'Jie Song', 'jie song')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('1681236', 'Luc Van Gool', 'luc van gool')<br/>('2531379', 'Otmar Hilliges', 'otmar hilliges')</td><td>yifan.wang@student.ethz.ch +<br/>jsong@inf.ethz.ch +<br/>07wanglimin@gmail.com +<br/>vangool@vision.ee.ethz.ch +<br/>otmar.hilliges@inf.ethz.ch +</td></tr><tr><td>b013cce42dd769db754a57351d49b7410b8e82ad</td><td>Automatic Point-based Facial Trait Judgments Evaluation +<br/>1Computer Vision Center, Edifici O, Campus UAB, Spain +<br/>2Universitat Oberta de Catalunya, Rambla del Poblenou 156, 08018, Barcelona, Spain +<br/><b>Princeton University, Princeton, New Jersey, USA</b><br/>4Department de Matematica Aplicada i Analisi, Universitat de Barcelona, Spain +</td><td>('1863902', 'David Masip', 'david masip')<br/>('2913698', 'Alexander Todorov', 'alexander todorov')</td><td>mrojas@cvc.uab.es, dmasipr@uoc.edu, atodorov@princeton.edu, jordi.vitria@ub.edu +</td></tr><tr><td>b07582d1a59a9c6f029d0d8328414c7bef64dca0</td><td>Employing Fusion of Learned and Handcrafted +<br/>Features for Unconstrained Ear Recognition +<br/>Maur´ıcio Pamplona Segundo∗† +<br/>October 24, 2017 +</td><td>('26977067', 'Earnest E. Hansley', 'earnest e. hansley')<br/>('1715991', 'Sudeep Sarkar', 'sudeep sarkar')</td><td></td></tr><tr><td>b017963d83b3edf71e1673d7ffdec13a6d350a87</td><td>View Independent Face Detection Based on +<br/>Combination of Local and Global Kernels +<br/><b>The University of Electro-Communications</b><br/>1-5-1 Chofugaoka, Chofu-shi, Tokyo 182-8585, JAPAN +</td><td>('2510362', 'Kazuhiro HOTTA', 'kazuhiro hotta')</td><td>hotta@ice.uec.ac.jp, +</td></tr><tr><td>b03d6e268cde7380e090ddaea889c75f64560891</td><td></td><td></td><td></td></tr><tr><td>b084683e5bab9b2bc327788e7b9a8e049d5fff8f</td><td>Using LIP to Gloss Over Faces in Single-Stage Face Detection +<br/>Networks +<br/><b>The University of Queensland, School of ITEE, QLD 4072, Australia</b></td><td>('1973322', 'Siqi Yang', 'siqi yang')<br/>('2331880', 'Arnold Wiliem', 'arnold wiliem')<br/>('3104113', 'Shaokang Chen', 'shaokang chen')<br/>('2270092', 'Brian C. Lovell', 'brian c. lovell')</td><td>{siqi.yang, a.wiliem, s.chen2}@uq.edu.au, lovell@itee.uq.edu.au +</td></tr><tr><td>b0c1615ebcad516b5a26d45be58068673e2ff217</td><td>How Image Degradations Affect Deep CNN-based Face +<br/>Recognition? +<br/>S¸amil Karahan1 Merve Kılınc¸ Yıldırım1 Kadir Kırtac¸1 Ferhat S¸ ¨ukr¨u Rende1 +<br/>G¨ultekin B¨ut¨un1Hazım Kemal Ekenel2 +</td><td></td><td></td></tr><tr><td>b03446a2de01126e6a06eb5d526df277fa36099f</td><td>A Torch Library for Action Recognition and Detection Using CNNs and LSTMs +<br/><b>Stanford University</b></td><td>('4910251', 'Helen Jiang', 'helen jiang')</td><td>{gthung, helennn}@stanford.edu +</td></tr><tr><td>b0de0892d2092c8c70aa22500fed31aa7eb4dd3f</td><td>(will be inserted by the editor) +<br/>A robust and efficient video representation for action recognition +<br/>Received: date / Accepted: date +</td><td>('1804138', 'Heng Wang', 'heng wang')</td><td></td></tr><tr><td>b018fa5cb9793e260b8844ae155bd06380988584</td><td>Project STAR IST-2000-28764 +<br/>Deliverable D6.3 Enhanced face and arm/hand +<br/>detector +<br/>Date: August 29th, 2003 +<br/><b>Katholieke Universiteit Leuven, ESAT/VISICS</b><br/>Kasteelpark Arenberg 10, 3001 Heverlee, Belgium +<br/>Tel. +32-16-32.10.61 and Fax. +32-16-32.17.23 +<br/>http://www.esat.kuleuven.ac.be/ knummiar/star/star.html +<br/>To: STAR project partners +<br/>Siemens CT PP6, +<br/>Otto-Hahn-Ring 6, 81730 Munich, Germany +<br/>Tel. +49-89-636.49.851, Fax. +49-89-636.481.00 +<br/>Introduction +<br/>KU Leuven is responsible for the work package number 6, Automated view selection and +<br/>camera hand-over. The main goal is to build an intelligent virtual editor that produces as +<br/>an output a single video stream from multiple input streams. The selection should be made +<br/>in such a way that the resulting stream is pleasant to watch and informative about what is +<br/>going on in the scene. Face detection and object tracking is needed to select the best camera +<br/>view from the multi-camera system. +<br/>KUL has delivered the STAR deliverables D6.1 Initial face detection software and D6.2 Initial +<br/>arm/hand tracking software from work package 6, July 2002 (month 12). The integration of +<br/>the detection and tracking has been needed to successfully provide this deliverable D6.3 +<br/>Enhanced face and arm/hand detector. +<br/>We explain (cid:12)rst the enhanced face detection, followed by the enhanced tracking software and +<br/>(cid:12)nally the integration. Also the hand tracking results with simple histogram-based detection +<br/>is presented. The results will be shown using the common STAR data sequencies, from +<br/>di(cid:11)erent Siemens factories, in Germany. +</td><td>('2381884', 'Katja Nummiaro', 'katja nummiaro')<br/>('2733505', 'Rik Fransens', 'rik fransens')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>fknummiar, fransen, vangoolg@esat.kuleuven.ac.be +<br/>artur.raczynski@mchp.siemens.de +</td></tr><tr><td>b073313325b6482e22032e259d7311fb9615356c</td><td>Robust and Accurate Cancer Classification with Gene Expression Profiling +<br/>Dept. of Computer Science +<br/><b>Human Interaction Research Lab</b><br/>Dept. of Computer Science +<br/><b>University of California</b><br/>Riverside, CA 92521 +<br/><b>Motorola, Inc</b><br/>Tempe, AZ 85282 +<br/><b>University of California</b><br/>Riverside, CA 92521 +</td><td>('31947043', 'Haifeng Li', 'haifeng li')<br/>('1749400', 'Keshu Zhang', 'keshu zhang')<br/>('6820989', 'Tao Jiang', 'tao jiang')</td><td>hli@cs.ucr.edu +<br/>keshu.zhang@motorola.com +<br/>jiang@cs.ucr.edu +</td></tr><tr><td>a6f81619158d9caeaa0863738ab400b9ba2d77c2</td><td>Face Recognition using Convolutional Neural Network +<br/>and Simple Logistic Classifier +<br/>Intelligent Systems Laboratory (ISLAB), +<br/>Faculty of Electrical & Computer Engineering +<br/><b>K.N. Toosi University of Technology, Tehran, Iran</b></td><td>('2040276', 'Hurieh Khalajzadeh', 'hurieh khalajzadeh')<br/>('10694774', 'Mohammad Mansouri', 'mohammad mansouri')<br/>('1709359', 'Mohammad Teshnehlab', 'mohammad teshnehlab')</td><td>hurieh.khalajzadeh@gmail.com, +<br/>mohammad.mansouri@ee.kntu.ac.ir, +<br/>teshnehlab@eetd.kntu.ac.ir +</td></tr><tr><td>a66d89357ada66d98d242c124e1e8d96ac9b37a0</td><td>Failure Detection for Facial Landmark Detectors +<br/>Computer Vision Lab, D-ITET, ETH Zurich, Switzerland +</td><td>('33028242', 'Andreas Steger', 'andreas steger')<br/>('1732855', 'Radu Timofte', 'radu timofte')</td><td>stegeran@ethz.ch, {radu.timofte, vangool}@vision.ee.ethz.ch +</td></tr><tr><td>a6d7cf29f333ea3d2aeac67cde39a73898e270b7</td><td>Gender Classification from Facial Images Using Texture Descriptors +<br/>801 +<br/>Gender Classification from Facial Images Using Texture Descriptors +<br/><b>King Saud University, KSA</b><br/><b>King Saud University, KSA</b><br/><b>King Saud University, KSA</b><br/><b>University of Nevada at Reno, USA</b></td><td>('1758125', 'Ihsan Ullah', 'ihsan ullah')<br/>('1966959', 'Hatim Aboalsamh', 'hatim aboalsamh')<br/>('2363759', 'Muhammad Hussain', 'muhammad hussain')<br/>('1758305', 'Ghulam Muhammad', 'ghulam muhammad')<br/>('1808451', 'George Bebis', 'george bebis')</td><td>{ihsanullah, hatim, mhussain, ghulam}@ksu.edu.sa, bebis@cse.unr.edu +</td></tr><tr><td>a611c978e05d7feab01fb8a37737996ad6e88bd9</td><td>Benchmarking 3D pose estimation for +<br/>face recognition +<br/><b>Computational Biomedicine Lab, University of Houston, TX, USA</b></td><td>('39634395', 'Pengfei Dou', 'pengfei dou')<br/>('2461369', 'Yuhang Wu', 'yuhang wu')<br/>('2700399', 'Shishir K. Shah', 'shishir k. shah')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td>{pengfei,yuhang}@cbl.uh.edu, {sshah,IKakadia}@central.uh.edu +</td></tr><tr><td>a608c5f8fd42af6e9bd332ab516c8c2af7063c61</td><td>2408 +<br/>Age Estimation via Grouping and Decision Fusion +</td><td>('3006921', 'Kuan-Hsien Liu', 'kuan-hsien liu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('9363144', 'C.-C. Jay Kuo', 'c.-c. jay kuo')</td><td></td></tr><tr><td>a6e8a8bb99e30a9e80dbf80c46495cf798066105</td><td>Ranking Generative Adversarial Networks: +<br/>Subjective Control over Semantic Image Attributes +<br/><b>University of Bath</b></td><td>('41020280', 'Yassir Saquil', 'yassir saquil')<br/>('1808255', 'Kwang In Kim', 'kwang in kim')</td><td></td></tr><tr><td>a6eb6ad9142130406fb4ffd4d60e8348c2442c29</td><td>Video Description: A Survey of Methods, +<br/>Datasets and Evaluation Metrics +</td><td>('50978260', 'Nayyer Aafaq', 'nayyer aafaq')<br/>('1746166', 'Syed Zulqarnain Gilani', 'syed zulqarnain gilani')<br/>('46641573', 'Wei Liu', 'wei liu')<br/>('46332747', 'Ajmal Mian', 'ajmal mian')</td><td></td></tr><tr><td>a6ffe238eaf8632b4a8a6f718c8917e7f3261546</td><td> Australasian Medical Journal [AMJ 2011, 4, 10, 555-562] +<br/>Dynamic Facial Prosthetics for Sufferers of Facial Paralysis +<br/><b>Nottingham Trent University, Nottingham, UK</b><br/><b>Nottingham University Hospital, Nottingham, UK</b><br/> RESEARCH +<br/> +<br/>Please cite this paper as: Coulter F, Breedon P, Vloeberghs +<br/>M. Dynamic facial prosthetics for sufferers of facial +<br/>paralysis. +<br/>AMJ 2011, 4, 10, 555-562 +<br/>http//dx.doi.org/10.4066/AMJ.2011.921 +<br/>Corresponding Author: +<br/><b>Nottingham Trent University</b><br/> +<br/>United Kingdom +</td><td>('6930559', 'Fergal Coulter', 'fergal coulter')<br/>('3214667', 'Philip Breedon', 'philip breedon')<br/>('40436855', 'Michael Vloeberghs', 'michael vloeberghs')<br/>('3214667', 'Philip Breedon', 'philip breedon')</td><td>philip.breedon@ntu.ac.uk +</td></tr><tr><td>a6583c8daa7927eedb3e892a60fc88bdfe89a486</td><td></td><td></td><td></td></tr><tr><td>a660390654498dff2470667b64ea656668c98ecc</td><td>FACIAL EXPRESSION RECOGNITION BASED ON GRAPH-PRESERVING SPARSE +<br/>NON-NEGATIVE MATRIX FACTORIZATION +<br/><b>Institute of Information Science</b><br/><b>Beijing Jiaotong University</b><br/>Beijing 100044, P.R. China +<br/>, Bastiaan Kleijn +<br/>ACCESS Linnaeus Center +<br/><b>KTH Royal Institute of Technology, Stockholm</b><br/>School of Electrical Engineering +</td><td>('3247912', 'Ruicong Zhi', 'ruicong zhi')<br/>('1749334', 'Markus Flierl', 'markus flierl')<br/>('1738408', 'Qiuqi Ruan', 'qiuqi ruan')</td><td>{05120370, qqruan}@bjtu.edu.cn +<br/>{ruicong, mflierl, bastiaan}@kth.se +</td></tr><tr><td>a60907b7ee346b567972074e3e03c82f64d7ea30</td><td>Head Motion Signatures from Egocentric Videos +<br/><b>The Hebrew University of Jerusalem, Israel</b><br/>2 IIIT Delhi, India +</td><td>('2926663', 'Yair Poleg', 'yair poleg')<br/>('1897733', 'Chetan Arora', 'chetan arora')<br/>('1796055', 'Shmuel Peleg', 'shmuel peleg')</td><td></td></tr><tr><td>a6e43b73f9f87588783988333997a81b4487e2d5</td><td>Facial Age Estimation by Total Ordering +<br/>Preserving Projection +<br/>National Key Laboratory for Novel Software Technology +<br/><b>Nanjing University, Nanjing 210023, China</b></td><td>('39527177', 'Xiao-Dong Wang', 'xiao-dong wang')<br/>('1692625', 'Zhi-Hua Zhou', 'zhi-hua zhou')</td><td>{wangxd,zhouzh}@lamda.nju.edu.cn +</td></tr><tr><td>a6496553fb9ab9ca5d69eb45af1bdf0b60ed86dc</td><td>Semi-supervised Neighborhood Preserving +<br/>Discriminant Embedding: +<br/>A Semi-supervised Subspace Learning +<br/>Algorithm +<br/>1 Department of Computer Science and Software Engineering, +<br/><b></b><br/><b>University of Western Australia</b></td><td>('2067346', 'Maryam Mehdizadeh', 'maryam mehdizadeh')<br/>('1766400', 'Cara MacNish', 'cara macnish')<br/>('39128433', 'R. Nazim Khan', 'r. nazim khan')<br/>('1698675', 'Mohammed Bennamoun', 'mohammed bennamoun')</td><td></td></tr><tr><td>a6b5ffb5b406abfda2509cae66cdcf56b4bb3837</td><td>One Shot Similarity Metric Learning +<br/>for Action Recognition +<br/><b>The Weizmann Institute of</b><br/><b>The Open University</b><br/>Science, Rehovot, Israel. +<br/>Raanana, Israel. +<br/><b>The Blavatnik School of Computer Science, Tel-Aviv University, Tel-Aviv, Israel</b></td><td>('3294355', 'Orit Kliper-Gross', 'orit kliper-gross')<br/>('1756099', 'Tal Hassner', 'tal hassner')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td>orit.kliper@weizmann.ac.il +<br/>hassner@openu.ac.il +<br/>wolf@cs.tau.ac.il +</td></tr><tr><td>a6590c49e44aa4975b2b0152ee21ac8af3097d80</td><td>https://doi.org/10.1007/s11263-018-1074-6 +<br/>3D Interpreter Networks for Viewer-Centered Wireframe Modeling +<br/>Received: date / Accepted: date +</td><td>('3045089', 'Jiajun Wu', 'jiajun wu')<br/>('1763295', 'Joshua B. Tenenbaum', 'joshua b. tenenbaum')</td><td></td></tr><tr><td>a694180a683f7f4361042c61648aa97d222602db</td><td>Face Recognition using Scattering Wavelet under Illicit Drug Abuse Variations +<br/>IIIT-Delhi India +</td><td>('2503967', 'Prateekshit Pandey', 'prateekshit pandey')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')</td><td>fprateekshit12078, rsingh, mayankg@iiitd.ac.in +</td></tr><tr><td>a6db73f10084ce6a4186363ea9d7475a9a658a11</td><td></td><td></td><td></td></tr><tr><td>a6e25cab2251a8ded43c44b28a87f4c62e3a548a</td><td>Let’s Dance: Learning From Online Dance Videos +<br/><b>Georgia Institute of Technology</b><br/>Irfan Essa +</td><td>('40333356', 'Daniel Castro', 'daniel castro')<br/>('2935619', 'Steven Hickson', 'steven hickson')<br/>('3430745', 'Patsorn Sangkloy', 'patsorn sangkloy')<br/>('40506496', 'Bhavishya Mittal', 'bhavishya mittal')<br/>('35459529', 'Sean Dai', 'sean dai')<br/>('1945508', 'James Hays', 'james hays')</td><td>shickson@gatech.edu +<br/>patsorn sangkloy@gatech.edu +<br/>dcastro9@gatech.edu +<br/>bmittal6@gatech.edu +<br/>sdai@gatech.edu +<br/>hays@gatech.edu +<br/>irfan@gatech.edu +</td></tr><tr><td>a6634ff2f9c480e94ed8c01d64c9eb70e0d98487</td><td></td><td></td><td></td></tr><tr><td>a6270914cf5f60627a1332bcc3f5951c9eea3be0</td><td>Joint Attention in Driver-Pedestrian Interaction: from +<br/>Theory to Practice +<br/>Department of Electrical Engineering and Computer Science +<br/><b>York University, Toronto, ON, Canada</b><br/>March 28, 2018 +</td><td>('26902477', 'Amir Rasouli', 'amir rasouli')<br/>('1727853', 'John K. Tsotsos', 'john k. tsotsos')</td><td>{aras,tsotsos}@eecs.yorku.ca +</td></tr><tr><td>a6ce2f0795839d9c2543d64a08e043695887e0eb</td><td>Driver Gaze Region Estimation +<br/>Without Using Eye Movement +<br/><b>Massachusetts Institute of Technology (MIT</b></td><td>('49925254', 'Philipp Langhans', 'philipp langhans')<br/>('7137846', 'Joonbum Lee', 'joonbum lee')<br/>('1901227', 'Bryan Reimer', 'bryan reimer')</td><td></td></tr><tr><td>a6b1d79bc334c74cde199e26a7ef4c189e9acd46</td><td>bioRxiv preprint first posted online Aug. 17, 2017; +<br/>doi: +<br/>http://dx.doi.org/10.1101/177196 +<br/>. +<br/>The copyright holder for this preprint (which was +<br/>not peer-reviewed) is the author/funder. It is made available under a +<br/>CC-BY-NC 4.0 International license +<br/>Deep Recurrent Neural Network Reveals a Hierarchy of +<br/>Process Memory during Dynamic Natural Vision +<br/>1Weldon School of Biomedical Engineering +<br/>2School of Electrical and Computer Engineering +<br/><b>Purdue Institute for Integrative Neuroscience</b><br/><b>Purdue University, West Lafayette, Indiana, 47906, USA</b><br/>*Correspondence +<br/>Assistant Professor of Biomedical Engineering +<br/>Assistant Professor of Electrical and Computer Engineering +<br/><b>College of Engineering, Purdue University</b><br/>206 S. Martin Jischke Dr. +<br/>West Lafayette, IN 47907, USA +<br/>Phone: +1 765 496 1872 +<br/>Fax: +1 765 496 1459 +</td><td>('4416237', 'Junxing Shi', 'junxing shi')<br/>('4431043', 'Haiguang Wen', 'haiguang wen')<br/>('3334748', 'Yizhen Zhang', 'yizhen zhang')<br/>('3418794', 'Kuan Han', 'kuan han')<br/>('1799110', 'Zhongming Liu', 'zhongming liu')<br/>('1799110', 'Zhongming Liu', 'zhongming liu')</td><td>Email: zmliu@purdue.edu +</td></tr><tr><td>a6ebe013b639f0f79def4c219f585b8a012be04f</td><td>Facial Expression Recognition Based on Hybrid +<br/>Approach +<br/><b>Graduate School of Science and Engineering, Saitama University</b><br/>255 Shimo-Okubo, Sakura-ku, Saitama-shi, Saitama 338-8570, Japan +<br/>E-mail +</td><td>('13403748', 'Md. Abdul Mannan', 'md. abdul mannan')<br/>('34949901', 'Antony Lam', 'antony lam')<br/>('2367471', 'Yoshinori Kobayashi', 'yoshinori kobayashi')<br/>('1737913', 'Yoshinori Kuno', 'yoshinori kuno')</td><td></td></tr><tr><td>a6e21438695dbc3a184d33b6cf5064ddf655a9ba</td><td>PKU-MMD: A Large Scale Benchmark for Continuous Multi-Modal Human +<br/>Action Understanding +<br/><b>Institiude of Computer Science and Technology, Peking University</b></td><td>('2994549', 'Jiaying Liu', 'jiaying liu')<br/>('1708754', 'Chunhui Liu', 'chunhui liu')</td><td>{liuchunhui, huyy, lyttonhao, ssj940929, liujiaying}@pku.edu.cn +</td></tr><tr><td>b9081856963ceb78dcb44ac410c6fca0533676a3</td><td>UntrimmedNets for Weakly Supervised Action Recognition and Detection +<br/>1Computer Vision Laboratory, ETH Zurich, Switzerland +<br/><b>The Chinese University of Hong Kong, Hong Kong</b></td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('1807606', 'Dahua Lin', 'dahua lin')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>b97f694c2a111b5b1724eefd63c8d64c8e19f6c9</td><td>Group Affect Prediction Using Multimodal Distributions +<br/>Aspiring Minds +<br/>Univeristy of Massachusetts, Amherst +<br/><b>Johns Hopkins University</b></td><td>('40997180', 'Saqib Nizam Shamsi', 'saqib nizam shamsi')<br/>('47679973', 'Bhanu Pratap Singh', 'bhanu pratap singh')<br/>('7341605', 'Manya Wadhwa', 'manya wadhwa')</td><td>shamsi.saqib@gmail.com +<br/>bhanupratap.mnit@gmail.com +<br/>mwadhwa1@jhu.edu +</td></tr><tr><td>b9d0774b0321a5cfc75471b62c8c5ef6c15527f5</td><td>Fishy Faces: Crafting Adversarial Images to Poison Face Authentication +<br/>imec-DistriNet, KU Leuven +<br/>imec-DistriNet, KU Leuven +<br/>imec-DistriNet, KU Leuven +<br/>imec-DistriNet, KU Leuven +<br/>imec-DistriNet, KU Leuven +</td><td>('4412412', 'Giuseppe Garofalo', 'giuseppe garofalo')<br/>('23974422', 'Vera Rimmer', 'vera rimmer')<br/>('19243432', 'Tim Van hamme', 'tim van hamme')<br/>('1722184', 'Davy Preuveneers', 'davy preuveneers')<br/>('1752104', 'Wouter Joosen', 'wouter joosen')</td><td></td></tr><tr><td>b9cad920a00fc0e997fc24396872e03f13c0bb9c</td><td>FACE LIVENESS DETECTION UNDER BAD ILLUMINATION CONDITIONS +<br/><b>University of Campinas (Unicamp</b><br/>Campinas, SP, Brazil +</td><td>('2826093', 'Bruno Peixoto', 'bruno peixoto')<br/>('34629204', 'Carolina Michelassi', 'carolina michelassi')<br/>('2145405', 'Anderson Rocha', 'anderson rocha')</td><td></td></tr><tr><td>b908edadad58c604a1e4b431f69ac8ded350589a</td><td>Deep Face Feature for Face Alignment +</td><td>('15679675', 'Boyi Jiang', 'boyi jiang')<br/>('2938279', 'Juyong Zhang', 'juyong zhang')<br/>('2964129', 'Bailin Deng', 'bailin deng')<br/>('8280113', 'Yudong Guo', 'yudong guo')<br/>('47968194', 'Ligang Liu', 'ligang liu')</td><td></td></tr><tr><td>b93bf0a7e449cfd0db91a83284d9eba25a6094d8</td><td>Supplementary Material for: Active Pictorial Structures +<br/>Epameinondas Antonakos +<br/>Joan Alabort-i-Medina +<br/>Stefanos Zafeiriou +<br/><b>Imperial College London</b><br/>180 Queens Gate, SW7 2AZ, London, U.K. +<br/>In the following sections, we provide additional material for the paper “Active Pictorial Structures”. Section 1 explains in +<br/>more detail the differences between the proposed Active Pictorial Structures (APS) and Pictorial Structures (PS). Section 2 +<br/>presents the proofs about the structure of the precision matrices of the Gaussian Markov Random Filed (GMRF) (Eqs. 10 +<br/>and 12 of the main paper). Section 3 gives an analysis about the forward Gauss-Newton optimization of APS and shows that +<br/>the inverse technique with fixed Jacobian and Hessian, which is used in the main paper, is much faster. Finally, Sec. 4 shows +<br/>additional experimental results and conducts new experiments on different objects (human eyes and cars). An open-source +<br/>implementation of APS is available within the Menpo Project [1] in http://www.menpo.org/. +<br/>1. Differences between Active Pictorial Structures and Pictorial Structures +<br/>As explained in the main paper, the proposed model is partially motivated by PS [4, 8]. In the original formulation of PS, +<br/>the cost function to be optimized has the form +<br/>(cid:88) +<br/>n(cid:88) +<br/>n(cid:88) +<br/>i=1 +<br/>arg min +<br/>= arg min +<br/>i=1 +<br/>mi((cid:96)i) + +<br/>dij((cid:96)i, (cid:96)j) = +<br/>i,j:(vi,vj )∈E +<br/>[A((cid:96)i) − µa +<br/>i ]T (Σa +<br/>i )−1[A((cid:96)i) − µa +<br/>i ] + +<br/>(cid:88) +<br/>i,j:(vi,vj )∈E +<br/>[(cid:96)i − (cid:96)j − µd +<br/>ij]T (Σd +<br/>ij)−1[(cid:96)i − (cid:96)j − µd +<br/>ij] +<br/>(1) +<br/>1 , . . . , (cid:96)T +<br/>n ]T is the vector of landmark coordinates ((cid:96)i = [xi, yi]T , ∀i = 1, . . . , n), A((cid:96)i) is a feature vector +<br/>where s = [(cid:96)T +<br/>ij} denote the mean +<br/>extracted from the image location (cid:96)i and we have assumed a tree G = (V, E). {µa +<br/>and covariances of the appearance and deformation respectively. In Eq. 1, mi((cid:96)i) is a function measuring the degree of +<br/>mismatch when part vi is placed at location (cid:96)i in the image. Moreover, dij((cid:96)i, (cid:96)j) denotes a function measuring the degree +<br/>of deformation of the model when part vi is placed at location (cid:96)i and part vj is placed at location (cid:96)j. The authors show +<br/>an inference algorithm based on distance transform [3] that can find a global minimum of Eq. 1 without any initialization. +<br/>However, this algorithm imposes two important restrictions: (1) appearance of each part is independent of the rest of them +<br/>and (2) G must always be acyclic (a tree). Additionally, the computation of mi((cid:96)i) for all parts (i = 1, . . . , n) and all possible +<br/>image locations (response maps) has a high computational cost, which makes the algorithm very slow. Finally, in [8], the +<br/>authors only use a diagonal covariance for the relative locations (deformation) of each edge of the graph, which restricts the +<br/>flexibility of the model. +<br/>i } and {µd +<br/>ij, Σd +<br/>i , Σa +<br/>In the proposed APS, we aim to minimize the cost function (Eq. 19 of the main paper) +<br/>(cid:107)A(S(¯s, p)) − ¯a(cid:107)2 +<br/>[A(S(¯s, p)) − ¯a]T Qa[A(S(¯s, p)) − ¯a] + [S(¯s, p) − ¯s]T Qd[S(¯s, p) − ¯s] +<br/>Qa + (cid:107)S(¯s, p) − ¯s(cid:107)2 +<br/>Qd = +<br/>arg min +<br/>= arg min +<br/>(2) +<br/>There are two main differences between APS and PS: (1) we employ a statistical shape model and optimize with respect +<br/>to its parameters and (2) we use the efficient Gauss-Newton optimization technique. However, these differences introduce +<br/>some important advantages, as also mentioned in the main paper. The proposed formulation allows to define a graph (not +<br/>only tree) between the object’s parts. This means that we can assume dependencies between any pair of landmarks for both +</td><td></td><td>{e.antonakos, ja310, s.zafeiriou}@imperial.ac.uk +</td></tr><tr><td>b9c9c7ef82f31614c4b9226e92ab45de4394c5f6</td><td>11 +<br/>Face Recognition under Varying Illumination +<br/><b>Nanyang Technological University</b><br/>Singapore +<br/>1. Introduction +<br/>Face Recognition by a robot or machine is one of the challenging research topics in the +<br/>recent years. It has become an active research area which crosscuts several disciplines such +<br/>as image processing, pattern recognition, computer vision, neural networks and robotics. +<br/>For many applications, the performances of face recognition systems in controlled +<br/>environments have achieved a satisfactory level. However, there are still some challenging +<br/>issues to address in face recognition under uncontrolled conditions. The variation in +<br/>illumination is one of the main challenging problems that a practical face recognition system +<br/>needs to deal with. It has been proven that in face recognition, differences caused by +<br/>illumination variations are more significant than differences between individuals (Adini et +<br/>al., 1997). Various methods have been proposed to solve the problem. These methods can be +<br/>classified into three categories, named face and illumination modeling, illumination +<br/>invariant feature extraction and preprocessing and normalization. In this chapter, an +<br/>extensive and state-of-the-art study of existing approaches to handle illumination variations +<br/>is presented. Several latest and representative approaches of each category are presented in +<br/>detail, as well as the comparisons between them. Moreover, to deal with complex +<br/>environment where illumination variations are coupled with other problems such as pose +<br/>and expression variations, a good feature representation of human face should not only be +<br/>illumination invariant, but also robust enough against pose and expression variations. Local +<br/>binary pattern (LBP) is such a local texture descriptor. In this chapter, a detailed study of the +<br/>LBP and its several important extensions is carried out, as well as its various combinations +<br/>with other techniques to handle illumination invariant face recognition under a complex +<br/>environment. By generalizing different strategies in handling illumination variations and +<br/>evaluating their performances, several promising directions for future research have been +<br/>suggested. +<br/>This chapter is organized as follows. Several famous methods of face and illumination +<br/>modeling are introduced in Section 2. In Section 3, latest and representative approaches of +<br/>illumination invariant feature extraction are presented in detail. More attentions are paid on +<br/>quotient-image-based methods. In Section 4, the normalization methods on discarding low +<br/>frequency coefficients in various transformed domains are introduced with details. In +<br/>Section 5, a detailed introduction of the LBP and its several important extensions is +<br/>presented, as well as its various combinations with other face recognition techniques. In +<br/>Section 6, comparisons between different methods and discussion of their advantages and +<br/>disadvantages are presented. Finally, several promising directions as the conclusions are +<br/>drawn in Section 7. +<br/>www.intechopen.com +</td><td>('9244425', 'Lian Zhichao', 'lian zhichao')<br/>('9224769', 'Er Meng Joo', 'er meng joo')</td><td></td></tr><tr><td>b9f2a755940353549e55690437eb7e13ea226bbf</td><td>Unsupervised Feature Learning from Videos for Discovering and Recognizing Actions +</td><td>('3296857', 'Carolina Redondo-Cabrera', 'carolina redondo-cabrera')<br/>('2941882', 'Roberto J. López-Sastre', 'roberto j. lópez-sastre')</td><td>carolina.redondoc@edu.uah.es +<br/>robertoj.lopez@uah.es +</td></tr><tr><td>b9cedd1960d5c025be55ade0a0aa81b75a6efa61</td><td>INEXACT KRYLOV SUBSPACE ALGORITHMS FOR LARGE +<br/>MATRIX EXPONENTIAL EIGENPROBLEM FROM +<br/>DIMENSIONALITY REDUCTION +</td><td>('1685951', 'Gang Wu', 'gang wu')<br/>('7139289', 'Ting-ting Feng', 'ting-ting feng')<br/>('9472022', 'Li-jia Zhang', 'li-jia zhang')<br/>('5828998', 'Meng Yang', 'meng yang')</td><td></td></tr><tr><td>b971266b29fcecf1d5efe1c4dcdc2355cb188ab0</td><td>MAI et al.: ON THE RECONSTRUCTION OF FACE IMAGES FROM DEEP FACE TEMPLATES +<br/>On the Reconstruction of Face Images from +<br/>Deep Face Templates +</td><td>('3391550', 'Guangcan Mai', 'guangcan mai')<br/>('1684684', 'Kai Cao', 'kai cao')<br/>('1768574', 'Pong C. Yuen', 'pong c. yuen')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>a1af7ec84472afba0451b431dfdb59be323e35b7</td><td>LikeNet: A Siamese Motion Estimation +<br/>Network Trained in an Unsupervised Way +<br/>Multimedia and Vision Research Group +<br/><b>Queen Mary University of London</b><br/>London, UK +</td><td>('49505678', 'Aria Ahmadi', 'aria ahmadi')<br/>('2000297', 'Ioannis Marras', 'ioannis marras')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')<br/>('49505678', 'Aria Ahmadi', 'aria ahmadi')<br/>('2000297', 'Ioannis Marras', 'ioannis marras')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')</td><td>a.ahmadi@qmul.ac.uk +<br/>i.marras@qmul.ac.uk +<br/>i.patras@qmul.ac.uk +</td></tr><tr><td>a1dd806b8f4f418d01960e22fb950fe7a56c18f1</td><td>Interactively Building a Discriminative Vocabulary of Nameable Attributes +<br/><b>Toyota Technological Institute, Chicago (TTIC</b><br/><b>University of Texas at Austin</b></td><td>('1713589', 'Devi Parikh', 'devi parikh')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>dparikh@ttic.edu +<br/>grauman@cs.utexas.edu +</td></tr><tr><td>a158c1e2993ac90a90326881dd5cb0996c20d4f3</td><td>OPEN ACCESS +<br/>ISSN 2073-8994 +<br/>Article +<br/>1 DMA, Università degli Studi di Palermo, via Archirafi 34, 90123 Palermo, Italy +<br/>2 CITC, Università degli Studi di Palermo, via Archirafi 34, 90123 Palermo, Itlay +<br/>3 Istituto Nazionale di Ricerche Demopolis, via Col. Romey 7, 91100 Trapani, Italy +<br/>† Deceased on 15 March 2009. +<br/>Received: 4 March 2010; in revised form: 23 March 2010 / Accepted: 29 March 2010 / +<br/>Published: 1 April 2010 +</td><td>('1716744', 'Bertrand Zavidovique', 'bertrand zavidovique')</td><td>4 IEF, Université Paris IX–Orsay, Paris, France; E-Mail: bertrand.zavidovique@u-psud.fr (B.Z.) +<br/>* Author to whom correspondence should be addressed; E-Mail: metabacchi@demopolis.it. +</td></tr><tr><td>a15d9d2ed035f21e13b688a78412cb7b5a04c469</td><td>Object Detection Using +<br/>Strongly-Supervised Deformable Part Models +<br/>1Computer Vision and Active Perception Laboratory (CVAP), KTH, Sweden +<br/>2INRIA, WILLOW, Laboratoire d’Informatique de l’Ecole Normale Superieure +</td><td>('2622491', 'Hossein Azizpour', 'hossein azizpour')<br/>('1785596', 'Ivan Laptev', 'ivan laptev')</td><td>azizpour@kth.se,ivan.laptev@inria.fr +</td></tr><tr><td>a1b1442198f29072e907ed8cb02a064493737158</td><td>456 +<br/>Crowdsourcing Facial Responses +<br/>to Online Videos +</td><td>('1801452', 'Daniel McDuff', 'daniel mcduff')<br/>('1754451', 'Rana El Kaliouby', 'rana el kaliouby')<br/>('1719389', 'Rosalind W. Picard', 'rosalind w. picard')</td><td></td></tr><tr><td>a14db48785d41cd57d4eac75949a6b79fc684e70</td><td>Fast High Dimensional Vector Multiplication Face Recognition +<br/><b>Tel Aviv University</b><br/><b>Tel Aviv University</b><br/><b>Tel Aviv University</b><br/>IBM Research +</td><td>('2109324', 'Oren Barkan', 'oren barkan')<br/>('40389676', 'Jonathan Weill', 'jonathan weill')<br/>('1776343', 'Lior Wolf', 'lior wolf')<br/>('2580470', 'Hagai Aronowitz', 'hagai aronowitz')</td><td>orenbarkan@post.tau.ac.il +<br/>yonathanw@post.tau.ac.il +<br/>wolf@cs.tau.ac.il +<br/>hagaia@il.ibm.com +</td></tr><tr><td>a15c728d008801f5ffc7898568097bbeac8270a4</td><td>Concise Preservation by Combining Managed Forgetting +<br/>and Contextualized Remembering +<br/>Grant Agreement No. 600826 +<br/>Deliverable D4.4 +<br/>Work-package +<br/>Deliverable +<br/>Deliverable Leader +<br/>Quality Assessor +<br/>Dissemination level +<br/>Delivery date in Annex I +<br/>Actual delivery date +<br/>Revisions +<br/>Status +<br/>Keywords +<br/>Information Consolidation and Con- +<br/>WP4: +<br/>centration +<br/>D4.4: +<br/>Information analysis, consolidation +<br/>and concentration techniques, and evalua- +<br/>tion - Final release. +<br/>Vasileios Mezaris (CERTH) +<br/>Walter Allasia (EURIX) +<br/>PU +<br/>31-01-2016 (M36) +<br/>31-01-2016 +<br/>Final +<br/>multidocument summarization, semantic en- +<br/>richment, +<br/>feature extraction, concept de- +<br/>tection, event detection, image/video qual- +<br/>ity, image/video aesthetic quality, face de- +<br/>tection/clustering, +<br/>im- +<br/>age/video summarization, image/video near +<br/>duplicate detection, data deduplication, con- +<br/>densation, consolidation +<br/>image clustering, +</td><td></td><td></td></tr><tr><td>a1b7bb2a4970b7c479aff3324cc7773c1daf3fc1</td><td>Longitudinal Study of Child Face Recognition +<br/><b>Michigan State University</b><br/>East Lansing, MI, USA +<br/><b>Malaviya National Institute of Technology</b><br/>Jaipur, India +<br/><b>Michigan State University</b><br/>East Lansing, MI, USA +</td><td>('32623642', 'Debayan Deb', 'debayan deb')<br/>('2117075', 'Neeta Nain', 'neeta nain')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>debdebay@msu.edu +<br/>nnain.cse@mnit.ac.in +<br/>jain@cse.msu.edu +</td></tr><tr><td>a14ed872503a2f03d2b59e049fd6b4d61ab4d6ca</td><td>Attentional Pooling for Action Recognition +<br/><b>The Robotics Institute, Carnegie Mellon University</b><br/>http://rohitgirdhar.github.io/AttentionalPoolingAction +</td><td>('3102850', 'Rohit Girdhar', 'rohit girdhar')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td></td></tr><tr><td>a1132e2638a8abd08bdf7fc4884804dd6654fa63</td><td>6 +<br/>Real-Time Video Face Recognition +<br/>for Embedded Devices +<br/>Tessera, Galway, +<br/>Ireland +<br/>1. Introduction +<br/>This chapter will address the challenges of real-time video face recognition systems +<br/>implemented in embedded devices. Topics to be covered include: the importance and +<br/>challenges of video face recognition in real life scenarios, describing a general architecture of +<br/>a generic video face recognition system and a working solution suitable for recognizing +<br/>faces in real-time using low complexity devices. Each component of the system will be +<br/>described together with the system’s performance on a database of video samples that +<br/>resembles real life conditions. +<br/>2. Video face recognition +<br/>Face recognition remains a very active topic in computer vision and receives attention from +<br/>a large community of researchers in that discipline. Many reasons feed this interest; the +<br/>main being the wide range of commercial, law enforcement and security applications that +<br/>require authentication. The progress made in recent years on the methods and algorithms +<br/>for data processing as well as the availability of new technologies makes it easier to study +<br/>these algorithms and turn them into commercially viable product. Biometric based security +<br/>systems are becoming more popular due to their non-invasive nature and their increasing +<br/>reliability. Surveillance applications based on face recognition are gaining increasing +<br/>attention after the United States’ 9/11 events and with the ongoing security threats. The +<br/>Face Recognition Vendor Test (FRVT) (Phillips et al., 2003) includes video face recognition +<br/>testing starting with the 2002 series of tests. +<br/>Recently, face recognition technology was deployed in consumer applications such as +<br/>organizing a collection of images using the faces present in the images (Picassa; Corcoran & +<br/>Costache, 2005), prioritizing family members for best capturing conditions when taking +<br/>pictures, or directly annotating the images as they are captured (Costache et al., 2006). +<br/>Video face recognition, compared with more traditional still face recognition, has the main +<br/>advantage of using multiple instances of the same individual in sequential frames for +<br/>recognition to occur. In still recognition case, the system has only one input image to make +<br/>the decision if the person is or is not in the database. If the image is not suitable for +<br/>recognition (due to face orientation, expression, quality or facial occlusions) the recognition +<br/>result will most likely be incorrect. In the video image there are multiple frames which can +<br/>www.intechopen.com +</td><td>('1706790', 'Petronel Bigioi', 'petronel bigioi')<br/>('1734172', 'Peter Corcoran', 'peter corcoran')</td><td></td></tr><tr><td>a125bc55bdf4bec7484111eea9ae537be314ec62</td><td>Real-time Facial Expression Recognition in Image +<br/>Sequences Using an AdaBoost-based Multi-classifier +<br/><b>National Taiwan University of Science and Technology, Taipei 10607, Taiwan</b><br/><b>National Taiwan University of Science and Technology, Taipei 10607, Taiwan</b><br/><b>National Taiwan University of Science and Technology, Taipei 10607, Taiwan</b><br/> To surmount the shortcomings as stated above, we +<br/>attempt to develop an automatic facial expression recognition +<br/>system that detects human faces and extracts facial features +<br/>from an image sequence. This system is employed for +<br/>recognizing six kinds of facial expressions: joy, anger, +<br/>surprise, fear, sadness, and neutral of a computer user. In the +<br/>expression classification procedure, we mainly compare the +<br/>performance of different classifiers using multi-layer +<br/>perceptions (MLPs), SVMs, and AdaBoost algorithms +<br/>(ABAs). Through evaluating experimental +<br/>the +<br/>performance of ABAs is superior to that of the other two. +<br/>According to this, we develop an AdaBoost-based multi- +<br/>classifier used in our facial expression recognition system. +<br/>results, +<br/>II. FACE AND FACIAL FEATURE DETECTION +<br/>In our system design philosophy, the skin color cue is an +<br/>obvious characteristic to detect human faces. To begin with, +<br/>we will execute skin color detection, then the morphological +<br/>dilation operation, and facial feature detection. Subsequently, +<br/>a filtering operation based on geometrical properties is +<br/>applied to eliminate the skin color regions that do not pertain +<br/>to human faces. +<br/>A. Color Space Transformation +<br/>Face detection is dependent on skin color detection +<br/>techniques which work in one of frequently used color spaces. +<br/>In the past, three color spaces YCbCr, HSI, and RGB have +<br/>been extensively applied for skin color detection. Accordingly, +<br/>we extract the common attribute from skin color regions to +<br/>perform face detection. +<br/>The color model of an image captured from the +<br/>experimental camera is composed of RGB values, but it’s +<br/>easy to be influenced by lighting. Herein, we adopt the HSI +<br/>color space to replace the traditional RGB color space for skin +<br/>color detection. We distinguish skin color regions from non- +<br/>skin color ones by means of lower and upper bound +<br/>thresholds. Via many experiments of detecting human faces, +<br/>we choose the H value between 3 and 38 as the range of skin +<br/>colors. +<br/>B. Connected Component Labeling +<br/>After the processing of skin color detection, we employ +<br/>linear-time connected-component +<br/>technique +<br/>labeling +<br/>the +</td><td>('2574621', 'Chin-Shyurng Fahn', 'chin-shyurng fahn')<br/>('2604646', 'Ming-Hui Wu', 'ming-hui wu')<br/>('2309647', 'Chang-Yi Kao', 'chang-yi kao')</td><td>E-mail: csfahn@mail.ntust.edu.tw Tel: +886-02-2730-1215 +<br/>E-mail: M9415054@mail.ntust.edu.tw Tel: +886-02-2733-3141 ext.7425 +<br/>E-mail: D9515011@mail.ntust.edu.tw Tel: +886-02-2733-3141 ext.7425 +</td></tr><tr><td>a14ae81609d09fed217aa12a4df9466553db4859</td><td>REVISED VERSION, JUNE 2011 +<br/>Face Identification Using Large Feature Sets +</td><td>('1679142', 'William Robson Schwartz', 'william robson schwartz')<br/>('2723427', 'Huimin Guo', 'huimin guo')<br/>('3826759', 'Jonghyun Choi', 'jonghyun choi')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td></td></tr><tr><td>a1f1120653bb1bd8bd4bc9616f85fdc97f8ce892</td><td>Latent Embeddings for Zero-shot Classification +<br/>1MPI for Informatics +<br/>2IIT Kanpur +<br/><b>Saarland University</b></td><td>('3370667', 'Yongqin Xian', 'yongqin xian')<br/>('2893664', 'Zeynep Akata', 'zeynep akata')<br/>('2515597', 'Gaurav Sharma', 'gaurav sharma')<br/>('33460941', 'Matthias Hein', 'matthias hein')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td></td></tr><tr><td>a1ee0176a9c71863d812fe012b5c6b9c15f9aa8a</td><td>Affective recommender systems: the role of emotions in +<br/>recommender systems +<br/>Jurij Tasiˇc +<br/><b>University of Ljubljana Faculty</b><br/><b>University of Ljubljana Faculty</b><br/><b>University of Ljubljana Faculty</b><br/>of electrical engineering +<br/>Tržaška 25, Ljubljana, +<br/>Slovenia +<br/>of electrical engineering +<br/>Tržaška 25, Ljubljana, +<br/>Slovenia +<br/>of electrical engineering +<br/>Tržaška 25, Ljubljana, +<br/>Slovenia +</td><td>('1717186', 'Andrej Košir', 'andrej košir')</td><td>marko.tkalcic@fe.uni-lj.si +<br/>andrej.kosir@fe.uni-lj.si +<br/>jurij.tasic@fe.uni-lj.si +</td></tr><tr><td>a1dd9038b1e1e59c9d564e252d3e14705872fdec</td><td>Attributes as Operators: +<br/>Factorizing Unseen Attribute-Object Compositions +<br/><b>The University of Texas at Austin</b><br/>2 Facebook AI Research +</td><td>('38661780', 'Tushar Nagarajan', 'tushar nagarajan')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>tushar@cs.utexas.edu, grauman@fb.com∗ +</td></tr><tr><td>a1e97c4043d5cc9896dc60ae7ca135782d89e5fc</td><td>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +<br/>Re-identification of Humans in Crowds using +<br/>Personal, Social and Environmental Constraints +</td><td>('2963501', 'Shayan Modiri Assari', 'shayan modiri assari')<br/>('1803711', 'Haroon Idrees', 'haroon idrees')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td></td></tr><tr><td>a16fb74ea66025d1f346045fda00bd287c20af0e</td><td>A Coupled Evolutionary Network for Age Estimation +<br/>National Laboratory of Pattern Recognition, CASIA, Beijing, China 100190 +<br/>Center for Research on Intelligent Perception and Computing, CASIA, Beijing, China 100190 +<br/><b>University of Chinese Academy of Sciences, Beijing, China</b></td><td>('2112221', 'Peipei Li', 'peipei li')<br/>('49995036', 'Yibo Hu', 'yibo hu')<br/>('1705643', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>Email: {peipei.li, yibo.hu}@cripac.ia.ac.cn, {rhe, znsun}@nlpr.ia.ac.cn +</td></tr><tr><td>ef940b76e40e18f329c43a3f545dc41080f68748</td><td> +<br/> +<br/>Research Article Volume 7 Issue No.3 +<br/>ISSN XXXX XXXX © 2017 IJESC +<br/> +<br/>A Face Recognition and Spoofing Detection Adapted to Visually- +<br/>Impaired People +<br/><b>K.K Wagh Institute of Engineering and Education Research, Nashik, India</b><br/>Depart ment of Co mputer Engineering +<br/>Abstrac t: +<br/>According to estimates by the world Health organization, about 285 million people suffer fro m so me kind of v isual disabilit ies of +<br/>which 39 million are blind, resulting in 0.7 of the word population. As many v isual impaired peoples in the word they are unable +<br/>to recognize the people who is standing in front of them and some peoples who have problem to re me mbe r na me of the person. +<br/>They can easily recognize the person using this system. A co mputer vision technique and image ana lysis can help v isually +<br/>the home using face identification and spoofing detection system. This system also provide feature to add newly known people +<br/>and keep records of all peoples visiting their ho me. +<br/>Ke ywor ds: face-recognition, spoofing detection, visually-impaired, system architecture. +<br/>I. +<br/> INTRODUCTION +<br/>The facia l ana lysis can be used to e xtract very useful and +<br/>relevant information in order to help people with visual +<br/>impairment in several of its tasks daily providing them with a +<br/>greater degree of autonomy and security. Facia l recognition +<br/>has received many improve ments recent years and today is +<br/>approaching perfection. The advances in facia l recognition +<br/>have not been outside the People with disab ilities. For +<br/>e xa mple , recently it has an intelligent walking stick for the +<br/>blind that uses facial recognition [5]. The cane co mes +<br/>equipped with a fac ial recognition system, GPS and Bluetooth. +<br/>at the sight the face of any acquaintance or friend whose +<br/>picture is stored on the SD card stick, this will v ibrate and give +<br/>to Bluetooth headset through a necessary instructions to reach +<br/>this person. The system works with anyone who is at 10 +<br/>meters or less. And thanks to the GPS, the user will rece ive +<br/>instructions for reach wherever, as with any GPS navigator. +<br/>However, in addition to the task of recognition today have +<br/>biometric systems to deal with other problems, such as +<br/>spoofing. In network security terms, this term re fers to Using +<br/>techniques through which an attacker, usually with malic ious +<br/>use, it is passed by a other than through the falsification of +<br/>data entity in a co mmun ication. Motivation of the p roject is to +<br/>propose, build and validate an architecture based on face +<br/>recognition and anti-spoofing system that both can be +<br/>integrated in a video entry as a mobile app. In this way, we +<br/>want to give the blind and visually impaired an instrument or +<br/>tool to allo w an ult imate goal to improve the quality of life +<br/>and increase both safety and the feel of it in your ho me or +<br/>when you +<br/>interact with other people. The p roposed +<br/>architecture has been validated with rea l users and a real +<br/>environment simulating the same conditions as could give +<br/>both the images captured by a video portero as images taken +<br/>by a person visually impa ired through their mobile device. +<br/>Contributions are d iscussed below: First an algorith m is +<br/>proposed for the normalization face robust user as to rotations +<br/>and misalignments in the face detection algorith m. It is shown +<br/>that a robust norma lizat ion algorithm you can significantly +<br/>increase the rate of success in a face detection algorithm +<br/>The organizat ion of this document is as follo ws. In Section 2 +<br/>gives literature survey, Section 3 gives details of system +<br/>architecture. In Section 4 gives imp le mentation details. +<br/>Section 5 presents research findings and your analysis of those +<br/>findings. Section 6 concludes the paper. +<br/>II. LITERATURE S URVEY +<br/>A. Facial Rec ognition oriente d visual i mpair ment +<br/>The proble m of face recognition adapted to visually impaired +<br/>people has been investigated in their d ifferent ways. Belo w are +<br/>summarized the work impo rtant, indicating for each the most +<br/>important features that have been motivating development of +<br/>the architecture proposed here. In [6] fac ia l recognition system +<br/>is presented in mobile devices for the visually impaired, but +<br/>meet ings main ly focused on what aspects as visual fie ld +<br/>captured by the mobile focus much of the subject. In [7] +<br/>system developed facial recognition based on Local Binary +<br/>Pattern (LBP) [8]. They co mpared this with other a lternatives +<br/>descriptor (Local Te rnary Pattern [9] or Histogram of +<br/>Gradients [10]) and arrived It concluded that the performance +<br/>is slightly LBP superior, its computational cost is lower and +<br/>representation information is more co mpact. As has been +<br/>mentioned above, in [5] it has developed a system fac ial +<br/>recognition integrated into a cane. In none of these methods is +<br/>carried out detection spoofing, making the system has a +<br/>vulnerability high against such attacks. We believe it is a point +<br/>very important especially in people with visual d isabilities. +<br/>Moreover, none of the alternatives above mentioned is video +<br/>porters oriented. +<br/>B. De tection S poofing +<br/>As none of the above has been studied spoofing detection to +<br/>help people with visual impairment, we will discuss the +<br/>results more significant as +<br/>refers. There are many different methods +<br/>for detecting +<br/>spoofing. However, one o f the key factors in an application +<br/>that must run in rea l time and in a device Embedded is what +<br/>the method be co mputationally lightweight. Most algorith ms +<br/>or proposed are very comple x and are therefo re unfit for rea l, +<br/>far as detecting spoofing +<br/>International Journal of Engineering Science and Computing, March 2017 6051 http://ijesc.org/ +</td><td></td><td></td></tr><tr><td>efd308393b573e5410455960fe551160e1525f49</td><td>Tracking Persons-of-Interest via +<br/>Unsupervised Representation Adaptation +</td><td>('2481388', 'Shun Zhang', 'shun zhang')<br/>('3068086', 'Jia-Bin Huang', 'jia-bin huang')<br/>('33047058', 'Jongwoo Lim', 'jongwoo lim')<br/>('1698965', 'Yihong Gong', 'yihong gong')<br/>('32014778', 'Jinjun Wang', 'jinjun wang')<br/>('1752333', 'Narendra Ahuja', 'narendra ahuja')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>ef230e3df720abf2983ba6b347c9d46283e4b690</td><td>Page 1 of 20 +<br/>QUIS-CAMPI: An Annotated Multi-biometrics Data Feed From +<br/>Surveillance Scenarios +<br/><b>IT - Instituto de Telecomunica es, University of Beira Interior</b><br/><b>University of Beira Interior</b><br/><b>IT - Instituto de Telecomunica es, University of Beira Interior</b></td><td>('1712429', 'Hugo Proença', 'hugo proença')</td><td>*jcneves@ubi.pt +</td></tr><tr><td>ef4ecb76413a05c96eac4c743d2c2a3886f2ae07</td><td>Modeling the Importance of Faces in Natural Images +<br/>Jin B.a, Yildirim G.a, Lau C.a, Shaji A.a, Ortiz Segovia M.b and S¨usstrunk S.a +<br/>aEPFL, Lausanne, Switzerland; +<br/>bOc´e, Paris, France +</td><td></td><td></td></tr><tr><td>efd28eabebb9815e34031316624e7f095c7dfcfe</td><td>A. Uhl and P. Wild. Combining Face with Face-Part Detectors under Gaussian Assumption. In A. Campilho and M. Kamel, +<br/>editors, Proceedings of the 9th International Conference on Image Analysis and Recognition (ICIAR’12), volume 7325 of +<br/>LNCS, pages 80{89, Aveiro, Portugal, June 25{27, 2012. c⃝ Springer. doi: 10.1007/978-3-642-31298-4 10. The original +<br/>publication is available at www.springerlink.com. +<br/>Combining Face with Face-part Detectors +<br/>under Gaussian Assumption⋆ +<br/>Multimedia Signal Processing and Security Lab +<br/><b>University of Salzburg, Austria</b></td><td>('1689850', 'Andreas Uhl', 'andreas uhl')<br/>('2242291', 'Peter Wild', 'peter wild')</td><td>fuhl,pwildg@cosy.sbg.ac.at +</td></tr><tr><td>eff87ecafed67cc6fc4f661cb077fed5440994bb</td><td>Evaluation of Expression Recognition +<br/>Techniques +<br/><b>Beckman Institute, University of Illinois at Urbana-Champaign, USA</b><br/><b>Faculty of Science, University of Amsterdam, The Netherlands</b><br/><b>Leiden Institute of Advanced Computer Science, Leiden University, The Netherlands</b></td><td>('1774778', 'Ira Cohen', 'ira cohen')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1840164', 'Yafei Sun', 'yafei sun')<br/>('1731570', 'Michael S. Lew', 'michael s. lew')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td></td></tr><tr><td>ef458499c3856a6e9cd4738b3e97bef010786adb</td><td>Learning Type-Aware Embeddings for Fashion +<br/>Compatibility +<br/>Department of Computer Science, +<br/><b>University of Illinois at Urbana-Champaign</b></td><td>('47087718', 'Mariya I. Vasileva', 'mariya i. vasileva')<br/>('2856622', 'Bryan A. Plummer', 'bryan a. plummer')<br/>('40895028', 'Krishna Dusad', 'krishna dusad')<br/>('9560882', 'Shreya Rajpal', 'shreya rajpal')<br/>('40439276', 'Ranjitha Kumar', 'ranjitha kumar')</td><td>{mvasile2,bplumme2,dusad2,srajpal2,ranjitha,daf}@illnois.edu +</td></tr><tr><td>ef032afa4bdb18b328ffcc60e2dc5229cc1939bc</td><td>Fang and Yuan EURASIP Journal on Image and Video +<br/>Processing (2018) 2018:44 +<br/>https://doi.org/10.1186/s13640-018-0282-x +<br/>EURASIP Journal on Image +<br/>and Video Processing +<br/>RESEARCH +<br/>Open Access +<br/>Attribute-enhanced metric learning for +<br/>face retrieval +</td><td>('8589942', 'Yuchun Fang', 'yuchun fang')<br/>('30438417', 'Qiulong Yuan', 'qiulong yuan')</td><td></td></tr><tr><td>ef2a5a26448636570986d5cda8376da83d96ef87</td><td>Recurrent Neural Networks and Transfer Learning for Action Recognition +<br/><b>Stanford University</b><br/><b>Stanford University</b></td><td>('11647121', 'Andrew Giel', 'andrew giel')<br/>('32426361', 'Ryan Diaz', 'ryan diaz')</td><td>agiel@stanford.edu +<br/>ryandiaz@stanford.edu +</td></tr><tr><td>ef5531711a69ed687637c48930261769465457f0</td><td>Studio2Shop: from studio photo shoots to fashion articles +<br/>Zalando Research, Muehlenstr. 25, 10243 Berlin, Germany +<br/>Keywords: +<br/>computer vision, deep learning, fashion, item recognition, street-to-shop +</td><td>('46928510', 'Julia Lasserre', 'julia lasserre')<br/>('1724791', 'Katharina Rasch', 'katharina rasch')<br/>('2742129', 'Roland Vollgraf', 'roland vollgraf')</td><td>julia.lasserre@zalando.de +</td></tr><tr><td>ef559d5f02e43534168fbec86707915a70cd73a0</td><td>DING, HUO, HU, LU: DEEPINSIGHT +<br/>DeepInsight: Multi-Task Multi-Scale Deep +<br/>Learning for Mental Disorder Diagnosis +<br/>1 School of Information +<br/><b>Renmin University of China</b><br/>Beijing, 100872, China +<br/>2 Beijing Key Laboratory +<br/>of Big Data Management +<br/>and Analysis Methods +<br/>Beijing, 100872, China +</td><td>('5535865', 'Mingyu Ding', 'mingyu ding')<br/>('4140493', 'Yuqi Huo', 'yuqi huo')<br/>('1745787', 'Jun Hu', 'jun hu')<br/>('1776220', 'Zhiwu Lu', 'zhiwu lu')</td><td>d130143597@163.com +<br/>bnhony@163.com +<br/>junhu@ruc.edu.cn +<br/>luzhiwu@ruc.edu.cn +</td></tr><tr><td>efa08283656714911acff2d5022f26904e451113</td><td>Active Object Localization in Visual Situations +</td><td>('3438473', 'Max H. Quinn', 'max h. quinn')<br/>('13739397', 'Anthony D. Rhodes', 'anthony d. rhodes')<br/>('4421478', 'Melanie Mitchell', 'melanie mitchell')</td><td></td></tr><tr><td>ef8de1bd92e9ee9d0d2dee73095d4d348dc54a98</td><td>Fine-grained Activity Recognition +<br/>with Holistic and Pose based Features +<br/><b>Max Planck Institute for Informatics, Germany</b><br/><b>Stanford University, USA</b></td><td>('2299109', 'Leonid Pishchulin', 'leonid pishchulin')<br/>('1906895', 'Mykhaylo Andriluka', 'mykhaylo andriluka')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')</td><td></td></tr><tr><td>ef999ab2f7b37f46445a3457bf6c0f5fd7b5689d</td><td>Calhoun: The NPS Institutional Archive +<br/>DSpace Repository +<br/>Theses and Dissertations +<br/>1. Thesis and Dissertation Collection, all items +<br/>2017-12 +<br/>Improving face verification in photo albums by +<br/>combining facial recognition and metadata +<br/>with cross-matching +<br/>Monterey, California: Naval Postgraduate School +<br/>http://hdl.handle.net/10945/56868 +<br/>Downloaded from NPS Archive: Calhoun +</td><td></td><td></td></tr><tr><td>c32fb755856c21a238857b77d7548f18e05f482d</td><td>Multimodal Emotion Recognition for Human- +<br/>Computer Interaction: A Survey +<br/><b>School of Computer and Communication Engineering, University of Science and Technology Beijing, 100083 Beijing, China</b></td><td>('10692633', 'Michele Mukeshimana', 'michele mukeshimana')<br/>('1714904', 'Xiaojuan Ban', 'xiaojuan ban')<br/>('17056027', 'Nelson Karani', 'nelson karani')<br/>('7247643', 'Ruoyi Liu', 'ruoyi liu')</td><td></td></tr><tr><td>c3beae515f38daf4bd8053a7d72f6d2ed3b05d88</td><td></td><td></td><td></td></tr><tr><td>c3dc4f414f5233df96a9661609557e341b71670d</td><td>Tao et al. EURASIP Journal on Advances in Signal Processing 2011, 2011:4 +<br/>http://asp.eurasipjournals.com/content/2011/1/4 +<br/>RESEARCH +<br/>Utterance independent bimodal emotion +<br/>recognition in spontaneous communication +<br/>Open Access +</td><td>('37670752', 'Jianhua Tao', 'jianhua tao')<br/>('48027528', 'Shifeng Pan', 'shifeng pan')<br/>('2740129', 'Minghao Yang', 'minghao yang')<br/>('3295988', 'Kaihui Mu', 'kaihui mu')<br/>('2253805', 'Jianfeng Che', 'jianfeng che')</td><td></td></tr><tr><td>c3b3636080b9931ac802e2dd28b7b684d6cf4f8b</td><td>International Journal of Security and Its Applications +<br/>Vol. 7, No. 2, March, 2013 +<br/>Face Recognition via Local Directional Pattern +<br/><b>Division of IT Convergence, Daegu Gyeongbuk Institute of Science and Technology</b><br/>50-1, Sang-ri, Hyeonpung-myeon, Dalseong-gun, Daegu, Korea. +</td><td>('2437301', 'Dong-Ju Kim', 'dong-ju kim')<br/>('38107412', 'Sang-Heon Lee', 'sang-heon lee')<br/>('2735120', 'Myoung-Kyu Sohn', 'myoung-kyu sohn')</td><td>*radioguy@dgist.ac.kr +</td></tr><tr><td>c398684270543e97e3194674d9cce20acaef3db3</td><td>Chapter 2 +<br/>Comparative Face Soft Biometrics for +<br/>Human Identification +</td><td>('19249411', 'Nawaf Yousef Almudhahka', 'nawaf yousef almudhahka')<br/>('1727698', 'Mark S. Nixon', 'mark s. nixon')<br/>('31534955', 'Jonathon S. Hare', 'jonathon s. hare')</td><td></td></tr><tr><td>c3285a1d6ec6972156fea9e6dc9a8d88cd001617</td><td></td><td></td><td></td></tr><tr><td>c3418f866a86dfd947c2b548cbdeac8ca5783c15</td><td></td><td></td><td></td></tr><tr><td>c3bcc4ee9e81ce9c5c0845f34e9992872a8defc0</td><td>MVA2005 IAPR Conference on Machine VIsion Applications, May 16-18, 2005 Tsukuba Science City, Japan +<br/>8-10 +<br/>A New Scheme for Image Recognition Using Higher-Order Local +<br/>Autocorrelation and Factor Analysis +<br/><b>yThe University of Tokyo</b><br/>Tokyo, Japan +<br/>yyyAIST +<br/>Tukuba, Japan +</td><td>('29737626', 'Naoyuki Nomoto', 'naoyuki nomoto')<br/>('2163494', 'Yusuke Shinohara', 'yusuke shinohara')<br/>('2981587', 'Takayoshi Shiraki', 'takayoshi shiraki')<br/>('1800592', 'Takumi Kobayashi', 'takumi kobayashi')<br/>('1809629', 'Nobuyuki Otsu', 'nobuyuki otsu')</td><td>f shiraki, takumi, otsug @isi.imi.i.u-tokyo.ac.jp +</td></tr><tr><td>c34532fe6bfbd1e6df477c9ffdbb043b77e7804d</td><td>A 3D Morphable Eye Region Model +<br/>for Gaze Estimation +<br/><b>University of Cambridge, Cambridge, UK</b><br/><b>Carnegie Mellon University, Pittsburgh, USA</b><br/><b>Max Planck Institute for Informatics, Saarbr ucken, Germany</b></td><td>('34399452', 'Erroll Wood', 'erroll wood')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')<br/>('39626495', 'Peter Robinson', 'peter robinson')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')</td><td>{eww23,pr10}@cl.cam.ac.uk +<br/>{tbaltrus,morency}@cs.cmu.edu +<br/>bulling@mpi-inf.mpg.de +</td></tr><tr><td>c394a5dfe5bea5fbab4c2b6b90d2d03e01fb29c0</td><td>Person Reidentification and Recognition in +<br/>Video +<br/>Computer Science and Engineering, +<br/><b>University of South Florida, Tampa, Florida, USA</b><br/>http://figment.csee.usf.edu/ +</td><td>('3110392', 'Rangachar Kasturi', 'rangachar kasturi')</td><td>R1K@cse.usf.edu,rajmadhan@mail.usf.edu +</td></tr><tr><td>c32383330df27625592134edd72d69bb6b5cff5c</td><td>422 +<br/>IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 42, NO. 2, APRIL 2012 +<br/>Intrinsic Illumination Subspace for Lighting +<br/>Insensitive Face Recognition +</td><td>('1686057', 'Chia-Ping Chen', 'chia-ping chen')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')</td><td></td></tr><tr><td>c3a3f7758bccbead7c9713cb8517889ea6d04687</td><td></td><td></td><td></td></tr><tr><td>c32f04ccde4f11f8717189f056209eb091075254</td><td>Analysis and Synthesis of Behavioural Specific +<br/>Facial Motion +<br/><b>A dissertation submitted to the University of Bristol in accordance with the requirements</b><br/>for the degree of Doctor of Philosophy in the Faculty of Engineering, Department of +<br/>Computer Science. +<br/>February 2007 +<br/>71657 words +</td><td>('2903159', 'Lisa Nanette Gralewski', 'lisa nanette gralewski')</td><td></td></tr><tr><td>c30982d6d9bbe470a760c168002ed9d66e1718a2</td><td>Multi-Camera Head Pose Estimation +<br/>Using an Ensemble of Exemplars +<br/><b>University City Blvd., Charlotte, NC</b><br/>Department of Computer Science +<br/><b>University of North Carolina at Charlotte</b></td><td>('1715594', 'Scott Spurlock', 'scott spurlock')<br/>('2549750', 'Peter Malmgren', 'peter malmgren')<br/>('1873911', 'Hui Wu', 'hui wu')<br/>('1690110', 'Richard Souvenir', 'richard souvenir')</td><td>{sspurloc, ptmalmyr, hwu13, souvenir}@uncc.edu +</td></tr><tr><td>c39ffc56a41d436748b9b57bdabd8248b2d28a32</td><td>Residual Attention Network for Image Classification +<br/><b>SenseTime Group Limited, 2Tsinghua University</b><br/><b>The Chinese University of Hong Kong, 4Beijing University of Posts and Telecommunications</b></td><td>('1682816', 'Fei Wang', 'fei wang')<br/>('9563639', 'Mengqing Jiang', 'mengqing jiang')<br/>('40110742', 'Chen Qian', 'chen qian')<br/>('1692609', 'Shuo Yang', 'shuo yang')<br/>('49672774', 'Cheng Li', 'cheng li')<br/>('1720776', 'Honggang Zhang', 'honggang zhang')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>1{wangfei, qianchen, chengli}@sensetime.com, 2jmq14@mails.tsinghua.edu.cn +<br/>3{ys014, xtang}@ie.cuhk.edu.hk, xgwang@ee.cuhk.edu.hk, 4zhhg@bupt.edu.cn +</td></tr><tr><td>c32cd207855e301e6d1d9ddd3633c949630c793a</td><td>On the Effect of Illumination and Face Recognition +<br/>Jeffrey Ho +<br/>Department of CISE +<br/><b>University of Florida</b><br/>Gainesville, FL 32611 +<br/>Department of Computer Science +<br/><b>University of California at San Diego</b><br/>La Jolla, CA 92093 +</td><td>('38998440', 'David Kriegman', 'david kriegman')</td><td>Email: jho@cise.ufl.edu +<br/>Email: kriegman@cs.ucsd.edu +</td></tr><tr><td>c317181fa1de2260e956f05cd655642607520a4f</td><td>Research Article +<br/>Research +<br/>Article for submission to journal +<br/>Subject Areas: +<br/>computer vision, pattern recognition, +<br/>feature descriptor +<br/>Keywords: +<br/>micro-facial expression, expression +<br/>recognition, action unit +<br/>Objective Classes for +<br/>Micro-Facial Expression +<br/>Recognition +<br/><b>Centre for Imaging Sciences, University of</b><br/>Manchester, Manchester, United Kingdom +<br/><b>Sudan University of Science and Technology</b><br/>Khartoum, Sudan +<br/>3School of Computing, Mathematics and Digital +<br/><b>Technology, Manchester Metropolitan University</b><br/>Manchester, United Kingdom +<br/>instead of predicted emotion, +<br/>Micro-expressions are brief spontaneous facial expressions +<br/>that appear on a face when a person conceals an emotion, +<br/>making them different +<br/>to normal facial expressions in +<br/>subtlety and duration. Currently, emotion classes within +<br/>the CASME II dataset are based on Action Units and +<br/>self-reports, creating conflicts during machine learning +<br/>training. We will show that classifying expressions using +<br/>Action Units, +<br/>removes +<br/>the potential bias of human reporting. The proposed +<br/>classes are tested using LBP-TOP, HOOF and HOG 3D +<br/>feature descriptors. The experiments are evaluated on +<br/>two benchmark FACS coded datasets: CASME II and +<br/>SAMM. The best result achieves 86.35% accuracy when +<br/>classifying the proposed 5 classes on CASME II using +<br/>HOG 3D, outperforming the result of the state-of-the- +<br/>art 5-class emotional-based classification in CASME II. +<br/>Results indicate that classification based on Action Units +<br/>provides an objective method to improve micro-expression +<br/>recognition. +<br/>1. Introduction +<br/>A micro-facial expression is revealed when someone attempts +<br/>to conceal their true emotion [1,2]. When they consciously +<br/>realise that a facial expression is occurring, the person may try +<br/>to suppress the facial expression because showing the emotion +<br/>may not be appropriate [3]. Once the suppression has occurred, +<br/>the person may mask over the original facial expression and +<br/>cause a micro-facial expression. In a high-stakes environment, +<br/>these expressions tend to become more likely as there is more +<br/>risk to showing the emotion. +</td><td>('3125772', 'Moi Hoon Yap', 'moi hoon yap')<br/>('36059631', 'Adrian K. Davison', 'adrian k. davison')<br/>('23986818', 'Walied Merghani', 'walied merghani')<br/>('3125772', 'Moi Hoon Yap', 'moi hoon yap')</td><td>e-mail: M.Yap@mmu.ac.uk +</td></tr><tr><td>c30e4e4994b76605dcb2071954eaaea471307d80</td><td></td><td></td><td></td></tr><tr><td>c37a971f7a57f7345fdc479fa329d9b425ee02be</td><td>A Novice Guide towards Human Motion Analysis and Understanding +</td><td>('40360970', 'Ahmed Nabil Mohamed', 'ahmed nabil mohamed')</td><td>dr.ahmed.mohamed@ieee.org +</td></tr><tr><td>c3638b026c7f80a2199b5ae89c8fcbedfc0bd8af</td><td></td><td></td><td></td></tr><tr><td>c32c8bfadda8f44d40c6cd9058a4016ab1c27499</td><td>Unconstrained Face Recognition From a Single +<br/>Image +<br/><b>Siemens Corporate Research, 755 College Road East, Princeton, NJ</b><br/><b>Center for Automation Research (CfAR), University of Maryland, College Park, MD</b><br/>I. INTRODUCTION +<br/>In most situations, identifying humans using faces is an effortless task for humans. Is this true for computers? +<br/>This very question defines the field of automatic face recognition [10], [38], [79], one of the most active research +<br/>areas in computer vision, pattern recognition, and image understanding. Over the past decade, the problem of face +<br/>recognition has attracted substantial attention from various disciplines and has witnessed a skyrocketing growth of +<br/>the literature. Below, we mainly emphasize some key perspectives of the face recognition problem. +<br/>A. Biometric perspective +<br/>Face is a biometric. As a consequence, face recognition finds wide applications in authentication, security, and +<br/>so on. One recent application is the US-VISIT system by the Department of Homeland Security (DHS), collecting +<br/>foreign passengers’ fingerprints and face images. +<br/>Biometric signatures of a person characterize the physiological or behavioral characteristics. Physiological bio- +<br/>metrics are innate or naturally occuring, while behavioral biometrics arise from mannerisms or traits that are learned +<br/>or acquired. Table I lists commonly used biometrics. Biometric technologies provide the foundation for an extensive +<br/>array of highly secure identification and personal verification solutions. Compared to conventional identification and +<br/>verification methods based on personal identification numbers (PINs) or passwords, biometric technologies offer +<br/>many advantages. First, biometrics are individualized traits while passwords may be used or stolen by someone +<br/>other than the authorized user. Also, biometric is very convenient since there is nothing to carry or remember. In +<br/>addition, biometric technologies are becoming more accurate and less expensive. +<br/>Among all biometrics listed in Table I, face is a very unique one because it is the only biometric belonging to +<br/>both physiological and behavioral categories. While the physiological part of the face has been widely exploited +<br/>for face recognition, the behavioral part has not yet been fully investigated. In addition, as reported in [23], [51], +<br/>face enjoys many advantages over other biometrics because it is a natural, non-intrusive, and easy-to-use biometric. +<br/>For example [23], among six biometrics of face, finger, hand, voice, eye, and signature, face biometric ranks the +<br/>June 10, 2008 +<br/>DRAFT +</td><td>('1682187', 'Shaohua Kevin Zhou', 'shaohua kevin zhou')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('34713849', 'Narayanan Ramanathan', 'narayanan ramanathan')</td><td>Email: {shaohua.zhou}@siemens.com, {rama, ramanath}@cfar.umd.edu +</td></tr><tr><td>c3fb2399eb4bcec22723715556e31c44d086e054</td><td>499 +<br/>2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +<br/>978-1-4799-2893-4/14/$31.00 ©2014 IEEE +<br/>1. INTRODUCTION +</td><td></td><td></td></tr><tr><td>c37de914c6e9b743d90e2566723d0062bedc9e6a</td><td>©2016 Society for Imaging Science and Technology +<br/>DOI: 10.2352/ISSN.2470-1173.2016.11.IMAWM-455 +<br/>Joint and Discriminative Dictionary Learning +<br/>Expression Recognition +<br/>for Facial +</td><td>('38611433', 'Sriram Kumar', 'sriram kumar')<br/>('3168309', 'Behnaz Ghoraani', 'behnaz ghoraani')<br/>('32219349', 'Andreas Savakis', 'andreas savakis')</td><td></td></tr><tr><td>c418a3441f992fea523926f837f4bfb742548c16</td><td>A Computer Approach for Face Aging Problems +<br/>Centre for Pattern Recognition and Machine Intelligence, +<br/><b>Concordia University, Canada</b></td><td>('1769788', 'Khoa Luu', 'khoa luu')</td><td>kh_lu@cenparmi.concordia.ca +</td></tr><tr><td>c4fb2de4a5dc28710d9880aece321acf68338fde</td><td>Interactive Generative Adversarial Networks for Facial Expression Generation +<br/>in Dyadic Interactions +<br/><b>University of Central Florida</b><br/>Educational Testing Service +<br/>Saad Khan +<br/>Educational Testing Service +</td><td>('2974242', 'Behnaz Nojavanasghari', 'behnaz nojavanasghari')<br/>('2224875', 'Yuchi Huang', 'yuchi huang')</td><td>behnaz@eecs.ucf.edu +<br/>yhuang001@ets.org +<br/>skhan002@ets.org +</td></tr><tr><td>c44c84540db1c38ace232ef34b03bda1c81ba039</td><td>Cross-Age Reference Coding for Age-Invariant +<br/>Face Recognition and Retrieval +<br/><b>Institute of Information Science, Academia Sinica, Taipei, Taiwan</b><br/><b>National Taiwan University, Taipei, Taiwan</b></td><td>('33970300', 'Bor-Chun Chen', 'bor-chun chen')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')<br/>('1716836', 'Winston H. Hsu', 'winston h. hsu')</td><td></td></tr><tr><td>c4f1fcd0a5cdaad8b920ee8188a8557b6086c1a4</td><td>Int J Comput Vis (2014) 108:3–29 +<br/>DOI 10.1007/s11263-014-0698-4 +<br/>The Ignorant Led by the Blind: A Hybrid Human–Machine Vision +<br/>System for Fine-Grained Categorization +<br/>Received: 7 March 2013 / Accepted: 8 January 2014 / Published online: 20 February 2014 +<br/>© Springer Science+Business Media New York 2014 +</td><td>('3251767', 'Steve Branson', 'steve branson')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td></td></tr><tr><td>c46a4db7247d26aceafed3e4f38ce52d54361817</td><td>A CNN Cascade for Landmark Guided Semantic +<br/>Part Segmentation +<br/><b>School of Computer Science, The University of Nottingham, Nottingham, UK</b></td><td>('34596685', 'Aaron S. Jackson', 'aaron s. jackson')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>{aaron.jackson, michel.valstar, yorgos.tzimiropoulos}@nottingham.ac.uk +</td></tr><tr><td>c43862db5eb7e43e3ef45b5eac4ab30e318f2002</td><td>Provable Self-Representation Based Outlier Detection in a Union of Subspaces +<br/><b>Johns Hopkins University, Baltimore, MD, 21218, USA</b></td><td>('1878841', 'Chong You', 'chong you')<br/>('1780452', 'Daniel P. Robinson', 'daniel p. robinson')</td><td></td></tr><tr><td>c4dcf41506c23aa45c33a0a5e51b5b9f8990e8ad</td><td> Understanding Activity: Learning the Language of Action +<br/> Univ. of Rochester and Maryland +<br/>1.1 Overview +<br/>Understanding observed activity is an important +<br/>problem, both from the standpoint of practical applications, +<br/>and as a central issue in attempting to describe the +<br/>phenomenon of intelligence. On the practical side, there are a +<br/>large number of applications that would benefit from +<br/>improved machine ability to analyze activity. The most +<br/>prominent are various surveillance scenarios. The current +<br/>emphasis on homeland security has brought this issue to the +<br/>forefront, and resulted in considerable work on mostly low- +<br/>level detection schemes. There are also applications in +<br/>medical diagnosis and household assistants that, in the long +<br/>run, may be even more important. In addition, there are +<br/>numerous scientific projects, ranging from monitoring of +<br/>weather conditions to observation of animal behavior that +<br/>would be facilitated by automatic understanding of activity. +<br/>From a scientific standpoint, understanding activity +<br/>understanding is central to understanding intelligence. +<br/>Analyzing what is happening in the environment, and acting +<br/>on the results of that analysis is, to a large extent, what +<br/>natural intelligent systems do, whether they are human or +<br/>animal. Artificial intelligences, if we want them to work with +<br/>people in the natural world, will need commensurate abilities. +<br/>The importance of the problem has not gone unrecognized. +<br/>There is a substantial body of work on various components of +<br/>the problem, most especially on change detection, motion +<br/>analysis, and tracking. More recently, in the context of +<br/>surveillance applications, there have been some preliminary +<br/>efforts to come up with a general ontology of human activity. +<br/>These efforts have largely been top-down in the classic AI +<br/>tradition, and, as with earlier analogous effort in areas such +<br/>as object recognition and scene understanding, have seen +<br/>limited practical application because of the difficulty in +<br/>robustly extracting the putative primitives on which the top- +<br/>down formalism is based. We propose a novel alternative +<br/>approach, where understanding activity is centered on +</td><td>('34344092', 'Randal Nelson', 'randal nelson')<br/>('1697493', 'Yiannis Aloimonos', 'yiannis aloimonos')</td><td></td></tr><tr><td>c42a8969cd76e9f54d43f7f4dd8f9b08da566c5f</td><td>Towards Unconstrained Face Recognition +<br/>Using 3D Face Model +<br/><b>Intelligent Autonomous Systems (IAS), Technical University of Munich, Garching</b><br/><b>Computer Vision Research Group, COMSATS Institute of Information</b><br/>Technology, Lahore +<br/>1Germany +<br/>2Pakistan +<br/>1. Introduction +<br/>Over the last couple of decades, many commercial systems are available to identify human +<br/>faces. However, face recognition is still an outstanding challenge against different kinds of +<br/>real world variations especially facial poses, non-uniform lightings and facial expressions. +<br/>Meanwhile the face recognition technology has extended its role from biometrics and security +<br/>applications to human robot interaction (HRI). Person identity is one of the key tasks while +<br/>interacting with intelligent machines/robots, exploiting the non intrusive system security +<br/>and authentication of the human interacting with the system. This capability further helps +<br/>machines to learn person dependent traits and interaction behavior to utilize this knowledge +<br/>for tasks manipulation. In such scenarios acquired face images contain large variations which +<br/>demands an unconstrained face recognition system. +<br/>Fig. 1. Biometric analysis of past few years has been shown in figure showing the +<br/>contribution of revenue generated by various biometrics. Although AFIS are getting popular +<br/>in current biometric industry but faces are still considered as one of the widely used +<br/>biometrics. +<br/>www.intechopen.com +</td><td>('1725709', 'Zahid Riaz', 'zahid riaz')<br/>('4241648', 'M. Saquib Sarfraz', 'm. saquib sarfraz')<br/>('1746229', 'Michael Beetz', 'michael beetz')</td><td></td></tr><tr><td>c41de506423e301ef2a10ea6f984e9e19ba091b4</td><td>Modeling Attributes from Category-Attribute Proportions +<br/><b>Columbia University</b><br/>2IBM Research +</td><td>('1815972', 'Felix X. Yu', 'felix x. yu')<br/>('29889388', 'Tao Chen', 'tao chen')</td><td>{yuxinnan, taochen, sfchang}@ee.columbia.edu +<br/>{liangliang.cao, mimerler, nccodell, jsmith}@us.ibm.com +</td></tr><tr><td>c4934d9f9c41dbc46f4173aad2775432fe02e0e6</td><td>Workshop track - ICLR 2017 +<br/>GENERALIZATION TO NEW COMPOSITIONS OF KNOWN +<br/>ENTITIES IN IMAGE UNDERSTANDING +<br/><b>Bar Ilan University, Israel</b><br/>Jonathan Berant & +<br/>Amir Globerson +<br/><b>Tel Aviv University</b><br/>Israel +<br/>Vahid Kazemi & +<br/>Gal Chechik +<br/>Google Research, +<br/>Mountain View CA, USA +</td><td>('34815079', 'Yuval Atzmon', 'yuval atzmon')</td><td>yuval.atzmon@biu.ac.il +</td></tr><tr><td>c40c23e4afc81c8b119ea361e5582aa3adecb157</td><td>Coupled Marginal Fisher Analysis for +<br/>Low-resolution Face Recognition +<br/><b>Carnegie Mellon University, Electrical and Computer Engineering</b><br/>5000 Forbes Avenue, Pittsburgh, Pennsylvania, USA 15213 +</td><td>('2883809', 'Stephen Siena', 'stephen siena')<br/>('2232940', 'Vishnu Naresh Boddeti', 'vishnu naresh boddeti')</td><td>ssiena@andrew.cmu.edu +<br/>naresh@cmu.edu +<br/>kumar@ece.cmu.edu +</td></tr><tr><td>c49aed65fcf9ded15c44f9cbb4b161f851c6fa88</td><td>Multiscale Facial Expression Recognition using Convolutional Neural Networks +<br/>IDIAP, Martigny, Switzerland +</td><td>('8745904', 'Beat Fasel', 'beat fasel')</td><td>Beat.Fasel@idiap.ch +</td></tr><tr><td>c466ad258d6262c8ce7796681f564fec9c2b143d</td><td>14-21 +<br/>MVA2013 IAPR International Conference on Machine Vision Applications, May 20-23, 2013, Kyoto, JAPAN +<br/>Pose-Invariant Face Recognition +<br/>Using A Single 3D Reference Model +<br/><b>National Taiwan University of Science and Technology</b><br/>No. 43, Sec.4, Keelung Rd., Taipei, 106, Taiwan +</td><td>('38801529', 'Gee-Sern Hsu', 'gee-sern hsu')<br/>('3329222', 'Hsiao-Chia Peng', 'hsiao-chia peng')</td><td>*jison@mail.ntust.edu.tw +</td></tr><tr><td>ea46951b070f37ad95ea4ed08c7c2a71be2daedc</td><td>Using phase instead of optical flow +<br/>for action recognition +<br/><b>Computer Vision Lab, Delft University of Technology, Netherlands</b><br/><b>Intelligent Sensory Interactive Systems, University of Amsterdam, Netherlands</b></td><td>('9179750', 'Omar Hommos', 'omar hommos')<br/>('37041694', 'Silvia L. Pintea', 'silvia l. pintea')<br/>('1738975', 'Jan C. van Gemert', 'jan c. van gemert')</td><td></td></tr><tr><td>eac6aee477446a67d491ef7c95abb21867cf71fc</td><td>JOURNAL +<br/>A survey of sparse representation: algorithms and +<br/>applications +</td><td>('38448016', 'Zheng Zhang', 'zheng zhang')<br/>('38649019', 'Yong Xu', 'yong xu')<br/>('37081450', 'Jian Yang', 'jian yang')<br/>('1720243', 'Xuelong Li', 'xuelong li')<br/>('1698371', 'David Zhang', 'david zhang')</td><td></td></tr><tr><td>ea079334121a0ba89452036e5d7f8e18f6851519</td><td>UNSUPERVISED INCREMENTAL LEARNING OF DEEP DESCRIPTORS +<br/>FROM VIDEO STREAMS +<br/><b>MICC University of Florence</b></td><td>('2619131', 'Federico Pernici', 'federico pernici')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')</td><td>federico.pernici@unifi.it, alberto.delbimbo@unifi.it +</td></tr><tr><td>eac1b644492c10546a50f3e125a1f790ec46365f</td><td>Chained Multi-stream Networks Exploiting Pose, Motion, and Appearance for +<br/>Action Classification and Detection +<br/><b>University of Freiburg</b><br/>Freiburg im Breisgau, Germany +</td><td>('2890820', 'Mohammadreza Zolfaghari', 'mohammadreza zolfaghari')<br/>('2371771', 'Gabriel L. Oliveira', 'gabriel l. oliveira')<br/>('31656404', 'Nima Sedaghat', 'nima sedaghat')<br/>('1710872', 'Thomas Brox', 'thomas brox')</td><td>{zolfagha,oliveira,nima,brox}@cs.uni-freiburg.de +</td></tr><tr><td>ea80a050d20c0e24e0625a92e5c03e5c8db3e786</td><td>Face Verification and Face Image Synthesis +<br/>under Illumination Changes +<br/>using Neural Networks +<br/>by +<br/>Under the supervision of +<br/>Prof. Daphna Weinshall +<br/>School of Computer Science and Engineering +<br/><b>The Hebrew University of Jerusalem</b><br/>Israel +<br/>Submitted in partial fulfillment of the +<br/>requirements of the degree of +<br/>Master of Science +<br/>December, 2017 +</td><td></td><td></td></tr><tr><td>eacba5e8fbafb1302866c0860fc260a2bdfff232</td><td>VOS-GAN: Adversarial Learning of Visual-Temporal +<br/>Dynamics for Unsupervised Dense Prediction in Videos +<br/>∗ Pattern Recognition and Computer Vision (PeRCeiVe) Lab +<br/><b>University of Catania, Italy</b><br/>www.perceivelab.com +<br/>§ Center for Research in Computer Vision +<br/><b>University of Central Florida, USA</b><br/>http://crcv.ucf.edu +</td><td>('31411067', 'C. Spampinato', 'c. spampinato')<br/>('35323264', 'S. Palazzo', 's. palazzo')<br/>('2004177', 'F. Murabito', 'f. murabito')<br/>('1690194', 'D. Giordano', 'd. giordano')<br/>('1797029', 'M. Shah', 'm. shah')</td><td></td></tr><tr><td>ea482bf1e2b5b44c520fc77eab288caf8b3f367a</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>2592 +</td><td></td><td></td></tr><tr><td>ea6f5c8e12513dbaca6bbdff495ef2975b8001bd</td><td>Applying a Set of Gabor Filter to 2D-Retinal Fundus Image +<br/>to Detect the Optic Nerve Head (ONH) +<br/>1Higher National School of engineering of Tunis, ENSIT, Laboratory LATICE (Information Technology and Communication and +<br/><b>Electrical Engineering LR11ESO4), University of Tunis EL Manar. Adress: ENSIT 5, Avenue Taha Hussein, B. P. : 56, Bab</b><br/><b>Menara, 1008 Tunis; 2University of Tunis El-Manar, Tunis with expertise in Mechanic, Optics, Biophysics, Conference Master</b><br/><b>ISTMT, Laboratory of Research in Biophysics and Medical Technologies LRBTM Higher Institute of Medical Technologies of Tunis</b><br/><b>ISTMT, University of Tunis El Manar Address: 9, Rue Docteur Zouhe r Safi 1006; 3Faculty of Medicine of Tunis; Address</b><br/>Rue Djebel Lakhdhar. La Rabta. 1007, Tunis - Tunisia +<br/>Corresponding author: +<br/><b>High Institute of Medical Technologies</b><br/>of Tunis, ISTMT, and High National +<br/>School Engineering of Tunis, +<br/>Information Technology and +<br/>Communication Technology and +<br/><b>Electrical Engineering, University of</b><br/>Tunis El-Manar, ENSIT 5, Avenue Taha +<br/>Hussein, B. P.: 56, Bab Menara, 1008 +<br/>Tunis, Tunisia, +<br/>Tel: 9419010363; +</td><td>('9304667', 'Hédi Trabelsi', 'hédi trabelsi')<br/>('2281259', 'Ines Malek', 'ines malek')<br/>('31649078', 'Imed Jabri', 'imed jabri')</td><td>E-mail: rabelg@live.fr +</td></tr><tr><td>eafda8a94e410f1ad53b3e193ec124e80d57d095</td><td>Jeffrey F. Cohn +<br/>13 +<br/>Observer-Based Measurement of Facial Expression +<br/>With the Facial Action Coding System +<br/>Facial expression has been a focus of emotion research for over +<br/>a hundred years (Darwin, 1872/1998). It is central to several +<br/>leading theories of emotion (Ekman, 1992; Izard, 1977; +<br/>Tomkins, 1962) and has been the focus of at times heated +<br/>debate about issues in emotion science (Ekman, 1973, 1993; +<br/>Fridlund, 1992; Russell, 1994). Facial expression figures +<br/>prominently in research on almost every aspect of emotion, +<br/>including psychophysiology (Levenson, Ekman, & Friesen, +<br/>1990), neural bases (Calder et al., 1996; Davidson, Ekman, +<br/>Saron, Senulis, & Friesen, 1990), development (Malatesta, +<br/>Culver, Tesman, & Shephard, 1989; Matias & Cohn, 1993), +<br/>perception (Ambadar, Schooler, & Cohn, 2005), social pro- +<br/>cesses (Hatfield, Cacioppo, & Rapson, 1992; Hess & Kirouac, +<br/>2000), and emotion disorder (Kaiser, 2002; Sloan, Straussa, +<br/>Quirka, & Sajatovic, 1997), to name a few. +<br/>Because of its importance to the study of emotion, a num- +<br/>ber of observer-based systems of facial expression measure- +<br/>ment have been developed (Ekman & Friesen, 1978, 1982; +<br/>Ekman, Friesen, & Tomkins, 1971; Izard, 1979, 1983; Izard +<br/>& Dougherty, 1981; Kring & Sloan, 1991; Tronick, Als, & +<br/>Brazelton, 1980). Of these various systems for describing +<br/>facial expression, the Facial Action Coding System (FACS; +<br/>Ekman & Friesen, 1978; Ekman, Friesen, & Hager, 2002) is +<br/>the most comprehensive, psychometrically rigorous, and +<br/>widely used (Cohn & Ekman, 2005; Ekman & Rosenberg, +<br/>2005). Using FACS and viewing video-recorded facial behav- +<br/>ior at frame rate and slow motion, coders can manually code +<br/>nearly all possible facial expressions, which are decomposed +<br/>into action units (AUs). Action units, with some qualifica- +<br/>tions, are the smallest visually discriminable facial move- +<br/>ments. By comparison, other systems are less thorough +<br/>(Malatesta et al., 1989), fail to differentiate between some +<br/>anatomically distinct movements (Oster, Hegley, & Nagel, +<br/>1992), consider movements that are not anatomically dis- +<br/>tinct as separable (Oster et al., 1992), and often assume a one- +<br/>to-one mapping between facial expression and emotion (for +<br/>a review of these systems, see Cohn & Ekman, in press). +<br/>Unlike systems that use emotion labels to describe ex- +<br/>pression, FACS explicitly distinguishes between facial actions +<br/>and inferences about what they mean. FACS itself is descrip- +<br/>tive and includes no emotion-specified descriptors. Hypoth- +<br/>eses and inferences about the emotional meaning of facial +<br/>actions are extrinsic to FACS. If one wishes to make emo- +<br/>tion-based inferences from FACS codes, a variety of related +<br/>resources exist. These include the FACS Investigators’ Guide +<br/>(Ekman et al., 2002), the FACS interpretive database (Ekman, +<br/>Rosenberg, & Hager, 1998), and a large body of empirical +<br/>research.(Ekman & Rosenberg, 2005). These resources sug- +<br/>gest combination rules for defining emotion-specified expres- +<br/>sions from FACS action units, but this inferential step remains +<br/>extrinsic to FACS. Because of its descriptive power, FACS +<br/>is regarded by many as the standard measure for facial be- +<br/>havior and is used widely in diverse fields. Beyond emo- +<br/>tion science, these include facial neuromuscular disorders +<br/>(Van Swearingen & Cohn, 2005), neuroscience (Bruce & +<br/>Young, 1998; Rinn, 1984, 1991), computer vision (Bartlett, +<br/>203 +<br/>UNPROOFED PAGES </td><td>('2059653', 'Zara Ambadar', 'zara ambadar')<br/>('21451088', 'Paul Ekman', 'paul ekman')</td><td></td></tr><tr><td>ea85378a6549bb9eb9bcc13e31aa6a61b655a9af</td><td>Diplomarbeit +<br/>Template Protection for PCA-LDA-based 3D +<br/>Face Recognition System +<br/>von +<br/>Technische Universität Darmstadt +<br/>Fachbereich Informatik +<br/>Fachgebiet Graphisch-Interaktive Systeme +<br/>Fraunhoferstraße 5 +<br/>64283 Darmstadt +</td><td>('1788102', 'Daniel Hartung', 'daniel hartung')<br/>('35069235', 'Xuebing Zhou', 'xuebing zhou')<br/>('1734569', 'Dieter W. Fellner', 'dieter w. fellner')</td><td></td></tr><tr><td>ea2ee5c53747878f30f6d9c576fd09d388ab0e2b</td><td>Viola-Jones based Detectors: How much affects +<br/>the Training Set? +<br/>SIANI +<br/>Edif. Central del Parque Cient´ıfico Tecnol´ogico +<br/>Universidad de Las Palmas de Gran Canaria +<br/>35017 - Spain +</td><td>('4643134', 'Javier Lorenzo-Navarro', 'javier lorenzo-navarro')</td><td></td></tr><tr><td>ea890846912f16a0f3a860fce289596a7dac575f</td><td>ORIGINAL RESEARCH ARTICLE +<br/>published: 09 October 2014 +<br/>doi: 10.3389/fpsyg.2014.01154 +<br/>Benefits of social vs. non-social feedback on learning and +<br/>generosity. Results from theTipping Game +<br/><b>Tilburg Center for Logic, General Ethics, and Philosophy of Science, Tilburg University, Tilburg, Netherlands</b><br/><b>Institute for Adaptive and Neural Computation, University of Edinburgh, Edinburgh, UK</b><br/>Edited by: +<br/><b>Giulia Andrighetto, Institute of</b><br/>Cognitive Science and Technologies – +<br/>National Research Council, Italy +<br/>Reviewed by: +<br/><b>David R. Simmons, University of</b><br/>Glasgow, UK +<br/><b>Aron Szekely, University of Oxford, UK</b><br/>*Correspondence: +<br/>Logic, General Ethics, and Philosophy +<br/><b>of Science, Tilburg University</b><br/>P. O. Box 90153, 5000 LE +<br/>Tilburg, Netherlands +<br/>Stankevicius have contributed equally +<br/>to this work. +<br/>Although much work has recently been directed at understanding social decision-making, +<br/>relatively little is known about how different types of feedback impact adaptive changes +<br/>in social behavior. To address this issue quantitatively, we designed a novel associative +<br/>learning task called the “Tipping Game,” in which participants had to learn a social norm +<br/>of tipping in restaurants. Participants were found to make more generous decisions +<br/>from feedback in the form of facial expressions, +<br/>in comparison to feedback in the +<br/>form of symbols such as ticks and crosses. Furthermore, more participants displayed +<br/>learning in the condition where they received social feedback than participants in the non- +<br/>social condition. Modeling results showed that the pattern of performance displayed by +<br/>participants receiving social feedback could be explained by a lower sensitivity to economic +<br/>costs. +<br/>Keywords: social/non-social feedback, facial expressions, social norms, tipping behavior, associative learning +<br/>INTRODUCTION +<br/>Several behavioral, neurobiological and theoretical studies have +<br/>shown that social norm compliance, and more generally adap- +<br/>tive changes in social behavior, often require the effective use and +<br/><b>weighing of different types of information, including expected</b><br/>economic costs and benefits, the potential impact of our behavior +<br/>on the welfare of others and our own reputation, as well as feed- +<br/>back information (Bicchieri, 2006; Adolphs, 2009; Frith and Frith, +<br/>2012). Relatively little attention has been paid to how different +<br/>types of feedback (or reward) may impact the way social norms +<br/>are learned. The present study addresses this issue with behavioral +<br/>and modeling results from a novel associative learning task called +<br/>the “Tipping Game.” We take the example of tipping and ask: how +<br/>do social feedback in the form of facial expressions, as opposed +<br/>to non-social feedback in the form of such conventional signs as +<br/>ticks and crosses, affect the way participants learn a social norm +<br/>of tipping? +<br/>Recent findings indicate that people’s decision-making is often +<br/>biased by social stimuli. For example, images of a pair of eyes can +<br/>significantly increase pro-social behavior in laboratory conditions +<br/>as well as in real-world contexts (Haley and Fessler, 2005; Bateson +<br/>et al., 2006; Rigdon et al., 2009; Ernest-Jones et al., 2011). Fur- +<br/>thermore, decision-making can be systematically biased by facial +<br/>emotional expressions used as predictors of monetary reward +<br/>(Averbeck and Duchaine, 2009; Evans et al., 2011; Shore and +<br/>Heerey, 2011). Facial expressions of happiness elicit approach- +<br/>ing behavior, whereas angry faces elicit avoidance (Seidel et al., +<br/>2010; for a review seeBlair, 2003). Because they can function as +<br/>signals to others, eliciting specific behavioral responses, emotional +<br/>facial expressions play a major role in socialization practices that +<br/>help individuals to adapt to the norms and values of their culture +<br/>(Keltner and Haidt, 1999; Frith, 2009). +<br/>Despite this body of findings, the literature does not pro- +<br/>vide an unambiguous answer to the question of how learning +<br/>performance is affected by social stimuli in comparison to differ- +<br/>ent types of non-social stimuli used as feedback about previous +<br/>decisions in a learning task (Ruff and Fehr, 2014). Consistent +<br/>with the view that social reinforcement is a powerful facili- +<br/>tator of human learning (Zajonc, 1965; Bandura, 1977), one +<br/>recent study using a feedback-guided item-category association +<br/>task found that learning performance in control groups was +<br/>improved when social (smiling or angry faces) instead of non- +<br/>social (green or red lights) reinforcement was used (Hurlemann +<br/>et al., 2010). +<br/>However, the paradigm used in this study did not distin- +<br/>guish between two conditions in which social-facilitative effects +<br/>on learning performance have been observed: first, a condition +<br/>characterized by the mere presence of others (Allport, 1920); and +<br/>second, a condition where others provide reinforcing feedback +<br/>(Zajonc, 1965). In the task used by Hurlemann et al. (2010), faces +<br/>were present onscreen throughout each trial, changing from a +<br/>neutral to a happy expression for correct responses or angry for +<br/>incorrect responses. So, this study could not identify the specific +<br/>effect of social feedback on learning. +<br/>Consistent with the assumption oft made in economics and +<br/>psychology that optimal decisions and learning are based on an +<br/>assessment of the evidence that is unbiased by the social or non- +<br/>social nature of the evidence itself (Becker, 1976; Oaksford and +<br/>Chater, 2007), Lin et al. (2012a) found that, instead of boosting +<br/>learning performance, social reward (smiling or angry faces) made +<br/>www.frontiersin.org +<br/>October 2014 | Volume 5 | Article 1154 | 1 +</td><td>('37157064', 'Matteo Colombo', 'matteo colombo')<br/>('25749361', 'Aistis Stankevicius', 'aistis stankevicius')<br/>('2771872', 'Peggy Seriès', 'peggy seriès')<br/>('37157064', 'Matteo Colombo', 'matteo colombo')<br/>('37157064', 'Matteo Colombo', 'matteo colombo')</td><td>e-mail: m.colombo@uvt.nl +</td></tr><tr><td>eaaed082762337e7c3f8a1b1dfea9c0d3ca281bf</td><td><b>VICTORIA UNIVERSITY OF WELLINGTON</b><br/>Te Whare Wananga o te Upoko o te Ika a Maui +<br/>School of Mathematics, Statistics and Computer Science +<br/>Computer Science +<br/>Algebraic Simplification of Genetic +<br/>Programs during Evolution +<br/>Technical Report CS-TR-06/7 +<br/>February 2006 +<br/>School of Mathematics, Statistics and Computer Science +<br/><b>Victoria University</b><br/>PO Box 600, Wellington +<br/>New Zealand +<br/>Tel: +64 4 463 5341 +<br/>Fax: +64 4 463 5045 +<br/>http://www.mcs.vuw.ac.nz/research +</td><td>('1679067', 'Mengjie Zhang', 'mengjie zhang')</td><td>Email: Tech.Reports@mcs.vuw.ac.nz +</td></tr><tr><td>ea218cebea2228b360680cb85ca133e8c2972e56</td><td>Recover Canonical-View Faces in the 明Tild with Deep +<br/>Neural Networks +<br/><b>Departm nt of Information Engin ering Th Chines University of Hong Kong</b><br/><b>The Chinese University ofHong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b><br/>zz 012 日 ie . cuh k. edu . h k +</td><td>('2042558', 'Zhenyao Zhu', 'zhenyao zhu')<br/>('1693209', 'Ping Luo', 'ping luo')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>xgwang@ee . cuhk . edu . hk +<br/>p 1 uo .1 h 工 @gm a i l . com +<br/>xtang@ i e . cuhk. edu . hk +</td></tr><tr><td>ea96bc017fb56593a59149e10d5f14011a3744a0</td><td></td><td></td><td></td></tr><tr><td>e1630014a5ae3d2fb7ff6618f1470a567f4d90f5</td><td>Look, Listen and Learn - A Multimodal LSTM for Speaker Identification +<br/>SenseTime Group Limited1 +<br/><b>The University of Hong Kong</b><br/>Project page: http://www.deeplearning.cc/mmlstm +</td><td>('46972608', 'Yongtao Hu', 'yongtao hu')</td><td>{rensijie, yuwing, xuli, sunwenxiu, yanqiong}@sensetime.com +<br/>{herohuyongtao, wangchuan2400}@gmail.com +</td></tr><tr><td>e19fb22b35c352f57f520f593d748096b41a4a7b</td><td>Modeling Context for Image +<br/>Understanding: +<br/>When, For What, and How? +<br/>Department of Electrical and Computer Engineering, +<br/><b>Carnegie Mellon University</b><br/>A thesis submitted for the degree of +<br/>Doctor of Philosophy +<br/>April 3, 2009 +</td><td>('1713589', 'Devi Parikh', 'devi parikh')</td><td></td></tr><tr><td>e10a257f1daf279e55f17f273a1b557141953ce2</td><td></td><td></td><td></td></tr><tr><td>e171fba00d88710e78e181c3e807c2fdffc6798a</td><td></td><td></td><td></td></tr><tr><td>e1c59e00458b4dee3f0e683ed265735f33187f77</td><td>Spectral Rotation versus K-Means in Spectral Clustering +<br/>Computer Science and Engineering Department +<br/><b>University of Texas at Arlington</b><br/>Arlington,TX,76019 +</td><td>('39122448', 'Jin Huang', 'jin huang')<br/>('1688370', 'Feiping Nie', 'feiping nie')<br/>('1748032', 'Heng Huang', 'heng huang')</td><td>huangjinsuzhou@gmail.com, feipingnie@gmail.com, heng@uta.edu +</td></tr><tr><td>e1f790bbedcba3134277f545e56946bc6ffce48d</td><td> +<br/>International Journal of Innovative Research in Science, +<br/>Engineering and Technology +<br/>(An ISO 3297: 2007 Certified Organization) +<br/> Vol. 3, Issue 5, May 2014 +<br/>Sparse Code Words +<br/> +<br/> +<br/> ISSN: 2319-8753 +<br/>Image Retrieval Using Attribute Enhanced +<br/><b>SRV Engineering College, sembodai, india</b><br/><b>P.G. Student, SRV Engineering College, sembodai, India</b></td><td>('5768860', 'M.Balaganesh', 'm.balaganesh')<br/>('14176059', 'N.Arthi', 'n.arthi')</td><td></td></tr><tr><td>e1ab3b9dee2da20078464f4ad8deb523b5b1792e</td><td>Pre-Training CNNs Using Convolutional +<br/>Autoencoders +<br/>TU Berlin +<br/>TU Berlin +<br/>Sabbir Ahmmed +<br/>TU Berlin +<br/>TU Berlin +</td><td>('16258861', 'Maximilian Kohlbrenner', 'maximilian kohlbrenner')<br/>('40805229', 'Russell Hofmann', 'russell hofmann')<br/>('3196053', 'Youssef Kashef', 'youssef kashef')</td><td>m.kohlbrenner@campus.tu-berlin.de +<br/>r.hofmann@campus.tu-berlin.de +<br/>ahmmed@campus.tu-berlin.de +<br/>kashefy@ni.tu-berlin.de +</td></tr><tr><td>e16efd2ae73a325b7571a456618bfa682b51aef8</td><td></td><td></td><td></td></tr><tr><td>e19ebad4739d59f999d192bac7d596b20b887f78</td><td>Learning Gating ConvNet for Two-Stream based Methods in Action +<br/>Recognition +</td><td>('1696573', 'Jiagang Zhu', 'jiagang zhu')<br/>('1726367', 'Wei Zou', 'wei zou')<br/>('48147901', 'Zheng Zhu', 'zheng zhu')</td><td></td></tr><tr><td>e13360cda1ebd6fa5c3f3386c0862f292e4dbee4</td><td></td><td></td><td></td></tr><tr><td>e1f6e2651b7294951b5eab5d2322336af1f676dc</td><td>Appl. Math. Inf. Sci. 9, No. 2L, 461-469 (2015) +<br/>461 +<br/>Applied Mathematics & Information Sciences +<br/>An International Journal +<br/>http://dx.doi.org/10.12785/amis/092L21 +<br/>Emotional Avatars: Appearance Augmentation and +<br/>Animation based on Facial Expression Analysis +<br/><b>Sejong University, 98 Gunja, Gwangjin, Seoul 143-747, Korea</b><br/>Received: 22 May 2014, Revised: 23 Jul. 2014, Accepted: 24 Jul. 2014 +<br/>Published online: 1 Apr. 2015 +</td><td>('2137943', 'Taehoon Cho', 'taehoon cho')<br/>('4027010', 'Jin-Ho Choi', 'jin-ho choi')<br/>('2849238', 'Hyeon-Joong Kim', 'hyeon-joong kim')<br/>('7236280', 'Soo-Mi Choi', 'soo-mi choi')</td><td></td></tr><tr><td>e1d726d812554f2b2b92cac3a4d2bec678969368</td><td>J Electr Eng Technol.2015; 10(?): 30-40 +<br/>http://dx.doi.org/10.5370/JEET.2015.10.2.030 +<br/>ISSN(Print) +<br/>1975-0102 +<br/>ISSN(Online) 2093-7423 +<br/>Human Action Recognition Bases on Local Action Attributes +<br/>and Mohan S Kankanhalli** +</td><td>('3132751', 'Weizhi Nie', 'weizhi nie')<br/>('3026404', 'Yongkang Wong', 'yongkang wong')</td><td></td></tr><tr><td>e1256ff535bf4c024dd62faeb2418d48674ddfa2</td><td>Towards Open-Set Identity Preserving Face Synthesis +<br/><b>University of Science and Technology of China</b><br/>2Microsoft Research +</td><td>('3093568', 'Jianmin Bao', 'jianmin bao')<br/>('39447786', 'Dong Chen', 'dong chen')<br/>('1716835', 'Fang Wen', 'fang wen')<br/>('7179232', 'Houqiang Li', 'houqiang li')<br/>('1745420', 'Gang Hua', 'gang hua')</td><td>{doch, fangwen, ganghua}@microsoft.com +<br/>lihq@ustc.edu.cn +<br/>jmbao@mail.ustc.edu.cn +</td></tr><tr><td>e1e6e6792e92f7110e26e27e80e0c30ec36ac9c2</td><td>TSINGHUA SCIENCE AND TECHNOLOGY +<br/>ISSNll1007-0214 +<br/>0?/?? pp???–??? +<br/>DOI: 10.26599/TST.2018.9010000 +<br/>Volume 1, Number 1, Septembelr 2018 +<br/>Ranking with Adaptive Neighbors +</td><td>('39021559', 'Muge Li', 'muge li')<br/>('2897748', 'Liangyue Li', 'liangyue li')<br/>('1688370', 'Feiping Nie', 'feiping nie')</td><td></td></tr><tr><td>cd9666858f6c211e13aa80589d75373fd06f6246</td><td>A Novel Time Series Kernel for +<br/>Sequences Generated by LTI Systems +<br/>V.le delle Scienze Ed.6, DIID, Universit´a degli studi di Palermo, Italy +</td><td>('1711610', 'Liliana Lo Presti', 'liliana lo presti')<br/>('9127836', 'Marco La Cascia', 'marco la cascia')</td><td></td></tr><tr><td>cdc7bd87a2c9983dab728dbc8aac74d8c9ed7e66</td><td>What Makes a Video a Video: Analyzing Temporal Information in Video +<br/>Understanding Models and Datasets +<br/><b>Stanford University, 2Facebook, 3Dartmouth College</b></td><td>('38485317', 'De-An Huang', 'de-an huang')<br/>('34066479', 'Vignesh Ramanathan', 'vignesh ramanathan')<br/>('49274550', 'Dhruv Mahajan', 'dhruv mahajan')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')<br/>('2210374', 'Manohar Paluri', 'manohar paluri')<br/>('9200530', 'Juan Carlos Niebles', 'juan carlos niebles')</td><td></td></tr><tr><td>cd4941cbef1e27d7afdc41b48c1aff5338aacf06</td><td>MovieGraphs: Towards Understanding Human-Centric Situations from Videos +<br/><b>University of Toronto</b><br/><b>Vector Institute</b><br/>Lluís Castrejón3 +<br/><b>Montreal Institute for Learning Algorithms</b><br/>http://moviegraphs.cs.toronto.edu +<br/>Figure 1: An example from the MovieGraphs dataset. Each of the 7637 video clips is annotated with: 1) a graph that captures the characters +<br/>in the scene and their attributes, interactions (with topics and reasons), relationships, and time stamps; 2) a situation label that captures the +<br/>overarching theme of the interactions; 3) a scene label showing where the action takes place; and 4) a natural language description of the +<br/>clip. The graphs at the bottom show situations that occur before and after the one depicted in the main panel. +</td><td>('2039154', 'Paul Vicol', 'paul vicol')<br/>('2103464', 'Makarand Tapaswi', 'makarand tapaswi')<br/>('37895334', 'Sanja Fidler', 'sanja fidler')</td><td>{pvicol, makarand, fidler}@cs.toronto.edu, lluis.enric.castrejon.subira@umontreal.ca +</td></tr><tr><td>cd4c047f4d4df7937aff8fc76f4bae7718004f40</td><td></td><td></td><td></td></tr><tr><td>cdef0eaff4a3c168290d238999fc066ebc3a93e8</td><td>CONTRASTIVE-CENTER LOSS FOR DEEP NEURAL NETWORKS +<br/>1School of Information and Communication Engineering +<br/>2Beijing Key Laboratory of Network System and Network Culture +<br/><b>Beijing University of Posts and Telecommunications, Beijing, China</b></td><td>('49712251', 'Ce Qi', 'ce qi')<br/>('1684263', 'Fei Su', 'fei su')</td><td></td></tr><tr><td>cd444ee7f165032b97ee76b21b9ff58c10750570</td><td><b>UNIVERSITY OF CALIFORNIA</b><br/>IRVINE +<br/>Relational Models for Human-Object Interactions and Object Affordances +<br/>DISSERTATION +<br/>submitted in partial satisfaction of the requirements +<br/>for the degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>in Computer Science +<br/>by +<br/>Dissertation Committee: +<br/>Professor Deva Ramanan, Chair +<br/>Professor Charless Fowlkes +<br/>Professor Padhraic Smyth +<br/>Professor Serge Belongie +<br/>2013 +</td><td>('40277674', 'Chaitanya Desai', 'chaitanya desai')</td><td></td></tr><tr><td>cd23dc3227ee2a3ab0f4de1817d03ca771267aeb</td><td>WU, KAMATA, BRECKON: FACE RECOGNITION VIA DSGNN +<br/>Face Recognition via Deep Sparse Graph +<br/>Neural Networks +<br/>Renjie WU1 +<br/>Toby Breckon2 +<br/>1 Graduate School of Information, +<br/>Production and Systems +<br/><b>Waseda University</b><br/>Kitakyushu-shi, Japan +<br/>2 Engineering and Computing Sciences +<br/><b>Durham University, Durham, UK</b></td><td>('35222422', 'Sei-ichiro Kamata', 'sei-ichiro kamata')</td><td>wurj-sjtu-waseda@ruri.waseda.jp +<br/>kam@waseda.jp +<br/>toby.breckon@durham.ac.uk +</td></tr><tr><td>cd596a2682d74bdfa7b7160dd070b598975e89d9</td><td>Mood Detection: Implementing a facial +<br/>expression recognition system +<br/>1. Introduction +<br/>Facial expressions play a significant role in human dialogue. As a result, there has been +<br/>considerable work done on the recognition of emotional expressions and the application of this +<br/>research will be beneficial in improving human-machine dialogue. One can imagine the +<br/>improvements to computer interfaces, automated clinical (psychological) research or even +<br/>interactions between humans and autonomous robots. +<br/>Unfortunately, a lot of the literature does not focus on trying to achieve high recognition rates +<br/>across multiple databases. In this project we develop our own mood detection system that +<br/>addresses this challenge. The system involves pre-processing image data by normalizing and +<br/>applying a simple mask, extracting certain (facial) features using PCA and Gabor filters and then +<br/>using SVMs for classification and recognition of expressions. Eigenfaces for each class are used +<br/>to determine class-specific masks which are then applied to the image data and used to train +<br/>multiple, one against the rest, SVMs. We find that simply using normalized pixel intensities +<br/>works well with such an approach. +<br/>Figure 1 – Overview of our system design +<br/>2. Image pre-processing +<br/>We performed pre-processing on the images used to train and test our algorithms as follows: +<br/>1. The location of the eyes is first selected manually +<br/>2. Images are scaled and cropped to a fixed size (170 x 130) keeping the eyes in all images +<br/>aligned +<br/>3. The image is histogram equalized using the mean histogram of all the training images to +<br/>make it invariant to lighting, skin color etc. +<br/>4. A fixed oval mask is applied to the image to extract face region. This serves to eliminate +<br/>the background, hair, ears and other extraneous features in the image which provide no +<br/>information about facial expression. +<br/>This approach works reasonably well in capturing expression-relevant facial information across +<br/>all databases. Examples of pre-processed images from the various datasets are shown in Figure- +<br/>2a below. +</td><td>('1906123', 'Neeraj Agrawal', 'neeraj agrawal')<br/>('2929557', 'Rob Cosgriff', 'rob cosgriff')<br/>('2594170', 'Ritvik Mudur', 'ritvik mudur')</td><td></td></tr><tr><td>cdb1d32bc5c1a9bb0d9a5b9c9222401eab3e9ca0</td><td>Functional Faces: Groupwise Dense Correspondence using Functional Maps +<br/><b>The University of York, UK</b><br/>2IMB/LaBRI, Universit´e de Bordeaux, France +</td><td>('1720735', 'Chao Zhang', 'chao zhang')<br/>('34895713', 'Arnaud Dessein', 'arnaud dessein')<br/>('1737428', 'Nick Pears', 'nick pears')<br/>('1694260', 'Hang Dai', 'hang dai')</td><td>{cz679, william.smith, nick.pears, hd816}@york.ac.uk +<br/>arnaud.dessein@u-bordeaux.fr +</td></tr><tr><td>cda4fb9df653b5721ad4fe8b4a88468a410e55ec</td><td>Gabor wavelet transform and its application +</td><td>('38784892', 'Wei-lun Chao', 'wei-lun chao')</td><td></td></tr><tr><td>cd3005753012409361aba17f3f766e33e3a7320d</td><td>Multilinear Biased Discriminant Analysis: A Novel Method for Facial +<br/>Action Unit Representation +</td><td>('1736464', 'Mahmoud Khademi', 'mahmoud khademi')<br/>('2179339', 'Mehran Safayani', 'mehran safayani')</td><td>†: Sharif University of Tech., DSP Lab, {khademi@ce, safayani@ce, manzuri@}.sharif.edu +</td></tr><tr><td>cd687ddbd89a832f51d5510c478942800a3e6854</td><td>A Game to Crowdsource Data for Affective Computing +<br/><b>Games Studio, Faculty of Engineering and IT, University of Technology, Sydney</b></td><td>('1733360', 'Chek Tien Tan', 'chek tien tan')<br/>('2117735', 'Hemanta Sapkota', 'hemanta sapkota')<br/>('2823535', 'Daniel Rosser', 'daniel rosser')<br/>('3141633', 'Yusuf Pisan', 'yusuf pisan')</td><td>chek@gamesstudio.org +<br/>hemanta.sapkota@student.uts.edu.au +<br/>daniel.j.rosser@gmail.com +<br/>yusuf.pisan@gamesstudio.org +</td></tr><tr><td>cd436f05fb4aeeda5d1085f2fe0384526571a46e</td><td>Information Bottleneck Domain Adaptation with +<br/>Privileged Information for Visual Recognition +<br/>Lane Department of Computer Science and Electrical Engineering +<br/><b>West Virginia University</b></td><td>('2897426', 'Saeid Motiian', 'saeid motiian')<br/>('1736352', 'Gianfranco Doretto', 'gianfranco doretto')</td><td>{samotiian,gidoretto}@mix.wvu.edu +</td></tr><tr><td>cd2c54705c455a4379f45eefdf32d8d10087e521</td><td>A Hybrid Model for Identity Obfuscation by +<br/>Face Replacement +<br/><b>Max Planck Institute for Informatics, Saarland Informatics Campus</b></td><td>('32222907', 'Qianru Sun', 'qianru sun')<br/>('1739548', 'Mario Fritz', 'mario fritz')</td><td>{qsun, atewari, wxu, mfritz, theobalt, schiele}@mpi-inf.mpg.de +</td></tr><tr><td>cd7a7be3804fd217e9f10682e0c0bfd9583a08db</td><td>Women also Snowboard: +<br/>Overcoming Bias in Captioning Models +</td><td>('40895688', 'Kaylee Burns', 'kaylee burns')</td><td></td></tr><tr><td>cd023d2d067365c83d8e27431e83e7e66082f718</td><td>Real-Time Rotation-Invariant Face Detection with +<br/>Progressive Calibration Networks +<br/>1 Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b><br/>3 CAS Center for Excellence in Brain Science and Intelligence Technology +</td><td>('41017549', 'Xuepeng Shi', 'xuepeng shi')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('3126238', 'Shuzhe Wu', 'shuzhe wu')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{xuepeng.shi, shiguang.shan, meina.kan, shuzhe.wu, xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>cca9ae621e8228cfa787ec7954bb375536160e0d</td><td>Learning to Collaborate for User-Controlled Privacy +<br/>Martin Bertran 1† +<br/>Natalia Martinez 1†* +<br/>Afroditi Papadaki 2 +<br/>Miguel Rodrigues 2 +<br/><b>Duke University, Durham, NC, USA</b><br/><b>University College London, London, UK</b><br/>†These authors contributed equally to this work. +<br/>Privacy is a human right. Tim Cook, Apple CEO. +</td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('1699339', 'Guillermo Sapiro', 'guillermo sapiro')</td><td>martin.bertran@duke.edu +<br/>natalia.martinez@duke.edu +<br/>a.papadaki.17@ucl.ac.uk +<br/>qiuqiang@gmail.com +<br/>m.rodrigues@ucl.ac.uk +<br/>guillermo.sapiro@duke.edu +</td></tr><tr><td>cc589c499dcf323fe4a143bbef0074c3e31f9b60</td><td>A 3D Facial Expression Database For Facial Behavior Research +<br/><b>State University of New York at Binghamton</b></td><td>('8072251', 'Lijun Yin', 'lijun yin')</td><td></td></tr><tr><td>ccfcbf0eda6df876f0170bdb4d7b4ab4e7676f18</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 6, NO. 1, JUNE 2011 +<br/>A Dynamic Appearance Descriptor Approach to +<br/>Facial Actions Temporal Modelling +</td><td>('39532631', 'Bihan Jiang', 'bihan jiang')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>cc2eaa182f33defbb33d69e9547630aab7ed9c9c</td><td>Surpassing Humans and Computers with JELLYBEAN: +<br/>Crowd-Vision-Hybrid Counting Algorithms +<br/><b>Stanford University</b><br/><b>University of Illinois</b><br/><b>The Ohio State University</b><br/>Aditya Parameswaran +<br/><b>University of Illinois</b></td><td>('32953042', 'Akash Das Sarma', 'akash das sarma')<br/>('2636295', 'Ayush Jain', 'ayush jain')<br/>('39393264', 'Arnab Nandi', 'arnab nandi')</td><td>akashds@stanford.edu +<br/>ajain42@illinois.edu +<br/>arnab@cse.osu.edu +<br/>adityagp@illinois.edu +</td></tr><tr><td>ccbfc004e29b3aceea091056b0ec536e8ea7c47e</td><td></td><td></td><td></td></tr><tr><td>ccdea57234d38c7831f1e9231efcb6352c801c55</td><td>Illumination Processing in Face Recognition +<br/>187 +<br/>11 +<br/>X +<br/>Illumination Processing in Face Recognition +<br/>Yongping Li, Chao Wang and Xinyu Ao +<br/><b>Shanghai Institute of Applied Physics, Chinese Academy of Sciences</b><br/>China +<br/>1. Introduction +<br/>Driven by the demanding of public security, face recognition has emerged as a viable +<br/>solution and achieved comparable accuracies to fingerprint system under controlled +<br/>lightning environment. In recent years, with wide installing of camera in open area, the +<br/>automatic face recognition in watch-list application is facing a serious problem. Under the +<br/>open environment, lightning changes is unpredictable, and the performance of face +<br/>recognition degrades seriously. +<br/>Illumination processing is a necessary step for face recognition to be useful in the +<br/>uncontrolled environment. NIST has started a test called FRGC to boost the research in +<br/>improving the performance under changing illumination. In this chapter, we will focus on +<br/>the research effort made in this direction and the influence on face recognition caused by +<br/>illumination. +<br/>First of all, we will discuss the quest on the image formation mechanism under various +<br/>illumination situations, and the corresponding mathematical modelling. The Lambertian +<br/>lighting model, bilinear illuminating model and some recent model are reviewed. Secondly, +<br/>under different state of face, like various head pose and different facial expression, how +<br/>illumination influences the recognition result, where the different pose and illuminating will +<br/>be examined carefully. Thirdly, the current methods researcher employ to counter the change +<br/>of illumination to maintain good performance on face recognition are assessed briefly. The +<br/>processing technique in video and how it will improve face recognition on video, where +<br/>Wang’s (Wang & Li, 2009) work will be discussed to give an example on the related +<br/>advancement in the fourth part. And finally, the current state-of-art of illumination +<br/>processing and its future trends will be discussed. +<br/>2. The formation of camera imaging and its difference from the human visual +<br/>system +<br/>With the camera invented in 1814 by Joseph N, recording of human face began its new era. +<br/>Since we do not need to hire a painter to draw our figures, as the nobles did in the middle +<br/>age. And the machine recorded our image as it is, if the camera is in good condition. +<br/>Currently, the imaging system is mostly to be digital format. The central part is CCD +<br/>(charge-coupled device) or CMOS (complimentary metal-oxide semiconductor). The +<br/>CCD/CMOS operates just like the human eyes. Both CCD and CMOS image sensors operate +<br/>www.intechopen.com +</td><td></td><td></td></tr><tr><td>cc38942825d3a2c9ee8583c153d2c56c607e61a7</td><td>Database Cross Matching: A Novel Source of +<br/>Fictitious Forensic Cases +<br/>Signals and Systems Group, EEMCS, +<br/><b>University of Twente, Netherlands</b></td><td>('34214663', 'Abhishek Dutta', 'abhishek dutta')<br/>('39128850', 'Raymond Veldhuis', 'raymond veldhuis')<br/>('1745742', 'Luuk Spreeuwers', 'luuk spreeuwers')</td><td>{a.dutta,r.n.j.veldhuis,l.j.spreeuwers}@utwente.nl +</td></tr><tr><td>cc3c273bb213240515147e8be68c50f7ea22777c</td><td>Gaining Insight Into Films +<br/>Via Topic Modeling & Visualization +<br/>KEYWORDS Collaboration, computer vision, cultural +<br/>analytics, economy of abundance, interactive data +<br/>visualization +<br/>We moved beyond misuse when the software actually +<br/>became useful for film analysis with the addition of audio +<br/>analysis, subtitle analysis, facial recognition, and topic +<br/>modeling. Using multiple types of visualizations and +<br/>a back-and-fourth workflow between people and AI +<br/>we arrived at an approach for cultural analytics that +<br/>can be used to review and develop film criticism. Finally, +<br/>we present ways to apply these techniques to Database +<br/>Cinema and other aspects of film and video creation. +<br/>PROJECT DATE 2014 +<br/>URL http://misharabinovich.com/soyummy.html +</td><td>('40462877', 'MISHA RABINOVICH', 'misha rabinovich')<br/>('1679896', 'Yogesh Girdhar', 'yogesh girdhar')</td><td></td></tr><tr><td>cc8e378fd05152a81c2810f682a78c5057c8a735</td><td>International Journal of Computer Sciences and Engineering Open Access +<br/> Research Paper Volume-5, Issue-12 E-ISSN: 2347-2693 +<br/>Expression Invariant Face Recognition System based on Topographic +<br/>Independent Component Analysis and Inner Product Classifier +<br/> +<br/>Department of Electrical Engineering, IIT Delhi, New Delhi, India +<br/>Available online at: www.ijcseonline.org +<br/>Received: 07/Nov/2017, Revised: 22/Nov/2017, Accepted: 14/Dec/2017, Published: 31/Dec/2017 +</td><td>('40258123', 'Aruna Bhat', 'aruna bhat')</td><td>*Corresponding Author: abigit06@yahoo.com +</td></tr><tr><td>ccf43c62e4bf76b6a48ff588ef7ed51e87ddf50b</td><td>American Journal of Food Science and Health +<br/>Vol. 2, No. 2, 2016, pp. 7-17 +<br/>http://www.aiscience.org/journal/ajfsh +<br/>ISSN: 2381-7216 (Print); ISSN: 2381-7224 (Online) +<br/>Nutraceuticals and Cosmeceuticals for Human +<br/>Beings–An Overview +<br/><b>Narayana Pharmacy College, Nellore, India</b></td><td>('40179150', 'R. Ramasubramania Raja', 'r. ramasubramania raja')</td><td></td></tr><tr><td>cc31db984282bb70946f6881bab741aa841d3a7c</td><td>ALBANIE, VEDALDI: LEARNING GRIMACES BY WATCHING TV +<br/>Learning Grimaces by Watching TV +<br/>http://www.robots.ox.ac.uk/~albanie +<br/>http://www.robots.ox.ac.uk/~vedaldi +<br/>Engineering Science Department +<br/>Univeristy of Oxford +<br/>Oxford, UK +</td><td>('7641268', 'Samuel Albanie', 'samuel albanie')<br/>('1687524', 'Andrea Vedaldi', 'andrea vedaldi')</td><td></td></tr><tr><td>cc8bf03b3f5800ac23e1a833447c421440d92197</td><td></td><td></td><td></td></tr><tr><td>cc91001f9d299ad70deb6453d55b2c0b967f8c0d</td><td>OPEN ACCESS +<br/>ISSN 2073-8994 +<br/>Article +<br/>Performance Enhancement of Face Recognition in Smart TV +<br/>Using Symmetrical Fuzzy-Based Quality Assessment +<br/><b>Division of Electronics and Electrical Engineering, Dongguk University, 26 Pil-dong 3-ga, Jung-gu</b><br/>Tel.: +82-10-3111-7022; Fax: +82-2-2277-8735. +<br/>Academic Editor: Christopher Tyler +<br/>Received: 31 March 2015 / Accepted: 21 August 2015 / Published: 25 August 2015 +</td><td>('3021526', 'Yeong Gon Kim', 'yeong gon kim')<br/>('2026806', 'Won Oh Lee', 'won oh lee')<br/>('1922686', 'Hyung Gil Hong', 'hyung gil hong')<br/>('4634733', 'Kang Ryoung Park', 'kang ryoung park')</td><td>Seoul 100-715, Korea; E-Mails: csokyg@dongguk.edu (Y.G.K.); 215p8@hanmail.net (W.O.L.); +<br/>yawara18@hotmail.com (K.W.K.); hell@dongguk.edu (H.G.H.) +<br/>* Author to whom correspondence should be addressed; E-Mail: parkgr@dgu.edu; +</td></tr><tr><td>cc96eab1e55e771e417b758119ce5d7ef1722b43</td><td>An Empirical Study of Recent +<br/>Face Alignment Methods +</td><td>('2966679', 'Heng Yang', 'heng yang')<br/>('34760532', 'Xuhui Jia', 'xuhui jia')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('39626495', 'Peter Robinson', 'peter robinson')</td><td></td></tr><tr><td>cc7e66f2ba9ac0c639c80c65534ce6031997acd7</td><td>Facial Descriptors for Identity-Preserving +<br/>Multiple People Tracking +<br/>CVLab, School of Computer and Communication Sciences +<br/><b>Swiss Federal Institute of Technology, Lausanne (EPFL</b><br/>EPFL-REPORT-187534 +<br/>July 2013 +</td><td></td><td>Michalis Zervos1 (michail.zervos@epfl.ch) +<br/>Horesh Ben Shitrit1 (horesh.benshitrit@epfl.ch) +<br/>Franc¸ois Fleuret(cid:63) (francois.fleuret@idiap.ch) +<br/>Pascal Fua (pascal.fua@epfl.ch) +</td></tr><tr><td>cc9057d2762e077c53e381f90884595677eceafa</td><td>On the Exploration of Joint Attribute Learning +<br/>for Person Re-identification +<br/><b>Michigan State University</b></td><td>('38993748', 'Joseph Roth', 'joseph roth')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td>{rothjos1,liuxm}@cse.msu.edu +</td></tr><tr><td>ccf16bcf458e4d7a37643b8364594656287f5bfc</td><td>A CNN Cascade for Landmark Guided Semantic +<br/>Part Segmentation +<br/><b>School of Computer Science, The University of Nottingham, Nottingham, UK</b></td><td>('34596685', 'Aaron S. Jackson', 'aaron s. jackson')<br/>('46637307', 'Michel Valstar', 'michel valstar')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>{aaron.jackson, michel.valstar, yorgos.tzimiropoulos}@nottingham.ac.uk +</td></tr><tr><td>e64b683e32525643a9ddb6b6af8b0472ef5b6a37</td><td>Face Recognition and Retrieval in Video +</td><td>('10795229', 'Caifeng Shan', 'caifeng shan')</td><td></td></tr><tr><td>e69ac130e3c7267cce5e1e3d9508ff76eb0e0eef</td><td>Research Article +<br/>Addressing the illumination challenge in two- +<br/>dimensional face recognition: a survey +<br/>ISSN 1751-9632 +<br/>Received on 31st March 2014 +<br/>Revised on 7th January 2015 +<br/>Accepted on 9th April 2015 +<br/>doi: 10.1049/iet-cvi.2014.0086 +<br/>www.ietdl.org +<br/><b>Computational Biomedicine Laboratory, University of Houston, Houston, Texas 77204, USA</b><br/>2Department of Computer Science, Cybersecurity Laboratory, Instituto Tecnológico y de Estudios Superiores de Monterrey, Monterrey, +<br/>NL 64840, Mexico +</td><td>('2899018', 'Miguel A. Ochoa-Villegas', 'miguel a. ochoa-villegas')<br/>('1905427', 'Olivia Barron-Cano', 'olivia barron-cano')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td>✉ E-mail: ioannisk@uh.edu +</td></tr><tr><td>e6b45d5a86092bbfdcd6c3c54cda3d6c3ac6b227</td><td>Pairwise Relational Networks for Face +<br/>Recognition +<br/>1 Department of Creative IT Engineering, POSTECH, Korea +<br/>2 Department of Computer Science and Engineering, POSTECH, Korea +</td><td>('2794366', 'Bong-Nam Kang', 'bong-nam kang')<br/>('50682377', 'Yonghyun Kim', 'yonghyun kim')<br/>('1695669', 'Daijin Kim', 'daijin kim')</td><td>{bnkang,gkyh0805,dkim}@postech.ac.kr +</td></tr><tr><td>e6865b000cf4d4e84c3fe895b7ddfc65a9c4aaec</td><td>Chapter 15. The critical role of the +<br/>cold-start problem and incentive systems +<br/>in emotional Web 2.0 services +</td><td>('2443050', 'Tobias Siebenlist', 'tobias siebenlist')<br/>('2153585', 'Kathrin Knautz', 'kathrin knautz')</td><td></td></tr><tr><td>e6d689054e87ad3b8fbbb70714d48712ad84dc1c</td><td>Robust Facial Feature Tracking +<br/><b>School of Computing, Staffordshire University</b><br/>Stafford ST18 0DG +</td><td>('2155770', 'Fabrice Bourel', 'fabrice bourel')<br/>('1919849', 'Claude C. Chibelushi', 'claude c. chibelushi')<br/>('32890308', 'Adrian A. Low', 'adrian a. low')</td><td>F.Bourel@staffs.ac.uk +<br/>C.C.Chibelushi@staffs.ac.uk +<br/>A.A.Low@staffs.ac.uk +</td></tr><tr><td>e6dc1200a31defda100b2e5ddb27fb7ecbbd4acd</td><td>1921 +<br/>Flexible Manifold Embedding: A Framework +<br/>for Semi-Supervised and Unsupervised +<br/>Dimension Reduction +<br/>0 = +<br/>, the linear regression function ( +</td><td>('1688370', 'Feiping Nie', 'feiping nie')<br/>('1714390', 'Dong Xu', 'dong xu')<br/>('1700883', 'Changshui Zhang', 'changshui zhang')</td><td></td></tr><tr><td>e6f20e7431172c68f7fce0d4595100445a06c117</td><td>Searching Action Proposals via Spatial +<br/>Actionness Estimation and Temporal Path +<br/>Inference and Tracking +<br/><b>cid:93)Peking University Shenzhen Graduate School, Shenzhen, P.R.China</b><br/><b>DISI, University of Trento, Trento, Italy</b></td><td>('40147776', 'Dan Xu', 'dan xu')<br/>('3238696', 'Zhihao Li', 'zhihao li')<br/>('1684933', 'Ge Li', 'ge li')</td><td></td></tr><tr><td>e6e5a6090016810fb902b51d5baa2469ae28b8a1</td><td>Title +<br/>Energy-Efficient Deep In-memory Architecture for NAND +<br/>Flash Memories +<br/>Archived version +<br/>Accepted manuscript: the content is same as the published +<br/>paper but without the final typesetting by the publisher +<br/>Published version +<br/>DOI +<br/>Published paper +<br/>URL +<br/>Authors (contact) +<br/>10.1109/ISCAS.2018.8351458 +</td><td></td><td></td></tr><tr><td>e6540d70e5ffeed9f447602ea3455c7f0b38113e</td><td></td><td></td><td></td></tr><tr><td>e6ee36444038de5885473693fb206f49c1369138</td><td></td><td></td><td></td></tr><tr><td>e6178de1ef15a6a973aad2791ce5fbabc2cb8ae5</td><td>Improving Facial Landmark Detection via a +<br/>Super-Resolution Inception Network +<br/><b>Institute for Human-Machine Communication</b><br/><b>Technical University of Munich, Germany</b></td><td>('38746426', 'Martin Knoche', 'martin knoche')<br/>('3044182', 'Daniel Merget', 'daniel merget')<br/>('1705843', 'Gerhard Rigoll', 'gerhard rigoll')</td><td></td></tr><tr><td>f913bb65b62b0a6391ffa8f59b1d5527b7eba948</td><td></td><td></td><td></td></tr><tr><td>f9784db8ff805439f0a6b6e15aeaf892dba47ca0</td><td>Comparing the performance of Emotion-Recognition Implementations +<br/>in OpenCV, Cognitive Services, and Google Vision APIs +<br/>Department of Informatics and Artificial Intelligence +<br/><b>Tomas Bata University in Zl n</b><br/>Nad Stráněmi 4511, 76005, Zlín +<br/>CZECH REPUBLIC +</td><td></td><td>beltran_prieto@fai.utb.cz +</td></tr><tr><td>f935225e7811858fe9ef6b5fd3fdd59aec9abd1a</td><td>www.elsevier.com/locate/ynimg +<br/>Spatiotemporal dynamics and connectivity pattern differences +<br/>between centrally and peripherally presented faces +<br/><b>Laboratory for Human Brain Dynamics, RIKEN Brain Science Institute (BSI), 2-1 Hirosawa, Wakoshi, Saitama, 351-0198, Japan</b><br/>Received 4 May 2005; revised 26 January 2006; accepted 6 February 2006 +<br/>Available online 24 March 2006 +<br/>Most neuroimaging studies on face processing used centrally presented +<br/>images with a relatively large visual field. Images presented in this way +<br/>activate widespread striate and extrastriate areas and make it difficult +<br/>to study spatiotemporal dynamics and connectivity pattern differences +<br/>from various parts of the visual field. Here we studied magneto- +<br/>encephalographic responses in humans to centrally and peripherally +<br/>presented faces for testing the hypothesis that processing of visual +<br/>stimuli with facial expressions of emotions depends on where the +<br/>stimuli are presented in the visual field. Using our tomographic and +<br/>statistical parametric mapping analyses, we identified occipitotemporal +<br/>areas activated by face stimuli more than by control conditions. V1/V2 +<br/>activity was significantly stronger for lower than central and upper +<br/>visual field presentation. Fusiform activity, however, was significantly +<br/>stronger for central than for peripheral presentation. Both the V1/V2 +<br/>and fusiform areas activated earlier for peripheral than for central +<br/>presentation. Fast responses in the fusiform were found at 70 – 80 ms +<br/>after image onset, as well as a response at 130 – 160 ms. For peripheral +<br/>presentation, contralateral V1/V2 and fusiform activated earlier (10 ms +<br/>and 23 ms, respectively) and significantly stronger than their ipsilateral +<br/>counterparts. Mutual +<br/>information analysis further showed linked +<br/>activity from bilateral V1/V2 to fusiform for central presentation and +<br/>from contralateral V1/V2 to fusiform for lower visual field presenta- +<br/>tion. In the upper visual field, the linkage was from fusiform to V1/V2. +<br/>Our results showed that face stimuli are processed predominantly in +<br/>the hemisphere contralateral to the stimulation and demonstrated for +<br/>the first time early fusiform activation leading V1/V2 activation for +<br/>upper visual field stimulation. +<br/>D 2006 Elsevier Inc. All rights reserved. +<br/>Keywords: Magnetoencephalography (MEG); Striate cortex; Extrastriate +<br/>cortex; Fusiform gyrus; Face perception; Connectivity +<br/>Introduction +<br/>It is well established that visual stimuli presented in one part of +<br/>the visual field are projected to the contralateral part of the visual +<br/>cortex such that images presented in the right visual field are +<br/>* Corresponding author. Fax: +81 48 467 9731. +<br/>Available online on ScienceDirect (www.sciencedirect.com). +<br/>1053-8119/$ - see front matter D 2006 Elsevier Inc. All rights reserved. +<br/>projected to the left visual cortex. It is, however, unclear whether +<br/>stimuli presented in different parts of the visual field are processed +<br/>differently in extrastriate areas that specialize for processing +<br/>complex properties of stimuli and whether different connectivity +<br/>patterns are produced between striate and extrastriate cortices when +<br/>such complex stimuli are presented to different quadrants. To +<br/>address these questions, one needs to incorporate three ingredients +<br/>in the experimental design and analysis. First, one must use stimuli +<br/>that are known to excite at least one specific extrastriate area well. +<br/>Second, one must present stimuli at positions in the visual field +<br/>known to project to specific parts of the visual cortex so that the +<br/>early entry into the visual system via V1 can be reliably extracted +<br/>for connectivity analysis. Third, one must use a technique that can +<br/>provide refined spatial and temporal +<br/>information about brain +<br/>activity. The information can then be used in follow-up analysis of +<br/>spatiotemporal dynamics and connectivity patterns in the brain. +<br/>The choice of faces is obvious because many studies have +<br/>shown that faces are effective stimuli for exciting extrastriate areas. +<br/>The posterior fusiform gyrus was first associated with cortical face +<br/>processing from lesion studies on patients with specific recognition +<br/>deficits of familiar faces (Meadows, 1974; Damasio et al., 1990; +<br/>Sergent and Poncet, 1990). Neuroimaging studies have shown that +<br/>extrastriate areas are involved in face processing in normal subjects +<br/>using techniques such as positron emission tomography (PET) +<br/>(Sergent et al., 1992; Haxby et al., 1994), functional magnetic +<br/>resonance imaging (fMRI) (Puce et al., 1995; McCarthy et al., +<br/>1997; Kanwisher et al., 1997; Halgren et al., 1999), electroen- +<br/>cephalography (EEG) (Allison et al., 1994; Bentin et al., 1996; +<br/>George et al., 1996) and magnetoencephalography (MEG) (Link- +<br/>enkaer-Hansen et al., 1998; Halgren et al., 2000). In the present +<br/>study, we chose the same face stimuli from our earlier MEG study +<br/>on complex object and face affect recognition that were shown to +<br/>activate extrastriate areas well (Liu et al., 1999; Ioannides et al., +<br/>2000). +<br/><b>Most of the earlier studies mentioned above, including ours</b><br/>have presented facial images centrally with a relatively large visual +<br/>field covering both the fovea and parafovea. Central presentation +<br/>of images activates widespread striate and extrastriate areas. Low +<br/>order visual areas (V1/V2) corresponding to left – right – upper – +<br/>lower visual field stimulation are therefore activated by the same +</td><td>('2259342', 'Lichan Liu', 'lichan liu')</td><td>E-mail address: ioannides@postman.riken.jp (A.A. Ioannides). +</td></tr><tr><td>f963967e52a5fd97fa3ebd679fd098c3cb70340e</td><td>Analysis, Interpretation, and Recognition of Facial +<br/>Action Units and Expressions Using Neuro-Fuzzy +<br/>Modeling +<br/>and Ali A. Kiaei1 +<br/><b>DSP Lab, Sharif University of Technology, Tehran, Iran</b><br/><b>Institute for Studies in Fundamental Sciences (IPM), Tehran, Iran</b></td><td>('1736464', 'Mahmoud Khademi', 'mahmoud khademi')<br/>('1702826', 'Mohammad Hadi Kiapour', 'mohammad hadi kiapour')</td><td>{khademi@ce.,kiapour@ee.,manzuri@,kiaei@ce.}sharif.edu +</td></tr><tr><td>f9e0209dc9e72d64b290d0622c1c1662aa2cc771</td><td>CONTRIBUTIONS TO BIOMETRIC RECOGNITION: +<br/>MATCHING IDENTICAL TWINS AND LATENT FINGERPRINTS +<br/>By +<br/>A DISSERTATION +<br/>Submitted +<br/><b>to Michigan State University</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>Computer Science– Doctor of Philosophy +<br/>2013 +</td><td>('31508481', 'Alessandra Aparecida Paulino', 'alessandra aparecida paulino')</td><td></td></tr><tr><td>f92ade569cbe54344ffd3bb25efd366dcd8ad659</td><td>EFFECT OF SUPER RESOLUTION ON HIGH DIMENSIONAL FEATURES FOR +<br/>UNSUPERVISED FACE RECOGNITION IN THE WILD +<br/><b>University of Bridgeport, Bridgeport, CT 06604, USA</b></td><td>('40373065', 'Ahmed ElSayed', 'ahmed elsayed')<br/>('37374395', 'Ausif Mahmood', 'ausif mahmood')</td><td>Emails: aelsayed@my.bridgeport.edu, {mahmood,sobh}@bridgeport.edu +</td></tr><tr><td>f96bdd1e2a940030fb0a89abbe6c69b8d7f6f0c1</td><td></td><td></td><td></td></tr><tr><td>f93606d362fcbe62550d0bf1b3edeb7be684b000</td><td>The Computer Journal Advance Access published February 1, 2012 +<br/><b>The Author 2012. Published by Oxford University Press on behalf of The British Computer Society. All rights reserved</b><br/>doi:10.1093/comjnl/bxs001 +<br/>Nearest Neighbor Classifier Based +<br/>on Nearest Feature Decisions +<br/><b>Machine Intelligence Group, School of Computer Science, Indian Institute of Information Technology and</b><br/><b>Queensland Micro- and Nanotechnology Centre and Grif th School of Engineering, Grif th University</b><br/>Management, Kerala, India +<br/>Nathan, Australia +<br/>High feature dimensionality of realistic datasets adversely affects the recognition accuracy of nearest +<br/>neighbor (NN) classifiers. To address this issue, we introduce a nearest feature classifier that shifts +<br/>the NN concept from the global-decision level to the level of individual features. Performance +<br/><b>comparisons with 12 instance-based classi ers on 13 benchmark University of California Irvine</b><br/>classification datasets show average improvements of 6 and 3.5% in recognition accuracy and +<br/>area under curve performance measures, respectively. The statistical significance of the observed +<br/>performance improvements is verified by the Friedman test and by the post hoc Bonferroni–Dunn +<br/>test. In addition, the application of the classifier is demonstrated on face recognition databases, a +<br/>character recognition database and medical diagnosis problems for binary and multi-class diagnosis +<br/>on databases including morphological and gene expression features. +<br/>Keywords: nearest neighbors; classification; local features; local ranking +<br/>Received 2 September 2011; revised 3 December 2011 +<br/>Handling editor: Ethem Alpaydin +<br/>1. +<br/>INTRODUCTION +<br/>Automatic classification of patterns has been continuously and +<br/>rigorously investigated for the last 30 years. Simple classifiers, +<br/>based on the nearest neighbor (NN) principle, have been used +<br/>to solve a wide range of classification problems [1–5]. The NN +<br/>classification works on the idea of calculating global distances +<br/>between patterns, followed by ranking to determine the NNs +<br/>that best represent the class of a test pattern. Usually, distance +<br/>metric measures are used to compute the distances between +<br/>feature vectors. The accuracy of the calculated distances is +<br/>affected by the quality of the features, which can be degraded by +<br/>natural variability and measurement noise. Furthermore, some +<br/>distance calculations are affected by falsely assumed correlation +<br/>between different features. For example, Mahalanobis distance +<br/>will include the comparison between poorly or uncorrelated +<br/>features. This problem is more pronounced when the number +<br/>of features in a pattern is very large because the irrelevant +<br/>distance calculations can accumulate to a large value (for +<br/>example, +<br/>there will be many false correlations in gene +<br/>expressions data that can have dimensionality higher than 104 +<br/>features). In addition to this problem, a considerable increase +<br/>in dimensionality complicates the classifier implementations +<br/>resulting in ‘curse of dimensionality’, where a possible +<br/>convergence to a classification solution becomes very slow +<br/>and inaccurate [6, 7]. The conventional solution to address +<br/>these problems is to rely on feature extraction and feature +<br/>selection methods [8–10]. However, unpredictability of natural +<br/>variability in patterns makes processing a specific feature +<br/>inapplicable to diverse pattern-recognition problems. Another +<br/>approach to improve the classifier performance is by using +<br/>machine learning techniques to learn the distance metrics +<br/>[11–13]. These methods attempt to reduce the inaccuracies +<br/>that occur with distance calculations. However, this solution +<br/>tends to include optimization problems that suffer from +<br/>high computational complexity and require reduced feature +<br/>dimensionality, resulting in low accuracies when the feature +<br/>vectors are highly dimensional and the number of intra-class +<br/>gallery objects is low. Learning distance metrics can completely +<br/>fail in high- and ultra-high-dimensional databases when the +<br/>relevance and redundancy of features often become impossible +<br/>to trace even with feature weighting or selection schemes. +<br/>Owing to these reasons, performance improvement of the NN +<br/>The Computer Journal, 2012 +</td><td>('1744784', 'Alex Pappachen James', 'alex pappachen james')<br/>('1697594', 'Sima Dimitrijev', 'sima dimitrijev')</td><td>For Permissions, please email: journals.permissions@oup.com +<br/>Corresponding author: apj@ieee.org +</td></tr><tr><td>f94f366ce14555cf0d5d34248f9467c18241c3ee</td><td>Deep Convolutional Neural Network in +<br/>Deformable Part Models for Face Detection +<br/><b>University of Science, Vietnam National University, HCMC</b><br/><b>School of Information Science, Japan Advanced Institute of Science and Technology</b></td><td>('2187730', 'Dinh-Luan Nguyen', 'dinh-luan nguyen')<br/>('34453615', 'Vinh-Tiep Nguyen', 'vinh-tiep nguyen')<br/>('1780348', 'Minh-Triet Tran', 'minh-triet tran')<br/>('2854896', 'Atsuo Yoshitaka', 'atsuo yoshitaka')</td><td>1212223@student.hcmus.edu.vn +<br/>{nvtiep,tmtriet}@fit.hcmus.edu.vn +<br/>ayoshi@jaist.ac.jp +</td></tr><tr><td>f997a71f1e54d044184240b38d9dc680b3bbbbc0</td><td>Deep Cross Modal Learning for Caricature Verification and +<br/>Identification(CaVINet) +<br/>https://lsaiml.github.io/CaVINet/ +<br/><b>Indian Institute of Technology Ropar</b><br/><b>Indian Institute of Technology Ropar</b><br/><b>Indian Institute of Technology Ropar</b><br/>Narayanan C Krishnan +<br/><b>Indian Institute of Technology Ropar</b></td><td>('6220011', 'Jatin Garg', 'jatin garg')<br/>('51152207', 'Himanshu Tolani', 'himanshu tolani')<br/>('41021778', 'Skand Vishwanath Peri', 'skand vishwanath peri')</td><td>2014csb1017@iitrpr.ac.in +<br/>2014csb1015@iitrpr.ac.in +<br/>pvskand@gmail.com +<br/>ckn@iitrpr.ac.in +</td></tr><tr><td>f909d04c809013b930bafca12c0f9a8192df9d92</td><td>Single Image Subspace for Face Recognition +<br/><b>Nanjing University of Aeronautics and Astronautics, China</b><br/>1 Department of Computer Science and Engineering, +<br/>2 National Key Laboratory for Novel Software Technology, +<br/><b>Nanjing University, China</b></td><td>('39497343', 'Jun Liu', 'jun liu')<br/>('1680768', 'Songcan Chen', 'songcan chen')<br/>('1692625', 'Zhi-Hua Zhou', 'zhi-hua zhou')<br/>('2248421', 'Xiaoyang Tan', 'xiaoyang tan')</td><td>{j.liu, s.chen, x.tan}@nuaa.edu.cn +<br/>zhouzh@nju.edu.cn +</td></tr><tr><td>f9d1f12070e5267afc60828002137af949ff1544</td><td>Maximum Entropy Binary Encoding for Face Template Protection +<br/>Rohit Kumar Pandey +<br/><b>University at Buffalo, SUNY</b></td><td>('34872128', 'Yingbo Zhou', 'yingbo zhou')<br/>('3352136', 'Bhargava Urala Kota', 'bhargava urala kota')<br/>('1723877', 'Venu Govindaraju', 'venu govindaraju')</td><td>{rpandey, yingbozh, buralako, govind}@buffalo.edu +</td></tr><tr><td>f9ccfe000092121a2016639732cdb368378256d5</td><td>Cognitive behaviour analysis based on facial +<br/>information using depth sensors +<br/><b>Kingston University London, University of Westminster London</b><br/><b>Imperial College London</b></td><td>('1686887', 'Juan Manuel Fernandez Montenegro', 'juan manuel fernandez montenegro')<br/>('2866802', 'Barbara Villarini', 'barbara villarini')<br/>('2140622', 'Athanasios Gkelias', 'athanasios gkelias')<br/>('1689047', 'Vasileios Argyriou', 'vasileios argyriou')</td><td>Juan.Fernandez@kingston.ac.uk,B.Villarini@westminster.ac.uk,A.Gkelias@ +<br/>imperial.ac.uk,Vasileios.Argyriou@kingston.ac.uk +</td></tr><tr><td>f08e425c2fce277aedb51d93757839900d591008</td><td>Neural Motifs: Scene Graph Parsing with Global Context +<br/><b>Paul G. Allen School of Computer Science and Engineering, University of Washington</b><br/><b>Allen Institute for Arti cial Intelligence</b><br/><b>School of Computer Science, Carnegie Mellon University</b><br/>https://rowanzellers.com/neuralmotifs +</td><td>('2545335', 'Rowan Zellers', 'rowan zellers')<br/>('38094552', 'Sam Thomson', 'sam thomson')</td><td>{rowanz, my89, yejin}@cs.washington.edu, sthomson@cs.cmu.edu +</td></tr><tr><td>f02f0f6fcd56a9b1407045de6634df15c60a85cd</td><td>Learning Low-shot facial representations via 2D warping +<br/><b>RWTH Aachen University</b></td><td>('35362682', 'Shen Yan', 'shen yan')</td><td>shen.yan@rwth-aachen.de +</td></tr><tr><td>f0cee87e9ecedeb927664b8da44b8649050e1c86</td><td></td><td></td><td></td></tr><tr><td>f0f4f16d5b5f9efe304369120651fa688a03d495</td><td>Temporal Generative Adversarial Nets +<br/>Preferred Networks inc., Japan +</td><td>('49160719', 'Masaki Saito', 'masaki saito')<br/>('8252749', 'Eiichi Matsumoto', 'eiichi matsumoto')</td><td>{msaito, matsumoto}@preferred.jp +</td></tr><tr><td>f0ca31fd5cad07e84b47d50dc07db9fc53482a46</td><td>Advances in Pure Mathematics, 2012, 2, 226-242 +<br/>http://dx.doi.org/10.4236/apm.2012.24033 Published Online July 2012 (http://www.SciRP.org/journal/apm) +<br/>Feature Patch Illumination Spaces and Karcher +<br/>Compression for Face Recognition via +<br/>Grassmannians +<br/><b>California State University, Long Beach, USA</b><br/><b>Colorado State University, Fort Collins, USA</b><br/>Received January 7, 2012; revised February 20, 2012; accepted February 27, 2012 +</td><td>('2640182', 'Jen-Mei Chang', 'jen-mei chang')<br/>('30383278', 'Chris Peterson', 'chris peterson')<br/>('41211081', 'Michael Kirby', 'michael kirby')</td><td>Email: jen-mei.chang@csulb.edu, {peterson, Kirby}@math.colostate.edu +</td></tr><tr><td>f0ae807627f81acb63eb5837c75a1e895a92c376</td><td>International Journal of Emerging Engineering Research and Technology +<br/>Volume 3, Issue 12, December 2015, PP 128-133 +<br/>ISSN 2349-4395 (Print) & ISSN 2349-4409 (Online) +<br/>Facial Landmark Detection using Ensemble of Cascaded +<br/>Regressions +<br/><b>Faculty of Telecommunications, Technical University, Sofia, Bulgaria</b><br/><b>Faculty of Telecommunications, Technical University, Sofia, Bulgaria</b></td><td>('6203133', 'Martin Penev', 'martin penev')<br/>('1734848', 'Ognian Boumbarov', 'ognian boumbarov')</td><td></td></tr><tr><td>f074e86e003d5b7a3b6e1780d9c323598d93f3bc</td><td>OPEN ACCESS +<br/>ISSN 2075-1680 +<br/>Article +<br/>Characteristic Number: Theory and Its Application to +<br/>Shape Analysis +<br/><b>School of Software, Dalian University of Technology, Tuqiang St. 321, Dalian 116620, China</b><br/><b>School of Mathematical Sciences, Dalian University of Technology, Linggong Rd. 2, Dalian</b><br/>Tel.: +86-411-87571777; Fax: +86-411-87571567. +<br/>Received: 27 March 2014; in revised form: 28 April 2014 / Accepted: 28 April 2014 / +<br/>Published: 15 May 2014 +</td><td>('1710408', 'Xin Fan', 'xin fan')<br/>('7864960', 'Zhongxuan Luo', 'zhongxuan luo')<br/>('1732068', 'Jielin Zhang', 'jielin zhang')<br/>('2758604', 'Xinchen Zhou', 'xinchen zhou')<br/>('2235253', 'Qi Jia', 'qi jia')<br/>('3136305', 'Daiyun Luo', 'daiyun luo')</td><td>E-Mails: xin.fan@ieee.org (X.F.); jiaqi7166@gmail.com (Q.J.) +<br/>China; E-Mails: jielinzh@dlut.edu.cn (J.Z.); dasazxc@gmail.com (X.Z.); 419524597@qq.com (D.L.) +<br/>* Author to whom correspondence should be addressed; E-Mail: zxluo@dlut.edu.cn; +</td></tr><tr><td>f0a4a3fb6997334511d7b8fc090f9ce894679faf</td><td>Generative Face Completion +<br/><b>University of California, Merced</b><br/>2Adobe Research +</td><td>('1754382', 'Yijun Li', 'yijun li')<br/>('2391885', 'Sifei Liu', 'sifei liu')<br/>('1768964', 'Jimei Yang', 'jimei yang')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td>{yli62,sliu32,mhyang}@ucmerced.edu +<br/>jimyang@adobe.com +</td></tr><tr><td>f0681fc08f4d7198dcde803d69ca62f09f3db6c5</td><td>Spatiotemporal Features for Effective Facial +<br/>Expression Recognition +<br/>Hatice C¸ ınar Akakın and B¨ulent Sankur +<br/><b>Bogazici University, Bebek</b><br/>Istanbul +<br/>http://www.ee.boun.edu.tr +</td><td></td><td>{hatice.cinar,bulent.sankur}@boun.edu.tr +</td></tr><tr><td>f0f501e1e8726148d18e70c8e9f6feea9360d119</td><td>OULU 2015 +<br/>C 537 +<br/>U N I V E R S I TAT I S O U L U E N S I S +<br/>U N I V E R S I TAT I S O U L U E N S I S +<br/>CTECHNICA +<br/>CTECHNICA +<br/>C537etukansi.kesken.fm Page 1 Thursday, June 18, 2015 3:57 PM +<br/><b>UNIVERSITY OF OULU P.O. Box 8000 FI-90014 UNIVERSITY OF OULU FINLAND</b><br/>A C T A U N I V E R S I T A T I S O U L U E N S I S +<br/>ACTA +<br/>ACTA +<br/>Professor Esa Hohtola +<br/><b>University Lecturer Veli-Matti Ulvinen</b><br/><b>University Lecturer Anu Soikkeli</b><br/>Publications Editor Kirsti Nurkkala +<br/>ISBN 978-952-62-0872-5 (Paperback) +<br/>ISBN 978-952-62-0873-2 (PDF) +<br/>ISSN 0355-3213 (Print) +<br/>ISSN 1796-2226 (Online) +<br/>SOFTWARE-BASED +<br/>COUNTERMEASURES TO 2D +<br/>FACIAL SPOOFING ATTACKS +<br/><b>UNIVERSITY OF OULU GRADUATE SCHOOL</b><br/><b>UNIVERSITY OF OULU</b><br/>FACULTY OF INFORMATION TECHNOLOGY AND ELECTRICAL ENGINEERING, +<br/>DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING; +<br/>INFOTECH OULU +</td><td>('6433503', 'Santeri Palviainen', 'santeri palviainen')<br/>('3797304', 'Sanna Taskila', 'sanna taskila')<br/>('5451992', 'Olli Vuolteenaho', 'olli vuolteenaho')<br/>('6238085', 'Sinikka Eskelinen', 'sinikka eskelinen')<br/>('2165962', 'Jari Juga', 'jari juga')<br/>('5451992', 'Olli Vuolteenaho', 'olli vuolteenaho')<br/>('35709493', 'Jukka Komulainen', 'jukka komulainen')</td><td></td></tr><tr><td>f0398ee5291b153b716411c146a17d4af9cb0edc</td><td>LEARNING OPTICAL FLOW VIA DILATED NETWORKS AND OCCLUSION REASONING +<br/><b>University of California, Merced</b><br/>5200 N Lake Rd, Merced, CA, US +</td><td>('1749901', 'Yi Zhu', 'yi zhu')</td><td>{yzhu25, snewsam}@ucmerced.edu +</td></tr><tr><td>f0f0e94d333b4923ae42ee195df17c0df62ea0b1</td><td>Scaling Manifold Ranking Based Image Retrieval +<br/>†NTT Software Innovation Center, 3-9-11 Midori-cho Musashino-shi, Tokyo, Japan +<br/>‡NTT Service Evolution Laboratories, 1-1 Hikarinooka Yokosuka-shi, Kanagawa, Japan +<br/><b>California Institute of Technology, 1200 East California Boulevard Pasadena, California, USA</b><br/><b>Osaka University, 1-5 Yamadaoka, Suita-shi, Osaka, Japan</b></td><td>('32130106', 'Yasuhiro Fujiwara', 'yasuhiro fujiwara')<br/>('32285163', 'Go Irie', 'go irie')<br/>('46593534', 'Shari Kuroyama', 'shari kuroyama')<br/>('48075831', 'Makoto Onizuka', 'makoto onizuka')</td><td>{fujiwara.yasuhiro, irie.go}@lab.ntt.co.jp, kuroyama@caltech.edu, oni@acm.org +</td></tr><tr><td>f06b015bb19bd3c39ac5b1e4320566f8d83a0c84</td><td></td><td></td><td></td></tr><tr><td>f0a3f12469fa55ad0d40c21212d18c02be0d1264</td><td>Sparsity Sharing Embedding for Face +<br/>Verification +<br/>Department of Electrical Engineering, KAIST, Daejeon, Korea +</td><td>('2350325', 'Donghoon Lee', 'donghoon lee')<br/>('2857402', 'Hyunsin Park', 'hyunsin park')<br/>('8270717', 'Junyoung Chung', 'junyoung chung')<br/>('2126465', 'Youngook Song', 'youngook song')</td><td></td></tr><tr><td>f05ad40246656a977cf321c8299158435e3f3b61</td><td>Face Recognition Using Face Patch Networks +<br/><b>The Chinese University of Hong Kong</b></td><td>('2312486', 'Chaochao Lu', 'chaochao lu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('1678783', 'Deli Zhao', 'deli zhao')</td><td>{cclu,dlzhao,xtang}@ie.cuhk.edu.hk +</td></tr><tr><td>f02a6bccdaee14ab55ad94263539f4f33f1b15bb</td><td>Article +<br/>Segment-Tube: Spatio-Temporal Action Localization +<br/>in Untrimmed Videos with Per-Frame Segmentation +<br/><b>Institute of Arti cial Intelligence and Robotics, Xi an Jiaotong University, Xi an, Shannxi 710049, China</b><br/>Received: 23 April 2018; Accepted: 16 May 2018; Published: 22 May 2018 +</td><td>('40367806', 'Le Wang', 'le wang')<br/>('46809347', 'Xuhuan Duan', 'xuhuan duan')<br/>('46324995', 'Qilin Zhang', 'qilin zhang')<br/>('1786361', 'Zhenxing Niu', 'zhenxing niu')<br/>('1745420', 'Gang Hua', 'gang hua')<br/>('1715389', 'Nanning Zheng', 'nanning zheng')</td><td>duanxuhuan0123@stu.xjtu.edu.cn (X.D.); nnzheng@xjtu.edu.cn (N.Z.) +<br/>2 HERE Technologies, Chicago, IL 60606, USA; qilin.zhang@here.com +<br/>3 Alibaba Group, Hangzhou 311121, China; zhenxing.nzx@alibaba-inc.com +<br/>4 Microsoft Research, Redmond, WA 98052, USA; ganghua@microsoft.com +<br/>* Correspondence: lewang@xjtu.edu.cn; Tel.: +86-29-8266-8672 +</td></tr><tr><td>f7dea4454c2de0b96ab5cf95008ce7144292e52a</td><td></td><td></td><td></td></tr><tr><td>f781e50caa43be13c5ceb13f4ccc2abc7d1507c5</td><td>MVA2005 IAPR Conference on Machine VIsion Applications, May 16-18, 2005 Tsukuba Science City, Japan +<br/>12-1 +<br/>Towards Flexible and Intelligent Vision Systems +<br/>– From Thresholding to CHLAC – +<br/><b>University of Tokyo</b><br/>AISTy +<br/><b>y National Institute of Advanced Industrial Science and Technology</b><br/>Umezono 1-1-1, Tsukuba-shi, Ibaraki-ken, 305-8568 Japan +</td><td>('1809629', 'Nobuyuki Otsu', 'nobuyuki otsu')</td><td>Email: otsu.n@aist.go.jp +</td></tr><tr><td>f7b4bc4ef14349a6e66829a0101d5b21129dcf55</td><td>LONG ET AL.: TOWARDS LIGHT-WEIGHT ANNOTATIONS: FIR FOR ZSL +<br/>Towards Light-weight Annotations: Fuzzy +<br/>Interpolative Reasoning for Zero-shot Image +<br/>Classification +<br/>1 Open Lab, School of Computing +<br/><b>Newcastle University, UK</b><br/>2 Department of Computer Science and +<br/>Digital Technologies, Northumbria Uni- +<br/>versity, UK +<br/><b>Inception Institute of Arti cial</b><br/>gence, UAE +<br/>Intelli- +</td><td>('50363618', 'Yang Long', 'yang long')<br/>('48272923', 'Yao Tan', 'yao tan')<br/>('34975328', 'Daniel Organisciak', 'daniel organisciak')<br/>('1706028', 'Longzhi Yang', 'longzhi yang')<br/>('40799321', 'Ling Shao', 'ling shao')</td><td>yang.long@ieee.org +<br/>yao.tan@northumbria.ac.uk +<br/>d.organisciak@gmail.com +<br/>longzhi.yang@northumbria.ac.uk +<br/>ling.shao@ieee.org +</td></tr><tr><td>f7b422df567ce9813926461251517761e3e6cda0</td><td>FACE AGING WITH CONDITIONAL GENERATIVE ADVERSARIAL NETWORKS +<br/>(cid:63) Orange Labs, 4 rue Clos Courtel, 35512 Cesson-S´evign´e, France +<br/>† Eurecom, 450 route des Chappes, 06410 Biot, France +</td><td>('3116433', 'Grigory Antipov', 'grigory antipov')<br/>('2341854', 'Moez Baccouche', 'moez baccouche')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td></td></tr><tr><td>f7824758800a7b1a386db5bd35f84c81454d017a</td><td>KEPLER: Keypoint and Pose Estimation of Unconstrained Faces by +<br/>Learning Efficient H-CNN Regressors +<br/>Department of Electrical and Computer Engineering, CFAR and UMIACS +<br/><b>University of Maryland-College Park, USA</b></td><td>('50333013', 'Amit Kumar', 'amit kumar')<br/>('2943431', 'Azadeh Alavi', 'azadeh alavi')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{akumar14,azadeh,rama}@umiacs.umd.edu +</td></tr><tr><td>f74917fc0e55f4f5682909dcf6929abd19d33e2e</td><td>Workshop track - ICLR 2018 +<br/>GAN QUALITY INDEX (GQI) BY GAN-INDUCED +<br/>CLASSIFIER +<br/><b>The City College and the Graduate Center</b><br/><b>The City University of New York</b><br/>Department of Electrical & Computer Engineering +<br/><b>Northeastern University</b><br/>Microsoft Research +</td><td>('3105254', 'Yuancheng Ye', 'yuancheng ye')<br/>('39092100', 'Yue Wu', 'yue wu')<br/>('1689145', 'Lijuan Wang', 'lijuan wang')<br/>('2249952', 'Yinpeng Chen', 'yinpeng chen')<br/>('3419208', 'Zicheng Liu', 'zicheng liu')</td><td>yye@gradcenter.cuny.edu +<br/>ytian@ccny.cuny.edu +<br/>yuewu@ece.neu.edu +<br/>{lijuanw, yiche, zliu, zhang}@microsoft.com +</td></tr><tr><td>f740bac1484f2f2c70777db6d2a11cf4280081d6</td><td>Soft Locality Preserving Map (SLPM) for Facial Expression +<br/>Recognition +<br/>a Centre for Signal Processing, Department of Electronic and Information Engineering, The Hong +<br/><b>Kong Polytechnic University, Kowloon, Hong Kong</b><br/><b>b Computer Science, School of Electrical and Data Engineering, University of Technology, Sydney</b><br/>Australia +</td><td>('13671251', 'Cigdem Turan', 'cigdem turan')<br/>('1703078', 'Kin-Man Lam', 'kin-man lam')<br/>('1706670', 'Xiangjian He', 'xiangjian he')</td><td>E-mail addresses: cigdem.turan@connect.polyu.hk (C. Turan), enkmlam@polyu.edu.hk (K.-M. Lam), +<br/>xiangjian.he@uts.edu.au (X. He) +</td></tr><tr><td>f78fe101b21be36e98cd3da010051bb9b9829a1e</td><td>Hindawi +<br/>Computational Intelligence and Neuroscience +<br/>Volume 2018, Article ID 7208794, 10 pages +<br/>https://doi.org/10.1155/2018/7208794 +<br/>Research Article +<br/>Unsupervised Domain Adaptation for Facial Expression +<br/>Recognition Using Generative Adversarial Networks +<br/>1,2 +<br/><b>State Key Laboratory of Precision Measuring Technology and Instruments, Tianjin University, 300072, China</b><br/><b>Key Laboratory of MOEMS of the Ministry of Education, Tianjin University, 300072, China</b><br/>Received 14 April 2018; Accepted 19 June 2018; Published 9 July 2018 +<br/>Academic Editor: Ant´onio D. P. Correia +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>In the facial expression recognition task, a good-performing convolutional neural network (CNN) model trained on one dataset +<br/>(source dataset) usually performs poorly on another dataset (target dataset). This is because the feature distribution of the same +<br/>emotion varies in different datasets. To improve the cross-dataset accuracy of the CNN model, we introduce an unsupervised +<br/>domain adaptation method, which is especially suitable for unlabelled small target dataset. In order to solve the problem of lack of +<br/>samples from the target dataset, we train a generative adversarial network (GAN) on the target dataset and use the GAN generated +<br/>samples to fine-tune the model pretrained on the source dataset. In the process of fine-tuning, we give the unlabelled GAN generated +<br/>samples distributed pseudolabels dynamically according to the current prediction probabilities. Our method can be easily applied +<br/>to any existing convolutional neural networks (CNN). We demonstrate the effectiveness of our method on four facial expression +<br/>recognition datasets with two CNN structures and obtain inspiring results. +<br/>1. Introduction +<br/>Facial expressions recognition (FER) has a wide spectrum of +<br/>application potentials in human-computer interaction, cog- +<br/>nitive psychology, computational neuroscience, and medical +<br/>healthcare. In recent years, convolutional neural networks +<br/>(CNN) have achieved many exciting results in artificial +<br/>intelligent and pattern recognition and have been successfully +<br/>used in facial expression recognition [1]. Jaiswal et al. [2] +<br/>present a novel approach to facial action unit detection +<br/>using a combination of Convolutional and Bidirectional +<br/>Long Short-Term Memory Neural Networks (CNN-BLSTM), +<br/>which jointly learns shape, appearance, and dynamics in a +<br/>deep learning manner. You et al. [3] introduce a new data +<br/>set, which contains more than 3 million weakly labelled +<br/>images of different emotions. Esser et al. [4] develop a model +<br/>for efficient neuromorphic computing using the Deep CNN +<br/>technique. H-W.Ng et al. [5] develop a cascading fine-tuning +<br/>approach for emotion recognition. Neagoe et al. [6] propose +<br/>a model for subject independent emotion recognition from +<br/>facial expressions using combined CNN and DBN. However, +<br/>these CNN models are often trained and tested on the +<br/>same dataset, whereas the cross-dataset performance is less +<br/>concerned. Although the basic emotions defined by Ekman +<br/>and Friesen [7], anger, disgust, fear, happy, sadness, and +<br/>surprise, are believed to be universal, the way of expressing +<br/>these emotions can be quite diverse across different cultures, +<br/>ages, and genders [8]. As a result, a well-trained CNN model, +<br/>having high recognition accuracy on the training dataset, +<br/>usually performs poorly on other datasets. In order to make +<br/>the facial expression recognition system more practical, it +<br/>is necessary to improve the generalization ability of the +<br/>recognition model. +<br/>In this paper, we aim at improving the cross-dataset +<br/>accuracy of a CNN model on facial expression recognition. +<br/>One way to solve this problem is to rebuild models from +<br/>scratch using large-scale newly collected samples. Large +<br/>amounts of training samples, such as the dataset ImageNet [9] +<br/>containing over 15 million images, can reduce the overfitting +<br/>problem and help to train a reliable model. However, for +<br/>facial expression recognition, +<br/>it is expensive and some- +<br/>times even impossible to get enough labelled training data. +<br/>Therefore, we proposed an unsupervised domain adaptation +<br/>method, which is especially suitable for unlabelled small +</td><td>('47119020', 'Xiaoqing Wang', 'xiaoqing wang')<br/>('36142058', 'Xiangjun Wang', 'xiangjun wang')<br/>('3332231', 'Yubo Ni', 'yubo ni')<br/>('47119020', 'Xiaoqing Wang', 'xiaoqing wang')</td><td>Correspondence should be addressed to Xiangjun Wang; tjuxjw@126.com +</td></tr><tr><td>f79c97e7c3f9a98cf6f4a5d2431f149ffacae48f</td><td>Provided by the author(s) and NUI Galway in accordance with publisher policies. Please cite the published +<br/>version when available. +<br/>Title +<br/>On color texture normalization for active appearance models +<br/>Author(s) +<br/>Ionita, Mircea C.; Corcoran, Peter M.; Buzuloiu, Vasile +<br/>Publication +<br/>Date +<br/>2009-05-12 +<br/>Publication +<br/>Information +<br/>Ionita, M. C., Corcoran, P., & Buzuloiu, V. (2009). On Color +<br/>Texture Normalization for Active Appearance Models. Image +<br/>Processing, IEEE Transactions on, 18(6), 1372-1378. +<br/>Publisher +<br/>IEEE +<br/>Link to +<br/>publisher's +<br/>version +<br/>http://dx.doi.org/10.1109/TIP.2009.2017163 +<br/>Item record +<br/>http://hdl.handle.net/10379/1350 +<br/>Some rights reserved. For more information, please see the item record link above. +<br/>Downloaded 2017-06-17T22:38:27Z +</td><td></td><td></td></tr><tr><td>f7452a12f9bd927398e036ea6ede02da79097e6e</td><td></td><td></td><td></td></tr><tr><td>f7a271acccf9ec66c9b114d36eec284fbb89c7ef</td><td>Open Access +<br/>Research +<br/>Does attractiveness influence condom +<br/>use intentions in heterosexual men? +<br/>An experimental study +<br/>To cite: Eleftheriou A, +<br/>Bullock S, Graham CA, et al. +<br/>Does attractiveness influence +<br/>condom use intentions in +<br/>heterosexual men? +<br/>An experimental study. BMJ +<br/>Open 2016;6:e010883. +<br/>doi:10.1136/bmjopen-2015- +<br/>010883 +<br/>▸ Prepublication history for +<br/>this paper is available online. +<br/>To view these files please +<br/>visit the journal online +<br/>(http://dx.doi.org/10.1136/ +<br/>bmjopen-2015-010883). +<br/>Received 17 December 2015 +<br/>Revised 1 March 2016 +<br/>Accepted 7 April 2016 +<br/>1Department of Electronics +<br/>and Computer Science, +<br/><b>University of Southampton</b><br/>Southampton, UK +<br/><b>Institute for Complex</b><br/>Systems Simulation, +<br/><b>University of Southampton</b><br/>Southampton, UK +<br/>3Department of Computer +<br/><b>Science, University of Bristol</b><br/>Bristol, UK +<br/>4Centre for Sexual Health +<br/>Research, Department of +<br/><b>Psychology, University of</b><br/>Southampton, Southampton, +<br/>UK +<br/>Correspondence to +</td><td>('6093065', 'Anastasia Eleftheriou', 'anastasia eleftheriou')<br/>('1733871', 'Seth Bullock', 'seth bullock')<br/>('4712904', 'Cynthia A Graham', 'cynthia a graham')<br/>('48479171', 'Nicole Stone', 'nicole stone')<br/>('50227141', 'Roger Ingham', 'roger ingham')<br/>('6093065', 'Anastasia Eleftheriou', 'anastasia eleftheriou')</td><td>ae2n12@soton.ac.uk +</td></tr><tr><td>f7093b138fd31956e30d411a7043741dcb8ca4aa</td><td>Hierarchical Clustering in Face Similarity Score +<br/>Space +<br/>Jason Grant and Patrick Flynn +<br/>Department of Computer Science and Engineering +<br/><b>University of Notre Dame</b><br/>Notre Dame, IN 46556 +</td><td></td><td></td></tr><tr><td>f7dcadc5288653ec6764600c7c1e2b49c305dfaa</td><td>Copyright +<br/>by +<br/>Adriana Ivanova Kovashka +<br/>2014 +</td><td></td><td></td></tr><tr><td>f7de943aa75406fe5568fdbb08133ce0f9a765d4</td><td>Project 1.5: Human Identification at a Distance - Hornak, Adjeroh, Cukic, Gautum, & Ross +<br/>Project 1.5 +<br/>Biometric Identification and Surveillance1 +<br/>Year 5 Deliverable +<br/>Technical Report: +<br/>and +<br/>Research Challenges in Biometrics +<br/>Indexed biography of relevant biometric research literature +<br/>Donald Adjeroh, Bojan Cukic, Arun Ross +<br/>April, 2014 +<br/> +<br/>1 "This research was supported by the United States Department of Homeland Security through the National Center for Border Security +<br/>and Immigration (BORDERS) under grant number 2008-ST-061-BS0002. However, any opinions, findings, and conclusions or +<br/>recommendations in this document are those of the authors and do not necessarily reflect views of the United States Department of +<br/>Homeland Security." +</td><td>('4800511', 'Don Adjeroh', 'don adjeroh')<br/>('1702603', 'Bojan Cukic', 'bojan cukic')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td>donald.adjeroh@mail.wvu.edu; bojan.cukic@mail.wvu.edu; arun.ross@mail.wvu.edu +</td></tr><tr><td>f75852386e563ca580a48b18420e446be45fcf8d</td><td>ILLUMINATION INVARIANT FACE RECOGNITION +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/> +<br/>ENEE 631: Digital Image and Video Processing +<br/>Instructor: Dr. K. J. Ray Liu +<br/>Term Project - Spring 2006 +<br/>1. +<br/>INTRODUCTION +<br/> +<br/> +<br/>The performance of the Face Recognition algorithms is severely affected by two +<br/>important factors: the change in Pose and Illumination conditions of the subjects. The +<br/>changes in Illumination conditions of the subjects can be so drastic that, the variation in +<br/>lighting will be of the similar order as that of the variation due to the change in subjects +<br/>[1] and this can result in misclassification. +<br/> +<br/> For example, in the acquisition of the face of a person from a real time video, the +<br/>ambient conditions will cause different lighting variations on the tracked face. Some +<br/>examples of images with different illumination conditions are shown in Fig. 1. In this +<br/>project, we study some algorithms that are capable of performing Illumination Invariant +<br/>Face Recognition. The performances of these algorithms were compared on the CMU- +<br/>Illumination dataset [13], by using the entire face as the input to the algorithms. Then, a +<br/>model of dividing the face into four regions is proposed and the performance of the +<br/>algorithms on these new features is analyzed. +<br/> +<br/> +</td><td>('33692583', 'Raghuraman Gopalan', 'raghuraman gopalan')</td><td>raghuram@umd.edu +</td></tr><tr><td>f7c50d2be9fba0e4527fd9fbe3095e9d9a94fdd3</td><td>Large Margin Multi-Metric Learning for Face +<br/>and Kinship Verification in the Wild +<br/><b>School of EEE, Nanyang Technological University, Singapore</b><br/>2Advanced Digital Sciences Center, Singapore +</td><td>('34651153', 'Junlin Hu', 'junlin hu')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('34316743', 'Junsong Yuan', 'junsong yuan')<br/>('1689805', 'Yap-Peng Tan', 'yap-peng tan')</td><td></td></tr><tr><td>f78863f4e7c4c57744715abe524ae4256be884a9</td><td></td><td></td><td></td></tr><tr><td>f77c9bf5beec7c975584e8087aae8d679664a1eb</td><td>Local Deep Neural Networks for Age and Gender Classification +<br/>March 27, 2017 +</td><td>('9949538', 'Zukang Liao', 'zukang liao')<br/>('2403354', 'Stavros Petridis', 'stavros petridis')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>f7ba77d23a0eea5a3034a1833b2d2552cb42fb7a</td><td>This is a pre-print of the original paper accepted at the International Joint Conference on Biometrics (IJCB) 2017. +<br/>LOTS about Attacking Deep Features +<br/>Vision and Security Technology (VAST) Lab +<br/><b>University of Colorado, Colorado Springs, USA</b></td><td>('2974221', 'Andras Rozsa', 'andras rozsa')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')</td><td>{arozsa,mgunther,tboult}@vast.uccs.edu +</td></tr><tr><td>e8686663aec64f4414eba6a0f821ab9eb9f93e38</td><td>IMPROVING SHAPE-BASED FACE RECOGNITION BY MEANS OF A SUPERVISED +<br/>DISCRIMINANT HAUSDORFF DISTANCE +<br/>J.L. Alba +<br/>, A. Pujol +<br/>†† +<br/>, A. L´opez +<br/>††† +<br/>and J.J. Villanueva +<br/>††† +<br/><b>University of Vigo, Spain</b><br/>†††Centre de Visio per Computador, Universitat Autonoma de Barcelona, Spain +<br/>††Digital Pointer MVT +</td><td></td><td></td></tr><tr><td>e82360682c4da11f136f3fccb73a31d7fd195694</td><td><b>AALTO UNIVERSITY</b><br/>SCHOOL OF SCIENCE AND TECHNOLOGY +<br/>Faculty of Information and Natural Science +<br/>Department of Information and Computer Science +<br/>Online Face Recognition with +<br/>Application to Proactive Augmented +<br/>Reality +<br/>Master’s Thesis submitted in partial fulfillment of the requirements for the +<br/>degree of Master of Science in Technology. +<br/>Espoo, May 25, 2010 +<br/>Supervisor: +<br/>Instructor: +<br/>Professor Erkki Oja +</td><td>('1700492', 'Jing Wu', 'jing wu')<br/>('1758971', 'Markus Koskela', 'markus koskela')</td><td></td></tr><tr><td>e8410c4cd1689829c15bd1f34995eb3bd4321069</td><td></td><td></td><td></td></tr><tr><td>e8fdacbd708feb60fd6e7843b048bf3c4387c6db</td><td>Deep Learning +<br/>Hinnerup Net A/S +<br/>www.hinnerup.net +<br/>July 4, 2014 +<br/>Introduction +<br/>Deep learning is a topic in the field of artificial intelligence (AI) and is a relatively +<br/>new research area although based on the popular artificial neural networks (supposedly +<br/>mirroring brain function). With the development of the perceptron in the 1950s and +<br/>1960s by Frank RosenBlatt, research began on artificial neural networks. To further +<br/>mimic the architectural depth of the brain, researchers wanted to train a deep multi- +<br/>layer neural network – this, however, did not happen until Geoffrey Hinton in 2006 +<br/>introduced Deep Belief Networks [1]. +<br/>Recently, the topic of deep learning has gained public interest. Large web companies such +<br/>as Google and Facebook have a focused research on AI and an ever increasing amount +<br/>of compute power, which has led to researchers finally being able to produce results +<br/>that are of interest to the general public. In July 2012 Google trained a deep learning +<br/>network on YouTube videos with the remarkable result that the network learned to +<br/>recognize humans as well as cats [6], and in January this year Google successfully used +<br/>deep learning on Street View images to automatically recognize house numbers with +<br/>an accuracy comparable to that of a human operator [5]. In March this year Facebook +<br/>announced their DeepFace algorithm that is able to match faces in photos with Facebook +<br/>users almost as accurately as a human can do [9]. +<br/>Deep learning and other AI are here to stay and will become more and more present in +<br/>our daily lives, so we had better make ourselves acquainted with the technology. Let’s +<br/>dive into the deep water and try not to drown! +<br/>Data Representations +<br/>Before presenting data to an AI algorithm, we would normally prepare the data to make +<br/>it feasible to work with. For instance, if the data consists of images, we would take each +</td><td></td><td></td></tr><tr><td>e8f0f9b74db6794830baa2cab48d99d8724e8cb6</td><td>Active Image Labeling and Its Application to +<br/>Facial Action Labeling +<br/><b>Electrical, Computer, Rensselaer Polytechnic Institute</b><br/><b>Visualization and Computer Vision Lab, GE Global Research Center</b></td><td>('40396543', 'Lei Zhang', 'lei zhang')<br/>('1686235', 'Yan Tong', 'yan tong')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>zhangl2@rpi.edu,tongyan@research.ge.com,qji@ecse.rpi.edu +</td></tr><tr><td>e8b2a98f87b7b2593b4a046464c1ec63bfd13b51</td><td>CMS-RCNN: Contextual Multi-Scale +<br/>Region-based CNN for Unconstrained Face +<br/>Detection +</td><td>('3117715', 'Chenchen Zhu', 'chenchen zhu')<br/>('3049981', 'Yutong Zheng', 'yutong zheng')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td></td></tr><tr><td>e87d6c284cdd6828dfe7c092087fbd9ff5091ee4</td><td>Unsupervised Creation of Parameterized Avatars +<br/>1Facebook AI Research +<br/><b>School of Computer Science, Tel Aviv University</b></td><td>('1776343', 'Lior Wolf', 'lior wolf')<br/>('2188620', 'Yaniv Taigman', 'yaniv taigman')<br/>('33964593', 'Adam Polyak', 'adam polyak')</td><td></td></tr><tr><td>e8523c4ac9d7aa21f3eb4062e09f2a3bc1eedcf7</td><td>Towards End-to-End Face Recognition through Alignment Learning +<br/><b>Tsinghua University</b><br/>Beijing, China, 100084 +</td><td>('8802368', 'Yuanyi Zhong', 'yuanyi zhong')<br/>('1752427', 'Jiansheng Chen', 'jiansheng chen')<br/>('39071060', 'Bo Huang', 'bo huang')</td><td>zhongyy13@mails.tsinghua.edu.cn, jschenthu@mail.tsinghua.edu.cn, huangb14@mails.tsinghua.edu.cn +</td></tr><tr><td>e85a255a970ee4c1eecc3e3d110e157f3e0a4629</td><td>Fusing Hierarchical Convolutional Features for Human Body Segmentation and +<br/>Clothing Fashion Classification +<br/><b>School of Computer Science, Wuhan University, P.R. China</b></td><td>('47294008', 'Zheng Zhang', 'zheng zhang')<br/>('3127916', 'Chengfang Song', 'chengfang song')<br/>('4793870', 'Qin Zou', 'qin zou')</td><td>E-mails: {zhangzheng, songchf, qzou}@whu.edu.cn +</td></tr><tr><td>e8c9dcbf56714db53063b9c367e3e44300141ff6</td><td>Automated FACS face analysis benefits from the addition of velocity +<br/>Get The FACS Fast: +<br/>Timothy R. Brick +<br/><b>University of Virginia</b><br/>Charlottesville, VA 22904 +<br/>Michael D. Hunter +<br/><b>University of Virginia</b><br/>Charlottesville, VA 22904 +<br/>Jeffrey F. Cohn +<br/><b>University of Pittsburgh</b><br/>Pittsburgh, PA 15260 +</td><td></td><td>tbrick@virginia.edu +<br/>mhunter@virginia.edu +<br/>jeffcohn@cs.cmu.edu +</td></tr><tr><td>e8d1b134d48eb0928bc999923a4e092537e106f6</td><td>WEIGHTED MULTI-REGION CONVOLUTIONAL NEURAL NETWORK FOR ACTION +<br/>RECOGNITION WITH LOW-LATENCY ONLINE PREDICTION +<br/><b>cid:63)University of Science and Technology of China, Hefei, Anhui, China</b><br/>†HERE Technologies, Chicago, Illinois, USA +</td><td>('49417387', 'Yunfeng Wang', 'yunfeng wang')<br/>('38272296', 'Wengang Zhou', 'wengang zhou')<br/>('46324995', 'Qilin Zhang', 'qilin zhang')<br/>('49897466', 'Xiaotian Zhu', 'xiaotian zhu')<br/>('7179232', 'Houqiang Li', 'houqiang li')</td><td></td></tr><tr><td>e8c6c3fc9b52dffb15fe115702c6f159d955d308</td><td>13 +<br/>Linear Subspace Learning for +<br/>Facial Expression Analysis +<br/>Philips Research +<br/>The Netherlands +<br/>1. Introduction +<br/>Facial expression, resulting from movements of the facial muscles, is one of the most +<br/>powerful, natural, and immediate means for human beings to communicate their emotions +<br/>and intentions. Some examples of facial expressions are shown in Fig. 1. Darwin (1872) was +<br/>the first to describe in detail the specific facial expressions associated with emotions in +<br/>animals and humans; he argued that all mammals show emotions reliably in their faces. +<br/>Psychological studies (Mehrabian, 1968; Ambady & Rosenthal, 1992) indicate that facial +<br/>expressions, with other non-verbal cues, play a major and fundamental role in face-to-face +<br/>communication. +<br/>Fig. 1. Facial expressions of George W. Bush. +<br/>Machine analysis of facial expressions, enabling computers to analyze and interpret facial +<br/>expressions as humans do, has many important applications including intelligent human- +<br/>computer interaction, computer animation, surveillance and security, medical diagnosis, +<br/>law enforcement, and awareness system (Shan, 2007). Driven by its potential applications +<br/>and theoretical interests of cognitive and psychological scientists, automatic facial +<br/>expression analysis has attracted much attention in last two decades (Pantic & Rothkrantz, +<br/>2000a; Fasel & Luettin, 2003; Tian et al, 2005; Pantic & Bartlett, 2007). It has been studied in +<br/>multiple disciplines such as psychology, cognitive science, computer vision, pattern +<br/>Source: Machine Learning, Book edited by: Abdelhamid Mellouk and Abdennacer Chebira, +<br/> ISBN 978-3-902613-56-1, pp. 450, February 2009, I-Tech, Vienna, Austria +<br/>www.intechopen.com +</td><td>('10795229', 'Caifeng Shan', 'caifeng shan')</td><td></td></tr><tr><td>e8b3a257a0a44d2859862cdec91c8841dc69144d</td><td>Liquid Pouring Monitoring via +<br/>Rich Sensory Inputs +<br/><b>National Tsing Hua University, Taiwan</b><br/><b>Stanford University, USA</b></td><td>('27555915', 'Tz-Ying Wu', 'tz-ying wu')<br/>('9618379', 'Juan-Ting Lin', 'juan-ting lin')<br/>('27538483', 'Chan-Wei Hu', 'chan-wei hu')<br/>('9200530', 'Juan Carlos Niebles', 'juan carlos niebles')<br/>('46611107', 'Min Sun', 'min sun')</td><td>{gina9726, brade31919, johnsonwang0810, huchanwei1204}@gmail.com, +<br/>sunmin@ee.nthu.edu.tw +<br/>jniebles@cs.stanford.edu +</td></tr><tr><td>fa90b825346a51562d42f6b59a343b98ea2e501a</td><td>Modeling Naive Psychology of Characters in Simple Commonsense Stories +<br/><b>Paul G. Allen School of Computer Science and Engineering, University of Washington</b><br/><b>Allen Institute for Arti cial Intelligence</b><br/><b>Information Sciences Institute and Computer Science, University of Southern California</b></td><td>('2516777', 'Hannah Rashkin', 'hannah rashkin')<br/>('2691021', 'Antoine Bosselut', 'antoine bosselut')<br/>('2729164', 'Maarten Sap', 'maarten sap')<br/>('1710034', 'Kevin Knight', 'kevin knight')<br/>('1699545', 'Yejin Choi', 'yejin choi')</td><td>{hrashkin,msap,antoineb,yejin}@cs.washington.edu +<br/>knight@isi.edu +</td></tr><tr><td>fab83bf8d7cab8fe069796b33d2a6bd70c8cefc6</td><td>Draft: Evaluation Guidelines for Gender +<br/>Classification and Age Estimation +<br/>July 1, 2011 +<br/>Introduction +<br/>In previous research on gender classification and age estimation did not use a +<br/>standardised evaluation procedure. This makes comparison the different ap- +<br/>proaches difficult. +<br/>Thus we propose here a benchmarking and evaluation protocol for gender +<br/>classification as well as age estimation to set a common ground for future re- +<br/>search in these two areas. +<br/>The evaluations are designed such that there is one scenario under controlled +<br/>labratory conditions and one under uncontrolled real life conditions. +<br/>The datasets were selected with the criteria of being publicly available for +<br/>research purposes. +<br/>File lists for the folds corresponding to the individual benchmarking proto- +<br/>cols will be provided over our website at http://face.cs.kit.edu/befit. We +<br/>will provide two kinds of folds for each of the tasks and conditions: one set of +<br/>folds using the whole dataset and one set of folds using a reduced dataset, which +<br/>is approximately balanced in terms of age, gender and ethnicity. +<br/>2 Gender Classification +<br/>In this task the goal is to determine the gender of the persons depicted in the +<br/>individual images. +<br/>2.1 Data +<br/>In previous works one of the most commonly used databases is the Feret database [1, +<br/>2]. We decided here not to take this database, because of its low number of im- +<br/>ages. +</td><td>('40303076', 'Tobias Gehrig', 'tobias gehrig')<br/>('39504159', 'Matthias Steiner', 'matthias steiner')</td><td>{tobias.gehrig, ekenel}@kit.edu +</td></tr><tr><td>faeefc5da67421ecd71d400f1505cfacb990119c</td><td>Original research +<br/>published: 20 November 2017 +<br/>doi: 10.3389/frobt.2017.00061 +<br/>PastVision+: Thermovisual inference +<br/>of recent Medicine intake by +<br/>Detecting heated Objects and +<br/>cooled lips +<br/><b>Intelligent Systems Laboratory, Halmstad University, Halmstad, Sweden</b><br/>This article addresses the problem of how a robot can infer what a person has done +<br/>recently, with a focus on checking oral medicine intake in dementia patients. We present +<br/>PastVision+, an approach showing how thermovisual cues in objects and humans can +<br/>be leveraged to infer recent unobserved human–object interactions. Our expectation +<br/>is that this approach can provide enhanced speed and robustness compared to exist- +<br/>ing methods, because our approach can draw inferences from single images without +<br/>needing to wait to observe ongoing actions and can deal with short-lasting occlusions; +<br/>when combined, we expect a potential improvement in accuracy due to the extra infor- +<br/>mation from knowing what a person has recently done. To evaluate our approach, we +<br/>obtained some data in which an experimenter touched medicine packages and a glass +<br/>of water to simulate intake of oral medicine, for a challenging scenario in which some +<br/>touches were conducted in front of a warm background. Results were promising, with +<br/>a detection accuracy of touched objects of 50% at the 15 s mark and 0% at the 60 s +<br/>mark, and a detection accuracy of cooled lips of about 100 and 60% at the 15 s mark +<br/>for cold and tepid water, respectively. Furthermore, we conducted a follow-up check for +<br/>another challenging scenario in which some participants pretended to take medicine or +<br/>otherwise touched a medicine package: accuracies of inferring object touches, mouth +<br/>touches, and actions were 72.2, 80.3, and 58.3% initially, and 50.0, 81.7, and 50.0% +<br/>at the 15 s mark, with a rate of 89.0% for person identification. The results suggested +<br/>some areas in which further improvements would be possible, toward facilitating robot +<br/>inference of human actions, in the context of medicine intake monitoring. +<br/>Keywords: thermovisual inference, touch detection, medicine intake, action recognition, monitoring, near past +<br/>inference +<br/>1. inTrODUcTiOn +<br/>This article addresses the problem of how a robot can detect what a person has touched recently, +<br/>with a focus on checking oral medicine intake in dementia patients. +<br/>Detecting recent touches would be useful because touch is a typical component of many human– +<br/>object interactions; moreover, knowing which objects have been touched allows inference into +<br/>what actions have been conducted, which is an important requirement for robots to collaborate +<br/>effectively with people (Vernon et al., 2016). For example, touches to a stove, door handle, or pill +<br/>bottle can occur as a result of cooking, leaving one’s house, or taking medicine, all of which could +<br/>potentially be dangerous for a person with dementia, if they forget to turn off the heat, lose their +<br/>way, or make a mistake. Here, we focus on the latter problem of medicine adherence—whose +<br/>Edited by: +<br/>Alberto Montebelli, +<br/><b>University of Sk vde, Sweden</b><br/>Reviewed by: +<br/>Sam Neymotin, +<br/><b>Brown University, United States</b><br/>Per Backlund, +<br/><b>University of Sk vde, Sweden</b><br/>Fernando Bevilacqua, +<br/><b>University of Sk vde, Sweden</b><br/>(in collaboration with Per Backlund) +<br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Computational Intelligence, +<br/>a section of the journal +<br/>Frontiers in Robotics and AI +<br/>Received: 15 May 2017 +<br/>Accepted: 02 November 2017 +<br/>Published: 20 November 2017 +<br/>Citation: +<br/>Cooney M and Bigun J (2017) +<br/>PastVision+: Thermovisual Inference +<br/>of Recent Medicine Intake by +<br/>Detecting Heated Objects +<br/>and Cooled Lips. +<br/>Front. Robot. AI 4:61. +<br/>doi: 10.3389/frobt.2017.00061 +<br/>Frontiers in Robotics and AI | www.frontiersin.org +<br/>November 2017 | Volume 4 | Article 61 +</td><td>('7149684', 'Martin Cooney', 'martin cooney')<br/>('5058247', 'Josef Bigun', 'josef bigun')<br/>('7149684', 'Martin Cooney', 'martin cooney')</td><td>martin.daniel.cooney@gmail.com +</td></tr><tr><td>fa4f59397f964a23e3c10335c67d9a24ef532d5c</td><td>DAP3D-Net: Where, What and How Actions Occur in Videos? +<br/>Department of Computer Science and Digital Technologies +<br/><b>Northumbria University, Newcastle upon Tyne, NE1 8ST, UK</b></td><td>('40241836', 'Li Liu', 'li liu')<br/>('47942896', 'Yi Zhou', 'yi zhou')<br/>('40799321', 'Ling Shao', 'ling shao')</td><td>li2.liu@northumbria.ac.uk, m.y.yu@ieee.org, ling.shao@ieee.org +</td></tr><tr><td>fa08a4da5f2fa39632d90ce3a2e1688d147ece61</td><td>Supplementary material for +<br/>“Unsupervised Creation of Parameterized Avatars” +<br/>1 Summary of Notations +<br/>Tab. 1 itemizes the symbols used in the submission. Fig. 2,3,4 of the main text illustrate many of these +<br/>symbols. +<br/>2 DANN results +<br/>Fig. 1 shows side by side samples of the original image and the emoji generated by the method of [1]. +<br/>As can be seen, these results do not preserve the identity very well, despite considerable effort invested in +<br/>finding suitable architectures. +<br/>3 Multiple Images Per Person +<br/>Following [4], we evaluate the visual quality that is obtained per person and not just per image, by testing +<br/>TOS on the Facescrub dataset [3]. For each person p, we considered the set of their images Xp, and selected +<br/>the emoji that was most similar to their source image, i.e., the one for which: +<br/>||f (x) − f (e(c(G(x))))||. +<br/>argmin +<br/>x∈Xp +<br/>(1) +<br/>Fig. 2 depicts the results obtained by this selection method on sample images form the Facescrub dataset +<br/>(it is an extension of Fig. 7 of the main text). The figure also shows, for comparison, the DTN [4] result for +<br/>the same image. +<br/>4 Detailed Architecture of the Various Networks +<br/>In this section we describe the architectures of the networks used in for the emoji and avatar experiments. +<br/>4.1 TOS +<br/>Network g maps DeepFace’s 256-dimensional representation [5] into 64 × 64 RGB emoji images. Follow- +<br/>ing [4], this is done through a network with 9 blocks, each consisting of a convolution, batch-normalization +<br/>and ReLU, except the last layer which employs Tanh activation. The odd blocks 1,3,5,7,9 perform upscaling +<br/>convolutions with 512-256-128-64-3 filters respectively of spatial size 4 × 4. The even ones perform 1 × 1 +<br/>convolutions [2]. The odd blocks use a stride of 2 and padding of 1, excluding the first one which does not +<br/>use stride or padding. +<br/>Network e maps emoji parameterization into the matching 64× 64 RGB emoji. The parameterization is +<br/>given as binary vectors in R813 for emojis; Avatar parameterization is in R354. While there are dependencies +<br/>among the various dimensions (an emoji cannot have two hairstyles at once), the binary representation is +<br/>chosen for its simplicity and generality. e is trained in a fully supervised way, using pairs of matching +<br/>parameterization vectors and images in a supervised manner. +<br/>The architecture of e employs five upscaling convolutions with 512-256-128-64-3 filters respectively, +<br/>each of spatial size 4×4. All layers except the last one are batch normalized followed by a ReLU activation. +<br/>The last layer is followed by Tanh activation, generating an RGB image with values in range [−1, 1]. All +<br/>the layers use a stride of 2 and padding of 1, excluding the first one which does not use stride or padding. +</td><td></td><td></td></tr><tr><td>fab2fc6882872746498b362825184c0fb7d810e4</td><td>RESEARCH ARTICLE +<br/>Right wing authoritarianism is associated with +<br/>race bias in face detection +<br/>1 Univ. Grenoble Alpes, LPNC, Grenoble, France, 2 CNRS, LPNC UMR 5105, Grenoble, France, 3 IPSY, +<br/><b>Universite Catholique de Louvain, Louvain-la-Neuve, Belgium, 4 The Queensland Brain Institute, The</b><br/><b>University of Queensland, St Lucia QLD Australia, 5 Institut Universitaire de France, Paris, France</b></td><td>('3128194', 'Brice Beffara', 'brice beffara')<br/>('2066203', 'Jessica McFadyen', 'jessica mcfadyen')<br/>('2634712', 'Martial Mermillod', 'martial mermillod')</td><td>* amelie.bret@univ-grenoble-alpes.fr +</td></tr><tr><td>faead8f2eb54c7bc33bc7d0569adc7a4c2ec4c3b</td><td></td><td></td><td></td></tr><tr><td>fa24bf887d3b3f6f58f8305dcd076f0ccc30272a</td><td>JMLR: Workshop and Conference Proceedings 39:189–204, 2014 +<br/>ACML 2014 +<br/>Interval Insensitive Loss for Ordinal Classification +<br/>Vojtˇech Franc +<br/>V´aclav Hlav´aˇc +<br/>Center for Machine Perception, Department of Cybernetics, Faculty of Electrical Engineering, Czech +<br/><b>Technical University in Prague, Technick a 2, 166 27 Prague 6 Czech Republic</b><br/>Editor: Dinh Phung and Hang Li +</td><td>('2742026', 'Kostiantyn Antoniuk', 'kostiantyn antoniuk')</td><td>antonkos@cmp.felk.cvut.cz +<br/>xfrancv@cmp.felk.cvut.cz +<br/>hlavac@fel.cvut.cz +</td></tr><tr><td>fac8cff9052fc5fab7d5ef114d1342daba5e4b82</td><td>(CV last updated Oct. 5th, 2009.) +<br/>www.stat.cmu.edu/~abrock +<br/>1-412-478-3609 +<br/>Citizenship: U.S., Australia (dual) +<br/>Education +<br/>1994-1998 +<br/>: Ph.D., Department of Statistics and Department of of Electrical Engineering at +<br/><b>Melbourne University, Advisors: K. Borovkov, R. Evans</b><br/>1993 +<br/>: Honours Science Degree (in the Department of Statistics) completed at Melbourne +<br/><b>University (H</b><br/>1988-92 +<br/>: Bachelor of Science and Bachelor of Engineering with Honours completed at Mel- +<br/><b>bourne University</b><br/>Employment +<br/>2007+ +<br/><b>Carnegie Mellon University</b><br/>2007-2009 +<br/>: Senior Analyst, Horton Point LLC (Hedge Fund Management Company) +<br/>2006-2007 +<br/>: Associate Professor, Department of Statistics, Carnegie Mellon Uniuversity +<br/>2005-2007 +<br/>: Affiliated faculty member, Machine Learning Department (formerly known as the +<br/><b>Center for Automated Learning and Discovery), Carnegie Mellon University</b><br/>2003-2007 +<br/><b>Faculty member, Parallel Data Lab (PDL), Carnegie Mellon University</b><br/>2002-2005 +<br/><b>Carnegie Mellon University</b><br/>1999-2002 +<br/><b>Carnegie Mellon University</b><br/>1998-1999 +<br/>: Research Fellow, Department of Electrical and Electronic Engineering, The Univer- +<br/>sity of Melbourne +<br/>1993-1995 +<br/><b>Sessional Tutor, The University of Melbourne</b></td><td>('1680307', 'Anthony Brockwell', 'anthony brockwell')</td><td>anthony.brockwell@gmail.com +</td></tr><tr><td>faa29975169ba3bbb954e518bc9814a5819876f6</td><td>Evolution-Preserving Dense Trajectory Descriptors +<br/><b>Stony Brook University, Stony Brook, NY 11794, USA</b></td><td>('2295608', 'Yang Wang', 'yang wang')<br/>('3482497', 'Vinh Tran', 'vinh tran')<br/>('2356016', 'Minh Hoai', 'minh hoai')</td><td>{wang33, tquangvinh, minhhoai}@cs.stonybrook.edu +</td></tr><tr><td>fafe69a00565895c7d57ad09ef44ce9ddd5a6caa</td><td>Applied Mathematics, 2012, 3, 2071-2079 +<br/>http://dx.doi.org/10.4236/am.2012.312A286 Published Online December 2012 (http://www.SciRP.org/journal/am) +<br/>Gaussian Mixture Models for Human Face Recognition +<br/>under Illumination Variations +<br/><b>Mihaylo College of Business and Economics</b><br/><b>California State University, Fullerton, USA</b><br/>Received August 18, 2012; revised September 18, 2012; accepted September 25, 2012 +</td><td>('2046854', 'Sinjini Mitra', 'sinjini mitra')</td><td>Email: smitra@fullerton.edu +</td></tr><tr><td>faf5583063682e70dedc4466ac0f74eeb63169e7</td><td></td><td></td><td>HolisticPersonProcessing:FacesWithBodiesTelltheWholeStoryHillelAviezerPrincetonUniversityandNewYorkUniversityYaacovTropeNewYorkUniversityAlexanderTodorovPrincetonUniversityFacesandbodiesaretypicallyencounteredsimultaneously,yetlittleresearchhasexploredthevisualprocessingofthefullperson.Specifically,itisunknownwhetherthefaceandbodyareperceivedasdistinctcomponentsorasanintegrated,gestalt-likeunit.Toexaminethisquestion,weinvestigatedwhetheremotionalface–bodycompositesareprocessedinaholistic-likemannerbyusingavariantofthecompositefacetask,ameasureofholisticprocessing.Participantsjudgedfacialexpressionscombinedwithemotionallycongruentorincongruentbodiesthathavebeenshowntoinfluencetherecognitionofemotionfromtheface.Critically,thefaceswereeitheralignedwiththebodyinanaturalpositionormisalignedinamannerthatbreakstheecologicalpersonform.Convergingdatafrom3experimentsconfirmthatbreakingthepersonformreducesthefacilitatinginfluenceofcongruentbodycontextaswellastheimpedinginfluenceofincongruentbodycontextontherecognitionofemotionfromtheface.Theseresultsshowthatfacesandbodiesareprocessedasasingleunitandsupportthenotionofacompositepersoneffectanalogoustotheclassiceffectdescribedforfaces.Keywords:emotionperception,contexteffects,facialandbodyexpressions,holisticperception,com-positeeffectAglanceisusuallysufficientforextractingagreatdealofsocialinformationfromotherpeople(Adolphs,2002).Perceptualcuestocharacteristicssuchasgender,sexualorientation,emotionalex-pression,attractiveness,andpersonalitytraitscanbefoundinboththefaceandthebody(e.g.,facecues,Adolphs,2003;Calder&Young,2005;Ekman,1993;Elfenbein&Ambady,2002;Haxby,Hoffman,&Gobbini,2000;Rule,Ambady,&Hallett,2009;Thornhill&Gangestad,1999;Todorov&Duchaine,2008;Todo-rov,Pakrashi,&Oosterhof,2009;Willis&Todorov,2006;Ze-browitz,Hall,Murphy,&Rhodes,2002;Zebrowitz&Montepare,2008;bodycues,deGelderetal.,2006;Johnson,Gill,Reichman,&Tassinary,2007;Peelen&Downing,2005;Stevenage,Nixon,&Vince,1999;Wallbott,1998).Todate,mostresearchershaveinvestigatedthefaceandthebodyasdiscreteperceptualunits,focusingontheprocessingofeachsourceinisolation.Althoughthisapproachhasprovedex-tremelyfruitfulforcharacterizingtheuniqueperceptualcontribu-tionsofthefaceandbody,surprisinglylittleisknownabouttheprocessingofbothsourcescombined.Theaimofthecurrentstudywastoshedlightontheperceptualprocessingofthefullpersonbyexaminingwhetherthefaceandbodyinconjunctionareprocessedasaholistic“personunit.”Onthebasisofpreviousaccounts,onemaypredictthatfacesandbodiesareprocessedastwovisualcomponentsofsocialinformation(Wallbott,1998).Theseviewsarguethatfacesandbodiesmaydifferinvalue,intensity,andclarity,andconsequentlytheinformationfromeachmustbeweightedandcombinedbythecognitivesysteminordertoreachaconclusionaboutthetarget(Ekman,Friesen,&Ellsworth,1982;Ellison&Massaro,1997;Trope,1986;Wallbott,1998).Accordingtothisapproach,thefaceandbodymayinfluenceeachother.However,theinfluenceisnotsynergistic,andtheperceptionofthefaceandbodyisequaltotheweightedsumoftheirparts(Wallbott,1998).Bycontrast,thehypothesisofferedhereisthatthefaceandbodyaresubcomponentsofalargerperceptualpersonunit.Fromanecologicalperspectivethisseemslikelybecauseundernaturalconditions,thevisualsystemrarelyencountersisolatedfacesandbodies(McArthur&Baron,1983;Russell,1997).Accordingtothisview,thefaceandbodyformaunitaryperceptthatmayencompassdifferentpropertiesthanthetwosourcesofinformationseparately.Inotherwords,theinformationreadoutfromthefullpersonmaybemorethanthesumofthefaceandbodyalone.HolisticProcessingandtheCompositeEffectPastresearchonsocialperceptionexaminingunitizedgestaltprocessinghasfocusedprimarilyontheface.Indeed,ahallmarkoffaceperceptionisholisticprocessingbywhichindividualfacialcomponentsbecomeintegratedintoawhole-faceunit(Farah,Wilson,Drain,&Tanaka,1995;Tanaka&Farah,1993).Althoughisolatedfacialcomponentsdobearspecificinformation(Smith,Cottrell,Gosselin,&Schyns,2005;Whalenetal.,2004),theirarrangementinthenaturalfaceconfigurationresultsinaninte-ThisarticlewaspublishedOnlineFirstFebruary20,2012.HillelAviezer,DepartmentofPsychology,PrincetonUniversity,andDepartmentofPsychology,NewYorkUniversity;YaacovTrope,Depart-mentofPsychology,NewYorkUniversity;AlexanderTodorov,Depart-mentofPsychology,PrincetonUniversity.CorrespondenceconcerningthisarticleshouldbeaddressedtoHillelAviezer,DepartmentofPsychology,PrincetonUniversity,Princeton,NJ08540-1010.E-mail:haviezer@princeton.eduJournalofPersonalityandSocialPsychology©2012AmericanPsychologicalAssociation2012,Vol.103,No.1,20–370022-3514/12/$12.00DOI:10.1037/a002741120 </td></tr><tr><td>faca1c97ac2df9d972c0766a296efcf101aaf969</td><td>Sympathy for the Details: Dense Trajectories and Hybrid +<br/>Classification Architectures for Action Recognition +<br/><b>Computer Vision Group, Xerox Research Center Europe, Meylan, France</b><br/>2Centre de Visi´o per Computador, Universitat Aut`onoma de Barcelona, Bellaterra, Spain +<br/>3German Aerospace Center, Wessling, Germany +</td><td>('1799820', 'Adrien Gaidon', 'adrien gaidon')<br/>('2286630', 'Eleonora Vig', 'eleonora vig')</td><td>{cesar.desouza, adrien.gaidon}@xrce.xerox.com, +<br/>eleonora.vig@dlr.de, antonio@cvc.uab.es +</td></tr><tr><td>fab60b3db164327be8588bce6ce5e45d5b882db6</td><td>Maximum A Posteriori Estimation of Distances +<br/>Between Deep Features in Still-to-Video Face +<br/>Recognition +<br/><b>National Research University Higher School of Economics</b><br/>Laboratory of Algorithms and Technologies for Network Analysis, +<br/>36 Rodionova St., Nizhny Novgorod, Russia +<br/><b>National Research University Higher School of Economics</b><br/>20 Myasnitskaya St., Moscow, Russia +<br/>September 2, 2018 +</td><td>('35153729', 'Andrey V. Savchenko', 'andrey v. savchenko')<br/>('2080292', 'Natalya S. Belova', 'natalya s. belova')</td><td>avsavchenko@hse.ru +<br/>nbelova@hse.ru +</td></tr><tr><td>fad895771260048f58d12158a4d4d6d0623f4158</td><td>Audio-Visual Emotion +<br/>Recognition For Natural +<br/>Human-Robot Interaction +<br/>Dissertation zur Erlangung des akademischen Grades +<br/>Doktor der Ingenieurwissenschaften (Dr.-Ing.) +<br/>vorgelegt von +<br/>an der Technischen Fakultät der Universität Bielefeld +<br/>15. März 2010 +</td><td>('32382494', 'Ahmad Rabie', 'ahmad rabie')</td><td></td></tr><tr><td>fae83b145e5eeda8327de9f19df286edfaf5e60c</td><td>Readings in Technology and Education: Proceedings of ICICTE 2010 +<br/>367 +<br/>TOWARDS AN INTERACTIVE E-LEARNING SYSTEM BASED ON +<br/>EMOTIONS AND AFFECTIVE COGNITION +<br/>Department of Informatics +<br/>Department of Audiovisual Arts +<br/>Department of Informatics +<br/>Konstantinos Ch. Drossos +<br/>Department of Audiovisual Arts +<br/><b>Ionian University</b><br/>Greece +</td><td>('25189167', 'Panagiotis Vlamos', 'panagiotis vlamos')<br/>('2284118', 'Andreas Floros', 'andreas floros')<br/>('1761403', 'Michail N. Giannakos', 'michail n. giannakos')</td><td></td></tr><tr><td>ffea8775fc9c32f573d1251e177cd283b4fe09c9</td><td>Accepted to be Published in Proceedings of the IEEE International Conference on Multimedia and Expo (ICME) 2018, San Diego, USA +<br/>TRANSFORMATION ON COMPUTER–GENERATED FACIAL IMAGE TO AVOID DETECTION +<br/>BY SPOOFING DETECTOR +<br/><b>Graduate University for Advanced Studies, Kanagawa, Japan</b><br/><b>National Institute of Informatics, Tokyo, Japan</b><br/><b>The University of Edinburgh, Edinburgh, UK</b></td><td>('47321045', 'Huy H. Nguyen', 'huy h. nguyen')<br/>('9328269', 'Ngoc-Dung T. Tieu', 'ngoc-dung t. tieu')<br/>('2912817', 'Hoang-Quoc Nguyen-Son', 'hoang-quoc nguyen-son')<br/>('1716857', 'Junichi Yamagishi', 'junichi yamagishi')<br/>('1678602', 'Isao Echizen', 'isao echizen')</td><td>{nhhuy, dungtieu, nshquoc, jyamagishi, iechizen}@nii.ac.jp +</td></tr><tr><td>ff8315c1a0587563510195356c9153729b533c5b</td><td>432 +<br/>Zapping Index:Using Smile to Measure +<br/>Advertisement Zapping Likelihood +</td><td>('1803478', 'Songfan Yang', 'songfan yang')<br/>('1784929', 'Mehran Kafai', 'mehran kafai')<br/>('39776603', 'Le An', 'le an')<br/>('1707159', 'Bir Bhanu', 'bir bhanu')</td><td></td></tr><tr><td>ff44d8938c52cfdca48c80f8e1618bbcbf91cb2a</td><td>Towards Video Captioning with Naming: a +<br/>Novel Dataset and a Multi-Modal Approach +<br/>Dipartimento di Ingegneria “Enzo Ferrari” +<br/>Universit`a degli Studi di Modena e Reggio Emilia +</td><td>('2035969', 'Stefano Pini', 'stefano pini')<br/>('3468983', 'Marcella Cornia', 'marcella cornia')<br/>('1843795', 'Lorenzo Baraldi', 'lorenzo baraldi')<br/>('1741922', 'Rita Cucchiara', 'rita cucchiara')</td><td>{name.surname}@unimore.it +</td></tr><tr><td>fffefc1fb840da63e17428fd5de6e79feb726894</td><td>Fine-Grained Age Estimation in the wild with +<br/>Attention LSTM Networks +</td><td>('47969038', 'Ke Zhang', 'ke zhang')<br/>('49229283', 'Na Liu', 'na liu')<br/>('3451660', 'Xingfang Yuan', 'xingfang yuan')<br/>('46910049', 'Xinyao Guo', 'xinyao guo')<br/>('35038034', 'Ce Gao', 'ce gao')<br/>('2626320', 'Zhenbing Zhao', 'zhenbing zhao')</td><td></td></tr><tr><td>ff398e7b6584d9a692e70c2170b4eecaddd78357</td><td></td><td></td><td></td></tr><tr><td>ffc5a9610df0341369aa75c0331ef021de0a02a9</td><td>Transferred Dimensionality Reduction +<br/>State Key Laboratory on Intelligent Technology and Systems +<br/>Tsinghua National Laboratory for Information Science and Technology (TNList) +<br/><b>Tsinghua University, Beijing 100084, China</b></td><td>('39747687', 'Zheng Wang', 'zheng wang')<br/>('1809614', 'Yangqiu Song', 'yangqiu song')<br/>('1700883', 'Changshui Zhang', 'changshui zhang')</td><td></td></tr><tr><td>ffd81d784549ee51a9b0b7b8aaf20d5581031b74</td><td>Performance Analysis of Retina and DoG +<br/>Filtering Applied to Face Images for Training +<br/>Correlation Filters +<br/>Everardo Santiago Ram(cid:19)(cid:16)rez1, Jos(cid:19)e (cid:19)Angel Gonz(cid:19)alez Fraga1, Omar (cid:19)Alvarez +<br/>1 Facultad de Ciencias, Universidad Aut(cid:19)onoma de Baja California, +<br/>Carretera Transpeninsular Tijuana-Ensenada, N(cid:19)um. 3917, Colonia Playitas, +<br/>Ensenada, Baja California, C.P. 22860 +<br/>{everardo.santiagoramirez,angel_fraga, +<br/>2 Facultad de Ingenier(cid:19)(cid:16)a, Arquitectura y Dise~no, Universidad Aut(cid:19)onoma de Baja +<br/>California, Carretera Transpeninsular Tijuana-Ensenada, N(cid:19)um. 3917, Colonia +<br/>Playitas, Ensenada, Baja California, C.P. 22860 +</td><td>('2973536', 'Sergio Omar Infante Prieto', 'sergio omar infante prieto')</td><td>aomar,everardo.gutierrez}@uabc.edu.mx +<br/>sinfante@uabc.edu.mx +</td></tr><tr><td>ff01bc3f49130d436fca24b987b7e3beedfa404d</td><td>Article +<br/>Fuzzy System-Based Face Detection Robust to +<br/>In-Plane Rotation Based on Symmetrical +<br/>Characteristics of a Face +<br/><b>Division of Electronics and Electrical Engineering, Dongguk University, 30 Pildong-ro 1-gil, Jung-gu</b><br/>Academic Editor: Angel Garrido +<br/>Received: 15 June 2016; Accepted: 29 July 2016; Published: 3 August 2016 +</td><td>('1922686', 'Hyung Gil Hong', 'hyung gil hong')<br/>('2026806', 'Won Oh Lee', 'won oh lee')<br/>('3021526', 'Yeong Gon Kim', 'yeong gon kim')<br/>('4634733', 'Kang Ryoung Park', 'kang ryoung park')</td><td>Seoul 100-715, Korea; hell@dongguk.edu (H.G.H.); 215p8@hanmail.net (W.O.L.); csokyg@dongguk.edu (Y.G.K.); +<br/>yawara18@hotmail.com (K.W.K.); nguyentiendat@dongguk.edu (D.T.N.) +<br/>* Correspondence: parkgr@dongguk.edu; Tel.: +82-10-3111-7022; Fax: +82-2-2277-8735 +</td></tr><tr><td>ff061f7e46a6213d15ac2eb2c49d9d3003612e49</td><td>Morphable Human Face Modelling +<br/>by +<br/>Thesis +<br/>for fulfillment of the Requirements for the Degree of +<br/>Doctor of Philosophy (0190) +<br/>Clayton School of Information Technology +<br/><b>Monash University</b><br/>February, 2008 +</td><td>('1695402', 'Nathan Faggian', 'nathan faggian')<br/>('1695402', 'Nathan Faggian', 'nathan faggian')<br/>('1728337', 'Andrew Paplinski', 'andrew paplinski')<br/>('2696169', 'Jamie Sherrah', 'jamie sherrah')</td><td></td></tr><tr><td>ff1f45bdad41d8b35435098041e009627e60d208</td><td>NAGRANI, ZISSERMAN: FROM BENEDICT CUMBERBATCH TO SHERLOCK HOLMES +<br/>From Benedict Cumberbatch to Sherlock +<br/>Holmes: Character Identification in TV +<br/>series without a Script +<br/>Visual Geometry Group, +<br/>Department of Engineering Science, +<br/><b>University of Oxford, UK</b></td><td>('19263506', 'Arsha Nagrani', 'arsha nagrani')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>arsha@robots.ox.ac.uk/ +<br/>az@robots.ox.ac.uk/ +</td></tr><tr><td>ff60d4601adabe04214c67e12253ea3359f4e082</td><td></td><td></td><td></td></tr><tr><td>ffe4bb47ec15f768e1744bdf530d5796ba56cfc1</td><td>AFIF4: Deep Gender Classification based on +<br/>AdaBoost-based Fusion of Isolated Facial Features and +<br/>Foggy Faces +<br/>aDepartment of Electrical Engineering and Computer Science, Lassonde School of +<br/><b>Engineering, York University, Canada</b><br/><b>bFaculty of Computers and Information, Assiut University, Egypt</b></td><td>('40239027', 'Abdelrahman Abdelhamed', 'abdelrahman abdelhamed')</td><td></td></tr><tr><td>ffc9d6a5f353e5aec3116a10cf685294979c63d9</td><td>Eigenphase-based face recognition: a comparison of phase- +<br/>information extraction methods +<br/>Faculty of Electrical Engineering and Computing, +<br/><b>University of Zagreb, Unska 3, 10 000 Zagreb</b></td><td>('35675021', 'Slobodan Ribarić', 'slobodan ribarić')<br/>('3069572', 'Marijo Maračić', 'marijo maračić')</td><td>E-mail: slobodan.ribaric@fer.hr +</td></tr><tr><td>ff8ef43168b9c8dd467208a0b1b02e223b731254</td><td>BreakingNews: Article Annotation by +<br/>Image and Text Processing +</td><td>('1780343', 'Arnau Ramisa', 'arnau ramisa')<br/>('47242882', 'Fei Yan', 'fei yan')<br/>('1994318', 'Francesc Moreno-Noguer', 'francesc moreno-noguer')<br/>('1712041', 'Krystian Mikolajczyk', 'krystian mikolajczyk')</td><td></td></tr><tr><td>ff9195f99a1a28ced431362f5363c9a5da47a37b</td><td>Journal of Vision (2016) 16(15):28, 1–8 +<br/>Serial dependence in the perception of attractiveness +<br/><b>University of California</b><br/>Berkeley, CA, USA +<br/><b>University of California</b><br/>Berkeley, CA, USA +<br/>David Whitney +<br/><b>University of California</b><br/>Berkeley, CA, USA +<br/><b>Helen Wills Neuroscience Institute, University of</b><br/>California, Berkeley, CA, USA +<br/><b>Vision Science Group, University of California</b><br/>Berkeley, CA, USA +<br/>The perception of attractiveness is essential for choices +<br/>of food, object, and mate preference. Like perception of +<br/>other visual features, perception of attractiveness is +<br/>stable despite constant changes of image properties due +<br/>to factors like occlusion, visual noise, and eye +<br/>movements. Recent results demonstrate that perception +<br/>of low-level stimulus features and even more complex +<br/>attributes like human identity are biased towards recent +<br/>percepts. This effect is often called serial dependence. +<br/>Some recent studies have suggested that serial +<br/>dependence also exists for perceived facial +<br/>attractiveness, though there is also concern that the +<br/>reported effects are due to response bias. Here we used +<br/>an attractiveness-rating task to test the existence of +<br/>serial dependence in perceived facial attractiveness. Our +<br/>results demonstrate that perceived face attractiveness +<br/>was pulled by the attractiveness level of facial images +<br/>encountered up to 6 s prior. This effect was not due to +<br/>response bias and did not rely on the previous motor +<br/>response. This perceptual pull increased as the difference +<br/>in attractiveness between previous and current stimuli +<br/>increased. Our results reconcile previously conflicting +<br/>findings and extend previous work, demonstrating that +<br/>sequential dependence in perception operates across +<br/>different levels of visual analysis, even at the highest +<br/>levels of perceptual interpretation. +<br/>Introduction +<br/>Humans make aesthetic judgments all the time about +<br/>the attractiveness or desirability of objects and scenes. +<br/>Aesthetic judgments are not merely about judging +<br/>works of art; they are constantly involved in our daily +<br/>activity, influencing or determining our choices of food, +<br/>object (Creusen & Schoormans, 2005), and mate +<br/>preference (Rhodes, Simmons, & Peters, 2005). +<br/>Aesthetic judgments are based on perceptual pro- +<br/>cessing (Arnheim, 1954; Livingstone & Hubel, 2002; +<br/>Solso, 1996). These judgments, like other perceptual +<br/>experiences, are thought to be relatively stable in spite +<br/>of fluctuations in the raw visual input we receive due to +<br/>factors like occlusion, visual noise, and eye movements. +<br/>One mechanism that allows the visual system to achieve +<br/>this stability is serial dependence. Recent results have +<br/>revealed that the perception of visual features such as +<br/>orientation (Fischer & Whitney, 2014), numerosity +<br/>(Cicchini, Anobile, & Burr, 2014), and facial identity +<br/>(Liberman, Fischer, & Whitney, 2014) are systemati- +<br/>cally assimilated toward visual input from the recent +<br/>past. This perceptual pull has been distinguished from +<br/>hysteresis in motor responses or decision processes, and +<br/>has been shown to be tuned by the magnitude of the +<br/>difference between previous and current visual inputs +<br/>(Fischer & Whitney, 2014; Liberman, Fischer, & +<br/>Whitney, 2014). +<br/>Is aesthetics perception similarly stable like feature +<br/>perception? Some previous studies have suggested that +<br/>the answer is yes. It has been shown that there is a +<br/>positive correlation between observers’ successive +<br/>attractiveness ratings of facial images (Kondo, Taka- +<br/>hashi, & Watanabe, 2012; Taubert, Van der Burg, & +<br/>Alais, 2016). This suggests that there is an assimilative +<br/>sequential dependence in attractiveness judgments. +<br/>Citation: Xia, Y., Leib, A. Y., & Whitney, D. (2016). Serial dependence in the perception of attractiveness. Journal of Vision, +<br/>16(15):28, 1–8, doi:10.1167/16.15.28. +<br/>doi: 10 .116 7 /1 6. 15 . 28 +<br/>Received July 13, 2016; published December 22, 2016 +<br/>ISSN 1534-7362 +<br/>This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. +</td><td>('27678837', 'Ye Xia', 'ye xia')<br/>('6931574', 'Allison Yamanashi Leib', 'allison yamanashi leib')</td><td></td></tr><tr><td>ffaad0204f4af763e3390a2f6053c0e9875376be</td><td>Article +<br/>Non-Convex Sparse and Low-Rank Based Robust +<br/>Subspace Segmentation for Data Mining +<br/><b>School of Information Science and Technology, Donghua University, Shanghai 200051, China</b><br/><b>City University of Hong Kong, Kowloon 999077, Hong Kong, China</b><br/><b>School of Mathematics and Computer Science, Northeastern State University, Tahlequah, OK 74464, USA</b><br/>Received: 16 June 2017; Accepted: 10 July 2017; Published: 15 July 2017 +</td><td>('1743434', 'Wenlong Cheng', 'wenlong cheng')<br/>('2482149', 'Mingbo Zhao', 'mingbo zhao')<br/>('1691742', 'Naixue Xiong', 'naixue xiong')<br/>('1977592', 'Kwok Tai Chui', 'kwok tai chui')</td><td>cheng.python@gmail.com +<br/>ktchui3-c@my.cityu.edu.hk +<br/>xiongnaixue@gmail.com +<br/>* Correspondence: mbzhao4@gmail.com; Tel.: +86-131-0684-8616 +</td></tr><tr><td>ffcbedb92e76fbab083bb2c57d846a2a96b5ae30</td><td></td><td></td><td></td></tr><tr><td>ff7bc7a6d493e01ec8fa2b889bcaf6349101676e</td><td>Facial expression recognition with spatiotemporal local +<br/>descriptors +<br/>Machine Vision Group, Infotech Oulu and Department of Electrical and +<br/><b>Information Engineering, P. O. Box 4500 FI-90014 University of Oulu, Finland</b></td><td>('1757287', 'Guoying Zhao', 'guoying zhao')<br/>('1714724', 'Matti Pietikäinen', 'matti pietikäinen')</td><td>{gyzhao, mkp}@ee.oulu.fi +</td></tr><tr><td>fffa2943808509fdbd2fc817cc5366752e57664a</td><td>Combined Ordered and Improved Trajectories for Large Scale Human Action +<br/>Recognition +<br/>1Vision & Sensing, HCC Lab, +<br/><b>ESTeM, University of Canberra</b><br/>2IHCC, RSCS, CECS, +<br/><b>Australian National University</b></td><td>('1793720', 'O. V. Ramana Murthy', 'o. v. ramana murthy')<br/>('1717204', 'Roland Goecke', 'roland goecke')</td><td>O.V.RamanaMurthy@ieee.org +<br/>roland.goecke@ieee.org +</td></tr><tr><td>ff46c41e9ea139d499dd349e78d7cc8be19f936c</td><td>International Journal of Modern Engineering Research (IJMER) +<br/>www.ijmer.com Vol.3, Issue.3, May-June. 2013 pp-1339-1342 ISSN: 2249-6645 +<br/>A Novel Method for Movie Character Identification and its +<br/>Facial Expression Recognition +<br/><b>M.Tech, Sri Sunflower College of Engineering and Technology, Lankapalli</b><br/><b>Sri Sunflower College of Engineering and Technology, Lankapalli</b></td><td>('6339174', 'N. Praveen', 'n. praveen')</td><td></td></tr><tr><td>ff5dd6f96e108d8233220cc262bc282229c1a582</td><td>Applications (IJERA) ISSN: 2248-9622 www.ijera.com +<br/>Vol. 2, Issue 6, November- December 2012, pp.708-715 +<br/>Robust Facial Marks Detection Method Using AAM And SURF +<br/><b>B.S. Abdur Rahman University, Chennai-48, India</b><br/><b>B.S. Abdur Rahman University, Chennai-48, India</b><br/> +</td><td>('9401261', 'Ziaul Haque Choudhury', 'ziaul haque choudhury')<br/>('9401261', 'Ziaul Haque Choudhury', 'ziaul haque choudhury')</td><td></td></tr><tr><td>c5468665d98ce7349d38afb620adbf51757ab86f</td><td>Pose-Encoded Spherical Harmonics for Robust Face +<br/>Recognition Using a Single Image +<br/><b>Center for Automation Research, University of Maryland, College Park, MD 20742, USA</b><br/>2 Vision Technologies Lab, Sarnoff Corporation, Princeton, NJ 08873, USA +</td><td>('39265975', 'Zhanfeng Yue', 'zhanfeng yue')<br/>('38480590', 'Wenyi Zhao', 'wenyi zhao')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>c588c89a72f89eed29d42f34bfa5d4cffa530732</td><td>Attributes2Classname: A discriminative model for attribute-based +<br/>unsupervised zero-shot learning +<br/><b>HAVELSAN Inc., 2Bilkent University, 3Hacettepe University</b></td><td>('9424554', 'Berkan Demirel', 'berkan demirel')<br/>('1939006', 'Ramazan Gokberk Cinbis', 'ramazan gokberk cinbis')<br/>('2011587', 'Nazli Ikizler-Cinbis', 'nazli ikizler-cinbis')</td><td>bdemirel@havelsan.com.tr, gcinbis@cs.bilkent.edu.tr, nazli@cs.hacettepe.edu.tr +</td></tr><tr><td>c5d13e42071813a0a9dd809d54268712eba7883f</td><td>Face Recognition Robust to Head Pose Changes Based on the RGB-D Sensor +<br/><b>West Virginia University, Morgantown, WV</b></td><td>('2997432', 'Cesare Ciaccio', 'cesare ciaccio')<br/>('2671284', 'Lingyun Wen', 'lingyun wen')<br/>('1822413', 'Guodong Guo', 'guodong guo')</td><td>cciaccio@mix.wvu.edu, lwen@mix.wvu.edu, guodong.guo@mail.wvu.edu +</td></tr><tr><td>c50d73557be96907f88b59cfbd1ab1b2fd696d41</td><td>JournalofElectronicImaging13(3),474–485(July2004). +<br/>Semiconductor sidewall shape estimation +<br/>Oak Ridge National Laboratory +<br/>Oak Ridge, Tennessee 37831-6010 +</td><td>('3078522', 'Philip R. Bingham', 'philip r. bingham')<br/>('3211433', 'Jeffery R. Price', 'jeffery r. price')<br/>('2019731', 'Kenneth W. Tobin', 'kenneth w. tobin')<br/>('1970334', 'Thomas P. Karnowski', 'thomas p. karnowski')</td><td>E-mail: binghampr@ornl.gov +</td></tr><tr><td>c54f9f33382f9f656ec0e97d3004df614ec56434</td><td></td><td></td><td></td></tr><tr><td>c574c72b5ef1759b7fd41cf19a9dcd67e5473739</td><td>Zlatintsi et al. EURASIP Journal on Image and Video Processing (2017) 2017:54 +<br/>DOI 10.1186/s13640-017-0194-1 +<br/>EURASIP Journal on Image +<br/>and Video Processing +<br/>RESEARCH +<br/>Open Access +<br/>COGNIMUSE: a multimodal video +<br/>database annotated with saliency, events, +<br/>semantics and emotion with application to +<br/>summarization +</td><td>('2641229', 'Athanasia Zlatintsi', 'athanasia zlatintsi')<br/>('27687205', 'Niki Efthymiou', 'niki efthymiou')<br/>('2861393', 'Katerina Pastra', 'katerina pastra')<br/>('1791187', 'Alexandros Potamianos', 'alexandros potamianos')<br/>('1750686', 'Petros Maragos', 'petros maragos')<br/>('2539459', 'Petros Koutras', 'petros koutras')<br/>('1710606', 'Georgios Evangelopoulos', 'georgios evangelopoulos')</td><td></td></tr><tr><td>c5a561c662fc2b195ff80d2655cc5a13a44ffd2d</td><td>Using Language to Learn Structured Appearance +<br/>Models for Image Annotation +</td><td>('37894231', 'Michael Jamieson', 'michael jamieson')<br/>('1775745', 'Afsaneh Fazly', 'afsaneh fazly')<br/>('1792908', 'Suzanne Stevenson', 'suzanne stevenson')<br/>('1724954', 'Sven Wachsmuth', 'sven wachsmuth')</td><td></td></tr><tr><td>c5fe40875358a286594b77fa23285fcfb7bda68e</td><td></td><td></td><td></td></tr><tr><td>c5c379a807e02cab2e57de45699ababe8d13fb6d</td><td> Facial Expression Recognition Using Sparse Representation +<br/>1School of Physics and Electronic Engineering +<br/><b>Taizhou University</b><br/>Taizhou 318000 +<br/>CHINA +<br/> 2Department of Computer Science +<br/><b>Taizhou University</b><br/>Taizhou 318000 +<br/>CHINA +</td><td>('1695589', 'SHIQING ZHANG', 'shiqing zhang')<br/>('1730594', 'XIAOMING ZHAO', 'xiaoming zhao')<br/>('38909691', 'BICHENG LEI', 'bicheng lei')</td><td>tzczsq@163.com, leibicheng@163.com +<br/>tzxyzxm@163.com +</td></tr><tr><td>c5ea084531212284ce3f1ca86a6209f0001de9d1</td><td>Audio-Visual Speech Processing for +<br/>Multimedia Localisation +<br/>by +<br/>Matthew Aaron Benatan +<br/>Submitted in accordance with the requirements +<br/>for the degree of Doctor of Philosophy +<br/><b>The University of Leeds</b><br/>School of Computing +<br/>September 2016 +</td><td></td><td></td></tr><tr><td>c5935b92bd23fd25cae20222c7c2abc9f4caa770</td><td>Spatiotemporal Multiplier Networks for Video Action Recognition +<br/><b>Graz University of Technology</b><br/><b>Graz University of Technology</b><br/><b>York University, Toronto</b></td><td>('2322150', 'Christoph Feichtenhofer', 'christoph feichtenhofer')<br/>('1718587', 'Axel Pinz', 'axel pinz')<br/>('1709096', 'Richard P. Wildes', 'richard p. wildes')</td><td>feichtenhofer@tugraz.at +<br/>axel.pinz@tugraz.at +<br/>wildes@cse.yorku.ca +</td></tr><tr><td>c5421a18583f629b49ca20577022f201692c4f5d</td><td>Facial Age Classification using Subpattern-based +<br/>Approaches +<br/><b>Eastern Mediterranean University, Gazima usa, Northern Cyprus</b><br/>Mersin 10, Turkey +<br/> +<br/>are +<br/>(mPCA) +<br/>examined +</td><td>('3437942', 'Fatemeh Mirzaei', 'fatemeh mirzaei')<br/>('2907423', 'Önsen Toygar', 'önsen toygar')</td><td>{fatemeh.mirzaei, onsen.toygar}@emu.edu.tr +</td></tr><tr><td>c5be0feacec2860982fbbb4404cf98c654142489</td><td>Semi-Qualitative Probabilistic Networks in Computer +<br/>Vision Problems +<br/>Troy, NY 12180, USA. +<br/>Troy, NY 12180, USA. +<br/>Troy, NY 12180, USA. +<br/>Troy, NY 12180, USA. +<br/>Received: *** +<br/>Revised: *** +</td><td>('1680860', 'Cassio P. de Campos', 'cassio p. de campos')<br/>('1684635', 'Lei Zhang', 'lei zhang')<br/>('1686235', 'Yan Tong', 'yan tong')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>Email: decamc@rpi.edu +<br/>Email: zhangl2@rpi.edu +<br/>Email: tongy2@rpi.edu +<br/>Email: jiq@rpi.edu +</td></tr><tr><td>c5844de3fdf5e0069d08e235514863c8ef900eb7</td><td>Lam S K et al. / (IJCSE) International Journal on Computer Science and Engineering +<br/>Vol. 02, No. 08, 2010, 2659-2665 +<br/>A Study on Similarity Computations in Template +<br/>Matching Technique for Identity Verification +<br/>Lam, S. K., Yeong, C. Y., Yew, C. T., Chai, W. S., Suandi, S. A. +<br/>Intelligent Biometric Group, School of Electrical and Electronic Engineering +<br/>Engineering Campus, Universiti Sains Malaysia +<br/>14300 Nibong Tebal, Pulau Pinang, MALAYSIA +</td><td></td><td>Email: shahrel@eng.usm.my +</td></tr><tr><td>c58b7466f2855ffdcff1bebfad6b6a027b8c5ee1</td><td>Ultra-Resolving Face Images by Discriminative +<br/>Generative Networks(cid:63) +<br/><b>Australian National University</b></td><td>('4092561', 'Xin Yu', 'xin yu')</td><td>{xin.yu, fatih.porikli}@anu.edu.au +</td></tr><tr><td>c590c6c171392e9f66aab1bce337470c43b48f39</td><td>Emotion Recognition by Machine Learning Algorithms using +<br/>Psychophysiological Signals +<br/>1, 2, 3 BT Convergence Technology Research Department, Electronics and Telecommunications +<br/><b>Research Institute, 138 Gajeongno, Yuseong-gu, Daejeon, 305-700, Republic of Korea</b><br/><b>Chungnam National University</b></td><td>('2329242', 'Eun-Hye Jang', 'eun-hye jang')<br/>('1696731', 'Byoung-Jun Park', 'byoung-jun park')<br/>('2030031', 'Sang-Hyeob Kim', 'sang-hyeob kim')<br/>('2615387', 'Jin-Hun Sohn', 'jin-hun sohn')</td><td>cleta4u@etri.re.kr, bj_park@etri.re.kr, shk1028@etri.re.kr +<br/>Gung-dong, Yuseong-gu, Daejeon, 305-765, Republic of Korea, jhsohn@cnu.ac.kr +</td></tr><tr><td>c5f1ae9f46dc44624591db3d5e9f90a6a8391111</td><td>Application of non-negative and local non negative matrix factorization to facial +<br/>expression recognition +<br/>Dept. of Informatics +<br/><b>Aristotle University of Thessaloniki</b><br/>GR-541 24, Thessaloniki, Box 451, Greece +</td><td>('2336758', 'Ioan Buciu', 'ioan buciu')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>{nelu,pitas}@zeus.csd.auth.gr +</td></tr><tr><td>c53352a4239568cc915ad968aff51c49924a3072</td><td>Transfer Representation-Learning for Anomaly Detection +<br/>Lewis D. Griffin† +<br/><b>University College London, UK</b><br/><b>University College London, UK</b><br/>(cid:63)Rapiscan Systems Ltd, USA +</td><td>('3451382', 'Thomas Tanay', 'thomas tanay')<br/>('13736095', 'Edward J. Morton', 'edward j. morton')</td><td>JERONE.ANDREWS@CS.UCL.AC.UK +<br/>THOMAS.TANAY.13@UCL.AC.UK +<br/>EMORTON@RAPISCANSYSTEMS.COM +<br/>L.GRIFFIN@CS.UCL.AC.UK +</td></tr><tr><td>c2c5206f6a539b02f5d5a19bdb3a90584f7e6ba4</td><td>Affective Computing: A Review +<br/><b>National Laboratory of Pattern Recognition (NLPR), Institute of Automation</b><br/>Chinese Academy of Sciences, P.O.X. 2728, Beijing 100080 +</td><td>('37670752', 'Jianhua Tao', 'jianhua tao')<br/>('1688870', 'Tieniu Tan', 'tieniu tan')</td><td>{jhtao, tnt}@nlpr.ia.ac.cn +</td></tr><tr><td>c2fa83e8a428c03c74148d91f60468089b80c328</td><td>Optimal Mean Robust Principal Component Analysis +<br/><b>University of Texas at Arlington, Arlington, TX</b></td><td>('1688370', 'Feiping Nie', 'feiping nie')<br/>('40034801', 'Jianjun Yuan', 'jianjun yuan')<br/>('1748032', 'Heng Huang', 'heng huang')</td><td>FEIPINGNIE@GMAIL.COM +<br/>WRIYJJ@GMAIL.COM +<br/>HENG@UTA.EDU +</td></tr><tr><td>c2c3ff1778ed9c33c6e613417832505d33513c55</td><td>Multimodal Biometric Person Authentication +<br/>Using Fingerprint, Face Features +<br/><b>University of Lac Hong 10 Huynh Van Nghe</b><br/>DongNai 71000, Viet Nam +<br/><b>Ho Chi Minh City University of Science</b><br/>227 Nguyen Van Cu, HoChiMinh 70000, Viet Nam +</td><td>('2009230', 'Tran Binh Long', 'tran binh long')<br/>('2710459', 'Le Hoang Thai', 'le hoang thai')<br/>('1971778', 'Tran Hanh', 'tran hanh')</td><td>tblong@lhu.edu.vn +<br/>lhthai@fit.hcmus.edu.vn +</td></tr><tr><td>c27f64eaf48e88758f650e38fa4e043c16580d26</td><td>Title of the proposed research project: Subspace analysis using Locality Preserving +<br/>Projection and its applications for image recognition +<br/>Research area: Data manifold learning for pattern recognition +<br/>Contact Details: +<br/><b>University: Dhirubhai Ambani Institute of Information and Communication Technology</b><br/>(DA-IICT), Gandhinagar. +<br/> +</td><td>('2050838', 'Gitam C Shikkenawis', 'gitam c shikkenawis')</td><td>Email Address: 201221004@daiict.ac.in +</td></tr><tr><td>c23153aade9be0c941390909c5d1aad8924821db</td><td>Efficient and Accurate Tracking +<br/>for Face Diarization via Periodical Detection +<br/>∗Ecole Polytechnique Federal de Lausanne, Switzerland +<br/><b>Idiap Research Institute, Martigny, Switzerland</b></td><td>('39560344', 'Nam Le', 'nam le')<br/>('30790014', 'Alexander Heili', 'alexander heili')<br/>('1719610', 'Jean-Marc Odobez', 'jean-marc odobez')</td><td>Email: { nle, aheili, dwu, odobez }@idiap.ch +</td></tr><tr><td>c207fd762728f3da4cddcfcf8bf19669809ab284</td><td>Face Alignment Using Boosting and Evolutionary +<br/>Search +<br/><b>College of Software Engineering, Southeast University, Nanjing 210096, China</b><br/><b>Lab of Science and Technology, Southeast University, Nanjing 210096, China</b><br/><b>Human Media Interaction, University of Twente, P.O. Box</b><br/>7500 AE Enschede, The Netherlands +</td><td>('39063774', 'Hua Zhang', 'hua zhang')<br/>('2779570', 'Duanduan Liu', 'duanduan liu')<br/>('1688157', 'Mannes Poel', 'mannes poel')<br/>('1745198', 'Anton Nijholt', 'anton nijholt')</td><td>reynzhang@sina.com +<br/>liuduanduan@seu.edu.cn +<br/>{anijholt,mpoel}@cs.utwente.nl +</td></tr><tr><td>c220f457ad0b28886f8b3ef41f012dd0236cd91a</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>Crystal Loss and Quality Pooling for +<br/>Unconstrained Face Verification and Recognition +</td><td>('40497884', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('2068427', 'Ankan Bansal', 'ankan bansal')<br/>('2680836', 'Hongyu Xu', 'hongyu xu')<br/>('2716670', 'Swami Sankaranarayanan', 'swami sankaranarayanan')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>c254b4c0f6d5a5a45680eb3742907ec93c3a222b</td><td>A Fusion-based Gender Recognition Method +<br/>Using Facial Images +</td><td>('24033665', 'Benyamin Ghojogh', 'benyamin ghojogh')<br/>('1779028', 'Saeed Bagheri Shouraki', 'saeed bagheri shouraki')<br/>('1782221', 'Hoda Mohammadzade', 'hoda mohammadzade')<br/>('22395643', 'Ensieh Iranmehr', 'ensieh iranmehr')</td><td></td></tr><tr><td>c2e03efd8c5217188ab685e73cc2e52c54835d1a</td><td>Deep Tree-structured Face: A Unified Representation for Multi-task Facial +<br/>Biometrics +<br/>Department of Electrical Engineering and Computer Science +<br/><b>University of Tennessee, Knoxville</b></td><td>('1691576', 'Rui Guo', 'rui guo')<br/>('9120475', 'Liu Liu', 'liu liu')<br/>('40560485', 'Wei Wang', 'wei wang')<br/>('2885826', 'Ali Taalimi', 'ali taalimi')<br/>('1690083', 'Chi Zhang', 'chi zhang')<br/>('1698645', 'Hairong Qi', 'hairong qi')</td><td>{rguo1, lliu25, wwang34, ataalimi, czhang24, hqi} @utk.edu +</td></tr><tr><td>c28461e266fe0f03c0f9a9525a266aa3050229f0</td><td>Automatic Detection of Facial Feature Points via +<br/>HOGs and Geometric Prior Models +<br/>1 Computer Vision Center , Universitat Aut`onoma de Barcelona +<br/>2 Universitat Oberta de Catalunya +<br/>3 Dept. de Matem`atica Aplicada i An`alisi +<br/>Universitat de Barcelona +</td><td>('1863902', 'David Masip', 'david masip')</td><td>mrojas@cvc.uab.es, dmasipr@uoc.edu, jordi.vitria@ub.edu +</td></tr><tr><td>c29e33fbd078d9a8ab7adbc74b03d4f830714cd0</td><td></td><td></td><td></td></tr><tr><td>c2e6daebb95c9dfc741af67464c98f1039127627</td><td>5-1 +<br/>MVA2013 IAPR International Conference on Machine Vision Applications, May 20-23, 2013, Kyoto, JAPAN +<br/>Efficient Measuring of Facial Action Unit Activation Intensities +<br/>using Active Appearance Models +<br/><b>Computer Vision Group, Friedrich Schiller University of Jena, Germany</b><br/><b>University Hospital Jena, Germany</b></td><td>('1708249', 'Daniel Haase', 'daniel haase')<br/>('8993584', 'Michael Kemmler', 'michael kemmler')<br/>('1814631', 'Orlando Guntinas-Lichius', 'orlando guntinas-lichius')<br/>('1728382', 'Joachim Denzler', 'joachim denzler')</td><td></td></tr><tr><td>f60a85bd35fa85739d712f4c93ea80d31aa7de07</td><td>VisDA: The Visual Domain Adaptation Challenge +<br/><b>Boston University</b><br/><b>EECS, University of California Berkeley</b></td><td>('2960713', 'Xingchao Peng', 'xingchao peng')<br/>('39058756', 'Ben Usman', 'ben usman')<br/>('34836903', 'Neela Kaushik', 'neela kaushik')<br/>('50196944', 'Judy Hoffman', 'judy hoffman')<br/>('2774612', 'Dequan Wang', 'dequan wang')<br/>('2903226', 'Kate Saenko', 'kate saenko')</td><td>xpeng,usmn,nkaushik,saenko@bu.edu, jhoffman,dqwang@eecs.berkeley.edu +</td></tr><tr><td>f6f06be05981689b94809130e251f9e4bf932660</td><td>An Approach to Illumination and Expression Invariant +<br/>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 91 – No.15, April 2014 +<br/>Multiple Classifier Face Recognition +<br/>Dalton Meitei Thounaojam +<br/><b>National Institute of Technology</b><br/>Silchar +<br/>Assam: 788010 +<br/>India +<br/><b>National Institute of Technology</b><br/>Silchar +<br/>Assam: 788010 +<br/>India +<br/>Romesh Laishram +<br/><b>Manipur Institute of Technology</b><br/>Imphal West: 795001 +<br/>India +</td><td></td><td></td></tr><tr><td>f68ed499e9d41f9c3d16d843db75dc12833d988d</td><td></td><td></td><td></td></tr><tr><td>f6742010372210d06e531e7df7df9c01a185e241</td><td>Dimensional Affect and Expression in +<br/>Natural and Mediated Interaction +<br/><b>Ritsumeikan, University</b><br/>Kyoto, Japan +<br/>October, 2007 +</td><td>('1709339', 'Michael J. Lyons', 'michael j. lyons')</td><td>lyons@im.ritsumei.ac.jp +</td></tr><tr><td>f69de2b6770f0a8de6d3ec1a65cb7996b3c99317</td><td>Research Journal of Applied Sciences, Engineering and Technology 8(22): 2265-2271, 2014 +<br/>ISSN: 2040-7459; e-ISSN: 2040-7467 +<br/>© Maxwell Scientific Organization, 2014 +<br/>Submitted: September 13, 2014 +<br/>Accepted: September 20, 2014 +<br/>Published: December 15, 2014 +<br/>Face Recognition System Based on Sparse Codeword Analysis +<br/><b>St.Joseph s College of Engineering, Old Mamallapuram Road, Kamaraj Nagar, Semmencherry, Chennai</b><br/><b>Anna University, Chennai</b><br/>Tamil Nadu 600119, India +</td><td>('2508896', 'P. Geetha', 'p. geetha')<br/>('40574934', 'Vasumathi Narayanan', 'vasumathi narayanan')</td><td></td></tr><tr><td>f6ca29516cce3fa346673a2aec550d8e671929a6</td><td>International Journal of Engineering and Advanced Technology (IJEAT) +<br/>ISSN: 2249 – 8958, Volume-2, Issue-4, April 2013 +<br/>Algorithm for Face Matching Using Normalized +<br/>Cross-Correlation +<br/> +</td><td>('2426695', 'C. Saravanan', 'c. saravanan')<br/>('14289238', 'M. Surender', 'm. surender')</td><td></td></tr><tr><td>f67a73c9dd1e05bfc51219e70536dbb49158f7bc</td><td>Journal of Computer Science 10 (11): 2292-2298, 2014 +<br/>ISSN: 1549-3636 +<br/>© 2014 Nithyashri and Kulanthaivel, This open access article is distributed under a Creative Commons Attribution +<br/>(CC-BY) 3.0 license +<br/>A GAUSSIAN MIXTURE MODEL FOR CLASSIFYING THE +<br/>HUMAN AGE USING DWT AND SAMMON MAP +<br/><b>Sathyabama University, Chennai, India</b><br/>2Department of Electronics Engineering, NITTTR, Chennai, India +<br/>Received 2014-05-08; Revised 2014-05-23; Accepted 2014-11-28 +</td><td>('9513864', 'J. Nithyashri', 'j. nithyashri')<br/>('5014650', 'G. Kulanthaivel', 'g. kulanthaivel')</td><td></td></tr><tr><td>f6c70635241968a6d5fd5e03cde6907022091d64</td><td></td><td></td><td></td></tr><tr><td>f6149fc5b39fa6b33220ccee32a8ee3f6bbcaf4a</td><td>Syn2Real: A New Benchmark for +<br/>Synthetic-to-Real Visual Domain Adaptation +<br/><b>Boston University1, University of Tokyo</b><br/><b>University of California Berkeley</b></td><td>('2960713', 'Xingchao Peng', 'xingchao peng')<br/>('39058756', 'Ben Usman', 'ben usman')<br/>('8915348', 'Kuniaki Saito', 'kuniaki saito')<br/>('34836903', 'Neela Kaushik', 'neela kaushik')<br/>('2903226', 'Kate Saenko', 'kate saenko')</td><td></td></tr><tr><td>f66f3d1e6e33cb9e9b3315d3374cd5f121144213</td><td>The Journal of Neuroscience, October 30, 2013 • 33(44):17435–17443 • 17435 +<br/>Behavioral/Cognitive +<br/>Top-Down Control of Visual Responses to Fear by the +<br/>Amygdala +<br/>1Medical Research Council Cognition and Brain Sciences Unit, Cambridge CB2 7EF, United Kingdom, and 2Wellcome Centre for Imaging Neuroscience, +<br/><b>University College London, London WC1N 3BG, United Kingdom</b><br/>The visual cortex is sensitive to emotional stimuli. This sensitivity is typically assumed to arise when amygdala modulates visual cortex +<br/>via backwards connections. Using human fMRI, we compared dynamic causal connectivity models of sensitivity with fearful faces. This +<br/>model comparison tested whether amygdala modulates distinct cortical areas, depending on dynamic or static face presentation. The +<br/>ventral temporal fusiform face area showed sensitivity to fearful expressions in static faces. However, for dynamic faces, we found fear +<br/>sensitivity in dorsal motion-sensitive areas within hMT⫹/V5 and superior temporal sulcus. The model with the greatest evidence +<br/>included connections modulated by dynamic and static fear from amygdala to dorsal and ventral temporal areas, respectively. According +<br/>to this functional architecture, amygdala could enhance encoding of fearful expression movements from video and the form of fearful +<br/>expressions from static images. The amygdala may therefore optimize visual encoding of socially charged and salient information. +<br/>Introduction +<br/>Emotional images enhance responses in visual areas, an effect +<br/>typically observed in the fusiform gyrus for static fearful faces and +<br/>ascribed to backwards connections from amygdala (Morris et al., +<br/>1998; Vuilleumier and Pourtois, 2007). Although support for +<br/>amygdala influence comes from structural connectivity (Amaral +<br/>and Price, 1984; Catani et al., 2003), functional connectivity +<br/>(Morris et al., 1998; Foley et al., 2012), and path analysis (Lim et +<br/>al., 2009), directed connectivity measures and formal model +<br/>comparison are still needed to show that backwards connections +<br/>from amygdala are more likely than other architectures to gener- +<br/>ate cortical emotion sensitivity. +<br/>Moreover, it is surprising that the putative amygdala feedback +<br/>would enhance fusiform cortex responses. According to the pre- +<br/>vailing view, a face-selective area in fusiform cortex, the fusiform +<br/>face area (FFA), is associated with processing facial identity, +<br/>whereas dorsal temporal regions, particularly in the superior +<br/>temporal sulcus (STS), are associated with processing facial ex- +<br/>pression (Haxby et al., 2000). An alternative position is that fusi- +<br/>form and STS areas both contribute to facial expression +<br/>processing but contribute to encoding structural forms and dy- +<br/>namic features, respectively (Calder and Young, 2005; Calder, +<br/>2011). In this case, static fearful expressions may enhance FFA +<br/>Received July 11, 2013; revised Sept. 7, 2013; accepted Sept. 12, 2013. +<br/>Author contributions: N.F., R.N.H., K.J.F., and A.J.C. designed research; N.F. performed research; N.F. analyzed +<br/>data; N.F., R.N.H., K.J.F., and A.J.C. wrote the paper. +<br/>This work was supported by the United Kingdom Economic and Social Research Council Grant RES-062-23-2925 +<br/>to N.F. and the Medical Research Council Grant MC_US_A060_5PQ50 to A.J.C. and Grant MC_US_A060_0046 to +<br/>R.N.H. We thank Christopher Fox for supplying the dynamic object stimuli and James Rowe and Francesca Carota for +<br/>contributing useful comments. +<br/>The authors declare no competing financial interests. +<br/>DOI:10.1523/JNEUROSCI.2992-13.2013 +<br/>Copyright © 2013 the authors +<br/>0270-6474/13/3317435-09$15.00/0 +<br/>encoding of structural cues associated with emotional expres- +<br/>sion. We therefore characterized the conditions under which +<br/>amygdala mediates fear sensitivity in fusiform cortex, compared +<br/>with dorsal temporal areas (Sabatinelli et al., 2011). +<br/>We asked whether dynamic and static fearful expressions en- +<br/>hance responses in dorsal temporal and ventral fusiform areas, re- +<br/>spectively. One dorsal temporal area, hMT⫹/V5, is sensitive to low +<br/>level and facial motion and may be homologous to the middle tem- +<br/>poral (MT), medial superior temporal (MST), and fundus of the +<br/>super temporal (FST) areas in the macaque (Kolster et al., 2010). +<br/>Another dorsal area, the posterior STS, is responsive generally to +<br/>biological motion (Giese and Poggio, 2003). Compared with dorsal +<br/>areas, the fusiform gyrus shows less sensitivity to facial motion +<br/>(Schultz and Pilz, 2009; Trautmann et al., 2009; Pitcher et al., 2011; +<br/>Foley et al., 2012; Schultz et al., 2012). Despite its association with +<br/>facial identity processing, many studies have shown that FFA con- +<br/>tributes to processing facial expressions (Ganel et al., 2005; Fox et al., +<br/>2009b; Cohen Kadosh et al., 2010; Harris et al., 2012) and may have +<br/>a general role in processing facial form (O’Toole et al., 2002; Calder, +<br/>2011). Sensitivity to static fearful expressions in the FFA may reflect +<br/>this role in processing static form. If so, then dynamic fearful expres- +<br/>sions may evoke fear sensitivity in dorsal temporal areas instead, +<br/>reflecting the role of these areas to processing motion. +<br/>Our fMRI results confirmed our hypothesis that dorsal +<br/>motion-sensitive areas showed fear sensitivity for dynamic facial +<br/>expressions, whereas the FFA showed fear sensitivity for static +<br/>expressions. To explore connectivity mechanisms that mediate +<br/>fear sensitivity, we used dynamic causal modeling (DCM) to ex- +<br/>plore 508 plausible connectivity architectures. Our Bayesian +<br/>model comparison identified the most likely model, which +<br/>showed that dynamic and static fear modulated connections +<br/>from amygdala to dorsal or ventral areas, respectively. Amygdala +<br/>therefore may control how behaviorally relevant information is +<br/>visually coded in a context-sensitive fashion. +</td><td>('3162581', 'Nicholas Furl', 'nicholas furl')<br/>('3162581', 'Nicholas Furl', 'nicholas furl')</td><td>Unit, 15 Chaucer Road, Cambridge, CB2 7EF, United Kingdom. E-mail: nick.furl@mrc-cbu.cam.ac.uk. +</td></tr><tr><td>f6ce34d6e4e445cc2c8a9b8ba624e971dd4144ca</td><td>Cross-label Suppression: A Discriminative and Fast +<br/>Dictionary Learning with Group Regularization +<br/>April 24, 2017 +</td><td>('9293691', 'Xiudong Wang', 'xiudong wang')<br/>('2080215', 'Yuantao Gu', 'yuantao gu')</td><td></td></tr><tr><td>f6abecc1f48f6ec6eede4143af33cc936f14d0d0</td><td></td><td></td><td></td></tr><tr><td>f61d5f2a082c65d5330f21b6f36312cc4fab8a3b</td><td>Multi-Level Variational Autoencoder: +<br/>Learning Disentangled Representations from +<br/>Grouped Observations +<br/>OVAL Group +<br/><b>University of Oxford</b><br/>Machine Intelligence and Perception Group +<br/>Microsoft Research +<br/>Cambridge, UK +</td><td>('3365029', 'Diane Bouchacourt', 'diane bouchacourt')<br/>('2870603', 'Ryota Tomioka', 'ryota tomioka')<br/>('2388416', 'Sebastian Nowozin', 'sebastian nowozin')</td><td>diane@robots.ox.ac.uk +<br/>{ryoto,Sebastian.Nowozin}@microsoft.com +</td></tr><tr><td>f6fa97fbfa07691bc9ff28caf93d0998a767a5c1</td><td>k2-means for fast and accurate large scale clustering +<br/>Computer Vision Lab +<br/>D-ITET +<br/>ETH Zurich +<br/>Computer Vision Lab +<br/>D-ITET +<br/>ETH Zurich +<br/>ESAT, KU Leuven +<br/>D-ITET, ETH Zurich +</td><td>('2794259', 'Eirikur Agustsson', 'eirikur agustsson')<br/>('1732855', 'Radu Timofte', 'radu timofte')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>aeirikur@vision.ee.ethz.ch +<br/>timofter@vision.ee.ethz.ch +<br/>vangool@vision.ee.ethz.ch +</td></tr><tr><td>f6cf2108ec9d0f59124454d88045173aa328bd2e</td><td>Robust user identification based on facial action units +<br/>unaffected by users’ emotions +<br/><b>Aalen University, Germany</b></td><td>('3114281', 'Ricardo Buettner', 'ricardo buettner')</td><td>ricardo.buettner@hs-aalen.de +</td></tr><tr><td>f68f20868a6c46c2150ca70f412dc4b53e6a03c2</td><td>157 +<br/>Differential Evolution to Optimize +<br/>Hidden Markov Models Training: +<br/>Application to Facial Expression +<br/>Recognition +<br/>Ars`ene Simbabawe +<br/><b>MISC Laboratory, Constantine 2 University, Constantine, Algeria</b><br/>The base system in this paper uses Hidden Markov +<br/>Models (HMMs) to model dynamic relationships among +<br/>facial features in facial behavior interpretation and un- +<br/>derstanding field. The input of HMMs is a new set +<br/>of derived features from geometrical distances obtained +<br/>from detected and automatically tracked facial points. +<br/>Numerical data representation which is in the form of +<br/>multi-time series is transformed to a symbolic repre- +<br/>sentation in order to reduce dimensionality, extract the +<br/>most pertinent information and give a meaningful repre- +<br/>sentation to humans. The main problem of the use of +<br/>HMMs is that the training is generally trapped in local +<br/>minima, so we used the Differential Evolution (DE) +<br/>algorithm to offer more diversity and so limit as much as +<br/>possible the occurrence of stagnation. For this reason, +<br/>this paper proposes to enhance HMM learning abilities +<br/>by the use of DE as an optimization tool, instead of the +<br/>classical Baum and Welch algorithm. Obtained results +<br/>are compared against the traditional learning approach +<br/>and significant improvements have been obtained. +<br/>Keywords: facial expressions, occurrence order, Hidden +<br/>Markov Model, Baum-Welch, optimization, differential +<br/>evolution +<br/>1. Introduction +<br/>Analyzing the dynamics of facial features and +<br/>(or) the changes in the appearance of facial fea- +<br/>tures (eyes, eyebrows and mouth) is a very im- +<br/>portant step in facial expression understanding +<br/>and interpretation. Many researchers attempt to +<br/>study the dynamic facial behavior. Timing, du- +<br/>ration, speed and occurrence order of face/body +<br/>actions are crucial parameters related to dy- +<br/>namic behavior (Ekman, & Rosenberg, 2005). +<br/>For instance, facial expression temporal dynam- +<br/>ics are essential for recognition of either full ex- +<br/>pressions (Kotsia & Pitas, 2007; Littlewort & +<br/>al, 2006), or components of expressions such +<br/>as facial Action Units (AUs) (Pantic & Patras, +<br/>2006; Valstar & Pantic, 2007). They are essen- +<br/>tial for categorization of complex psychologi- +<br/>cal states like various types of pain and mood +<br/>(Williams, 2002) and are highly important cues +<br/>for distinguishing posed from spontaneous fa- +<br/>cial expressions (Cohn & Schmidt, 2004; Val- +<br/>star & al, 2006). Timing, duration and speed +<br/>have been analyzed in several studies (Cohn & +<br/>Schmidt, 2004; Valstar & al, 2006; Valstar & al +<br/>2007). However, little attention has been given +<br/>to occurrence order (Valstar & al, 2006; Valstar +<br/>& al 2007). +<br/>Several efforts have been recently reported on +<br/>automatic analysis of facial expression data +<br/>(Zeng & al, 2009; Sandbach & al, 2012; Gunes +<br/>that most recent methods employ probabilistic +<br/>(Hidden Markov Models, Dynamic Bayesian +<br/>Network), statistical (Support Vector Machine), +<br/>and ensemble learning techniques (Gentle- +<br/>-Boost), which seem to be particularly suitable +<br/>for automatic facial expression recognition from +<br/>face image sequences. Because we want to ex- +<br/>HMM (Koelstra & al, 2010; Cohen & al, 2003) +<br/>and DBN (Tong & al, 2007; Tong & al, 2010) +<br/>can be used. +<br/>The presented work in this paper is a part of +<br/>a project which aims to construct “An Optimal +</td><td>('2654160', 'Khadoudja Ghanem', 'khadoudja ghanem')<br/>('1749675', 'Amer Draa', 'amer draa')<br/>('2483552', 'Elvis Vyumvuhore', 'elvis vyumvuhore')</td><td></td></tr><tr><td>f6e00d6430cbbaa64789d826d093f7f3e323b082</td><td>Visual Object Recognition +<br/><b>University of Texas at Austin</b><br/><b>RWTH Aachen University</b><br/>SYNTHESIS LECTURES ON COMPUTER +<br/>VISION # 1 +</td><td>('1794409', 'Kristen Grauman', 'kristen grauman')<br/>('1789756', 'Bastian Leibe', 'bastian leibe')</td><td></td></tr><tr><td>e9a5a38e7da3f0aa5d21499149536199f2e0e1f7</td><td>Article +<br/>A Bayesian Scene-Prior-Based Deep Network Model +<br/>for Face Verification +<br/><b>North China University of Technology</b><br/><b>Curtin University, Perth, WA 6102, Australia</b><br/>† These authors contributed equally to this work. +<br/>Received: 12 May 2018; Accepted: 8 June 2018 ; Published: 11 June 2018 +</td><td>('2104779', 'Huafeng Wang', 'huafeng wang')<br/>('2239474', 'Haixia Pan', 'haixia pan')<br/>('3229158', 'Wenfeng Song', 'wenfeng song')<br/>('1713220', 'Wanquan Liu', 'wanquan liu')<br/>('47311804', 'Ning Song', 'ning song')<br/>('2361868', 'Yuehai Wang', 'yuehai wang')</td><td>Beijing 100144, China; wangyuehai@ncut.edu.cn +<br/>2 Department of Software, Beihang University, Beijing 100191, China; swfbuaa@163.com +<br/>* Correspondence: wanghuafeng@ncut.edu.cn (H.W.); W.Liu@curtin.edu.au (W.L.); zy1621125@buaa.edu.cn +<br/>(N.S.); haixiapan@buaa.edu.cn (H.P.); Tel.: +86-189-1192-4121 (H.W.) +</td></tr><tr><td>e9ed17fd8bf1f3d343198e206a4a7e0561ad7e66</td><td>International Journal of Enhanced Research in Science Technology & Engineering, ISSN: 2319-7463 +<br/>Vol. 3 Issue 1, January-2014, pp: (362-365), Impact Factor: 1.252, Available online at: www.erpublications.com +<br/>Cognitive Learning for Social Robot through +<br/>Facial Expression from Video Input +<br/>1Department of Automation & Robotics, 2Department of Computer Science & Engg. +</td><td>('26944751', 'Neeraj Rai', 'neeraj rai')<br/>('2586264', 'Deepak Rai', 'deepak rai')<br/>('26477055', 'Ajay Kumar Garg', 'ajay kumar garg')</td><td></td></tr><tr><td>e988be047b28ba3b2f1e4cdba3e8c94026139fcf</td><td>Multi-Task Convolutional Neural Network for +<br/>Pose-Invariant Face Recognition +</td><td>('2399004', 'Xi Yin', 'xi yin')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td></td></tr><tr><td>e9d43231a403b4409633594fa6ccc518f035a135</td><td>Deformable Part Models with CNN Features +<br/>Kokkinos1,2 +<br/>1 Ecole Centrale Paris,2 INRIA, 3TTI-Chicago (cid:63) +</td><td>('2381485', 'Stavros Tsogkas', 'stavros tsogkas')<br/>('2776496', 'George Papandreou', 'george papandreou')</td><td></td></tr><tr><td>e90e12e77cab78ba8f8f657db2bf4ae3dabd5166</td><td>Nonconvex Sparse Spectral Clustering by Alternating Direction Method of +<br/>Multipliers and Its Convergence Analysis +<br/><b>National University of Singapore</b><br/><b>Key Laboratory of Machine Perception (MOE), School of EECS, Peking University</b><br/><b>Cooperative Medianet Innovation Center, Shanghai Jiao Tong University</b><br/><b>AI Institute</b></td><td>('33224509', 'Canyi Lu', 'canyi lu')<br/>('33221685', 'Jiashi Feng', 'jiashi feng')<br/>('33383055', 'Zhouchen Lin', 'zhouchen lin')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>canyilu@gmail.com, elefjia@nus.edu.sg, zlin@pku.edu.cn, eleyans@nus.edu.sg +</td></tr><tr><td>e9c008d31da38d9eef67a28d2c77cb7daec941fb</td><td>Noisy Softmax: Improving the Generalization Ability of DCNN via Postponing +<br/>the Early Softmax Saturation +<br/><b>School of Information and Communication Engineering, Beijing University of Posts and Telecommunications</b><br/><b>School of Computer Science, Beijing University of Posts and Telecommunications, Beijing China</b></td><td>('3450321', 'Binghui Chen', 'binghui chen')<br/>('1774956', 'Weihong Deng', 'weihong deng')<br/>('8491162', 'Junping Du', 'junping du')</td><td>chenbinghui@bupt.edu.cn, whdeng@bupt.edu.cn, junpingd@bupt.edu.cn +</td></tr><tr><td>e9e40e588f8e6510fa5537e0c9e083ceed5d07ad</td><td>Fast Face Detection Using Graphics Processor +<br/><b>National Institute of Technology Karnataka</b><br/>Surathkal, India +</td><td>('36598334', 'K.Vinay Kumar', 'k.vinay kumar')</td><td></td></tr><tr><td>e9bb045e702ee38e566ce46cc1312ed25cb59ea7</td><td>Integrating Geometric and Textural Features for +<br/>Facial Emotion Classification using SVM +<br/>Frameworks +<br/>1 Department of Computer Science and Engineering, +<br/><b>Indian Institute of Technology, Roorkee</b><br/>2 Department of Electronics and Electrical Communication Engineering, +<br/><b>Indian Institute of Technology, Kharagpur</b></td><td>('19200118', 'Samyak Datta', 'samyak datta')<br/>('3165117', 'Debashis Sen', 'debashis sen')<br/>('1726184', 'R. Balasubramanian', 'r. balasubramanian')</td><td></td></tr><tr><td>e9fcd15bcb0f65565138dda292e0c71ef25ea8bb</td><td>Repositorio Institucional de la Universidad Autónoma de Madrid +<br/>https://repositorio.uam.es +<br/>Esta es la versión de autor de la comunicación de congreso publicada en: +<br/>This is an author produced version of a paper published in: +<br/>Highlights on Practical Applications of Agents and Multi-Agent Systems: +<br/>International Workshops of PAAMS. Communications in Computer and +<br/>Information Science, Volumen 365. Springer, 2013. 223-230 +<br/>DOI: http://dx.doi.org/10.1007/978-3-642-38061-7_22 +<br/>Copyright: © 2013 Springer-Verlag +<br/>El acceso a la versión del editor puede requerir la suscripción del recurso +<br/>Access to the published version may require subscription +</td><td></td><td></td></tr><tr><td>e9f1cdd9ea95810efed306a338de9e0de25990a0</td><td>FEPS: An Easy-to-Learn Sensory Substitution System to +<br/>Perceive Facial Expressions +<br/>Electrical and Computer Engineering +<br/><b>University of Memphis</b><br/>Memphis, TN 38152, USA +</td><td>('2497319', 'M. Iftekhar Tanveer', 'm. iftekhar tanveer')<br/>('2464507', 'Sreya Ghosh', 'sreya ghosh')<br/>('33019079', 'A.K.M. Mahbubur Rahman', 'a.k.m. mahbubur rahman')<br/>('1828610', 'Mohammed Yeasin', 'mohammed yeasin')</td><td>{mtanveer,aanam,sghosh,arahman,myeasin}@memphis.edu +</td></tr><tr><td>e9363f4368b04aeaa6d6617db0a574844fc59338</td><td>BENCHIP: Benchmarking Intelligence +<br/>Processors +<br/>1ICT CAS,2Cambricon,3Alibaba Infrastructure Service, Alibaba Group +<br/>4IFLYTEK,5JD,6RDA Microelectronics,7AMD +</td><td>('2631042', 'Jinhua Tao', 'jinhua tao')<br/>('1678776', 'Zidong Du', 'zidong du')<br/>('50770616', 'Qi Guo', 'qi guo')<br/>('4304175', 'Huiying Lan', 'huiying lan')<br/>('48571185', 'Lei Zhang', 'lei zhang')<br/>('7523063', 'Shengyuan Zhou', 'shengyuan zhou')<br/>('49046597', 'Cong Liu', 'cong liu')<br/>('49343896', 'Shan Tang', 'shan tang')<br/>('38253244', 'Allen Rush', 'allen rush')<br/>('47482936', 'Willian Chen', 'willian chen')<br/>('39419985', 'Shaoli Liu', 'shaoli liu')<br/>('7377735', 'Yunji Chen', 'yunji chen')<br/>('7934735', 'Tianshi Chen', 'tianshi chen')</td><td></td></tr><tr><td>f1250900074689061196d876f551ba590fc0a064</td><td>Learning to Recognize Actions from Limited Training +<br/>Examples Using a Recurrent Spiking Neural Model +<br/><b>School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA</b><br/>2Intel Labs, Hillsboro, OR, USA 97124 +</td><td>('9352814', 'Priyadarshini Panda', 'priyadarshini panda')<br/>('1753812', 'Narayan Srinivasa', 'narayan srinivasa')</td><td>*Correspondence: narayan.srinivasa@intel.com +</td></tr><tr><td>f1b4583c576d6d8c661b4b2c82bdebf3ba3d7e53</td><td>Faster Than Real-time Facial Alignment: A 3D Spatial Transformer Network +<br/>Approach in Unconstrained Poses +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA +</td><td>('47894545', 'Chenchen Zhu', 'chenchen zhu')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td>cbhagava@andrew.cmu.edu, zcckernel@cmu.edu, kluu@andrew.cmu.edu, msavvid@ri.cmu.edu +</td></tr><tr><td>f16a605abb5857c39a10709bd9f9d14cdaa7918f</td><td>Fast greyscale road sign model matching +<br/>and recognition +<br/>Centre de Visió per Computador +<br/>Edifici O – Campus UAB, 08193 Bellaterra, Barcelona, Catalonia, Spain +</td><td>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('1724155', 'Petia Radeva', 'petia radeva')</td><td>{sescalera,petia}@cvc.uab.es +</td></tr><tr><td>f1aa120fb720f6cfaab13aea4b8379275e6d40a2</td><td>InverseFaceNet: Deep Single-Shot Inverse Face Rendering From A Single Image +<br/><b>Max-Planck-Institute for Informatics</b><br/><b>University of Erlangen-Nuremberg 3 University of Bath</b><br/>Figure 1. Our single-shot deep inverse face renderer InverseFaceNet obtains a high-quality geometry, reflectance and illumination estimate +<br/>from just a single input image. We jointly recover the face pose, shape, expression, reflectance and incident scene illumination. From left to +<br/>right: input photo, our estimated face model, its geometry, and the pointwise Euclidean error compared to Garrido et al. [14]. +</td><td>('3022958', 'Hyeongwoo Kim', 'hyeongwoo kim')<br/>('34105638', 'Justus Thies', 'justus thies')<br/>('1699058', 'Michael Zollhöfer', 'michael zollhöfer')<br/>('1819028', 'Christian Richardt', 'christian richardt')<br/>('1680185', 'Christian Theobalt', 'christian theobalt')<br/>('9102722', 'Ayush Tewari', 'ayush tewari')</td><td></td></tr><tr><td>f1748303cc02424704b3a35595610890229567f9</td><td></td><td></td><td></td></tr><tr><td>f1ba2fe3491c715ded9677862fea966b32ca81f0</td><td>ISSN: 2321-7782 (Online) +<br/>Volume 1, Issue 7, December 2013 +<br/>International Journal of Advance Research in +<br/>Computer Science and Management Studies +<br/>Research Paper +<br/>Available online at: www.ijarcsms.com +<br/>Face Tracking and Recognition in Videos: +<br/>HMM Vs KNN +<br/>Assistant Professor +<br/>Department of Computer Engineering +<br/><b>MIT College of Engineering (Pune University</b><br/>Pune - India +</td><td></td><td></td></tr><tr><td>f1d090fcea63d9f9e835c49352a3cd576ec899c1</td><td>Iosifidis, A., Tefas, A., & Pitas, I. (2015). Single-Hidden Layer Feedforward +<br/>Neual Network Training Using Class Geometric Information. In . J. J. +<br/>Computational Intelligence: International Joint Conference, IJCCI 2014 +<br/>Rome, Italy, October 22-24, 2014 Revised Selected Papers. (Vol. III, pp. +<br/>351-364). (Studies in Computational Intelligence; Vol. 620). Springer. DOI: +<br/>10.1007/978-3-319-26393-9_21 +<br/>Peer reviewed version +<br/>Link to published version (if available): +<br/>10.1007/978-3-319-26393-9_21 +<br/>Link to publication record in Explore Bristol Research +<br/>PDF-document +<br/><b>University of Bristol - Explore Bristol Research</b><br/>General rights +<br/>This document is made available in accordance with publisher policies. Please cite only the published +<br/>version using the reference above. Full terms of use are available: +<br/>http://www.bristol.ac.uk/pure/about/ebr-terms.html +<br/> </td><td>('1685469', 'A. Rosa', 'a. rosa')<br/>('9246794', 'J. M. Cadenas', 'j. m. cadenas')<br/>('2092535', 'A. Dourado', 'a. dourado')<br/>('39545211', 'K. Madani', 'k. madani')</td><td></td></tr><tr><td>f113aed343bcac1021dc3e57ba6cc0647a8f5ce1</td><td>International Journal of Science and Research (IJSR) +<br/>ISSN (Online): 2319-7064 +<br/>Index Copernicus Value (2013): 6.14 | Impact Factor (2014): 5.611 +<br/>A Survey on Mining of Weakly Labeled Web Facial +<br/>Images and Annotation +<br/><b>Pune Institute of Computer Technology, Pune, India</b><br/><b>Pune Institute of Computer Technology, Pune, India</b><br/>the +<br/>the proposed system which +</td><td></td><td></td></tr><tr><td>f19777e37321f79e34462fc4c416bd56772031bf</td><td>International Journal of Scientific & Engineering Research, Volume 3, Issue 6, June-2012 1 +<br/>ISSN 2229-5518 +<br/>Literature Review of Image Compression Algorithm +<br/> Dr. B. Chandrasekhar +<br/>Padmaja.V.K +<br/><b>Jawaharlal Technological University, Anantapur</b></td><td></td><td>email: padmaja_vk@yahoo.co.in email:: drchandrasekhar@gmail.com +</td></tr><tr><td>f19ab817dd1ef64ee94e94689b0daae0f686e849</td><td>TECHNISCHE UNIVERSIT¨AT M ¨UNCHEN +<br/>Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +<br/>Blickrichtungsunabh¨angige Erkennung von +<br/>Personen in Bild- und Tiefendaten +<br/>Andre St¨ormer +<br/>Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +<br/>der Technischen Universit¨at M¨unchen zur Erlangung des akademischen Grades eines +<br/>Doktor-Ingenieurs (Dr.-Ing.) +<br/>genehmigten Dissertation. +<br/>Vorsitzender: +<br/>Univ.-Prof. Dr.-Ing. Thomas Eibert +<br/>Pr¨ufer der Dissertation: +<br/>1. Univ.-Prof. Dr.-Ing. habil. Gerhard Rigoll +<br/>2. Univ.-Prof. Dr.-Ing. Horst-Michael Groß, +<br/>Technische Universit¨at Ilmenau +<br/>Die Dissertation wurde am 16.06.2009 bei der Technischen Universit¨at M¨unchen einge- +<br/>reicht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik am 30.10.2009 +<br/>angenommen. +</td><td></td><td></td></tr><tr><td>e76798bddd0f12ae03de26b7c7743c008d505215</td><td></td><td></td><td></td></tr><tr><td>e7cac91da51b78eb4a28e194d3f599f95742e2a2</td><td>RESEARCH ARTICLE +<br/>Positive Feeling, Negative Meaning: +<br/>Visualizing the Mental Representations of In- +<br/>Group and Out-Group Smiles +<br/><b>Saarland University, Saarbr cken, Germany, 2 Utrecht University, Utrecht, the Netherlands</b><br/><b>Behavioural Science Institute, Radboud University, Nijmegen, the Netherlands</b><br/>☯ These authors contributed equally to this work. +</td><td>('34533048', 'Andrea Paulus', 'andrea paulus')<br/>('40358273', 'Michaela Rohr', 'michaela rohr')<br/>('2365875', 'Ron Dotsch', 'ron dotsch')<br/>('3905267', 'Dirk Wentura', 'dirk wentura')</td><td>* a.paulus@mx.uni-saarland.de +</td></tr><tr><td>e793f8644c94b81b7a0f89395937a7f8ad428a89</td><td>LPM for Action Recognition in Temporally +<br/>Untrimmed Videos +<br/>School of Electrical Engineering and Computer Scinece +<br/><b>University of Ottawa, Ottawa, On, Canada</b></td><td>('36047295', 'Feng Shi', 'feng shi')<br/>('1745632', 'Emil Petriu', 'emil petriu')</td><td>{fshi098, laganier, petriu}@site.uottawa.ca +</td></tr><tr><td>e726174d516605f80ff359e71f68b6e8e6ec6d5d</td><td><b>Institute of Information Science</b><br/><b>Beijing Jiaotong University</b><br/>Beijing, 100044 P.R. China +<br/>A novel Patched Locality Preserving Projections for 3D face recognition was pre- +<br/>sented in this paper. In this paper, we firstly patched each image to get the spatial infor- +<br/>mation, and then Gabor filter was used extract intrinsic discriminative information em- +<br/>bedded in each patch. Finally Locality Preserving Projections, which was improved by +<br/>Principle Components Analysis, was utilized to the corresponding patches to obtain lo- +<br/>cality preserving information. The feature was constructed by connecting all these pro- +<br/>jections. Recognition was achieved by using a Nearest Neighbor classifier finally. The +<br/>novelty of this paper came from: (1) The method was robust to changes in facial expres- +<br/>sions and poses, because Gabor filters promoted their useful properties, such as invari- +<br/>ance to rotation, scale and translations, in feature extraction; (2) The method not only +<br/>preserved spatial information, but also preserved locality information of the correspond- +<br/>ing patches. Experiments demonstrated the efficiency and effectiveness of the new +<br/>method. The experimental results showed that the new algorithm outperformed the other +<br/>popular approaches reported in the literature and achieved a much higher accurate recog- +<br/>nition rate. +<br/>Keywords: 3D face recognition, Gabor filters, locality preserving projections, principle +<br/>components analysis, nearest neighbor +<br/>1. INTRODUCTION +<br/>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 26, 2297-2307 (2010) +<br/>Short Paper__________________________________________________ +<br/>3D Face Recognition Using Patched +<br/>Locality Preserving Projections* +<br/>Face recognition is a very challenging subject. So far, studies in 2D face recognition +<br/>have gained significant development, such as Principal Component Analysis (PCA) [1], +<br/>Linear Discriminant Analysis (LDA) [2], and Independent Component Analysis (ICA) [3] +<br/>and so on. But it still bears limitations mostly due to pose variation, illumination, and +<br/>facial expression. 3D face recognition stood out due to the use of face depth information +<br/>which can overcome such limitations. Recently with the development of 3D acquisition +<br/>system, 3D face recognition has attracted more and more interest and a great deal of re- +<br/>search effort has been devoted to this topic [4-7]. +<br/>Many methods have been proposed for 3D face recognition over the last two dec- +<br/>ades. Beumier et al. [8] proposed two methods of surface matching. Central and lateral +<br/>profiles were compared in the curvature space to achieve recognition. However, the me- +<br/>Received October 19, 2009; revised January 8, 2010; accepted March 5, 2010. +<br/>Communicated by Tyng-Luh Liu. +<br/>* This work was also partially supported by the National Natural Science Foundation of China under Grant No. +<br/>60973060 and the Doctorial Foundation of Ministry of Education of China under Grant No. 200800040008. +<br/>2297 +</td><td>('3282147', 'Xue-Qiao Wang', 'xue-qiao wang')<br/>('2383779', 'Qiu-Qi Ruan', 'qiu-qi ruan')</td><td></td></tr><tr><td>e78394213ae07b682ce40dc600352f674aa4cb05</td><td>Expression-invariant three-dimensional face recognition +<br/>Computer Science Department, +<br/><b>Technion Israel Institute of Technology</b><br/>Haifa 32000, Israel +<br/>One of the hardest problems in face recognition is dealing with facial expressions. Finding an +<br/>expression-invariant representation of the face could be a remedy for this problem. We suggest +<br/>treating faces as deformable surfaces in the context of Riemannian geometry, and propose to ap- +<br/>proximate facial expressions as isometries of the facial surface. This way, we can define geometric +<br/>invariants of a given face under different expressions. One such invariant is constructed by iso- +<br/>metrically embedding the facial surface structure into a low-dimensional flat space. Based on this +<br/>approach, we built an accurate three-dimensional face recognition system that is able to distinguish +<br/>between identical twins under various facial expressions. In this chapter we show how under the +<br/>near-isometric model assumption, the difficult problem of face recognition in the presence of facial +<br/>expressions can be solved in a relatively simple way. +<br/>0.1 Introduction +<br/>It is well-known that some characteristics or behavior patterns of the human body are strictly +<br/>individual and can be observed in two different people with a very low probability – a few such +<br/>examples include the DNA code, fingerprints, structure of retinal veins and iris, individual’s written +<br/>signature or face. The term biometrics refers to a variety of methods that attempt to uniquely +<br/>identify a person according to a set of such features. +<br/>While many of today’s biometric technologies are based on the discoveries of the last century (like +<br/>the DNA, for example), some of them have been exploited from the dawn of the human civilization +<br/>[17]. One of the oldest written testimonies of a biometric technology and the first identity theft +<br/>dates back to biblical times, when Jacob fraudulently used the identity of his twin brother Esau to +<br/>benefit from their father’s blessing. The Genesis book describes a combination of hand scan and +<br/>voice recognition that Isaac used to attempt to verify his son’s identity, without knowing that the +<br/>smooth-skinned Jacob had wrapped his hands in kidskin: +<br/>“And Jacob went near unto Isaac his father; and he felt him, and said, ’The voice is Jacob’s +<br/>voice, but the hands are the hands of Esau’. And he recognized him not, because his hands +<br/>were hairy, as his brother Esau’s hands.” +<br/>The false acceptance which resulted from this very inaccurate biometric test had historical conse- +<br/>quences of unmatched proportions. +<br/>Face recognition is probably the most natural biometric method. The remarkable ability of the +<br/>human vision to recognize faces is widely used for biometric authentication from prehistoric times. +<br/>These days, almost every identification document contains a photograph of its bearer, which allows +<br/>the respective officials to verify a person’s identity by comparing his actual face with the one on the +<br/>photo. +<br/>Unlike many other biometrics, face recognition does not require physical contact with the individ- +<br/>ual (like fingerprint recognition) or taking samples of the body (like DNA-based identification) or the +<br/>individual’s behavior (like signature recognition). For these reasons, face recognition is considered a +<br/>natural, less intimidating, and widely accepted biometric identification method [4, 47], and as such, +<br/>has the potential of becoming the leading biometric technology. The great technological challenge is +<br/>to perform face recognition automatically, by means of computer algorithms that work without any +</td><td>('1731883', 'Alexander M. Bronstein', 'alexander m. bronstein')<br/>('1732570', 'Michael M. Bronstein', 'michael m. bronstein')<br/>('1692832', 'Ron Kimmel', 'ron kimmel')</td><td>Email: alexbron@ieee.org +<br/>bronstein@ieee.org +<br/>ron@cs.technion.ac.il +</td></tr><tr><td>e7b2b0538731adaacb2255235e0a07d5ccf09189</td><td>Learning Deep Representations with +<br/>Probabilistic Knowledge Transfer +<br/><b>Aristotle University of Thessaloniki, Thessaloniki 541 24, Greece</b></td><td>('3200630', 'Nikolaos Passalis', 'nikolaos passalis')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')</td><td>passalis@csd.auth.gr, tefas@aiia.csd.auth.gr +</td></tr><tr><td>e726acda15d41b992b5a41feabd43617fab6dc23</td><td></td><td></td><td></td></tr><tr><td>e74816bc0803460e20edbd30a44ab857b06e288e</td><td>Semi-Automated Annotation of Discrete States +<br/>in Large Video Datasets +<br/>Lex Fridman +<br/><b>Massachusetts Institute of Technology</b><br/><b>Massachusetts Institute of Technology</b></td><td>('1901227', 'Bryan Reimer', 'bryan reimer')</td><td>fridman@mit.edu +<br/>reimer@mit.edu +</td></tr><tr><td>e7b6887cd06d0c1aa4902335f7893d7640aef823</td><td>Modelling of Facial Aging and Kinship: A Survey +</td><td>('34291068', 'Markos Georgopoulos', 'markos georgopoulos')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>e73b9b16adcf4339ff4d6723e61502489c50c2d9</td><td>Informatics Engineering, an International Journal (IEIJ) ,Vol.2, No.1, March 2014 +<br/>AN EFFICIENT FEATURE EXTRACTION METHOD WITH +<br/>LOCAL REGION ZERNIKE MOMENT FOR FACIAL +<br/>RECOGNITION OF IDENTICAL TWINS +<br/>1Department of Electrical,Computer and Biomedical Engineering, Qazvin branch, Islamic +<br/><b>Amirkabir University of Technology, Tehran</b><br/><b>Azad University, Qazvin, Iran</b><br/>Iran +</td><td>('1692435', 'Karim Faez', 'karim faez')</td><td></td></tr><tr><td>cbca355c5467f501d37b919d8b2a17dcb39d3ef9</td><td>CANSIZOGLU, JONES: SUPER-RESOLUTION OF VERY LR FACES FROM VIDEOS +<br/>Super-resolution of Very Low-Resolution +<br/>Faces from Videos +<br/>Esra Ataer-Cansizoglu +<br/><b>Mitsubishi Electric Research Labs</b><br/>(MERL) +<br/>Cambridge, MA, USA +</td><td>('1961683', 'Michael Jones', 'michael jones')</td><td>cansizoglu@merl.com +<br/>mjones@merl.com +</td></tr><tr><td>cbbd13c29d042743f0139f1e044b6bca731886d0</td><td>Not-So-CLEVR: learning same–different relations strains +<br/>feedforward neural networks +<br/>†equal contributions +<br/>Department of Cognitive, Linguistic & Psychological Sciences +<br/><b>Carney Institute for Brain Science</b><br/><b>Brown University, Providence, RI 02912, USA</b></td><td>('5546699', 'Junkyung Kim', 'junkyung kim')</td><td></td></tr><tr><td>cbcf5da9f09b12f53d656446fd43bc6df4b2fa48</td><td>ISSN: 2277-3754 +<br/>ISO 9001:2008 Certified +<br/>International Journal of Engineering and Innovative Technology (IJEIT) +<br/>Volume 2, Issue 6, December 2012 +<br/> Face Recognition using Gray level Co-occurrence +<br/>Matrix and Snap Shot Method of the Eigen Face +<br/><b>Sri Chandrasekharendra Saraswathi Viswa Mahavidyalaya University, Kanchipuram, India</b><br/>M. Madhu, R. Amutha +<br/><b>SSN College of Engineering, Chennai, India</b></td><td></td><td></td></tr><tr><td>cba45a87fc6cf12b3b0b6f57ba1a5282ef7fee7a</td><td>Emotion AI, Real-Time Emotion Detection using CNN +<br/>M.S. Computer Science +<br/><b>Stanford University</b><br/>B.S. Computer Science +<br/><b>Stanford University</b></td><td></td><td>tanner12@stanford.edu +<br/>bakis@stanford.edu +</td></tr><tr><td>cb004e9706f12d1de83b88c209ac948b137caae0</td><td>Face Aging Effect Simulation using Hidden Factor +<br/>Analysis Joint Sparse Representation +</td><td>('1787137', 'Hongyu Yang', 'hongyu yang')<br/>('31454775', 'Di Huang', 'di huang')<br/>('40013375', 'Yunhong Wang', 'yunhong wang')<br/>('46506697', 'Heng Wang', 'heng wang')<br/>('2289713', 'Yuanyan Tang', 'yuanyan tang')</td><td></td></tr><tr><td>cb2917413c9b36c3bb9739bce6c03a1a6eb619b3</td><td>MiCT: Mixed 3D/2D Convolutional Tube for Human Action Recognition +<br/><b>University of Science and Technology of China</b><br/>2Microsoft Research Asia +</td><td>('49455479', 'Yizhou Zhou', 'yizhou zhou')<br/>('48305246', 'Xiaoyan Sun', 'xiaoyan sun')<br/>('2057216', 'Zheng-Jun Zha', 'zheng-jun zha')<br/>('8434337', 'Wenjun Zeng', 'wenjun zeng')</td><td>zyz0205@mail.ustc.edu.cn, zhazj@ustc.edu.cn +<br/>{xysun,wezeng}@microsoft.com +</td></tr><tr><td>cb9092fe74ea6a5b2bb56e9226f1c88f96094388</td><td></td><td></td><td></td></tr><tr><td>cb13e29fb8af6cfca568c6dc523da04d1db1fff5</td><td>Paper accepted to Frontiers in Psychology +<br/>Received: 02 Dec 2017 +<br/>Accepted: 12 June 2018 +<br/>DOI: 10.3389/fpsyg.2018.01128 +<br/>A Survey of Automatic Facial +<br/>Micro-expression Analysis: +<br/>Databases, Methods and Challenges +<br/><b>Multimedia University, Faculty of Engineering, Cyberjaya, 63100 Selangor, Malaysia</b><br/><b>Multimedia University, Faculty of Computing and Informatics, Cyberjaya</b><br/>Selangor, Malaysia +<br/><b>University of Nottingham, School of Psychology, University Park, Nottingham NG</b><br/>2RD, United Kingdom +<br/><b>Multimedia University, Research Institute for Digital Security, Cyberjaya</b><br/>Selangor, Malaysia +<br/><b>Monash University Malaysia, School of Information Technology, Sunway</b><br/>Selangor, Malaysia +<br/>Correspondence*: +</td><td>('2154760', 'Yee-Hui Oh', 'yee-hui oh')<br/>('2339975', 'John See', 'john see')<br/>('35256518', 'Anh Cat Le Ngo', 'anh cat le ngo')<br/>('6633183', 'Raphael C.-W. Phan', 'raphael c.-w. phan')<br/>('34287833', 'Vishnu Monn Baskaran', 'vishnu monn baskaran')<br/>('2339975', 'John See', 'john see')</td><td>johnsee@mmu.edu.my +</td></tr><tr><td>cb08f679f2cb29c7aa972d66fe9e9996c8dfae00</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 13, NO. 9, SEPTEMBER 2014 +<br/>Action Understanding +<br/>with Multiple Classes of Actors +</td><td>('2026123', 'Chenliang Xu', 'chenliang xu')<br/>('2228109', 'Caiming Xiong', 'caiming xiong')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')</td><td></td></tr><tr><td>cb84229e005645e8623a866d3d7956c197f85e11</td><td>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. X, NO. X, MONTH 201X +<br/>Disambiguating Visual Verbs +</td><td>('2921001', 'Spandana Gella', 'spandana gella')<br/>('2505673', 'Frank Keller', 'frank keller')<br/>('1747893', 'Mirella Lapata', 'mirella lapata')</td><td></td></tr><tr><td>cb1b5e8b35609e470ce519303915236b907b13b6</td><td>On the Vulnerability of ECG Verification to Online Presentation Attacks +<br/><b>University of Connecticut</b><br/>Electrical & Computer Engineering +<br/><b>University of Florida</b><br/>Electrical & Computer Engineering +</td><td>('3445153', 'Nima Karimian', 'nima karimian')<br/>('2171076', 'Damon L. Woodard', 'damon l. woodard')<br/>('2925373', 'Domenic Forte', 'domenic forte')</td><td>nima@engr.uconn.edu +<br/>dwoodard, dforte@ece.ufl.edu +</td></tr><tr><td>cbb27980eb04f68d9f10067d3d3c114efa9d0054</td><td>An Attention Model for group-level emotion recognition +<br/><b>Indian Institute of Technology</b><br/>Roorkee +<br/>Roorkee, India +<br/><b>Indian Institute of Technology</b><br/>Roorkee +<br/>Roorkee, India +<br/><b>Indian Institute of Technology</b><br/>Roorkee +<br/>Roorkee, India +<br/>École de Technologie Supérieure +<br/>Montreal, Canada +<br/>École de Technologie Supérieure +<br/>Montreal, Canada +</td><td>('51127375', 'Aarush Gupta', 'aarush gupta')<br/>('51134535', 'Dakshit Agrawal', 'dakshit agrawal')<br/>('51118849', 'Hardik Chauhan', 'hardik chauhan')<br/>('3055538', 'Jose Dolz', 'jose dolz')<br/>('3048367', 'Marco Pedersoli', 'marco pedersoli')</td><td>agupta1@cs.iitr.ac.in +<br/>dagrawal@cs.iitr.ac.in +<br/>haroi.uee2014@iitr.ac.in +<br/>jose.dolz@livia.etsmtl.ca +<br/>Marco.Pedersoli@etsmtl.ca +</td></tr><tr><td>cbe859d151466315a050a6925d54a8d3dbad591f</td><td>GAZE SHIFTS AS DYNAMICAL RANDOM SAMPLING +<br/>Dipartimento di Scienze dell’Informazione +<br/>Universit´a di Milano +<br/>Via Comelico 39/41 +<br/>20135 Milano, Italy +</td><td>('1715361', 'Giuseppe Boccignone', 'giuseppe boccignone')<br/>('3241931', 'Mario Ferraro', 'mario ferraro')</td><td>boccignone@dsi.unimi.it +</td></tr><tr><td>f86ddd6561f522d115614c93520faad122eb3b56</td><td>PACS2016 +<br/>Beyond AlphaGo +<br/>October 27-28, 2016 +<br/>Visual Imagination from Texts +<br/>School of Computer Science and Engineering +<br/><b>Seoul National University</b><br/>Seoul 151-744, Korea +</td><td>('3434480', 'Hanock Kwak', 'hanock kwak')<br/>('1692756', 'Byoung-Tak Zhang', 'byoung-tak zhang')</td><td>Email: (hnkwak, btzhang)@bi.snu.ac.kr +</td></tr><tr><td>f8015e31d1421f6aee5e17fc3907070b8e0a5e59</td><td>April 19, 2016 +<br/>DRAFT +<br/>Towards Usable Multimedia Event Detection +<br/>from Web Videos +<br/>April, 2016 +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>Thesis Committee: +<br/>Alexander G. Hauptmann, Chair +<br/>Submitted in partial fulfillment of the requirements +<br/>for the degree of Doctor of Philosophy. +</td><td>('34692532', 'Zhenzhong Lan', 'zhenzhong lan')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')<br/>('14517812', 'Leonid Sigal', 'leonid sigal')<br/>('34692532', 'Zhenzhong Lan', 'zhenzhong lan')</td><td></td></tr><tr><td>f842b13bd494be1bbc1161dc6df244340b28a47f</td><td>An Improved Face Recognition Technique Based +<br/>on Modular Multi-directional Two-dimensional +<br/>Principle Component Analysis Approach +<br/><b>Hanshan Normal University, Chaozhou, 521041, China</b><br/><b>Hanshan Normal University, Chaozhou, 521041, China</b></td><td>('48477766', 'Xiaoqing Dong', 'xiaoqing dong')<br/>('2747115', 'Hongcai Chen', 'hongcai chen')</td><td>Email: dxqzq110@163.com +<br/>Email: czhschc@126.com +</td></tr><tr><td>f83dd9ff002a40228bbe3427419b272ab9d5c9e4</td><td>Facial Features Matching using a Virtual Structuring Element +<br/>Intelligent Systems Lab Amsterdam, +<br/><b>University of Amsterdam</b><br/>Kruislaan 403, 1098 SJ Amsterdam, The Netherlands +</td><td>('9301018', 'Roberto Valenti', 'roberto valenti')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td></td></tr><tr><td>f8c94afd478821681a1565d463fc305337b02779</td><td> +<br/>www.semargroup.org, +<br/>www.ijsetr.com +<br/> +<br/>ISSN 2319-8885 +<br/>Vol.03,Issue.25 +<br/>September-2014, +<br/>Pages:5079-5085 +<br/>Design and Implementation of Robust Face Recognition System for +<br/>Uncontrolled Pose and Illumination Changes +<br/>2 +</td><td></td><td>1PG Scholar, Dept of ECE, LITAM, JNTUK, Andhrapradesh, India, Email: bhaskar.t60@gmail.com. +<br/>2Assistant Professor, Dept of ECE, LITAM, JNTUK, Andhrapradesh, India, Email: venky999v@gmail.com. +</td></tr><tr><td>f8f2d2910ce8b81cb4bbf84239f9229888158b34</td><td>Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-16) +<br/>A Generative Model for Recognizing +<br/>Mixed Group Activities in Still Images +<br/><b>School of Computer, Beijing Institute of Technology, Beijing, China</b><br/><b>School of Computing and Communications, University of Technology Sydney, Sydney, Australia</b></td><td>('32056779', 'Zheng Zhou', 'zheng zhou')<br/>('1780081', 'Kan Li', 'kan li')<br/>('1706670', 'Xiangjian He', 'xiangjian he')<br/>('3225703', 'Mengmeng Li', 'mengmeng li')</td><td>{zz24, likan}@bit.edu.cn, xiangjian.he@uts.edu.au, limengmeng93@163.com +</td></tr><tr><td>f8ec92f6d009b588ddfbb47a518dd5e73855547d</td><td>J Inf Process Syst, Vol.10, No.3, pp.443~458, September 2014 +<br/> +<br/>ISSN 1976-913X (Print) +<br/>ISSN 2092-805X (Electronic) +<br/>Extreme Learning Machine Ensemble Using +<br/>Bagging for Facial Expression Recognition +</td><td>('32322842', 'Deepak Ghimire', 'deepak ghimire')<br/>('2034182', 'Joonwhoan Lee', 'joonwhoan lee')</td><td></td></tr><tr><td>f869601ae682e6116daebefb77d92e7c5dd2cb15</td><td></td><td></td><td></td></tr><tr><td>f8ddb2cac276812c25021b5b79bf720e97063b1e</td><td>A Comprehensive Empirical Study on Linear Subspace Methods for Facial +<br/>Expression Analysis +<br/><b>Queen Mary, University of London</b><br/>Mile End Road, London E1 4NS +</td><td>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('2803283', 'Peter W. McOwan', 'peter w. mcowan')</td><td>{cfshan, sgg, pmco}@dcs.qmul.ac.uk +</td></tr><tr><td>f8ed5f2c71e1a647a82677df24e70cc46d2f12a8</td><td>International Journal of Scientific & Engineering Research, Volume 2, Issue 12, December-2011 1 +<br/>ISSN 2229-5518 +<br/>Artificial Neural Network Design and Parameter +<br/>Optimization for Facial Expressions Recognition +</td><td></td><td></td></tr><tr><td>f8f872044be2918de442ba26a30336d80d200c42</td><td>IJSRD - International Journal for Scientific Research & Development| Vol. 3, Issue 03, 2015 | ISSN (online): 2321-0613 +<br/>Facial Emotion Recognition Techniques: A Survey +<br/>1,2Department of Computer Science and Engineering +<br/><b>Dr C V Raman Institute of Science and Technology</b><br/>defense +<br/>systems, +<br/>surveillance +</td><td></td><td></td></tr><tr><td>f8a5bc2bd26790d474a1f6cc246b2ba0bcde9464</td><td>ORIGINAL RESEARCH +<br/>published: 19 December 2017 +<br/>doi: 10.3389/fpsyg.2017.02181 +<br/>KDEF-PT: Valence, Emotional +<br/>Intensity, Familiarity and +<br/>Attractiveness Ratings of Angry, +<br/>Neutral, and Happy Faces +<br/>Instituto Universitário de Lisboa (ISCTE-IUL), CIS – IUL, Lisboa, Portugal +<br/>The Karolinska Directed Emotional Faces (KDEF) +<br/>is one of the most widely used +<br/>human facial expressions database. Almost a decade after the original validation study +<br/>(Goeleven et al., 2008), we present subjective rating norms for a sub-set of 210 pictures +<br/>which depict 70 models (half female) each displaying an angry, happy and neutral facial +<br/>expressions. Our main goals were to provide an additional and updated validation +<br/>to this database, using a sample from a different nationality (N = 155 Portuguese +<br/>students, M = 23.73 years old, SD = 7.24) and to extend the number of subjective +<br/>dimensions used to evaluate each image. Specifically, participants reported emotional +<br/>labeling (forced-choice task) and evaluated the emotional intensity and valence of the +<br/>expression, as well as the attractiveness and familiarity of the model (7-points rating +<br/>scales). Overall, results show that happy faces obtained the highest ratings across +<br/>evaluative dimensions and emotion labeling accuracy. Female (vs. male) models were +<br/>perceived as more attractive, familiar and positive. The sex of the model also moderated +<br/>the accuracy of emotional +<br/>labeling and ratings of different facial expressions. Each +<br/>picture of the set was categorized as low, moderate, or high for each dimension. +<br/>Normative data for each stimulus (hits proportion, means, standard deviations, and +<br/>confidence intervals per evaluative dimension) is available as supplementary material +<br/>(available at https://osf.io/fvc4m/). +<br/>Keywords: facial expressions, normative data, subjective ratings, emotion labeling, sex differences +<br/>INTRODUCTION +<br/>The human face conveys important information for social interaction. For example, it is a major +<br/>source for forming first impressions, and to make fast and automatic personality trait inferences +<br/>(for a review, see Zebrowitz, 2017). Indeed, facial expressions have been the most studied non- +<br/>verbal emotional cue (for a review, see Schirmer and Adolphs, 2017). In addition to their physical +<br/>component (i.e., morphological changes in the face such as frowning or opening the mouth), +<br/>emotional facial expressions also have an affective component that conveys information about the +<br/>internal feelings of the person expressing it (for a review, see Calvo and Nummenmaa, 2016). +<br/>Moreover, facial expressions communicate a social message that informs about the behavioral +<br/>intentions of the expresser, which in turn prompt responses in the perceiver such approach and +<br/>avoidance reactions (for a review, see Paulus and Wentura, 2016). +<br/>Edited by: +<br/>Sergio Machado, +<br/><b>Salgado de Oliveira University, Brazil</b><br/>Reviewed by: +<br/>Pietro De Carli, +<br/>Dipartimento di Psicologia dello +<br/>Sviluppo e della Socializzazione, +<br/>Università degli Studi di Padova, Italy +<br/>Sylvie Berthoz, +<br/>Institut National de la Santé et de la +<br/>Recherche Médicale, France +<br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Quantitative Psychology +<br/>and Measurement, +<br/>a section of the journal +<br/>Frontiers in Psychology +<br/>Received: 18 July 2017 +<br/>Accepted: 30 November 2017 +<br/>Published: 19 December 2017 +<br/>Citation: +<br/>Garrido MV and Prada M (2017) +<br/>KDEF-PT: Valence, Emotional +<br/>Intensity, Familiarity +<br/>and Attractiveness Ratings of Angry, +<br/>Neutral, and Happy Faces. +<br/>Front. Psychol. 8:2181. +<br/>doi: 10.3389/fpsyg.2017.02181 +<br/>Frontiers in Psychology | www.frontiersin.org +<br/>December 2017 | Volume 8 | Article 2181 +</td><td>('28239829', 'Margarida V. Garrido', 'margarida v. garrido')<br/>('38831356', 'Marília Prada', 'marília prada')<br/>('28239829', 'Margarida V. Garrido', 'margarida v. garrido')</td><td>margarida.garrido@iscte-iul.pt +</td></tr><tr><td>f87b22e7f0c66225824a99cada71f9b3e66b5742</td><td>Robust Emotion Recognition from Low Quality and Low Bit Rate Video: +<br/>A Deep Learning Approach +<br/><b>Beckman Institute, University of Illinois at Urbana-Champaign</b><br/><b>Texas AandM University</b><br/><b>University of Missouri, Kansas City</b><br/>§ Snap Inc, USA +<br/><b>University of Washington</b></td><td>('50563570', 'Bowen Cheng', 'bowen cheng')<br/>('2969311', 'Zhangyang Wang', 'zhangyang wang')<br/>('4622305', 'Zhaobin Zhang', 'zhaobin zhang')<br/>('49970050', 'Zhu Li', 'zhu li')<br/>('1771885', 'Ding Liu', 'ding liu')<br/>('1706007', 'Jianchao Yang', 'jianchao yang')<br/>('47156875', 'Shuai Huang', 'shuai huang')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>{bcheng9, dingliu2, t-huang1}@illinois.edu +<br/>atlaswang@tamu.edu +<br/>{zzktb@mail., lizhu@}umkc.edu +<br/>jianchao.yang@snap.com +<br/>shuaih@uw.edu +</td></tr><tr><td>cef841f27535c0865278ee9a4bc8ee113b4fb9f3</td><td></td><td></td><td></td></tr><tr><td>ce6d60b69eb95477596535227958109e07c61e1e</td><td>Unconstrained Face Verification Using Fisher Vectors +<br/>Computed From Frontalized Faces +<br/>Center for Automation Research +<br/><b>University of Maryland, College Park, MD</b></td><td>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')<br/>('2716670', 'Swami Sankaranarayanan', 'swami sankaranarayanan')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{pullpull, swamiviv, pvishalm, rama}@umiacs.umd.edu +</td></tr><tr><td>ceb763d6657a07b47e48e8a2956bcfdf2cf10818</td><td>International Journal of Computational Science and Information Technology (IJCSITY) Vol.2, No.1, February 2014 +<br/>AN EFFICIENT FEATURE EXTRACTION METHOD +<br/>WITH PSEUDO-ZERNIKE MOMENT FOR FACIAL +<br/>RECOGNITION OF IDENTICAL TWINS +<br/>1Department of Electrical, Computer and Biomedical Engineering, Qazvin branch, +<br/><b>Amirkabir University of Technology, Tehran</b><br/><b>Islamic Azad University, Qazvin, Iran</b><br/>Iran +</td><td>('13302047', 'Hoda Marouf', 'hoda marouf')<br/>('1692435', 'Karim Faez', 'karim faez')</td><td></td></tr><tr><td>cefd9936e91885ba7af9364d50470f6cb54315a4</td><td>The Journal of Neuroscience, December 8, 2010 • 30(49):16601–16608 • 16601 +<br/>Behavioral/Systems/Cognitive +<br/>Expectation and Surprise Determine Neural Population +<br/>Responses in the Ventral Visual Stream +<br/><b>and 2Center for Cognitive Neuroscience, Duke University, Durham, North Carolina 27708</b><br/><b>Psychology, University of Illinois, Beckman Institute, Urbana-Champaign, Illinois 61801, University of</b><br/>Oxford, Oxford OX1 3UD, United Kingdom +<br/>Visual cortex is traditionally viewed as a hierarchy of neural feature detectors, with neural population responses being driven by +<br/>bottom-up stimulus features. Conversely, “predictive coding” models propose that each stage of the visual hierarchy harbors two +<br/>computationally distinct classes of processing unit: representational units that encode the conditional probability of a stimulus and +<br/>provide predictions to the next lower level; and error units that encode the mismatch between predictions and bottom-up evidence, and +<br/>forward prediction error to the next higher level. Predictive coding therefore suggests that neural population responses in category- +<br/>selective visual regions, like the fusiform face area (FFA), reflect a summation of activity related to prediction (“face expectation”) and +<br/>prediction error (“face surprise”), rather than a homogenous feature detection response. We tested the rival hypotheses of the feature +<br/>detection and predictive coding models by collecting functional magnetic resonance imaging data from the FFA while independently +<br/>varying both stimulus features (faces vs houses) and subjects’ perceptual expectations regarding those features (low vs medium vs high +<br/>face expectation). The effects of stimulus and expectation factors interacted, whereby FFA activity elicited by face and house stimuli was +<br/>indistinguishable under high face expectation and maximally differentiated under low face expectation. Using computational modeling, +<br/>we show that these data can be explained by predictive coding but not by feature detection models, even when the latter are augmented +<br/>with attentional mechanisms. Thus, population responses in the ventral visual stream appear to be determined by feature expectation +<br/>and surprise rather than by stimulus features per se. +<br/>Introduction +<br/>“Predictive coding” models of visual cognition propose that per- +<br/>ceptual inference proceeds as an iterative matching process of +<br/>top-down predictions against bottom-up evidence along the vi- +<br/>sual cortical hierarchy (Mumford, 1992; Rao and Ballard, 1999; +<br/>Lee and Mumford, 2003; Friston, 2005; Spratling, 2008). Specif- +<br/>ically, each stage of the visual cortical hierarchy is thought to +<br/>harbor two computationally distinct classes of processing unit: +<br/>representational units that encode the conditional probability of +<br/>a stimulus (“expectation”) and provide predictions regarding ex- +<br/>pected inputs to the next lower level; and error units that encode +<br/>the mismatch between predictions and bottom-up evidence +<br/>(“surprise”), and forward this prediction error to the next higher +<br/>level, where representations are adjusted to eliminate prediction +<br/>error (Friston, 2005). These assumptions contrast sharply with +<br/>more traditional views that cast visual neurons primarily as fea- +<br/>ture detectors (Hubel and Wiesel, 1965; Riesenhuber and Poggio, +<br/>2000), but explicit empirical tests adjudicating between these ri- +<br/>val conceptions are lacking. +<br/>Received June 1, 2010; revised Sept. 21, 2010; accepted Sept. 28, 2010. +<br/>This work was supported by funds granted by the Cognitive Neurology and Alzheimer’s Disease Center +<br/><b>Northwestern University) to T.E. We thank Vincent De Gardelle for helpful comments on an earlier version of</b><br/>this manuscript. +<br/>DOI:10.1523/JNEUROSCI.2770-10.2010 +<br/>Copyright © 2010 the authors +<br/>0270-6474/10/3016601-08$15.00/0 +<br/>Here, we exploited the fact that the two models make diver- +<br/>gent predictions regarding determinants of neural population +<br/>responses in category-selective visual regions, like the fusiform +<br/>face area (FFA) (Kanwisher et al., 1997). Predictive coding sug- +<br/>gests that FFA population responses should reflect a summation +<br/>of activity related to representational units (“face expectation”) +<br/>and error units (“face surprise”), whereas feature detection mod- +<br/>els suppose the population response to be driven by physical +<br/>stimulus characteristics (“face features”) alone. We adjudicated +<br/>between these hypotheses by acquiring functional magnetic res- +<br/>onance imaging (fMRI) data from the FFA while independently +<br/>varying both stimulus features (faces vs houses) and subjects’ +<br/>perceptual expectations regarding those features (low vs medium +<br/>vs high face expectation) (Fig. 1A,C). Of note, both the feature +<br/>detection and predictive coding views also allow for visual neural +<br/>responses to be scaled by attention. Therefore, the above manip- +<br/>ulations were orthogonal to the task demands (the detection of +<br/>occasional inverted “target” stimuli) (Fig. 1B) to control for po- +<br/>tential differences in attention across the conditions of interest. +<br/>According to predictive coding, FFA activity in this experi- +<br/>ment should vary as an additive function of face expectation +<br/>(high ⬎ low) (Fig. 2A, left) and face surprise (unexpected ⬎ +<br/>expected faces) (Fig. 2A, middle). This would result in an inter- +<br/>action between stimulus and expectation factors (Fig. 2A right +<br/>panel), whereby FFA responses to face and house stimuli should +<br/>be similar under high face expectation, because both of these +<br/>conditions would be associated with activity related to face ex- +</td><td>('1900710', 'Tobias Egner', 'tobias egner')<br/>('2372244', 'Christopher Summerfield', 'christopher summerfield')<br/>('1900710', 'Tobias Egner', 'tobias egner')</td><td>Box 90999, Durham, NC 27708. E-mail: tobias.egner@duke.edu. +</td></tr><tr><td>ce85d953086294d989c09ae5c41af795d098d5b2</td><td>This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. +<br/>Bilinear Analysis for Kernel Selection and +<br/>Nonlinear Feature Extraction +</td><td>('1718245', 'Shu Yang', 'shu yang')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('1720735', 'Chao Zhang', 'chao zhang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>ce5eac297174c17311ee28bda534faaa1d559bae</td><td>Automatic analysis of malaria infected red +<br/>blood cell digitized microscope images +<br/>A dissertation submitted in partial fulfilment +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>of +<br/><b>University College London</b><br/>Department of Computer Science +<br/><b>University College London</b><br/>Supervisor: Prof. Bernard F. Buxton +<br/>February 2016 +</td><td>('2768033', 'Houari Abdallahi', 'houari abdallahi')</td><td></td></tr><tr><td>ce691a37060944c136d2795e10ed7ba751cd8394</td><td></td><td></td><td></td></tr><tr><td>ce3f3088d0c0bf236638014a299a28e492069753</td><td></td><td></td><td></td></tr><tr><td>ceaa5eb51f761b5f84bd88b58c8f484fcd2a22d6</td><td>UC San Diego +<br/>UC San Diego Electronic Theses and Dissertations +<br/>Title +<br/>Inhibitions of ascorbate fatty acid derivatives on three rabbit muscle glycolytic enzymes +<br/>Permalink +<br/>https://escholarship.org/uc/item/8x33n1gj +<br/>Author +<br/>Pham, Duyen-Anh +<br/>Publication Date +<br/>2011-01-01 +<br/>Peer reviewed|Thesis/dissertation +<br/>eScholarship.org +<br/>Powered by the California Digital Library +<br/><b>University of California</b></td><td></td><td></td></tr><tr><td>ce450e4849490924488664b44769b4ca57f1bc1a</td><td>Procedural Generation of Videos to Train Deep Action Recognition Networks +<br/>1Computer Vision Group, NAVER LABS Europe, Meylan, France +<br/>2Centre de Visi´o per Computador, Universitat Aut`onoma de Barcelona, Bellaterra, Spain +<br/><b>Toyota Research Institute, Los Altos, CA, USA</b></td><td>('1799820', 'Adrien Gaidon', 'adrien gaidon')<br/>('3407519', 'Yohann Cabon', 'yohann cabon')</td><td>{cesar.desouza, yohann.cabon}@europe.naverlabs.com, adrien.gaidon@tri.global, antonio@cvc.uab.es +</td></tr><tr><td>ceeb67bf53ffab1395c36f1141b516f893bada27</td><td>Face Alignment by Local Deep Descriptor Regression +<br/><b>University of Maryland</b><br/><b>College Park, MD</b><br/><b>University of Maryland</b><br/><b>College Park, MD</b><br/><b>University of Maryland</b><br/><b>College Park, MD</b><br/><b>Rutgers University</b><br/>New Brunswick, NJ 08901 +</td><td>('40080979', 'Amit Kumar', 'amit kumar')<br/>('26988560', 'Rajeev Ranjan', 'rajeev ranjan')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')</td><td>akumar14@umd.edu +<br/>rranjan1@umd.edu +<br/>rama@umiacs.umd.edu +<br/>vishal.m.patel@rutgers.edu +</td></tr><tr><td>ce032dae834f383125cdd852e7c1bc793d4c3ba3</td><td>Motion Interchange Patterns for Action +<br/>Recognition in Unconstrained Videos +<br/><b>The Weizmann Institute of Science, Israel</b><br/><b>Tel-Aviv University, Israel</b><br/><b>The Open University, Israel</b></td><td>('3294355', 'Orit Kliper-Gross', 'orit kliper-gross')<br/>('2916582', 'Yaron Gurovich', 'yaron gurovich')<br/>('1756099', 'Tal Hassner', 'tal hassner')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td></td></tr><tr><td>ce9e1dfa7705623bb67df3a91052062a0a0ca456</td><td>Deep Feature Interpolation for Image Content Changes +<br/>Kilian Weinberger1 +<br/><b>Cornell University</b><br/><b>George Washington University</b><br/>*Authors contributed equally +</td><td>('3222840', 'Paul Upchurch', 'paul upchurch')<br/>('1791337', 'Kavita Bala', 'kavita bala')</td><td></td></tr><tr><td>ce9a61bcba6decba72f91497085807bface02daf</td><td>Eigen-Harmonics Faces: Face Recognition under Generic Lighting +<br/>1Graduate School, CAS, Beijing, China, 100080 +<br/>2ICT-ISVISION Joint R&D Laboratory for Face Recognition, CAS, Beijing, China, 100080 +<br/>Emails: {lyqing, sgshan, wgao}jdl.ac.cn +</td><td>('2343895', 'Laiyun Qing', 'laiyun qing')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1698902', 'Wen Gao', 'wen gao')</td><td></td></tr><tr><td>cef6cffd7ad15e7fa5632269ef154d32eaf057af</td><td>Emotion Detection Through Facial Feature +<br/>Recognition +<br/>through consistent +</td><td>('4959365', 'James Pao', 'james pao')</td><td>jpao@stanford.edu +</td></tr><tr><td>cebfafea92ed51b74a8d27c730efdacd65572c40</td><td>JANUARY 2006 +<br/>31 +<br/>Matching 2.5D Face Scans to 3D Models +</td><td>('2637547', 'Xiaoguang Lu', 'xiaoguang lu')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')<br/>('2205218', 'Dirk Colbry', 'dirk colbry')</td><td></td></tr><tr><td>ce56be1acffda599dec6cc2af2b35600488846c9</td><td>Inferring Sentiment from Web Images with Joint Inference on Visual and Social +<br/>Cues: A Regulated Matrix Factorization Approach +<br/><b>Arizona State University, Tempe AZ</b><br/><b>IBM Almaden Research Center, San Jose CA</b></td><td>('33513248', 'Yilin Wang', 'yilin wang')</td><td>{ywang370,rao,baoxin.li}@asu.edu yuhenghu@us.ibm.com +</td></tr><tr><td>ce54e891e956d5b502a834ad131616786897dc91</td><td>International Journal of Science and Research (IJSR) +<br/>ISSN (Online): 2319-7064 +<br/>Index Copernicus Value (2013): 6.14 | Impact Factor (2014): 5.611 +<br/>Face Recognition Using LTP Algorithm +<br/>1ECE & KUK +<br/>2Assistant Professor (ECE) +<br/>Volume 4 Issue 12, December 2015 +<br/>Licensed Under Creative Commons Attribution CC BY +<br/>www.ijsr.net +<br/> Variation in luminance: Third main challenge that +<br/>appears in face recognition process is the luminance. Due +<br/>to variation in the luminance the representation get varied +<br/>from the original image. The person with same poses +<br/>expression and seen from same viewpoint can be appear +<br/>very different due to variation in lightening. +</td><td>('1781253', 'Richa Sharma', 'richa sharma')<br/>('1887206', 'Rohit Arora', 'rohit arora')</td><td></td></tr><tr><td>ce6f459462ea9419ca5adcc549d1d10e616c0213</td><td>A Survey on Face Identification Methodologies in +<br/>Videos +<br/>Student, M.Tech CSE ,Department of Computer Science +<br/><b>Engineering, G.H.Raisoni College of Engineering</b><br/>Technology for Women, Nagpur, Maharashtra, India. +</td><td>('2776196', 'Deepti Yadav', 'deepti yadav')</td><td></td></tr><tr><td>ce933821661a0139a329e6c8243e335bfa1022b1</td><td>Temporal Modeling Approaches for Large-scale +<br/>Youtube-8M Video Understanding +<br/><b>Baidu IDL and Tsinghua University</b></td><td>('9921390', 'Fu Li', 'fu li')<br/>('2551285', 'Chuang Gan', 'chuang gan')<br/>('3025977', 'Xiao Liu', 'xiao liu')<br/>('38812373', 'Yunlong Bian', 'yunlong bian')<br/>('1716690', 'Xiang Long', 'xiang long')<br/>('2653177', 'Yandong Li', 'yandong li')<br/>('2027571', 'Zhichao Li', 'zhichao li')<br/>('1743129', 'Jie Zhou', 'jie zhou')<br/>('35247507', 'Shilei Wen', 'shilei wen')</td><td></td></tr><tr><td>e03bda45248b4169e2a20cb9124ae60440cad2de</td><td>Learning a Dictionary of Shape-Components in Visual Cortex: +<br/>Comparison with Neurons, Humans and Machines +<br/>by +<br/>Ing´enieur de l’Ecole Nationale Sup´erieure +<br/>des T´el´ecommunications de Bretagne, 2000 +<br/>and +<br/>MS, Universit´e de Rennes, 2000 +<br/>Submitted to the Department of Brain and Cognitive Sciences +<br/>in partial fulfillment of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>at the +<br/><b>MASSACHUSETTS INSTITUTE OF TECHNOLOGY</b><br/>June 2006 +<br/><b>c(cid:13) Massachusetts Institute of Technology 2006. All rights reserved</b><br/>Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Department of Brain and Cognitive Sciences +<br/>April 24, 2006 +<br/>Certified by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Tomaso Poggio +<br/>Eugene McDermott Professor in the Brain Sciences and Human Behavior +<br/>Thesis Supervisor +<br/>Accepted by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Matt Wilson +<br/>Professor of Neurobiology and +<br/>Chairman, Department Graduate Committee +</td><td>('1981539', 'Thomas Serre', 'thomas serre')</td><td></td></tr><tr><td>e03e86ac61cfac9148b371d75ce81a55e8b332ca</td><td>Unsupervised Learning using Sequential +<br/>Verification for Action Recognition +<br/><b>cid:63)The Robotics Institute, Carnegie Mellon University</b><br/>†Facebook AI Research +</td><td>('1806773', 'Ishan Misra', 'ishan misra')<br/>('1709305', 'Martial Hebert', 'martial hebert')<br/>('1699161', 'C. Lawrence Zitnick', 'c. lawrence zitnick')</td><td></td></tr><tr><td>e0dedb6fc4d370f4399bf7d67e234dc44deb4333</td><td>Supplementary Material: Multi-Task Video Captioning with Video and +<br/>Entailment Generation +<br/>UNC Chapel Hill +<br/>1 Experimental Setup +<br/>1.1 Datasets +<br/>1.1.1 Video Captioning Datasets +<br/>YouTube2Text or MSVD The Microsoft Re- +<br/>search Video Description Corpus (MSVD) or +<br/>YouTube2Text (Chen and Dolan, 2011) is used +<br/>for our primary video captioning experiments. It +<br/>has 1970 YouTube videos in the wild with many +<br/>diverse captions in multiple languages for each +<br/>video. Caption annotations to these videos are +<br/>collected using Amazon Mechanical Turk (AMT). +<br/>All our experiments use only English captions. On +<br/>average, each video has 40 captions, and the over- +<br/>all dataset has about 80, 000 unique video-caption +<br/>pairs. The average clip duration is roughly 10 sec- +<br/>onds. We used the standard split as stated in Venu- +<br/>gopalan et al. (2015), i.e., 1200 videos for training, +<br/>100 videos for validation, and 670 for testing. +<br/>MSR-VTT MSR-VTT is a recent collection of +<br/>10, 000 video clips of 41.2 hours duration (i.e., +<br/>average duration of 15 seconds), which are an- +<br/>notated by AMT workers. It has 200, 000 video +<br/>clip-sentence pairs covering diverse content from +<br/>a commercial video search engine. On average, +<br/>each clip is annotated with 20 natural language +<br/>captions. We used the standard split as provided +<br/>in (Xu et al., 2016), i.e., 6, 513 video clips for +<br/>training, 497 for validation, and 2, 990 for testing. +<br/>M-VAD M-VAD is a movie description dataset +<br/>with 49, 000 video clips collected from 92 movies, +<br/>with the average clip duration being 6 seconds. +<br/>Alignment of descriptions to video clips is done +<br/>through an automatic procedure using Descrip- +<br/>tive Video Service (DVS) provided for the movies. +<br/>Each video clip description has only 1 or 2 sen- +<br/>tences, making most evaluation metrics (except +<br/>paraphrase-based METEOR) infeasible. Again, +<br/>we used the standard train/val/test split as pro- +<br/>vided in Torabi et al. (2015). +<br/>1.1.2 Video Prediction Dataset +<br/>For our unsupervised video representation learn- +<br/>ing task, we use the UCF-101 action videos +<br/>dataset (Soomro et al., 2012), which contains +<br/>13, 320 video clips of 101 action categories and +<br/>with an average clip length of 7.21 seconds each. +<br/>This dataset suits our video captioning task well +<br/>because both contain short video clips of a sin- +<br/>gle action or few actions, and hence using future +<br/>frame prediction on UCF-101 helps learn more ro- +<br/>bust and context-aware video representations for +<br/>our short clip video captioning task. We use the +<br/>standard split of 9, 500 videos for training (we +<br/>don’t need any validation set in our setup because +<br/>we directly tune on the validation set of the video +<br/>captioning task). +<br/>the +<br/>three +<br/>video +<br/>captioning +<br/>1.2 Pre-trained Visual Frame Features +<br/>For +<br/>datasets +<br/>(Youtube2Text, MSR-VTT, M-VAD) and the +<br/>unsupervised video prediction dataset (UCF-101), +<br/>we fix our sampling rate to 3f ps to bring uni- +<br/>formity in the temporal representation of actions +<br/>across all videos. These sampled frames are then +<br/>converted into features using several state-of-the- +<br/>art pre-trained models on ImageNet (Deng et al., +<br/>2009) – VGGNet +<br/>(Simonyan and Zisserman, +<br/>2015), GoogLeNet (Szegedy et al., 2015; Ioffe +<br/>and Szegedy, 2015), and Inception-v4 (Szegedy +<br/>et al., 2016). For VGGNet, we use its f c7 layer +<br/>features with dimension 4096. For GoogLeNet +<br/>and Inception-v4, we use the layer before the fully +<br/>connected layer with dimensions 1024 and 1536, +<br/>respectively. We follow standard preprocessing +<br/>and convert all the natural language descriptions +<br/>to lower case and tokenize the sentences and +<br/>remove punctuations. +</td><td>('10721120', 'Ramakanth Pasunuru', 'ramakanth pasunuru')<br/>('7736730', 'Mohit Bansal', 'mohit bansal')</td><td>{ram, mbansal}@cs.unc.edu +</td></tr><tr><td>e096b11b3988441c0995c13742ad188a80f2b461</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>DeepProposals: Hunting Objects and Actions by Cascading +<br/>Deep Convolutional Layers +<br/>Van Gool +<br/>Received: date / Accepted: date +</td><td>('3060081', 'Amir Ghodrati', 'amir ghodrati')</td><td></td></tr><tr><td>e0638e0628021712ac76e3472663ccc17bd8838c</td><td> VOL. 9, NO. 2, FEBRUARY 2014 ISSN 1819-6608 +<br/>ARPN Journal of Engineering and Applied Sciences +<br/>©2006-2014 Asian Research Publishing Network (ARPN). All rights reserved. +<br/>www.arpnjournals.com +<br/>SIGN LANGUAGE RECOGNITION: STATE OF THE ART +<br/><b>Sharda University, Greater Noida, India</b></td><td>('27105713', 'Ashok K Sahoo', 'ashok k sahoo')<br/>('40867787', 'Gouri Sankar Mishra', 'gouri sankar mishra')<br/>('3017041', 'Kiran Kumar Ravulakollu', 'kiran kumar ravulakollu')</td><td>E-Mail: ashoksahoo2000@yahoo.com +</td></tr><tr><td>e0c081a007435e0c64e208e9918ca727e2c1c44e</td><td></td><td></td><td></td></tr><tr><td>e0d878cc095eaae220ad1f681b33d7d61eb5e425</td><td>Article +<br/>Temporal and Fine-Grained Pedestrian Action +<br/>Recognition on Driving Recorder Database +<br/><b>National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba 305-8560, Japan</b><br/><b>Keio University, Yokohama 223-8522, Japan</b><br/>Received: 5 January 2018; Accepted: 8 February 2018; Published: 20 February 2018 +</td><td>('1730200', 'Hirokatsu Kataoka', 'hirokatsu kataoka')<br/>('1732705', 'Yutaka Satoh', 'yutaka satoh')<br/>('1716469', 'Yoshimitsu Aoki', 'yoshimitsu aoki')<br/>('6881850', 'Shoko Oikawa', 'shoko oikawa')<br/>('1720770', 'Yasuhiro Matsui', 'yasuhiro matsui')</td><td>yu.satou@aist.go.jp +<br/>aoki@elec.keio.ac.jp +<br/>Tokyo Metropolitan University, Tokyo 192-0364, Japan; shoko_o@hotmail.com +<br/>4 National Traffic Safety and Environment Laboratory, Tokyo 182-0012, Japan; ymatsui@ntsel.go.jp +<br/>* Correspondence: hirokatsu.kataoka@aist.go.jp; Tel.: +81-29-861-2267 +</td></tr><tr><td>e00d4e4ba25fff3583b180db078ef962bf7d6824</td><td>Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 20 March 2017 doi:10.20944/preprints201703.0152.v1 +<br/>Article +<br/>Face Verification with Multi-Task and Multi-Scale +<br/>Features Fusion +</td><td>('39198322', 'Xiaojun Lu', 'xiaojun lu')<br/>('39683642', 'Yue Yang', 'yue yang')<br/>('8030754', 'Weilin Zhang', 'weilin zhang')<br/>('36286794', 'Qi Wang', 'qi wang')<br/>('37622915', 'Yang Wang', 'yang wang')</td><td>1 College of Sciences, Northeastern University, Shenyang 110819, China; luxiaojun@mail.neu.edu.cn (X.L.); +<br/>YangY1503@163.com (Y.Y.); wangy_neu@163.com (Y.W.) +<br/>2 New York University Shanghai, 1555 Century Ave, Pudong, Shanghai 200122, China; wz723@nyu.edu +<br/>* Correspondence: wangqimath@mail.neu.edu.cn; Tel.: +86-024-8368-7680 +</td></tr><tr><td>e01bb53b611c679141494f3ffe6f0b91953af658</td><td>FSRNet: End-to-End Learning Face Super-Resolution with Facial Priors +<br/><b>Nanjing University of Science and Technology</b><br/>2Youtu Lab, Tencent +<br/><b>Michigan State University</b><br/><b>University of Adelaide</b><br/>Figure 1: Visual results of different super-resolution methods on scale factor 8. +</td><td>('50579509', 'Yu Chen', 'yu chen')<br/>('49499405', 'Jian Yang', 'jian yang')</td><td></td></tr><tr><td>e0bfcf965b402f3f209f26ae20ee88bc4d0002ab</td><td>AI Thinking for Cloud Education Platform with Personalized Learning +<br/><b>University of Texas at San Antonio</b><br/><b>University of Texas at San Antonio</b><br/><b>University of Texas at San Antonio</b><br/><b>University of Texas at San Antonio</b><br/><b>University of Texas at San Antonio</b></td><td>('2055316', 'Paul Rad', 'paul rad')<br/>('2918902', 'Mehdi Roopaei', 'mehdi roopaei')<br/>('1716725', 'Nicole Beebe', 'nicole beebe')<br/>('9324267', 'Mehdi Shadaram', 'mehdi shadaram')<br/>('1839489', 'Yoris A. Au', 'yoris a. au')</td><td>Paul.rad@utsa.edu +<br/> Mehdi.roopaei@utsa.edu +<br/>Nicole.beebe@utsa.edu +<br/>Mehdi.shadaram@utsa.edu +<br/>Yoris.au@utsa.edu +</td></tr><tr><td>e0939b4518a5ad649ba04194f74f3413c793f28e</td><td>Technical Report +<br/>UCAM-CL-TR-636 +<br/>ISSN 1476-2986 +<br/>Number 636 +<br/>Computer Laboratory +<br/>Mind-reading machines: +<br/>automated inference +<br/>of complex mental states +<br/>July 2005 +<br/>15 JJ Thomson Avenue +<br/>Cambridge CB3 0FD +<br/>United Kingdom +<br/>phone +44 1223 763500 +<br/>http://www.cl.cam.ac.uk/ +</td><td></td><td></td></tr><tr><td>e0ed0e2d189ff73701ec72e167d44df4eb6e864d</td><td>Recognition of static and dynamic facial expressions: a study review +<br/>Estudos de Psicologia, 18(1), janeiro-março/2013, 125-130 +<br/><b>Federal University of Para ba</b></td><td>('39169435', 'Nelson Torro Alves', 'nelson torro alves')</td><td></td></tr><tr><td>e00d391d7943561f5c7b772ab68e2bb6a85e64c4</td><td>Robust continuous clustering +<br/><b>University of Maryland, College Park, MD 20740; and bIntel Labs, Santa Clara, CA</b><br/><b>Edited by David L. Donoho, Stanford University, Stanford, CA, and approved August 7, 2017 (received for review January</b><br/>Clustering is a fundamental procedure in the analysis of scientific +<br/>data. It is used ubiquitously across the sciences. Despite decades +<br/>of research, existing clustering algorithms have limited effective- +<br/>ness in high dimensions and often require tuning parameters for +<br/>different domains and datasets. We present a clustering algo- +<br/>rithm that achieves high accuracy across multiple domains and +<br/>scales efficiently to high dimensions and large datasets. The pre- +<br/>sented algorithm optimizes a smooth continuous objective, which +<br/>is based on robust statistics and allows heavily mixed clusters to +<br/>be untangled. The continuous nature of the objective also allows +<br/>clustering to be integrated as a module in end-to-end feature +<br/>learning pipelines. We demonstrate this by extending the algo- +<br/>rithm to perform joint clustering and dimensionality reduction +<br/>by efficiently optimizing a continuous global objective. The pre- +<br/>sented approach is evaluated on large datasets of faces, hand- +<br/>written digits, objects, newswire articles, sensor readings from +<br/>the Space Shuttle, and protein expression levels. Our method +<br/>achieves high accuracy across all datasets, outperforming the best +<br/>prior algorithm by a factor of 3 in average rank. +<br/>clustering | data analysis | unsupervised learning +<br/>Clustering is one of the fundamental experimental procedures +<br/>in data analysis. It is used in virtually all natural and social +<br/>sciences and has played a central role in biology, astronomy, +<br/>psychology, medicine, and chemistry. Data-clustering algorithms +<br/>have been developed for more than half a century (1). Significant +<br/>advances in the last two decades include spectral clustering (2–4), +<br/>generalizations of classic center-based methods (5, 6), mixture +<br/>models (7, 8), mean shift (9), affinity propagation (10), subspace +<br/>clustering (11–13), nonparametric methods (14, 15), and feature +<br/>selection (16–20). +<br/>Despite these developments, no single algorithm has emerged +<br/>to displace the k-means scheme and its variants (21). This +<br/>is despite the known drawbacks of such center-based meth- +<br/><b>ods, including sensitivity to initialization, limited effectiveness in</b><br/>high-dimensional spaces, and the requirement that the number +<br/>of clusters be set in advance. The endurance of these methods +<br/>is in part due to their simplicity and in part due to difficulties +<br/>associated with some of the new techniques, such as additional +<br/>hyperparameters that need to be tuned, high computational cost, +<br/>and varying effectiveness across domains. Consequently, scien- +<br/>tists who analyze large high-dimensional datasets with unknown +<br/>distribution must maintain and apply multiple different cluster- +<br/>ing algorithms in the hope that one will succeed. Books have +<br/>been written to guide practitioners through the landscape of +<br/>data-clustering techniques (22). +<br/>We present a clustering algorithm that is fast, easy to use, and +<br/>effective in high dimensions. The algorithm optimizes a clear +<br/>continuous objective, using standard numerical methods that +<br/>scale to massive datasets. The number of clusters need not be +<br/>known in advance. +<br/>The operation of the algorithm can be understood by contrast- +<br/>ing it with other popular clustering techniques. In center-based +<br/>algorithms such as k-means (1, 24), a small set of putative cluster +<br/>centers is initialized from the data and then iteratively refined. In +<br/>affinity propagation (10), data points communicate over a graph +<br/>structure to elect a subset of the points as representatives. In the +<br/>presented algorithm, each data point has a dedicated representa- +<br/>tive, initially located at the data point. Over the course of the algo- +<br/>rithm, the representatives move and coalesce into easily separable +<br/>clusters. The progress of the algorithm is visualized in Fig. 1. +<br/>Our formulation is based on recent convex relaxations for clus- +<br/>tering (25, 26). However, our objective is deliberately not convex. +<br/>We use redescending robust estimators that allow even heavily +<br/>mixed clusters to be untangled by optimizing a single contin- +<br/>uous objective. Despite the nonconvexity of the objective, the +<br/>optimization can still be performed using standard linear least- +<br/>squares solvers, which are highly efficient and scalable. Since the +<br/>algorithm expresses clustering as optimization of a continuous +<br/>objective based on robust estimation, we call it robust continu- +<br/>ous clustering (RCC). +<br/>One of the characteristics of the presented formulation is that +<br/>clustering is reduced to optimization of a continuous objective. +<br/>This enables the integration of clustering in end-to-end fea- +<br/>ture learning pipelines. We demonstrate this by extending RCC +<br/>to perform joint clustering and dimensionality reduction. The +<br/>extended algorithm, called RCC-DR, learns an embedding of +<br/>the data into a low-dimensional space in which it is clustered. +<br/>Embedding and clustering are performed jointly, by an algorithm +<br/>that optimizes a clear global objective. +<br/>We evaluate RCC and RCC-DR on a large number of datasets +<br/>from a variety of domains. These include image datasets, docu- +<br/>ment datasets, a dataset of sensor readings from the Space Shut- +<br/>tle, and a dataset of protein expression levels in mice. Exper- +<br/>iments demonstrate that our method significantly outperforms +<br/>prior state-of-the-art techniques. RCC-DR is particularly robust +<br/>across datasets from different domains, outperforming the best +<br/>prior algorithm by a factor of 3 in average rank. +<br/>Formulation +<br/>We consider the problem of clustering a set of n data points. +<br/>The input is denoted by X = [x1, x2, . . . , xn ], where xi ∈ RD. +<br/>Our approach operates on a set of representatives U = +<br/>[u1, u2, . . . , un ], where ui ∈ RD. The representatives U are ini- +<br/>tialized at the corresponding data points X. The optimization +<br/>operates on the representation U, which coalesces to reveal the +<br/>cluster structure latent in the data. Thus, the number of clusters +<br/>Significance +<br/>Clustering is a fundamental experimental procedure in data +<br/>analysis. It is used in virtually all natural and social sciences +<br/>and has played a central role in biology, astronomy, psychol- +<br/>ogy, medicine, and chemistry. Despite the importance and +<br/>ubiquity of clustering, existing algorithms suffer from a vari- +<br/>ety of drawbacks and no universal solution has emerged. We +<br/>present a clustering algorithm that reliably achieves high accu- +<br/>racy across domains, handles high data dimensionality, and +<br/>scales to large datasets. The algorithm optimizes a smooth +<br/>global objective, using efficient numerical methods. Experi- +<br/>ments demonstrate that our method outperforms state-of- +<br/>the-art clustering algorithms by significant factors in multiple +<br/>domains. +<br/>Author contributions: S.A.S. and V.K. designed research, performed research, analyzed +<br/>data, and wrote the paper. +<br/>The authors declare no conflict of interest. +<br/>This article is a PNAS Direct Submission. +<br/>Freely available online through the PNAS open access option. +<br/>This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. +<br/>1073/pnas.1700770114/-/DCSupplemental. +<br/>9814–9819 | PNAS | September 12, 2017 | vol. 114 | no. 37 +<br/>www.pnas.org/cgi/doi/10.1073/pnas.1700770114 +</td><td>('49485254', 'Sohil Atul Shah', 'sohil atul shah')<br/>('1770944', 'Vladlen Koltun', 'vladlen koltun')</td><td>1To whom correspondence should be addressed. Email: sohilas@umd.edu. +</td></tr><tr><td>e0765de5cabe7e287582532456d7f4815acd74c1</td><td></td><td></td><td></td></tr><tr><td>e065a2cb4534492ccf46d0afc81b9ad8b420c5ec</td><td>SFace: An Efficient Network for Face Detection +<br/>in Large Scale Variations +<br/><b>College of Software, Beihang University</b><br/>Megvii Inc. (Face++)† +</td><td>('38504661', 'Jianfeng Wang', 'jianfeng wang')<br/>('48009795', 'Ye Yuan', 'ye yuan')<br/>('2789329', 'Boxun Li', 'boxun li')<br/>('2352391', 'Gang Yu', 'gang yu')<br/>('2017810', 'Sun Jian', 'sun jian')</td><td>{wjfwzzc}@buaa.edu.cn, {yuanye,liboxun,yugang,sunjian}@megvii.com +</td></tr><tr><td>e00241f00fb31c660df6c6f129ca38370e6eadb3</td><td>What have we learned from deep representations for action recognition? +<br/>TU Graz +<br/>TU Graz +<br/><b>York University, Toronto</b><br/><b>University of Oxford</b></td><td>('2322150', 'Christoph Feichtenhofer', 'christoph feichtenhofer')<br/>('1718587', 'Axel Pinz', 'axel pinz')<br/>('1709096', 'Richard P. Wildes', 'richard p. wildes')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td>feichtenhofer@tugraz.at +<br/>axel.pinz@tugraz.at +<br/>wildes@cse.yorku.ca +<br/>az@robots.ox.ac.uk +</td></tr><tr><td>e013c650c7c6b480a1b692bedb663947cd9d260f</td><td>860 +<br/>Robust Image Analysis With Sparse Representation +<br/>on Quantized Visual Features +</td><td>('8180253', 'Bing-Kun Bao', 'bing-kun bao')<br/>('36601906', 'Guangyu Zhu', 'guangyu zhu')<br/>('38203359', 'Jialie Shen', 'jialie shen')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>e0244a8356b57a5721c101ead351924bcfb2eef4</td><td>Journal of Experimental Psychology: General +<br/>2017, Vol. 146, No. 10, 1379 –1401 +<br/>0096-3445/17/$12.00 +<br/>© 2017 American Psychological Association +<br/>http://dx.doi.org/10.1037/xge0000292 +<br/>Power as an Emotional Liability: Implications for Perceived Authenticity +<br/>and Trust After a Transgression +<br/><b>University of Southern California</b><br/><b>Webster University</b><br/><b>University of Haifa</b><br/>Alexandra Mislin +<br/><b>American University</b><br/><b>University of Washington, Seattle</b><br/>Gerben A. van Kleef +<br/><b>University of Amsterdam</b><br/>People may express a variety of emotions after committing a transgression. Through 6 empirical studies and +<br/>a meta-analysis, we investigate how the perceived authenticity of such emotional displays and resulting levels +<br/>of trust are shaped by the transgressor’s power. Past findings suggest that individuals with power tend to be +<br/>more authentic because they have more freedom to act on the basis of their own personal inclinations. Yet, +<br/>our findings reveal that (a) a transgressor’s display of emotion is perceived to be less authentic when that +<br/>party’s power is high rather than low; (b) this perception of emotional authenticity, in turn, directly influences +<br/>(and mediates) the level of trust in that party; and (c) perceivers ultimately exert less effort when asked to make +<br/>a case for leniency toward high rather than low-power transgressors. This tendency to discount the emotional +<br/>authenticity of the powerful was found to arise from power increasing the transgressor’s perceived level of +<br/>emotional control and strategic motivation, rather than a host of alternative mechanisms. These results were +<br/>also found across different types of emotions (sadness, anger, fear, happiness, and neutral), expressive +<br/>modalities, operationalizations of the transgression, and participant populations. Altogether, our findings +<br/>demonstrate that besides the wealth of benefits power can afford, it also comes with a notable downside. The +<br/>findings, furthermore, extend past research on perceived emotional authenticity, which has focused on how +<br/>and when specific emotions are expressed, by revealing how this perception can depend on considerations that +<br/>have nothing to do with the expression itself. +<br/>Keywords: trust, emotion, power, authenticity, perception +<br/>Supplemental materials: http://dx.doi.org/10.1037/xge0000292.supp +<br/>Research suggests that those who attain positions of power tend +<br/>to be more emotionally skilled (Côté, Lopes, Salovey, & Miners, +<br/>2010; George, 2000). Indeed, it is the very possession of such +<br/>skills that has been suggested to help these parties attain and +<br/>succeed in leadership positions (e.g., Lewis, 2000; Rubin, Munz, +<br/><b>School of Business, University of Southern California; Alexandra Mislin</b><br/>Department of Management, Kogod School of Business, American Uni- +<br/><b>chael G. Foster School of Business, University of Washington, Seattle</b><br/><b>A. van Kleef, University of Amsterdam</b><br/>This research was supported in part by a faculty research grant from +<br/><b>Webster University</b><br/>Correspondence concerning this article should be addressed to Peter H. +<br/>Kim, Marshall School of Business, Department of Management and Or- +<br/><b>ganization, University of Southern California, Hoffman Hall 515, Los</b><br/>1379 +<br/>& Bommer, 2005). Yet, this tendency for the powerful to be +<br/>emotionally skilled may not necessarily prove beneficial, to the +<br/>extent that those evaluating such powerful individuals subscribe to +<br/>this notion as well, and may even undermine the effectiveness of +<br/>high-power parties’ emotional expressions when they might need +<br/>them most. In particular, through six empirical studies and a +<br/>meta-analysis, we investigate the possibility that perceivers’ gen- +<br/>eral beliefs about the powerful as emotionally skilled would lead +<br/>perceivers to discount the authenticity of the emotions the power- +<br/>ful express, and that this would ultimately impair the effectiveness +<br/>of those emotional displays for addressing a transgression. +<br/>Theoretical Background +<br/>Power, which has been defined as an individual’s capacity to +<br/>modify others’ states by providing or withholding resources or +<br/>administering punishments (Keltner, Gruenfeld, & Anderson, +<br/>2003), has been widely recognized to offer numerous benefits to +<br/><b>those who possess it, including the ability to act based on one s</b><br/>own inclinations, perceive greater choice, and obtain greater ben- +<br/>efits from both work and nonwork interactions (e.g., Galinsky, +</td><td>('34770901', 'Peter H. Kim', 'peter h. kim')<br/>('47847686', 'Ece Tuncel', 'ece tuncel')<br/>('3198839', 'Arik Cheshin', 'arik cheshin')<br/>('50222018', 'Ryan Fehr', 'ryan fehr')<br/>('34770901', 'Peter H. Kim', 'peter h. kim')<br/>('47847686', 'Ece Tuncel', 'ece tuncel')<br/>('50222018', 'Ryan Fehr', 'ryan fehr')<br/>('3198839', 'Arik Cheshin', 'arik cheshin')</td><td>Angeles, CA 90089-1421. E-mail: kimpeter@usc.edu +</td></tr><tr><td>e0dc6f1b740479098c1d397a7bc0962991b5e294</td><td>快速人脸检测技术综述 +<br/>李月敏 1 陈杰 2 高文 1,2,3 尹宝才 1 +<br/>1(北京工业大学计算机学院多媒体与智能软件技术实验室 北京 100022) +<br/>2(哈尔滨工业大学计算机科学与技术学院 哈尔滨 150001) +<br/>3(中国科学院计算技术研究所先进人机通信技术联合实验室 北京 100080) +<br/>摘 要 人脸检测问题研究具有很重要的意义，可以应用到人脸识别、新一代的人机界 +<br/>面、安全访问和视觉监控以及基于内容的检索等领域，近年来受到研究者的普遍重视。人脸 +<br/>检测要走向实际应用，精度和速度是亟需解决的两个关键问题。经过 20 世纪 90 年代以来十 +<br/>多年的发展，人脸检测的精度得到了大幅度的提高，但是速度却一直是阻挠人脸检测走向实 +<br/>用的绊脚石。为此研究者们也作了艰辛的努力。直到 21 世纪 Viola 基于 AdaBoost 算法的人 +<br/>脸检测器的发表，人脸检测的速度才得到了实质性的提高。该算法的发表也促进了人脸检测 +<br/>研究的进一步蓬勃发展，在这方面先后涌现出了一批优秀的文献。基于此，本文在系统地整 +<br/>理分析了人脸检测领域内的相关文献之后，从速度的角度将人脸检测的各种算法大致划分为 +<br/>初始期，发展期，转折点和综合期等四类，并在此基础上进行了全新的总结和论述，最后给 +<br/>出了人脸检测研究的一些可能的发展方向。 +<br/>关键词 人脸检测，速度，人脸识别，模式识别，Boosting +<br/>图法分类号:TP391.4 +<br/>Face Detection: a Survey +<br/>1(Multimedia and Intelligent Software Technology Laboratory +<br/><b>Beijing University of Technology, Beijing 100022, China</b><br/><b>School of Computer Science and Technology, Harbin Institute of</b><br/>Technology, Harbin, 150001, China) +<br/><b>Institute of Computing Technology, Chinese Academy of Sciences</b><br/>Beijing, 100080, China) +</td><td>('7771395', 'Yuemin Li', 'yuemin li')<br/>('1714354', 'Baocai Yin', 'baocai yin')</td><td>ymli@jdl.ac.cn, chenjie@jdl.ac.cn, +<br/>wgao@jdl.ac.cn, ybc@bjut.edu.cn +</td></tr><tr><td>468c8f09d2ad8b558b65d11ec5ad49208c4da2f2</td><td>MSR-CNN: Applying Motion Salient Region Based +<br/>Descriptors for Action Recognition +<br/>School of Computing, Informatics, +<br/>Decision System Engineering +<br/><b>Arizona State University</b><br/>Tempe, USA +<br/>Intel Corp. +<br/>Tempe, USA +<br/>School of Computing, Informatics, +<br/>Decision System Engineering +<br/><b>Arizona State University</b><br/>Tempe, USA +</td><td>('3334478', 'Zhigang Tu', 'zhigang tu')<br/>('4244188', 'Jun Cao', 'jun cao')<br/>('2180892', 'Yikang Li', 'yikang li')<br/>('2913552', 'Baoxin Li', 'baoxin li')</td><td>Email: Zhigang.Tu@asu.edu +<br/>Email: jun.cao@intel.com +<br/>Email: YikangLi,Baoxin.Li@asu.edu +</td></tr><tr><td>46a4551a6d53a3cd10474ef3945f546f45ef76ee</td><td>2014 IEEE Intelligent Vehicles Symposium (IV) +<br/>June 8-11, 2014. Dearborn, Michigan, USA +<br/>978-1-4799-3637-3/14/$31.00 ©2014 IEEE +<br/>344 +</td><td></td><td></td></tr><tr><td>4686bdcee01520ed6a769943f112b2471e436208</td><td>Utsumi et al. IPSJ Transactions on Computer Vision and +<br/>Applications (2017) 9:11 +<br/>DOI 10.1186/s41074-017-0024-5 +<br/>IPSJ Transactions on Computer +<br/>Vision and Applications +<br/>EXPRESS PAPER +<br/>Open Access +<br/>Fast search based on generalized +<br/>similarity measure +</td><td>('40142989', 'Yuzuko Utsumi', 'yuzuko utsumi')<br/>('4629425', 'Tomoya Mizuno', 'tomoya mizuno')<br/>('35613969', 'Masakazu Iwamura', 'masakazu iwamura')<br/>('3277321', 'Koichi Kise', 'koichi kise')</td><td></td></tr><tr><td>4688787d064e59023a304f7c9af950d192ddd33e</td><td>Investigating the Discriminative Power of Keystroke +<br/>Sound +<br/>and Dimitris Metaxas, Member, IEEE +</td><td>('38993748', 'Joseph Roth', 'joseph roth')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td></td></tr><tr><td>466184b10fb7ce9857e6b5bd6b4e5003e09a0b16</td><td>Extended Grassmann Kernels for +<br/>Subspace-Based Learning +<br/>GRASP Laboratory +<br/><b>University of Pennsylvania</b><br/>Philadelphia, PA 19104 +<br/>GRASP Laboratory +<br/><b>University of Pennsylvania</b><br/>Philadelphia, PA 19104 +</td><td>('2720935', 'Jihun Ham', 'jihun ham')<br/>('1732066', 'Daniel D. Lee', 'daniel d. lee')</td><td>jhham@seas.upenn.edu +<br/>ddlee@seas.upenn.edu +</td></tr><tr><td>46e86cdb674440f61b6658ef3e84fea95ea51fb4</td><td></td><td></td><td></td></tr><tr><td>46f2611dc4a9302e0ac00a79456fa162461a8c80</td><td>for Action Classification +<br/><b>ESAT-PSI, KU Leuven, 2CV:HCI, KIT, Karlsruhe, 3University of Bonn, 4Sensifai</b></td><td>('3310120', 'Ali Diba', 'ali diba')<br/>('3169187', 'Mohsen Fayyaz', 'mohsen fayyaz')<br/>('50633941', 'Vivek Sharma', 'vivek sharma')<br/>('2946643', 'Juergen Gall', 'juergen gall')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>1{firstname.lastname}@kuleuven.be, 2{firstname.lastname}@kit.edu, +<br/>3{lastname}@iai.uni-bonn.de, 4{firstname.lastname}@sensifai.com +</td></tr><tr><td>46b7ee97d7dfbd61cc3745e8dfdd81a15ab5c1d4</td><td>3D FACIAL GEOMETRIC FEATURES FOR CONSTRAINED LOCAL MODEL +<br/><b>cid:2) Imperial College London, United Kingdom</b><br/><b>University of Twente, EEMCS, Netherlands</b></td><td>('1694605', 'Maja Pantic', 'maja pantic')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('3183108', 'Akshay Asthana', 'akshay asthana')<br/>('1902288', 'Shiyang Cheng', 'shiyang cheng')</td><td>{shiyang.cheng11, s.zafeiriou, a.asthana, m.pantic}@imperial.ac.uk +</td></tr><tr><td>46ae4d593d89b72e1a479a91806c39095cd96615</td><td>A CONDITIONAL RANDOM FIELD APPROACH FOR FACE IDENTIFICATION IN +<br/>BROADCAST NEWS USING OVERLAID TEXT +<br/>(1,2)Gay Paul, 1Khoury Elie, 2Meignier Sylvain, 1Odobez Jean-Marc, 2Deleglise Paul +<br/><b>Idiap Research Institute, Martigny, Switzerland, 2LIUM, University of Maine, Le Mans, France</b></td><td></td><td></td></tr><tr><td>467b602a67cfd7c347fe7ce74c02b38c4bb1f332</td><td>Large Margin Local Metric Learning +<br/><b>University College London, London, UK</b><br/>2 Safran Morpho, Issy-les-Moulineaux, France +<br/><b>University of Exceter, Exceter, UK</b></td><td>('38954213', 'Yiming Ying', 'yiming ying')<br/>('1704699', 'Massimiliano Pontil', 'massimiliano pontil')</td><td>m.pontil@cs.ucl.ac.uk +<br/>{julien.bohne,stephane.gentric}@morpho.com +<br/>y.ying@exeter.ac.uk +</td></tr><tr><td>466f80b066215e85da63e6f30e276f1a9d7c843b</td><td>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>Joint Head Pose Estimation and Face Alignment Framework +<br/>Using Global and Local CNN Features +<br/>Computational Biomedicine Lab +<br/><b>University of Houston, Houston, TX, USA</b></td><td>('5084124', 'Xiang Xu', 'xiang xu')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td>{xxu18, ikakadia}@central.uh.edu +</td></tr><tr><td>464de30d3310123644ab81a1f0adc51598586fd2</td><td></td><td></td><td></td></tr><tr><td>466a5add15bb5f91e0cfd29a55f5fb159a7980e5</td><td>Video Repeat Recognition and Mining by Visual +<br/>Features +</td><td>('4052001', 'Xianfeng Yang', 'xianfeng yang')</td><td></td></tr><tr><td>46f3b113838e4680caa5fc8bda6e9ae0d35a038c</td><td>Cancers 2010, 2, 262-273; doi:10.3390/cancers2020262 +<br/>OPEN ACCESS +<br/>cancers +<br/>ISSN 2072-6694 +<br/>www.mdpi.com/journal/cancers +<br/>Review +<br/>Automated Dermoscopy Image Analysis of Pigmented Skin +<br/>Lesions +<br/><b>Section of Pathology, Second University of Naples, Via L. Armanni</b><br/>5, 80138 Naples, Italy +<br/>3 ACS, Advanced Computer Systems, Via della Bufalotta 378, 00139 Rome, Italy +<br/>Fax: +390815569693. +<br/>Received: 23 February 2010; in revised form: 15 March 2010 / Accepted: 25 March 2010 / +<br/>Published: 26 March 2010 +</td><td>('32152948', 'Alfonso Baldi', 'alfonso baldi')<br/>('1705562', 'Marco Quartulli', 'marco quartulli')<br/>('3899127', 'Raffaele Murace', 'raffaele murace')<br/>('5703272', 'Emanuele Dragonetti', 'emanuele dragonetti')<br/>('38220535', 'Mario Manganaro', 'mario manganaro')<br/>('2237329', 'Oscar Guerra', 'oscar guerra')<br/>('4108084', 'Stefano Bizzi', 'stefano bizzi')</td><td>2 Futura-onlus, Via Pordenone 2, 00182 Rome, Italy; E-Mail: raffaele@murace.it +<br/>* Author to whom correspondence should be addressed; E-Mail: alfonsobaldi@tiscali.it; +</td></tr><tr><td>465d5bb11912005f0a4f0569c6524981df18a7de</td><td>IMOTION – Searching for Video Sequences +<br/>using Multi-Shot Sketch Queries +<br/>Metin Sezgin3, Ozan Can Altıok3, and Yusuf Sahillio˘glu3 +<br/>1 Databases and Information Systems Research Group, +<br/><b>University of Basel, Switzerland</b><br/><b>Research Center in Information Technologies, Universit e de Mons, Belgium</b><br/><b>Intelligent User Interfaces Lab, Ko c University, Turkey</b></td><td>('27401642', 'Luca Rossetto', 'luca rossetto')<br/>('2155883', 'Ivan Giangreco', 'ivan giangreco')<br/>('34588610', 'Silvan Heller', 'silvan heller')<br/>('1806643', 'Heiko Schuldt', 'heiko schuldt')<br/>('3272087', 'Omar Seddati', 'omar seddati')</td><td>{luca.rossetto|ivan.giangreco|c.tanase|silvan.heller|heiko.schuldt}@unibas.ch +<br/>{stephane.dupont|omar.seddati}@umons.ac.be +<br/>{mtsezgin|oaltiok15|ysahillioglu}@ku.edu.tr +</td></tr><tr><td>46c87fded035c97f35bb991fdec45634d15f9df2</td><td>Spatial-Aware Object Embeddings for Zero-Shot Localization +<br/>and Classification of Actions +<br/><b>University of Amsterdam</b></td><td>('2606260', 'Pascal Mettes', 'pascal mettes')</td><td></td></tr><tr><td>46e72046a9bb2d4982d60bcf5c63dbc622717f0f</td><td>Learning Discriminative Features with Class Encoder +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/><b>University of Chinese Academy of Science</b></td><td>('1704812', 'Hailin Shi', 'hailin shi')<br/>('8362374', 'Xiangyu Zhu', 'xiangyu zhu')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{hailin.shi, xiangyu.zhu, zlei, scliao, szli}@nlpr.ia.ac.cn +</td></tr><tr><td>46f32991ebb6235509a6d297928947a8c483f29e</td><td>In Proc. IEEE Computer Vision and Pattern Recognition (CVPR), Madison (WI), June 2003 +<br/>Recognizing Expression Variant Faces +<br/>from a Single Sample Image per Class +<br/>Aleix M. Mart(cid:19)(cid:16)nez +<br/>Department of Electrical Engineering +<br/><b>The Ohio State University, OH</b></td><td></td><td>aleix@ee.eng.ohio-state.edu +</td></tr><tr><td>46538b0d841654a0934e4c75ccd659f6c5309b72</td><td>Signal & Image Processing : An International Journal (SIPIJ) Vol.5, No.1, February 2014 +<br/>A NOVEL APPROACH TO GENERATE FACE +<br/>BIOMETRIC TEMPLATE USING BINARY +<br/>DISCRIMINATING ANALYSIS +<br/>1P.G. Student, Department of Computer Engineering, MCERC, Nashik (M.S.), India. +<br/>2Associate Professor, Department of Computer Engineering, +<br/>MCERC, Nashik (M.S.), India +</td><td>('40075681', 'Shraddha S. Shinde', 'shraddha s. shinde')<br/>('2590072', 'Anagha P. Khedkar', 'anagha p. khedkar')</td><td></td></tr><tr><td>4641986af5fc8836b2c883ea1a65278d58fe4577</td><td>Scene Graph Generation by Iterative Message Passing +<br/><b>Stanford University</b><br/><b>Stanford University</b></td><td>('2068265', 'Danfei Xu', 'danfei xu')</td><td>{danfei, yukez, chrischoy, feifeili}@cs.stanford.edu +</td></tr><tr><td>464b3f0824fc1c3a9eaf721ce2db1b7dfe7cb05a</td><td>Deep Adaptive Temporal Pooling for Activity Recognition +<br/><b>Singapore University of Technology and Design</b><br/><b>Singapore University of Technology and Design</b><br/>Singapore, Singapore +<br/>Singapore, Singapore +<br/><b>Institute for Infocomm Research</b><br/>Singapore, Singapore +<br/><b>Keele University</b><br/>Keele, Staffordshire, United Kingdom +</td><td>('1729827', 'Ngai-Man Cheung', 'ngai-man cheung')<br/>('2527741', 'Sibo Song', 'sibo song')<br/>('1802086', 'Vijay Chandrasekhar', 'vijay chandrasekhar')<br/>('1709001', 'Bappaditya Mandal', 'bappaditya mandal')</td><td>ngaiman_cheung@sutd.edu.sg +<br/>sibo_song@mymail.sutd.edu.sg +<br/>vijay@i2r.a-star.edu.sg +<br/>b.mandal@keele.ac.uk +</td></tr><tr><td>469ee1b00f7bbfe17c698ccded6f48be398f2a44</td><td>MIT International Journal of Computer Science and Information Technology, Vol. 4, No. 2, August 2014, pp. 82-88 +<br/>ISSN 2230-7621©MIT Publications +<br/>82 +<br/>SURVEy: Techniques for +<br/>Aging Problems in Face Recognition +<br/>Aashmi +<br/>Scholar, Computer Science Engg. Dept. +<br/><b>Moradabad Institute of Technology</b><br/>Scholar, Computer Science Engg. Dept. +<br/><b>Moradabad Institute of Technology</b><br/>Scholar, Computer Science Engg. Dept. +<br/><b>Moradabad Institute of Technology</b><br/>Moradabad, U.P., INDIA +<br/>Moradabad, U.P., INDIA +<br/>Moradabad, U.P., INDIA +</td><td>('40062749', 'Sakshi Sahni', 'sakshi sahni')<br/>('9186211', 'Sakshi Saxena', 'sakshi saxena')</td><td>E-mail: aashmichaudhary@gmail.com +<br/>E-mail: sakshisahni92@gmail.com +<br/>E-mail: saxena.sakshi2511992@gmail.com +</td></tr><tr><td>46196735a201185db3a6d8f6e473baf05ba7b68f</td><td></td><td></td><td></td></tr><tr><td>4682fee7dc045aea7177d7f3bfe344aabf153bd5</td><td>Tabula Rasa: Model Transfer for +<br/>Object Category Detection +<br/>Department of Engineering Science +<br/>Oxford +<br/>(Presented by Elad Liebman) +</td><td>('3152281', 'Yusuf Aytar', 'yusuf aytar')</td><td></td></tr><tr><td>4657d87aebd652a5920ed255dca993353575f441</td><td>Image Normalization for +<br/>Illumination Compensation in Facial Images +<br/>by +<br/>Department of Electrical & Computer Engineering +<br/>& Center for Intelligent Machines +<br/><b>McGill University, Montreal, Canada</b><br/>August 2004 +</td><td>('3631473', 'Martin D. Levine', 'martin d. levine')<br/>('35712223', 'Jisnu Bhattacharyya', 'jisnu bhattacharyya')</td><td></td></tr><tr><td>4622b82a8aff4ac1e87b01d2708a333380b5913b</td><td>Multi-label CNN Based Pedestrian Attribute Learning for Soft Biometrics +<br/>Center for Biometrics and Security Research, +<br/><b>Institute of Automation, Chinese Academy of Sciences</b><br/>95 Zhongguancun Donglu, Beijing 100190, China +</td><td>('1739258', 'Jianqing Zhu', 'jianqing zhu')<br/>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('1716143', 'Dong Yi', 'dong yi')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>jianqingzhu@foxmail.com, {scliao, dyi, zlei, szli}@nlpr.ia.ac.cn +</td></tr><tr><td>46e866f58419ff4259c65e8256c1d4f14927b2c6</td><td>On the Generalization Power of Face and Gait Gender +<br/>Recognition Methods +<br/><b>University of Warwick</b><br/>Gibbet Hill Road, Coventry, CV4 7AL, UK +</td><td>('1735787', 'Yu Guan', 'yu guan')<br/>('1799504', 'Chang-Tsun Li', 'chang-tsun li')</td><td>{g.yu, x.wei, c-t.li}@warwick.ac.uk +</td></tr><tr><td>46072f872eee3413f9d05482be6446f6b96b6c09</td><td>Trace Quotient Problems Revisited +<br/>1 Department of Information Engineering, +<br/><b>The Chinese University of Hong Kong, Hong Kong</b><br/>2 Microsoft Research Asia, Beijing, China +</td><td>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>4698a599425c3a6bae1c698456029519f8f2befe</td><td>Transferring Rich Deep Features +<br/>for Facial Beauty Prediction +<br/><b>College of Informatics</b><br/><b>College of Informatics</b><br/>Department of Computer Science and Engineering +<br/><b>Huazhong Agricultural University</b><br/><b>Huazhong Agricultural University</b><br/>Wuhan, China +<br/>Wuhan, China +<br/><b>University of North Texas</b><br/>Denton, USA +</td><td>('40557104', 'Lu Xu', 'lu xu')<br/>('2697879', 'Jinhai Xiang', 'jinhai xiang')<br/>('1982703', 'Xiaohui Yuan', 'xiaohui yuan')</td><td>Email: xulu coi@webmail.hzau.edu.cn +<br/>Email: jimmy xiang@mail.hzau.edu.cn +<br/>Email: Xiaohui.Yuan@unt.edu +</td></tr><tr><td>2c424f21607ff6c92e640bfe3da9ff105c08fac4</td><td>Learning Structured Output Representation +<br/>using Deep Conditional Generative Models +<br/><b>NEC Laboratories America, Inc</b><br/><b>University of Michigan, Ann Arbor</b></td><td>('1729571', 'Kihyuk Sohn', 'kihyuk sohn')<br/>('3084614', 'Xinchen Yan', 'xinchen yan')<br/>('1697141', 'Honglak Lee', 'honglak lee')</td><td>ksohn@nec-labs.com, {xcyan,honglak}@umich.edu +</td></tr><tr><td>2c258eec8e4da9e65018f116b237f7e2e0b2ad17</td><td>Deep Quantization: Encoding Convolutional Activations +<br/>with Deep Generative Model ∗ +<br/><b>University of Science and Technology of China, Hefei, China</b><br/>Microsoft Research, Beijing, China +</td><td>('3430743', 'Zhaofan Qiu', 'zhaofan qiu')<br/>('2053452', 'Ting Yao', 'ting yao')<br/>('1724211', 'Tao Mei', 'tao mei')</td><td>zhaofanqiu@gmail.com, {tiyao, tmei}@microsoft.com +</td></tr><tr><td>2cbb4a2f8fd2ddac86f8804fd7ffacd830a66b58</td><td></td><td></td><td></td></tr><tr><td>2c8743089d9c7df04883405a31b5fbe494f175b4</td><td>Washington State Convention Center +<br/>Seattle, Washington, May 26-30, 2015 +<br/>978-1-4799-6922-7/15/$31.00 ©2015 IEEE +<br/>3039 +</td><td></td><td></td></tr><tr><td>2c61a9e26557dd0fe824909adeadf22a6a0d86b0</td><td></td><td></td><td></td></tr><tr><td>2c93c8da5dfe5c50119949881f90ac5a0a4f39fe</td><td>Advanced local motion patterns for macro and micro facial +<br/>expression recognition +<br/>B. Allaerta,∗, IM. Bilascoa, C. Djerabaa +<br/>aUniv. Lille, CNRS, Centrale Lille, UMR 9189 - CRIStAL - +<br/>Centre de Recherche en Informatique Signal et Automatique de Lille, F-59000 Lille, France +</td><td></td><td></td></tr><tr><td>2c34bf897bad780e124d5539099405c28f3279ac</td><td>Robust Face Recognition via Block Sparse Bayesian Learning +<br/><b>School of Financial Information Engineering, Southwestern University of Finance and Economics, Chengdu</b><br/>China +<br/><b>Institute of Chinese Payment System, Southwestern University of Finance and Economics, Chengdu 610074, China</b><br/><b>University of California at San Diego, La Jolla, CA</b><br/>USA +<br/><b>Samsung RandD Institute America - Dallas, 1301 East Lookout Drive, Richardson, TX 75082, USA</b></td><td>('2775350', 'Taiyong Li', 'taiyong li')<br/>('1791667', 'Zhilin Zhang', 'zhilin zhang')</td><td></td></tr><tr><td>2c203050a6cca0a0bff80e574bda16a8c46fe9c2</td><td>Discriminative Deep Hashing for Scalable Face Image Retrieval +<br/><b>School of Computer Science and Engineering, Nanjing University of Science and Technology</b></td><td>('1699053', 'Jie Lin', 'jie lin')<br/>('3233021', 'Zechao Li', 'zechao li')<br/>('8053308', 'Jinhui Tang', 'jinhui tang')</td><td>jinhuitang@njust.edu.cn +</td></tr><tr><td>2cc4ae2e864321cdab13c90144d4810464b24275</td><td>23 +<br/>Face Recognition Using Optimized 3D +<br/>Information from Stereo Images +<br/>1Advanced Technology R&D Center, Samsung Thales Co., Ltd., 2Graduate School of +<br/><b>Advanced Imaging Science, Multimedia, and Film Chung-Ang University, Seoul</b><br/>Korea +<br/>1. Introduction +<br/>Human biometric characteristics are unique, so it can not be easily duplicated [1]. Such +<br/>information +<br/>includes; facial, hands, torso, fingerprints, etc. Potential applications, +<br/>economical efficiency, and user convenience make the face detection and recognition +<br/>technique an important commodity compared to other biometric features [2], [3]. It can also +<br/>use a low-cost personal computer (PC) camera instead of expensive equipments, and require +<br/>minimal user interface. Recently, extensive research using 3D face data has been carried out +<br/>in order to overcome the limits of 2D face detection and feature extraction [2], which +<br/>includes PCA [3], neural networks (NN) [4], support vector machines (SVM) [5], hidden +<br/>markov models (HMM) [6], and linear discriminant analysis (LDA) [7]. Among them, PCA +<br/>and LDA methods with self-learning method are most widely used [3]. The frontal face +<br/>image database provides fairly high recognition rate. However, if the view data of facial +<br/>rotation, illumination and pose change is not acquired, the correct recognition rate +<br/>remarkably drops because of the entire face modeling. Such performance degradation +<br/>problem can be solved by using a new recognition method based on the optimized 3D +<br/>information in the stereo face images. +<br/>This chapter presents a new face detection and recognition method using optimized 3D +<br/>information from stereo images. The proposed method can significantly improve the +<br/>recognition rate and is robust against object’s size, distance, motion, and depth using the +<br/>PCA algorithm. By using the optimized 3D information, we estimate the position of the eyes +<br/>in the stereo face images. As a result, we can accurately detect the facial size, depth, and +<br/>rotation in the stereo face images. For efficient detection of face area, we adopt YCbCr color +<br/>format. The biggest object can be chosen as a face candidate among the candidate areas +<br/>which are extracted by the morphological opening for the Cb and Cr components [8]. In +<br/>order to detect the face characteristics such as eyes, nose, and mouth, a pre-processing is +<br/>performed, which utilizes brightness information in the estimated face area. For fast +<br/>processing, we train the partial face region segmented by estimating the position of eyes, +<br/>instead of the entire face region. Figure 1. shows the block diagram of proposed algorithm. +<br/>This chapter is organized as follows: Section 2 and 3 describe proposed stereo vision system +<br/>and pos estimation for face recognition, respectively. Section 4 presents experimental, and +<br/>section 5 concludes the chapter. +<br/>Source: Face Recognition, Book edited by: Kresimir Delac and Mislav Grgic, ISBN 978-3-902613-03-5, pp.558, I-Tech, Vienna, Austria, June 2007 +</td><td>('1727735', 'Changhan Park', 'changhan park')<br/>('1684329', 'Joonki Paik', 'joonki paik')</td><td></td></tr><tr><td>2c3430e0cbe6c8d7be3316a88a5c13a50e90021d</td><td>Multi-feature Spectral Clustering with Minimax Optimization +<br/>School of Electrical and Electronic Engineering +<br/><b>Nanyang Technological University, Singapore</b></td><td>('19172541', 'Hongxing Wang', 'hongxing wang')<br/>('1764228', 'Chaoqun Weng', 'chaoqun weng')<br/>('34316743', 'Junsong Yuan', 'junsong yuan')</td><td>{hwang8, weng0018}@e.ntu.edu.sg, jsyuan@ntu.edu.sg +</td></tr><tr><td>2cac8ab4088e2bdd32dcb276b86459427355085c</td><td>A Face-to-Face Neural Conversation Model +<br/>Hang Chu1 +<br/><b>University of Toronto 2Vector Institute</b></td><td>('46598920', 'Daiqing Li', 'daiqing li')</td><td>{chuhang1122, daiqing, fidler}@cs.toronto.edu +</td></tr><tr><td>2cde051e04569496fb525d7f1b1e5ce6364c8b21</td><td>Sparse 3D convolutional neural networks +<br/><b>University of Warwick</b><br/>August 26, 2015 +</td><td>('39294240', 'Ben Graham', 'ben graham')</td><td>b.graham@warwick.ac.uk +</td></tr><tr><td>2c2786ea6386f2d611fc9dbf209362699b104f83</td><td></td><td>('31914125', 'Mohammad Shahidul Islam', 'mohammad shahidul islam')</td><td></td></tr><tr><td>2c92839418a64728438c351a42f6dc5ad0c6e686</td><td>Pose-Aware Face Recognition in the Wild +<br/>Prem Natarajan2 +<br/><b>USC Institute for Robotics and Intelligent Systems (IRIS), Los Angeles, CA</b><br/>G´erard Medioni1 +<br/><b>USC Information Sciences Institute (ISI), Marina Del Rey, CA</b></td><td>('11269472', 'Iacopo Masi', 'iacopo masi')<br/>('38696444', 'Stephen Rawls', 'stephen rawls')</td><td>{srawls,pnataraj}@isi.edu +<br/>{iacopo.masi,medioni}@usc.edu +</td></tr><tr><td>2c848cc514293414d916c0e5931baf1e8583eabc</td><td>An automatic facial expression recognition system +<br/>evaluated by different classifiers +<br/>∗Programa de P´os-Graduac¸˜ao em Mecatrˆonica +<br/>Universidade Federal da Bahia, +<br/>†Department of Electrical Engineering - EESC/USP +</td><td>('3797834', 'Caroline Silva', 'caroline silva')<br/>('2105008', 'Raissa Tavares Vieira', 'raissa tavares vieira')</td><td>Email: lolyne.pacheco@gmail.com +<br/>Email: andrewssobral@gmail.com +<br/>Email: raissa@ieee.org, +</td></tr><tr><td>2c883977e4292806739041cf8409b2f6df171aee</td><td>Aalborg Universitet +<br/>Are Haar-like Rectangular Features for Biometric Recognition Reducible? +<br/>Nasrollahi, Kamal; Moeslund, Thomas B. +<br/>Published in: +<br/>Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications +<br/>DOI (link to publication from Publisher): +<br/>10.1007/978-3-642-41827-3_42 +<br/>Publication date: +<br/>2013 +<br/>Document Version +<br/>Early version, also known as pre-print +<br/><b>Link to publication from Aalborg University</b><br/>Citation for published version (APA): +<br/>Nasrollahi, K., & Moeslund, T. B. (2013). Are Haar-like Rectangular Features for Biometric Recognition +<br/>Reducible? In J. Ruiz-Shulcloper, & G. Sanniti di Baja (Eds.), Progress in Pattern Recognition, Image Analysis, +<br/>Computer Vision, and Applications (Vol. 8259, pp. 334-341). Springer Berlin Heidelberg: Springer Publishing +<br/>Company. Lecture Notes in Computer Science, DOI: 10.1007/978-3-642-41827-3_42 +<br/>General rights +<br/>Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners +<br/>and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. +<br/> ? Users may download and print one copy of any publication from the public portal for the purpose of private study or research. +<br/> ? You may not further distribute the material or use it for any profit-making activity or commercial gain +<br/> ? You may freely distribute the URL identifying the publication in the public portal ? +<br/>Take down policy +<br/>the work immediately and investigate your claim. +<br/>Downloaded from vbn.aau.dk on: oktober 28, 2017 +<br/> </td><td></td><td>If you believe that this document breaches copyright please contact us at vbn@aub.aau.dk providing details, and we will remove access to +</td></tr><tr><td>2cdd9e445e7259117b995516025fcfc02fa7eebb</td><td>Title +<br/>Temporal Exemplar-based Bayesian Networks for facial +<br/>expression recognition +<br/>Author(s) +<br/>Shang, L; Chan, KP +<br/>Citation +<br/>Proceedings - 7Th International Conference On Machine +<br/>Learning And Applications, Icmla 2008, 2008, p. 16-22 +<br/>Issued Date +<br/>2008 +<br/>URL +<br/>http://hdl.handle.net/10722/61208 +<br/>Rights +<br/>This work is licensed under a Creative Commons Attribution- +<br/>NonCommercial-NoDerivatives 4.0 International License.; +<br/>International Conference on Machine Learning and Applications +<br/>Proceedings. Copyright © IEEE.; ©2008 IEEE. Personal use of +<br/>this material is permitted. However, permission to +<br/>reprint/republish this material for advertising or promotional +<br/>purposes or for creating new collective works for resale or +<br/>redistribution to servers or lists, or to reuse any copyrighted +<br/>component of this work in other works must be obtained from +<br/>the IEEE. +</td><td></td><td></td></tr><tr><td>2c1ffb0feea5f707c890347d2c2882be0494a67a</td><td>Learning to learn high capacity generative models from few examples +<br/>The Variational Homoencoder: +<br/>Tommi Jaakkola1 +<br/><b>Massachusetts Institute of Technology</b><br/>2MIT-IBM Watson AI Lab +</td><td>('51152627', 'Luke B. Hewitt', 'luke b. hewitt')<br/>('51150953', 'Maxwell I. Nye', 'maxwell i. nye')<br/>('3071104', 'Andreea Gane', 'andreea gane')<br/>('1763295', 'Joshua B. Tenenbaum', 'joshua b. tenenbaum')</td><td></td></tr><tr><td>2cdc40f20b70ca44d9fd8e7716080ee05ca7924a</td><td>Real-time Convolutional Neural Networks for +<br/>Emotion and Gender Classification +<br/>Hochschule Bonn-Rhein-Sieg +<br/>Sankt Augustin Germany +<br/>Paul G. Pl¨oger +<br/>Hochschule Bonn-Rhein-Sieg +<br/>Sankt Augustin Germany +<br/>Matias Valdenegro +<br/><b>Heriot-Watt University</b><br/>Edinburgh, UK +</td><td>('27629437', 'Octavio Arriaga', 'octavio arriaga')</td><td>Email: octavio.arriaga@smail.inf.h-brs.de +<br/>Email: paul.ploeger@h-brs.de +<br/>Email: m.valdenegro@hw.ac.uk +</td></tr><tr><td>2cac70f9c8140a12b6a55cef834a3d7504200b62</td><td>Reconstructing High Quality Face-Surfaces using Model Based Stereo +<br/><b>University of Basel, Switzerland</b><br/>Contribution +<br/>We present a method to fit a detailed 3D morphable +<br/>model to multiple images. Our formulation allows +<br/>the fitting of the model without determining the +<br/>lighting conditions and albedo of the face, mak- +<br/>ing the system robust against difficult lighting sit- +<br/>uations and unmodelled albedo variations such as +<br/>skin colour, moles, freckles and cast shadows. +<br/>The cost function employs +<br/>Microsoft Research, Cambridge‡ +<br/>Ambient Lighting +<br/>Evaluation: Gold Standard +<br/>Ambient Only Dataset (20 Subjects) +<br/>Stereo: Landmarks + Silhouette + Colour +<br/>Stereo: Landmarks + Silhouette +<br/>Stereo: Landmarks +<br/>Monocular +<br/>The model shape prior +<br/>A small number of landmarks for initialization +<br/>A monocular silhouette distance cost +<br/>A stereo colour cost +<br/>The optimisation consists of multiple runs of a non- +<br/>linear minimizer. During each run the visibility of +<br/>all sample points is assumed to stay constant. After +<br/>some iterations the minimizer is stopped and visi- +<br/>bility is reevaluated. +<br/>Model +<br/>The linear morphable face model was created by +<br/>registering 200 face scans and performing a PCA on +<br/>the data matrix to fit a Gaussian probability to the +<br/>data and reduce the dimensionality of the model. +<br/>Input Images +<br/>Multiview +<br/>Landmarks +<br/>Multiview +<br/>L.+Silhouette +<br/>Multiview +<br/>L.+S.+Colour +<br/>Ground Truth +<br/>Monocular [1] +<br/>Each cue increases the reconstruction accuracy, lead- +<br/>ing to significantly better result than possible with +<br/>the state of the art monocular system [1]. Recon- +<br/>structions of the face surface are compared to ground +<br/>truth data acquired with a structured light system. +<br/>The point wise distance from the reconstruction to +<br/>the ground truth is shown in the inset head render- +<br/>ings. Here green is a perfect match, and red denotes +<br/>a distance of 3mm or more. +<br/>The best of the three monocular results is shown. +<br/>Silhouette Cost +<br/>Directed Lighting +<br/>The silhouette cost measures +<br/>the distance of the silhouette +<br/>to image edges. An edge cost +<br/>surface is created from the im- +<br/>age, by combining the distance +<br/>transforms of edge detections +<br/>with different thresholds. The +<br/>cost ist integrated over the pro- +<br/>jection of 3D sample points at +<br/>the silhouette of the hypotheses. +<br/>Edge Cost Surface +<br/>Colour Reprojection Cost +<br/>The colour +<br/>reprojection cost +<br/>measures the image colour dif- +<br/>ference between the projected +<br/>positions of sample points in +<br/>two images. The sample points +<br/>are spaced out regularly in the +<br/>projected images. +<br/>Multiview Ground Truth Monocular +<br/>Input Images +<br/>The new stereo algorithm is robust under directed +<br/>lighting and yields significantly more accurate sur- +<br/>face reconstructions than the monocular algorithm. +<br/>Again the distance to the ground truth is shown +<br/>Funding +<br/>This work was supported in part by Microsoft Research through +<br/>the European PhD Scholarship Programme. +<br/>Multiview Ground Truth Monocular +<br/>Input Images +<br/>for green=0mm and red=3mm in the insets. Future +<br/>work will include a skin and lighting model, hope- +<br/>fully improving speed and accuracy of the method. +<br/>All cues were used. +<br/>References +<br/>[1] S. Romdhani and T. Vetter. Estimating 3D Shape and Texture +<br/>Using Pixel Intensity, Edges, Specular Highlights, Texture +<br/>Constraints and a Prior. In CVPR 2005 +<br/>Distance to Ground Truth (mm) +<br/>Directed Light Dataset (5 Subjects) +<br/>Stereo: Landmarks + Silhouette + Colour +<br/>Stereo: Landmarks + Silhouette +<br/>Stereo: Landmarks +<br/>Monocular +<br/>Distance to Ground Truth (mm) +<br/>The use of multi-view information results in a +<br/>much higher accuracy than achievable by the +<br/>monocular method. A higher frequency of lower +<br/>residuals is better. +<br/>Evaluation: Face Recognition +<br/>To test the method on a difficult dataset, a face +<br/>recognition experiment on the PIE dataset was per- +<br/>formed. The results show, that the extracted sur- +<br/>faces are consistent over variations in viewpoint +<br/>and that the reconstruction quality increases with +<br/>an increasing number of images. +<br/>View- +<br/>points +<br/>Landmark +<br/>+ Silhouette +<br/>+ Colour +<br/>2nd +<br/>2nd +<br/>1st +<br/>1st +<br/>2nd +<br/>1st +<br/>68% 63% 82% +<br/>10% 18% 50% +<br/>74% 74% 85% +<br/>7% 18% 62% +<br/>82% 87% 94% +<br/>19% 37% 76% +<br/>The columns labelled “1st” show the frequency of +<br/>correct results, “2nd” is the frequency with which +<br/>the correct result was within the first two subjects +<br/>returned. The angle between the shape coefficients +<br/>was used as the distance measure. +<br/>Texture information should be used to achieve state +<br/>of the art recognition results. +<br/>FaceCamera1Camera2SamplePoint </td><td>('1994157', 'Brian Amberg', 'brian amberg')<br/>('1745076', 'Andrew Blake', 'andrew blake')<br/>('3293655', 'Sami Romdhani', 'sami romdhani')<br/>('1687079', 'Thomas Vetter', 'thomas vetter')</td><td></td></tr><tr><td>2c5d1e0719f3ad7f66e1763685ae536806f0c23b</td><td>AENet: Learning Deep Audio Features for Video +<br/>Analysis +</td><td>('47893464', 'Naoya Takahashi', 'naoya takahashi')<br/>('3037160', 'Michael Gygli', 'michael gygli')<br/>('7329802', 'Luc van Gool', 'luc van gool')</td><td></td></tr><tr><td>2c8f24f859bbbc4193d4d83645ef467bcf25adc2</td><td>845 +<br/>Classification in the Presence of +<br/>Label Noise: a Survey +</td><td>('1786603', 'Benoît Frénay', 'benoît frénay')<br/>('1782629', 'Michel Verleysen', 'michel verleysen')</td><td></td></tr><tr><td>2c1f8ddbfbb224271253a27fed0c2425599dfe47</td><td>Understanding and Comparing Deep Neural Networks +<br/>for Age and Gender Classification +<br/><b>Fraunhofer Heinrich Hertz Institute</b><br/><b>Singapore University of Technology and Design</b><br/>10587 Berlin, Germany +<br/>Klaus-Robert M¨uller +<br/><b>Berlin Institute of Technology</b><br/>10623 Berlin, Germany +<br/>Singapore 487372, Singapore +<br/><b>Fraunhofer Heinrich Hertz Institute</b><br/>10587 Berlin, Germany +</td><td>('3633358', 'Sebastian Lapuschkin', 'sebastian lapuschkin')<br/>('40344011', 'Alexander Binder', 'alexander binder')<br/>('1699054', 'Wojciech Samek', 'wojciech samek')</td><td>sebastian.lapuschkin@hhi.fraunhofer.de +<br/>alexander binder@sutd.edu.sg +<br/>klaus-robert.mueller@tu-berlin.de +<br/>wojciech.samek@hhi.fraunhofer.de +</td></tr><tr><td>2ca43325a5dbde91af90bf850b83b0984587b3cc</td><td>For Your Eyes Only – Biometric Protection of PDF Documents +<br/><b>Faculty of ETI, Gdansk University of Technology, Gdansk, Poland</b></td><td>('2026734', 'J. Siciarek', 'j. siciarek')</td><td></td></tr><tr><td>2cfc28a96b57e0817cc9624a5d553b3aafba56f3</td><td>P2F2: Privacy-Preserving Face Finder +<br/><b>New Jersey Institute of Technology</b></td><td>('9037517', 'Nora Almalki', 'nora almalki')<br/>('1692516', 'Reza Curtmola', 'reza curtmola')<br/>('34645435', 'Xiaoning Ding', 'xiaoning ding')<br/>('1690806', 'Cristian Borcea', 'cristian borcea')</td><td>Email: {naa34, crix, xiaoning.ding, narain.gehani, borcea}@njit.edu +</td></tr><tr><td>2cdd5b50a67e4615cb0892beaac12664ec53b81f</td><td>To appear in ACM TOG 33(6). +<br/>Mirror Mirror: Crowdsourcing Better Portraits +<br/>Jun-Yan Zhu1 +<br/>Aseem Agarwala2 +<br/>Jue Wang2 +<br/><b>University of California, Berkeley1 Adobe</b><br/>Figure 1: We collect thousands of portraits by capturing video of a subject while they watch movie clips designed to elicit a range of positive +<br/>emotions. We use crowdsourcing and machine learning to train models that can predict attractiveness scores of different expressions. These +<br/>models can be used to select a subject’s best expressions across a range of emotions, from more serious professional portraits to big smiles. +</td><td>('1763086', 'Alexei A. Efros', 'alexei a. efros')<br/>('2177801', 'Eli Shechtman', 'eli shechtman')</td><td></td></tr><tr><td>2cae619d0209c338dc94593892a787ee712d9db0</td><td>Selective Hidden Random Fields: Exploiting Domain-Specific Saliency for Event +<br/>Classification +<br/><b>University of Massachusetts Amherst</b><br/>Amherst MA USA +</td><td>('2246870', 'Vidit Jain', 'vidit jain')</td><td>vidit@cs.umass.edu +</td></tr><tr><td>2c0acaec54ab2585ff807e18b6b9550c44651eab</td><td>Face Quality Assessment for Face Verification in Video +<br/><b>Lomonosov Moscow State University, 2Video Analysis Technologies, LLC</b><br/>fusion of +<br/>facial +</td><td>('38982797', 'M. Nikitin', 'm. nikitin')<br/>('2943115', 'V. Konushin', 'v. konushin')<br/>('1934937', 'A. Konushin', 'a. konushin')</td><td>mnikitin@graphics.cs.msu.ru, vadim@tevian.ru, ktosh@graphics.cs.msu.ru +</td></tr><tr><td>2cdde47c27a8ecd391cbb6b2dea64b73282c7491</td><td>ORDER-AWARE CONVOLUTIONAL POOLING FOR VIDEO BASED ACTION RECOGNITION +<br/>Order-aware Convolutional Pooling for Video Based +<br/>Action Recognition +</td><td>('1722767', 'Peng Wang', 'peng wang')<br/>('2161037', 'Lingqiao Liu', 'lingqiao liu')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')<br/>('1724393', 'Heng Tao Shen', 'heng tao shen')</td><td></td></tr><tr><td>2c62b9e64aeddf12f9d399b43baaefbca8e11148</td><td>Evaluation of Dense 3D Reconstruction from 2D Face Images in the Wild +<br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, Guildford GU2 7XH, UK</b><br/><b>Faculty of Natural Sciences, University of Stirling, Stirling FK9 4LA, UK</b><br/><b>School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China</b><br/><b>Biometrics Research Lab, College of Computer Science, Sichuan University, Chengdu 610065, China</b><br/><b>Image Understanding and Interactive Robotics, Reutlingen University, 72762 Reutlingen, Germany</b></td><td>('2976854', 'Zhen-Hua Feng', 'zhen-hua feng')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('7345195', 'Qijun Zhao', 'qijun zhao')</td><td>{z.feng, j.kittler, p.koppen}@surrey.ac.uk, patrikhuber@gmail.com, +<br/>wu_xiaojun@jiangnan.edu.cn, p.j.b.hancock@stir.ac.uk, qjzhao@scu.edu.cn +</td></tr><tr><td>2c7c3a74da960cc76c00965bd3e343958464da45</td><td></td><td></td><td></td></tr><tr><td>2cf5f2091f9c2d9ab97086756c47cd11522a6ef3</td><td>MPIIGaze: Real-World Dataset and Deep +<br/>Appearance-Based Gaze Estimation +</td><td>('2520795', 'Xucong Zhang', 'xucong zhang')<br/>('1751242', 'Yusuke Sugano', 'yusuke sugano')<br/>('1739548', 'Mario Fritz', 'mario fritz')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')</td><td></td></tr><tr><td>2c19d3d35ef7062061b9e16d040cebd7e45f281d</td><td>End-to-end Video-level Representation Learning for Action Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences (CASIA</b><br/><b>University of Chinese Academy of Sciences (UCAS</b></td><td>('1696573', 'Jiagang Zhu', 'jiagang zhu')<br/>('1726367', 'Wei Zou', 'wei zou')<br/>('48147901', 'Zheng Zhu', 'zheng zhu')</td><td>{zhujiagang2015, wei.zou}@ia.ac.cn, zhuzheng14@mails.ucas.ac.cn +</td></tr><tr><td>2c17d36bab56083293456fe14ceff5497cc97d75</td><td>Unconstrained Face Alignment via Cascaded Compositional Learning +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b><br/>2SenseTime Group Limited +</td><td>('2226254', 'Shizhan Zhu', 'shizhan zhu')<br/>('40475617', 'Cheng Li', 'cheng li')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>zs014@ie.cuhk.edu.hk, chengli@sensetime.com, ccloy@ie.cuhk.edu.hk, xtang@ie.cuhk.edu.hk +</td></tr><tr><td>2c4b96f6c1a520e75eb37c6ee8b844332bc0435c</td><td>Automatic Emotion Recognition in Robot-Children Interaction for ASD +<br/>Treatment +<br/>ISASI UOS Lecce +<br/>Campus Universitario via Monteroni sn, 73100 Lecce Italy +<br/>ISASI UOS Messina +<br/>Univerisita’ di Bari +<br/><b>Marine Institute, via Torre Bianca, 98164 Messina Italy</b><br/>Via Orabona 4, 70126 Bari, Italy +</td><td>('4730472', 'Marco Leo', 'marco leo')<br/>('33097940', 'Marco Del Coco', 'marco del coco')<br/>('1741861', 'Cosimo Distante', 'cosimo distante')<br/>('3049247', 'Giovanni Pioggia', 'giovanni pioggia')<br/>('2235498', 'Giuseppe Palestra', 'giuseppe palestra')</td><td>marco.leo@cnr.it +</td></tr><tr><td>2cd7821fcf5fae53a185624f7eeda007434ae037</td><td>Exploring the Geo-Dependence of Human Face Appearance +<br/>Computer Science +<br/><b>University of Kentucky</b><br/>Computer Science +<br/>UNC Charlotte +<br/>Computer Science +<br/><b>University of Kentucky</b></td><td>('2142962', 'Mohammad T. Islam', 'mohammad t. islam')<br/>('38792670', 'Scott Workman', 'scott workman')<br/>('1873911', 'Hui Wu', 'hui wu')<br/>('1690110', 'Richard Souvenir', 'richard souvenir')<br/>('1990750', 'Nathan Jacobs', 'nathan jacobs')</td><td>{tarik,scott}@cs.uky.edu +<br/>{hwu13,souvenir}@uncc.edu +<br/>jacobs@cs.uky.edu +</td></tr><tr><td>79581c364cefe53bff6bdd224acd4f4bbc43d6d4</td><td></td><td></td><td></td></tr><tr><td>794ddb1f3b7598985d4d289b5b0664be736a50c4</td><td>Exploiting Competition Relationship for Robust Visual Recognition +<br/>Center for Data Analytics and Biomedical Informatics +<br/>Department of Computer and Information Science +<br/><b>Temple University</b><br/>Philadelphia, PA, 19122, USA +</td><td>('38909760', 'Liang Du', 'liang du')<br/>('1805398', 'Haibin Ling', 'haibin ling')</td><td>{liang.du, hbling}@temple.edu +</td></tr><tr><td>790aa543151312aef3f7102d64ea699a1d15cb29</td><td>Confidence-Weighted Local Expression Predictions for +<br/>Occlusion Handling in Expression Recognition and Action +<br/>Unit detection +<br/>1 Sorbonne Universités, UPMC Univ Paris 06, CNRS, ISIR UMR 7222 +<br/>4 place Jussieu 75005 Paris +</td><td>('3190846', 'Arnaud Dapogny', 'arnaud dapogny')<br/>('2521061', 'Kevin Bailly', 'kevin bailly')<br/>('1701986', 'Séverine Dubuisson', 'séverine dubuisson')</td><td>arnaud.dapogny@isir.upmc.fr +<br/>kevin.bailly@isir.upmc.fr +<br/>severine.dubuisson@isir.upmc.fr +</td></tr><tr><td>79f6a8f777a11fd626185ab549079236629431ac</td><td>Copyright +<br/>by +<br/>2013 +</td><td>('35788904', 'Sung Ju Hwang', 'sung ju hwang')</td><td></td></tr><tr><td>795ea140df2c3d29753f40ccc4952ef24f46576c</td><td></td><td></td><td></td></tr><tr><td>79dc84a3bf76f1cb983902e2591d913cee5bdb0e</td><td></td><td></td><td></td></tr><tr><td>79744fc71bea58d2e1918c9e254b10047472bd76</td><td>Disentangling 3D Pose in A Dendritic CNN +<br/>for Unconstrained 2D Face Alignment +<br/>Department of Electrical and Computer Engineering, CFAR and UMIACS +<br/><b>University of Maryland-College Park, USA</b></td><td>('50333013', 'Amit Kumar', 'amit kumar')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>akumar14@umiacs.umd.edu, rama@umiacs.umd.edu +</td></tr><tr><td>79b669abf65c2ca323098cf3f19fa7bdd837ff31</td><td> Deakin Research Online +<br/>This is the published version: +<br/>Rana, Santu, Liu, Wanquan, Lazarescu, Mihai and Venkatesh, Svetha 2008, Efficient tensor +<br/>based face recognition, in ICPR 2008 : Proceedings of the 19th International Conference on +<br/>Pattern Recognition, IEEE, Washington, D. C., pp. 1-4. +<br/>Available from Deakin Research Online: +<br/>http://hdl.handle.net/10536/DRO/DU:30044585 +<br/> +<br/>Reproduced with the kind permissions of the copyright owner. +<br/>Personal use of this material is permitted. However, permission to reprint/republish this +<br/>material for advertising or promotional purposes or for creating new collective works for +<br/>resale or redistribution to servers or lists, or to reuse any copyrighted component of this work +<br/>in other works must be obtained from the IEEE. +<br/>Copyright : 2008, IEEE +</td><td></td><td></td></tr><tr><td>794c0dc199f0bf778e2d40ce8e1969d4069ffa7b</td><td>Odd Leaf Out +<br/>Improving visual recognition with games +<br/>Preece +<br/>School of Information +<br/><b>University of Maryland</b><br/><b>College Park, United States</b></td><td>('6519022', 'Darcy Lewis', 'darcy lewis')<br/>('2662457', 'Dana Rotman', 'dana rotman')</td><td></td></tr><tr><td>79c3a7131c6c176b02b97d368cd0cd0bc713ff7e</td><td></td><td></td><td></td></tr><tr><td>79dd787b2877cf9ce08762d702589543bda373be</td><td>Face Detection Using SURF Cascade +<br/>Intel Labs China +</td><td>('35423937', 'Jianguo Li', 'jianguo li')<br/>('40279370', 'Tao Wang', 'tao wang')<br/>('2470865', 'Yimin Zhang', 'yimin zhang')</td><td></td></tr><tr><td>799c02a3cde2c0805ea728eb778161499017396b</td><td>PersonRank: Detecting Important People in Images +<br/><b>School of Electronics and Information Technology, Sun Yat-Sen University, GuangZhou, China</b><br/><b>School of Data and Computer Science, Sun Yat-Sen University, GuangZhou, China</b></td><td>('9186191', 'Benchao Li', 'benchao li')<br/>('3333315', 'Wei-Shi Zheng', 'wei-shi zheng')</td><td></td></tr><tr><td>7966146d72f9953330556baa04be746d18702047</td><td>Harnessing Human Manipulation +<br/>NSF/ARL Workshop on Cloud Robotics: Challenges and Opportunities +<br/>February 27-28, 2013 +<br/><b>The Robotics Institute Carnegie Mellon University</b><br/><b>Georgia Institute of Technology</b></td><td>('1781040', 'Matthew T. Mason', 'matthew t. mason')<br/>('1735665', 'Nancy Pollard', 'nancy pollard')<br/>('1760708', 'Alberto Rodriguez', 'alberto rodriguez')<br/>('38637733', 'Ryan Kerwin', 'ryan kerwin')</td><td><matt.mason, nsp, albertor>@cs.cmu.edu +<br/>ryankerwin@gatech.edu +</td></tr><tr><td>79fa57dedafddd3f3720ca26eb41c82086bfb332</td><td>Modeling Facial Expression Space for Recognition * +<br/>National Lab. on Machine Perception +<br/><b>Peking University</b><br/>Beijing, China +<br/>National Lab. on Machine Perception +<br/><b>Peking University</b><br/>Beijing, China +<br/>National Lab. on Machine Perception +<br/><b>Peking University</b><br/>Beijing, China +</td><td>('2086289', 'Hong Liu', 'hong liu')<br/>('1687248', 'Hongbin Zha', 'hongbin zha')<br/>('2976781', 'Yuwen Wu', 'yuwen wu')</td><td>wuyw@cis.pku.edu.cn +<br/>liuhong@cis.pku.edu.cn +<br/>zha@cis.pku.edu.cn +</td></tr><tr><td>793e7f1ba18848908da30cbad14323b0389fd2a8</td><td></td><td></td><td></td></tr><tr><td>79db191ca1268dc88271abef3179c4fe4ee92aed</td><td>Facial Expression Based Automatic Album +<br/>Creation +<br/><b>School of Computer Science, CECS, Australian National University, Canberra</b><br/><b>School of Engineering, CECS, Australian National University, Canberra, Australia</b><br/>3 Vision & Sensing, Faculty of Information Sciences and Engineering, +<br/>Australia +<br/><b>University of Canberra, Australia</b></td><td>('1735697', 'Abhinav Dhall', 'abhinav dhall')<br/>('3183108', 'Akshay Asthana', 'akshay asthana')<br/>('1717204', 'Roland Goecke', 'roland goecke')</td><td>abhinav.dhall@anu.edu.au, aasthana@rsise.anu.edu.au, +<br/>roland.goecke@ieee.org +</td></tr><tr><td>2d990b04c2bd61d3b7b922b8eed33aeeeb7b9359</td><td>Discriminative Dictionary Learning with +<br/>Pairwise Constraints +<br/><b>University of Maryland, College Park, MD</b></td><td>('2723427', 'Huimin Guo', 'huimin guo')<br/>('34145947', 'Zhuolin Jiang', 'zhuolin jiang')<br/>('1693428', 'Larry S. Davis', 'larry s. davis')</td><td>{hmguo,zhuolin,lsd}@umiacs.umd.edu +</td></tr><tr><td>2d25045ec63f9132371841c0beccd801d3733908</td><td>Sensors 2015, 15, 6719-6739; doi:10.3390/s150306719 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>Multi-Layer Sparse Representation for Weighted LBP-Patches +<br/>Based Facial Expression Recognition +<br/><b>School of Software, Dalian University of Technology, Dalian 116621, China</b><br/>Tel.: +86-411-8757-1516. +<br/>Academic Editor: Vittorio M.N. Passaro +<br/>Received: 15 December 2014 / Accepted: 10 March 2015 / Published: 19 March 2015 +</td><td>('2235253', 'Qi Jia', 'qi jia')<br/>('3459398', 'Xinkai Gao', 'xinkai gao')<br/>('2736880', 'He Guo', 'he guo')<br/>('7864960', 'Zhongxuan Luo', 'zhongxuan luo')<br/>('1734275', 'Yi Wang', 'yi wang')</td><td>E-Mails: jiaqi7166@gmail.com (Q.J.); gaoxinkai@mail.dlut.edu.cn (X.G.); zxluo@dlut.edu.cn (Z.L.); +<br/>wangyi_dlut@126.com (Y.W.) +<br/>* Author to whom correspondence should be addressed; E-Mail: guohe@dlut.edu.cn; +</td></tr><tr><td>2dd6c988b279d89ab5fb5155baba65ce4ce53c1e</td><td></td><td></td><td></td></tr><tr><td>2d080662a1653f523321974a57518e7cb67ecb41</td><td>On Constrained Local Model Feature +<br/>Normalization for Facial Expression Recognition +<br/><b>School of Computing and Info. Sciences, Florida International University</b><br/>11200 SW 8th St, Miami, FL 33199, USA +<br/>http://ascl.cis.fiu.edu/ +</td><td>('3489972', 'Zhenglin Pan', 'zhenglin pan')<br/>('2008564', 'Mihai Polceanu', 'mihai polceanu')</td><td>zpan004@fiu.edu,{mpolcean,lisetti}@cs.fiu.edu +</td></tr><tr><td>2d4b9fe3854ccce24040074c461d0c516c46baf4</td><td>Temporal Action Localization by Structured Maximal Sums +<br/><b>State Key Laboratory for Novel Software Technology, Nanjing University, China</b><br/><b>University of Michigan, Ann Arbor</b></td><td>('40188401', 'Jonathan C. Stroud', 'jonathan c. stroud')<br/>('2285916', 'Tong Lu', 'tong lu')<br/>('8342699', 'Jia Deng', 'jia deng')</td><td></td></tr><tr><td>2d294c58b2afb529b26c49d3c92293431f5f98d0</td><td>4413 +<br/>Maximum Margin Projection Subspace Learning +<br/>for Visual Data Analysis +</td><td>('1793625', 'Symeon Nikitidis', 'symeon nikitidis')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td></td></tr><tr><td>2d1f86e2c7ba81392c8914edbc079ac64d29b666</td><td></td><td></td><td></td></tr><tr><td>2d9e58ea582e054e9d690afca8b6a554c3687ce6</td><td>Learning Local Feature Aggregation Functions +<br/>with Backpropagation +<br/>Multimedia Understanding Group +<br/><b>Aristotle University of Thessaloniki, Greece</b></td><td>('3493855', 'Angelos Katharopoulos', 'angelos katharopoulos')<br/>('3493472', 'Despoina Paschalidou', 'despoina paschalidou')<br/>('1789830', 'Christos Diou', 'christos diou')<br/>('1708199', 'Anastasios Delopoulos', 'anastasios delopoulos')</td><td>{katharas, pdespoin}@auth.gr; diou@mug.ee.auth.gr; adelo@eng.auth.gr +</td></tr><tr><td>2d164f88a579ba53e06b601d39959aaaae9016b7</td><td>Dynamic Facial Expression Recognition Using +<br/>A Bayesian Temporal Manifold Model +<br/>Department of Computer Science +<br/><b>Queen Mary University of London</b><br/>Mile End Road, London E1 4NS, UK +</td><td>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('2803283', 'Peter W. McOwan', 'peter w. mcowan')</td><td>{cfshan, sgg, pmco}@dcs.qmul.ac.uk +</td></tr><tr><td>2d8001ffee6584b3f4d951d230dc00a06e8219f8</td><td>Feature Agglomeration Networks for Single Stage Face Detection +<br/><b>School of Information Systems, Singapore Management University, Singapore</b><br/><b>College of Computer Science and Technology, Zhejiang University, Hangzhou, China</b><br/>§DeepIR Inc., Beijing, China +</td><td>('1826176', 'Jialiang Zhang', 'jialiang zhang')<br/>('2791484', 'Xiongwei Wu', 'xiongwei wu')<br/>('1704030', 'Jianke Zhu', 'jianke zhu')</td><td>{chhoi,xwwu.2015@phdis}@smu.edu.sg;{zjialiang,jkzhu}@zju.edu.cn +</td></tr><tr><td>2d23fa205acca9c21e3e1a04674f1e5a9528550e</td><td>The Fast and the Flexible: +<br/>Extended Pseudo Two-Dimensional Warping for +<br/>Face Recognition +<br/>1Computer Vision and Multimodal Computing +<br/>2 Computer Vision Laboratory +<br/>MPI Informatics, Saarbruecken +<br/>ETH Zurich +<br/>3Human Language Technology and Pattern Recognition Group, +<br/><b>RWTH Aachen University</b></td><td>('2299109', 'Leonid Pishchulin', 'leonid pishchulin')<br/>('1948162', 'Tobias Gass', 'tobias gass')<br/>('1967060', 'Philippe Dreuw', 'philippe dreuw')<br/>('1685956', 'Hermann Ney', 'hermann ney')</td><td>leonid@mpi-inf.mpg.de +<br/>gasst@vision.ee.ethz.ch +<br/><last name>@cs.rwth-aachen.de +</td></tr><tr><td>2d244d70ed1a2ba03d152189f1f90ff2b4f16a79</td><td>An Analytical Mapping for LLE and Its +<br/>Application in Multi-Pose Face Synthesis +<br/>State Key Lab of Intelligent Technology and Systems +<br/><b>Tsinghua University</b><br/>Beijing, 100084, China +</td><td>('1715001', 'Jun Wang', 'jun wang')</td><td>wangjun00@mails.tsinghua.edu.cn +<br/>zcs@mail.tsinghua.edu.cn +<br/>kzb98@mails.tsinghua.edu.cn +</td></tr><tr><td>2d88e7922d9f046ace0234f9f96f570ee848a5b5</td><td>Building Better Detection with Privileged Information +<br/>Department of CSE +<br/>The Pennsylvania State +<br/><b>University</b><br/>Department of CSE +<br/>The Pennsylvania State +<br/><b>University</b><br/>Applied Communication +<br/>Sciences +<br/>Basking Ridge, NJ, US +<br/>Department of CSE +<br/>The Pennsylvania State +<br/><b>University</b><br/>Army Research +<br/>Laboratory +<br/>Adelphi, MD, USA +</td><td>('2950892', 'Z. Berkay Celik', 'z. berkay celik')<br/>('4108832', 'Patrick McDaniel', 'patrick mcdaniel')<br/>('1804289', 'Rauf Izmailov', 'rauf izmailov')<br/>('1967156', 'Nicolas Papernot', 'nicolas papernot')<br/>('1703726', 'Ananthram Swami', 'ananthram swami')</td><td>zbc102@cse.psu.edu +<br/>mcdaniel@cse.psu.edu +<br/>rizmailov@appcomsci.com +<br/>npg5056@cse.psu.edu +<br/>ananthram.swami.civ@mail.mil +</td></tr><tr><td>2d31ab536b3c8a05de0d24e0257ca4433d5a7c75</td><td>Materials Discovery: Fine-Grained Classification of X-ray Scattering Images +<br/>Kevin Yager† +<br/><b>University of North Carolina at Chapel Hill, NC, USA</b><br/>†Brookhaven National Lab, NY, USA +</td><td>('1772294', 'M. Hadi Kiapour', 'm. hadi kiapour')<br/>('39668247', 'Alexander C. Berg', 'alexander c. berg')<br/>('1685538', 'Tamara L. Berg', 'tamara l. berg')</td><td>{hadi,aberg,tlberg}@cs.unc.edu +<br/>kyager@bnl.gov +</td></tr><tr><td>2dbde64ca75e7986a0fa6181b6940263bcd70684</td><td>Pose Independent Face Recognition by Localizing +<br/>Local Binary Patterns via Deformation Components +<br/><b>MICC, University of Florence</b><br/>Italy +<br/>http://www.micc.unifi.it/vim +<br/>G´erard Medioni +<br/><b>USC IRIS Lab, University of Southern California</b><br/>Los Angeles, USA +<br/>http://iris.usc.edu/USC-Computer-Vision.html +</td><td>('11269472', 'Iacopo Masi', 'iacopo masi')<br/>('35220006', 'Claudio Ferrari', 'claudio ferrari')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')</td><td></td></tr><tr><td>2d0363a3ebda56d91d704d5ff5458a527775b609</td><td>Attribute2Image: Conditional Image Generation from Visual Attributes +<br/>1Computer Science and Engineering Division +<br/>2Adobe Research +<br/>3NEC Labs +<br/><b>University of Michigan, Ann Arbor</b></td><td>('3084614', 'Xinchen Yan', 'xinchen yan')<br/>('1768964', 'Jimei Yang', 'jimei yang')<br/>('1729571', 'Kihyuk Sohn', 'kihyuk sohn')<br/>('1697141', 'Honglak Lee', 'honglak lee')</td><td>{xcyan,kihyuks,honglak}@umich.edu +<br/>jimyang@adobe.com +<br/>ksohn@nec-labs.com +</td></tr><tr><td>2d93a9aa8bed51d0d1b940c73ac32c046ebf1eb8</td><td>Perceptual Reward Functions +<br/><b>College of Computing, Georgia Institute of Technology, Atlanta, GA, USA</b><br/><b>Waseda University, Tokyo, Japan</b></td><td>('1737432', 'Atsuo Takanishi', 'atsuo takanishi')</td><td>aedwards8@gatech.edu, isbell@cc.gatech.edu +<br/>takanisi@waseda.jp +</td></tr><tr><td>2dd2c7602d7f4a0b78494ac23ee1e28ff489be88</td><td>Large Scale Metric Learning from Equivalence Constraints ∗ +<br/><b>Institute for Computer Graphics and Vision, Graz University of Technology</b></td><td>('2918450', 'Martin Hirzer', 'martin hirzer')<br/>('3202367', 'Paul Wohlhart', 'paul wohlhart')<br/>('1791182', 'Peter M. Roth', 'peter m. roth')<br/>('3628150', 'Horst Bischof', 'horst bischof')</td><td>{koestinger,hirzer,wohlhart,pmroth,bischof}@icg.tugraz.at +</td></tr><tr><td>2d84e30c61281d3d7cdd11676683d6e66a68aea6</td><td>Automatic Construction of Action Datasets +<br/>using Web videos with Density-based Cluster +<br/>Analysis and Outlier Detection +<br/><b>The University of Electro-Communications</b><br/>185-8585 , Japan Tokyo Chofu Chofugaoka 1-5-1 +</td><td>('1681659', 'Keiji Yanai', 'keiji yanai')</td><td></td></tr><tr><td>2d98a1cb0d1a37c79a7ebcb727066f9ccc781703</td><td>Coupled Support Vector Machines for Supervised +<br/>Domain Adaptation +<br/>∗Center for Cognitive Ubiquitous Computing, Arizona State Univeristy +<br/>† Bosch Research and Technology Center, Palo Alto +<br/><b>University of Michigan, Ann Arbor</b></td><td>('3151995', 'Hemanth Venkateswara', 'hemanth venkateswara')<br/>('2929090', 'Prasanth Lade', 'prasanth lade')<br/>('37513601', 'Jieping Ye', 'jieping ye')<br/>('1743991', 'Sethuraman Panchanathan', 'sethuraman panchanathan')</td><td>hemanthv@asu.edu, prasanth.lade@us.bosch.com, jpye@umich.edu, +<br/>panch@asu.edu +</td></tr><tr><td>2dced31a14401d465cd115902bf8f508d79de076</td><td>ORIGINAL RESEARCH +<br/>published: 26 May 2015 +<br/>doi: 10.3389/fbioe.2015.00064 +<br/>Can a humanoid face be expressive? +<br/>A psychophysiological investigation +<br/><b>Research Center E. Piaggio , University of Pisa, Pisa, Italy, 2 Faculty of Psychology, University of Florence, Florence, Italy</b><br/><b>University of Pisa, Pisa, Italy</b><br/>Non-verbal signals expressed through body language play a crucial role in multi-modal +<br/>human communication during social relations. Indeed, in all cultures, facial expressions +<br/>are the most universal and direct signs to express innate emotional cues. A human face +<br/>conveys important information in social interactions and helps us to better understand +<br/>our social partners and establish empathic links. Latest researches show that humanoid +<br/>and social robots are becoming increasingly similar to humans, both esthetically and +<br/>expressively. However, their visual expressiveness is a crucial issue that must be improved +<br/>to make these robots more realistic and intuitively perceivable by humans as not different +<br/>from them. This study concerns the capability of a humanoid robot to exhibit emotions +<br/>through facial expressions. More specifically, emotional signs performed by a humanoid +<br/>robot have been compared with corresponding human facial expressions in terms of +<br/>recognition rate and response time. The set of stimuli +<br/>included standardized human +<br/>expressions taken from an Ekman-based database and the same facial expressions +<br/>performed by the robot. Furthermore, participants’ psychophysiological responses have +<br/>been explored to investigate whether there could be differences induced by interpreting +<br/>robot or human emotional stimuli. Preliminary results show a trend to better recognize +<br/>expressions performed by the robot than 2D photos or 3D models. Moreover, no +<br/>significant differences in the subjects’ psychophysiological state have been found during +<br/>the discrimination of facial expressions performed by the robot in comparison with the +<br/>same task performed with 2D photos and 3D models. +<br/>Keywords: facial expressions, emotion perception, humanoid robot, expression recognition, social robots, +<br/>psychophysiological signals, affective computing +<br/>1. Introduction +<br/>Human beings communicate in a rich and sophisticated way through many different channels, +<br/>e.g., sound, vision, and touch. In human social relationships, visual information plays a crucial +<br/>role. Human faces convey important information both from static features, such as identity, age, +<br/>and gender, and from dynamic changes, such as expressions, eye blinking, and muscular micro- +<br/>movements. The ability to recognize and understand facial expressions of the social partner allows +<br/>us to establish and manage the empathic links that drive our social relationships. +<br/>Charles Darwin was the first to observe that basic expressions, such as anger, disgust, contempt, +<br/>fear, surprise, sadness, and happiness, are universal and innate (Darwin, 1872). Since the publication +<br/>of his book “The Expression of the Emotions in Man and Animals” in 1872, a strong debate over the +<br/>Edited by: +<br/>Cecilia Laschi, +<br/>Scuola Superiore Sant’Anna, Italy +<br/>Reviewed by: +<br/>John-John Cabibihan, +<br/><b>Qatar University, Qatar</b><br/>Egidio Falotico, +<br/>Scuola Superiore Sant’Anna, Italy +<br/>*Correspondence: +<br/><b>Research Center E. Piaggio</b><br/><b>University of Pisa, Largo Lucio</b><br/>Lazzarino 1, Pisa 56122, Italy +<br/>Specialty section: +<br/>This article was submitted to Bionics +<br/>and Biomimetics, a section of the +<br/>journal Frontiers in Bioengineering and +<br/>Biotechnology +<br/>Received: 24 November 2014 +<br/>Accepted: 27 April 2015 +<br/>Published: 26 May 2015 +<br/>Citation: +<br/>Lazzeri N, Mazzei D, Greco A, Rotesi +<br/>A, Lanatà A and De Rossi DE (2015) +<br/>Can a humanoid face be expressive? +<br/>A psychophysiological investigation. +<br/>Front. Bioeng. Biotechnol. 3:64. +<br/>doi: 10.3389/fbioe.2015.00064 +<br/>Frontiers in Bioengineering and Biotechnology | www.frontiersin.org +<br/>May 2015 | Volume 3 | Article 64 +</td><td>('35440863', 'Nicole Lazzeri', 'nicole lazzeri')<br/>('34573296', 'Daniele Mazzei', 'daniele mazzei')<br/>('32070391', 'Alberto Greco', 'alberto greco')<br/>('6284325', 'Annalisa Rotesi', 'annalisa rotesi')<br/>('1730665', 'Antonio Lanatà', 'antonio lanatà')<br/>('20115987', 'Danilo Emilio De Rossi', 'danilo emilio de rossi')<br/>('34573296', 'Daniele Mazzei', 'daniele mazzei')</td><td>mazzei@di.unipi.it +</td></tr><tr><td>2d05e768c64628c034db858b7154c6cbd580b2d5</td><td>Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/> A Monthly Journal of Computer Science and Information Technology +<br/> IJCSMC, Vol. 4, Issue. 8, August 2015, pg.431 – 446 +<br/> RESEARCH ARTICLE +<br/>ISSN 2320–088X +<br/>FACIAL EXPRESSION RECOGNITION: +<br/>Machine Learning using C# +</td><td></td><td>Author: Neda Firoz (nedafiroz1910@gmail.com) +<br/>Advisor: Dr. Prashant Ankur Jain (prashant.jain@shiats.edu.in) +</td></tr><tr><td>2dfe0e7e81f65716b09c590652a4dd8452c10294</td><td>ORIGINAL RESEARCH +<br/>published: 06 June 2018 +<br/>doi: 10.3389/fpsyg.2018.00864 +<br/>Incongruence Between Observers’ +<br/>and Observed Facial Muscle +<br/>Activation Reduces Recognition of +<br/>Emotional Facial Expressions From +<br/>Video Stimuli +<br/><b>Centre for Applied Autism Research, University of Bath, Bath, United Kingdom, 2 Social and</b><br/><b>Cognitive Neuroscience Laboratory, Centre of Biology and Health Sciences, Mackenzie Presbyterian University, S o Paulo</b><br/><b>Brazil, University Hospital Zurich, Z rich</b><br/><b>Switzerland, Psychosomatic Medicine, and Psychotherapy, University Hospital Frankfurt</b><br/>Frankfurt, Germany +<br/>According to embodied cognition accounts, viewing others’ facial emotion can elicit +<br/>the respective emotion representation in observers which entails simulations of sensory, +<br/>motor, and contextual experiences. In line with that, published research found viewing +<br/>others’ +<br/>facial emotion to elicit automatic matched facial muscle activation, which +<br/>was further found to facilitate emotion recognition. Perhaps making congruent facial +<br/>muscle activity explicit produces an even greater recognition advantage. If there is +<br/><b>con icting sensory information, i.e., incongruent facial muscle activity, this might impede</b><br/>recognition. The effects of actively manipulating facial muscle activity on facial emotion +<br/>recognition from videos were investigated across three experimental conditions: (a) +<br/>explicit imitation of viewed facial emotional expressions (stimulus-congruent condition), +<br/>(b) pen-holding with the lips (stimulus-incongruent condition), and (c) passive viewing +<br/>(control condition). It was hypothesised that (1) experimental condition (a) and (b) result +<br/>in greater facial muscle activity than (c), (2) experimental condition (a) increases emotion +<br/>recognition accuracy from others’ faces compared to (c), (3) experimental condition (b) +<br/>lowers recognition accuracy for expressions with a salient facial feature in the lower, +<br/>but not the upper face area, compared to (c). Participants (42 males, 42 females) +<br/>underwent a facial emotion recognition experiment (ADFES-BIV) while electromyography +<br/>(EMG) was recorded from five facial muscle sites. The experimental conditions’ order +<br/>was counter-balanced. Pen-holding caused stimulus-incongruent facial muscle activity +<br/>for expressions with facial feature saliency in the lower face region, which reduced +<br/>recognition of lower face region emotions. Explicit imitation caused stimulus-congruent +<br/>facial muscle activity without modulating recognition. Methodological +<br/>implications are +<br/>discussed. +<br/>Keywords: facial emotion recognition, imitation, facial muscle activity, facial EMG, embodiment, videos, dynamic +<br/>stimuli, facial expressions of emotion +<br/>Edited by: +<br/>Eva G. Krumhuber, +<br/><b>University College London</b><br/>United Kingdom +<br/>Reviewed by: +<br/>Sebastian Korb, +<br/>Universität Wien, Austria +<br/>Michal Olszanowski, +<br/><b>SWPS University of Social Sciences</b><br/>and Humanities, Poland +<br/>*Correspondence: +<br/>Tanja S. H. Wingenbach +<br/>Specialty section: +<br/>This article was submitted to +<br/>Emotion Science, +<br/>a section of the journal +<br/>Frontiers in Psychology +<br/>Received: 15 December 2017 +<br/>Accepted: 14 May 2018 +<br/>Published: 06 June 2018 +<br/>Citation: +<br/>Wingenbach TSH, Brosnan M, +<br/>Pfaltz MC, Plichta MM and Ashwin C +<br/>(2018) Incongruence Between +<br/>Observers’ and Observed Facial +<br/>Muscle Activation Reduces +<br/>Recognition of Emotional Facial +<br/>Expressions From Video Stimuli. +<br/>Front. Psychol. 9:864. +<br/>doi: 10.3389/fpsyg.2018.00864 +<br/>Frontiers in Psychology | www.frontiersin.org +<br/>June 2018 | Volume 9 | Article 864 +</td><td>('39455300', 'Mark Brosnan', 'mark brosnan')<br/>('34495803', 'Monique C. Pfaltz', 'monique c. pfaltz')<br/>('2976177', 'Michael M. Plichta', 'michael m. plichta')<br/>('2708124', 'Chris Ashwin', 'chris ashwin')</td><td>tanja.wingenbach@bath.edu +</td></tr><tr><td>2d072cd43de8d17ce3198fae4469c498f97c6277</td><td>Random Cascaded-Regression Copse for Robust +<br/>Facial Landmark Detection +<br/>and Xiao-Jun Wu +</td><td>('2976854', 'Zhen-Hua Feng', 'zhen-hua feng')<br/>('39976184', 'Patrik Huber', 'patrik huber')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td></td></tr><tr><td>2dd5f1d69e0e8a95a10f3f07f2c0c7fa172994b3</td><td>20 +<br/>Machine Analysis of Facial Expressions +<br/><b>Imperial College London</b><br/><b>Inst. Neural Computation, University of California</b><br/>1 UK, 2 USA +<br/>1. Human Face and Its Expression +<br/>The human face is the site for major sensory inputs and major communicative outputs. It +<br/>houses the majority of our sensory apparatus as well as our speech production apparatus. It +<br/>is used to identify other members of our species, to gather information about age, gender, +<br/>attractiveness, and personality, and to regulate conversation by gazing or nodding. +<br/>Moreover, the human face is our preeminent means of communicating and understanding +<br/>somebody’s affective state and intentions on the basis of the shown facial expression +<br/>(Keltner & Ekman, 2000). Thus, the human face +<br/>input-output +<br/>communicative system capable of tremendous flexibility and specificity (Ekman & Friesen, +<br/>1975). In general, the human face conveys information via four kinds of signals. +<br/>(a) Static facial signals represent relatively permanent features of the face, such as the bony +<br/>structure, the soft tissue, and the overall proportions of the face. These signals +<br/>contribute to an individual’s appearance and are usually exploited for person +<br/>identification. +<br/>is a multi-signal +<br/>(b) Slow facial signals represent changes in the appearance of the face that occur gradually +<br/>over time, such as development of permanent wrinkles and changes in skin texture. +<br/>These signals can be used for assessing the age of an individual. Note that these signals +<br/>might diminish the distinctness of the boundaries of the facial features and impede +<br/>recognition of the rapid facial signals. +<br/>(c) Artificial signals are exogenous features of the face such as glasses and cosmetics. These +<br/>signals provide additional information that can be used for gender recognition. Note +<br/>that these signals might obscure facial features or, conversely, might enhance them. +<br/>(d) Rapid facial signals represent temporal changes in neuromuscular activity that may lead +<br/><b>to visually detectable changes in facial appearance, including blushing and tears. These</b><br/>(atomic facial) signals underlie facial expressions. +<br/>All four classes of signals contribute to person identification, gender recognition, +<br/>attractiveness assessment, and personality prediction. In Aristotle’s time, a theory was +<br/>proposed about mutual dependency between static facial signals (physiognomy) and +<br/>personality: “soft hair reveals a coward, strong chin a stubborn person, and a smile a happy +<br/>person”. Today, few psychologists share the belief about the meaning of soft hair and strong +<br/>chin, but many believe that rapid facial signals (facial expressions) communicate emotions +<br/>(Ekman & Friesen, 1975; Ambady & Rosenthal, 1992; Keltner & Ekman, 2000) and +<br/>personality traits (Ambady & Rosenthal, 1992). More specifically, types of messages +<br/>Source: Face Recognition, Book edited by: Kresimir Delac and Mislav Grgic, ISBN 978-3-902613-03-5, pp.558, I-Tech, Vienna, Austria, June 2007 +</td><td>('1694605', 'Maja Pantic', 'maja pantic')<br/>('2218905', 'Marian Stewart Bartlett', 'marian stewart bartlett')</td><td></td></tr><tr><td>2d71e0464a55ef2f424017ce91a6bcc6fd83f6c3</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>National Conference on Advancements in Computer & Information Technology (NCACIT-2016) +<br/>A Survey on: Image Process using Two- Stage Crawler +<br/>Assistant Professor +<br/>SPPU, Pune +<br/>Department of Computer Engg +<br/>Department of Computer Engg +<br/>Department of Computer Engg +<br/>BE Student +<br/>SPPU, Pune +<br/>BE Student +<br/>SPPU, Pune +<br/>BE Student +<br/>Department of Computer Engg +<br/>SPPU, Pune +<br/>additional +<br/>analysis +<br/>for +<br/>information +</td><td>('15156505', 'Nilesh Wani', 'nilesh wani')<br/>('1936852', 'Savita Gunjal', 'savita gunjal')</td><td></td></tr><tr><td>2d38fd1df95f5025e2cee5bc439ba92b369a93df</td><td>Scalable Object-Class Search +<br/>via Sparse Retrieval Models and Approximate Ranking +<br/>Dartmouth Computer Science Technical Report TR2011-700 +<br/>Computer Science Department +<br/><b>Dartmouth College</b><br/>Hanover, NH 03755, U.S.A. +<br/>July 5, 2011 +</td><td>('2563325', 'Mohammad Rastegari', 'mohammad rastegari')<br/>('2442612', 'Chen Fang', 'chen fang')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')</td><td>{mrastegari, chenfang, lorenzo}@cs.dartmouth.edu +</td></tr><tr><td>2d83ba2d43306e3c0587ef16f327d59bf4888dc3</td><td>Large-scale Video Classification with Convolutional Neural Networks +<br/><b>Stanford University</b><br/>1Google Research +<br/>http://cs.stanford.edu/people/karpathy/deepvideo +</td><td>('2354728', 'Andrej Karpathy', 'andrej karpathy')<br/>('1805076', 'George Toderici', 'george toderici')<br/>('24792872', 'Sanketh Shetty', 'sanketh shetty')<br/>('1893833', 'Thomas Leung', 'thomas leung')<br/>('1694199', 'Rahul Sukthankar', 'rahul sukthankar')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td>karpathy@cs.stanford.edu +<br/>gtoderici@google.com +<br/>sanketh@google.com +<br/>leungt@google.com +<br/>sukthankar@google.com +<br/>feifeili@cs.stanford.edu +</td></tr><tr><td>2d84c0d96332bb4fbd8acced98e726aabbf15591</td><td><b>UNIVERSITY OF CALIFORNIA</b><br/>RIVERSIDE +<br/>Investigating the Role of Saliency for Face Recognition +<br/>A Dissertation submitted in partial satisfaction +<br/>of the requirements for the degree of +<br/>Doctor of Philosophy +<br/>in +<br/>Electrical Engineering +<br/>by +<br/>March 2015 +<br/>Dissertation Committee: +<br/>Professor Conrad Rudolph +</td><td>('11012197', 'Ramya Malur Srinivasan', 'ramya malur srinivasan')<br/>('1688416', 'Amit K Roy-Chowdhury', 'amit k roy-chowdhury')<br/>('1686303', 'Ertem Tuncel', 'ertem tuncel')<br/>('2357146', 'Tamar Shinar', 'tamar shinar')</td><td></td></tr><tr><td>2d8d089d368f2982748fde93a959cf5944873673</td><td>Proceedings of NAACL-HLT 2018, pages 788–794 +<br/>New Orleans, Louisiana, June 1 - 6, 2018. c(cid:13)2018 Association for Computational Linguistics +<br/>788 +</td><td></td><td></td></tr><tr><td>2d79d338c114ece1d97cde1aa06ab4cf17d38254</td><td>iLab-20M: A large-scale controlled object dataset to investigate deep learning +<br/><b>Center for Research in Computer Vision, University of Central Florida</b><br/><b>Amirkabir University of Technology, University of Southern California</b></td><td>('3177797', 'Ali Borji', 'ali borji')<br/>('2391309', 'Saeed Izadi', 'saeed izadi')<br/>('7326223', 'Laurent Itti', 'laurent itti')</td><td>aborji@crcv.ucf.edu, sizadi@aut.ac.ir, itti@usc.edu +</td></tr><tr><td>2df4d05119fe3fbf1f8112b3ad901c33728b498a</td><td>Facial landmark detection using structured output deep +<br/>neural networks +<br/>Soufiane Belharbi ∗1, Cl´ement Chatelain∗1, Romain H´erault∗1, and S´ebastien +<br/>Adam∗2 +<br/>1LITIS EA 4108, INSA de Rouen, Saint ´Etienne du Rouvray 76800, France +<br/>2LITIS EA 4108, UFR des Sciences, Universit´e de Rouen, France. +<br/>September 24, 2015 +</td><td></td><td></td></tr><tr><td>2d3482dcff69c7417c7b933f22de606a0e8e42d4</td><td>Labeled Faces in the Wild: Updates and +<br/>New Reporting Procedures +<br/><b>University of Massachusetts, Amherst Technical Report UM-CS</b></td><td>('3219900', 'Gary B. Huang', 'gary b. huang')<br/>('1714536', 'Erik Learned-Miller', 'erik learned-miller')</td><td></td></tr><tr><td>2d4a3e9361505616fa4851674eb5c8dd18e0c3cf</td><td>Towards Privacy-Preserving Visual Recognition +<br/>via Adversarial Training: A Pilot Study +<br/><b>Texas AandM University, College Station TX 77843, USA</b><br/>2 Adobe Research, San Jose CA 95110, USA +</td><td>('1733940', 'Zhenyu Wu', 'zhenyu wu')<br/>('2969311', 'Zhangyang Wang', 'zhangyang wang')<br/>('8056043', 'Zhaowen Wang', 'zhaowen wang')<br/>('39909162', 'Hailin Jin', 'hailin jin')</td><td>{wuzhenyu sjtu,atlaswang}@tamu.edu +<br/>{zhawang,hljin}@adobe.com +</td></tr><tr><td>2d748f8ee023a5b1fbd50294d176981ded4ad4ee</td><td>TRIPLET SIMILARITY EMBEDDING FOR FACE VERIFICATION +<br/><b>Center for Automation Research, UMIACS, University of Maryland, College Park, MD</b><br/>1Department of Electrical and Computer Engineering, +</td><td>('2716670', 'Swami Sankaranarayanan', 'swami sankaranarayanan')<br/>('2943431', 'Azadeh Alavi', 'azadeh alavi')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{swamiviv, azadeh, rama}@umiacs.umd.edu +</td></tr><tr><td>2d3c17ced03e4b6c4b014490fe3d40c62d02e914</td><td>COMPUTER ANIMATION AND VIRTUAL WORLDS +<br/>Comp.Anim.VirtualWorlds2012; 23:167–178 +<br/>Published online 30 May 2012 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/cav.1455 +<br/>SPECIAL ISSUE PAPER +<br/>Video-driven state-aware facial animation +<br/><b>State Key Lab of CADandCG, Zhejiang University, Hangzhou, Zhejiang, China</b><br/>2 Microsoft Corporation, Seattle, WA, USA +</td><td>('2894564', 'Ming Zeng', 'ming zeng')<br/>('1680293', 'Lin Liang', 'lin liang')<br/>('3227032', 'Xinguo Liu', 'xinguo liu')<br/>('1679542', 'Hujun Bao', 'hujun bao')</td><td></td></tr><tr><td>41f26101fed63a8d149744264dd5aa79f1928265</td><td>Spot On: Action Localization from +<br/>Pointly-Supervised Proposals +<br/><b>University of Amsterdam</b><br/><b>Delft University of Technology</b></td><td>('2606260', 'Pascal Mettes', 'pascal mettes')<br/>('1738975', 'Jan C. van Gemert', 'jan c. van gemert')</td><td></td></tr><tr><td>4188bd3ef976ea0dec24a2512b44d7673fd4ad26</td><td>1050 +<br/>Nonlinear Non-Negative Component +<br/>Analysis Algorithms +</td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('2871609', 'Maria Petrou', 'maria petrou')</td><td></td></tr><tr><td>416b559402d0f3e2b785074fcee989d44d82b8e5</td><td>Multi-View Super Vector for Action Recognition +<br/>1Shenzhen Key Lab of Computer Vision and Pattern Recognition, +<br/><b>Shenzhen Institutes of Advanced Technology, CAS, China</b><br/><b>The Chinese University of Hong Kong, Hong Kong</b></td><td>('2985266', 'Zhuowei Cai', 'zhuowei cai')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('1766837', 'Xiaojiang Peng', 'xiaojiang peng')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td>{iamcaizhuowei, 07wanglimin, xiaojiangp}@gmail.com, yu.qiao@siat.ac.cn +</td></tr><tr><td>416364cfdbc131d6544582e552daf25f585c557d</td><td>Synthesis and Recognition of Facial Expressions in Virtual 3D Views +<br/><b>Queen Mary, University of London, E1 4NS, UK</b></td><td>('34780294', 'Lukasz Zalewski', 'lukasz zalewski')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')</td><td>[lukas|sgg]@dcs.qmul.ac.uk +</td></tr><tr><td>41000c3a3344676513ef4bfcd392d14c7a9a7599</td><td>A NOVEL APPROACH FOR GENERATING FACE +<br/>TEMPLATE USING BDA +<br/>1P.G. Student, Department of Computer Engineering, MCERC, Nashik (M.S.), India. +<br/>2Associate Professor, Department of Computer Engineering, MCERC, Nashik (M.S.), +<br/>India +</td><td>('40075681', 'Shraddha S. Shinde', 'shraddha s. shinde')<br/>('2590072', 'Anagha P. Khedkar', 'anagha p. khedkar')</td><td>shraddhashinde@gmail.com +<br/>anagha_p2@yahoo.com +</td></tr><tr><td>411ee9236095f8f5ca3b9ef18fd3381c1c68c4b8</td><td>Vol.59: e16161057, January-December 2016 +<br/>http://dx.doi.org/10.1590/1678-4324-2016161057 +<br/>ISSN 1678-4324 Online Edition +<br/>1 +<br/>Biological and Applied Sciences +<br/>BRAZILIAN ARCHIVES OF +<br/>BIOLOGY AND TECHNOLOGY +<br/>A N I N T E R N A T I O N A L J O U R N A L +<br/>An Empirical Evaluation of the Local Texture Description +<br/>Framework-Based Modified Local Directional Number +<br/>Pattern with Various Classifiers for Face Recognition +<br/><b>St. Xavier s Catholic College of Engineering, Nagercoil, India</b><br/><b>VelTech Dr. R.R. and Dr. S.R. Technical University, Chennai</b><br/><b>Manonmaniam Sundaranar University, Tirunelveli</b><br/>India. +</td><td>('9375880', 'R. Reena Rose', 'r. reena rose')</td><td></td></tr><tr><td>411318684bd2d42e4b663a37dcf0532a48f0146d</td><td>Improved Face Verification with Simple +<br/>Weighted Feature Combination +<br/><b>College of Electronics and Information Engineering, Tongji University</b><br/>4800 Cao’an Highway, Shanghai 201804, People’s Republic of China +</td><td>('1775391', 'Xinyu Zhang', 'xinyu zhang')<br/>('48566761', 'Jiang Zhu', 'jiang zhu')<br/>('34647494', 'Mingyu You', 'mingyu you')</td><td>{1510464,zhujiang,myyou}@tongji.edu.cn +</td></tr><tr><td>4140498e96a5ff3ba816d13daf148fffb9a2be3f</td><td>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>2017 IEEE 12th International Conference on Automatic Face & Gesture Recognition +<br/>Constrained Ensemble Initialization for Facial Landmark +<br/>Tracking in Video +<br/><b>Language Technology Institute, Carnegie Mellon University, Pittsburgh, PA, USA</b></td><td>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td></td></tr><tr><td>41f8477a6be9cd992a674d84062108c68b7a9520</td><td>An Automated System for Visual Biometrics +<br/>Dept. of Electrical Engineering and Computer Science +<br/><b>Northwestern University</b><br/>Evanston, IL 60208-3118 +</td><td>('2563314', 'Derek J. Shiell', 'derek j. shiell')<br/>('3271105', 'Louis H. Terry', 'louis h. terry')<br/>('2691927', 'Petar S. Aleksic', 'petar s. aleksic')<br/>('1695338', 'Aggelos K. Katsaggelos', 'aggelos k. katsaggelos')</td><td>d-shiell@northwestern.edu, l-terry@northwestern.edu, +<br/>apetar@eecs.northwestern.edu, aggk@eecs.northwestern.edu +</td></tr><tr><td>414715421e01e8c8b5743c5330e6d2553a08c16d</td><td>PoTion: Pose MoTion Representation for Action Recognition +<br/>1Inria∗ +<br/>2NAVER LABS Europe +</td><td>('2492127', 'Philippe Weinzaepfel', 'philippe weinzaepfel')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>41aa8c1c90d74f2653ef4b3a2e02ac473af61e47</td><td>Compositional Structure Learning for Action Understanding +<br/>1Department of Computer Science and Engineering, SUNY at Buffalo +<br/>2Department of Statistics, UCLA +<br/><b>University of Michigan</b><br/>October 23, 2014 +</td><td>('1856629', 'Ran Xu', 'ran xu')<br/>('1690235', 'Gang Chen', 'gang chen')<br/>('2228109', 'Caiming Xiong', 'caiming xiong')<br/>('1728624', 'Wei Chen', 'wei chen')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')</td><td></td></tr><tr><td>41ab4939db641fa4d327071ae9bb0df4a612dc89</td><td>Interpreting Face Images by Fitting a Fast +<br/>Illumination-Based 3D Active Appearance +<br/>Model +<br/>Instituto Nacional de Astrof´ısica, ´Optica y Electr´onica, +<br/>Luis Enrique Erro #1, 72840 Sta Ma. Tonantzintla. Pue., M´exico +<br/>Coordinaci´on de Ciencias Computacionales +</td><td>('2349309', 'Salvador E. Ayala-Raggi', 'salvador e. ayala-raggi')</td><td>{saraggi, robles, jcruze}@ccc.inaoep.mx +</td></tr><tr><td>41971dfbf404abeb8cf73fea29dc37b9aae12439</td><td>Detection of Facial Feature Points Using +<br/>Anthropometric Face Model +<br/> +<br/><b>Concordia University</b><br/>1455 de Maisonneuve Blvd. West, Montréal, Québec H3G 1M8, Canada +</td><td>('8018736', 'Abu Sayeed', 'abu sayeed')<br/>('1715620', 'Prabir Bhattacharya', 'prabir bhattacharya')</td><td>E-mails: a_sohai@encs.concordia.ca, prabir@ciise.concordia.ca +</td></tr><tr><td>4157e45f616233a0874f54a59c3df001b9646cd7</td><td>elifesciences.org +<br/>RESEARCH ARTICLE +<br/>Diagnostically relevant facial gestalt +<br/>information from ordinary photos +<br/><b>University of Oxford, Oxford, United Kingdom</b><br/>2Medical Research Council Functional Genomics Unit, Department of Physiology, +<br/><b>Anatomy and Genetics, University of Oxford, Oxford, United Kingdom; 3The Wellcome</b><br/><b>Trust Centre for Human Genetics, University of Oxford, Oxford, United Kingdom</b><br/><b>Medical Research Council Human Genetics Unit, Institute of Genetics and Molecular</b><br/>Medicine, Edinburgh, United Kingdom +</td><td>('4569459', 'Quentin Ferry', 'quentin ferry')<br/>('1985983', 'Julia Steinberg', 'julia steinberg')<br/>('39722750', 'Caleb Webber', 'caleb webber')<br/>('1880309', 'David R FitzPatrick', 'david r fitzpatrick')<br/>('2500371', 'Chris P Ponting', 'chris p ponting')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')<br/>('2204967', 'Christoffer Nellåker', 'christoffer nellåker')</td><td></td></tr><tr><td>41a6196f88beced105d8bc48dd54d5494cc156fb</td><td>2015 International Conference on +<br/>Communications, Signal +<br/>Processing, and their Applications +<br/>(ICCSPA 2015) +<br/>Sharjah, United Arab Emirates +<br/>17-19 February 2015 +<br/>IEEE Catalog Number: +<br/>ISBN: +<br/>CFP1574T-POD +<br/>978-1-4799-6533-5 +</td><td></td><td></td></tr><tr><td>41de109bca9343691f1d5720df864cdbeeecd9d0</td><td>Article +<br/>Facial Emotion Recognition: A Survey and +<br/>Real-World User Experiences in Mixed Reality +<br/>Received: 10 December 2017; Accepted: 26 January 2018; Published: 1 Febuary 2018 +</td><td>('38085139', 'Dhwani Mehta', 'dhwani mehta')<br/>('3655354', 'Mohammad Faridul Haque Siddiqui', 'mohammad faridul haque siddiqui')<br/>('39803999', 'Ahmad Y. Javaid', 'ahmad y. javaid')</td><td>EECS Department, The University of Toledo, Toledo, OH 43606, USA; dhwani.mehta@utoledo.edu (D.M.); +<br/>mohammadfaridulhaque.siddiqui@utoledo.edu (M.F.H.S.) +<br/>* Correspondence: ahmad.javaid@utoledo.edu; Tel.: +1-419-530-8260 +</td></tr><tr><td>41d9a240b711ff76c5448d4bf4df840cc5dad5fc</td><td>JOURNAL DRAFT, VOL. X, NO. X, APR 2013 +<br/>Image Similarity Using Sparse Representation +<br/>and Compression Distance +</td><td>('1720741', 'Tanaya Guha', 'tanaya guha')</td><td></td></tr><tr><td>419a6fca4c8d73a1e43003edc3f6b610174c41d2</td><td>A Component Based Approach Improves Classification of Discrete +<br/>Facial Expressions Over a Holistic Approach +</td><td>('2370974', 'Kenny Hong', 'kenny hong')<br/>('1716539', 'Stephan K. Chalup', 'stephan k. chalup')</td><td></td></tr><tr><td>4136a4c4b24c9c386d00e5ef5dffdd31ca7aea2c</td><td>MULTI-MODAL PERSON-PROFILES FROM BROADCAST NEWS VIDEO +<br/><b>Beckman Institute for Advanced Science and Technology</b><br/><b>University of Illinois at Urbana-Champaign</b><br/>Urbana, IL 61801 +</td><td>('1804874', 'Charlie K. Dagli', 'charlie k. dagli')<br/>('25639435', 'Sharad V. Rao', 'sharad v. rao')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>{dagli,svrao,huang}@ifp.uiuc.edu +</td></tr><tr><td>4180978dbcd09162d166f7449136cb0b320adf1f</td><td>Real-time head pose classification in uncontrolled environments +<br/>with Spatio-Temporal Active Appearance Models +<br/>∗ Matematica Aplicada i Analisi ,Universitat de Barcelona, Barcelona, Spain +<br/>+ Matematica Aplicada i Analisi, Universitat de Barcelona, Barcelona, Spain +<br/>+ Matematica Aplicada i Analisi, Universitat de Barcelona, Barcelona, Spain +</td><td>('3276130', 'Miguel Reyes', 'miguel reyes')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('1724155', 'Petia Radeva', 'petia radeva')</td><td>E-mail:mreyes@cvc.uab.es +<br/>E-mail:sergio@maia.ub.es +<br/>E-mail:petia@cvc.uab.es +</td></tr><tr><td>41b997f6cec7a6a773cd09f174cb6d2f036b36cd</td><td></td><td></td><td></td></tr><tr><td>41aa209e9d294d370357434f310d49b2b0baebeb</td><td>BEYOND CAPTION TO NARRATIVE: +<br/>VIDEO CAPTIONING WITH MULTIPLE SENTENCES +<br/><b>Grad. School of Information Science and Technology, The University of Tokyo, Japan</b></td><td>('2518695', 'Andrew Shin', 'andrew shin')<br/>('8197937', 'Katsunori Ohnishi', 'katsunori ohnishi')<br/>('1790553', 'Tatsuya Harada', 'tatsuya harada')</td><td></td></tr><tr><td>413a184b584dc2b669fbe731ace1e48b22945443</td><td>Human Pose Co-Estimation and Applications +</td><td>('31786895', 'Marcin Eichner', 'marcin eichner')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')</td><td></td></tr><tr><td>83b7578e2d9fa60d33d9336be334f6f2cc4f218f</td><td>The S-HOCK Dataset: Analyzing Crowds at the Stadium +<br/><b>University of Verona. 2Vienna Institute of Technology. 3ISTC CNR (Trento). 4University of Trento</b><br/>The topic of crowd modeling in computer vision usually assumes a sin- +<br/>gle generic typology of crowd, which is very simplistic. In this paper we +<br/>adopt a taxonomy that is widely accepted in sociology, focusing on a partic- +<br/>ular category, the spectator crowd, which is formed by people “interested in +<br/>watching something specific that they came to see” [1]. This can be found +<br/>at the stadiums, amphitheaters, cinema, etc. +<br/>In particular, we propose a +<br/>novel dataset, the Spectators Hockey (S-HOCK), which deals with 4 hockey +<br/>matches during an international tournament. +<br/>The dataset is unique in the crowd literature, and in general in the +<br/>surveillance realm. The dataset analyzes the crowd at different levels of +<br/>detail. At the highest level, it models the network of social connections +<br/>among the public (who knows whom in the neighborhood), what is the sup- +<br/>ported team and what has been the best action in the match; all of this has +<br/>been obtained by interviews at the stadium. At a medium level, spectators +<br/>are localized, and information regarding the pose of their heads and body is +<br/>given. Finally, at a lowest level, a fine grained specification of all the actions +<br/>performed by each single person is available. This information is summa- +<br/>rized by a large number of annotations collected over a year of work: more +<br/>than 100 millions of double checked annotations. This permits potentially +<br/>to deal with hundreds of tasks, some of which are documented in the full +<br/>paper. +<br/>Furthermore, the dataset is multidimensional, in the sense that offers +<br/>not only the view of the crowd (at different resolutions, with 4 cameras) but +<br/>also on the matches. This multiplies the number of possible applications that +<br/>could be assessed, investigating the reactions of the crowd to the actions of +<br/>the game, opening up to applications of summarization and content analysis. +<br/>Besides these figures, S-HOCK is significantly different from all the other +<br/>crowd datasets, since the crowd as a whole is mostly static and the motion +<br/>of each spectator is constrained within a limited space in the surrounding of +<br/>his position. +<br/>Annotation +<br/>People detection +<br/>Head detection +<br/>Head pose∗ +<br/>Body position +<br/>Posture +<br/>Locomotion +<br/>Action / Interaction +<br/>Supported team +<br/>Best action +<br/>Social relation +<br/>Typical Values +<br/>full body bounding box [x,y,width,height] +<br/>head bounding box [x,y,width,height] +<br/>left, frontal, right, away, down +<br/>sitting, standing, (locomotion) +<br/>crossed arms, hands in pocket, crossed legs . . . +<br/>walking, jumping (each jump), rising pelvis slightly up +<br/>waving arms, pointing toward game, applauding, . . . +<br/>the team supported in this game +<br/>the most exciting action of the game +<br/>If he/she did know the person seated at his/her right +<br/>Table 1: Some of the annotations provided for each person and each frame +<br/>of the videos. +<br/>Together with the annotations, in the paper we discuss issues related to +<br/>low and high level detail of the crowd analysis, namely, people detection +<br/>and head pose estimation for the low level analysis, and the spectator cate- +<br/>gorization for the high level analysis. For all of these applications, we define +<br/>the experimental protocols, promoting future comparisons. +<br/>For people detection task we provide five different baselines, from the +<br/>simplest algorithms to the state of the art method for object detection, show- +<br/>ing how in this scenario the simplest method gets very high scores. +<br/>Regarding head pose estimation, we tested two state of the art methods +<br/>which work in a low resolution domain. Furthermore, we propose two novel +<br/>approaches based on Deep Learning. In particular, we evaluate the perfor- +<br/>mance of the Convolutional Neural Network and the Stacked Auto-encoder +<br/>Neural Network architecture. Here the results are comparable with state of +<br/>the art but are obtainable at a much higher speed. +<br/>Spectator categorization is a kind of crowd segmentation, where the goal +<br/>is to find the team supported by each spectator. This task is intuitively use- +</td><td>('1843683', 'Davide Conigliaro', 'davide conigliaro')<br/>('39337007', 'Paolo Rota', 'paolo rota')<br/>('2793423', 'Francesco Setti', 'francesco setti')<br/>('1919464', 'Chiara Bassetti', 'chiara bassetti')<br/>('3058987', 'Nicola Conci', 'nicola conci')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')<br/>('1723008', 'Marco Cristani', 'marco cristani')</td><td></td></tr><tr><td>839a2155995acc0a053a326e283be12068b35cb8</td><td>Under review as a conference paper at ICLR 2016 +<br/>HANDCRAFTED LOCAL FEATURES ARE CONVOLU- +<br/>TIONAL NEURAL NETWORKS +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213, USA +</td><td>('2927024', 'Shoou-I Yu', 'shoou-i yu')<br/>('2735055', 'Ming Lin', 'ming lin')<br/>('1681921', 'Bhiksha Raj', 'bhiksha raj')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td>{lanzhzh, iyu, minglin, bhiksha, alex}@cs.cmu.edu +</td></tr><tr><td>83fd2d2d5ad6e4e153672c9b6d1a3785f754b60e</td><td>RESEARCH ARTICLE +<br/>Neuropsychiatric Genetics +<br/>Quantifying Naturalistic Social Gaze in Fragile X +<br/>Syndrome Using a Novel Eye Tracking Paradigm +<br/>and Allan L. Reiss1 +<br/>1Center for Interdisciplinary Brain Sciences Research, Department of Psychiatry and Behavioral Sciences, Stanford, California +<br/><b>Stanford University, Stanford, California</b><br/>Manuscript Received: 7 November 2014; Manuscript Accepted: 22 May 2015 +<br/>A hallmark behavioral feature of fragile X syndrome (FXS) is +<br/>the propensity for individuals with the syndrome to exhibit +<br/>significant impairments in social gaze during interactions +<br/>with others. However, previous studies employing eye tracking +<br/>methodology to investigate this phenomenon have been limited +<br/>to presenting static photographs or videos of social interactions +<br/>rather than employing a real-life social partner. To improve +<br/>upon previous studies, we used a customized eye tracking +<br/>configuration to quantify the social gaze of 51 individuals +<br/>with FXS and 19 controls, aged 14–28 years, while they engaged +<br/>in a naturalistic face-to-face social interaction with a female +<br/>experimenter. Importantly, our control group was matched to +<br/>the FXS group on age, developmental functioning, and degree of +<br/>autistic symptomatology. Results showed that participants with +<br/>FXS spent significantly less time looking at the face and had +<br/>shorter episodes (and longer inter-episodes) of social gaze than +<br/>controls. Regression analyses indicated that communication +<br/>ability predicted higher levels of social gaze in individuals +<br/>with FXS, but not in controls. Conversely, degree of autistic +<br/>symptoms predicted lower levels of social gaze in controls, but +<br/>not in individuals with FXS. Taken together, these data indicate +<br/>that naturalistic social gaze in FXS can be measured objectively +<br/>using existing eye tracking technology during face-to-face social +<br/>interactions. Given that impairments in social gaze were specific +<br/>to FXS, this paradigm could be employed as an objective and +<br/>ecologically valid outcome measure in ongoing Phase II/Phase +<br/>III clinical trials of FXS-specific interventions. +<br/><b>2015 Wiley Periodicals, Inc</b><br/>Key words: eye tracking; social gaze; autism; +<br/>syndrome +<br/>fragile X +<br/>INTRODUCTION +<br/>Children diagnosed with genetic syndromes associated with intel- +<br/>lectual and developmental disability (e.g., fragile X syndrome, +<br/>Williams syndrome) often engage in highly specific forms of aber- +<br/>rant social behavior that can interfere with everyday functioning. For +<br/>How to Cite this Article: +<br/>Hall SS, Frank MC, Pusiol GT, Farzin F, +<br/>Lightbody AA, Reiss AL. 2015. Quantifying +<br/>Naturalistic Social Gaze in Fragile X +<br/>Syndrome Using a Novel Eye Tracking +<br/>Paradigm. +<br/>Am J Med Genet Part B 9999:1–9. +<br/>example, individuals diagnosed with Williams syndrome show a +<br/>particular form of hypersociability in which they actively seek out +<br/>social interactions with others [Jones et al., 2000; Frigerio et al., +<br/>2006]. Conversely, children with fragile X syndrome (FXS) com- +<br/>monly show deficits in social gaze behavior in which interactions +<br/>with others are actively avoided [Cohen et al., 1988; Cohen et al., +<br/>1989; Cohen et al., 1991; Hall et al., 2006; Hall et al., 2009]. These +<br/>contrasting behavioral phenotypes have been considered useful +<br/>and important models for investigations examining the interplay +<br/>between genes and environment [Kennedy et al., 2001; Schroeder +<br/>et al., 2001]. +<br/>FXS is a particularly interesting model of potential gene-envi- +<br/>ronment interactions because it is a “single-gene” disorder. The +<br/>disease affects approximately 1 in 3,000 individuals in the United +<br/>States (approx. 100,000 people) and is the most common known +<br/>form of inherited intellectual disability [Hagerman, 2008]. First +<br/>described by Martin and Bell in 1943 as a “pedigree of mental defect +<br/>showing sex linkage” [Martin and Bell, 1943], FXS is caused by +<br/>mutations to the FMR1 gene at locus 27.3 on the long arm of the X +<br/>chromosome [Verkerk et al., 1991]. Excessive methylation of the +<br/>gene results in reduced or absent Fragile X Mental Retardation +<br/>Protein (FMRP), a key protein involved in synaptic plasticity and +<br/>Grant sponsor: NIH grants; Grant numbers: MH050047, MH081998. +<br/>Correspondence to: +<br/>Scott S. Hall, PhD, Department of Psychiatry and Behavioral Sciences, +<br/><b>Rm 1365, Stanford University, 401 Quarry Road, Stanford, CA</b><br/>Article first published online in Wiley Online Library +<br/>(wileyonlinelibrary.com): 00 Month 2015 +<br/>DOI 10.1002/ajmg.b.32331 +<br/><b>2015 Wiley Periodicals, Inc</b></td><td>('4708625', 'Faraz Farzin', 'faraz farzin')</td><td>E-mail: hallss@stanford.edu +</td></tr><tr><td>83ca4cca9b28ae58f461b5a192e08dffdc1c76f3</td><td>DETECTING EMOTIONAL STRESS FROM FACIAL EXPRESSIONS FOR DRIVING SAFETY +<br/>Signal Processing Laboratory (LTS5), +<br/>´Ecole Polytechnique F´ed´erale de Lausanne, Switzerland +</td><td>('1697965', 'Hua Gao', 'hua gao')<br/>('1710257', 'Jean-Philippe Thiran', 'jean-philippe thiran')</td><td></td></tr><tr><td>8356832f883207187437872742d6b7dc95b51fde</td><td>Adversarial Perturbations Against Real-Time Video +<br/>Classification Systems +<br/><b>University of California, Riverside</b><br/><b>University of California, Riverside</b><br/><b>University of California, Riverside</b><br/>Riverside, California +<br/>Riverside, California +<br/><b>University of California, Riverside</b><br/>Riverside, California +<br/>Riverside, California +<br/><b>University of California, Riverside</b><br/>Riverside, California +<br/>Amit K. Roy Chowdhury +<br/><b>University of California, Riverside</b><br/>Riverside, California +<br/>United States Army Research +<br/>Laboratory +</td><td>('26576993', 'Shasha Li', 'shasha li')<br/>('2252367', 'Chengyu Song', 'chengyu song')<br/>('1718484', 'Ajaya Neupane', 'ajaya neupane')<br/>('49616225', 'Sujoy Paul', 'sujoy paul')<br/>('38774813', 'Srikanth V. Krishnamurthy', 'srikanth v. krishnamurthy')<br/>('1703726', 'Ananthram Swami', 'ananthram swami')</td><td>sli057@ucr.edu +<br/>csong@cs.ucr.edu +<br/>ajaya@ucr.edu +<br/>spaul003@ucr.edu +<br/>krish@cs.ucr.edu +<br/>amitrc@ece.ucr.edu +<br/>ananthram.swami.civ@mail.mil +</td></tr><tr><td>831fbef657cc5e1bbf298ce6aad6b62f00a5b5d9</td><td></td><td></td><td></td></tr><tr><td>835e510fcf22b4b9097ef51b8d0bb4e7b806bdfd</td><td>Unsupervised Learning of Sequence Representations by +<br/>Autoencoders +<br/><b>aPattern Recognition Laboratory, Delft University of Technology</b></td><td>('1678473', 'Wenjie Pei', 'wenjie pei')</td><td></td></tr><tr><td>832e1d128059dd5ed5fa5a0b0f021a025903f9d5</td><td>Pairwise Conditional Random Forests for Facial Expression Recognition +<br/>S´everine Dubuisson1 +<br/>1 Sorbonne Universit´es, UPMC Univ Paris 06, CNRS, ISIR UMR 7222, 4 place Jussieu 75005 Paris +</td><td>('3190846', 'Arnaud Dapogny', 'arnaud dapogny')<br/>('2521061', 'Kevin Bailly', 'kevin bailly')</td><td>arnaud.dapogny@isir.upmc.fr +<br/>kevin.bailly@isir.upmc.fr +<br/>severine.dubuisson@isir.upmc.fr +</td></tr><tr><td>83e093a07efcf795db5e3aa3576531d61557dd0d</td><td>Facial Landmark Localization using Robust +<br/>Relationship Priors and Approximative Gibbs +<br/>Sampling +<br/>Institut f¨ur Informationsverarbeitung (tnt) +<br/>Leibniz Universit¨at Hannover, Germany +</td><td>('35033145', 'Karsten Vogt', 'karsten vogt')</td><td>{vogt, omueller, ostermann}@tnt.uni-hannover.de +</td></tr><tr><td>831d661d657d97a07894da8639a048c430c5536d</td><td>Weakly Supervised Facial Analysis with Dense Hyper-column Features +<br/>CyLab Biometrics Center and the Department of Electrical and Computer Engineering, +<br/><b>Carnegie Mellon University, Pittsburgh, PA, USA</b></td><td>('3117715', 'Chenchen Zhu', 'chenchen zhu')<br/>('3049981', 'Yutong Zheng', 'yutong zheng')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('6131978', 'T. Hoang Ngan Le', 't. hoang ngan le')<br/>('2043374', 'Chandrasekhar Bhagavatula', 'chandrasekhar bhagavatula')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td>{chenchez, yutongzh, kluu, thihoanl, cbhagava}@andrew.cmu.edu, msavvid@ri.cmu.edu +</td></tr><tr><td>83b4899d2899dd6a8d956eda3c4b89f27f1cd308</td><td>1-4244-1437-7/07/$20.00 ©2007 IEEE +<br/>I - 377 +<br/>ICIP 2007 +</td><td></td><td></td></tr><tr><td>83295bce2340cb87901499cff492ae6ff3365475</td><td>Deep Multi-Center Learning for Face Alignment +<br/><b>Shanghai Jiao Tong University, China</b><br/><b>School of Computer Science and Software Engineering, East China Normal University, China</b></td><td>('3403352', 'Zhiwen Shao', 'zhiwen shao')<br/>('7296339', 'Hengliang Zhu', 'hengliang zhu')<br/>('1767677', 'Xin Tan', 'xin tan')<br/>('2107352', 'Yangyang Hao', 'yangyang hao')<br/>('8452947', 'Lizhuang Ma', 'lizhuang ma')</td><td>{shaozhiwen, hengliang zhu, tanxin2017, haoyangyang2014}@sjtu.edu.cn, ma-lz@cs.sjtu.edu.cn +</td></tr><tr><td>83e96ed8a4663edaa3a5ca90b7ce75a1bb595b05</td><td>ARANDJELOVI´C:RECOGNITIONFROMAPPEARANCESUBSPACESACROSSSCALE +<br/>Recognition from Appearance Subspaces +<br/>Across Image Sets of Variable Scale +<br/>Ognjen Arandjelovi´c +<br/>http://mi.eng.cam.ac.uk/~oa214 +<br/><b>Trinity College</b><br/><b>University of Cambridge</b><br/>CB2 1TQ, UK +</td><td></td><td></td></tr><tr><td>830e5b1043227fe189b3f93619ef4c58868758a7</td><td></td><td></td><td></td></tr><tr><td>8323af714efe9a3cadb31b309fcc2c36c8acba8f</td><td>Automatic Real-Time +<br/>Facial Expression Recognition +<br/>for Signed Language Translation +<br/>A thesis submitted in partial fulfillment of the requirements for the de- +<br/>gree of Magister Scientiae in the Department of Computer Science, +<br/><b>University of the Western Cape</b><br/>May 2006 +</td><td>('1775637', 'Jacob Richard Whitehill', 'jacob richard whitehill')</td><td></td></tr><tr><td>831226405bb255527e9127b84e8eaedd7eb8e9f9</td><td>ORIGINAL RESEARCH +<br/>published: 04 January 2017 +<br/>doi: 10.3389/fnins.2016.00594 +<br/>A Motion-Based Feature for +<br/>Event-Based Pattern Recognition +<br/>Centre National de la Recherche Scientifique, Institut National de la Santé Et de la Recherche Médicale, Institut de la Vision, +<br/><b>Sorbonne Universit s, UPMC University Paris 06, Paris, France</b><br/>This paper introduces an event-based luminance-free feature from the output of +<br/>asynchronous event-based neuromorphic retinas. The feature consists in mapping the +<br/>distribution of the optical flow along the contours of the moving objects in the visual +<br/>scene into a matrix. Asynchronous event-based neuromorphic retinas are composed +<br/>of autonomous pixels, each of them asynchronously generating “spiking” events that +<br/>encode relative changes in pixels’ illumination at high temporal resolutions. The optical +<br/>flow is computed at each event, and is integrated locally or globally in a speed and +<br/>direction coordinate frame based grid, using speed-tuned temporal kernels. The latter +<br/>ensures that the resulting feature equitably represents the distribution of the normal +<br/>motion along the current moving edges, whatever their respective dynamics. The +<br/>usefulness and the generality of the proposed feature are demonstrated in pattern +<br/>recognition applications: local corner detection and global gesture recognition. +<br/>Keywords: neuromorphic sensor, event-driven vision, pattern recognition, motion-based feature, speed-tuned +<br/>integration time, histogram of oriented optical flow, corner detection, gesture recognition +<br/>1. INTRODUCTION +<br/>In computer vision, a feature is a more or less compact representation of visual information that is +<br/>relevant to solve a task related to a given application (see Laptev, 2005; Mikolajczyk and Schmid, +<br/>2005; Mokhtarian and Mohanna, 2006; Moreels and Perona, 2007; Gil et al., 2010; Dickscheid et al., +<br/>2011; Gauglitz et al., 2011). Building a feature consists in encoding information contained in the +<br/>visual scene (global approach) or in a neighborhood of a point (local approach). It can represent +<br/>static information (e.g., shape of an object, contour, etc.), dynamic information (e.g., speed and +<br/>direction at the point, dynamic deformations, etc.) or both simultaneously. +<br/>In this article, we propose a motion-based feature computed on visual information provided by +<br/>asynchronous image sensors known as neuromorphic retinas (see Delbrück et al., 2010; Posch, +<br/>2015). These cameras provide visual information as asynchronous event-based streams while +<br/>conventional cameras output it as synchronous frame-based streams. The ATIS (“Asynchronous +<br/>Time-based Image Sensor,” Posch et al., 2010; Posch, 2015), one of the neuromorphic visual +<br/>sensors used in this work, is a time-domain encoding image sensor with QVGA resolution. It +<br/>contains an array of fully autonomous pixels that combine an illuminance change detector circuit, +<br/>associated to the PD1 photodiode, see Figure 1A and a conditional exposure measurement block, +<br/>associated to the PD2 photodiode. The change detector individually and asynchronously initiates +<br/>the measurement of an exposure/gray scale value only if a brightness change of a certain magnitude +<br/>has been detected in the field-of-view of the respective pixel, as shown in the functional diagram +<br/>of the ATIS pixel in Figures 1B, 2. The exposure measurement circuit encodes the absolute +<br/>instantaneous pixel illuminance into the timing of asynchronous event pulses, more precisely +<br/>Edited by: +<br/>Tobi Delbruck, +<br/>ETH Zurich, Switzerland +<br/>Reviewed by: +<br/>Dan Hammerstrom, +<br/><b>Portland State University, USA</b><br/>Rodrigo Alvarez-Icaza, +<br/>IBM, USA +<br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Neuromorphic Engineering, +<br/>a section of the journal +<br/>Frontiers in Neuroscience +<br/>Received: 07 September 2016 +<br/>Accepted: 13 December 2016 +<br/>Published: 04 January 2017 +<br/>Citation: +<br/>Clady X, Maro J-M, Barré S and +<br/>Benosman RB (2017) A Motion-Based +<br/>Feature for Event-Based Pattern +<br/>Recognition. Front. Neurosci. 10:594. +<br/>doi: 10.3389/fnins.2016.00594 +<br/>Frontiers in Neuroscience | www.frontiersin.org +<br/>January 2017 | Volume 10 | Article 594 +</td><td>('1804748', 'Xavier Clady', 'xavier clady')<br/>('24337536', 'Jean-Matthieu Maro', 'jean-matthieu maro')<br/>('2133648', 'Sébastien Barré', 'sébastien barré')<br/>('1750848', 'Ryad B. Benosman', 'ryad b. benosman')<br/>('1804748', 'Xavier Clady', 'xavier clady')</td><td>xavier.clady@upmc.fr +</td></tr><tr><td>83fd5c23204147844a0528c21e645b757edd7af9</td><td>USDOT Number Localization and Recognition From Vehicle Side-View NIR +<br/>Images +<br/><b>Palo Alto Research Center (PARC</b><br/>800 Phillips Rd. Webster NY 14580 +</td><td>('2415287', 'Orhan Bulan', 'orhan bulan')<br/>('1732789', 'Safwan Wshah', 'safwan wshah')<br/>('3195726', 'Ramesh Palghat', 'ramesh palghat')<br/>('2978081', 'Vladimir Kozitsky', 'vladimir kozitsky')<br/>('34801919', 'Aaron Burry', 'aaron burry')</td><td>orhan.bulan,safwan.wshah,ramesh.palghat,vladimir.kozitsky,aaron.burry@parc.com +</td></tr><tr><td>8384e104796488fa2667c355dd15b65d6d5ff957</td><td>A Discriminative Latent Model of Image Region and +<br/>Object Tag Correspondence +<br/>Department of Computer Science +<br/><b>University of Illinois at Urbana-Champaign</b><br/>School of Computing Science +<br/><b>Simon Fraser University</b></td><td>('40457160', 'Yang Wang', 'yang wang')<br/>('10771328', 'Greg Mori', 'greg mori')</td><td>yangwang@uiuc.edu +<br/>mori@cs.sfu.ca +</td></tr><tr><td>8323529cf37f955fb3fc6674af6e708374006a28</td><td>Evaluation of Face Resolution for Expression Analysis +<br/><b>IBM T. J. Watson Research Center</b><br/>PO Box 704, Yorktown Heights, NY 10598 +</td><td>('40383812', 'Ying-li Tian', 'ying-li tian')</td><td>Email: yltian@us.ibm.com +</td></tr><tr><td>8395cf3535a6628c3bdc9b8d0171568d551f5ff0</td><td>Entropy Non-increasing Games for the +<br/>Improvement of Dataflow Programming +<br/>Norbert B´atfai, Ren´at´o Besenczi, Gerg˝o Bogacsovics, +<br/>February 16, 2017 +</td><td>('9544536', 'Fanny Monori', 'fanny monori')</td><td></td></tr><tr><td>83ac942d71ba908c8d76fc68de6173151f012b38</td><td></td><td></td><td></td></tr><tr><td>834f5ab0cb374b13a6e19198d550e7a32901a4b2</td><td>Face Translation between Images and Videos using Identity-aware CycleGAN +<br/>†Computer Vision Lab, ETH Zurich, Switzerland +<br/>‡VISICS, KU Leuven, Belgium +</td><td>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('2208488', 'Bernhard Kratzwald', 'bernhard kratzwald')<br/>('35268081', 'Danda Pani Paudel', 'danda pani paudel')<br/>('1839268', 'Jiqing Wu', 'jiqing wu')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{zhiwu.huang, paudel, jwu, vangool}@vision.ee.ethz.ch, bkratzwald@ethz.ch +</td></tr><tr><td>8320dbdd3e4712cca813451cd94a909527652d63</td><td>EAR BIOMETRICS +<br/>and Wilhelm Burger +<br/><b>Johannes Kepler University(cid:1) Institute of Systems Science(cid:1) A(cid:2) +</td><td>('12811570', 'Mark Burge', 'mark burge')</td><td></td></tr><tr><td>837e99301e00c2244023a8a48ff98d7b521c93ac</td><td>Local Feature Evaluation for a Constrained +<br/>Local Model Framework +<br/><b>Graduate School of Engineering, Tottori University</b><br/>101 Minami 4-chome, Koyama-cho, Tottori 680-8550, Japan +</td><td>('1770332', 'Maiya Hori', 'maiya hori')<br/>('48532779', 'Shogo Kawai', 'shogo kawai')<br/>('2020088', 'Hiroki Yoshimura', 'hiroki yoshimura')<br/>('1679437', 'Yoshio Iwai', 'yoshio iwai')</td><td>hori@ike.tottori-u.ac.jp +</td></tr><tr><td>834b15762f97b4da11a2d851840123dbeee51d33</td><td>Landmark-free smile intensity estimation +<br/>IMAGO Research Group - Universidade Federal do Paran´a +<br/>Fig. 1. Overview of our method for smile intensity estimation +</td><td>('1800955', 'Olga R. P. Bellon', 'olga r. p. bellon')</td><td>{julio.batista,olga,luciano}@ufpr.br +</td></tr><tr><td>833f6ab858f26b848f0d747de502127406f06417</td><td>978-1-4244-5654-3/09/$26.00 ©2009 IEEE +<br/>61 +<br/>ICIP 2009 +</td><td></td><td></td></tr><tr><td>8334da483f1986aea87b62028672836cb3dc6205</td><td>Fully Associative Patch-based 1-to-N Matcher for Face Recognition +<br/>Computational Biomedicine Lab +<br/><b>University of Houston</b></td><td>('39089616', 'Lingfeng Zhang', 'lingfeng zhang')<br/>('1706204', 'Ioannis A. Kakadiaris', 'ioannis a. kakadiaris')</td><td>{lzhang34, ioannisk}@uh.edu +</td></tr><tr><td>831b4d8b0c0173b0bac0e328e844a0fbafae6639</td><td>Consensus-Driven Propagation in +<br/>Massive Unlabeled Data for Face Recognition +<br/><b>CUHK - SenseTime Joint Lab, The Chinese University of Hong Kong</b><br/>2 SenseTime Group Limited +<br/><b>Nanyang Technological University</b></td><td>('31818765', 'Xiaohang Zhan', 'xiaohang zhan')<br/>('3243969', 'Ziwei Liu', 'ziwei liu')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('1807606', 'Dahua Lin', 'dahua lin')<br/>('1717179', 'Chen Change Loy', 'chen change loy')</td><td>{zx017, zwliu, dhlin}@ie.cuhk.edu.hk +<br/>yanjunjie@sensetime.com +<br/>ccloy@ieee.org +</td></tr><tr><td>8309e8f27f3fb6f2ac1b4343a4ad7db09fb8f0ff</td><td>Generic versus Salient Region-based Partitioning +<br/>for Local Appearance Face Recognition +<br/>Computer Science Depatment, Universit¨at Karlsruhe (TH) +<br/>Am Fasanengarten 5, Karlsruhe 76131, Germany +<br/>http://isl.ira.uka.de/cvhci +</td><td>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>{ekenel,stiefel}@ira.uka.de +</td></tr><tr><td>1b02b9413b730b96b91d16dcd61b2420aef97414</td><td>Détection de marqueurs affectifs et attentionnels de +<br/>personnes âgées en interaction avec un robot +<br/>To cite this version: +<br/>avec un robot. +<br/>Intelligence artificielle [cs.AI]. Université Paris-Saclay, 2015. Français. <NNT : +<br/>2015SACLS081>. <tel-01280505> +<br/>HAL Id: tel-01280505 +<br/>https://tel.archives-ouvertes.fr/tel-01280505 +<br/>Submitted on 29 Feb 2016 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('47829802', 'Fan Yang', 'fan yang')<br/>('47829802', 'Fan Yang', 'fan yang')</td><td></td></tr><tr><td>1b55c4e804d1298cbbb9c507497177014a923d22</td><td>Incremental Class Representation +<br/>Learning for Face Recognition +<br/>Degree’s Thesis +<br/>Audiovisual Systems Engineering +<br/>Author: +<br/>Universitat Politècnica de Catalunya (UPC) +<br/>2016 - 2017 +</td><td>('2470219', 'Elisa Sayrol', 'elisa sayrol')<br/>('2585946', 'Josep Ramon Morros', 'josep ramon morros')</td><td></td></tr><tr><td>1b635f494eff2e5501607ebe55eda7bdfa8263b8</td><td>USC at THUMOS 2014 +<br/><b>University of Southern California, Institute for Robotics and Intelligent Systems</b><br/>Los Angeles, CA 90089, USA +</td><td>('1726241', 'Chen Sun', 'chen sun')<br/>('27735100', 'Ram Nevatia', 'ram nevatia')</td><td></td></tr><tr><td>1b6394178dbc31d0867f0b44686d224a19d61cf4</td><td>EPML: Expanded Parts based Metric Learning for +<br/>Occlusion Robust Face Verification +<br/>To cite this version: +<br/>for Occlusion Robust Face Verification. Asian Conference on Computer Vision, Nov 2014, -, +<br/>Singapore. pp.1-15, 2014. <hal-01070657> +<br/>HAL Id: hal-01070657 +<br/>https://hal.archives-ouvertes.fr/hal-01070657 +<br/>Submitted on 2 Oct 2014 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('2515597', 'Gaurav Sharma', 'gaurav sharma')<br/>('2515597', 'Gaurav Sharma', 'gaurav sharma')</td><td></td></tr><tr><td>1bd50926079e68a6e32dc4412e9d5abe331daefb</td><td></td><td></td><td></td></tr><tr><td>1bdef21f093c41df2682a07f05f3548717c7a3d1</td><td>Towards Automated Classification of Emotional Facial Expressions +<br/>1Department of Mathematics and Computer Science, 2Department of Psychology +<br/><b>Rutgers University Newark, 101 Warren St., Newark, NJ, 07102 USA</b></td><td></td><td>Lewis J. Baker (lewis.j.baker@rutgers.edu)1, Vanessa LoBue (vlobue@rutgers.edu)2, +<br/>Elizabeth Bonawitz (elizabeth.bonawitz@rutgers.edu)2, & Patrick Shafto (patrick.shafto@gmail.com)1 +</td></tr><tr><td>1b150248d856f95da8316da868532a4286b9d58e</td><td>Analyzing 3D Objects in Cluttered Images +<br/>UC Irvine +<br/>UC Irvine +</td><td>('1888731', 'Mohsen Hejrati', 'mohsen hejrati')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td>shejrati@ics.uci.edu +<br/>dramanan@ics.uci.edu +</td></tr><tr><td>1be498d4bbc30c3bfd0029114c784bc2114d67c0</td><td>Age and Gender Estimation of Unfiltered Faces +</td><td>('2037829', 'Eran Eidinger', 'eran eidinger')<br/>('1792038', 'Roee Enbar', 'roee enbar')<br/>('1756099', 'Tal Hassner', 'tal hassner')</td><td></td></tr><tr><td>1bbec7190ac3ba34ca91d28f145e356a11418b67</td><td>Action Recognition with Dynamic Image Networks +<br/>Citation for published version: +<br/>Bilen, H, Fernando, B, Gravves, E & Vedaldi, A 2017, 'Action Recognition with Dynamic Image Networks' +<br/>IEEE Transactions on Pattern Analysis and Machine Intelligence. DOI: 10.1109/TPAMI.2017.2769085 +<br/>Digital Object Identifier (DOI): +<br/>10.1109/TPAMI.2017.2769085 +<br/>Link: +<br/>Link to publication record in Edinburgh Research Explorer +<br/>Document Version: +<br/>Peer reviewed version +<br/>Published In: +<br/>IEEE Transactions on Pattern Analysis and Machine Intelligence +<br/>General rights +<br/>Copyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s) +<br/>and / or other copyright owners and it is a condition of accessing these publications that users recognise and +<br/>abide by the legal requirements associated with these rights. +<br/>Take down policy +<br/><b>The University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer</b><br/>content complies with UK legislation. If you believe that the public display of this file breaches copyright please +<br/>investigate your claim. +<br/>Download date: 25. Dec. 2017 +<br/> Edinburgh Research Explorer </td><td></td><td>contact openaccess@ed.ac.uk providing details, and we will remove access to the work immediately and +</td></tr><tr><td>1b3587363d37dd197b6adbcfa79d49b5486f27d8</td><td>Multimodal Grounding for Language Processing +<br/><b>Language Technology Lab, University of Duisburg-Essen</b><br/>(cid:52) Ubiquitous Knowledge Processing Lab (UKP) and Research Training Group AIPHES +<br/>Department of Computer Science, Technische Universit¨at Darmstadt +<br/>www.ukp.tu-darmstadt.de +</td><td>('2752573', 'Lisa Beinborn', 'lisa beinborn')<br/>('25080314', 'Teresa Botschen', 'teresa botschen')<br/>('1730400', 'Iryna Gurevych', 'iryna gurevych')</td><td></td></tr><tr><td>1b5875dbebc76fec87e72cee7a5263d325a77376</td><td>Learnt Quasi-Transitive Similarity for Retrieval from Large Collections of Faces +<br/>Ognjen Arandjelovi´c +<br/><b>University of St Andrews, United Kingdom</b></td><td></td><td>ognjen.arandjelovic@gmail.com +</td></tr><tr><td>1bdfb3deae6e6c0df6537efcd1d7edcb4d7a96e9</td><td>Groupwise Constrained Reconstruction for Subspace Clustering +<br/>Ke Zhang† +<br/><b>School of Computer Science, Fudan University, Shanghai, 200433, China</b><br/><b>QCIS Centre, FEIT, University of Technology, Sydney, NSW 2007, Australia</b></td><td>('1736607', 'Ruijiang Li', 'ruijiang li')<br/>('1713520', 'Bin Li', 'bin li')<br/>('1751513', 'Cheng Jin', 'cheng jin')<br/>('1713721', 'Xiangyang Xue', 'xiangyang xue')</td><td>rjli@fudan.edu.cn +<br/>bin.li-1@uts.edu.au +<br/>k_zhang@fudan.edu.cn +<br/>jc@fudan.edu.cn +<br/>xyxue@fudan.edu.cn +</td></tr><tr><td>1b300a7858ab7870d36622a51b0549b1936572d4</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TIP.2016.2537215, IEEE +<br/>Transactions on Image Processing +<br/>Dynamic Facial Expression Recognition with Atlas +<br/>Construction and Sparse Representation +</td><td>('1734663', 'Yimo Guo', 'yimo guo')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')</td><td></td></tr><tr><td>1b90507f02967ff143fce993a5abbfba173b1ed0</td><td>Image Processing Theory, Tools and Applications +<br/>Gradient-DCT (G-DCT) Descriptors +<br/><b>Technical University of Ostrava, FEECS</b><br/>17. listopadu 15, 708 33 Ostrava-Poruba, Czech Republic +</td><td>('2467747', 'Radovan Fusek', 'radovan fusek')<br/>('2557877', 'Eduard Sojka', 'eduard sojka')</td><td>e-mail: radovan.fusek@vsb.cz, eduard.sojka@vsb.cz +</td></tr><tr><td>1b794b944fd462a2742b6c2f8021fecc663004c9</td><td>A Hierarchical Probabilistic Model for Facial Feature Detection +<br/><b>Rensselaer Polytechnic Institute</b></td><td>('1746738', 'Yue Wu', 'yue wu')<br/>('2860279', 'Ziheng Wang', 'ziheng wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>{wuy9,wangz10,jiq}@rpi.edu +</td></tr><tr><td>1b7ae509c8637f3c123cf6151a3089e6b8a0d5b2</td><td>From Few to Many: Generative Models for Recognition +<br/>Under Variable Pose and Illumination +<br/>Departments of Electrical Engineering +<br/><b>Beckman Institute</b><br/>and Computer Science +<br/><b>Yale University</b><br/>New Haven, CT +<br/><b>University of Illinois, Urbana-Champaign</b><br/>Urbana, IL +</td><td>('3230391', 'Athinodoros S. Georghiades', 'athinodoros s. georghiades')<br/>('1765887', 'David J. Kriegman', 'david j. kriegman')</td><td></td></tr><tr><td>1b41d4ffb601d48d7a07dbbae01343f4eb8cc38c</td><td>Exploiting Temporal Information for DCNN-based Fine-Grained Object Classification +<br/>Australian Centre for Robotic Vision, Australia +<br/><b>Queensland University of Technology, Australia</b><br/>Data61, CSIRO, Australia +<br/><b>University of Queensland, Australia</b><br/><b>University of Adelaide, Australia</b></td><td>('1808390', 'ZongYuan Ge', 'zongyuan ge')<br/>('1763662', 'Chris McCool', 'chris mccool')<br/>('1781182', 'Conrad Sanderson', 'conrad sanderson')<br/>('1722767', 'Peng Wang', 'peng wang')<br/>('2161037', 'Lingqiao Liu', 'lingqiao liu')</td><td></td></tr><tr><td>1b1173a3fb33f9dfaf8d8cc36eb0bf35e364913d</td><td>DICTA +<br/>#147 +<br/>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>045 +<br/>046 +<br/>047 +<br/>048 +<br/>049 +<br/>050 +<br/>051 +<br/>052 +<br/>053 +<br/>DICTA 2010 Submission #147. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +<br/>Registration Invariant Representations for Expression Detection +<br/>Anonymous DICTA submission +<br/>Paper ID 147 +</td><td></td><td></td></tr><tr><td>1b0a071450c419138432c033f722027ec88846ea</td><td>Windsor Oceanico Hotel, Rio de Janeiro, Brazil, November 1-4, 2016 +<br/>978-1-5090-1889-5/16/$31.00 ©2016 IEEE +<br/>649 +</td><td></td><td></td></tr><tr><td>1b60b8e70859d5c85ac90510b370b501c5728620</td><td>Using Detailed Independent 3D Sub-models to Improve +<br/>Facial Feature Localisation and Pose Estimation +<br/><b>Imaging Science and Biomedical Engineering, The University of Manchester, UK</b></td><td>('1753123', 'Angela Caunce', 'angela caunce')</td><td></td></tr><tr><td>1b3b01513f99d13973e631c87ffa43904cd8a821</td><td>HMM RECOGNITION OF EXPRESSIONS IN UNRESTRAINED VIDEO INTERVALS +<br/>Universitat Politècnica de Catalunya, Barcelona, Spain +</td><td>('3067467', 'José Luis Landabaso', 'josé luis landabaso')<br/>('1767549', 'Montse Pardàs', 'montse pardàs')<br/>('2868058', 'Antonio Bonafonte', 'antonio bonafonte')</td><td></td></tr><tr><td>1bc214c39536c940b12c3a2a6b78cafcbfddb59a</td><td></td><td></td><td></td></tr><tr><td>1bc9aaa41c08bbd0c01dd5d7d7ebf3e48ae78113</td><td>Article +<br/>k-Same-Net: k-Anonymity with Generative Deep +<br/>Neural Networks for Face Deidentification † +<br/><b>Faculty of Computer and Information Science, University of Ljubljana, Ve cna pot 113, SI-1000 Ljubljana</b><br/><b>Faculty of Electrical Engineering, University of Ljubljana, Tr a ka cesta 25, SI-1000 Ljubljana, Slovenia</b><br/>† This paper is an extended version of our paper published in Meden B.; Emeršiˇc Ž.; Štruc V.; Peer P. +<br/>k-Same-Net: Neural-Network-Based Face Deidentification. In the Proceedings of the International +<br/>Conference and Workshop on Bioinspired Intelligence (IWOBI), Funchal Madeira, Portugal, 10–12 July 2017. +<br/>Received: 1 December 2017 ; Accepted: 9 January 2018; Published: 13 January 2018 +</td><td>('34862665', 'Peter Peer', 'peter peer')</td><td>Slovenia; ziga.emersic@fri.uni-lj.si (Z.E.); peter.peer@fri.uni-lj.si (P.P.) +<br/>vitomir.struc@fe.uni-lj.si +<br/>* Correspondence: blaz.meden@fri.uni-lj.si; Tel.: +386-1-479-8245 +</td></tr><tr><td>1be18a701d5af2d8088db3e6aaa5b9b1d54b6fd3</td><td>ENHANCEMENT OF FAST FACE DETECTION ALGORITHM BASED ON A CASCADE OF +<br/>DECISION TREES +<br/>Commission II, WG II/5 +<br/>KEY WORDS: Face Detection, Cascade Algorithm, Decision Trees. +</td><td>('40293010', 'V. V. Khryashchev', 'v. v. khryashchev')<br/>('32423989', 'A. A. Lebedev', 'a. a. lebedev')<br/>('3414890', 'A. L. Priorov', 'a. l. priorov')</td><td>a YSU, Yaroslavl, Russia - lebedevdes@gmail.com, (vhr, andcat)@yandex.ru +</td></tr><tr><td>1b79628af96eb3ad64dbb859dae64f31a09027d5</td><td></td><td></td><td></td></tr><tr><td>1bcbf2a4500d27d036e0f9d36d7af71c72f8ab61</td><td>Computer Vision and Pattern Recognition 2005 +<br/>Recognizing Facial Expression: Machine Learning and Application to +<br/>Spontaneous Behavior +<br/><b>Institute for Neural Computation, University of California, San Diego</b><br/>Ian Fasel1, Javier Movellan1 +<br/><b>Rutgers University, New Brunswick, NJ</b></td><td>('2218905', 'Marian Stewart Bartlett', 'marian stewart bartlett')<br/>('2724380', 'Gwen Littlewort', 'gwen littlewort')<br/>('2767464', 'Claudia Lainscsek', 'claudia lainscsek')</td><td>mbartlett@ucsd.edu +</td></tr><tr><td>1b70bbf7cdfc692873ce98dd3c0e191580a1b041</td><td> International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056 +<br/> Volume: 03 Issue: 10 | Oct -2016 www.irjet.net p-ISSN: 2395-0072 +<br/>Enhancing Performance of Face Recognition +<br/>System Using Independent Component Analysis +<br/><b>Student, Computer Science, Shah and Anchor Kuttchi Engineering College, Mumbai, India</b><br/><b>Guide, HOD, Computer Science, Shah and Anchor Kuttchi Engineering College, Mumbai, India</b><br/><b>Co-Guide, Computer Science, Shah and Anchor Kuttchi Engineering College, Mumbai, India</b><br/>---------------------------------------------------------------------***--------------------------------------------------------------------- +<br/>cards, tokens and keys. Biometric based methods examine +</td><td>('32330340', 'Manimala Mahato', 'manimala mahato')</td><td></td></tr><tr><td>1b71d3f30238cb6621021a95543cce3aab96a21b</td><td>Fine-grained Video Classification and Captioning +<br/><b>University of Toronto1, Twenty Billion Neurons</b></td><td>('2454800', 'Farzaneh Mahdisoltani', 'farzaneh mahdisoltani')<br/>('40586522', 'Guillaume Berger', 'guillaume berger')<br/>('3462264', 'Waseem Gharbieh', 'waseem gharbieh')<br/>('1710604', 'Roland Memisevic', 'roland memisevic')</td><td>1 {farzaneh, fleet}@cs.toronto.edu, {firstname.lastname}@twentybn.com +</td></tr><tr><td>1b4f6f73c70353869026e5eec1dd903f9e26d43f</td><td>Robust Subjective Visual Property Prediction +<br/>from Crowdsourced Pairwise Labels +</td><td>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('1697755', 'Timothy M. Hospedales', 'timothy m. hospedales')<br/>('1700927', 'Tao Xiang', 'tao xiang')<br/>('3081531', 'Jiechao Xiong', 'jiechao xiong')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('1717863', 'Yizhou Wang', 'yizhou wang')<br/>('1746280', 'Yuan Yao', 'yuan yao')</td><td></td></tr><tr><td>1bc23c771688109bed9fd295ce82d7e702726327</td><td></td><td>('1706007', 'Jianchao Yang', 'jianchao yang')</td><td></td></tr><tr><td>1bad8a9640cdbc4fe7de12685651f44c4cff35ce</td><td>THETIS: THree Dimensional Tennis Shots +<br/>A human action dataset +<br/>Sofia Gourgari +<br/>Konstantinos Karpouzis +<br/>Stefanos Kollias +<br/><b>National Technical University of Athens</b><br/>Image Video and Multimedia Systems Laboratory +</td><td>('2123731', 'Georgios Goudelis', 'georgios goudelis')</td><td></td></tr><tr><td>1b589016fbabe607a1fb7ce0c265442be9caf3a9</td><td></td><td></td><td></td></tr><tr><td>1be0ce87bb5ba35fa2b45506ad997deef6d6a0a8</td><td>EXMOVES: Classifier-based Features for Scalable Action Recognition +<br/><b>Dartmouth College, NH 03755 USA</b></td><td>('1687325', 'Du Tran', 'du tran')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')</td><td>{DUTRAN,LORENZO}@CS.DARTMOUTH.EDU +</td></tr><tr><td>1b4bc7447f500af2601c5233879afc057a5876d8</td><td>Facial Action Unit Classification with Hidden Knowledge +<br/>under Incomplete Annotation +<br/><b>University of Science and</b><br/>Technology of China +<br/>Hefei, Anhui +<br/><b>University of Science and</b><br/>Technology of China +<br/>Hefei, Anhui +<br/>Rensselaer Polytechnic +<br/><b>Institute</b><br/>Troy, NY +<br/>P.R.China, 230027 +<br/>P.R.China, 230027 +<br/>USA, 12180 +</td><td>('1715001', 'Jun Wang', 'jun wang')<br/>('1791319', 'Shangfei Wang', 'shangfei wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>junwong@mail.ustc.edu.cn +<br/>sfwang@ustc.edu.cn +<br/>qji@ecse.rpi.edu +</td></tr><tr><td>1b27ca161d2e1d4dd7d22b1247acee5c53db5104</td><td></td><td></td><td></td></tr><tr><td>1badfeece64d1bf43aa55c141afe61c74d0bd25e</td><td>OL ´E: Orthogonal Low-rank Embedding, +<br/>A Plug and Play Geometric Loss for Deep Learning +<br/>1Universidad de la Rep´ublica +<br/>Uruguay +<br/><b>Duke University</b><br/>USA +</td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('1699339', 'Guillermo Sapiro', 'guillermo sapiro')</td><td></td></tr><tr><td>7711a7404f1f1ac3a0107203936e6332f50ac30c</td><td>Action Classification and Highlighting in Videos +<br/>Disney Research Pittsburgh +<br/>Disney Research Pittsburgh +</td><td>('1730844', 'Atousa Torabi', 'atousa torabi')<br/>('14517812', 'Leonid Sigal', 'leonid sigal')</td><td>atousa.torabi@disneyresearch.com +<br/>lsigal@disneyresearch.com +</td></tr><tr><td>778c9f88839eb26129427e1b8633caa4bd4d275e</td><td>Pose Pooling Kernels for Sub-category Recognition +<br/>ICSI & UC Berkeley +<br/>ICSI & UC Berkeley +<br/>Trever Darrell +<br/>ICSI & UC Berkeley +</td><td>('40565777', 'Ning Zhang', 'ning zhang')<br/>('2071606', 'Ryan Farrell', 'ryan farrell')</td><td>nzhang@eecs.berkeley.edu +<br/>farrell@eecs.berkeley.edu +<br/>trevor@eecs.berkeley.edu +</td></tr><tr><td>7735f63e5790006cb3d989c8c19910e40200abfc</td><td>Multispectral Imaging For Face +<br/>Recognition Over Varying +<br/>Illumination +<br/>A Dissertation +<br/>Presented for the +<br/>Doctor of Philosophy Degree +<br/><b>The University of Tennessee, Knoxville</b><br/>December 2008 +</td><td>('21051127', 'Hong Chang', 'hong chang')</td><td></td></tr><tr><td>7789a5d87884f8bafec8a82085292e87d4e2866f</td><td>A Unified Tensor-based Active Appearance Face +<br/>Model +<br/>Member, IEEE +</td><td>('2976854', 'Zhen-Hua Feng', 'zhen-hua feng')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td></td></tr><tr><td>77b1db2281292372c38926cc4aca32ef056011dc</td><td>451492 EMR0010.1177/1754073912451492Widen Children’s Interpretation of Facial ExpressionsEmotion Review +<br/>2012 +<br/>SPECIAL SECTION: FACIAL EXPRESSIONS +<br/>Children’s Interpretation of Facial Expressions: +<br/>The Long Path from Valence-Based to Specific +<br/>Discrete Categories +<br/>Emotion Review +<br/>Vol. 0, No. 0 (2012) 1 –6 +<br/>© The Author(s) 2012 +<br/>ISSN 1754-0739 +<br/>DOI: 10.1177/1754073912451492 +<br/>er.sagepub.com +<br/><b>Boston College, USA</b></td><td>('3947094', 'Sherri C. Widen', 'sherri c. widen')</td><td></td></tr><tr><td>776835eb176ed4655d6e6c308ab203126194c41e</td><td></td><td></td><td></td></tr><tr><td>77c53ec6ea448db4dad586e002a395c4a47ecf66</td><td>Research Journal of Applied Sciences, Engineering and Technology 4(17): 2879-2886, 2012 +<br/>ISSN: 2040-7467 +<br/>© Maxwell Scientific Organization, 2012 +<br/>Submitted: November 25, 2011 +<br/>Accepted: January 13, 2012 +<br/>Published: September 01, 2012 +<br/>Face Recognition Based on Facial Features +<br/><b>COMSATS Institute of Information Technology Wah Cantt</b><br/>47040, Pakistan +<br/><b>National University of Science and Technology</b><br/>Peshawar Road, Rawalpindi, 46000, Pakistan +</td><td>('33088042', 'Muhammad Sharif', 'muhammad sharif')<br/>('3349608', 'Muhammad Younas Javed', 'muhammad younas javed')<br/>('32805529', 'Sajjad Mohsin', 'sajjad mohsin')</td><td></td></tr><tr><td>778bff335ae1b77fd7ec67404f71a1446624331b</td><td>Hough Forest-based Facial Expression Recognition from +<br/>Video Sequences +<br/>BIWI, ETH Zurich http://www.vision.ee.ethz.ch +<br/>VISICS, K.U. Leuven http://www.esat.kuleuven.be/psi/visics +</td><td>('3092828', 'Gabriele Fanelli', 'gabriele fanelli')<br/>('2569989', 'Angela Yao', 'angela yao')<br/>('40324831', 'Pierre-Luc Noel', 'pierre-luc noel')<br/>('2946643', 'Juergen Gall', 'juergen gall')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td>{gfanelli,yaoa,gall,vangool}@vision.ee.ethz.ch +<br/>noelp@student.ethz.ch +</td></tr><tr><td>7726a6ab26a1654d34ec04c0b7b3dd80c5f84e0d</td><td>CONTENT-AWARE COMPRESSION USING SALIENCY-DRIVEN IMAGE RETARGETING +<br/>*Disney Research Zurich +<br/>†ETH Zurich +</td><td>('1782328', 'Yael Pritch', 'yael pritch')<br/>('2893744', 'Alexander Sorkine-Hornung', 'alexander sorkine-hornung')<br/>('1712877', 'Stefan Mangold', 'stefan mangold')</td><td></td></tr><tr><td>774cbb45968607a027ae4729077734db000a1ec5</td><td>I. KWAK ET AL.: VISUAL RECOGNITION OF URBAN TRIBES +<br/>From Bikers to Surfers: +<br/>Visual Recognition of Urban Tribes +<br/>Ana C. Murillo2 +<br/>David Kriegman1 +<br/>Serge Belongie1 +<br/>1 Dept. of Computer Science and +<br/>Engineering +<br/><b>University of California, San Diego</b><br/>San Diego, CA, USA +<br/>2 Dpt. Informática e Ing. Sistemas - Inst. +<br/>Investigación en Ingeniería de Aragón. +<br/><b>University of Zaragoza, Spain</b><br/>3 Department of Computer Science +<br/><b>Columbia University, USA</b></td><td>('2064392', 'Iljung S. Kwak', 'iljung s. kwak')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')</td><td>iskwak@cs.ucsd.edu +<br/>acm@unizar.es +<br/>belhumeur@cs.columbia.edu +<br/>kriegman@cs.ucsd.edu +<br/>sjb@cs.ucsd.edu +</td></tr><tr><td>7754b708d6258fb8279aa5667ce805e9f925dfd0</td><td>Facial Action Unit Recognition by Exploiting +<br/>Their Dynamic and Semantic Relationships +</td><td>('1686235', 'Yan Tong', 'yan tong')<br/>('2460793', 'Wenhui Liao', 'wenhui liao')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td></td></tr><tr><td>77db171a523fc3d08c91cea94c9562f3edce56e1</td><td>Poursaberi et al. EURASIP Journal on Image and Video Processing 2012, 2012:17 +<br/>http://jivp.eurasipjournals.com/content/2012/1/17 +<br/>R ES EAR CH +<br/>Open Access +<br/>Gauss–Laguerre wavelet textural feature fusion +<br/>with geometrical information for facial expression +<br/>identification +</td><td>('1786383', 'Ahmad Poursaberi', 'ahmad poursaberi')<br/>('1870195', 'Hossein Ahmadi', 'hossein ahmadi')</td><td></td></tr><tr><td>77037a22c9b8169930d74d2ce6f50f1a999c1221</td><td>Robust Face Recognition With Kernelized +<br/>Locality-Sensitive Group Sparsity Representation +</td><td>('1907978', 'Shoubiao Tan', 'shoubiao tan')<br/>('2796142', 'Xi Sun', 'xi sun')<br/>('2710497', 'Wentao Chan', 'wentao chan')<br/>('33306018', 'Lei Qu', 'lei qu')</td><td></td></tr><tr><td>779ad364cae60ca57af593c83851360c0f52c7bf</td><td>Steerable Pyramids Feature Based Classification Using Fisher +<br/>Linear Discriminant for Face Recognition +<br/>EL HASSOUNI MOHAMMED12 +<br/><b>GSCM-LRIT, Faculty of Sciences, Mohammed V University-Agdal, Rabat, Morocco</b><br/><b>DESTEC, FLSHR Mohammed V University-Agdal, Rabat, Morocco</b><br/>PO.Box 1014, Rabat, Morocco +</td><td>('37917405', 'ABOUTAJDINE DRISS', 'aboutajdine driss')</td><td>moha387@yahoo.fr +</td></tr><tr><td>7792fbc59f3eafc709323cdb63852c5d3a4b23e9</td><td>Pose from Action: Unsupervised Learning of +<br/>Pose Features based on Motion +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b></td><td>('3234247', 'Senthil Purushwalkam', 'senthil purushwalkam')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')</td><td>{spurushw@andrew,abhinavg@cs}.cmu.edu +</td></tr><tr><td>77fbbf0c5729f97fcdbfdc507deee3d388cd4889</td><td>SMITH & DYER: 3D FACIAL LANDMARK ESTIMATION +<br/>Pose-Robust 3D Facial Landmark Estimation +<br/>from a Single 2D Image +<br/>http://www.cs.wisc.edu/~bmsmith +<br/>http://www.cs.wisc.edu/~dyer +<br/>Department of Computer Sciences +<br/><b>University of Wisconsin-Madison</b><br/>Madison, WI USA +</td><td>('2721523', 'Brandon M. Smith', 'brandon m. smith')<br/>('1724754', 'Charles R. Dyer', 'charles r. dyer')</td><td></td></tr><tr><td>776362314f1479f5319aaf989624ac604ba42c65</td><td>Attribute learning in large-scale datasets +<br/><b>Stanford University</b></td><td>('2192178', 'Olga Russakovsky', 'olga russakovsky')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td>{olga,feifeili}@cs.stanford.edu +</td></tr><tr><td>77d31d2ec25df44781d999d6ff980183093fb3de</td><td>The Multiverse Loss for Robust Transfer Learning +<br/>Supplementary +<br/>1. Omitted proofs +<br/>for which the joint loss: +<br/>m(cid:88) +<br/>r=1 +<br/>L(F r, br, D, y) +<br/>(2) +<br/>J(F 1, b1...F m, bm, D, y) = +<br/>is bounded by: +<br/>mL∗(D, y) ≤ J(F 1, b1...F m, bm, D, y) +<br/>m−1(cid:88) +<br/>≤ mL∗(D, y) + +<br/>Alλd−j+1 +<br/>(3) +<br/>l=1 +<br/>where [A1 . . . Am−1] are bounded parameters. +<br/>We provide proofs that were omitted from the paper for +<br/>lack of space. We follow the same theorem numbering as in +<br/>the paper. +<br/>Lemma 1. The minimizers F ∗, b∗ of L are not unique, and +<br/>it holds that for any vector v ∈ Rc and scalar s, the solu- +<br/>tions F ∗ + v1(cid:62) +<br/>Proof. denoting V = v1(cid:62) +<br/>c , b∗ + s1c are also minimizers of L. +<br/>c , s = s1c, +<br/>i v+byi +s +<br/>i v+bj +s +<br/>i fyi +byi +<br/>i v+sed(cid:62) +<br/>i fj +bj +<br/>i=1 +<br/>log( +<br/>L(F ∗ + V, b∗ + s, D, y) = +<br/>i fyi +d(cid:62) +<br/>ed(cid:62) +<br/>i fj +d(cid:62) +<br/>j=1 ed(cid:62) +<br/>i v+sed(cid:62) +<br/>ed(cid:62) +<br/>j=1 ed(cid:62) +<br/>i v+sed(cid:62) +<br/>ed(cid:62) +<br/>(cid:80)c +<br/>(cid:80)c +<br/>i v+s(cid:80)c +<br/>− n(cid:88) +<br/>= − n(cid:88) +<br/>= − n(cid:88) +<br/>(cid:80)c +<br/>= − n(cid:88) +<br/>ed(cid:62) +<br/>i fyi +byi +<br/>j=1 ed(cid:62) +<br/>i fj +bj +<br/>ed(cid:62) +<br/>log( +<br/>log( +<br/>log( +<br/>i=1 +<br/>i=1 +<br/>i=1 +<br/>i fj +bj +<br/>i fyi +byi +<br/>j=1 ed(cid:62) +<br/>) = L(F ∗, b∗, D, y) +<br/>The following simple lemma was not part of the paper. +<br/>However, it is the reasoning behind the statement at the end +<br/>of the proof of Thm. 1. “Since ∀i, j pi(j) > 0 and since +<br/>rank(D) is full,(cid:80)n +<br/>Lemma 2. Let K =(cid:80)n +<br/>such that ∀i qi > 0, the matrix ˆK =(cid:80)n +<br/>i be a full rank d×d matrix, +<br/>i.e., it is PD and not just PSD, then for all vector q ∈ Rn +<br/>is also +<br/>i pi(j)pi(j(cid:48)) is PD.” +<br/>i=1 did(cid:62) +<br/>i=1 did(cid:62) +<br/>i=1 qidid(cid:62) +<br/>full rank. +<br/>Proof. For +<br/>(miniqi)v(cid:62)Kv > 0. +<br/>every vector v +<br/>(cid:2)f 1 +<br/>(cid:3) , b1, F 2 = (cid:2)f 2 +<br/>Theorem 3. There exist a set of weights F 1 = +<br/>j ⊥ f s +<br/>C ] , bm which are orthogonal ∀jrs f r +<br/>2 , ..., f 1 +<br/>2 , ..., f m +<br/>1 , f 1 +<br/>1 , f m +<br/>2 , ..., f 2 +<br/>1 , f 2 +<br/>[f m +<br/>(cid:3) , b2...F m = +<br/>Proof. We again prove the theorem by constructing such a +<br/>solution. Denoting by vd−m+2...vd the eigenvectors of K +<br/>corresponding to λd−m+2 . . . λd. Given F 1 = F ∗, b1 = b∗, +<br/>we can construct each pair F r, br as follows: +<br/>(1) +<br/>∀j, r +<br/>fj +<br/>r = f1 +<br/>1 + +<br/>m−1(cid:88) +<br/>l=1 +<br/>αjlrvd−l+1 +<br/>br = b1 +<br/>(4) +<br/>The tensor of parameters αjlr is constructed to insure the +<br/>orthogonality condition. Formally, αjlr has to satisfy: +<br/>Rd, +<br/>v(cid:62) ˆKv +<br/>∀j, r (cid:54)= s +<br/>(f 1 +<br/>j + +<br/>m−1(cid:88) +<br/>l=1 +<br/>αjlrvd−l+1)(cid:62)f s +<br/>j = 0 +<br/>(5) +<br/>2 m(m− 1) equations, it +<br/>Noticing that 5 constitutes a set of 1 +<br/>can be satisfied by the tensor αjlr which contains m(m − +<br/>c ] = F r − +<br/>1)c parameters. Defining Ψr = [ψr +<br/>1, ψr +<br/>2, . . . , ψr +</td><td></td><td></td></tr><tr><td>77fb9e36196d7bb2b505340b6b94ba552a58b01b</td><td>Detecting the Moment of Completion: +<br/>Temporal Models for Localising Action Completion +<br/><b>University of Bristol, Bristol, BS8 1UB, UK</b></td><td>('10007321', 'Farnoosh Heidarivincheh', 'farnoosh heidarivincheh')<br/>('1728108', 'Majid Mirmehdi', 'majid mirmehdi')<br/>('1728459', 'Dima Damen', 'dima damen')</td><td>farnoosh.heidarivincheh@bristol.ac.uk +</td></tr><tr><td>486840f4f524e97f692a7f6b42cd19019ee71533</td><td>DeepVisage: Making face recognition simple yet with powerful generalization +<br/>skills +<br/>1Laboratoire LIRIS, ´Ecole centrale de Lyon, 69134 Ecully, France. +<br/>2Safran Identity & Security, 92130 Issy-les-Moulineaux, France. +</td><td>('34767162', 'Jonathan Milgram', 'jonathan milgram')<br/>('34086868', 'Liming Chen', 'liming chen')</td><td>md-abul.hasnat@ec-lyon.fr, {julien.bohne, stephane.gentric, jonathan.milgram}@safrangroup.com, +<br/>liming.chen@ec-lyon.fr +</td></tr><tr><td>48463a119f67ff2c43b7c38f0a722a32f590dfeb</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 52– No.4, August 2012 +<br/>Intelligent Method for Face Recognition of Infant +<br/>Department of Computer +<br/>Engineering +<br/><b>Indian Institute of Technology</b><br/><b>Banaras Hindu University</b><br/>Varanasi, India-221005 +<br/>Department of Computer +<br/>Engineering +<br/><b>Indian Institute of Technology</b><br/><b>Banaras Hindu University</b><br/>Varanasi, India-221005 +<br/> +<br/>Department of Computer +<br/>Engineering +<br/><b>Indian Institute of Technology</b><br/><b>Banaras Hindu University</b><br/>Varanasi, India-221005 +</td><td>('2829597', 'Shrikant Tiwari', 'shrikant tiwari')<br/>('1920426', 'Aruni Singh', 'aruni singh')<br/>('32120516', 'Sanjay Kumar Singh', 'sanjay kumar singh')</td><td></td></tr><tr><td>488d3e32d046232680cc0ba80ce3879f92f35cac</td><td>Journal of Information Systems and Telecommunication, Vol. 2, No. 4, October-December 2014 +<br/>205 +<br/>Facial Expression Recognition Using Texture Description of +<br/>Displacement Image +<br/><b>Amirkabir University of Technology, Tehran. Iran</b><br/>Abolghasem-Asadollah Raie* +<br/><b>Amirkabir University of Technology, Tehran. Iran</b><br/><b>Sharif University of Technology, Tehran. Iran</b><br/>Received: 14/Sep/2013 Revised: 15/Mar/2014 Accepted: 10/Aug/2014 +</td><td>('3295771', 'Hamid Sadeghi', 'hamid sadeghi')<br/>('1697809', 'Mohammad-Reza Mohammadi', 'mohammad-reza mohammadi')</td><td>hamid.sadeghi@aut.ac.ir +<br/>raie@aut.ac.ir +<br/>mrmohammadi@ee.sharif.edu +</td></tr><tr><td>48186494fc7c0cc664edec16ce582b3fcb5249c0</td><td>P-CNN: Pose-based CNN Features for Action Recognition +<br/>Guilhem Ch´eron∗ † +<br/>INRIA +</td><td>('1785596', 'Ivan Laptev', 'ivan laptev')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>48499deeaa1e31ac22c901d115b8b9867f89f952</td><td>Interim Report of Final Year Project +<br/>HKU-Face: A Large Scale Dataset for +<br/>Deep Face Recognition +<br/>3035140108 +<br/>Haoyu Li +<br/>3035141841 +<br/>COMP4801 Final Year Project +<br/>Project Code: 17007 +</td><td>('3347561', 'Haicheng Wang', 'haicheng wang')</td><td></td></tr><tr><td>486a82f50835ea888fbc5c6babf3cf8e8b9807bc</td><td>MSU TECHNICAL REPORT MSU-CSE-15-11, JULY 24, 2015 +<br/>Face Search at Scale: 80 Million Gallery +</td><td>('7496032', 'Dayong Wang', 'dayong wang')<br/>('40653304', 'Charles Otto', 'charles otto')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>48fea82b247641c79e1994f4ac24cad6b6275972</td><td>Mining Discriminative Components With Low-Rank And +<br/>Sparsity Constraints for Face Recognition +<br/>Computer Science and Engineering +<br/><b>Arizona State University</b><br/>Tempe, AZ, 85281 +</td><td>('1689161', 'Qiang Zhang', 'qiang zhang')<br/>('2913552', 'Baoxin Li', 'baoxin li')</td><td>qzhang53, baoxin.li@asu.edu +</td></tr><tr><td>48734cb558b271d5809286447ff105fd2e9a6850</td><td>Facial Expression Recognition Using Enhanced Deep 3D Convolutional Neural +<br/>Networks +<br/>Department of Electrical and Computer Engineering +<br/><b>University of Denver, Denver, CO</b></td><td>('3093835', 'Mohammad H. Mahoor', 'mohammad h. mahoor')</td><td>behzad.hasani@du.edu and mmahoor@du.edu +</td></tr><tr><td>48a417cfeba06feb4c7ab30f06c57ffbc288d0b5</td><td>Robust Dictionary Learning by Error Source Decomposition +<br/><b>Northwestern University</b><br/>2145 Sheridan Road, Evanston, IL 60208 +</td><td>('2240134', 'Zhuoyuan Chen', 'zhuoyuan chen')<br/>('39955137', 'Ying Wu', 'ying wu')</td><td>zhuoyuanchen2014@u.northwestern.edu,yingwu@eecs.northwestern.edu +</td></tr><tr><td>4850af6b54391fc33c8028a0b7fafe05855a96ff</td><td>Discovering Useful Parts for Pose Estimation in Sparsely Annotated Datasets +<br/>1Department of Computer Science and 2Department of Biology +<br/><b>Boston University and 2University of North Carolina</b></td><td>('2025025', 'Mikhail Breslav', 'mikhail breslav')<br/>('1711465', 'Tyson L. Hedrick', 'tyson l. hedrick')<br/>('1749590', 'Stan Sclaroff', 'stan sclaroff')<br/>('1723703', 'Margrit Betke', 'margrit betke')</td><td>breslav@bu.edu, thedrick@bio.unc.edu, sclaroff@bu.edu, betke@bu.edu +</td></tr><tr><td>48c41ffab7ff19d24e8df3092f0b5812c1d3fb6e</td><td>Multi-Modal Embedding for Main Product Detection in Fashion +<br/>1Institut de Robtica i Informtica Industrial (CSIC-UPC) +<br/>2Wide Eyes Technologies +<br/><b>Waseda University</b></td><td>('1737881', 'Antonio Rubio', 'antonio rubio')<br/>('9072783', 'LongLong Yu', 'longlong yu')<br/>('3114470', 'Edgar Simo-Serra', 'edgar simo-serra')<br/>('1994318', 'Francesc Moreno-Noguer', 'francesc moreno-noguer')</td><td>arubio@iri.upc.edu, longyu@wide-eyes.it, esimo@aoni.waseda.jp, fmoreno@iri.upc.edu +</td></tr><tr><td>488a61e0a1c3768affdcd3c694706e5bb17ae548</td><td>FITTING A 3D MORPHABLE MODEL TO EDGES: +<br/>A COMPARISON BETWEEN HARD AND SOFT CORRESPONDENCES +<br/><b>Multimodal Computing and Interaction, Saarland University, Germany</b><br/><b>University of York, UK</b><br/>‡ Morpheo Team, INRIA Grenoble Rhˆone-Alpes, France +</td><td>('39180407', 'Anil Bas', 'anil bas')<br/>('1687021', 'William A. P. Smith', 'william a. p. smith')<br/>('1780750', 'Timo Bolkart', 'timo bolkart')<br/>('1792200', 'Stefanie Wuhrer', 'stefanie wuhrer')</td><td></td></tr><tr><td>48910f9b6ccc40226cd4f105ed5291571271b39e</td><td>Learning Discriminative Fisher Kernels +<br/><b>Pattern Recognition and Bio-informatics Laboratory, Delft University of Technology, THE NETHERLANDS</b></td><td>('1803520', 'Laurens van der Maaten', 'laurens van der maaten')</td><td>lvdmaaten@gmail.com +</td></tr><tr><td>48a9241edda07252c1aadca09875fabcfee32871</td><td>Convolutional Experts Network for Facial Landmark Detection +<br/><b>Carnegie Mellon University</b><br/>Tadas Baltruˇsaitis∗ +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +<br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave, Pittsburgh, PA 15213, USA +</td><td>('1783029', 'Amir Zadeh', 'amir zadeh')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td>abagherz@cs.cmu.edu +<br/>tbaltrus@cs.cmu.edu +<br/>morency@cs.cmu.edu +</td></tr><tr><td>48f0055295be7b175a06df5bc6fa5c6b69725785</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 96– No.19, June 2014 +<br/>Facial Action Unit Recognition from Video Streams +<br/>with Recurrent Neural Networks +<br/><b>University of the Witwatersrand</b><br/>Braamfontein, Johannesburg +<br/>South Africa +</td><td>('3122515', 'Hima Vadapalli', 'hima vadapalli')</td><td></td></tr><tr><td>48729e4de8aa478ee5eeeb08a72a446b0f5367d5</td><td>COMPRESSED FACE HALLUCINATION +<br/>Electrical Engineering and Computer Science +<br/><b>University of California, Merced, CA 95344, USA</b></td><td>('2391885', 'Sifei Liu', 'sifei liu')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>48e6c6d981efe2c2fb0ae9287376fcae59da9878</td><td>Sidekick Policy Learning +<br/>for Active Visual Exploration +<br/><b>The University of Texas at Austin, Austin, TX</b><br/>2 Facebook AI Research, 300 W. Sixth St. Austin, TX 78701 +</td><td>('21810992', 'Santhosh K. Ramakrishnan', 'santhosh k. ramakrishnan')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>srama@cs.utexas.edu, grauman@fb.com(cid:63) +</td></tr><tr><td>48174c414cfce7f1d71c4401d2b3d49ba91c5338</td><td>Robust Performance-driven 3D Face Tracking in Long Range Depth Scenes +<br/><b>Rutgers University, USA</b><br/><b>Hong Kong Polytechnic University, Hong Kong</b><br/><b>School of Computer Engineering, Nanyang Technological University, Singapore</b></td><td>('1965812', 'Chongyu Chen', 'chongyu chen')<br/>('40643777', 'Luc N. Dao', 'luc n. dao')<br/>('1736042', 'Vladimir Pavlovic', 'vladimir pavlovic')<br/>('1688642', 'Jianfei Cai', 'jianfei cai')<br/>('1775268', 'Tat-Jen Cham', 'tat-jen cham')</td><td>{hxp1,vladimir}@cs.rutgers.edu +<br/>{nldao,asjfcai,astfcham}@ntu.edu.sg +<br/>cscychen@comp.polyu.edu.hk +</td></tr><tr><td>48a5b6ee60475b18411a910c6084b3a32147b8cd</td><td>Pedestrian attribute recognition with part-based CNN +<br/>and combined feature representations +<br/>Baskurt +<br/>To cite this version: +<br/>recognition with part-based CNN and combined feature representations. VISAPP2018, Jan 2018, +<br/>Funchal, Portugal. <hal-01625470> +<br/>HAL Id: hal-01625470 +<br/>https://hal.archives-ouvertes.fr/hal-01625470 +<br/>Submitted on 21 Jun 2018 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destinée au dépôt et à la diffusion de documents +<br/>scientifiques de niveau recherche, publiés ou non, +<br/>émanant des établissements d’enseignement et de +<br/>recherche français ou étrangers, des laboratoires +<br/>publics ou privés. +</td><td>('1705461', 'Yiqiang Chen', 'yiqiang chen')<br/>('1762557', 'Stefan Duffner', 'stefan duffner')<br/>('10469201', 'Andrei Stoian', 'andrei stoian')<br/>('1733569', 'Jean-Yves Dufour', 'jean-yves dufour')<br/>('1705461', 'Yiqiang Chen', 'yiqiang chen')<br/>('1762557', 'Stefan Duffner', 'stefan duffner')<br/>('10469201', 'Andrei Stoian', 'andrei stoian')<br/>('1733569', 'Jean-Yves Dufour', 'jean-yves dufour')<br/>('1739898', 'Atilla Baskurt', 'atilla baskurt')</td><td></td></tr><tr><td>488375ae857a424febed7c0347cc9590989f01f7</td><td>Convolutional neural networks for the analysis of broadcasted +<br/>tennis games +<br/><b>Institute of Computer Science, Foundation for Research and Technology - Hellas (FORTH), Crete, 73100, Greece</b><br/>(cid:63) NantVision Inc., Culver City, CA, 90230, USA. +<br/><b>University of Crete, Crete, 73100, Greece</b></td><td>('2272443', 'Grigorios Tsagkatakis', 'grigorios tsagkatakis')<br/>('40495798', 'Mustafa Jaber', 'mustafa jaber')<br/>('1694755', 'Panagiotis Tsakalides', 'panagiotis tsakalides')</td><td></td></tr><tr><td>4836b084a583d2e794eb6a94982ea30d7990f663</td><td>Cascaded Face Alignment via Intimacy Definition Feature +<br/><b>The Hong Kong Polytechnic University</b><br/><b>Hong Kong Applied Science and Technology Research Institute Company Limited</b><br/>Hong Kong, China +<br/> +</td><td>('2116302', 'Hailiang Li', 'hailiang li')<br/>('1703078', 'Kin-Man Lam', 'kin-man lam')<br/>('2233216', 'Kangheng Wu', 'kangheng wu')<br/>('1982263', 'Zhibin Lei', 'zhibin lei')</td><td>harley.li@connect.polyu.hk,{harleyli, edmondchiu, khwu, lei}@astri.org, enkmlam@polyu.edu.hk +</td></tr><tr><td>4866a5d6d7a40a26f038fc743e16345c064e9842</td><td></td><td></td><td></td></tr><tr><td>488e475eeb3bb39a145f23ede197cd3620f1d98a</td><td>Pedestrian Attribute Classification in Surveillance: Database and Evaluation +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences (CASIA</b><br/>95 Zhongguancun East Road, 100190, Beijing, China +</td><td>('1739258', 'Jianqing Zhu', 'jianqing zhu')<br/>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1716143', 'Dong Yi', 'dong yi')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>{jqzhu, scliao, zlei, dyi, szli}@cbsr.ia.ac.cn +</td></tr><tr><td>487df616e981557c8e1201829a1d0ec1ecb7d275</td><td>Acoustic Echo Cancellation Using a Vector-Space-Based +<br/>Adaptive Filtering Algorithm +</td><td>('1742704', 'Yu Tsao', 'yu tsao')<br/>('1757214', 'Shih-Hau Fang', 'shih-hau fang')<br/>('40466874', 'Yao Shiao', 'yao shiao')</td><td></td></tr><tr><td>48f211a9764f2bf6d6dda4a467008eda5680837a</td><td></td><td></td><td></td></tr><tr><td>4858d014bb5119a199448fcd36746c413e60f295</td><td></td><td></td><td></td></tr><tr><td>48319e611f0daaa758ed5dcf5a6496b4c6ef45f2</td><td>Non Binary Local Gradient Contours for Face Recognition +<br/><b>P.A. College of Engnineering, Mangalore</b><br/>bSenior IEEE Member, Department of Electrical and Electronics Engineering, Aligarh Muslim +<br/><b>P A College of Engineering, Nadupadavu</b><br/>As the features from the traditional Local Binary patterns (LBP) and Local Directional Patterns (LDP) are +<br/>found to be ineffective for face recognition, we have proposed a new approach derived on the basis of Information +<br/>sets whereby the loss of information that occurs during the binarization is eliminated. The information sets +<br/>as a product. Since face is having smooth texture in a limited area, the extracted features must be highly +<br/>discernible. To limit the number of features, we consider only the non overlapping windows. By the application +<br/>of the information set theory we can reduce the number of feature of an image. The derived features are shown +<br/>to work fairly well over eigenface, fisherface and LBP methods. +<br/>Keywords: Local Binary Pattern, Local Directional Pattern, Information Sets, Gradient Contour, Support +<br/>Vector Machine, KNN, Face Recognition. +<br/>1. INTRODUCTION +<br/>In face recognition, the major issue to be ad- +<br/>dressed is the extraction of features which are +<br/>discriminating in nature [1], [2]. The accuracy +<br/>of classification depends upon which texture fea- +<br/>ture of the face are extracted e.g., geometrical, +<br/>statistical, local or global features in addition to +<br/>representation of these features and the design +<br/>extraction algorithm should produce little vari- +<br/>ance of features within the class and large vari- +<br/>ance between the classes. There are typically +<br/>two common approaches to extract facial fea- +<br/>tures: geometric-feature-based and appearance- +<br/>based methods. The geometric-feature-based [[3], +<br/>[4]] method encodes the shape and locations of +<br/>different facial components, which are combined +<br/>into a feature vector that represents the face. +<br/>An illustration of this method is the graph-based +<br/>method [5], that uses several facial components +<br/>to create a representation of the face and pro- +<br/>cess it. The Local-Global Graph algorithm [5] ap- +<br/>proach makes use Voronoi tessellation and Delau- +<br/>nay graphs to segment local features and builds +<br/>a graph. These features are combined into a lo- +<br/>cal graph, and then the skeleton (global graph) +<br/>is created by interrelating the local graphs to +<br/>represent the topology of the face. The major +<br/>requirements of geometric-feature-based methods +<br/>is accurate and reliable facial feature detection +<br/>and tracking, which is difficult to accommodate +<br/>in many situations. +<br/>In the case of appearance +<br/>based methods, there are many methods for the +<br/>holistic classes such as, Eigenfaces [6] and Fisher- +<br/>faces [7], which are built on Principal Component +<br/>Analysis (PCA) [6], to the more recent 2D-PCA +<br/>[8], and Linear Discriminant Analysis [9] are also +<br/>examples of holistic methods. The [10] and [11] +<br/>makes use of image filters, either on the whole +<br/>face to create holistic features, or some specific +<br/>face-region to create local features, to extract the +</td><td>('1913846', 'Abdullah Gubbi', 'abdullah gubbi')<br/>('2093112', 'Mohammad Fazle Azeem', 'mohammad fazle azeem')</td><td>Nadupadavu, Mangalore, India, Contact: abdullahgubbi@yahoo.com +<br/>University, India, Contact: mf.azeem@gmail.com +<br/>Mangalore, India. Contact: sharmilabp@gmail.com +</td></tr><tr><td>4896909796f9bd2f70a2cb24bf18daacd6a12128</td><td>Spatial Bag of Features Learning for Large Scale +<br/>Face Image Retrieval +<br/><b>Aristotle University of Thessaloniki, Thessaloniki, Greece</b></td><td>('3200630', 'Nikolaos Passalis', 'nikolaos passalis')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')</td><td>passalis@csd.auth.gr, tefas@aiia.csd.auth.gr +</td></tr><tr><td>48cfc5789c246c6ad88ff841701204fc9d6577ed</td><td>J Inf Process Syst, Vol.12, No.3, pp.392~409, September 2016 +<br/> +<br/> +<br/>ISSN 1976-913X (Print) +<br/>ISSN 2092-805X (Electronic) +<br/>Age Invariant Face Recognition Based on DCT +<br/>Feature Extraction and Kernel Fisher Analysis +</td><td>('17349931', 'Leila Boussaad', 'leila boussaad')<br/>('2411455', 'Mohamed Benmohammed', 'mohamed benmohammed')<br/>('2123013', 'Redha Benzid', 'redha benzid')</td><td></td></tr><tr><td>481fb0a74528fa7706669a5cce6a212ac46eaea3</td><td>Recognizing RGB Images by Learning from RGB-D Data +<br/><b>Institute for Infocomm Research, Agency for Science, Technology and Research, Singapore</b><br/><b>School of Computer Engineering, Nanyang Technological University, Singapore</b></td><td>('39253009', 'Lin Chen', 'lin chen')<br/>('38188040', 'Dong Xu', 'dong xu')</td><td></td></tr><tr><td>70f189798c8b9f2b31c8b5566a5cf3107050b349</td><td>The Challenge of Face Recognition from Digital Point-and-Shoot Cameras +<br/>David Bolme‡ +</td><td>('1757322', 'J. Ross Beveridge', 'j. ross beveridge')<br/>('1750370', 'Geof H. Givens', 'geof h. givens')<br/>('2067993', 'W. Todd Scruggs', 'w. todd scruggs')<br/>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')<br/>('1733571', 'Yui Man Lui', 'yui man lui')<br/>('1799014', 'Kevin W. Bowyer', 'kevin w. bowyer')<br/>('36903861', 'Mohammad Nayeem Teli', 'mohammad nayeem teli')<br/>('1704876', 'Patrick J. Flynn', 'patrick j. flynn')<br/>('1694404', 'Bruce A. Draper', 'bruce a. draper')<br/>('40370804', 'Hao Zhang', 'hao zhang')<br/>('9099328', 'Su Cheng', 'su cheng')</td><td></td></tr><tr><td>70580ed8bc482cad66e059e838e4a779081d1648</td><td>Acta Polytechnica Hungarica +<br/>Vol. 10, No. 4, 2013 +<br/>Gender Classification using Multi-Level +<br/>Wavelets on Real World Face Images +<br/><b>Shaheed Zulfikar Ali Bhutto Institute of</b><br/>Science and Technology, Plot # 67, Street # 9, H/8-4 Islamabad, 44000, Pakistan +<br/>isb.edu.pk +</td><td>('35332495', 'Sajid Ali Khan', 'sajid ali khan')<br/>('1723986', 'Muhammad Nazir', 'muhammad nazir')<br/>('2521631', 'Naveed Riaz', 'naveed riaz')</td><td>sajid.ali@szabist-isb.edu.pk, nazir@szabist-isb.edu.pk, n.r.ansari@szabist- +</td></tr><tr><td>70109c670471db2e0ede3842cbb58ba6be804561</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Zero-Shot Visual Recognition via Bidirectional Latent Embedding +<br/>Received: date / Accepted: date +</td><td>('47599321', 'Qian Wang', 'qian wang')</td><td></td></tr><tr><td>703890b7a50d6535900a5883e8d2a6813ead3a03</td><td></td><td></td><td></td></tr><tr><td>703dc33736939f88625227e38367cfb2a65319fe</td><td>Labeling Temporal Bounds for Object Interactions in Egocentric Video +<br/>Trespassing the Boundaries: +<br/><b>University of Bristol, United Kingdom</b><br/>Walterio Mayol-Cuevas +</td><td>('3420479', 'Davide Moltisanti', 'davide moltisanti')<br/>('2052236', 'Michael Wray', 'michael wray')<br/>('1728459', 'Dima Damen', 'dima damen')</td><td><FirstName>.<LastName>@bristol.ac.uk +</td></tr><tr><td>70db3a0d2ca8a797153cc68506b8650908cb0ada</td><td>An Overview of Research Activities in Facial +<br/>Age Estimation Using the FG-NET Aging +<br/>Database +<br/>Visual Media Computing Lab, +<br/>Dept. of Multimedia and Graphic Arts, +<br/><b>Cyprus University of Technology, Cyprus</b></td><td>('31950370', 'Gabriel Panis', 'gabriel panis')<br/>('1830709', 'Andreas Lanitis', 'andreas lanitis')</td><td>gpanis@gmail.com, andreas.lanitis@cut.ac.cy +</td></tr><tr><td>706236308e1c8d8b8ba7749869c6b9c25fa9f957</td><td>Crowdsourced Data Collection of Facial Responses +<br/>MIT Media Lab +<br/>Cambridge +<br/>02139, USA +<br/>Rosalind Picard +<br/>MIT Media Lab +<br/>Cambridge +<br/>02139, USA +<br/>MIT Media Lab +<br/>Cambridge +<br/>02139, USA +</td><td>('1801452', 'Daniel McDuff', 'daniel mcduff')<br/>('1754451', 'Rana El Kaliouby', 'rana el kaliouby')</td><td>djmcduff@media.mit.edu +<br/>kaliouby@media.mit.edu +<br/>picard@media.mit.edu +</td></tr><tr><td>701f56f0eac9f88387de1f556acef78016b05d52</td><td>Direct Shape Regression Networks for End-to-End Face Alignment +<br/>1 ∗ +<br/><b>University of Texas at Arlington, TX, USA, 2Beihang University, Beijing, China</b><br/><b>Xidian University, Xi an, China, 4 University of Pittsburgh, PA, USA</b></td><td>('6050999', 'Xin Miao', 'xin miao')<br/>('34798935', 'Xiantong Zhen', 'xiantong zhen')<br/>('1720747', 'Vassilis Athitsos', 'vassilis athitsos')<br/>('6820648', 'Xianglong Liu', 'xianglong liu')<br/>('1748032', 'Heng Huang', 'heng huang')<br/>('50542664', 'Cheng Deng', 'cheng deng')</td><td>xin.miao@mavs.uta.edu, zhenxt@gmail.com, xlliu@nlsde.edu.cn, chdeng.xd@gmail.com +<br/>athitsos@uta.edu, heng.huang@pitt.edu +</td></tr><tr><td>7002d6fc3e0453320da5c863a70dbb598415e7aa</td><td>Electrical Engineering +<br/><b>University of California, Riverside</b><br/>Date: Friday, October 21, 2011 +<br/>Location: EBU2 Room 205/206 +<br/>Time: 12:10am +<br/>Understanding Discrete Facial +<br/>Expression in Video Using Emotion +<br/>Avatar Image +</td><td>('1803478', 'Songfan Yang', 'songfan yang')</td><td></td></tr><tr><td>7071cd1ee46db4bc1824c4fd62d36f6d13cad08a</td><td>Face Detection through Scale-Friendly Deep Convolutional Networks +<br/><b>The Chinese University of Hong Kong</b></td><td>('1692609', 'Shuo Yang', 'shuo yang')<br/>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')<br/>('1717179', 'Chen Change Loy', 'chen change loy')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{ys014, yjxiong, ccloy, xtang}@ie.cuhk,edu.hk +</td></tr><tr><td>706b9767a444de4fe153b2f3bff29df7674c3161</td><td>Fast Metric Learning For Deep Neural Networks +<br/><b>University of Waikato, Hamilton, New Zealand</b><br/><b>School of Engineering, University of Waikato, Hamilton, New Zealand</b></td><td>('2319565', 'Henry Gouk', 'henry gouk')<br/>('1737420', 'Bernhard Pfahringer', 'bernhard pfahringer')</td><td>hgrg1@students.waikato.ac.nz, bernhard@waikato.ac.nz +<br/>cree@waikato.ac.nz +</td></tr><tr><td>70c58700eb89368e66a8f0d3fc54f32f69d423e1</td><td>INCORPORATING SCALABILITY IN UNSUPERVISED SPATIO-TEMPORAL FEATURE +<br/>LEARNING +<br/><b>University of California, Riverside, CA</b></td><td>('49616225', 'Sujoy Paul', 'sujoy paul')<br/>('2177805', 'Sourya Roy', 'sourya roy')<br/>('1688416', 'Amit K. Roy-Chowdhury', 'amit k. roy-chowdhury')</td><td></td></tr><tr><td>707a542c580bcbf3a5a75cce2df80d75990853cc</td><td>Disentangled Variational Representation for Heterogeneous Face Recognition +<br/>1 Center for Research on Intelligent Perception and Computing (CRIPAC), CASIA, Beijing, China +<br/>2 National Laboratory of Pattern Recognition (NLPR), CASIA, Beijing, China +<br/><b>School of Arti cial Intelligence, University of Chinese Academy of Sciences, Beijing, China</b><br/><b>Johns Hopkins University, 3400 N. Charles St, Baltimore, MD 21218, USA</b></td><td>('2225749', 'Xiang Wu', 'xiang wu')<br/>('32885778', 'Huaibo Huang', 'huaibo huang')<br/>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('1705643', 'Ran He', 'ran he')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>alfredxiangwu@gmail.com, huaibo.huang@cripac.ia.ac.cn, +<br/>vpatel36@jhu.edu, {rhe, znsun}@nlpr.ia.ac.cn +</td></tr><tr><td>70569810e46f476515fce80a602a210f8d9a2b95</td><td>Apparent Age Estimation from Face Images Combining General and +<br/>Children-Specialized Deep Learning Models +<br/>1Orange Labs – France Telecom, 4 rue Clos Courtel, 35512 Cesson-S´evign´e, France +<br/>2Eurecom, 450 route des Chappes, 06410 Biot, France +</td><td>('3116433', 'Grigory Antipov', 'grigory antipov')<br/>('2341854', 'Moez Baccouche', 'moez baccouche')<br/>('1708844', 'Sid-Ahmed Berrani', 'sid-ahmed berrani')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td>{grigory.antipov,moez.baccouche,sidahmed.berrani}@orange.com, jean-luc.dugelay@eurecom.fr +</td></tr><tr><td>704d88168bdfabe31b6ff484507f4a2244b8c52b</td><td>MLtuner: System Support for Automatic Machine Learning Tuning +<br/><b>Carnegie Mellon University</b></td><td>('1874200', 'Henggang Cui', 'henggang cui')<br/>('1707164', 'Gregory R. Ganger', 'gregory r. ganger')<br/>('1974678', 'Phillip B. Gibbons', 'phillip b. gibbons')</td><td></td></tr><tr><td>70e79d7b64f5540d309465620b0dab19d9520df1</td><td>International Journal of Scientific & Engineering Research, Volume 8, Issue 3, March-2017 +<br/>ISSN 2229-5518 +<br/>Facial Expression Recognition System +<br/>Using Extreme Learning Machine +</td><td>('3274320', 'Firoz Mahmud', 'firoz mahmud')<br/>('2376022', 'Md. Al Mamun', 'md. al mamun')</td><td></td></tr><tr><td>7003d903d5e88351d649b90d378f3fc5f211282b</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 68– No.23, April 2013 +<br/>Facial Expression Recognition using Gabor Wavelet +<br/>ENTC SVERI’S COE (Poly), +<br/>Pandharpur, +<br/>Solapur, India +<br/>ENTC SVERI’S COE, +<br/>Pandharpur, +<br/>Solapur, India +<br/>ENTC SVERI’S COE (Poly), +<br/>Pandharpur, +<br/>Solapur, India +</td><td>('2730988', 'Mahesh Kumbhar', 'mahesh kumbhar')<br/>('10845943', 'Manasi Patil', 'manasi patil')<br/>('2707920', 'Ashish Jadhav', 'ashish jadhav')</td><td></td></tr><tr><td>703c9c8f20860a1b1be63e6df1622b2021b003ca</td><td>Flip-Invariant Motion Representation +<br/><b>National Institute of Advanced Industrial Science and Technology</b><br/>Umezono 1-1-1, Tsukuba, Japan +</td><td>('1800592', 'Takumi Kobayashi', 'takumi kobayashi')</td><td>takumi.kobayashi@aist.go.jp +</td></tr><tr><td>70a69569ba61f3585cd90c70ca5832e838fa1584</td><td>Friendly Faces: +<br/>Weakly supervised character identification +<br/><b>CVSSP, University of Surrey, UK</b></td><td>('2735914', 'Matthew Marter', 'matthew marter')<br/>('1695195', 'Richard Bowden', 'richard bowden')</td><td>{m.marter, s.hadfield, r.bowden} @surrey.ac.uk +</td></tr><tr><td>70bf1769d2d5737fc82de72c24adbb7882d2effd</td><td>Face detection in intelligent ambiences with colored illumination +<br/>Department of Intelligent Systems +<br/>TU Delft +<br/>Delft, The Netherlands +</td><td>('3137870', 'Christina Katsimerou', 'christina katsimerou')<br/>('1728396', 'Ingrid Heynderickx', 'ingrid heynderickx')</td><td></td></tr><tr><td>70c9d11cad12dc1692a4507a97f50311f1689dbf</td><td>Video Frame Synthesis using Deep Voxel Flow +<br/><b>The Chinese University of Hong Kong</b><br/>3Pony.AI Inc. +<br/><b>University of Illinois at Urbana-Champaign</b><br/>4Google Inc. +</td><td>('3243969', 'Ziwei Liu', 'ziwei liu')</td><td>{lz013,xtang}@ie.cuhk.edu.hk +<br/>yiming@pony.ai +<br/>yeh17@illinois.edu +<br/>aseemaa@google.com +</td></tr><tr><td>1e5ca4183929929a4e6f09b1e1d54823b8217b8e</td><td>Classification in the Presence of Heavy +<br/>Label Noise: A Markov Chain Sampling +<br/>Framework +<br/>by +<br/><b>B.Eng., Nankai University</b><br/>Thesis Submitted in Partial Fulfillment of the +<br/>Requirements for the Degree of +<br/>Master of Science +<br/>in the +<br/>School of Computing Science +<br/>Faculty of Applied Sciences +<br/><b>SIMON FRASER UNIVERSITY</b><br/>Summer 2017 +<br/>However, in accordance with the Copyright Act of Canada, this work may be +<br/>reproduced without authorization under the conditions for “Fair Dealing.” +<br/>Therefore, limited reproduction of this work for the purposes of private study, +<br/>research, education, satire, parody, criticism, review and news reporting is likely +<br/>All rights reserved. +<br/>to be in accordance with the law, particularly if cited appropriately. +</td><td>('3440173', 'Zijin Zhao', 'zijin zhao')<br/>('3440173', 'Zijin Zhao', 'zijin zhao')</td><td></td></tr><tr><td>1e058b3af90d475bf53b3f977bab6f4d9269e6e8</td><td>Manifold Relevance Determination +<br/><b>University of Shef eld, UK</b><br/><b>KTH Royal Institute of Technology, CVAP Lab, Stockholm, Sweden</b><br/>Wellcome Trust Centre for Human Genetics, Roosevelt Drive, Oxford OX3 7BN, UK +<br/><b>University of Shef eld, UK</b></td><td>('3106771', 'Andreas C. Damianou', 'andreas c. damianou')<br/>('2484138', 'Carl Henrik Ek', 'carl henrik ek')<br/>('1722732', 'Michalis K. Titsias', 'michalis k. titsias')<br/>('1739851', 'Neil D. Lawrence', 'neil d. lawrence')</td><td>ANDREAS.DAMIANOU@SHEFFIELD.AC.UK +<br/>CHEK@CSC.KTH.SE +<br/>MTITSIAS@WELL.OX.AC.UK +<br/>N.LAWRENCE@SHEFFIELD.AC.UK +</td></tr><tr><td>1e799047e294267087ec1e2c385fac67074ee5c8</td><td>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 21, NO. 12, DECEMBER 1999 +<br/>1357 +<br/>Short Papers___________________________________________________________________________________________________ +<br/>Automatic Classification of +<br/>Single Facial Images +</td><td>('1709339', 'Michael J. Lyons', 'michael j. lyons')<br/>('2240088', 'Julien Budynek', 'julien budynek')<br/>('34801422', 'Shigeru Akamatsu', 'shigeru akamatsu')</td><td></td></tr><tr><td>1ef4815f41fa3a9217a8a8af12cc385f6ed137e1</td><td>Rendering of Eyes for Eye-Shape Registration and Gaze Estimation +</td><td>('34399452', 'Erroll Wood', 'erroll wood')<br/>('2520795', 'Xucong Zhang', 'xucong zhang')<br/>('1751242', 'Yusuke Sugano', 'yusuke sugano')<br/>('39626495', 'Peter Robinson', 'peter robinson')<br/>('3194727', 'Andreas Bulling', 'andreas bulling')</td><td>University of Cambridge, United Kingdom {eww23,tb346,pr10}@cam.ac.uk +<br/>Max Planck Institute for Informatics, Germany {xczhang,sugano,bulling}@mpi-inf.mpg.de +</td></tr><tr><td>1eb4ea011a3122dc7ef3447e10c1dad5b69b0642</td><td>Contextual Visual Recognition from Images and Videos +<br/>Jitendra Malik +<br/>Electrical Engineering and Computer Sciences +<br/><b>University of California at Berkeley</b><br/>Technical Report No. UCB/EECS-2016-132 +<br/>http://www.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-132.html +<br/>July 19, 2016 +</td><td>('2082991', 'Georgia Gkioxari', 'georgia gkioxari')</td><td></td></tr><tr><td>1e7ae86a78a9b4860aa720fb0fd0bdc199b092c3</td><td>Article +<br/>A Brief Review of Facial Emotion Recognition Based +<br/>on Visual Information +<br/>Byoung Chul Ko ID +<br/>Tel.: +82-10-3559-4564 +<br/>Received: 6 December 2017; Accepted: 25 January 2018; Published: 30 January 2018 +</td><td></td><td>Department of Computer Engineering, Keimyung University, Daegu 42601, Korea; niceko@kmu.ac.kr; +</td></tr><tr><td>1e8eee51fd3bf7a9570d6ee6aa9a09454254689d</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TPAMI.2016.2582166, IEEE +<br/>Transactions on Pattern Analysis and Machine Intelligence +<br/>Face Search at Scale +</td><td>('7496032', 'Dayong Wang', 'dayong wang')<br/>('40653304', 'Charles Otto', 'charles otto')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>1ea8085fe1c79d12adffb02bd157b54d799568e4</td><td></td><td></td><td></td></tr><tr><td>1ea74780d529a458123a08250d8fa6ef1da47a25</td><td>Videos from the 2013 Boston Marathon: +<br/>An Event Reconstruction Dataset for +<br/>Synchronization and Localization +<br/>CMU-LTI-018 +<br/><b>Language Technologies Institute</b><br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Ave., Pittsburgh, PA 15213 +<br/>www.lti.cs.cmu.edu +<br/>© October 1, 2016 +</td><td>('49252656', 'Jia Chen', 'jia chen')<br/>('1915796', 'Junwei Liang', 'junwei liang')<br/>('47896638', 'Han Lu', 'han lu')<br/>('2927024', 'Shoou-I Yu', 'shoou-i yu')<br/>('7661726', 'Alexander G. Hauptmann', 'alexander g. hauptmann')</td><td></td></tr><tr><td>1ebdfceebad642299e573a8995bc5ed1fad173e3</td><td></td><td></td><td></td></tr><tr><td>1eec03527703114d15e98ef9e55bee5d6eeba736</td><td>UNIVERSITÄT KARLSRUHE (TH) +<br/>FAKULTÄT FÜR INFORMATIK +<br/>INTERACTIVE SYSTEMS LABS +<br/>DIPLOMA THESIS +<br/>Automatic identification +<br/>of persons in TV series +<br/>SUBMITTED BY +<br/>MAY 2008 +<br/>ADVISORS +</td><td>('12141635', 'A. Waibel', 'a. waibel')<br/>('2284204', 'Mika Fischer', 'mika fischer')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td></td></tr><tr><td>1e07500b00fcd0f65cf30a11f9023f74fe8ce65c</td><td>WHOLE SPACE SUBCLASS DISCRIMINANT ANALYSIS FOR FACE RECOGNITION +<br/><b>Institute for Infocomm Research, A*STAR, Singapore</b></td><td>('1709001', 'Bappaditya Mandal', 'bappaditya mandal')<br/>('35718875', 'Liyuan Li', 'liyuan li')<br/>('1802086', 'Vijay Chandrasekhar', 'vijay chandrasekhar')</td><td>Email: {bmandal, lyli, vijay, joohwee}@i2r.a-star.edu.sg +</td></tr><tr><td>1e19ea6e7f1c04a18c952ce29386252485e4031e</td><td>International Association of Scientific Innovation and Research (IASIR) +<br/>(An Association Unifying the Sciences, Engineering, and Applied Research) +<br/>ISSN (Print): 2279-0047 +<br/>ISSN (Online): 2279-0055 +<br/> International Journal of Emerging Technologies in Computational +<br/>and Applied Sciences (IJETCAS) +<br/>www.iasir.net +<br/>MATLAB Based Face Recognition System Using PCA and Neural Network +<br/>1Faculty of Computer Science & Engineering, 2Research Scholar +<br/><b>University Institute of Engineering and Technology</b><br/><b>Kurukshetra University, Kurukshetra-136 119, Haryana, INDIA</b></td><td>('1989126', 'Sanjeev Dhawan', 'sanjeev dhawan')<br/>('7940433', 'Himanshu Dogra', 'himanshu dogra')</td><td>E-mail (s): rsdhawan@rediffmail.com, himanshu.dogra.13@gmail.com +</td></tr><tr><td>1ec98785ac91808455b753d4bc00441d8572c416</td><td>Curriculum Learning for Facial Expression Recognition +<br/><b>Language Technologies Institute, School of Computer Science</b><br/><b>Carnegie Mellon University, USA</b><br/>few years, +</td><td>('1970583', 'Liangke Gui', 'liangke gui')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td></td></tr><tr><td>1ed6c7e02b4b3ef76f74dd04b2b6050faa6e2177</td><td>Face Detection with a 3D Model +<br/>Department of Statistics +<br/><b>Florida State University</b><br/><b>National Institutes of Health</b></td><td>('2455529', 'Adrian Barbu', 'adrian barbu')<br/>('2230628', 'Nathan Lay', 'nathan lay')</td><td>abarbu@stat.fsu.edu +<br/>nathan.lay@nih.gov +</td></tr><tr><td>1efacaa0eaa7e16146c34cd20814d1411b35538e</td><td>HEIDARIVINCHEHET AL: ACTIONCOMPLETION:A TEMPORALMODEL.. +<br/>Action Completion: +<br/>A Temporal Model for Moment Detection +<br/>Department of Computer Science +<br/><b>University of Bristol</b><br/>Bristol, UK +</td><td>('10007321', 'Farnoosh Heidarivincheh', 'farnoosh heidarivincheh')<br/>('1728108', 'Majid Mirmehdi', 'majid mirmehdi')<br/>('1728459', 'Dima Damen', 'dima damen')</td><td>Farnoosh.Heidarivincheh@bristol.ac.uk +<br/>M.Mirmehdi@bristol.ac.uk +<br/>Dima.Damen@bristol.ac.uk +</td></tr><tr><td>1eba6fc35a027134aa8997413647b49685f6fbd1</td><td>Superpower Glass: Delivering +<br/>Unobtrusive Real-time Social Cues +<br/>in Wearable Systems +<br/>Dennis Wall +<br/><b>Stanford University</b><br/>Stanford, CA 94305, USA +<br/>Permission to make digital or hard copies of part or all of this work for +<br/>personal or classroom use is granted without fee provided that copies are +<br/>not made or distributed for profit or commercial advantage and that copies +<br/>bear this notice and the full citation on the first page. Copyrights for third- +<br/>party components of this work must be honored. For all other uses, contact +<br/>the Owner/Author. +<br/>Copyright is held by the owner/author(s). +<br/>Ubicomp/ISWC'16 Adjunct , September 12-16, 2016, Heidelberg, Germany +<br/>ACM 978-1-4503-4462-3/16/09. +<br/>http://dx.doi.org/10.1145/2968219.2968310 +</td><td>('21701693', 'Catalin Voss', 'catalin voss')<br/>('40026202', 'Peter Washington', 'peter washington')<br/>('32551479', 'Nick Haber', 'nick haber')<br/>('40494635', 'Aaron Kline', 'aaron kline')<br/>('34240128', 'Jena Daniels', 'jena daniels')<br/>('3407835', 'Azar Fazel', 'azar fazel')<br/>('3457025', 'Titas De', 'titas de')<br/>('3456914', 'Beth McCarthy', 'beth mccarthy')<br/>('34925386', 'Carl Feinstein', 'carl feinstein')<br/>('1699245', 'Terry Winograd', 'terry winograd')</td><td>catalin@cs.stanford.edu +<br/>peterwashington@stanford.edu +<br/>nhaber@stanford.edu +<br/>akline@stanford.edu +<br/>danielsj@stanford.edu +<br/>azarf@stanford.edu +<br/>titasde@stanford.edu +<br/>bethmac@stanford.edu +<br/>carlf@stanford.edu +<br/>winograd@cs.stanford.edu +<br/>dpwall@stanford.edu +</td></tr><tr><td>1e1d7cbbef67e9e042a3a0a9a1bcefcc4a9adacf</td><td>A Multi-Level Contextual Model For Person Recognition in Photo Albums +<br/><b>Stevens Institute of Technology</b><br/>‡Adobe Research +<br/>(cid:92)Microsoft Research +</td><td>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')<br/>('1720987', 'Xiaohui Shen', 'xiaohui shen')<br/>('1745420', 'Gang Hua', 'gang hua')</td><td>†hli18@stevens.edu +<br/>‡{jbrandt, zlin, xshen}@adobe.com +<br/>(cid:92)ganghua@microsoft.com +</td></tr><tr><td>1ef1f33c48bc159881c5c8536cbbd533d31b0e9a</td><td>Z. ZHANG ET AL.: ADVERSARIAL TRAINING FOR ACTION UNIT RECOGNITION +<br/>Identity-based Adversarial Training of Deep +<br/>CNNs for Facial Action Unit Recognition +<br/>Department of Computer Science +<br/><b>State University of New York at</b><br/>Binghamton +<br/>NY, USA. +</td><td>('47294008', 'Zheng Zhang', 'zheng zhang')<br/>('2443456', 'Shuangfei Zhai', 'shuangfei zhai')<br/>('8072251', 'Lijun Yin', 'lijun yin')</td><td>zzhang27@cs.binghamton.edu +<br/>szhai2@cs.binghamton.edu +<br/>lijun@cs.binghamton.edu +</td></tr><tr><td>1ef5ce743a44d8a454dbfc2657e1e2e2d025e366</td><td>Global Journal of Computer Science & Technology +<br/>Volume 11 Issue Version 1.0 April 2011 +<br/>Type: Double Blind Peer Reviewed International Research Journal +<br/>Publisher: Global Journals Inc. (USA) +<br/>Online ISSN: 0975-4172 & Print ISSN: 0975-4350 +<br/> +<br/>Accurate Corner Detection Methods using Two Step Approach +<br/><b>Thapar University</b></td><td>('1765523', 'Nitin Bhatia', 'nitin bhatia')<br/>('9180065', 'Megha Chhabra', 'megha chhabra')</td><td></td></tr><tr><td>1e58d7e5277288176456c66f6b1433c41ca77415</td><td>Bootstrapping Fine-Grained Classifiers: +<br/>Active Learning with a Crowd in the Loop +<br/><b>Brown University, 2University of California, San Diego, 3California Institute of Technology</b></td><td>('40541456', 'Genevieve Patterson', 'genevieve patterson')</td><td>{gen, hays}@cs.brown.edu gvanhorn@ucsd.edu sjb@cs.ucsd.edu +<br/>perona@caltech.edu +</td></tr><tr><td>1e5a1619fe5586e5ded2c7a845e73f22960bbf5a</td><td>Group Membership Prediction +<br/><b>Boston University</b></td><td>('7969330', 'Ziming Zhang', 'ziming zhang')<br/>('9772059', 'Yuting Chen', 'yuting chen')<br/>('1699322', 'Venkatesh Saligrama', 'venkatesh saligrama')</td><td>{zzhang14, yutingch, srv}@bu.edu +</td></tr><tr><td>1e9f1bbb751fe538dde9f612f60eb946747defaa</td><td>Journal of Systems Engineering and Electronics +<br/>Vol. 28, No. 4, August 2017, pp.784 – 792 +<br/>Identity-aware convolutional neural networks for +<br/>facial expression recognition +<br/><b>The Big Data Research Center, Henan University, Kaifeng 475001, China</b><br/><b>Tampere University of Technology, Tampere 33720, Finland</b></td><td>('34461878', 'Chongsheng Zhang', 'chongsheng zhang')<br/>('39720477', 'Pengyou Wang', 'pengyou wang')<br/>('40611812', 'Ke Chen', 'ke chen')</td><td></td></tr><tr><td>1e917fe7462445996837934a7e46eeec14ebc65f</td><td>Expression Classification using +<br/>Wavelet Packet Method +<br/>on Asymmetry Faces +<br/>CMU-RI-TR-06-03 +<br/>January 2006 +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, Pennsylvania 15213 +<br/><b>Carnegie Mellon University</b></td><td>('1689241', 'Yanxi Liu', 'yanxi liu')</td><td></td></tr><tr><td>1e8394cc9fe7c2392aa36fb4878faf7e78bbf2de</td><td>TO APPEAR IN IEEE THMS +<br/>Zero-Shot Object Recognition System +<br/>based on Topic Model +</td><td>('2800072', 'Wai Lam Hoo', 'wai lam hoo')<br/>('2863960', 'Chee Seng Chan', 'chee seng chan')</td><td></td></tr><tr><td>1ef4aac0ebc34e76123f848c256840d89ff728d0</td><td></td><td></td><td></td></tr><tr><td>1ecb56e7c06a380b3ce582af3a629f6ef0104457</td><td>List of Contents Vol.8 +<br/>Contents of +<br/>Journal of Advanced Computational +<br/> Intelligence and Intelligent Informatics +<br/>Volume 8 +<br/>Vol.8 No.1, January 2004 +<br/>Editorial: +<br/>o Special Issue on Selected Papers from Humanoid, +<br/>Papers: +<br/>o Dynamic Color Object Recognition Using Fuzzy +<br/>Nano-technology, Information Technology, +<br/>Communication and Control, Environment, and +<br/>Management (HNICEM’03). +<br/>. 1 +<br/>Elmer P. Dadios +<br/>Papers: +<br/>o A New Way of Discovery of Belief, Desire and +<br/>Intention in the BDI Agent-Based Software +<br/>Modeling . +<br/>. 2 +<br/>o Integration of Distributed Robotic Systems +<br/>. 7 +<br/>Fakhri Karray, Rogelio Soto, Federico Guedea, +<br/>and Insop Song +<br/>o A Searching and Tracking Framework for +<br/>Multi-Robot Observation of Multiple Moving +<br/>Targets . +<br/>. 14 +<br/>Zheng Liu, Marcelo H. Ang Jr., and Winston +<br/>Khoon Guan Seah +<br/>Development Paper: +<br/>o Possibilistic Uncertainty Propagation and +<br/>Compromise Programming in the Life Cycle +<br/>Analysis of Alternative Motor Vehicle Fuels +<br/>Raymond R. Tan, Alvin B. Culaba, and +<br/>Michael R. I. Purvis +<br/>. 23 +<br/>Logic . +<br/>Napoleon H. Reyes, and Elmer P. Dadios +<br/>. 29 +<br/>o A Optical Coordinate Measuring Machine for +<br/>Nanoscale Dimensional Metrology . +<br/>. 39 +<br/>Eric Kirkland, Thomas R. Kurfess, and Steven +<br/>Y. Liang +<br/>o Humanoid Robot HanSaRam: Recent Progress +<br/>and Developments . +<br/>. 45 +<br/>Jong-Hwan Kim, Dong-Han Kim, Yong-Jae +<br/>Kim, Kui-Hong Park, Jae-Ho Park, +<br/>Choon-Kyoung Moon, Jee-Hwan Ryu, Kiam +<br/>Tian Seow, and Kyoung-Chul Koh +<br/>o Generalized Associative Memory Models: Their +<br/>Memory Capacities and Potential Application +<br/>. 56 +<br/>Teddy N. Yap, Jr., and Arnulfo P. Azcarraga +<br/>o Hybrid Fuzzy Logic Strategy for Soccer Robot +<br/>Game. +<br/>. 65 +<br/>Elmer A. Maravillas , Napoleon H. Reyes, and +<br/>Elmer P. Dadios +<br/>o Image Compression and Reconstruction Based on +<br/>Fuzzy Relation and Soft Computing +<br/>Technology . +<br/>. 72 +<br/>Kaoru Hirota, Hajime Nobuhara, Kazuhiko +<br/>Kawamoto, and Shin’ichi Yoshida +<br/>Vol.8 No.2, March 2004 +<br/>Editorial: +<br/>o Special Issue on Pattern Recognition . +<br/>. 83 +<br/>Papers: +<br/>o Operation of Spatiotemporal Patterns Stored in +<br/>Osamu Hasegawa +<br/>Review: +<br/>o Support Vector Machine and Generalization . 84 +<br/>Takio Kurita +<br/>o Bayesian Network: Probabilistic Reasoning, +<br/>Statistical Learning, and Applications . +<br/>. 93 +<br/>Yoichi Motomura +<br/>Living Neuronal Networks Cultured on a +<br/>Microelectrode Array . +<br/>Suguru N. Kudoh, and Takahisa Taguchi +<br/>o Rapid Discriminative Learning . +<br/>. 100 +<br/>. 108 +<br/>Jun Rokui +<br/>o Robust Fuzzy Clustering Based on Similarity +<br/>between Data . +<br/>Kohei Inoue, and Kiichi Urahama +<br/>Vol.8 No.6, 2004 +<br/>Journal of Advanced Computational Intelligence +<br/>and Intelligent Informatics +<br/>. 115 +<br/>I-1 +</td><td>('33358236', 'Chang-Hyun Jo', 'chang-hyun jo')</td><td></td></tr><tr><td>1e64b2d2f0a8a608d0d9d913c4baee6973995952</td><td>DOMINANT AND +<br/>COMPLEMENTARY MULTI- +<br/>EMOTIONAL FACIAL +<br/>EXPRESSION RECOGNITION +<br/>USING C-SUPPORT VECTOR +<br/>CLASSIFICATION +</td><td>('19172816', 'Christer Loob', 'christer loob')<br/>('2303909', 'Pejman Rasti', 'pejman rasti')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('2531522', 'Tomasz Sapinski', 'tomasz sapinski')<br/>('34969391', 'Dorota Kaminska', 'dorota kaminska')<br/>('3087532', 'Gholamreza Anbarjafari', 'gholamreza anbarjafari')</td><td></td></tr><tr><td>1e21b925b65303ef0299af65e018ec1e1b9b8d60</td><td>Under review as a conference paper at ICLR 2017 +<br/>UNSUPERVISED CROSS-DOMAIN IMAGE GENERATION +<br/>Facebook AI Research +<br/>Tel-Aviv, Israel +</td><td>('2188620', 'Yaniv Taigman', 'yaniv taigman')<br/>('33964593', 'Adam Polyak', 'adam polyak')</td><td>{yaniv,adampolyak,wolf}@fb.com +</td></tr><tr><td>1ee27c66fabde8ffe90bd2f4ccee5835f8dedbb9</td><td>Entropy Regularization +<br/>The problem of semi-supervised induction consists in learning a decision rule from +<br/>labeled and unlabeled data. This task can be undertaken by discriminative methods, +<br/>provided that learning criteria are adapted consequently. In this chapter, we moti- +<br/>vate the use of entropy regularization as a means to bene(cid:12)t from unlabeled data in +<br/>the framework of maximum a posteriori estimation. The learning criterion is derived +<br/>from clearly stated assumptions and can be applied to any smoothly parametrized +<br/>model of posterior probabilities. The regularization scheme favors low density sep- +<br/>aration, without any modeling of the density of input features. The contribution +<br/>of unlabeled data to the learning criterion induces local optima, but this problem +<br/>can be alleviated by deterministic annealing. For well-behaved models of posterior +<br/>probabilities, deterministic annealing EM provides a decomposition of the learning +<br/>problem in a series of concave subproblems. Other approaches to the semi-supervised +<br/>problem are shown to be close relatives or limiting cases of entropy regularization. +<br/>A series of experiments illustrates the good behavior of the algorithm in terms of +<br/>performance and robustness with respect to the violation of the postulated low den- +<br/>sity separation assumption. The minimum entropy solution bene(cid:12)ts from unlabeled +<br/>data and is able to challenge mixture models and manifold learning in a number of +<br/>situations. +<br/>9.1 Introduction +<br/>semi-supervised +<br/>induction +<br/>This chapter addresses semi-supervised induction, which refers to the learning of +<br/>a decision rule, on the entire input domain X, from labeled and unlabeled data. +<br/>The objective is identical to the one of supervised classi(cid:12)cation: generalize from +<br/>examples. The problem di(cid:11)ers in the respect that the supervisor’s responses are +<br/>missing for some training examples. This characteristic is shared with transduction, +<br/>which has however a di(cid:11)erent goal, that is, of predicting labels on a set of prede(cid:12)ned +</td><td>('1802711', 'Yves Grandvalet', 'yves grandvalet')<br/>('1751762', 'Yoshua Bengio', 'yoshua bengio')</td><td></td></tr><tr><td>1ee3b4ba04e54bfbacba94d54bf8d05fd202931d</td><td>Indonesian Journal of Electrical Engineering and Computer Science +<br/>Vol. 12, No. 2, November 2018, pp. 476~481 +<br/>ISSN: 2502-4752, DOI: 10.11591/ijeecs.v12.i2.pp476-481 +<br/> 476 +<br/>Celebrity Face Recognition using Deep Learning +<br/>1,2,3Faculty of Computer and Mathematical Sciences, UniversitiTeknologi MARA (UiTM), +<br/>4Faculty of Computer and Mathematical Sciences, UniversitiTeknologi MARA (UiTM), +<br/> Shah Alam, Selangor, Malaysia +<br/>Campus Jasin, Melaka, Malaysia +<br/>Article Info +<br/>Article history: +<br/>Received May 29, 2018 +<br/>Revised Jul 30, 2018 +<br/>Accepted Aug 3, 2018 +<br/>Keywords: +<br/>AlexNet +<br/>Convolutional neural network +<br/>Deep learning +<br/>Face recognition +<br/>GoogLeNet +</td><td>('2743254', 'Zaidah Ibrahim', 'zaidah ibrahim')</td><td></td></tr><tr><td>1e41a3fdaac9f306c0ef0a978ae050d884d77d2a</td><td>411 +<br/>Robust Object Recognition with +<br/>Cortex-Like Mechanisms +<br/>Tomaso Poggio, Member, IEEE +</td><td>('1981539', 'Thomas Serre', 'thomas serre')<br/>('1776343', 'Lior Wolf', 'lior wolf')<br/>('1996960', 'Maximilian Riesenhuber', 'maximilian riesenhuber')</td><td></td></tr><tr><td>1e94cc91c5293c8fc89204d4b881552e5b2ce672</td><td>Unsupervised Alignment of Actions in Video with Text Descriptions +<br/><b>University of Rochester, Rochester, NY, USA</b><br/><b>Indian Institute of Technology Delhi, New Delhi, India</b></td><td>('3193978', 'Young Chol Song', 'young chol song')<br/>('2296971', 'Iftekhar Naim', 'iftekhar naim')<br/>('1782355', 'Abdullah Al Mamun', 'abdullah al mamun')<br/>('38370357', 'Kaustubh Kulkarni', 'kaustubh kulkarni')<br/>('35108153', 'Parag Singla', 'parag singla')<br/>('33642939', 'Jiebo Luo', 'jiebo luo')<br/>('1793218', 'Daniel Gildea', 'daniel gildea')</td><td></td></tr><tr><td>1e1e66783f51a206509b0a427e68b3f6e40a27c8</td><td>SEMI-SUPERVISED ESTIMATION OF PERCEIVED AGE +<br/>FROM FACE IMAGES +<br/>VALWAY Technology Center, NEC Soft, Ltd., Tokyo, Japan +<br/>Keywords: +</td><td>('2163491', 'Kazuya Ueki', 'kazuya ueki')<br/>('1719221', 'Masashi Sugiyama', 'masashi sugiyama')</td><td>ueki@mxf.nes.nec.co.jp +</td></tr><tr><td>1efaa128378f988965841eb3f49d1319a102dc36</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +<br/>Hierarchical binary CNNs for landmark +<br/>localization with limited resources +</td><td>('3458121', 'Adrian Bulat', 'adrian bulat')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td></td></tr><tr><td>1e8eec6fc0e4538e21909ab6037c228547a678ba</td><td><b>IMPERIAL COLLEGE</b><br/><b>University of London</b><br/>enVisage : Face Recognition in +<br/>Videos +<br/>Supervisor : Dr. Stefan Rüeger +<br/>June 14, 2006 +</td><td>('23558890', 'Ashwin Venkatraman', 'ashwin venkatraman')<br/>('35805861', 'Ian Harries', 'ian harries')</td><td>(av102@doc.ic.ac.uk) +</td></tr><tr><td>1e6ed6ca8209340573a5e907a6e2e546a3bf2d28</td><td>Pooling Faces: Template based Face Recognition with Pooled Face Images +<br/>Prem Natarajan1 +<br/>Gérard Medioni3 +<br/><b>Information Sciences Institute, USC, CA, USA</b><br/><b>The Open University of Israel, Israel</b><br/><b>Institute for Robotics and Intelligent Systems, USC, CA, USA</b></td><td>('1756099', 'Tal Hassner', 'tal hassner')<br/>('11269472', 'Iacopo Masi', 'iacopo masi')<br/>('5911467', 'Jungyeon Kim', 'jungyeon kim')<br/>('1689391', 'Jongmoo Choi', 'jongmoo choi')<br/>('35840854', 'Shai Harel', 'shai harel')</td><td></td></tr><tr><td>8451bf3dd6bcd946be14b1a75af8bbb65a42d4b2</td><td>Consensual and Privacy-Preserving Sharing of +<br/>Multi-Subject and Interdependent Data +<br/>EPFL, UNIL–HEC Lausanne +<br/>K´evin Huguenin +<br/>UNIL–HEC Lausanne +<br/>EPFL +<br/>EPFL +</td><td>('1862343', 'Alexandra-Mihaela Olteanu', 'alexandra-mihaela olteanu')<br/>('2461431', 'Italo Dacosta', 'italo dacosta')<br/>('1757221', 'Jean-Pierre Hubaux', 'jean-pierre hubaux')</td><td>alexandramihaela.olteanu@epfl.ch +<br/>kevin.huguenin@unil.ch +<br/>italo.dacosta@epfl.ch +<br/>jean-pierre.hubaux@epfl.ch +</td></tr><tr><td>841855205818d3a6d6f85ec17a22515f4f062882</td><td>Low Resolution Face Recognition in the Wild +<br/>Patrick Flynn1 +<br/>1Department of Computer Science and Engineering +<br/><b>University of Notre Dame</b><br/>2Department of Computer Science +<br/>Pontificia Universidad Cat´olica de Chile +</td><td>('50492554', 'Pei Li', 'pei li')<br/>('47522390', 'Loreto Prieto', 'loreto prieto')<br/>('1797475', 'Domingo Mery', 'domingo mery')</td><td></td></tr><tr><td>84c0f814951b80c3b2e39caf3925b56a9b2e1733</td><td>Manifesto from Dagstuhl Perspectives Workshop 12382 +<br/>Computation and Palaeography: Potentials and Limits∗ +<br/>Edited by +<br/><b>The Open University of</b><br/><b>University of Nebraska Lincoln, USA</b><br/><b>King s College London, UK</b><br/><b>The Blavatnik School of Computer Science, Tel Aviv University, IL</b></td><td>('1756099', 'Tal Hassner', 'tal hassner')<br/>('34564710', 'Malte Rehbein', 'malte rehbein')<br/>('34876976', 'Peter A. Stokes', 'peter a. stokes')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td>Israel, IL, hassner@openu.ac.il +<br/>malte.rehbein@unl.edu +<br/>peter.stokes@kcl.ac.uk +<br/>wolf@cs.tau.ac.il +</td></tr><tr><td>84fe5b4ac805af63206012d29523a1e033bc827e</td><td></td><td></td><td></td></tr><tr><td>84e4b7469f9c4b6c9e73733fa28788730fd30379</td><td>Duong et al. EURASIP Journal on Advances in Signal Processing (2018) 2018:10 +<br/>DOI 10.1186/s13634-017-0521-9 +<br/>EURASIP Journal on Advances +<br/>in Signal Processing +<br/>R ES EAR CH +<br/>Projective complex matrix factorization for +<br/>facial expression recognition +<br/>Open Access +</td><td>('2345136', 'Viet-Hang Duong', 'viet-hang duong')<br/>('2033188', 'Yuan-Shan Lee', 'yuan-shan lee')<br/>('1782417', 'Jian-Jiun Ding', 'jian-jiun ding')<br/>('34759060', 'Bach-Tung Pham', 'bach-tung pham')<br/>('30065390', 'Manh-Quan Bui', 'manh-quan bui')<br/>('35196812', 'Pham The Bao', 'pham the bao')<br/>('3205648', 'Jia-Ching Wang', 'jia-ching wang')</td><td></td></tr><tr><td>84dcf04802743d9907b5b3ae28b19cbbacd97981</td><td></td><td></td><td></td></tr><tr><td>841bf196ee0086c805bd5d1d0bddfadc87e424ec</td><td>International Journal of Signal Processing, Image Processing and Pattern Recognition +<br/>Vol. 5, No. 4, December, 2012 +<br/>Locally Kernel-based Nonlinear Regression for Face Recognition +<br/>South Tehran Branch, Electrical Engineering Department, Tehran, Iran +<br/><b>Islamic Azad University</b><br/><b>Amirkabir University of Technology</b><br/>Electrical Engineering Department,Tehran, Iran +</td><td>('3345810', 'Yaser Arianpour', 'yaser arianpour')<br/>('2630546', 'Sedigheh Ghofrani', 'sedigheh ghofrani')<br/>('1685153', 'Hamidreza Amindavar', 'hamidreza amindavar')</td><td>st_y_arianpour@azad.ac.ir, s_ghofrani@azad.ac.ir and hamidami@aut.ac.ir +</td></tr><tr><td>842d82081f4b27ca2d4bc05c6c7e389378f0c7b8</td><td>ELEKTROTEHNI ˇSKI VESTNIK 78(1-2): 12–17, 2011 +<br/>ENGLISH EDITION +<br/>Usage of affective computing in recommender systems +<br/>Marko Tkalˇciˇc, Andrej Koˇsir, Jurij Tasiˇc +<br/><b>University of Ljubljana, Faculty of Electrical Engineering, Trzaska 25, 1000 Ljubljana, Slovenia</b></td><td></td><td>E-mail: marko.tkalcic@fe.uni-lj.si +</td></tr><tr><td>84fa126cb19d569d2f0147bf6f9e26b54c9ad4f1</td><td>Improved Boosting Performance by Explicit +<br/>Handling of Ambiguous Positive Examples +</td><td>('1750517', 'Miroslav Kobetski', 'miroslav kobetski')<br/>('1736906', 'Josephine Sullivan', 'josephine sullivan')</td><td></td></tr><tr><td>84508e846af3ac509f7e1d74b37709107ba48bde</td><td>Use of the Septum as a Reference Point in a Neurophysiologic Approach to +<br/>Facial Expression Recognition +<br/>Department of Computer Engineering, Faculty of Engineering, +<br/><b>Prince of Songkla University, Hat Yai, Songkhla, 90112 Thailand</b><br/>Telephone: (66)080-7045015, (66)074-287-357 +</td><td>('38928684', 'Igor Stankovic', 'igor stankovic')<br/>('2799130', 'Montri Karnjanadecha', 'montri karnjanadecha')</td><td>E-mail: bizmut@neobee.net, montri@coe.psu.ac.th +</td></tr><tr><td>841a5de1d71a0b51957d9be9d9bebed33fb5d9fa</td><td>5017 +<br/>PCANet: A Simple Deep Learning Baseline for +<br/>Image Classification? +</td><td>('1926757', 'Tsung-Han Chan', 'tsung-han chan')<br/>('2370507', 'Kui Jia', 'kui jia')<br/>('1702868', 'Shenghua Gao', 'shenghua gao')<br/>('1697700', 'Jiwen Lu', 'jiwen lu')<br/>('1920683', 'Zinan Zeng', 'zinan zeng')<br/>('1700297', 'Yi Ma', 'yi ma')</td><td></td></tr><tr><td>84e6669b47670f9f4f49c0085311dce0e178b685</td><td>Face frontalization for Alignment and Recognition +<br/>∗Department of Computing, +<br/><b>Imperial College London</b><br/>180 Queens Gate, +<br/>†EEMCS, +<br/><b>University of Twente</b><br/>Drienerlolaan 5, +<br/>London SW7 2AZ, U.K. +<br/>7522 NB Enschede, The Netherlands +</td><td>('3320415', 'Christos Sagonas', 'christos sagonas')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{c.sagonas, i.panagakis, s.zafeiriou, m.pantic}@imperial.ac.uk +</td></tr><tr><td>847e07387142c1bcc65035109ccce681ef88362c</td><td>Feature Synthesis Using Genetic Programming for Face +<br/>Expression Recognition +<br/>Center for research in intelligent systems +<br/><b>University of California, Riverside CA 92521-0425, USA</b></td><td>('1707159', 'Bir Bhanu', 'bir bhanu')<br/>('1723555', 'Jiangang Yu', 'jiangang yu')<br/>('1711543', 'Xuejun Tan', 'xuejun tan')<br/>('1742735', 'Yingqiang Lin', 'yingqiang lin')</td><td>{bhanu, jyu, xtan, yqlin}@cris.ucr.edu +</td></tr><tr><td>8411fe1142935a86b819f065cd1f879f16e77401</td><td>International Journal of Artificial Intelligence & Applications (IJAIA), Vol. 4, No. 6, November 2013 +<br/>Facial Recognition using Modified Local Binary +<br/>Pattern and Random Forest +<br/>Department of Computer Science, +<br/><b>North Carolina AandT State University</b><br/>Greensboro, NC 27411 +</td><td>('3536162', 'Brian O’Connor', 'brian o’connor')<br/>('34999544', 'Kaushik Roy', 'kaushik roy')</td><td></td></tr><tr><td>843e6f1e226480e8a6872d8fd7b7b2cd74b637a4</td><td>Research Journal of Applied Sciences, Engineering and Technology 4(22): 4724-4728, 2012 +<br/>ISSN: 2040-7467 +<br/>© Maxwell Scientific Organization, 2012 +<br/>Submitted: March 31, 2012 +<br/>Accepted: April 30, 2012 +<br/> Published: November 15, 2012 +<br/>Palmprint Recognition Using Directional Representation and +<br/>Compresses Sensing +<br/>1Shandong Provincial Key Laboratory of computer Network, Shandong Computer +<br/>Science Center, Jinan 250014, China +<br/><b>School of Mechanical Engineering, Southwest Jiaotong University, Chengdu 610031, China</b></td><td>('2112738', 'Hengjian Li', 'hengjian li')</td><td></td></tr><tr><td>84f904a71bee129a1cf00dc97f6cdbe1011657e6</td><td>Fashioning with Networks: Neural Style Transfer to Design +<br/>Clothes +<br/><b>University Of Maryland</b><br/>Baltimore County (UMBC), +<br/><b>University Of Maryland</b><br/>Baltimore County (UMBC), +<br/><b>University Of Maryland</b><br/>Baltimore County (UMBC), +<br/>Baltimore, MD, +<br/>USA +<br/>Baltimore, MD, +<br/>USA +<br/>Baltimore, MD, +<br/>USA +</td><td>('30834050', 'Prutha Date', 'prutha date')<br/>('2116290', 'Ashwinkumar Ganesan', 'ashwinkumar ganesan')<br/>('1756624', 'Tim Oates', 'tim oates')</td><td>dprutha1@umbc.edu +<br/>gashwin1@umbc.edu +<br/>oates@cs.umbc.edu +</td></tr><tr><td>849f891973ad2b6c6f70d7d43d9ac5805f1a1a5b</td><td>Detecting Faces Using Region-based Fully +<br/>Convolutional Networks +<br/>Tencent AI Lab, China +</td><td>('1996677', 'Yitong Wang', 'yitong wang')</td><td>{yitongwang,denisji,encorezhou,hawelwang,michaelzfli}@tencent.com +</td></tr><tr><td>846c028643e60fefc86bae13bebd27341b87c4d1</td><td>Face Recognition Under Varying Illumination +<br/>Based on MAP Estimation Incorporating +<br/>Correlation Between Surface Points +<br/>1 Panasonic Tokyo (Matsushita Electric Industrial Co., Ltd.,) +<br/>4–3–1 Tsunashima-higashi, Kohoku-ku, Yokohama City, Kanagawa 223–8639, Japan +<br/><b>Institute of Industrial Science, The University of Tokyo</b><br/>4–6–1 Komaba, Meguro-ku Tokyo 153–8505, Japan +<br/><b>National Institute of Informatics</b><br/>2–1–2 Hitotsubashi, Chiyoda-ku Tokyo 101–8430, Japan +</td><td>('20877506', 'Mihoko Shimano', 'mihoko shimano')<br/>('1977815', 'Kenji Nagao', 'kenji nagao')<br/>('1706742', 'Takahiro Okabe', 'takahiro okabe')<br/>('1746794', 'Imari Sato', 'imari sato')<br/>('9467266', 'Yoichi Sato', 'yoichi sato')</td><td>shimano.mhk@jp.panasonic.com +<br/>{takahiro, ysato}@iis.u-tokyo.ac.jp +<br/>imarik@nii.ac.jp +</td></tr><tr><td>4a14a321a9b5101b14ed5ad6aa7636e757909a7c</td><td>Learning Semi-Supervised Representation Towards a Unified Optimization +<br/>Framework for Semi-Supervised Learning +<br/><b>School of Info. and Commu. Engineering, Beijing University of Posts and Telecommunications</b><br/><b>Key Laboratory of Machine Perception (MOE), School of EECS, Peking University</b><br/><b>Cooperative Medianet Innovation Center, Shanghai Jiaotong University</b></td><td>('9171002', 'Chun-Guang Li', 'chun-guang li')<br/>('33383055', 'Zhouchen Lin', 'zhouchen lin')<br/>('1720776', 'Honggang Zhang', 'honggang zhang')<br/>('39954962', 'Jun Guo', 'jun guo')</td><td>{lichunguang, zhhg, guojun}@bupt.edu.cn; zlin@pku.edu.cn +</td></tr><tr><td>4adca62f888226d3a16654ca499bf2a7d3d11b71</td><td>Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics, pages 572–582, +<br/>Sofia, Bulgaria, August 4-9 2013. c(cid:13)2013 Association for Computational Linguistics +<br/>572 +</td><td></td><td></td></tr><tr><td>4aa286914f17cd8cefa0320e41800a99c142a1cd</td><td>Leveraging Context to Support Automated Food Recognition in Restaurants +<br/>School of Interactive Computing +<br/><b>Georgia Institute of Technology, Atlanta, Georgia, USA</b><br/>http://www.vbettadapura.com/egocentric/food +</td><td>('3115428', 'Vinay Bettadapura', 'vinay bettadapura')<br/>('39642711', 'Edison Thomaz', 'edison thomaz')<br/>('2943897', 'Aman Parnami', 'aman parnami')<br/>('9267108', 'Gregory D. Abowd', 'gregory d. abowd')</td><td></td></tr><tr><td>4a9d906935c9de019c61aedc10b77ee10e3aec63</td><td>Cross Modal Distillation for Supervision Transfer +<br/><b>University of California, Berkeley</b></td><td>('3134457', 'Saurabh Gupta', 'saurabh gupta')<br/>('4742485', 'Judy Hoffman', 'judy hoffman')<br/>('1689212', 'Jitendra Malik', 'jitendra malik')</td><td>{sgupta, jhoffman, malik}@eecs.berkeley.edu +</td></tr><tr><td>4a2d54ea1da851151d43b38652b7ea30cdb6dfb2</td><td>Direct Recognition of Motion Blurred Faces +</td><td>('39487011', 'Kaushik Mitra', 'kaushik mitra')<br/>('2715270', 'Priyanka Vageeswaran', 'priyanka vageeswaran')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>4ae59d2a28abd76e6d9fb53c9e7ece833dce7733</td><td>A Survey on Mobile Affective Computing +<br/>Shengkai Zhang and Pan Hui +<br/>Department of Computer Science and Engineering +<br/><b>The Hong Kong University of Science and Technology</b></td><td></td><td>{szhangaj, panhui}@cse.ust.hk +</td></tr><tr><td>4ab10174a4f98f7e2da7cf6ccfeb9bc64c8e7da8</td><td><b>Graz University of Technology</b><br/><b>Institute for Computer Graphics and Vision</b><br/>Dissertation +<br/>Efficient Metric Learning for +<br/>Real-World Face Recognition +<br/>Graz, Austria, December 2013 +<br/>Thesis supervisors +<br/>Prof. Dr. Horst Bischof +<br/>Prof. Dr. Fernando De la Torre +</td><td>('1993853', 'Martin Köstinger', 'martin köstinger')</td><td></td></tr><tr><td>4ab84f203b0e752be83f7f213d7495b04b1c4c79</td><td>CONCAVE LOSSES FOR ROBUST DICTIONARY LEARNING +<br/><b>University of S ao Paulo</b><br/><b>Institute of Mathematics and Statistics</b><br/>Rua do Mat˜ao, 1010 – 05508-090 – S˜ao Paulo-SP, Brazil +<br/>Universit´e de Rouen Normandie +<br/>LITIS EA 4108 +<br/>76800 Saint- ´Etienne-du-Rouvray, France +</td><td>('30146203', 'Rafael Will M. de Araujo', 'rafael will m. de araujo')<br/>('1792962', 'Alain Rakotomamonjy', 'alain rakotomamonjy')</td><td></td></tr><tr><td>4a484d97e402ed0365d6cf162f5a60a4d8000ea0</td><td>A Crowdsourcing Approach for Finding Misidentifications of Bibliographic Records +<br/><b>University of Tsukuba</b><br/>2 National Diet Library +<br/>3 Doshisha Univeristy +</td><td>('34573158', 'Atsuyuki Morishima', 'atsuyuki morishima')<br/>('32857584', 'Takanori Kawashima', 'takanori kawashima')<br/>('23161591', 'Takashi Harada', 'takashi harada')<br/>('2406721', 'Sho Sato', 'sho sato')</td><td></td></tr><tr><td>4a3758f283b7c484d3f164528d73bc8667eb1591</td><td>Attribute Enhanced Face Aging with Wavelet-based Generative Adversarial +<br/>Networks +<br/>Center for Research on Intelligent Perception and Computing, CASIA +<br/>National Laboratory of Pattern Recognition, CASIA +</td><td>('1860829', 'Yunfan Liu', 'yunfan liu')<br/>('1682467', 'Qi Li', 'qi li')<br/>('1757186', 'Zhenan Sun', 'zhenan sun')</td><td>yunfan,liu@cripac.ia.ac.cn, {qli, znsun}@nlpr.ia.ac.cn +</td></tr><tr><td>4a4da3d1bbf10f15b448577e75112bac4861620a</td><td>FACE, EXPRESSION, AND IRIS RECOGNITION +<br/>USING LEARNING-BASED APPROACHES +<br/>by +<br/>A dissertation submitted in partial fulfillment of +<br/>the requirements for the degree of +<br/>Doctor of Philosophy +<br/>(Computer Sciences) +<br/>at the +<br/><b>UNIVERSITY OF WISCONSIN MADISON</b><br/>2006 +</td><td>('1822413', 'Guodong Guo', 'guodong guo')</td><td></td></tr><tr><td>4abd49538d04ea5c7e6d31701b57ea17bc349412</td><td>Recognizing Fine-Grained and Composite Activities +<br/>using Hand-Centric Features and Script Data +</td><td>('34849128', 'Marcus Rohrbach', 'marcus rohrbach')<br/>('40404576', 'Sikandar Amin', 'sikandar amin')</td><td></td></tr><tr><td>4aa093d1986b4ad9b073ac9edfb903f62c00e0b0</td><td>Facial Recognition with +<br/>Encoded Local Projections +<br/>Mechanincal and Mechatronics Engineering +<br/><b>University of Waterloo</b><br/>Waterloo, Canada +<br/>Kimia Lab +<br/><b>University of Waterloo</b><br/>Waterloo, Canada +</td><td>('34139904', 'Dhruv Sharma', 'dhruv sharma')<br/>('7641396', 'Sarim Zafar', 'sarim zafar')<br/>('38685017', 'Morteza Babaie', 'morteza babaie')</td><td></td></tr><tr><td>4a0f98d7dbc31497106d4f652968c708f7da6692</td><td>Real-time Eye Gaze Direction Classification Using +<br/>Convolutional Neural Network +</td><td>('3110004', 'Anjith George', 'anjith george')<br/>('2680543', 'Aurobinda Routray', 'aurobinda routray')</td><td></td></tr><tr><td>4aabd6db4594212019c9af89b3e66f39f3108aac</td><td><b>University of Colorado, Boulder</b><br/>CU Scholar +<br/>Undergraduate Honors Theses +<br/>Honors Program +<br/>Spring 2015 +<br/>The Mere Exposure Effect and Classical +<br/>Conditioning +<br/>Follow this and additional works at: http://scholar.colorado.edu/honr_theses +<br/>Part of the Cognition and Perception Commons, and the Cognitive Psychology Commons +<br/>Recommended Citation +<br/>Wong, Rosalyn, "The Mere Exposure Effect and Classical Conditioning" (2015). Undergraduate Honors Theses. Paper 937. +<br/>This Thesis is brought to you for free and open access by Honors Program at CU Scholar. It has been accepted for inclusion in Undergraduate Honors +</td><td>('10191508', 'Rosalyn Wong', 'rosalyn wong')</td><td>University of Colorado Boulder, Rosalyn.Wong@Colorado.EDU +<br/>Theses by an authorized administrator of CU Scholar. For more information, please contact cuscholaradmin@colorado.edu. +</td></tr><tr><td>4adb97b096b700af9a58d00e45a2f980136fcbb5</td><td>Exploring Temporal Preservation Networks for Precise Temporal Action +<br/>Localization +<br/>National Laboratory for Parallel and Distributed Processing, +<br/><b>National University of Defense Technology</b><br/>Changsha, China +</td><td>('40520103', 'Ke Yang', 'ke yang')<br/>('2292038', 'Peng Qiao', 'peng qiao')<br/>('40252278', 'Dongsheng Li', 'dongsheng li')<br/>('1893776', 'Shaohe Lv', 'shaohe lv')<br/>('1791001', 'Yong Dou', 'yong dou')</td><td>{yangke13,pengqiao,dongshengli,yongdou,shaohelv}@nudt.edu.cn +</td></tr><tr><td>4a5592ae1f5e9fa83d9fa17451c8ab49608421e4</td><td>Multi-modal Social Signal Analysis for Predicting +<br/>Agreement in Conversation Settings +<br/><b>IN3, Open University of</b><br/>Catalonia, Roc Boronat, 117, +<br/>08018 Barcelona, Spain. +<br/><b>University of</b><br/>Barcelona, Gran Via, 585, +<br/>08007 Barcelona, Spain. +<br/>Computer Vision Center, UAB, +<br/>08193 Barcelona, Spain. +<br/><b>University of</b><br/>Barcelona, Gran Via, 585, +<br/>08007 Barcelona, Spain. +<br/>Computer Vision Center, UAB, +<br/>08193 Barcelona, Spain. +<br/><b>EIMT, Open University of</b><br/>Catalonia, Rbla. Poblenou, +<br/>156, 08018 Barcelona, Spain. +<br/>Computer Vision Center, UAB, +<br/>08193 Barcelona, Spain. +</td><td>('1960768', 'Víctor Ponce-López', 'víctor ponce-lópez')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('1857280', 'Xavier Baró', 'xavier baró')</td><td>vponcel@uoc.edu +<br/>sergio@maia.ub.es +<br/>xbaro@uoc.edu +</td></tr><tr><td>4a1a5316e85528f4ff7a5f76699dfa8c70f6cc5c</td><td> MVA2005 IAPR Conference on Machine VIsion Applications, May 16-18, 2005 Tsukuba Science City, Japan +<br/>3-22 +<br/>Face Recognition using Local Features based on Two-layer Block M odel +<br/>W onjun Hwang1 Ji-Yeun Kim Seokcheol Kee +<br/>Computing Lab., +<br/><b>Samsung Advanced Institute of Technology</b><br/>combined by Yang and etc [7]. The sparsification of LFA +<br/>helps the reduction of dimension of image in LDA scheme +<br/>and local topological property is more useful than holistic +<br/>property of PCA in recognition, but there is still structural +<br/>problem because the method to select the features is +<br/>designed for minimization of reconstruction error, not for +<br/>increasing discriminability in face model. +<br/>In this paper, we proposed the novel recognition +<br/>algorithm to merge LFA and LDA method. We do not use +<br/>the existing sparsification method for selecting features but +<br/>adopt the two-layer block model to make several groups +<br/>with topographic local features in similar position. Each +<br/>local block, flocked local features, can represent its own +<br/>local property and at +<br/>time holistic face +<br/>information. Flocks of local features can easily solve the +<br/>small sample size problem in LDA without discarding +<br/>unselected local features, and LDA scheme can extract the +<br/>important information for recognition not in focus of +<br/>representation. M oreover, we can extract lots of vectors on +<br/>separated viewpoint from different layer model in one face +<br/>image and they have the property robust to environmental +<br/>changes and overfitting problem as compared with limited +<br/>number of features vectors. +<br/>the same +<br/>The rest of this paper is organized as follows: the brief +<br/>description on LFA and LDA is explained in Section 2.1 +<br/>and Section 2.2, respectively and proposed algorithm - +<br/>local feature based on two-layer block model is given in +<br/>Section 2.3. The experimental results are given in Section 3. +<br/>Conclusion is summarized in Section 4. +<br/>2 LFA and LDA M ethod based on Two- +<br/>Layer Block M odel +<br/>2.1 Theory of local feature analysis +<br/>A topographic representation based on second-order +<br/>image dependencies called local features analysis (LFA) +<br/>was developed by Penev and Atick [4]. Local feature +<br/>analysis can makes a set of topographic and local kernels +<br/>that are optimally matched to the second-order statistics of +<br/>the input ensemble. Local features are basically derived +<br/>from principal component eigenvectors, and consist of +<br/>sphering principal component eigenvalues to equalize their +<br/>variance. +<br/>Suppose that we are given a set of M training +<br/>M , each represented by an +<br/>images, +<br/>i(cid:77) , =1,(cid:133) , +<br/>dimensional vector obtained by a raster scan. The mean +</td><td></td><td></td></tr><tr><td>4ae291b070ad7940b3c9d3cb10e8c05955c9e269</td><td>Automatic Detection of Naturalistic Hand-over-Face +<br/>Gesture Descriptors +<br/><b>University of Cambridge, Computer Laboratory, UK</b></td><td>('2022940', 'Marwa Mahmoud', 'marwa mahmoud')<br/>('39626495', 'Peter Robinson', 'peter robinson')</td><td>{marwa.mahmoud, tadas.baltrusaitis, peter.robinson}@cl.cam.ac.uk +</td></tr><tr><td>4aa8db1a3379f00db2403bba7dade5d6e258b9e9</td><td>Recognizing Combinations of Facial Action Units with +<br/>Different Intensity Using a Mixture of Hidden Markov +<br/>Models and Neural Network +<br/><b>DSP Lab, Sharif University of Technology, Tehran, Iran</b></td><td>('1736464', 'Mahmoud Khademi', 'mahmoud khademi')<br/>('1702826', 'Mohammad Hadi Kiapour', 'mohammad hadi kiapour')<br/>('1707281', 'Ali Akbar Kiaei', 'ali akbar kiaei')</td><td>{khademi@ce.,manzuri@,kiapour@ee.,kiaei@ce.}sharif.edu +</td></tr><tr><td>4a2062ba576ca9e9a73b6aa6e8aac07f4d9344b9</td><td>Fusing Deep Convolutional Networks for Large +<br/>Scale Visual Concept Classification +<br/>Department of Computer Engineering +<br/><b>Bas kent University</b><br/>06810 Ankara, TURKEY +</td><td>('2140386', 'Hilal Ergun', 'hilal ergun')<br/>('1700011', 'Mustafa Sert', 'mustafa sert')</td><td>21020005@mail.baskent.edu.tr, Bmsert@baskent.edu.tr +</td></tr><tr><td>4ac4e8d17132f2d9812a0088594d262a9a0d339b</td><td>Rank Constrained Recognition under Unknown Illuminations +<br/>Center for Automation Research (CfAR) +<br/>Department of Electrical and Computer Engineering +<br/><b>University of Maryland, College Park, MD</b></td><td>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{shaohua, rama}@cfar.umd.edu +</td></tr><tr><td>4ac3cd8b6c50f7a26f27eefc64855134932b39be</td><td>Robust Facial Landmark Detection +<br/>via a Fully-Convolutional Local-Global Context Network +<br/><b>Technical University of Munich</b></td><td>('3044182', 'Daniel Merget', 'daniel merget')<br/>('28096417', 'Matthias Rock', 'matthias rock')<br/>('46343645', 'Gerhard Rigoll', 'gerhard rigoll')</td><td>daniel.merget@tum.de +<br/>matthias.rock@tum.de +<br/>mmk@ei.tum.de +</td></tr><tr><td>4abaebe5137d40c9fcb72711cdefdf13d9fc3e62</td><td>Dimension Reduction for Regression +<br/>with Bottleneck Neural Networks +<br/><b>BECS, Aalto University School of Science and Technology, Finland</b></td><td>('2504988', 'Elina Parviainen', 'elina parviainen')</td><td></td></tr><tr><td>4acd683b5f91589002e6f50885df51f48bc985f4</td><td>BRIDGING COMPUTER VISION AND SOCIAL SCIENCE : A MULTI-CAMERA VISION +<br/>SYSTEM FOR SOCIAL INTERACTION TRAINING ANALYSIS +<br/>Peter Tu +<br/>GE Global Research, Niskayuna NY USA +</td><td>('1713712', 'Jixu Chen', 'jixu chen')<br/>('39643145', 'Ming-Ching Chang', 'ming-ching chang')<br/>('2095482', 'Tai-Peng Tian', 'tai-peng tian')<br/>('1689202', 'Ting Yu', 'ting yu')</td><td></td></tr><tr><td>4a1d640f5e25bb60bb2347d36009718249ce9230</td><td>Towards Multi-view and Partially-occluded Face Alignment +<br/><b>National Laboratory of Pattern Recognition, Institute of Automation, CAS, Beijing 100190, P. R. China</b><br/><b>National University of Singapore, Singapore</b></td><td>('1757173', 'Junliang Xing', 'junliang xing')<br/>('1773437', 'Zhiheng Niu', 'zhiheng niu')<br/>('1753492', 'Junshi Huang', 'junshi huang')<br/>('40506509', 'Weiming Hu', 'weiming hu')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>{jlxing,wmhu}@nlpr.ia.ac.cn +<br/>{niuzhiheng,junshi.huang,eleyans}@nus.edu.sg +</td></tr><tr><td>4aeb87c11fb3a8ad603311c4650040fd3c088832</td><td>Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +<br/>1816 +</td><td></td><td></td></tr><tr><td>4a3d96b2a53114da4be3880f652a6eef3f3cc035</td><td>2666 +<br/>A Dictionary Learning-Based +<br/>3D Morphable Shape Model +</td><td>('35220006', 'Claudio Ferrari', 'claudio ferrari')<br/>('2973738', 'Giuseppe Lisanti', 'giuseppe lisanti')<br/>('2507859', 'Stefano Berretti', 'stefano berretti')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')</td><td></td></tr><tr><td>4a6fcf714f663618657effc341ae5961784504c7</td><td>Scaling up Class-Specific Kernel Discriminant +<br/>Analysis for large-scale Face Verification +</td><td>('9219875', 'Moncef Gabbouj', 'moncef gabbouj')</td><td></td></tr><tr><td>24b37016fee57057cf403fe2fc3dda78476a8262</td><td>Automatic Recognition of Eye Blinking in Spontaneously Occurring Behavior +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA</b><br/><b>University of Pittsburgh, Pittsburgh</b></td><td>('1683262', 'Tsuyoshi Moriyama', 'tsuyoshi moriyama')<br/>('1724419', 'Jing Xiao', 'jing xiao')</td><td></td></tr><tr><td>24115d209e0733e319e39badc5411bbfd82c5133</td><td>Long-term Recurrent Convolutional Networks for +<br/>Visual Recognition and Description +</td><td>('7408951', 'Jeff Donahue', 'jeff donahue')<br/>('2234342', 'Lisa Anne Hendricks', 'lisa anne hendricks')<br/>('34849128', 'Marcus Rohrbach', 'marcus rohrbach')<br/>('1811430', 'Subhashini Venugopalan', 'subhashini venugopalan')<br/>('1687120', 'Sergio Guadarrama', 'sergio guadarrama')<br/>('2903226', 'Kate Saenko', 'kate saenko')<br/>('1753210', 'Trevor Darrell', 'trevor darrell')</td><td></td></tr><tr><td>24c442ac3f6802296d71b1a1914b5d44e48b4f29</td><td>Pose and expression-coherent face recovery in the wild +<br/>Technicolor, Cesson-S´evign´e, France +<br/>Franc¸ois Le Clerc +<br/>Patrick P´erez +</td><td>('2232848', 'Xavier P. Burgos-Artizzu', 'xavier p. burgos-artizzu')<br/>('2045531', 'Joaquin Zepeda', 'joaquin zepeda')</td><td>xavier.burgos,joaquin.zepeda,francois.leclerc,patrick.perez@technicolor.com +</td></tr><tr><td>247cab87b133bd0f4f9e8ce5e7fc682be6340eac</td><td>RESEARCH ARTICLE +<br/>Robust Eye Center Localization through Face +<br/>Alignment and Invariant Isocentric Patterns +<br/><b>School of Physics and Engineering, Sun Yat-Sen University, Guangzhou, China, 2 School of Information</b><br/><b>Science and Technology, Sun Yat-Sen University, Guangzhou, China, 3 SYSU-CMU Shunde International</b><br/><b>Joint Research Institute, Foshan, China</b><br/>☯ These authors contributed equally to this work. +</td><td>('36721307', 'Zhiyong Pang', 'zhiyong pang')<br/>('2940388', 'Chuansheng Wei', 'chuansheng wei')<br/>('2127322', 'Dongdong Teng', 'dongdong teng')<br/>('2547930', 'Dihu Chen', 'dihu chen')<br/>('31912378', 'Hongzhou Tan', 'hongzhou tan')</td><td>* issthz@mail.sysu.edu.cn (HT); stspzy@mail.sysu.edu.cn (ZP) +</td></tr><tr><td>245f8ec4373e0a6c1cae36cd6fed5a2babed1386</td><td>J. Appl. Environ. Biol. Sci., 7(3S)1-10, 2017 +<br/>© 2017, TextRoad Publication +<br/>ISSN: 2090-4274 +<br/>Journal of Applied Environmental +<br/>and Biological Sciences +<br/>www.textroad.com +<br/>Lucas Kanade Optical Flow Computation from Superpixel based Intensity +<br/>Region for Facial Expression Feature Extraction +<br/>1Intelligent Biometric Group, School of Electrical and Electronics Engineering, Universiti Sains Malaysia, +<br/><b>Electrical, Electronics and Automation Section, Universiti Kuala Lumpur Malaysian Spanish Institute</b><br/>Engineering Campus, 14300 Nibong Tebal, Pulau Pinang, Malaysia +<br/>Kulim Hi-Tech Park, Kedah, Malaysia +<br/>Received: February 21, 2017 +<br/>Accepted: May 14, 2017 +</td><td>('9114862', 'Halina Hassan', 'halina hassan')<br/>('2583099', 'Abduljalil Radman', 'abduljalil radman')<br/>('2612367', 'Shahrel Azmin Suandi', 'shahrel azmin suandi')<br/>('1685966', 'Sazali Yaacob', 'sazali yaacob')</td><td></td></tr><tr><td>24cb375a998f4af278998f8dee1d33603057e525</td><td>Projection Metric Learning on Grassmann Manifold with Application to Video based Face Recognition +<br/>1Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b><br/>seek to learn a generic mapping f : G(q,D) → G(q,d) that is defined as +<br/>f (YYY iYYY T +<br/>i ) = WWW TYYY iYYY T +<br/>i WWW = (WWW TYYY i)(WWW TYYY i)T . +<br/>(1) +<br/>where WWW ∈ RD×d (d ≤ D), is a transformation matrix of column full rank. +<br/>With this mapping, the original Grassmann manifold G(q,D) can be trans- +<br/>formed into a lower-dimensional Grassmann manifold G(q,d). However, +<br/>except the case WWW is an orthogonal matrix, WWW TYYY i is not generally an or- +<br/>thonormal basis matrix. Note that only the linear subspaces spanned by or- +<br/>thonormal basis matrix can form a valid Grassmann manifold. To tackle this +<br/>problem, we temporarily use the orthonormal components of WWW TYYY i defined +<br/>(cid:48) +<br/>by WWW TYYY +<br/>i to represent an orthonormal basis matrix of the transformed pro- +<br/>(cid:48) +<br/>jection matrices. As for the approach to get the WWW TYYY +<br/>i, we give more details +<br/>in the original paper. Here, we briefly describe the formulation of the Pro- +<br/>jection Metric on the new Grassmann manifold and the proposed objection +<br/>function in the following. +<br/>Learned Projection Metric. The Projection Metric of any pair of trans- +<br/>formed projection operators WWW TYYY +<br/>(cid:48)T +<br/>j WWW is defined by: +<br/>(cid:48) +<br/>jYYY +<br/>(cid:48) +<br/>iYYY +<br/>(cid:48) +<br/>iYYY +<br/>(cid:48) +<br/>jYYY +<br/>(cid:48)T +<br/>i WWW ,WWW TYYY +<br/>(cid:48)T +<br/>i WWW , WWW TYYY +<br/>(cid:48)T +<br/>p(WWW TYYY +<br/>d2 +<br/>j WWW ) +<br/>= 2−1/2(cid:107)WWW TYYY +<br/>(cid:48)T +<br/>(cid:48) +<br/>i WWW −WWW TYYY +<br/>iYYY +<br/>= 2−1/2tr(PPPAAAi jAAAT +<br/>i jPPP). +<br/>i −YYY +<br/>(cid:48)T +<br/>(cid:48) +<br/>jYYY +<br/>(cid:48)T +<br/>j WWW(cid:107)2 +<br/>(2) +<br/>(cid:48) +<br/>iYYY +<br/>(cid:48) +<br/>jYYY +<br/>(cid:48)T +<br/>j and PPP = WWWWWW T . Since WWW is required to be a +<br/>where AAAi j = YYY +<br/>matrix with column full rank, PPP is a rank-d symmetric positive semidefinite +<br/>matrix of size D× D, which has a similar form as Mahalanobis matrix. +<br/>Discriminant Function. The discriminant function is designed to minimize +<br/>the projection distances of any within-class subspace pairs while to maxi- +<br/>mize the projection distances of between-class subspace pairs. The matrix +<br/>PPP is thus achieved by the objective function J(PPP) as: +<br/>PPP∗ = argmin +<br/>PPP +<br/>J(PPP) = argmin +<br/>PPP +<br/>(Jw(PPP)− αJb(PPP)). +<br/>(3) +<br/>where α reflects the trade-off between the within-class compactness term +<br/>Jw(PPP) and between-class dispersion term Jb(PPP), which are measured by av- +<br/>erage within-class scatter and average between-class scatter respectively as: +<br/>Jw(PPP) = +<br/>Jb(PPP) = +<br/>Nw +<br/>i=1 +<br/>Nb +<br/>i=1 +<br/>j:Ci=Cj +<br/>j:Ci(cid:54)=Cj +<br/>2−1/2tr(PPPAAAi jAAAT +<br/>i jPPP). +<br/>2−1/2tr(PPPAAAi jAAAT +<br/>i jPPP). +<br/>(4) +<br/>(5) +<br/>where Nw is the number of pairs of samples from the same class, Nb is the +<br/>(cid:48)T +<br/>number of pairs of samples from different classes, AAAi j = YYY +<br/>j and +<br/>PPP is the PSD matrix that needs to be learned. +<br/>i −YYY +<br/>(cid:48)T +<br/>(cid:48) +<br/>jYYY +<br/>(cid:48) +<br/>iYYY +<br/>[1] J. Hamm and D. D. Lee. Grassmann discriminant analysis: a unifying +<br/>view on subspace-based learning. In ICML, 2008. +<br/>[2] Jihun Hamm and Daniel D Lee. Extended grassmann kernels for +<br/>subspace-based learning. In NIPS, 2008. +<br/>[3] Mehrtash Tafazzoli Harandi, C. Sanderson, S. Shirazi, and B. C. Lovell. +<br/>Graph embedding discriminant analysis on grassmannian manifolds for +<br/>improved image set matching. In CVPR, 2011. +<br/>[4] Mehrtash Tafazzoli Harandi, Mathieu Salzmann, Sadeep Jayasumana, +<br/>Richard Hartley, and Hongdong Li. Expanding the family of grassman- +<br/>nian kernels: An embedding perspective. In ECCV. 2014. +<br/>[5] R. Vemulapalli, J. Pillai, and R. Chellappa. Kernel learning for extrinsic +<br/>classification of manifold features. In CVPR, 2013. +<br/>Figure 1: Conceptual illustration of the proposed Projection Metric Learn- +<br/>ing (PML) on the Grassmann Manifold. Traditional Grassmann discrimi- +<br/>nant analysis methods take the away (a)-(b)-(d)-(e) to first embed the origi- +<br/>nal Grassmann manifold G(q,D) (b) into high dimensional Hilbert space H +<br/>(d) and then learn a map from the Hilbert space to a lower-dimensional, op- +<br/>tionally more discriminative space Rd (e). In contrast, the newly proposed +<br/>approach goes the way (a)-(b)-(c) to learn the metric/mapping from the orig- +<br/>inal Grassmann manifold G(q,D) (b) to a new more discriminant Grssmann +<br/>manifold G(q,d) (c). +<br/>In video based face recognition, great success has been made by represent- +<br/>ing videos as linear subspaces, which typically reside on Grassmann mani- +<br/>fold endowed with the well-studied projection metric. Under the projection +<br/>metric framework, most of recent studies [1, 2, 3, 4, 5] exploited a series of +<br/>positive definite kernel functions on Grassmann manifold to first embed the +<br/>manifold into a high dimensional Hilbert space, and then map the flattened +<br/>manifold into a lower-dimensional Euclidean space (see Fig.1 (a)-(b)-(d)- +<br/>(e)). Although these methods can be employed for supervised classification, +<br/>they are limited to the Mercer kernels which yields implicit projection, and +<br/>thus restricted to use only kernel-based classifiers. Moreover, the computa- +<br/>tional complexity of these kernel-based methods increases with the number +<br/>of training sample. +<br/>To overcome the limitations of existing Grassmann discriminant anal- +<br/>ysis methods, by endowing the well-studied Projection Metric with Grass- +<br/>mann manifold, this paper attempt to learn a Mahalanobis-like matrix on the +<br/>Grassmann manifold without resorting to kernel Hilbert space embedding. +<br/>In contrast to the kernelization scheme, our approach directly works on the +<br/>original manifold and exploits its geometry to learn a representation that stil- +<br/>l benefits from useful properties of the Grassmann manifold. Furthermore, +<br/>the learned Mahalanobis-like matrix can be decomposed into the transfor- +<br/>mation for dimensionality reduction, which maps the original Grassmann +<br/>manifold to a lower-dimensional, more discriminative Grassmann manifold +<br/>(see Fig.1 (a)-(b)-(c)). +<br/>Formally, assume m video sequences are given as {XXX 1,XXX 2, . . . ,XXX m}, +<br/>where XXX i ∈ RD×ni describes a data matrix of the i-th video containing ni +<br/>frames, each frame being expressed as a D-dimensional feature vector. In +<br/>these data, each video belongs to one of face classes denoted by Ci. The +<br/>i-th video XXX i is represented by a q-dimensional linear subspace spanned by +<br/>an orthonormal basis matrix YYY i ∈ RD×q, s.t. XXX iXXX T +<br/>i , where ΛΛΛi, +<br/>YYY i correspond to the matrices of the q largest eigenvalues and eigenvectors +<br/>respectively. +<br/>i (cid:39) YYY iΛΛΛiYYY T +<br/>Given a linear subspace span(YYY i) on Grassmann manifold (as discussed +<br/>i as the elements on the manifold), we +<br/>in the original paper, we denote YYY iYYY T +</td><td>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>24aac045f1e1a4c13a58eab4c7618dccd4c0e671</td><td></td><td></td><td></td></tr><tr><td>240d5390af19bb43761f112b0209771f19bfb696</td><td></td><td></td><td></td></tr><tr><td>24f9248f01df3020351347c2a3f632e01de72090</td><td>Reconstructing a Fragmented Face from a +<br/>Cryptographic Identification Protocol +<br/><b>The University of Texas at Austin</b></td><td>('39573884', 'Andy Luong', 'andy luong')<br/>('2499821', 'Michael Gerbush', 'michael gerbush')<br/>('1715120', 'Brent Waters', 'brent waters')<br/>('1794409', 'Kristen Grauman', 'kristen grauman')</td><td>aluong,mgerbush,bwaters,grauman@cs.utexas.edu +</td></tr><tr><td>24e099e77ae7bae3df2bebdc0ee4e00acca71250</td><td>Robust face alignment under occlusion via regional predictive power +<br/>estimation. +<br/>© 2015 IEEE +<br/>For additional information about this publication click this link. +<br/>http://qmro.qmul.ac.uk/xmlui/handle/123456789/22467 +<br/>Information about this research object was correct at the time of download; we occasionally +<br/>make corrections to records, please therefore check the published record when citing. For +</td><td>('2966679', 'Heng Yang', 'heng yang')</td><td>more information contact scholarlycommunications@qmul.ac.uk +</td></tr><tr><td>24959d1a9c9faf29238163b6bcaf523e2b05a053</td><td>High accuracy head pose tracking survey +<br/><b>Warsaw University of Technology, Poland</b></td><td>('1899063', 'Adam Strupczewski', 'adam strupczewski')</td><td></td></tr><tr><td>24f1febcdf56cd74cb19d08010b6eb5e7c81c362</td><td>Synergistic Methods for using Language in Robotics +<br/>Ching L. Teo +<br/><b>University of Maryland</b><br/>Dept of Computer Science +<br/><b>College Park, Maryland</b><br/>+01 3014051762 +<br/><b>University of Maryland</b><br/>Dept of Computer Science +<br/><b>College Park, Maryland</b><br/>+01 3014051762 +<br/><b>University of Maryland</b><br/><b>Institute for Advanced</b><br/>Computer Studies +<br/><b>College Park, Maryland</b><br/>+01 3014051743 +<br/><b>University of Maryland</b><br/>Dept of Computer Science +<br/><b>College Park, Maryland</b><br/>+01 3014051768 +</td><td>('7607499', 'Yezhou Yang', 'yezhou yang')<br/>('1759899', 'Cornelia Fermüller', 'cornelia fermüller')<br/>('1697493', 'Yiannis Aloimonos', 'yiannis aloimonos')</td><td>cteo@cs.umd.edu +<br/>yzyang@cs.umd.edu +<br/>fer@umiacs.umd.edu +<br/>yiannis@cs.umd.edu +</td></tr><tr><td>2450c618cca4cbd9b8cdbdb05bb57d67e63069b1</td><td>A Connexionist Approach for Robust and Precise Facial Feature Detection in +<br/>Complex Scenes +<br/>Stefan Duffner and Christophe Garcia +<br/>France Telecom Research & Development +<br/>4, rue du Clos Courtel +<br/>35512 Cesson-S´evign´e, France +</td><td></td><td>fstefan.duffner, christophe.garciag@francetelecom.com +</td></tr><tr><td>244b57cc4a00076efd5f913cc2833138087e1258</td><td>Warped Convolutions: Efficient Invariance to Spatial Transformations +</td><td>('1687524', 'Andrea Vedaldi', 'andrea vedaldi')</td><td></td></tr><tr><td>24cf9fe9045f50c732fc9c602358af89ae40a9f7</td><td>YANG et al.: ATTRIBUTE RECOGNITION FROM ADAPTIVE PARTS +<br/>Attribute Recognition from Adaptive Parts +<br/>Ligeng Zhu2 +<br/><b>Simon Fraser University</b><br/>Vancouver, Canada +<br/><b>Zhejiang University</b><br/>Hangzhou, China +<br/>3 Microsoft Research Asia +<br/>Beijing, China +<br/><b>Tongji University</b><br/>Shanghai, China +</td><td>('3202074', 'Luwei Yang', 'luwei yang')<br/>('1732264', 'Yichen Wei', 'yichen wei')<br/>('1729017', 'Shuang Liang', 'shuang liang')<br/>('37291674', 'Ping Tan', 'ping tan')</td><td>luweiy@sfu.ca +<br/>zhuligeng@zju.edu.cn +<br/>yichenw@microsoft.com +<br/>shuangliang@tongji.edu.cn +<br/>pingtan@sfu.ca +</td></tr><tr><td>24f022d807352abf071880877c38e53a98254dcd</td><td>Are screening methods useful in feature selection? An +<br/>empirical study +<br/><b>Florida State University, Tallahassee, Florida, U.S.A</b></td><td>('6693611', 'Mingyuan Wang', 'mingyuan wang')<br/>('2455529', 'Adrian Barbu', 'adrian barbu')</td><td>* abarbu@stat.fsu.edu +</td></tr><tr><td>241d2c517dbc0e22d7b8698e06ace67de5f26fdf</td><td>Online, Real-Time Tracking +<br/>Using a Category-to-Individual Detector(cid:2) +<br/><b>California Institute of Technology, USA</b></td><td>('1990633', 'David Hall', 'david hall')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>{dhall,perona}@vision.caltech.edu +</td></tr><tr><td>24869258fef8f47623b5ef43bd978a525f0af60e</td><td><b>UNIVERSITÉDEGRENOBLENoattribuéparlabibliothèqueTHÈSEpourobtenirlegradedeDOCTEURDEL’UNIVERSITÉDEGRENOBLESpécialité:MathématiquesetInformatiquepréparéeauLaboratoireJeanKuntzmanndanslecadredel’ÉcoleDoctoraleMathématiques,SciencesetTechnologiesdel’Information,InformatiqueprésentéeetsoutenuepubliquementparMatthieuGuillauminle27septembre2010ExploitingMultimodalDataforImageUnderstandingDonnéesmultimodalespourl’analysed’imageDirecteursdethèse:CordeliaSchmidetJakobVerbeekJURYM.ÉricGaussierUniversitéJosephFourierPrésidentM.AntonioTorralbaMassachusettsInstituteofTechnologyRapporteurMmeTinneTuytelaarsKatholiekeUniversiteitLeuvenRapporteurM.MarkEveringhamUniversityofLeedsExaminateurMmeCordeliaSchmidINRIAGrenobleExaminatriceM.JakobVerbeekINRIAGrenobleExaminateur</b></td><td></td><td></td></tr><tr><td>24e6a28c133b7539a57896393a79d43dba46e0f6</td><td>ROBUST BAYESIAN METHOD FOR SIMULTANEOUS BLOCK SPARSE SIGNAL +<br/>RECOVERY WITH APPLICATIONS TO FACE RECOGNITION +<br/>Department of Electrical and Computer Engineering +<br/><b>University of California, San Diego</b></td><td>('32352411', 'Igor Fedorov', 'igor fedorov')<br/>('3291075', 'Ritwik Giri', 'ritwik giri')<br/>('1748319', 'Bhaskar D. Rao', 'bhaskar d. rao')<br/>('1690269', 'Truong Q. Nguyen', 'truong q. nguyen')</td><td></td></tr><tr><td>248db911e3a6a63ecd5ff6b7397a5d48ac15e77a</td><td>Enriching Texture Analysis with Semantic Data +<br/>Communications, Signal Processing and Control Group +<br/>School of Electronics and Computer Science +<br/><b>University of Southampton</b></td><td>('28637223', 'Tim Matthews', 'tim matthews')<br/>('1727698', 'Mark S. Nixon', 'mark s. nixon')<br/>('1697360', 'Mahesan Niranjan', 'mahesan niranjan')</td><td>{tm1e10,msn,mn}@soton.ac.uk +</td></tr><tr><td>24d376e4d580fb28fd66bc5e7681f1a8db3b6b78</td><td></td><td></td><td></td></tr><tr><td>24f1e2b7a48c2c88c9e44de27dc3eefd563f6d39</td><td>Recognition of Action Units in the Wild +<br/>with Deep Nets and a New Global-Local Loss +<br/>C. Fabian Benitez-Quiroz +<br/>Aleix M. Martinez +<br/>Dept. Electrical and Computer Engineering +<br/><b>The Ohio State University</b></td><td>('1678691', 'Yan Wang', 'yan wang')</td><td>{benitez-quiroz.1,wang.9021,martinez.158}@osu.edu +</td></tr><tr><td>243e9d490fe98d139003bb8dc95683b366866c57</td><td>Distinctive Parts for Relative attributes +<br/>Thesis submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/>Master of science( by research) +<br/>in +<br/>Computer Science Engineering +<br/>by +<br/>Ramachandruni Naga Sandeep +<br/>201207582 +<br/>Center for Visual Information Technology +<br/><b>International Institute of Information Technology</b><br/>Hyderabad - 500 032, INDIA +<br/>December 2014 +</td><td></td><td>nsandeep.ramachandruni@research.iiit.ac.in +</td></tr><tr><td>2465fc22e03faf030e5a319479a95ef1dfc46e14</td><td>______________________________________________________PROCEEDING OF THE 20TH CONFERENCE OF FRUCT ASSOCIATION +<br/>Influence of Different Feature Selection Approaches +<br/>on the Performance of Emotion Recognition +<br/>Methods Based on SVM +<br/><b>Ural Federal University (UrFU</b><br/>Yekaterinburg, Russia +</td><td>('11063038', 'Daniil Belkov', 'daniil belkov')<br/>('3457868', 'Konstantin Purtov', 'konstantin purtov')</td><td>d.d.belkov, k.s.purtov@gmail.com, kublanov@mail.ru +</td></tr><tr><td>24ff832171cb774087a614152c21f54589bf7523</td><td>Beat-Event Detection in Action Movie Franchises +<br/>Jerome Revaud +<br/>Zaid Harchaoui +</td><td>('2319574', 'Danila Potapov', 'danila potapov')<br/>('3271933', 'Matthijs Douze', 'matthijs douze')<br/>('2462253', 'Cordelia Schmid', 'cordelia schmid')</td><td></td></tr><tr><td>247a6b0e97b9447850780fe8dbc4f94252251133</td><td>Facial Action Unit Detection: 3D versus 2D Modality +<br/>Electrical and Electronics Engineering +<br/><b>Bo gazic i University, Istanbul, Turkey</b><br/>B¨ulent Sankur +<br/>Electrical and Electronics Engineering +<br/><b>Bo gazic i University, Istanbul, Turkey</b><br/>Department of Psychology +<br/><b>Bo gazic i University, Istanbul, Turkey</b></td><td>('1839621', 'Arman Savran', 'arman savran')<br/>('27414819', 'M. Taha Bilge', 'm. taha bilge')</td><td>arman.savran@boun.edu.tr +<br/>bulent.sankur@boun.edu.tr +<br/>taha.bilge@boun.edu.tr +</td></tr><tr><td>24bf94f8090daf9bda56d54e42009067839b20df</td><td></td><td></td><td></td></tr><tr><td>240eb0b34872c431ecf9df504671281f59e7da37</td><td>Cutout-Search: Putting a Name to the Picture +<br/><b>Carnegie Mellon University</b><br/><b>Cornell University</b></td><td>('1746610', 'Dhruv Batra', 'dhruv batra')<br/>('2371390', 'Adarsh Kowdle', 'adarsh kowdle')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')<br/>('1713589', 'Devi Parikh', 'devi parikh')</td><td>batradhruv@cmu.edu +<br/>apk64@cornell.edu dparikh@cmu.edu tsuhan@ece.cornell.edu +</td></tr><tr><td>230527d37421c28b7387c54e203deda64564e1b7</td><td>Person Re-identification: System Design and +<br/>Evaluation Overview +</td><td>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('40156369', 'Rui Zhao', 'rui zhao')</td><td></td></tr><tr><td>23fdbef123bcda0f07d940c72f3b15704fd49a98</td><td></td><td></td><td></td></tr><tr><td>23ebbbba11c6ca785b0589543bf5675883283a57</td><td></td><td></td><td></td></tr><tr><td>23aef683f60cb8af239b0906c45d11dac352fb4e</td><td>Incorporating Context Information into Deep +<br/>Neural Network Acoustic Models +<br/>July 2016 +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>Thesis Committee: +<br/><b>Florian Metze, Chair (Carnegie Mellon University</b><br/><b>Alan W Black (Carnegie Mellon University</b><br/><b>Alex Waibel (Carnegie Mellon University</b><br/>Jinyu Li (Microsoft) +<br/>Submitted in partial fulfillment of the requirements +<br/>for the degree of Doctor of Philosophy. +</td><td>('37467623', 'Yajie Miao', 'yajie miao')<br/>('37467623', 'Yajie Miao', 'yajie miao')</td><td></td></tr><tr><td>235d5620d05bb7710f5c4fa6fceead0eb670dec5</td><td>Who’s Doing What: Joint Modeling of Names and +<br/>Verbs for Simultaneous Face and Pose Annotation +<br/>Luo Jie +<br/>Idiap and EPF Lausanne +<br/><b>Idiap Research Institute</b><br/>ETH Zurich +</td><td>('3033284', 'Barbara Caputo', 'barbara caputo')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')</td><td>jluo@idiap.ch +<br/>bcaputo@idiap.ch +<br/>ferrari@vision.ee.ethz.ch +</td></tr><tr><td>23ce6f404c504592767b8bec7d844d87b462de71</td><td>A Deep Face Identification Network Enhanced by Facial Attributes Prediction +<br/><b>West Virginia University</b></td><td>('34708406', 'Fariborz Taherkhani', 'fariborz taherkhani')<br/>('8147588', 'Nasser M. Nasrabadi', 'nasser m. nasrabadi')</td><td>ft0009@mix.wvu.edu, nasser.nasrabadi@mail.wvu.edu, Jeremy.Dawson@mail.wvu.edu +</td></tr><tr><td>23fd653b094c7e4591a95506416a72aeb50a32b5</td><td>Emotion Recognition using Fuzzy Rule-based System +<br/>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 93 – No.11, May 2014 +<br/>Department of Computer Science +<br/><b>Amity University, Lucknow, India</b><br/>Faculty in Department Of Computer Science +<br/><b>Amity University, Lucknow, India</b><br/> +<br/> +</td><td>('14559473', 'Akanksha Chaturvedi', 'akanksha chaturvedi')</td><td></td></tr><tr><td>23172f9a397f13ae1ecb5793efd81b6aba9b4537</td><td>Proceedings of the 2015 Workshop on Vision and Language (VL’15), pages 10–17, +<br/>Lisbon, Portugal, 18 September 2015. c(cid:13)2015 Association for Computational Linguistics. +<br/>10 +</td><td></td><td></td></tr><tr><td>231a6d2ee1cc76f7e0c5912a530912f766e0b459</td><td>Shape Primitive Histogram: A Novel Low-Level Face Representation for Face +<br/>Recognition +<br/><b>aCollege of Computer Science at Chongqing University, 400044, Chongqing, P.R.C</b><br/>bSchool of Software Engineering at Chongqing Univeristy,400044,Chongqing,P.R.C +<br/><b>cSchool of Astronautics at Beihang University, 100191, Beijing, P.R.C</b><br/>dState Key Laboratory of Management and Control for Complex Systems +<br/><b>Institute of Automation, Chinese Academy of Sciences, 100190, Beijing, P.R.C</b><br/>eMinistry of Education Key Laboratory of Dependable Service Computing in Cyber Physical Society, 400044, Chongqing, P.R.C +</td><td>('1786011', 'Sheng Huang', 'sheng huang')<br/>('1698431', 'Dan Yang', 'dan yang')<br/>('1737368', 'Haopeng Zhang', 'haopeng zhang')</td><td></td></tr><tr><td>236a4f38f79a4dcc2183e99b568f472cf45d27f4</td><td>1632 +<br/>Randomized Clustering Forests +<br/>for Image Classification +<br/>Frederic Jurie, Member, IEEE Computer Society +</td><td>('3128253', 'Frank Moosmann', 'frank moosmann')<br/>('1975110', 'Eric Nowak', 'eric nowak')</td><td></td></tr><tr><td>230c4a30f439700355b268e5f57d15851bcbf41f</td><td>EM Algorithms for Weighted-Data Clustering +<br/>with Application to Audio-Visual Scene Analysis +</td><td>('1780201', 'Xavier Alameda-Pineda', 'xavier alameda-pineda')<br/>('1785817', 'Florence Forbes', 'florence forbes')<br/>('1794229', 'Radu Horaud', 'radu horaud')</td><td></td></tr><tr><td>237fa91c8e8098a0d44f32ce259ff0487aec02cf</td><td>IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 36, NO. 4, AUGUST 2006 +<br/>863 +<br/>Bidirectional PCA With Assembled Matrix +<br/>Distance Metric for Image Recognition +</td><td>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('1711542', 'Kuanquan Wang', 'kuanquan wang')</td><td></td></tr><tr><td>23fc83c8cfff14a16df7ca497661264fc54ed746</td><td><b>The Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA, USA 15213 +<br/>http://www.cs.cmu.edu/~face +<br/>Department of Psychology +<br/><b>University of Pittsburgh</b><br/><b>The Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>4015 O'Hara Street +<br/>Pittsburgh, PA, USA 15260 +<br/>Yingli Tian +<br/><b>The Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA, USA 15213 +<br/>Proceedings of the Fourth IEEE International Conference on Automatic Face and Gesture Recognition +<br/>(FG'00), pp. 484-490, Grenoble, France. +<br/>Comprehensive Database for Facial Expression Analysis +</td><td>('1733113', 'Takeo Kanade', 'takeo kanade')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td>tk@cs.cmu.edu +<br/>yltian@cs.cmu.edu +<br/>jeffcohn+@pitt.edu +</td></tr><tr><td>2331df8ca9f29320dd3a33ce68a539953fa87ff5</td><td>Extended Isomap for Pattern Classification +<br/><b>Honda Fundamental Research Labs</b><br/>Mountain View, CA 94041 +</td><td>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td>myang@hra.com +</td></tr><tr><td>232b6e2391c064d483546b9ee3aafe0ba48ca519</td><td>Optimization problems for fast AAM fitting in-the-wild +<br/>1. School of Computer Science +<br/><b>University of Lincoln, U.K</b><br/>2. Department of Computing +<br/><b>Imperial College London, U.K</b></td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')</td><td>gtzimiropoulos@lincoln.ac.uk +</td></tr><tr><td>23ba9e462151a4bf9dfc3be5d8b12dbcfb7fe4c3</td><td>CS 229 Project, Fall 2014 +<br/>Determining Mood from Facial Expressions +<br/>Introduction +<br/>I +<br/>Facial expressions play an extremely important role in human communication. As +<br/>society continues to make greater use of human-machine interactions, it is important for +<br/>machines to be able to interpret facial expressions in order to improve their +<br/>authenticity. If machines can be trained to determine mood to a better extent than +<br/>humans can, especially for more subtle moods, then this could be useful in fields such as +<br/>counseling. This could also be useful for gauging reactions of large audiences in various +<br/>contexts, such as political talks. +<br/>The results of this project could also be applied to recognizing other features of facial +<br/>expressions, such as determining when people are purposefully suppressing emotions or +<br/>lying. The ability to recognize different facial expressions could also improve technology +<br/>that recognizes to whom specific faces belong. This could in turn be used to search a +<br/>large number of pictures for a specific photo, which is becoming increasingly difficult, as +<br/>storing photos digitally has been extremely common in the past decade. The possibilities +<br/>are endless. +<br/>II Data and Features +<br/>2.1 Data +<br/>Our data consists of 1166 frontal images of +<br/>people’s faces from three databases, with each +<br/>image labeled with one of eight emotions: +<br/>anger, contempt, disgust, fear, happiness, +<br/>neutral, sadness, and surprise. The TFEID [1], +<br/>CK+ [2], and JAFFE [3] databases primarily +<br/>consist of Taiwanese, Caucasian, and Japanese +<br/>subjects, respectively. The TFEID and JAFFE +<br/>images are both cropped with the faces +<br/>centered. Each image has a subject posing with +<br/>one of the emotions. The JAFFE database does +<br/>not have any images for contempt. +<br/>2.2 Features +<br/>On each face, there are many different facial landmarks. While some of these landmarks +<br/>(pupil position, nose tip, and face contour) are not as indicative of emotion, others +<br/>(eyebrow, mouth, and eye shape) are. To extract landmark data from images, we used +<br/>Happiness +<br/>Figure 1 +<br/>Anger +</td><td>('34482382', 'Matthew Wang', 'matthew wang')</td><td>mmwang@stanford.edu +<br/>spencery@stanford.edu +</td></tr><tr><td>237eba4822744a9eabb121fe7b50fd2057bf744c</td><td>Facial Expression Synthesis Using PAD Emotional +<br/>Parameters for a Chinese Expressive Avatar +<br/>1 Department of Computer Science and Technology +<br/><b>Tsinghua University, 100084 Beijing, China</b><br/>2 Department of Systems Engineering and Engineering Management +<br/><b>The Chinese University of Hong Kong, HKSAR, China</b></td><td>('2180849', 'Shen Zhang', 'shen zhang')<br/>('3860920', 'Zhiyong Wu', 'zhiyong wu')<br/>('1702243', 'Helen M. Meng', 'helen m. meng')<br/>('7239047', 'Lianhong Cai', 'lianhong cai')</td><td>zhangshen05@mails.tsinghua.edu.cn, john.zy.wu@gmail.com +<br/>hmmeng@se.cuhk.edu.hk, clh-dcs@tsinghua.edu.cn +</td></tr><tr><td>238fc68b2e0ef9f5ec043d081451902573992a03</td><td>2656 +<br/>Enhanced Local Gradient Order Features and +<br/>Discriminant Analysis for Face Recognition +<br/>role in robust face recognition [5]. Many algorithms have +<br/>been proposed to deal with the effectiveness of feature design +<br/>and extraction [6], [7]; however, the performance of many +<br/>existing methods is still highly sensitive to variations of +<br/>imaging conditions, such as outdoor illumination, exaggerated +<br/>expression, and continuous occlusion. These complex varia- +<br/>tions are significantly affecting the recognition accuracy in +<br/>recent years [8]–[10]. +<br/>Appearance-based subspace learning is one of the sim- +<br/>plest approach for feature extraction, and many methods +<br/>are usually based on linear correlation of pixel intensities. +<br/>For example, Eigenface [11] uses eigen system of pixel +<br/>intensities to estimate the lower rank linear subspace of +<br/>a set of training face images by minimizing the (cid:2)2 dis- +<br/>tance metric. The solution enjoys optimality properties when +<br/>noise is independent +<br/>identically distributed Gaussian only. +<br/>Fisherface [12] will suffer more due to the estimation of +<br/>inverse within-class covariance matrix [13], +<br/>thus the per- +<br/>formance will degenerate rapidly in the cases of occlusion +<br/>and small sample size. Laplacianfaces [14] refer to another +<br/>appearance-based approach which learns a locality preserv- +<br/>ing subspace and seeks to capture the intrinsic geometry +<br/>and local structure of the data. Other methods such as those +<br/>in [5] and [15] also provide valuable approaches to supervised +<br/>or unsupervised dimension reduction tasks. +<br/>A fundamental problem of appearance-based methods for +<br/>face recognition, however, is that they are sensitive to imag- +<br/>ing conditions [10]. As for data corrupted by illumination +<br/>changes, occlusions, and inaccurate alignment, the estimated +<br/>subspace will be biased, thus much of the efforts concentrate +<br/>on removing/shrinking the noise components. In contrast, local +<br/>feature descriptors [15]–[19] have certain advantages as they +<br/>are more stable to local changes. In the view of image pro- +<br/>cessing and vision, the basic imaging system can be simply +<br/>formulated as +<br/>(x, y) = A(x, y) × L(x, y) +<br/>(1) +</td><td>('1688667', 'Chuan-Xian Ren', 'chuan-xian ren')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('1726138', 'Dao-Qing Dai', 'dao-qing dai')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td></td></tr><tr><td>2322ec2f3571e0ddc593c4e2237a6a794c61251d</td><td>Jack, R. E. , Sun, W., Delis, I., Garrod, O. G. B. and Schyns, P. G. (2016) +<br/>Four not six: revealing culturally common facial expressions of +<br/>emotion.Journal of Experimental Psychology: General, 145(6), pp. 708- +<br/>730. (doi:10.1037/xge0000162) +<br/>This is the author’s final accepted version. +<br/>There may be differences between this version and the published version. +<br/>You are advised to consult the publisher’s version if you wish to cite from +<br/>it. +<br/>http://eprints.gla.ac.uk/116592/ +<br/> +<br/>Deposited on: 20 April 2016 +<br/><b>Enlighten Research publications by members of the University of Glasgow</b><br/>http://eprints.gla.ac.uk +</td><td></td><td></td></tr><tr><td>23e75f5ce7e73714b63f036d6247fa0172d97cb6</td><td>BioMed Central +<br/>Research +<br/>Facial expression (mood) recognition from facial images using +<br/>committee neural networks +<br/>Open Access +<br/><b>University of Akron, Akron</b><br/><b>Engineering, University of Akron, Akron, OH 44325-3904, USA</b><br/>* Corresponding author +<br/>Published: 5 August 2009 +<br/>doi:10.1186/1475-925X-8-16 +<br/>Received: 24 September 2008 +<br/>Accepted: 5 August 2009 +<br/>This article is available from: http://www.biomedical-engineering-online.com/content/8/1/16 +<br/>© 2009 Kulkarni et al; licensee BioMed Central Ltd. +<br/>This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +</td><td>('39890387', 'Saket S Kulkarni', 'saket s kulkarni')<br/>('2484370', 'Narender P Reddy', 'narender p reddy')<br/>('32173165', 'SI Hariharan', 'si hariharan')</td><td>Email: Saket S Kulkarni - saketkulkarni@gmail.com; Narender P Reddy* - npreddy@uakron.edu; SI Hariharan - hari@uakron.edu +</td></tr><tr><td>23429ef60e7a9c0e2f4d81ed1b4e47cc2616522f</td><td>A Domain Based Approach to Social Relation Recognition +<br/><b>Max Planck Institute for Informatics, Saarland Informatics Campus</b><br/>Figure 1: We investigate the recognition of social relations in a domain-based approach. Our study is based on Bugental’s +<br/>social psychology theory [1] that partitions social life into 5 domains from which we derive 16 social relations. +</td><td>('32222907', 'Qianru Sun', 'qianru sun')<br/>('1697100', 'Bernt Schiele', 'bernt schiele')<br/>('1739548', 'Mario Fritz', 'mario fritz')</td><td>{qsun, schiele, mfritz}@mpi-inf.mpg.de +</td></tr><tr><td>23aba7b878544004b5dfa64f649697d9f082b0cf</td><td>Locality-Constrained Discriminative Learning and Coding +<br/>1Department of Electrical & Computer Engineering, +<br/><b>College of Computer and Information Science</b><br/><b>Northeastern University, Boston, MA, USA</b></td><td>('7489165', 'Shuyang Wang', 'shuyang wang')<br/>('37771688', 'Yun Fu', 'yun fu')</td><td>{shuyangwang, yunfu}@ece.neu.edu +</td></tr><tr><td>23120f9b39e59bbac4438bf4a8a7889431ae8adb</td><td>Aalborg Universitet +<br/>Improved RGB-D-T based Face Recognition +<br/>Nikisins, Olegs; Sun, Yunlian; Li, Haiqing; Sun, Zhenan; Moeslund, Thomas B.; Greitans, +<br/>Modris +<br/>Published in: +<br/>DOI (link to publication from Publisher): +<br/>10.1049/iet-bmt.2015.0057 +<br/>Publication date: +<br/>2016 +<br/>Document Version +<br/>Accepted manuscript, peer reviewed version +<br/><b>Link to publication from Aalborg University</b><br/>Citation for published version (APA): +<br/>Oliu Simon, M., Corneanu, C., Nasrollahi, K., Guerrero, S. E., Nikisins, O., Sun, Y., ... Greitans, M. (2016). +<br/>General rights +<br/>Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners +<br/>and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. +<br/> ? Users may download and print one copy of any publication from the public portal for the purpose of private study or research. +<br/> ? You may not further distribute the material or use it for any profit-making activity or commercial gain +<br/> ? You may freely distribute the URL identifying the publication in the public portal ? +<br/>Take down policy +<br/>the work immediately and investigate your claim. +<br/>Downloaded from vbn.aau.dk on: October 11, 2016 +<br/> </td><td>('7855312', 'Sergio Escalera', 'sergio escalera')</td><td>If you believe that this document breaches copyright please contact us at vbn@aub.aau.dk providing details, and we will remove access to +</td></tr><tr><td>2303d07d839e8b20f33d6e2ec78d1353cac256cf</td><td>Squeeze-and-Excitation on Spatial and Temporal +<br/>Deep Feature Space for Action Recognition +<br/><b>Institute of Information Science, Beijing Jiaotong University, Beijing 100044, China</b><br/>Beijing Key Laboratory of Advanced Information Science and Network Technology, Beijing 100044, China +</td><td>('2896701', 'Gaoyun An', 'gaoyun an')<br/>('3027947', 'Wen Zhou', 'wen zhou')<br/>('47095962', 'Yuxuan Wu', 'yuxuan wu')<br/>('4464686', 'ZhenXing Zheng', 'zhenxing zheng')<br/>('46398737', 'Yongwen Liu', 'yongwen liu')</td><td>Email:{gyan, 16125155, 16120307, zhxzheng, 17120314}@bjtu.edu.cn +</td></tr><tr><td>23d55061f7baf2ffa1c847d356d8f76d78ebc8c1</td><td>Solmaz et al. IPSJ Transactions on Computer Vision and +<br/>Applications (2017) 9:22 +<br/>DOI 10.1186/s41074-017-0033-4 +<br/>IPSJ Transactions on Computer +<br/>Vision and Applications +<br/>RESEARCH PAPER +<br/>Open Access +<br/>Generic and attribute-specific deep +<br/>representations for maritime vessels +</td><td>('2827750', 'Berkan Solmaz', 'berkan solmaz')<br/>('2131286', 'Erhan Gundogdu', 'erhan gundogdu')<br/>('32499620', 'Aykut Koc', 'aykut koc')</td><td></td></tr><tr><td>23c3eb6ad8e5f18f672f187a6e9e9b0d94042970</td><td>Deep Domain Adaptation for Describing People Based on +<br/>Fine-Grained Clothing Attributes +<br/><b>IBM Research, Australia, 2 IBM T.J. Watson Research Center, 3 National University of Singapore</b><br/>Source domain +<br/>RCNN +<br/>body +<br/>detection +<br/>Alignment +<br/>cost layer +<br/>Multi-label +<br/>attributes +<br/>objective +<br/>Target domain +<br/>Alignment cost layer +<br/>Extra Info +<br/>(e.g. Labels) +<br/>Figure 1: Our proposed Deep Domain Adaptation Network (DDAN). +<br/>Source and target domains are modeled jointly with knowledge transfer oc- +<br/>curring at multiple levels of the hierarchy through alignment cost layers. +<br/>Describing people in detail is an important task for many applications. +<br/>For instance, criminal investigation processes often involve searching for +<br/>suspects based on detailed descriptions provided by eyewitnesses or com- +<br/>piled from images captured by surveillance cameras. The FBI list of na- +<br/>tionwide wanted bank robbers (https://bankrobbers.fbi.gov/) has clear exam- +<br/><b>ples of such ne-grained descriptions, including attributes covering detailed</b><br/>color information (e.g., “light blue” “khaki”, “burgundy”), a variety of cloth- +<br/>ing types (e.g., ‘leather jacket”, “polo-style shirt”, “zip-up windbreaker”) +<br/>and also detailed clothing patterns (e.g., “narrow horizontal stripes”, “LA +<br/>printed text”, “checkered”). +<br/>Traditional computer vision methods for describing people, however, +<br/>have only focused on a small set of coarse-grained attributes. As an exam- +<br/>ple, the recent work of Zhang et al. [7] achieves impressive attribute predic- +<br/>tion performance in unconstrained scenarios, but only considers nine human +<br/>attributes. Existing systems for fashion analysis [1, 4, 6] and people search +<br/>in surveillance videos [2, 5] also rely on a relatively small set of clothing +<br/>attributes. Our work instead addresses the problem of describing people +<br/>with very fine-grained clothing attributes. A natural question that arises in +<br/>this setting is how to obtain a sufficient number of training samples for each +<br/>attribute without significant annotation cost. +<br/>Data collection: We observe that online shopping stores such as Ama- +<br/>zon.com and TMALL.com have a large set of garment images with associ- +<br/>ated descriptions. We created a huge dataset of clothing images with fine- +<br/>grained attribute labels by crawling data from these shopping websites. Our +<br/>dataset contains 1,108,013 clothing images with 25 different kinds attribute +<br/>categories (e.g. +<br/>type, color, pattern, season, occasion). The attribute la- +<br/>bels are very fine-detailed. For instance, we can find thousands of different +<br/>values for the “color” category. After data curation, we considered a subset +<br/>of this data that is meaningful from our application perspective. +<br/>Deep Domain Adaptation: Although we have collected a large-scale +<br/>dataset with fine-grained attributes, these images are taken in ideal pose / +<br/>lighting / background conditions, so it is unreliable to directly use them as +<br/>training data for attribute prediction in the domain of unconstrained images +<br/>captured, for example, by mobile phones or surveillance cameras. In or- +<br/>der to bridge this gap, we design a specific double-path deep convolutional +<br/>neural network for the domain adaptation problem. Each path receives one +<br/>domain image as the input, i.e., the street domain and the shop domain im- +<br/>ages. Each path consists of several convolutional layers which are stacked +<br/>layer-by-layer and normally higher layers represent higher-level concept ab- +<br/>stractions. Both of the two network paths share the same architecture, e.g., +<br/>the same number of convolutional filters and number of middle layers. This +</td><td>('35370244', 'Qiang Chen', 'qiang chen')<br/>('1753492', 'Junshi Huang', 'junshi huang')<br/>('2106286', 'Jian Dong', 'jian dong')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td></td></tr><tr><td>23dd8d17ce09c22d367e4d62c1ccf507bcbc64da</td><td>Deep Density Clustering of Unconstrained Faces +<br/>(Supplementary Material) +<br/><b>University of Maryland, College Park</b><br/>A. Mathematical Details +<br/>Let S = {i | 0 < αi < C}. We have the following results: +<br/>nV(cid:88) +<br/>nV(cid:88) +<br/>i=1 +<br/>c∗ = +<br/>w∗ = +<br/>αiΨθ(xi), +<br/>¯R∗ = (cid:107)Ψθ(xs) − c∗(cid:107)2 , +<br/>αiΨθ(xi), +<br/>ρ∗ = w∗T Ψθ(xs), +<br/>where s ∈ S. Substituting into (3) and (4), we obtain +<br/>hSVDD(x) = 2 · hOC-SVM(x) = 2 +<br/>αiK(xi, x) − ρ∗ +<br/>(cid:34) nV(cid:88) +<br/>i=1 +<br/>(1) +<br/>(2) +<br/>(5) +<br/>(6) +<br/>(cid:35) +<br/>(7) +<br/>A.2. Proof of Theorem 1 +<br/>Theorem 1. If 1/nV < ν ≤ 1 and c∗T Ψθ(xs) (cid:54)= 0 for +<br/>some support vector xs, hSVDD(x) defined in (3) is asymp- +<br/>totically a Parzen window density estimator in the feature +<br/>space with Epanechnikov kernel. +<br/>Proof. Given the condition, according to Lemma 1, +<br/>hSVDD(x) is equivalent to hOC-SVM(x) with ρ∗ (cid:54)= 0. From +<br/>the results in [10] and the fact that(cid:80) αi = 1, we obtain: +<br/>(cid:21) +<br/>(cid:20) +<br/>hOC-SVM(x) = +<br/>αi +<br/>1 − 1 +<br/>(cid:107)Ψθ(x) − Ψθ(xi)(cid:107)2 +<br/>(cid:18)(cid:107)Ψθ(x) − Ψθ(xi)(cid:107) +<br/>(cid:19) +<br/>− ρ∗ +<br/>− ρ∗ − 1, +<br/>αiKE +<br/>nV(cid:88) +<br/>nV(cid:88) +<br/>i=1 +<br/>i=1 +<br/>4 (1 − u2), |u| ≤ 1 is the Epanechnikov +<br/>where KE(u) = 3 +<br/>kernel. As a consequence of Proposition 4 in [10] and the +<br/>proof of Proposition 1 in [11], as nV → ∞, the fraction +<br/>of support vector is ν, and the fraction of points with 0 < +<br/>αi < 1/(ν · nV ) vanishes. Therefore, either αi = 0 or +<br/>αi = 1/(ν · nV ). We introduce the notation ¯S = {i | αi = +<br/>ξ(z) +<br/>i=1 +<br/>In this section, we first provide the two core mathe- +<br/>matical formulations and then present detailed proofs for +<br/>Lemma 1 and Theorem 1. +<br/>SVDD formulation: +<br/>(cid:88) +<br/>z∈V (x) +<br/>¯R + +<br/>ν · nV +<br/>min +<br/>c, ¯R, ξ +<br/>s.t. +<br/>(cid:107)Ψθ(z) − c(cid:107)2 ≤ ¯R + ξ(z), +<br/>ξ ≥ 0, ∀z ∈ V (x), +<br/>OC-SVM formulation: +<br/>(cid:88) +<br/>min +<br/>w, ρ, ξ +<br/>s.t. +<br/>(cid:107)w(cid:107)2 + +<br/>ν · nV +<br/>wT Ψθ(z) ≥ ρ − ξz, +<br/>z∈V (x) +<br/>ξz − ρ +<br/>ξz ≥ 0, ∀z ∈ V (x). +<br/>A.1. Proof of Lemma 1 +<br/>Lemma 1. If 1/nV < ν ≤ 1, the SVDD formulation in (1) +<br/>is equivalent to the OC-SVM formulation in (2) when the +<br/>evaluation functions for the two are given by +<br/>hSVDD(x) = ¯R∗ − (cid:107)Ψθ(x) − c∗(cid:107)2 , +<br/>hOC-SVM(x) = w∗T Ψθ(x) − ρ∗, +<br/>(3) +<br/>(4) +<br/>with the correspondence w∗ = c∗, and ρ∗ = c∗T Ψθ(xs), +<br/>where xs is a support vector in (1) that lies on the learned +<br/>enclosing sphere. +<br/>Proof. The condition corresponds to the case 1/nV ≤ C < +<br/>1 in [1] with C = 1/(ν · nV ). We introduce the kernel +<br/>function K(xi, xj) = Ψθ(xi)T Ψθ(xj). Since K(xi, xi) +<br/>is constant in our setting, the same dual formulation for (1) +<br/>and (2) can be written as: +<br/>(cid:88) +<br/>min +<br/>αiαjK(xi, xj) +<br/>s.t. +<br/>0 ≤ αi ≤ C, +<br/>ij +<br/>i=1 +<br/>nV(cid:88) +<br/>αi = 1. +</td><td>('3329881', 'Wei-An Lin', 'wei-an lin')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')</td><td>walin@umd.edu pullpull@cs.umd.edu carlos@cs.umd.edu rama@umiacs.umd.edu +</td></tr><tr><td>23a8d02389805854cf41c9e5fa56c66ee4160ce3</td><td>Multimed Tools Appl +<br/>DOI 10.1007/s11042-013-1568-8 +<br/>Influence of low resolution of images on reliability +<br/>of face detection and recognition +<br/>© The Author(s) 2013. This article is published with open access at SpringerLink.com +</td><td>('2553748', 'Tomasz Marciniak', 'tomasz marciniak')<br/>('2009993', 'Radoslaw Weychan', 'radoslaw weychan')<br/>('40397247', 'Adam Dabrowski', 'adam dabrowski')</td><td></td></tr><tr><td>23b37c2f803a2d4b701e2f39c5f623b2f3e14d8e</td><td>Available Online at www.ijcsmc.com +<br/>International Journal of Computer Science and Mobile Computing +<br/>A Monthly Journal of Computer Science and Information Technology +<br/>ISSN 2320–088X +<br/> IJCSMC, Vol. 2, Issue. 4, April 2013, pg.646 – 649 +<br/>RESEARCH ARTICLE +<br/>Modified Approaches on Face Recognition +<br/>By using Multisensory Image +<br/><b>Bharath University, India</b><br/><b>Bharath University, India</b></td><td></td><td></td></tr><tr><td>4f9e00aaf2736b79e415f5e7c8dfebda3043a97d</td><td>Machine Audition: +<br/>Principles, Algorithms +<br/>and Systems +<br/><b>University of Surrey, UK</b><br/>InformatIon scIence reference +<br/>Hershey • New York +</td><td>('46314841', 'WenWu Wang', 'wenwu wang')</td><td></td></tr><tr><td>4fd29e5f4b7186e349ba34ea30738af7860cf21f</td><td></td><td></td><td></td></tr><tr><td>4f0d9200647042e41dea71c35eb59e598e6018a7</td><td><b></b><br/>Experiments of Image Retrieval Using Weak Attributes +<br/><b>Columbia University, New York, NY</b></td><td>('1815972', 'Felix X. Yu', 'felix x. yu')<br/>('1725599', 'Rongrong Ji', 'rongrong ji')<br/>('3138710', 'Ming-Hen Tsai', 'ming-hen tsai')<br/>('35984288', 'Guangnan Ye', 'guangnan ye')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>yfyuxinnan, rrji, yegng@ee.columbia.edu +<br/>xfminghen, sfchangg@cs.columbia.edu +</td></tr><tr><td>4f051022de100241e5a4ba8a7514db9167eabf6e</td><td>Face Parsing via a Fully-Convolutional Continuous +<br/>CRF Neural Network +</td><td>('48207414', 'Lei Zhou', 'lei zhou')<br/>('36300239', 'Zhi Liu', 'zhi liu')<br/>('1706670', 'Xiangjian He', 'xiangjian he')</td><td></td></tr><tr><td>4faded442b506ad0f200a608a69c039e92eaff11</td><td><b>STANBUL TECHNICAL UNIVERSITY INSTITUTE OF SCIENCE AND TECHNOLOGY</b><br/>FACE RECOGNITION UNDER VARYING +<br/>ILLUMINATION +<br/>Master Thesis by +<br/>Department : Computer Engineering +<br/>Programme: Computer Engineering +<br/>JUNE 2006 +</td><td>('1968256', 'Erald VUÇINI', 'erald vuçini')<br/>('1766445', 'Muhittin GÖKMEN', 'muhittin gökmen')</td><td></td></tr><tr><td>4f7967158b257e86d66bdabfdc556c697d917d24</td><td>Guaranteed Parameter Estimation of Discrete Energy +<br/>Minimization for 3D Scene Parsing +<br/>CMU-RI-TR-16-49 +<br/>July 2016 +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>Thesis Committee: +<br/>Daniel Huber, Advisor +<br/>Submitted in partial fulfillment of the requirements +<br/>for the degree of Master of Science in Robotics. +</td><td>('3439037', 'Mengtian Li', 'mengtian li')<br/>('1691629', 'Alexander J. Smola', 'alexander j. smola')<br/>('1786435', 'David Fouhey', 'david fouhey')<br/>('3439037', 'Mengtian Li', 'mengtian li')</td><td></td></tr><tr><td>4fc936102e2b5247473ea2dd94c514e320375abb</td><td>Guess Where? Actor-Supervision for Spatiotemporal Action Localization +<br/><b>KAUST1, University of Amsterdam2, Qualcomm Technologies, Inc</b></td><td>('2795139', 'Victor Escorcia', 'victor escorcia')<br/>('3409955', 'Cuong D. Dao', 'cuong d. dao')<br/>('40027484', 'Mihir Jain', 'mihir jain')<br/>('2931652', 'Bernard Ghanem', 'bernard ghanem')<br/>('1706203', 'Cees Snoek', 'cees snoek')</td><td></td></tr><tr><td>4f6adc53798d9da26369bea5a0d91ed5e1314df2</td><td>IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. , NO. , 2016 +<br/>Online Nonnegative Matrix Factorization with +<br/>General Divergences +</td><td>('2345985', 'Renbo Zhao', 'renbo zhao')<br/>('1678675', 'Huan Xu', 'huan xu')</td><td></td></tr><tr><td>4fbef7ce1809d102215453c34bf22b5f9f9aab26</td><td></td><td></td><td></td></tr><tr><td>4fa0d73b8ba114578744c2ebaf610d2ca9694f45</td><td></td><td></td><td></td></tr><tr><td>4fcd19b0cc386215b8bd0c466e42934e5baaa4b7</td><td>Human Action Recognition using Factorized Spatio-Temporal +<br/>Convolutional Networks +<br/><b>Hong Kong University of Science and Technology</b><br/><b>Hong Kong University of Science and Technology</b><br/><b>cid:93) Faculty of Science and Technology, University of Macau</b><br/>§ Lenovo Corporate Research Hong Kong Branch +</td><td>('1750501', 'Lin Sun', 'lin sun')<br/>('2370507', 'Kui Jia', 'kui jia')<br/>('1739816', 'Dit-Yan Yeung', 'dit-yan yeung')<br/>('2131088', 'Bertram E. Shi', 'bertram e. shi')</td><td>lsunece@ust.hk, kuijia@gmail.com, dyyeung@cse.ust.hk, eebert@ust.hk +</td></tr><tr><td>4f591e243a8f38ee3152300bbf42899ac5aae0a5</td><td>SUBMITTED TO TPAMI +<br/>Understanding Higher-Order Shape +<br/>via 3D Shape Attributes +</td><td>('1786435', 'David F. Fouhey', 'david f. fouhey')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>4f9958946ad9fc71c2299847e9ff16741401c591</td><td>Facial Expression Recognition with Recurrent Neural Networks +<br/>Robotics and Embedded Systems Lab, Department of Computer Science +<br/>Image Understanding and Knowledge-Based Systems, Department of Computer Science +<br/>Technische Universit¨at M¨unchen, Germany +</td><td>('1753223', 'Alex Graves', 'alex graves')<br/>('1685773', 'Christoph Mayer', 'christoph mayer')<br/>('32131501', 'Matthias Wimmer', 'matthias wimmer')<br/>('1699132', 'Bernd Radig', 'bernd radig')</td><td>[graves,juergen.schmidhuber]@in.tum.de +<br/>[mayerc,wimmerm,radig]@informatik.tu-muenchen.de +</td></tr><tr><td>4f773c8e7ca98ece9894ba3a22823127a70c6e6c</td><td>A Real-Time System for Head Tracking +<br/>and Pose Estimation +<br/><b>Robotics Institute, Carnegie Mellon University</b><br/>2 Electrical & Controls Integration Lab, General Motors R&D +</td><td>('29915644', 'Zengyin Zhang', 'zengyin zhang')<br/>('2918263', 'Minyoung Kim', 'minyoung kim')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')<br/>('9399514', 'Wende Zhang', 'wende zhang')</td><td></td></tr><tr><td>4ff11512e4fde3d1a109546d9c61a963d4391add</td><td>Proceedings of the Twenty-Ninth International +<br/>Florida Artificial Intelligence Research Society Conference +<br/>Selecting Vantage Points for an Autonomous Quadcopter Videographer +<br/>Google +<br/>Mountain View, CA +<br/>Gita Sukthankar +<br/><b>University of Central Florida</b><br/>Orlando, FL +<br/>Google +<br/>Mountain View, CA +</td><td>('3391381', 'Rey Coaguila', 'rey coaguila')<br/>('1694199', 'Rahul Sukthankar', 'rahul sukthankar')</td><td>reyc@google.com +<br/>gitars@eecs.ucf.edu +<br/>sukthankar@google.com +</td></tr><tr><td>4f028efe6708fc252851eee4a14292b7ce79d378</td><td>An Integrated Shape and Intensity Coding Scheme for Face Recognition +<br/>Department of Computer Science +<br/><b>George Mason University</b><br/>Fairfax, VA 22030-4444 +</td><td>('39664966', 'Chengjun Liu', 'chengjun liu')<br/>('1781577', 'Harry Wechsler', 'harry wechsler')</td><td>fcliu, wechslerg@cs.gmu.edu +</td></tr><tr><td>4f0bf2508ae801aee082b37f684085adf0d06d23</td><td></td><td></td><td></td></tr><tr><td>4ff4c27e47b0aa80d6383427642bb8ee9d01c0ac</td><td>Deep Convolutional Neural Networks and Support +<br/>Vector Machines for Gender Recognition +<br/><b>Institute of Arti cial Intelligence and Cognitive Engineering</b><br/>Faculty of Mathematics and Natural Sciences +<br/><b>University of Groningen, The Netherlands</b></td><td>('3405120', 'Jos van de Wolfshaar', 'jos van de wolfshaar')</td><td></td></tr><tr><td>4fefd1bc8dc4e0ab37ee3324ddfa43ad9d6a04a7</td><td>Fashion Landmark Detection in the Wild +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Key Lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced</b><br/>Technology, CAS, China +</td><td>('3243969', 'Ziwei Liu', 'ziwei liu')<br/>('1979911', 'Sijie Yan', 'sijie yan')<br/>('1693209', 'Ping Luo', 'ping luo')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{lz013,siyan,pluo,xtang}@ie.cuhk.edu.hk, xgwang@ee.cuhk.edu.hk +</td></tr><tr><td>4f4f920eb43399d8d05b42808e45b56bdd36a929</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 123 – No.4, August 2015 +<br/>A Novel Method for 3D Image Segmentation with Fusion +<br/>of Two Images using Color K-means Algorithm +<br/>Neelam Kushwah +<br/>Dept. of CSE +<br/>ITM Universe +<br/>Gwalior +<br/>Priusha Narwariya +<br/>Dept. of CSE +<br/>ITM Universe +<br/>Gwalior +<br/>two +</td><td></td><td></td></tr><tr><td>4f0b8f730273e9f11b2bfad2415485414b96299f</td><td>BDD100K: A Diverse Driving Video Database with +<br/>Scalable Annotation Tooling +<br/>1UC Berkeley +<br/><b>Georgia Institute of Technology</b><br/><b>Peking University</b><br/>4Uber AI Labs +</td><td>('1807197', 'Fisher Yu', 'fisher yu')<br/>('32324034', 'Fangchen Liu', 'fangchen liu')<br/>('8309711', 'Vashisht Madhavan', 'vashisht madhavan')<br/>('1753210', 'Trevor Darrell', 'trevor darrell')</td><td></td></tr><tr><td>4f77a37753c03886ca9c9349723ec3bbfe4ee967</td><td>Localizing Facial Keypoints with Global Descriptor Search, +<br/>Neighbour Alignment and Locally Linear Models +<br/>1 ´Ecole Polytechnique de Montr´eal, Universit´e de Montr´eal +<br/><b>University of Toronto and Recognyz Systems Technologies</b><br/>also focused on emotion recognition in the wild [9]. +</td><td>('1972076', 'Christopher Pal', 'christopher pal')<br/>('9422894', 'Sharon Moalem', 'sharon moalem')</td><td>md-kamrul.hasan@polymtl.ca, christohper.pal@polymtl.ca, sharon@recognyz.com +</td></tr><tr><td>4f7b92bd678772552b3c3edfc9a7c5c4a8c60a8e</td><td>Deep Density Clustering of Unconstrained Faces +<br/><b>University of Maryland, College Park</b></td><td>('3329881', 'Wei-An Lin', 'wei-an lin')<br/>('36407236', 'Jun-Cheng Chen', 'jun-cheng chen')</td><td>walin@umd.edu pullpull@cs.umd.edu carlos@cs.umd.edu rama@umiacs.umd.edu +</td></tr><tr><td>4f36c14d1453fc9d6481b09c5a09e91d8d9ee47a</td><td>DU,CHELLAPPA: VIDEO-BASED FACE RECOGNITION +<br/>Video-Based Face Recognition Using the +<br/>Intra/Extra-Personal Difference Dictionary +<br/>Department of Electrical and Computer +<br/>Engineering +<br/><b>University of Maryland</b><br/><b>College Park, USA</b></td><td>('35554856', 'Ming Du', 'ming du')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>mingdu@umd.edu +<br/>rama@umiacs.umd.edu +</td></tr><tr><td>8d71872d5877c575a52f71ad445c7e5124a4b174</td><td></td><td></td><td></td></tr><tr><td>8de06a584955f04f399c10f09f2eed77722f6b1c</td><td>Author manuscript, published in "International Conference on Computer Vision Theory and Applications (VISAPP 2013) (2013)" +</td><td></td><td></td></tr><tr><td>8d4f0517eae232913bf27f516101a75da3249d15</td><td>ARXIV SUBMISSION, MARCH 2018 +<br/>Event-based Dynamic Face Detection and +<br/>Tracking Based on Activity +</td><td>('2500521', 'Gregor Lenz', 'gregor lenz')<br/>('1773138', 'Sio-Hoi Ieng', 'sio-hoi ieng')<br/>('1750848', 'Ryad Benosman', 'ryad benosman')</td><td></td></tr><tr><td>8de2dbe2b03be8a99628ffa000ac78f8b66a1028</td><td>´Ecole Nationale Sup´erieure dInformatique et de Math´ematiques Appliqu´ees de Grenoble +<br/>INP Grenoble – ENSIMAG +<br/>UFR Informatique et Math´ematiques Appliqu´ees de Grenoble +<br/>Rapport de stage de Master 2 et de projet de fin d’´etudes +<br/>Effectu´e au sein de l’´equipe LEAR, I.N.R.I.A., Grenoble +<br/>Action Recognition in Videos +<br/>3e ann´ee ENSIMAG – Option I.I.I. +<br/>M2R Informatique – sp´ecialit´e I.A. +<br/>04 f´evrier 2008 – 04 juillet 2008 +<br/>LEAR, +<br/>I.N.R.I.A., Grenoble +<br/>655 avenue de l’Europe +<br/>38 334 Montbonnot +<br/>France +<br/>Responsable de stage +<br/>Mme. Cordelia Schmid +<br/>Tuteur ´ecole +<br/>Jury +</td><td>('16585941', 'Gaidon Adrien', 'gaidon adrien')<br/>('31899928', 'M. Augustin Lux', 'm. augustin lux')<br/>('12844736', 'Roger Mohr', 'roger mohr')<br/>('40419740', 'M. James Crowley', 'm. james crowley')</td><td></td></tr><tr><td>8d3fbdb9783716c1832a0b7ab1da6390c2869c14</td><td>12 +<br/>Discriminant Subspace Analysis for Uncertain +<br/>Situation in Facial Recognition +<br/><b>School of Computing and Communications University of Technology, Sydney</b><br/>Australia +<br/>1. Introduction +<br/>Facial analysis and recognition have received substential attention from researchers in +<br/>biometrics, pattern recognition, and computer vision communities. They have a large +<br/>number of applications, such as security, communication, and entertainment. Although a +<br/>great deal of efforts has been devoted to automated face recognition systems, it still remains +<br/>a challenging uncertainty problem. This is because human facial appearance has potentially +<br/>of very large intra-subject variations of head pose, illumination, facial expression, occlusion +<br/>due to other objects or accessories, facial hair and aging. These misleading variations may +<br/>cause classifiers to degrade generalization performance. +<br/>It is important for face recognition systems to employ an effective feature extraction scheme +<br/>to enhance separability between pattern classes which should maintain and enhance +<br/>features of the input data that make distinct pattern classes separable (Jan, 2004). In general, +<br/>there exist a number of different feature extraction methods. The most common feature +<br/>extraction methods are subspace analysis methods such as principle component analysis +<br/>(PCA) (Kirby & Sirovich, 1990) (Jolliffe, 1986) (Turk & Pentland, 1991b), kernel principle +<br/>component analysis (KPCA) (Schölkopf et al., 1998) (Kim et al., 2002) (all of which extract +<br/>the most informative features and reduce the feature dimensionality), Fisher’s linear +<br/>discriminant analysis (FLD) (Duda et al., 2000) (Belhumeur et al., 1997), and kernel Fisher’s +<br/>discriminant analysis (KFLD) (Mika et al., 1999) (Scholkopf & Smola, 2002) (which +<br/>discriminate different patterns; that is, they minimize the intra-class pattern compactness +<br/>while enhancing the extra-class separability). The discriminant analysis is necessary because +<br/>the patterns may overlap in decision space. +<br/>Recently, Lu et al. (Lu et al., 2003) stated that PCA and LDA are the most widely used +<br/>conventional tools for dimensionality reduction and feature extraction in the appearance- +<br/>based face recognition. However, because facial features are naturally non-linear and the +<br/>inherent linear nature of PCA and LDA, there are some limitations when applying these +<br/>methods to the facial data distribution (Bichsel & Pentland, 1994) (Lu et al., 2003). To +<br/>overcome such problems, nonlinear methods can be applied to better construct the most +<br/>discriminative subspace. +<br/>In real world applications, overlapping classes and various environmental variations can +<br/>significantly impact face recognition accuracy and robustness. Such misleading information +<br/>make Machine Learning difficult in modelling facial data. According to Adini et al. (Adini et +<br/>al., 1997), it is desirable to have a recognition system which is able to recognize a face +<br/>insensitive to these within-personal variations. +</td><td>('3333820', 'Pohsiang Tsai', 'pohsiang tsai')<br/>('2184946', 'Tich Phuoc Tran', 'tich phuoc tran')<br/>('1801256', 'Tom Hintz', 'tom hintz')<br/>('2567343', 'Tony Jan', 'tony jan')</td><td></td></tr><tr><td>8d42a24d570ad8f1e869a665da855628fcb1378f</td><td>CVPR +<br/>#987 +<br/>000 +<br/>001 +<br/>002 +<br/>003 +<br/>004 +<br/>005 +<br/>006 +<br/>007 +<br/>008 +<br/>009 +<br/>010 +<br/>011 +<br/>012 +<br/>013 +<br/>014 +<br/>015 +<br/>016 +<br/>017 +<br/>018 +<br/>019 +<br/>020 +<br/>021 +<br/>022 +<br/>023 +<br/>024 +<br/>025 +<br/>026 +<br/>027 +<br/>028 +<br/>029 +<br/>030 +<br/>031 +<br/>032 +<br/>033 +<br/>034 +<br/>035 +<br/>036 +<br/>037 +<br/>038 +<br/>039 +<br/>040 +<br/>041 +<br/>042 +<br/>043 +<br/>044 +<br/>045 +<br/>046 +<br/>047 +<br/>048 +<br/>049 +<br/>050 +<br/>051 +<br/>052 +<br/>053 +<br/>CVPR 2009 Submission #987. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +<br/>An Empirical Study of Context in Object Detection +<br/>Anonymous CVPR submission +<br/>Paper ID 987 +</td><td></td><td></td></tr><tr><td>8d8461ed57b81e05cc46be8e83260cd68a2ebb4d</td><td>Age identification of Facial Images using Neural +<br/>Network +<br/>CSE Department,CSVTU +<br/>RIT, Raipur, Chhattisgarh , INDIA +</td><td>('7530203', 'Sneha Thakur', 'sneha thakur')</td><td></td></tr><tr><td>8d4f12ed7b5a0eb3aa55c10154d9f1197a0d84f3</td><td>Cascaded Pose Regression +<br/>Piotr Doll´ar +<br/><b>California Institute of Technology</b></td><td>('2930640', 'Peter Welinder', 'peter welinder')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td>{pdollar,welinder,perona}@caltech.edu +</td></tr><tr><td>8de6deefb90fb9b3f7d451b9d8a1a3264b768482</td><td>Multibiometric Systems: Fusion Strategies and +<br/>Template Security +<br/>By +<br/>A Dissertation +<br/>Submitted to +<br/><b>Michigan State University</b><br/>in partial fulfillment of the requirements +<br/>for the degree of +<br/>Doctor of Philosophy +<br/>Department of Computer Science and Engineering +<br/>2008 +</td><td>('34633765', 'Karthik Nandakumar', 'karthik nandakumar')</td><td></td></tr><tr><td>8d2c0c9155a1ed49ba576ac0446ec67725468d87</td><td>A Study of Two Image Representations for Head Pose Estimation +<br/>Dept. of Computer Science and Technology, +<br/><b>Tsinghua University, Beijing, China</b></td><td>('1968464', 'Ligeng Dong', 'ligeng dong')<br/>('3265275', 'Linmi Tao', 'linmi tao')<br/>('1797002', 'Guangyou Xu', 'guangyou xu')</td><td>dongligeng99@mails.thu.edu.cn, +<br/>{linmi, xgy-dcs}@tsinghua.edu.cn +</td></tr><tr><td>8d384e8c45a429f5c5f6628e8ba0d73c60a51a89</td><td>Temporal Dynamic Graph LSTM for Action-driven Video Object Detection +<br/><b>The Hong Kong University of Science and Technology 2 Carneige Mellon University</b></td><td>('38937910', 'Yuan Yuan', 'yuan yuan')</td><td>yyuanad@ust.hk, xiaodan1@cs.cmu.edu, xiaolonw@cs.cmu.edu, dyyeung@cse.ust.hk, abhinavg@cs.cmu.edu +</td></tr><tr><td>8d0243b8b663ca0ab7cbe613e3b886a5d1c8c152</td><td>Development of Optical Computer Recognition (OCR) for Monitoring Stress and Emotions in Space +<br/><b>Center for Computational Biomedicine Imaging and Modeling Center, Rutgers University, New Brunswick, NJ</b><br/><b>USA, 2Unit for Experimental Psychiatry, University of Pennsylvania School of Medicine</b><br/>Philadelphia, PA, USA +<br/>INTRODUCTION. While in space, astronauts are required to perform mission-critical tasks on very expensive +<br/>equipment at a high level of functional capability. Stressors can compromise their ability to do so, thus it is very +<br/>important to have a system that can unobtrusively and objectively detect neurobehavioral problems involving +<br/>elevated levels of behavioral stress and negative emotions. Computerized approaches involving inexpensive cameras +<br/>offer an unobtrusive way to detect distress and to monitor observable emotions of astronauts during critical +<br/>operations in space, by tracking and analyzing facial expressions and body gestures in video streams. Such systems +<br/>can have applications beyond space flight, e.g., surveillance, law enforcement and human computer interaction. +<br/>TECHNOLOGY DEVELOPMENT. We developed a framework [1-9] that is capable of real time tracking of faces +<br/>and skin blobs of heads and hands. Face tracking uses a group of deformable statistical models of facial shape +<br/>variation and local texture distribution to robustly track facial landmarks (e.g., eyes, eyebrows, nose, mouth). The +<br/>model tolerates partial occlusions, it automatically detects and recovers from lost track, and it handles head rotations +<br/>up to full profile view. The skin blob tracker is initialized with a generic skin color model, dynamically learning the +<br/>specific color distribution online for adaptive tracking. Detected facial landmarks and blobs are filtered online, both +<br/>in terms of shape and motion, using eigenspace analysis and temporal dynamical models to prune false detections. +<br/>We then extract geometric and appearance features to learn models that detect relevant gestures and facial +<br/>expressions. In particular, our method utilizes the relative intensity ordering of facial expressions (i.e., neutral, onset, +<br/>apex, offset) found in the training set to learn a ranking model (Rankboost) for their recognition and intensity +<br/>estimation, which improves our average recognition rate (~87.5% on CMU benchmark database [4,10]). In relation +<br/>to stress detection, we piloted an experiment to learn subject-specific models of deception detection using behavioral +<br/>cues to discriminate stressed and relaxed behaviors. We video recorded 147 subjects in 12-question interviews after +<br/>a mock crime scenario, tracking their facial expressions and body gestures using our algorithm. Using leave-one-out +<br/>cross validation we acquired separate Nearest Neighbor models per subject, discriminating deceptive from truthful +<br/>responses with an average accuracy of 81.6% [7, 9]. We are currently experimenting with structured sparsity [14] +<br/>and super-resolution [11-13] techniques to obtain better quality image features to improve tracking and recognition +</td><td>('11788023', 'N. Michael', 'n. michael')<br/>('1748881', 'F. Yang', 'f. yang')<br/>('29384491', 'D. Metaxas', 'd. metaxas')</td><td></td></tr><tr><td>8d6c4af9d4c01ff47fe0be48155174158a9a5e08</td><td>Labeling, Discovering, and Detecting Objects in +<br/>Images +<br/>by +<br/>Bryan Christopher Russell +<br/>A.B., Computer Science +<br/><b>Dartmouth College</b><br/>S.M., Electrical Engineering and Computer Science +<br/><b>Massachusetts Institute of Technology</b><br/>Submitted to the Department of Electrical Engineering and Computer +<br/>in partial fulfillment of the requirements for the degree of +<br/>Doctor of Philosophy in Electrical Engineering and Computer Science +<br/>Science +<br/>at the +<br/><b>MASSACHUSETTS INSTITUTE OF TECHNOLOGY</b><br/>February 2008 +<br/>c(cid:13) Bryan Christopher Russell, MMVIII. All rights reserved. +<br/>The author hereby grants to MIT permission to reproduce and +<br/>distribute publicly paper and electronic copies of this thesis document +<br/>in whole or in part. +<br/>Author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Department of Electrical Engineering and Computer Science +<br/>January 28, 2007 +<br/>Certified by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>William T. Freeman +<br/>Professor +<br/>Thesis Supervisor +<br/>Accepted by . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +<br/>Terry P. Orlando +<br/>Chairman, Department Committee on Graduate Students +</td><td></td><td></td></tr><tr><td>8dcc95debd07ebab1721c53fa50d846fef265022</td><td>MicroExpNet: An Extremely Small and Fast Model For Expression Recognition +<br/>From Frontal Face Images +<br/>˙Ilke C¸ u˘gu, Eren S¸ener, Emre Akbas¸ +<br/><b>Middle East Technical University</b><br/>06800 Ankara, Turkey +</td><td></td><td>{cugu.ilke, sener.eren}@metu.edu.tr, emre@ceng.metu.edu.tr +</td></tr><tr><td>8dbe79830713925affc48d0afa04ed567c54724b</td><td></td><td></td><td></td></tr><tr><td>8d1adf0ac74e901a94f05eca2f684528129a630a</td><td>Facial Expression Recognition Using Facial +<br/>Movement Features +</td><td></td><td></td></tr><tr><td>8d91f06af4ef65193f3943005922f25dbb483ee4</td><td>Facial Expression Classification Using Rotation +<br/>Slepian-based Moment Invariants +<br/><b>Faculty of Science and Technology, University of Macau</b><br/>Macao, China +</td><td>('2888882', 'Cuiming Zou', 'cuiming zou')<br/>('3369665', 'Kit Ian Kou', 'kit ian kou')</td><td></td></tr><tr><td>8dc9de0c7324d098b537639c8214543f55392a6b</td><td>Pose-invariant 3d object recognition using linear combination of 2d views and +<br/>evolutionary optimisation +<br/>Department of Computer Science, +<br/><b>University College London</b><br/>Malet Place, London, WC1E 6BT +</td><td>('1797883', 'Vasileios Zografos', 'vasileios zografos')<br/>('31557997', 'Bernard F. Buxton', 'bernard f. buxton')</td><td>{v.zografos, b.buxton}@cs.ucl.ac.uk +</td></tr><tr><td>8d712cef3a5a8a7b1619fb841a191bebc2a17f15</td><td></td><td></td><td></td></tr><tr><td>8d646ac6e5473398d668c1e35e3daa964d9eb0f6</td><td>MEMORY-EFFICIENT GLOBAL REFINEMENT OF DECISION-TREE ENSEMBLES AND +<br/>ITS APPLICATION TO FACE ALIGNMENT +<br/>Nenad Markuˇs† +<br/>Ivan Gogi´c† +<br/>Igor S. Pandˇzi´c† +<br/>J¨orgen Ahlberg‡ +<br/><b>University of Zagreb, Faculty of Electrical Engineering and Computing, Unska 3, 10000 Zagreb, Croatia</b><br/><b>Computer Vision Laboratory, Link oping University, SE-581 83 Link oping, Sweden</b></td><td></td><td></td></tr><tr><td>8dffbb6d75877d7d9b4dcde7665888b5675deee1</td><td>Emotion Recognition with Deep-Belief +<br/>Networks +<br/>Introduction +<br/>For our CS229 project, we studied the problem of +<br/>reliable computerized emotion recognition in images of +<br/>human +<br/>faces. First, we performed a preliminary +<br/>exploration using SVM classifiers, and then developed an +<br/>approach based on Deep Belief Nets. Deep Belief Nets, or +<br/>DBNs, are probabilistic generative models composed of +<br/>multiple layers of stochastic latent variables, where each +<br/>“building block” layer is a Restricted Boltzmann Machine +<br/>(RBM). DBNs have a greedy layer-wise unsupervised +<br/>learning algorithm as well as a discriminative fine-tuning +<br/>procedure for optimizing performance on classification +<br/>tasks. [1]. +<br/>We trained our classifier on three databases: the +<br/>Cohn-Kanade Extended Database (CK+) [2], the Japanese +<br/>Female Facial Expression Database (JAFFE) [3], and the +<br/>Yale Face Database (YALE) [4]. We tested several +<br/>different database configurations, image pre-processing +<br/>settings, and DBN parameters, and obtained test errors as +<br/>low as 20% on a limited subset of the emotion labels. +<br/>Finally, we created a real-time system which takes +<br/>images of a single subject using a computer webcam and +<br/>classifies the emotion shown by the subject. +<br/>Part 1: Exploration of SVM-based approaches +<br/>To set a baseline for comparison, we applied an +<br/>SVM classifier to the emotion images in the CK+ +<br/>database, using the LIBLINEAR library and its MATLAB +<br/>interface [5]. This database contains 593 image sequences +<br/>across 123 human subjects, beginning with a “neutral +<br/>“expression and showing the progression to one of seven +<br/>“peak” emotions. When given both a neutral and an +<br/>expressive face to compare, the SVM obtained accuracy +<br/>as high as 90%. This +<br/>the +<br/>implementation of the SVM classifier. For additional +<br/>details on this stage of the project, please see our +<br/>Milestone document. +<br/>Part 1.1 Choice of labels (emotion numbers vs. FACS +<br/>features) +<br/>The CK+ database offers two sets of emotion +<br/>features: “emotion numbers” and FACS features. Emotion +<br/>numbers are integer values representing the main emotion +<br/>shown in the “peak emotion” image. The emotions are +<br/>coded as follows: 1=anger, 2=contempt, 3=disgust, +<br/>4=fear, 5=happiness, 6=sadness, and 7=surprise. +<br/>The other labeling option is called FACS, or the +<br/>Facial Action Coding System. FACS decomposes every +<br/>summarizes +<br/>section +<br/>facial emotion into a set of Action Units (AUs), which +<br/>describe the specific muscle groups involved in forming +<br/>the emotion. We chose not to use FACS because accurate +<br/>labeling currently requires trained human experts [8], and +<br/>we are interesting in creating an automated system. +<br/> +<br/>Part 1.2 Features +<br/>Part 1.2.1 Norm of differences between neutral face +<br/>and full emotion +<br/>Each of the CK+ images has been hand-labeled with +<br/>68 standard Active Appearance Models (AAM) face +<br/>landmarks that describe the X and Y position of these +<br/>landmarks on the image (Figure 1). +<br/>Figure 1. AAM Facial Landmarks +<br/>We initially trained the SVM on the norm of the +<br/>vector differences in landmark positions between the +<br/>neutral and peak expressions. With this approach, the +<br/>training error was approximately 35% for hold out cross +<br/>validation (see Figure 2). +<br/>with +<br/>Figure 3. Accuracy of +<br/>SVM with separate X, Y +<br/>displacement features. +<br/>Figure 2. Accuracy of +<br/>SVM +<br/>norm- +<br/>displacement features. +<br/>Part 1.2.2 Separate X and Y differences between +<br/>neutral face and full emotion +<br/>Because the initial approach did not differentiate +<br/>between displacements of +<br/>in different +<br/>directions, we also provided the differences in the X and +<br/>Y components of each landmark separately. This doubled +<br/>the size of our feature vector, and resulting in a significant +<br/>(about 20%) improvement in accuracy (Figure 3). +<br/>Part 1.2.3 Feature Selection +<br/>landmarks +<br/>Finally, we visualized which features were the most +<br/>important for classifying each emotion; the results can be +<br/>seen in Figure 4. The figure shows the X and Y +</td><td>('39818775', 'Tom McLaughlin', 'tom mclaughlin')</td><td></td></tr><tr><td>8d5998cd984e7cce307da7d46f155f9db99c6590</td><td>ChaLearn Looking at People: +<br/>A Review of Events and Resources +<br/>1 Dept. Mathematics and Computer Science, UB, Spain, +<br/>2 Computer Vision Center, UAB, Barcelona, Spain, +<br/><b>EIMT, Open University of Catalonia, Barcelona, Spain</b><br/>4 ChaLearn, California, USA, 5 INAOE, Puebla, Mexico, +<br/>6 Universit´e Paris-Saclay, Paris, France, +<br/>http://chalearnlap.cvc.uab.es +</td><td>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('1742688', 'Hugo Jair Escalante', 'hugo jair escalante')<br/>('1743797', 'Isabelle Guyon', 'isabelle guyon')</td><td>sergio.escalera.guerrero@gmail.com +</td></tr><tr><td>8dce38840e6cf5ab3e0d1b26e401f8143d2a6bff</td><td>Towards large scale multimedia indexing: +<br/>A case study on person discovery in broadcast news +<br/><b>Idiap Research Institute and EPFL, 2 LIMSI, CNRS, Univ. Paris-Sud, Universit Paris-Saclay</b><br/>3 CNRS, Irisa & Inria Rennes, 4 PUC de Minas Gerais, Belo Horizonte, +<br/><b>Universitat Polit cnica de Catalunya, 6 University of Vigo, 7 LIUM, University of Maine</b></td><td>('39560344', 'Nam Le', 'nam le')<br/>('2578933', 'Hervé Bredin', 'hervé bredin')<br/>('2710421', 'Gabriel Sargent', 'gabriel sargent')<br/>('2613332', 'Miquel India', 'miquel india')<br/>('1794658', 'Paula Lopez-Otero', 'paula lopez-otero')<br/>('1802247', 'Claude Barras', 'claude barras')<br/>('1804407', 'Camille Guinaudeau', 'camille guinaudeau')<br/>('1708671', 'Guillaume Gravier', 'guillaume gravier')<br/>('23556030', 'Gabriel Barbosa da Fonseca', 'gabriel barbosa da fonseca')<br/>('32255257', 'Izabela Lyon Freire', 'izabela lyon freire')<br/>('37401316', 'Gerard Martí', 'gerard martí')<br/>('2585946', 'Josep Ramon Morros', 'josep ramon morros')<br/>('1726311', 'Javier Hernando', 'javier hernando')<br/>('2446815', 'Sylvain Meignier', 'sylvain meignier')<br/>('1719610', 'Jean-Marc Odobez', 'jean-marc odobez')</td><td>nle@idiap.ch,bredin@limsi.fr,gabriel.sargent@irisa.fr,miquel.india@tsc.upc.edu,plopez@gts.uvigo.es +</td></tr><tr><td>153f5ad54dd101f7f9c2ae17e96c69fe84aa9de4</td><td>Overview of algorithms for face detection and +<br/>tracking +<br/>Nenad Markuˇs +</td><td></td><td></td></tr><tr><td>155199d7f10218e29ddaee36ebe611c95cae68c4</td><td>Towards Scalable Visual Navigation of +<br/>Micro Aerial Vehicles +<br/><b>Robotics Institute</b><br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/>April 2016 +<br/>Thesis Supervisors: +<br/>Prof. Dr. Martial Hebert +<br/>Prof. Dr. J. Andrew Bagnell +<br/>Submitted in partial fulfillment of the requirements +<br/>for the degree of Master of Science in Robotics. +<br/>CMU-RI-TR-16-07 +</td><td>('2739544', 'Shreyansh Daftry', 'shreyansh daftry')<br/>('2739544', 'Shreyansh Daftry', 'shreyansh daftry')</td><td>daftry@cmu.edu +</td></tr><tr><td>15cd05baa849ab058b99a966c54d2f0bf82e7885</td><td>Structured Sparse Subspace Clustering: A Unified Optimization Framework +<br/><b>SICE, Beijing University of Posts and Telecommunications. 2Center for Imaging Science, Johns Hopkins University</b><br/>In many real-world applications, we need to deal with high-dimensional +<br/>datasets, such as images, videos, text, and more. In practice, such high- +<br/>dimensional datasets can be well approximated by multiple low-dimensional +<br/>subspaces corresponding to multiple classes or categories. For example, the +<br/>feature point trajectories associated with a rigidly moving object in a video +<br/>lie in an affine subspace (of dimension up to 4), and face images of a subject +<br/>under varying illumination lie in a linear subspace (of dimension up to 9). +<br/>Therefore, the task, known in the literature as subspace clustering [6], is +<br/>to segment the data into the corresponding subspaces and finds multiple +<br/>applications in computer vision. +<br/>State of the art approaches [1, 2, 3, 4, 5, 7] for solving this problem fol- +<br/>low a two-stage approach: a) Construct an affinity matrix between points by +<br/>exploiting the ‘self-expressiveness’ property of the data, which allows any +<br/>data point to be represented as a linear (or affine) combination of the other +<br/>data points; b) Apply spectral clustering on the affinity matrix to recover +<br/>the data segmentation. Dividing the problem in two steps is, on the one +<br/>hand, appealing because the first step can be solved using convex optimiza- +<br/>tion techniques, while the second one can be solved using existing spectral +<br/>clustering techniques. On the other hand, its major disadvantage is that the +<br/>natural relationship between the affinity matrix and the segmentation of the +<br/>data is not explicitly captured. +<br/>In this paper, we attempt to integrate the two separate stages into one +<br/>unified optimization framework. One important motivating observation is +<br/>that a perfect subspace clustering can often be obtained from an imperfec- +<br/>t affinity matrix. In other words, the spectral clustering step can clean up +<br/>the disturbance in the affinity matrix – which can be viewed as a process of +<br/>information gain by denoising. Because of this, if we feed back the infor- +<br/>mation gain properly, it may help the self-expressiveness model to yield a +<br/>better affinity matrix. +<br/>To jointly estimate the clustering and affinity matrix, we define a sus- +<br/>pace structured (cid:96)1 norm as follows: +<br/>(cid:107)Z(cid:107)1,Q +<br/>= (cid:107)(11(cid:62) + αΘ)(cid:12) Z(cid:107)1 +<br/>(1) +<br/>where α > 0 is a tradeoff parameter, Θi j ∈ {0,1} indicates whether two data +<br/>points belong to the same subspace in which Θi j = 0 if point i and j lie in +<br/>the same subspace and otherwise Θi j = 1, and 1 is the vector of all ones of +<br/>appropriate dimension. +<br/>Equipped with the subspace structured (cid:96)1 norm of Z, we then define the +<br/>unified optimization framework for subspace clustering as follows: +<br/>min +<br/>Z,E,Q +<br/>(cid:107)Z(cid:107)1,Q + λ(cid:107)E(cid:107)(cid:96) s.t. X = XZ + E, diag(Z) = 0, Q ∈ Q, +<br/>where Q is the set of all valid binary segmentation matrices defined as +<br/>(2) +<br/>Q = {Q ∈ {0,1}N×k : Q1 = 1 and rank(Q) = k}, +<br/>(3) +<br/>and the norm (cid:107)·(cid:107)(cid:96) on the error term E depends upon the prior knowledge +<br/>about the pattern of noise or corruptions. We call problem (2) Structured +<br/>Sparse Subspace Clustering (SSSC or S3C). +<br/>The solution to the optimization problem in (2) is based on solving the +<br/>following two subproblems alternatively: a) Find Z and E given Q by solv- +<br/>ing a weighted sparse representation problem; b) Find Q given Z and E by +<br/>spectral clustering. We solve this problem efficiently via a combination of an +<br/>alternating direction method of multipliers with spectral clustering. Experi- +<br/>ments on a synthetic data, the Hopkins 155 motion segmentation database, +<br/>and the Extended Yale B data set demonstrate its effectiveness. +<br/>Some results are presented in Figure 1, Table 1 and 2. Figure 1 shows +<br/>the improvement in both the affinity matrix and the subspace clustering us- +<br/>ing S3C over SSC on a subset of face images of three subjects from the +</td><td>('9171002', 'Chun-Guang Li', 'chun-guang li')<br/>('1745721', 'René Vidal', 'rené vidal')</td><td></td></tr><tr><td>15136c2f94fd29fc1cb6bedc8c1831b7002930a6</td><td>Deep Learning Architectures for Face +<br/>Recognition in Video Surveillance +</td><td>('2805645', 'Saman Bashbaghi', 'saman bashbaghi')<br/>('1697195', 'Eric Granger', 'eric granger')<br/>('1744351', 'Robert Sabourin', 'robert sabourin')<br/>('3046171', 'Mostafa Parchami', 'mostafa parchami')</td><td></td></tr><tr><td>15affdcef4bb9d78b2d3de23c9459ee5b7a43fcb</td><td>Semi-Supervised Classification Using Linear +<br/>Neighborhood Propagation +<br/><b>Tsinghua University, Beijing 100084, P.R.China</b><br/><b>The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong</b><br/>Semi-Supervised Classi(cid:12)cation +<br/>A Toy Example +<br/>Shape Ranking +<br/>Digits Ranking +<br/>(a) +<br/>(b) +<br/>Interactive Image Segmentation +<br/>1.2 +<br/>0.8 +<br/>0.6 +<br/>0.4 +<br/>0.2 +<br/>−0.2 +<br/>−0.4 +<br/>−0.6 +<br/>−0.8 +<br/>−1.5 +<br/>1.2 +<br/>0.8 +<br/>0.6 +<br/>0.4 +<br/>0.2 +<br/>−0.2 +<br/>−0.4 +<br/>−0.6 +<br/>−0.8 +<br/>−1.5 +<br/>4−NN Connected Graph +<br/>−1 +<br/>−0.5 +<br/>0.5 +<br/>(a) +<br/>1.5 +<br/>2.5 +<br/>Classification Results By Nearst Neighbor +<br/>class 1 +<br/>class 2 +<br/>−1 +<br/>−0.5 +<br/>0.5 +<br/>(c) +<br/>1.5 +<br/>2.5 +<br/>1.2 +<br/>0.8 +<br/>0.6 +<br/>0.4 +<br/>0.2 +<br/>−0.2 +<br/>−0.4 +<br/>−0.6 +<br/>−0.8 +<br/>−1.5 +<br/>1.2 +<br/>0.8 +<br/>0.6 +<br/>0.4 +<br/>0.2 +<br/>−0.2 +<br/>−0.4 +<br/>−0.6 +<br/>−0.8 +<br/>−1.5 +<br/>Classification Results By LNP +<br/>class 1 +<br/>class 2 +<br/>−1 +<br/>−0.5 +<br/>0.5 +<br/>(b) +<br/>1.5 +<br/>2.5 +<br/>Classification Results By Transductive SVM +<br/>class 1 +<br/>class 2 +<br/>−1 +<br/>−0.5 +<br/>0.5 +<br/>(d) +<br/>1.5 +<br/>2.5 +<br/>Multi-Class Semi-Supervised Classi(cid:12)cation +<br/>(cid:15) Label set: L = f1; 2; (cid:1) (cid:1) (cid:1) ; cg +<br/>(cid:15) M be a set of n (cid:2) c matrices with non-negative real-value entries +<br/>(cid:15) F = [f1; f2; (cid:1) (cid:1) (cid:1) ; fc] 2 M corresponds to a speci(cid:12)c classi(cid:12)cation on X +<br/>(cid:15) The entry of Fij can be regarded as the likelihood that xi belongs to +<br/>class j +<br/>(cid:15) The label of xi can be computed by yi = arg maxj6c Fij +<br/>Induction +<br/>minimize (cid:17)?(xt) = (cid:13)(cid:13)(cid:13)(cid:13) +<br/>ft (cid:0) Xxj2N (xt) +<br/>(cid:15) plug a test example xt into the cost function +<br/>(cid:15) keep the labels of all xi 2 X (cid:12)xed when inducing the label of xt +<br/>w(xt; xj)fj(cid:13)(cid:13)(cid:13)(cid:13) +<br/>(5) +<br/>Learning from partially labeled data +<br/>(cid:15) Face/object recognition +<br/>(cid:15) Image / video retrieval +<br/>(cid:15) Interactive image segmentation +<br/>Graph-Based Semi-Supervised Classi(cid:12)cation +<br/>Represent the dataset as an weighted undirected graph G =< V; E > +<br/>(cid:15) V: the node set, corresponding to the dataset +<br/>(cid:15) E: the edge set, corresponding to the pairwise relationships +<br/>wij = expn(cid:0)2(cid:12)kxi (cid:0) xjk2o +<br/>(1) +<br/>Cluster Assumption +<br/>(cid:15) nearby points are likely to have the same label +<br/>(cid:15) points on the same structure (such as a cluster or a submanifold) are +<br/>prone to have the same label +<br/>=) Similar to manifold analysis (ISOMAP, LLE, Laplacian Eigen- +<br/>map(cid:1) (cid:1) (cid:1) ) +<br/>=) Incorporate the neighborhood information into graph construction +<br/>Linear Neighborhoods +<br/>The data point can be linearly reconstructed from its k-nearest neigh- +<br/>bors. +<br/>minimize "i = (cid:13)(cid:13)(cid:13)(cid:13) +<br/>s:t: Xj +<br/>xi (cid:0) Xxj2N (xi) +<br/>wij = 1; wij > 0 +<br/>wij xj(cid:13)(cid:13)(cid:13)(cid:13) +<br/>(2) +<br/>(cid:15) wij re(cid:13)ects the similarity between xj and xi +<br/>(cid:15) How to solve it?=)Quadratic programming. +<br/>Collaborative Label Prediction +<br/>The label of an unlabeled point can be linearly reconstructed from its +<br/>neighbors’ labels +<br/>minimize (cid:17) = Xn +<br/>i=1 +<br/>fi (cid:0) Xxj2N (xi) +<br/>(cid:13)(cid:13)(cid:13)(cid:13) +<br/>wijfj(cid:13)(cid:13)(cid:13)(cid:13) +<br/>s:t: +<br/>fi = li (f or all labeled point xi) +<br/>(3) +<br/>(cid:15) wij is calculated through solving Eq.(2). +<br/>(cid:15) The neighborhood information are incorporated into label prediction. +<br/>How to solve Eq.(3)? +<br/>i=1 +<br/>fi (cid:0) Xxj2N (xi) +<br/>(cid:17) = Xn +<br/>=) minimize (cid:17) ()(I (cid:0) W)f = 0; s:t: fi = li +<br/>(cid:15) Refer to the following paper +<br/>wijfj(cid:13)(cid:13)(cid:13)(cid:13) +<br/>(cid:13)(cid:13)(cid:13)(cid:13) +<br/>= f T (I (cid:0) W)T (I (cid:0) W)f (4) +<br/>Recognition +<br/>Recognition accuracies on ORL database +<br/>LNP +<br/>Consistency +<br/>Kernel Eigenface +<br/>Fisherface +<br/>Eigenface +<br/>0.9 +<br/>0.8 +<br/>0.7 +<br/>0.6 +<br/>0.5 +<br/>0.4 +<br/>Recognition accuracies on COIL database +<br/>LNP +<br/>Consistency +<br/>Kernel PCA +<br/>PCA+LDA +<br/>PCA +<br/>10 +<br/>12 +<br/>14 +<br/>16 +<br/>18 +<br/>References +<br/>(cid:15) S.T. Roweis and L.K. Saul, Noninear Dimensionality Reduction by +<br/>Locally Linear Embedding. Science: vol. 290, 2323-2326. 2000. +<br/>(cid:15) O. Chapelle, et al. (eds.): Semi-Supervised Learning. MIT Press: +<br/>Cambridge, MA. 2006. +<br/>(cid:15) A. Levin D. Lischinski and Y. Weiss. Colorization using Optimization. +<br/>SIGGRAPH, ACM Transactions on Graphics, Aug 2004. +<br/>Data Ranking +<br/>Ranking Result by Euclidean Distance +<br/>1.5 +<br/>0.5 +<br/>−0.5 +<br/>Ranking Result by LNP +<br/>0.95 +<br/>0.9 +<br/>0.85 +<br/>0.8 +<br/>0.75 +<br/>0.7 +<br/>0.65 +<br/>1.5 +<br/>0.5 +<br/>−0.5 +<br/>Zhu, X., Ghahramani, Z., & La(cid:11)erty, J.(2003). Semi-Supervised Learn- +<br/>ing Using Gaussian Fields and Harmonic Functions. In Proceedings of +<br/>the 20th International Conference on Machine Learning +<br/>−1 +<br/>−1.5 +<br/>−1 +<br/>−0.5 +<br/>0.5 +<br/>(a) +<br/>1.5 +<br/>2.5 +<br/>−1 +<br/>−1.5 +<br/>−1 +<br/>−0.5 +<br/>0.5 +<br/>(b) +<br/>1.5 +<br/>2.5 +</td><td>('34410258', 'Fei Wang', 'fei wang')<br/>('1688516', 'Jingdong Wang', 'jingdong wang')<br/>('1700883', 'Changshui Zhang', 'changshui zhang')<br/>('7969645', 'Helen C. Shen', 'helen c. shen')</td><td></td></tr><tr><td>15d653972d176963ef0ad2cc582d3b35ca542673</td><td>CSVideoNet: A Real-time End-to-end Learning Framework for +<br/>High-frame-rate Video Compressive Sensing +<br/>School of Computing, Informatics, and Decision Systems Engineering +<br/><b>Arizona State University, Tempe AZ</b></td><td>('47831601', 'Kai Xu', 'kai xu')<br/>('40615963', 'Fengbo Ren', 'fengbo ren')</td><td>{kaixu, renfengbo}@asu.edu +</td></tr><tr><td>159e792096756b1ec02ec7a980d5ef26b434ff78</td><td>Proceedings of the Twenty-Eighth AAAI Conference on Artificial Intelligence +<br/>Signed Laplacian Embedding for Supervised Dimension Reduction +<br/><b>Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong University</b><br/><b>Centre for Quantum Computation and Intelligent Systems, University of Technology Sydney</b></td><td>('1710691', 'Chen Gong', 'chen gong')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('39264954', 'Jie Yang', 'jie yang')<br/>('1847070', 'Keren Fu', 'keren fu')</td><td>{goodgongchen, jieyang, fkrsuper}@sjtu.edu.cn +<br/>dacheng.tao@uts.edu.au +</td></tr><tr><td>153e5cddb79ac31154737b3e025b4fb639b3c9e7</td><td>PREPRINT SUBMITTED TO IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS +<br/>Active Dictionary Learning in Sparse +<br/>Representation Based Classification +</td><td>('1935596', 'Jin Xu', 'jin xu')<br/>('2198278', 'Haibo He', 'haibo he')<br/>('1881104', 'Hong Man', 'hong man')</td><td></td></tr><tr><td>1586871a1ddfe031b885b94efdbff647cf03eff1</td><td>A Visual Historical Record of American High School Yearbooks +<br/>A Century of Portraits: +<br/><b>University of California Berkeley</b><br/><b>Brown University</b><br/><b>University of California Berkeley</b></td><td>('2361255', 'Shiry Ginosar', 'shiry ginosar')<br/>('2660664', 'Kate Rakelly', 'kate rakelly')<br/>('33385802', 'Sarah Sachs', 'sarah sachs')<br/>('2130100', 'Brian Yin', 'brian yin')<br/>('1763086', 'Alexei A. Efros', 'alexei a. efros')</td><td></td></tr><tr><td>15cf7bdc36ec901596c56d04c934596cf7b43115</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 8, No. 9, 2017 +<br/>Face Extraction from Image based on K-Means +<br/>Clustering Algorithms +<br/><b>Faculty of Computer, Khoy Branch, Islamic Azad University, Khoy, Iran</b></td><td>('2062871', 'Yousef Farhang', 'yousef farhang')</td><td></td></tr><tr><td>1576ed0f3926c6ce65e0ca770475bca6adcfdbb4</td><td>Keep it Accurate and Diverse: Enhancing Action Recognition Performance by +<br/>Ensemble Learning +<br/><b>Faculty of Computer Science, Dalhousie University, Halifax, Canada</b><br/>Computer Vision Center, UAB +<br/>Edificio O, Campus UAB, 08193, Bellaterra (Cerdanyola), Barcelona, Spain +<br/><b>University of Barcelona</b><br/>Gran Via de les Corts Catalanes, 585, 08007, Barcelona +<br/>Visual Analysis of People (VAP) Laboratory +<br/>Rendsburggade 14, 9000 Aalborg, Denmark +</td><td>('1921285', 'Mohammad Ali Bagheri', 'mohammad ali bagheri')<br/>('3212432', 'Qigang Gao', 'qigang gao')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')<br/>('1803459', 'Kamal Nasrollahi', 'kamal nasrollahi')<br/>('1876184', 'Michael B. Holte', 'michael b. holte')<br/>('1700569', 'Thomas B. Moeslund', 'thomas b. moeslund')</td><td>bagheri@cs.dal.ca +<br/>sergio@maia.ub.es, aclapes@cvc.uab.cat +<br/>{kn, mbh, tbm}@create.aau.dk +</td></tr><tr><td>156cd2a0e2c378e4c3649a1d046cd080d3338bca</td><td>Exemplar based approaches on Face Fiducial Detection and +<br/>Frontalization +<br/>Thesis submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/>MS by Research +<br/>in +<br/>Computer Science & Engineering +<br/>by +<br/>Mallikarjun B R +<br/>201307681 +<br/><b>International Institute of Information Technology</b><br/>Hyderabad - 500 032, India +<br/>May 2017 +</td><td></td><td>mallikarjun.br@research.iiit.ac.in +</td></tr><tr><td>157eb982da8fe1da4c9e07b4d89f2e806ae4ceb6</td><td><b>MITSUBISHI ELECTRIC RESEARCH LABORATORIES</b><br/>http://www.merl.com +<br/>Connecting the Dots in Multi-Class Classification: From +<br/>Nearest Subspace to Collaborative Representation +<br/>Chi, Y.; Porikli, F. +<br/>TR2012-043 +<br/>June 2012 +</td><td></td><td></td></tr><tr><td>15e0b9ba3389a7394c6a1d267b6e06f8758ab82b</td><td>Xu et al. IPSJ Transactions on Computer Vision and +<br/>Applications (2017) 9:24 +<br/>DOI 10.1186/s41074-017-0035-2 +<br/>IPSJ Transactions on Computer +<br/>Vision and Applications +<br/>TECHNICAL NOTE +<br/>Open Access +<br/>The OU-ISIR Gait Database comprising the +<br/>Large Population Dataset with Age and +<br/>performance evaluation of age estimation +</td><td>('7513255', 'Chi Xu', 'chi xu')<br/>('1689334', 'Yasushi Makihara', 'yasushi makihara')<br/>('12881056', 'Gakuto Ogi', 'gakuto ogi')<br/>('1737850', 'Xiang Li', 'xiang li')<br/>('1715071', 'Yasushi Yagi', 'yasushi yagi')<br/>('6120396', 'Jianfeng Lu', 'jianfeng lu')</td><td></td></tr><tr><td>151481703aa8352dc78e2577f0601782b8c41b34</td><td>Appearance Manifold of Facial Expression +<br/><b>Queen Mary, University of London, London E1 4NS, UK</b><br/>Department of Computer Science +</td><td>('10795229', 'Caifeng Shan', 'caifeng shan')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')<br/>('2803283', 'Peter W. McOwan', 'peter w. mcowan')</td><td>{cfshan,sgg,pmco}@dcs.qmul.ac.uk +</td></tr><tr><td>15aa6c457678e25f6bc0e818e5fc39e42dd8e533</td><td></td><td></td><td></td></tr><tr><td>15cf1f17aeba62cd834116b770f173b0aa614bf4</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 77 – No.5, September 2013 +<br/>Facial Expression Recognition using Neural Network with +<br/>Regularized Back-propagation Algorithm +<br/>Research Scholar +<br/>Department of ECE, +<br/><b></b><br/>Phagwara, India +<br/>Assistant Professor +<br/>Department of ECE, +<br/><b></b><br/>Phagwara, India +<br/>Research Scholar +<br/>Department of ECE, +<br/><b>Gyan Ganga Institute of</b><br/>Technology & Sciences, +<br/>Jabalpur, India +</td><td>('35358999', 'Ashish Kumar Dogra', 'ashish kumar dogra')<br/>('50227570', 'Nikesh Bajaj', 'nikesh bajaj')</td><td></td></tr><tr><td>1565721ebdbd2518224f54388ed4f6b21ebd26f3</td><td>Face and Landmark Detection by Using Cascade of Classifiers +<br/><b>Eskisehir Osmangazi University</b><br/>Eskisehir, Turkey +<br/>Laboratoire Jean Kuntzmann +<br/>Grenoble Cedex 9, France +<br/><b>Czech Technical University</b><br/>Praha, Czech Republic +</td><td>('2277308', 'Hakan Cevikalp', 'hakan cevikalp')<br/>('1756114', 'Bill Triggs', 'bill triggs')<br/>('1778663', 'Vojtech Franc', 'vojtech franc')</td><td>hakan.cevikalp@gmail.com +<br/>Bill.Triggs@imag.fr +<br/>xfrancv@cmp.felk.cvut.cz +</td></tr><tr><td>15f3d47b48a7bcbe877f596cb2cfa76e798c6452</td><td>Automatic face analysis tools for interactive digital games +<br/>Anonymised for blind review +<br/>Anonymous +<br/>Anonymous +<br/>Anonymous +</td><td></td><td></td></tr><tr><td>15728d6fd5c9fc20b40364b733228caf63558c31</td><td></td><td>('2831988', 'IAN N. ENDRES', 'ian n. endres')</td><td></td></tr><tr><td>15252b7af081761bb00535aac6bd1987391f9b79</td><td>ESTIMATION OF EYE GAZE DIRECTION ANGLES BASED ON ACTIVE APPEARANCE +<br/>MODELS +<br/><b>School of E.C.E., National Technical University of Athens, 15773 Athens, Greece</b></td><td>('2539459', 'Petros Koutras', 'petros koutras')<br/>('1750686', 'Petros Maragos', 'petros maragos')</td><td>Email: {pkoutras, maragos}@cs.ntua.gr +</td></tr><tr><td>1513949773e3a47e11ab87d9a429864716aba42d</td><td></td><td></td><td></td></tr><tr><td>15ee80e86e75bf1413dc38f521b9142b28fe02d1</td><td>Towards a Deep Learning Framework for +<br/>Unconstrained Face Detection +<br/>CyLab Biometrics Center and the Department of Electrical and Computer Engineering, +<br/><b>Carnegie Mellon University, Pittsburgh, PA, USA</b></td><td>('3049981', 'Yutong Zheng', 'yutong zheng')<br/>('3117715', 'Chenchen Zhu', 'chenchen zhu')<br/>('6131978', 'T. Hoang Ngan Le', 't. hoang ngan le')<br/>('1769788', 'Khoa Luu', 'khoa luu')<br/>('2043374', 'Chandrasekhar Bhagavatula', 'chandrasekhar bhagavatula')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td>{yutongzh, chenchez, kluu, cbhagava, thihoanl}@andrew.cmu.edu, msavvid@ri.cmu.edu +</td></tr><tr><td>153c8715f491272b06dc93add038fae62846f498</td><td></td><td>('33047058', 'JONGWOO LIM', 'jongwoo lim')</td><td></td></tr><tr><td>15e27f968458bf99dd34e402b900ac7b34b1d575</td><td>8362 +<br/>2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +<br/>978-1-4799-2893-4/14/$31.00 ©2014 IEEE +<br/><b>University of Toronto</b><br/>1. INTRODUCTION +</td><td>('2030736', 'Mohammad Shahin Mahanta', 'mohammad shahin mahanta')<br/>('1705037', 'Konstantinos N. Plataniotis', 'konstantinos n. plataniotis')</td><td>Email: {mahanta, kostas} @ece.utoronto.ca +</td></tr><tr><td>15f70a0ad8903017250927595ae2096d8b263090</td><td>Learning Robust Deep Face Representation +<br/><b>University of Science and Technology Beijing</b><br/>Beijing, China +</td><td>('2225749', 'Xiang Wu', 'xiang wu')</td><td>alfredxiangwu@gmail.com +</td></tr><tr><td>1564bf0a268662df752b68bee5addc4b08868739</td><td>With Whom Do I Interact? +<br/>Detecting Social Interactions in Egocentric +<br/>Photo-streams +<br/><b>University of Barcelona</b><br/>Barcelona, Spain +<br/>Computer Vision Center and +<br/><b>University of Barcelona</b><br/>Barcelona, Spain +<br/>Computer Vision Center and +<br/><b>University of Barcelona</b><br/>Barcelona, Spain +</td><td>('2084534', 'Maedeh Aghaei', 'maedeh aghaei')<br/>('2837527', 'Mariella Dimiccoli', 'mariella dimiccoli')<br/>('1724155', 'Petia Radeva', 'petia radeva')</td><td>Email:maghaeigavari@ub.edu +</td></tr><tr><td>158e32579e38c29b26dfd33bf93e772e6211e188</td><td>Automated Real Time Emotion Recognition using +<br/>Facial Expression Analysis +<br/>by +<br/>A thesis submitted to the Faculty of Graduate and Postdoctoral +<br/>Affairs in partial fulfillment of the requirements for the degree of +<br/>Master +<br/>of +<br/>Computer Science +<br/><b>Carleton University</b><br/>Ottawa, Ontario +</td><td></td><td></td></tr><tr><td>122f51cee489ba4da5ab65064457fbe104713526</td><td>Long Short Term Memory Recurrent Neural Network based +<br/>Multimodal Dimensional Emotion Recognition +<br/>Recognition +<br/>Recognition +<br/>Recognition +<br/>National Laboratory of Pattern +<br/>National Laboratory of Pattern +<br/>National Laboratory of Pattern +<br/><b>Institute of Automation</b><br/>Chinese Academy of Sciences +<br/><b>Institute of Automation</b><br/>Chinese Academy of Sciences +<br/><b>Institute of Automation</b><br/>Chinese Academy of Sciences +<br/>National Laboratory of Pattern Recognition +<br/>National Laboratory of Pattern Recognition +<br/><b>Institute of Automation</b><br/>Chinese Academy of Sciences +<br/> +</td><td>('1850313', 'Linlin Chao', 'linlin chao')<br/>('37670752', 'Jianhua Tao', 'jianhua tao')<br/>('2740129', 'Minghao Yang', 'minghao yang')<br/>('1704841', 'Ya Li', 'ya li')</td><td>linlin.chao@nlpr.ia.ac.cn +<br/>jhtao@nlpr.ia.ac.cn +<br/>mhyang@nlpr.ia.ac.cn +<br/>yli@nlpr.ia.ac.cn +</td></tr><tr><td>121503705689f46546cade78ff62963574b4750b</td><td>We don’t need no bounding-boxes: +<br/>Training object class detectors using only human verification +<br/><b>University of Edinburgh</b></td><td>('1749373', 'Dim P. Papadopoulos', 'dim p. papadopoulos')<br/>('1823362', 'Jasper R. R. Uijlings', 'jasper r. r. uijlings')<br/>('48716849', 'Frank Keller', 'frank keller')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')</td><td>dim.papadopoulos@ed.ac.uk +<br/>jrr.uijlings@ed.ac.uk +<br/>keller@inf.ed.ac.uk +<br/>vferrari@inf.ed.ac.uk +</td></tr><tr><td>125d82fee1b9fbcc616622b0977f3d06771fc152</td><td>Hierarchical Face Parsing via Deep Learning +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('1693209', 'Ping Luo', 'ping luo')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>pluo.lhi@gmail.com +<br/>xgwang@ee.cuhk.edu.hk +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>1255afbf86423c171349e874b3ac297de19f00cd</td><td>Robust Face Recognition by Computing Distances +<br/>from Multiple Histograms of Oriented Gradients +<br/><b>Institute of Arti cial Intelligence and Cognitive Engineering (ALICE), University of Groningen</b><br/>Nijenborgh 9, Groningen, The Netherlands +</td><td>('3351361', 'Mahir Faik Karaaba', 'mahir faik karaaba')<br/>('1728531', 'Olarik Surinta', 'olarik surinta')<br/>('1799278', 'Lambert Schomaker', 'lambert schomaker')</td><td>Email: {m.f.karaaba, o.surinta, l.r.b.schomaker, m.a.wiering}@rug.nl +</td></tr><tr><td>1275d6a800f8cf93c092603175fdad362b69c191</td><td>Deep Face Recognition: A Survey +<br/>School of Information and Communication Engineering, +<br/><b>Beijing University of Posts and Telecommunications, Beijing, China</b><br/>still have an inevitable limitation on robustness against the +<br/>complex nonlinear facial appearance variations. +<br/>In general, traditional methods attempted to solve FR prob- +<br/>lem by one or two layer representation, such as filtering +<br/>responses or histogram of the feature codes. The research com- +<br/>munity studied intensively to separately improve the prepro- +<br/>cessing, local descriptors, and feature transformation, which +<br/>improve face recognition accuracy slowly. By the continuous +<br/>improvement of a decade, “shallow” methods only improve the +<br/>accuracy of the LFW benchmark to about 95% [26], which +<br/>indicates that “shallow” methods are insufficient to extract +<br/>stable identity feature against unconstrained facial variations. +<br/>Due to the technical insufficiency, facial recognition systems +<br/>were often reported with unstable performance or failures with +<br/>countless false alarms in real-world applications. +</td><td>('2285767', 'Mei Wang', 'mei wang')<br/>('1774956', 'Weihong Deng', 'weihong deng')</td><td>wm0245@126.com, whdeng@bupt.edu.cn +</td></tr><tr><td>126535430845361cd7a3a6f317797fe6e53f5a3b</td><td>Robust Photometric Stereo via Low-Rank Matrix +<br/>Completion and Recovery (cid:63) +<br/><b>School of Optics and Electronics, Beijing Institute of Technology, Beijing</b><br/><b>Coordinated Science Lab, University of Illinois at Urbana-Champaign</b><br/><b>Key Laboratory of Machine Perception, Peking University, Beijing</b><br/>§Visual Computing Group, Microsoft Research Asia, Beijing +</td><td>('2417838', 'Lun Wu', 'lun wu')<br/>('1701028', 'Arvind Ganesh', 'arvind ganesh')<br/>('35580784', 'Boxin Shi', 'boxin shi')<br/>('1774618', 'Yasuyuki Matsushita', 'yasuyuki matsushita')<br/>('1692621', 'Yongtian Wang', 'yongtian wang')<br/>('1700297', 'Yi Ma', 'yi ma')</td><td>lun.wu@hotmail.com, abalasu2@illinois.edu, shiboxin@cis.pku.edu.cn, +<br/>yasumat@microsoft.com, wyt@bit.edu.cn, mayi@microsoft.com +</td></tr><tr><td>122ee00cc25c0137cab2c510494cee98bd504e9f</td><td>The Application of +<br/>Active Appearance Models to +<br/>Comprehensive Face Analysis +<br/>Technical Report +<br/>TU M¨unchen +<br/>April 5, 2007 +</td><td>('2866162', 'Simon Kriegel', 'simon kriegel')</td><td>kriegel@mmer-systems.eu +</td></tr><tr><td>1287bfe73e381cc8042ac0cc27868ae086e1ce3b</td><td></td><td></td><td></td></tr><tr><td>121fe33daf55758219e53249cf8bcb0eb2b4db4b</td><td>CHAKRABARTI et al.: EMPIRICAL CAMERA MODEL +<br/>An Empirical Camera Model +<br/>for Internet Color Vision +<br/>http://www.eecs.harvard.edu/~ayanc/ +<br/>http://www.cs.middlebury.edu/~schar/ +<br/>Todd Zickler1 +<br/>http://www.eecs.harvard.edu/~zickler/ +<br/>1 Harvard School of Engineering and +<br/>Applied Sciences +<br/>Cambridge, MA, USA 02139 +<br/>2 Department of Computer Science +<br/><b>Middlebury College</b><br/>Middlebury, VT, USA 05753 +</td><td>('38534744', 'Ayan Chakrabarti', 'ayan chakrabarti')<br/>('1709053', 'Daniel Scharstein', 'daniel scharstein')</td><td></td></tr><tr><td>12408baf69419409d228d96c6f88b6bcde303505</td><td>Temporal Tessellation: A Unified Approach for Video Analysis +<br/><b>The Blavatnik School of Computer Science, Tel Aviv University, Israel</b><br/><b>Information Sciences Institute, USC, CA, USA</b><br/><b>The Open University of Israel, Israel</b><br/>4Facebook AI Research +</td><td>('48842639', 'Dotan Kaufman', 'dotan kaufman')<br/>('36813724', 'Gil Levi', 'gil levi')<br/>('1756099', 'Tal Hassner', 'tal hassner')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td></td></tr><tr><td>120bcc9879d953de7b2ecfbcd301f72f3a96fb87</td><td>Report on the FG 2015 Video Person Recognition Evaluation +<br/>Zhenhua Feng +<br/><b>Colorado State University</b><br/>Fort Collins, CO, USA +<br/><b>University of Notre Dame</b><br/>Notre Dame, IN, USA +<br/><b>University of Surrey</b><br/>United Kingdom +<br/>1 Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>Institute of Computing Technology, CAS, Beijing, 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b><br/><b>Stevens Institute of Technology</b><br/>Hoboken, NJ, USA +<br/>Vitomir ˇStruc +<br/>Janez Kriˇzaj +<br/><b>University of Ljubljana</b><br/>Ljubljana, Slovenia +<br/><b>University of Technology, Sydney</b><br/>Sydney, Australia +<br/><b>National Institute of Standards and Technology</b><br/>Gaithersburg, MD, USA +</td><td>('1757322', 'J. Ross Beveridge', 'j. ross beveridge')<br/>('1694404', 'Bruce A. Draper', 'bruce a. draper')<br/>('1704876', 'Patrick J. Flynn', 'patrick j. flynn')<br/>('39976184', 'Patrik Huber', 'patrik huber')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('7945869', 'Zhiwu Huang', 'zhiwu huang')<br/>('1688086', 'Shaoxin Li', 'shaoxin li')<br/>('38751558', 'Yan Li', 'yan li')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('3373117', 'Ruiping Wang', 'ruiping wang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('37990555', 'Changxing Ding', 'changxing ding')<br/>('32028519', 'P. Jonathon Phillips', 'p. jonathon phillips')</td><td>ross@cs.colostate.edu +</td></tr><tr><td>12cb3bf6abf63d190f849880b1703ccc183692fe</td><td>Guess Who?: A game to crowdsource the labeling of affective facial +<br/>expressions is comparable to expert ratings. +<br/>Graduation research project, june 2012 +<br/>Supervised by: Dr. Joost Broekens +<br/><b></b></td><td></td><td>mail@barryborsboom.nl +</td></tr><tr><td>12095f9b35ee88272dd5abc2d942a4f55804b31e</td><td>DenseReg: Fully Convolutional Dense Shape Regression In-the-Wild +<br/>Rıza Alp G¨uler1 +<br/>1INRIA-CentraleSup´elec, France +<br/><b>Imperial College London, UK</b><br/>Stefanos Zafeiriou2 +<br/>3Amazon, Berlin, Germany +<br/><b>University College London, UK</b></td><td>('2814229', 'George Trigeorgis', 'george trigeorgis')<br/>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('2796644', 'Patrick Snape', 'patrick snape')<br/>('48111527', 'Iasonas Kokkinos', 'iasonas kokkinos')</td><td>riza.guler@inria.fr +<br/>2{g.trigeorgis, p.snape, s.zafeiriou}@imperial.ac.uk +<br/>antonak@amazon.com +<br/>i.kokkinos@cs.ucl.ac.uk +</td></tr><tr><td>12cd96a419b1bd14cc40942b94d9c4dffe5094d2</td><td>29 +<br/>Proceedings of the 5th Workshop on Vision and Language, pages 29–38, +<br/>Berlin, Germany, August 12 2016. c(cid:13)2016 Association for Computational Linguistics +</td><td></td><td></td></tr><tr><td>1275852f2e78ed9afd189e8b845fdb5393413614</td><td>A Transfer Learning based Feature-Weak-Relevant Method for +<br/>Image Clustering +<br/><b>Dalian Maritime University</b><br/>Dalian, China +</td><td>('3852923', 'Bo Dong', 'bo dong')<br/>('2860808', 'Xinnian Wang', 'xinnian wang')</td><td>{dukedong,wxn}@dlmu.edu.cn +</td></tr><tr><td>1297ee7a41aa4e8499c7ddb3b1fed783eba19056</td><td><b>University of Nebraska - Lincoln</b><br/>US Army Research +<br/>2015 +<br/>U.S. Department of Defense +<br/>Effects of emotional expressions on persuasion +<br/>Gale Lucas +<br/><b>University of Southern California</b><br/><b>University of Southern California</b><br/><b>University of Southern California</b><br/><b>University of Southern California</b><br/>Follow this and additional works at: http://digitalcommons.unl.edu/usarmyresearch +<br/>Wang, Yuqiong; Lucas, Gale; Khooshabeh, Peter; de Melo, Celso; and Gratch, Jonathan, "Effects of emotional expressions on +<br/>persuasion" (2015). US Army Research. 340. +<br/>http://digitalcommons.unl.edu/usarmyresearch/340 +</td><td>('49416640', 'Yuqiong Wang', 'yuqiong wang')<br/>('2635945', 'Peter Khooshabeh', 'peter khooshabeh')<br/>('1977901', 'Celso de Melo', 'celso de melo')<br/>('1730824', 'Jonathan Gratch', 'jonathan gratch')</td><td>DigitalCommons@University of Nebraska - Lincoln +<br/>University of Southern California, wangyuqiong@ymail.com +<br/>This Article is brought to you for free and open access by the U.S. Department of Defense at DigitalCommons@University of Nebraska - Lincoln. It has +<br/>been accepted for inclusion in US Army Research by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln. +</td></tr><tr><td>12055b8f82d5411f9ad196b60698d76fbd07ac1e</td><td>1475 +<br/>Multiview Facial Landmark Localization in RGB-D +<br/>Images via Hierarchical Regression +<br/>With Binary Patterns +</td><td>('3152448', 'Zhanpeng Zhang', 'zhanpeng zhang')<br/>('40647981', 'Wei Zhang', 'wei zhang')<br/>('7137861', 'Jianzhuang Liu', 'jianzhuang liu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>126214ef0dcef2b456cb413905fa13160c73ec8e</td><td>Modelling human perception of static facial expressions +<br/>M.Sorci,J.Ph.Thiran +<br/>J.Cruz,T.Robin,M.Bierlaire +<br/><b>Electrical Engineering Institute, EPFL</b><br/>Transport and Mobility Laboratory,EPFL +<br/>Station 11, CH-1015, Lausanne +<br/>Station 11, CH-1015, Lausanne +<br/>G.Antonini +<br/>IBM Zurich Lab +<br/>Saumerstrasse 4 ,Ruschlikon +<br/>B.Cerretani +<br/><b>University of Siena</b><br/>DII,Siena +</td><td></td><td>{matteo.sorci,JP.Thiran}@epfl.ch +<br/>{javier.cruz,thomas.robin,michel.bierlaire}@epfl.ch +<br/>gan@zurich.ibm.com +<br/>barbara.cerretani@gmail.com +</td></tr><tr><td>120785f9b4952734818245cc305148676563a99b</td><td>Diagnostic automatique de l’état dépressif +<br/>S. Cholet +<br/>H. Paugam-Moisy +<br/>Laboratoire de Mathématiques Informatique et Applications (LAMIA - EA 4540) +<br/>Université des Antilles, Campus de Fouillole - Guadeloupe +<br/>Résumé +<br/>Les troubles psychosociaux sont un problème de santé pu- +<br/>blique majeur, pouvant avoir des conséquences graves sur +<br/>le court ou le long terme, tant sur le plan professionnel que +<br/>personnel ou familial. Le diagnostic de ces troubles doit +<br/>être établi par un professionnel. Toutefois, l’IA (l’Intelli- +<br/>gence Artificielle) peut apporter une contribution en four- +<br/>nissant au praticien une aide au diagnostic, et au patient +<br/>un suivi permanent rapide et peu coûteux. Nous proposons +<br/>une approche vers une méthode de diagnostic automatique +<br/>de l’état dépressif à partir d’observations du visage en +<br/>temps réel, au moyen d’une simple webcam. A partir de +<br/>vidéos du challenge AVEC’2014, nous avons entraîné un +<br/>classifieur neuronal à extraire des prototypes de visages +<br/>selon différentes valeurs du score de dépression de Beck +<br/>(BDI-II). +</td><td></td><td>Stephane.Cholet@univ-antilles.fr +</td></tr><tr><td>12692fbe915e6bb1c80733519371bbb90ae07539</td><td>Object Bank: A High-Level Image Representation for Scene +<br/>Classification & Semantic Feature Sparsification +<br/><b>Stanford University</b><br/><b>Carnegie Mellon University</b></td><td>('33642044', 'Li-Jia Li', 'li-jia li')<br/>('2888806', 'Hao Su', 'hao su')<br/>('1752601', 'Eric P. Xing', 'eric p. xing')<br/>('3216322', 'Li Fei-Fei', 'li fei-fei')</td><td></td></tr><tr><td>1251deae1b4a722a2155d932bdfb6fe4ae28dd22</td><td>A Large-scale Attribute Dataset for Zero-shot Learning +<br/>1 National Engineering Laboratory for Video Technology, +<br/>Key Laboratory of Machine Perception (MoE), +<br/>Cooperative Medianet Innovation Center, Shanghai, +<br/><b>School of EECS, Peking University, Beijing, 100871, China</b><br/><b>School of Data Science, Fudan University</b><br/>3 Sinovation Ventures +</td><td>('49217762', 'Bo Zhao', 'bo zhao')<br/>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('1705512', 'Rui Liang', 'rui liang')<br/>('3165417', 'Jiahong Wu', 'jiahong wu')<br/>('47904050', 'Yonggang Wang', 'yonggang wang')<br/>('36637369', 'Yizhou Wang', 'yizhou wang')</td><td>bozhao, yizhou.wang@pku.edu.cn, yanweifu@fudan.edu.cn +<br/>liangrui, wujiahong, wangyonggang@chuangxin.com +</td></tr><tr><td>12ccfc188de0b40c84d6a427999239c6a379cd66</td><td>Sparse Adversarial Perturbations for Videos +<br/>1 Tsinghua National Lab for Information Science and Technology +<br/>1 State Key Lab of Intelligent Technology and Systems +<br/><b>Tsinghua University</b><br/>1 Center for Bio-Inspired Computing Research +</td><td>('2769710', 'Xingxing Wei', 'xingxing wei')<br/>('40062221', 'Jun Zhu', 'jun zhu')<br/>('37409747', 'Hang Su', 'hang su')</td><td>{xwei11, dcszj, suhangss}@mail.tsinghua.edu.cn +</td></tr><tr><td>12c713166c46ac87f452e0ae383d04fb44fe4eb2</td><td></td><td></td><td></td></tr><tr><td>12ebeb2176a5043ad57bc5f3218e48a96254e3e9</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 120 – No.24, June 2015 +<br/>Traffic Road Sign Detection and Recognition for +<br/>Automotive Vehicles +<br/>Zakir Hyder +<br/>Department of Electrical Engineering and +<br/>Department of Electrical Engineering and +<br/><b>Computer Science North South University, Dhaka</b><br/><b>Computer Science North South University, Dhaka</b><br/>Bangladesh +<br/>Bangladesh +</td><td></td><td></td></tr><tr><td>1270044a3fa1a469ec2f4f3bd364754f58a1cb56</td><td>Video-Based Face Recognition Using Probabilistic Appearance Manifolds +<br/>yComputer Science +<br/>Urbana, IL 61801 +<br/>zComputer Science & Engineering +<br/><b>University of Illinois, Urbana-Champaign University of California, San Diego</b><br/>La Jolla, CA 92093 +<br/>David Kriegmanz +<br/><b>Honda Research Institute</b><br/>800 California Street +<br/>Mountain View, CA 94041 +</td><td>('2457452', 'Kuang-chih Lee', 'kuang-chih lee')<br/>('1788818', 'Jeffrey Ho', 'jeffrey ho')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td>klee10@uiuc.edu +<br/>jho@cs.ucsd.edu myang@honda-ri.com kriegman@cs.ucsd.edu +</td></tr><tr><td>12150d8b51a2158e574e006d4fbdd3f3d01edc93</td><td>Deep End2End Voxel2Voxel Prediction +<br/>Presented by: Ahmed Osman +<br/>Ahmed Osman +</td><td>('1687325', 'Du Tran', 'du tran')<br/>('2276554', 'Rob Fergus', 'rob fergus')<br/>('2210374', 'Manohar Paluri', 'manohar paluri')</td><td></td></tr><tr><td>12003a7d65c4f98fb57587fd0e764b44d0d10125</td><td>Face Recognition in the Wild with the Probabilistic Gabor-Fisher +<br/>Classifier +<br/>Simon Dobriˇsek, Vitomir ˇStruc, Janez Kriˇzaj, France Miheliˇc +<br/><b>Faculty of Electrical Engineering, University of Ljubljana, Tr za ska cesta 25, SI-1000 Ljubljana, Slovenia</b></td><td></td><td></td></tr><tr><td>124538b3db791e30e1b62f81d4101be435ee12ef</td><td>ORIGINAL RESEARCH ARTICLE +<br/>published: 29 August 2013 +<br/>doi: 10.3389/fpsyg.2013.00506 +<br/>Basic level scene understanding: categories, attributes and +<br/>structures +<br/><b>Computer Science, Princeton University, Princeton, NJ, USA</b><br/><b>Computer Science, Brown University, Providence, RI, USA</b><br/><b>Computer Science and Engineering, University of Washington, Seattle, WA, USA</b><br/><b>Brain and Cognitive Sciences, Massachusetts Institute of Technology, Cambridge, MA, USA</b><br/><b>Computer Science and Arti cial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA, USA</b><br/><b>Computer Science and Arti cial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA, USA</b><br/>Edited by: +<br/>Tamara Berg, Stony Brook +<br/><b>University, USA</b><br/>Reviewed by: +<br/>Andrew M. Haun, Harvard Medical +<br/>School, USA +<br/>Devi Parikh, Virginia Tech, USA +<br/>*Correspondence: +<br/><b>Brown University</b><br/>115 Waterman Street, Box 1910, +<br/>Providence, RI 02912, USA +<br/>A longstanding goal of computer vision is to build a system that can automatically +<br/>understand a 3D scene from a single image. This requires extracting semantic concepts +<br/>and 3D information from 2D images which can depict an enormous variety of +<br/>environments that comprise our visual world. This paper summarizes our recent efforts +<br/>toward these goals. First, we describe the richly annotated SUN database which is a +<br/>collection of annotated images spanning 908 different scene categories with object, +<br/>attribute, and geometric labels for many scenes. This database allows us to systematically +<br/>study the space of scenes and to establish a benchmark for scene and object recognition. +<br/>We augment the categorical SUN database with 102 scene attributes for every image and +<br/>explore attribute recognition. Finally, we present an integrated system to extract the 3D +<br/>structure of the scene and objects depicted in an image. +<br/>Keywords: SUN database, basic level scene understanding, scene recognition, scene attributes, geometry +<br/>recognition, 3D context +<br/>1. INTRODUCTION +<br/>The ability to understand a 3D scene depicted in a static 2D image +<br/>goes to the very heart of the computer vision problem. By “scene” +<br/>we mean a place in which a human can act within or navigate. +<br/>What does it mean to understand a scene? There is no univer- +<br/>sal answer as it heavily depends on the task involved, and this +<br/>seemingly simple question hides a lot of complexity. +<br/>The dominant view in the current computer vision literature +<br/>is to name the scene and objects present in an image. However, +<br/>this level of understanding is rather superficial. If we can reason +<br/>about a larger variety of semantic properties and structures of +<br/>scenes it will enable richer applications. Furthermore, working on +<br/>an over-simplified task may distract us from exploiting the natu- +<br/>ral structures of the problem (e.g., relationships between objects +<br/>and 3d surfaces or the relationship between scene attributes and +<br/>object presence), which may be critical for a complete scene +<br/>understanding solution. +<br/>What is the ultimate goal of computational scene under- +<br/>standing? One goal might be to pass the Turing test for scene +<br/>understanding: Given an image depicting a static scene, a human +<br/>judge will ask a human or a machine questions about the picture. +<br/>If the judge cannot reliably tell the machine from the human, the +<br/>machine is said to have passed the test. This task is beyond the +<br/>current state-of-the-art as humans could ask a huge variety of +<br/>meaningful visual questions about an image, e.g., Is it safe to cross +<br/>this road? Who ate the last cupcake? Is this a fun place to vacation? +<br/>Are these people frustrated? Where can I set these groceries? etc. +<br/>Therefore, we propose a set of goals that are suitable for the +<br/>current state of research in computer vision that are not too +<br/>simplistic nor challenging and lead to a natural representation of +<br/>scenes. Based on these considerations, we define the task of scene +<br/>understanding as predicting the scene category, scene attributes, +<br/>the 3D enclosure of the space, and all the objects in the images. +<br/>For each object, we want to know its category and 3D bound- +<br/>ing box, as well as its 3D orientation relative to the scene. As an +<br/>image is a viewer-centric observation of the space, we also want +<br/>to recover the camera parameters, such as observer viewpoint +<br/>and field of view. We call this taskbasic level scene understand- +<br/>ing, with analogy to basic level in cognitive categorization (Rosch, +<br/>1978). It has practical applications for providing sufficient infor- +<br/>mation for simple interaction with the scene, such as navigation +<br/>and object manipulation. +<br/>1.1. OUTLINE +<br/>In this paper we discuss several aspects of basic level scene under- +<br/>standing. First, we quickly review our recent work on categorical +<br/>(section 2) and attribute-based scene representations (section 3). +<br/>Finally, we go into greater detail about novel work in 3d scene +<br/>understanding using structured learning to simultaneously rea- +<br/>son about many aspects of scenes (section 4). +<br/>Supporting these research efforts is the Scene UNderstanding +<br/>(SUN) database. By modern standards, the SUN database is not +<br/>especially large, containing on the order of 100,000 scenes. But +<br/>the SUN database is, instead, richly annotated with scene cat- +<br/>egories, scene attributes, geometric properties, “memorability” +<br/>measurements (Isola et al., 2011), and object segmentations. +<br/>There are 326,582 manually segmented objects for the 5650 +<br/>object categories labeled (Barriuso and Torralba, 2012). Object +<br/>www.frontiersin.org +<br/>August 2013 | Volume 4 | Article 506 | 1 +</td><td>('40599257', 'Jianxiong Xiao', 'jianxiong xiao')<br/>('12532254', 'James Hays', 'james hays')<br/>('2537592', 'Bryan C. Russell', 'bryan c. russell')<br/>('40541456', 'Genevieve Patterson', 'genevieve patterson')<br/>('1865091', 'Krista A. Ehinger', 'krista a. ehinger')<br/>('38611723', 'Antonio Torralba', 'antonio torralba')<br/>('31735139', 'Aude Oliva', 'aude oliva')<br/>('12532254', 'James Hays', 'james hays')</td><td>e-mail: hays@cs.brown.edu +</td></tr><tr><td>12d8730da5aab242795bdff17b30b6e0bac82998</td><td>Persistent Evidence of Local Image Properties in Generic ConvNets +<br/><b>CVAP, KTH (Royal Institute of Technology), Stockholm, SE</b></td><td>('2835963', 'Ali Sharif Razavian', 'ali sharif razavian')<br/>('2622491', 'Hossein Azizpour', 'hossein azizpour')<br/>('1801052', 'Atsuto Maki', 'atsuto maki')<br/>('1736906', 'Josephine Sullivan', 'josephine sullivan')<br/>('2484138', 'Carl Henrik Ek', 'carl henrik ek')<br/>('1826607', 'Stefan Carlsson', 'stefan carlsson')</td><td>{razavian,azizpour,atsuto,sullivan,chek,stefanc}@csc.kth.se +</td></tr><tr><td>8c13f2900264b5cf65591e65f11e3f4a35408b48</td><td>A GENERIC FACE REPRESENTATION APPROACH FOR +<br/>LOCAL APPEARANCE BASED FACE VERIFICATION +<br/>Interactive Systems Labs, Universität Karlsruhe (TH) +<br/>76131 Karlsruhe, Germany +<br/>web: http://isl.ira.uka.de/face_recognition/ +</td><td>('3025777', 'Hazim Kemal Ekenel', 'hazim kemal ekenel')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>{ekenel, stiefel}@ira.uka.de +</td></tr><tr><td>8cb3f421b55c78e56c8a1c1d96f23335ebd4a5bf</td><td></td><td></td><td></td></tr><tr><td>8c955f3827a27e92b6858497284a9559d2d0623a</td><td>Buletinul Ştiinţific al Universităţii "Politehnica" din Timişoara +<br/>Seria ELECTRONICĂ şi TELECOMUNICAŢII +<br/>TRANSACTIONS on ELECTRONICS and COMMUNICATIONS +<br/>Tom 53(67), Fascicola 1-2, 2008 +<br/>Facial Expression Recognition under Noisy Environment +<br/>Using Gabor Filters +</td><td>('2336758', 'Ioan Buciu', 'ioan buciu')<br/>('2526319', 'I. Nafornita', 'i. nafornita')<br/>('29835181', 'I. Pitas', 'i. pitas')</td><td></td></tr><tr><td>8c8525e626c8857a4c6c385de34ffea31e7e41d1</td><td>Cross-domain Image Retrieval with a Dual Attribute-aware Ranking Network +<br/><b>National University of Singapore, Singapore</b><br/>2IBM Research +</td><td>('1753492', 'Junshi Huang', 'junshi huang')<br/>('35370244', 'Qiang Chen', 'qiang chen')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>{a0092558, eleyans}@nus.edu.sg +<br/>rsferis@us.ibm.com +<br/>qiangchen@au1.ibm.com +</td></tr><tr><td>8c66378df977606d332fc3b0047989e890a6ac76</td><td>Hierarchical-PEP Model for Real-world Face Recognition +<br/><b>Stevens Institute of Technology</b><br/>Pose variation remains one of the major factors adversely affect the accuracy +<br/>of real-world face recognition systems. The same face in different poses +<br/>can look drastically different to each other. Belhumeur et al. [1] empiri- +<br/>cally demonstrate that frontal faces can be projected to a low-dimensional +<br/>subspace invariant to variation in illumination and facial expressions. This +<br/>observation highlights the importance of addressing pose variation because +<br/>it can greatly help relieve the adverse effects of the other visual variations. +<br/>A set of methods build pose-invariant face representations by locating +<br/>the facial landmarks. For example, Chen et al. [2] concatenate dense fea- +<br/>tures around the facial landmarks to build the face representation. The pose- +<br/>invariance is achieved in this way, because it always extracts features from +<br/>the face part surrounded around the facial landmarks regardless of their loca- +<br/>tions in the image. The elastic matching methods [5] generalize this design +<br/>with a probabilistic elastic part (PEP) model unsupervisedly learned from +<br/>face image patches. +<br/>While this procedure – locating the face parts and stacking the face part +<br/>features to build face representation – is empirically demonstrated to be ef- +<br/>fective by both Chen et al. [2] and Li et al. [5], we argue that directly de- +<br/>scribing the face parts with naive dense extraction of low-level features may +<br/>not be optimal. +<br/>In this work, we propose to build a better face part model to construct +<br/>an improved face representation. Inspired by the probabilistic elastic part +<br/>(PEP) model and the success of the deep hierarchical architecture in a num- +<br/>ber of visual tasks, we propose the Hierarchical-PEP model to approach the +<br/>unconstrained face recognition problem. +<br/>As shown in Figure 1, we apply the PEP model hierarchically to decom- +<br/>pose a face image into face parts at different levels of details to build pose- +<br/>invariant part-based face representations. Following the hierarchy from bottom- +<br/>up, we stack the face part representations at each layer, discriminatively re- +<br/>duce its dimensionality, and hence aggregate the face part representations +<br/>layer-by-layer to build a compact and invariant face representation. The +<br/>Hierarchical-PEP model exploits the fine-grained structures of the face parts +<br/>at different levels of details to address the pose variations. It is also guided +<br/>by supervised information in constructing the face part/face representations. +<br/>We empirically verify the Hierarchical-PEP model on two public bench- +<br/>marks and a face recognition challenge for image-based and video-based +<br/>face verification. The state-of-the-art performance demonstrates the poten- +<br/>tial of our method. We show the performance comparison on the YouTube +<br/>faces dataset [9] in Table 1. +<br/>Table 1: Performance comparison on YouTube Faces dataset under the re- +<br/>stricted with no outside data protocol. +<br/>Algorithm +<br/>MBGS [9] +<br/>MBGS+SVM- [8] +<br/>STFRD+PMML [10] +<br/>VF2 [7] +<br/>DDML (combined) [3] +<br/>Eigen-PEP [6] +<br/>LM3L [4] +<br/>Hierarchical-PEP (layers fusion) +<br/>Accuracy ± Error(%) +<br/>76.4± 1.8 +<br/>78.9± 1.9 +<br/>79.5± 2.5 +<br/>84.7± 1.4 +<br/>82.3± 1.5 +<br/>84.8± 1.4 +<br/>81.3± 1.2 +<br/>87.00± 1.50 +<br/>[1] Peter N. Belhumeur, Jo ˜ao P. Hespanha, and David J. Kriegman. Eigen- +<br/>faces vs. Fisherfaces: Recognition using class specific linear projec- +<br/>tion. PAMI, 1997. +<br/>[2] Dong Chen, Xudong Cao, Fang Wen, and Jian Sun. Blessing of di- +<br/>mensionality: High dimensional feature and its efficient compression +<br/>for face verification. In CVPR, 2013. +<br/>[3] Junlin Hu, Jiwen Lu, and Yap-Peng Tan. Discriminative deep metric +<br/>learning for face verification in the wild. In CVPR, 2014. +<br/>[4] Junlin Hu, Jiwen Lu, Junsong Yuan, and Yap-Peng Tan. Large margin +<br/>multi-metric learning for face and kinship verification in the wild. In +<br/>ACCV, 2014. +<br/>Yang. Probabilistic elastic matching for pose variant face verification. +<br/>In CVPR, 2013. +<br/>Eigen-pep for video face recognition. In ACCV, 2014. +<br/>[7] O. M. Parkhi, K. Simonyan, A. Vedaldi, and A. Zisserman. A compact +<br/>and discriminative face track descriptor. In CVPR, 2014. +<br/>[8] Lior Wolf and Noga Levy. The svm-minus similarity score for video +<br/>face recognition. In CVPR, 2013. +<br/>[9] Lior Wolf, Tal Hassner, and Itay Maoz. Face recognition in uncon- +<br/>strained videos with matched background similarity. In CVPR, 2011. +<br/>[10] Cui Zhen, Wen Li, Dong Xu, Shiguang Shan, and Xilin Chen. Fus- +<br/>ing robust face region descriptors via multiple metric learning for face +<br/>recognition in the wild. In CVPR, 2013. +<br/>Figure 1: Construction of the face representation with an example 2-layer Hierarchical-PEP model: PCA at layer t keeps dt dimensions. +</td><td>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('1745420', 'Gang Hua', 'gang hua')<br/>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('1745420', 'Gang Hua', 'gang hua')<br/>('3131569', 'Haoxiang Li', 'haoxiang li')<br/>('1745420', 'Gang Hua', 'gang hua')</td><td></td></tr><tr><td>8c9c8111e18f8798a612e7386e88536dfe26455e</td><td>COMPARING BAYESIAN NETWORKS TO CLASSIFY FACIAL +<br/>EXPRESSIONS +<br/><b>Institute of Systems and Robotics</b><br/><b>University of Coimbra, Portugal</b><br/><b>Institute Polythechnic of Leiria, Portugal</b><br/>Jorge Dias +<br/><b>Institute of Systems and Robotics</b><br/><b>University of Coimbra, Portugal</b><br/><b>Institute of Systems and Robotics</b><br/><b>University of Coimbra, Portugal</b></td><td>('2700157', 'Carlos Simplício', 'carlos simplício')<br/>('40031257', 'José Prado', 'josé prado')</td><td>carlos.simplicio@ipleiria.pt +<br/>jaugusto@isr.uc.pt +<br/>jorge@isr.uc.pt +</td></tr><tr><td>8c7f4c11b0c9e8edf62a0f5e6cf0dd9d2da431fa</td><td>Dataset Augmentation for Pose and Lighting +<br/>Invariant Face Recognition +<br/><b>Vision Systems, Inc</b><br/>†Systems and Technology Research +</td><td>('2103732', 'Octavian Biris', 'octavian biris')<br/>('3390731', 'Nate Crosswhite', 'nate crosswhite')<br/>('36067742', 'Jeffrey Byrne', 'jeffrey byrne')<br/>('3453447', 'Joseph L. Mundy', 'joseph l. mundy')</td><td></td></tr><tr><td>8c81705e5e4a1e2068a5bd518adc6955d49ae434</td><td>3D Object Recognition with Enhanced +<br/>Grassmann Discriminant Analysis +<br/>Graduate School of Systems and Information Engineering, +<br/><b>University of Tsukuba, Japan</b></td><td>('9641567', 'Lincon Sales de Souza', 'lincon sales de souza')<br/>('34581814', 'Hideitsu Hino', 'hideitsu hino')<br/>('1770128', 'Kazuhiro Fukui', 'kazuhiro fukui')</td><td>lincons@cvlab.cs.tsukuba.ac.jp, {hinohide, kfukui}@cs.tsukuba.ac.jp +</td></tr><tr><td>8cb403c733a5f23aefa6f583a17cf9b972e35c90</td><td>Learning the semantic structure of objects +<br/>from Web supervision +<br/>David Novotny1 +<br/>1Visual Geometry Group +<br/><b>University of Oxford</b><br/>2Computer Vision Group +<br/>Xerox Research Centre Europe +</td><td>('2295553', 'Diane Larlus', 'diane larlus')<br/>('1687524', 'Andrea Vedaldi', 'andrea vedaldi')</td><td>{david,andrea}@robots.ox.ac.uk +<br/>diane.larlus@xrce.xerox.com +</td></tr><tr><td>8ccde9d80706a59e606f6e6d48d4260b60ccc736</td><td>RotDCF: Decomposition of Convolutional Filters for +<br/>Rotation-Equivariant Deep Networks +<br/><b>Duke University</b><br/><b>Duke University</b></td><td>('1823644', 'Xiuyuan Cheng', 'xiuyuan cheng')<br/>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('1699339', 'Guillermo Sapiro', 'guillermo sapiro')</td><td></td></tr><tr><td>8c6b9c9c26ead75ce549a57c4fd0a12b46142848</td><td>Facial expression recognition using shape and +<br/>texture information +<br/>I. Kotsia1 and I. Pitas1 +<br/><b>Aristotle University of Thessaloniki</b><br/>Department of Informatics +<br/>Box 451 54124 +<br/>Thessaloniki, Greece +<br/>Summary. A novel method based on shape and texture information is proposed in +<br/>this paper for facial expression recognition from video sequences. The Discriminant +<br/>Non-negative Matrix Factorization (DNMF) algorithm is applied at the image cor- +<br/>responding to the greatest intensity of the facial expression (last frame of the video +<br/>sequence), extracting that way the texture information. A Support Vector Machines +<br/>(SVMs) system is used for the classi(cid:12)cation of the shape information derived from +<br/>tracking the Candide grid over the video sequence. The shape information consists +<br/>of the di(cid:11)erences of the node coordinates between the (cid:12)rst (neutral) and last (fully +<br/>expressed facial expression) video frame. Subsequently, fusion of texture and shape +<br/>information obtained is performed using Radial Basis Function (RBF) Neural Net- +<br/>works (NNs). The accuracy achieved is equal to 98,2% when recognizing the six +<br/>basic facial expressions. +<br/>1.1 Introduction +<br/>During the past two decades, many studies regarding facial expression recog- +<br/>nition, which plays a vital role in human centered interfaces, have been +<br/>conducted. Psychologists have de(cid:12)ned the following basic facial expressions: +<br/>anger, disgust, fear, happiness, sadness and surprise [?]. A set of muscle move- +<br/>ments, known as Action Units, was created. These movements form the so +<br/>called F acial Action Coding System (F ACS) [?]. A survey on auto- +<br/>matic facial expression recognition can be found in [?]. +<br/>In the current paper, a novel method for video based facial expression +<br/>recognition by fusing texture and shape information is proposed. The texture +<br/>information is obtained by applying the DNMF algorithm [?] on the last +<br/>frame of the video sequence, i.e. the one that corresponds to the greatest +<br/>intensity of the facial expression depicted. The shape information is calculated +<br/>as the di(cid:11)erence of Candide facial model grid node coordinates between the +<br/>(cid:12)rst and the last frame of a video sequence [?]. The decision made regarding +</td><td></td><td>pitas@aiia.csd.auth.gr +</td></tr><tr><td>8ce9b7b52d05701d5ef4a573095db66ce60a7e1c</td><td>Structured Sparse Subspace Clustering: A Joint +<br/>Affinity Learning and Subspace Clustering +<br/>Framework +</td><td>('9171002', 'Chun-Guang Li', 'chun-guang li')<br/>('1878841', 'Chong You', 'chong you')</td><td></td></tr><tr><td>8cb6daba2cb1e208e809633133adfee0183b8dd2</td><td>Know Before You Do: Anticipating Maneuvers +<br/>via Learning Temporal Driving Models +<br/><b>Cornell University and Stanford University</b></td><td>('1726066', 'Ashesh Jain', 'ashesh jain')<br/>('3282281', 'Bharad Raghavan', 'bharad raghavan')<br/>('1681995', 'Ashutosh Saxena', 'ashutosh saxena')</td><td>{ashesh,hema,asaxena}@cs.cornell.edu {bharad,shanesoh}@stanford.edu +</td></tr><tr><td>8c4ea76e67a2a99339a8c4decd877fe0aa2d8e82</td><td>Article +<br/>Gated Convolutional Neural Network for Semantic +<br/>Segmentation in High-Resolution Images +<br/><b>National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences</b><br/><b>University of Chinese Academy of Sciences, Beijing 101408, China</b><br/>Academic Editors: Qi Wang, Xiaofeng Li and Prasad S. Thenkabail +<br/>Received: 2 April 2017; Accepted: 1 May 2017; Published: 5 May 2017 +</td><td>('2206625', 'Hongzhen Wang', 'hongzhen wang')<br/>('1738352', 'Ying Wang', 'ying wang')<br/>('1737486', 'Qian Zhang', 'qian zhang')<br/>('1683738', 'Shiming Xiang', 'shiming xiang')<br/>('3364363', 'Chunhong Pan', 'chunhong pan')</td><td>95 Zhongguancun East Road, Beijing 100190, China; hongzhen.wang@nlpr.ia.ac.cn (H.W.); +<br/>ywang@nlpr.ia.ac.cn (Y.W.); chpan@nlpr.ia.ac.cn (C.P.) +<br/>3 Alibaba Group, Beijing 100102, China; zhangqiancsuia@163.com +<br/>* Correspondence: smxiang@nlpr.ia.ac.cn; Tel.: +86-136-7118-9070 +</td></tr><tr><td>8c6c0783d90e4591a407a239bf6684960b72f34e</td><td>SESSION +<br/>KNOWLEDGE ENGINEERING AND +<br/>MANAGEMENT + KNOWLEDGE ACQUISITION +<br/>Chair(s) +<br/>TBA +<br/>Int'l Conf. Information and Knowledge Engineering | IKE'13 |1 </td><td></td><td></td></tr><tr><td>8cb55413f1c5b6bda943697bba1dc0f8fc880d28</td><td>Video-based Face Recognition on Real-World Data +<br/>Hazım K. Ekenel +<br/>Interactive System Labs +<br/><b>University of Karlsruhe, Germany</b></td><td>('1842921', 'Johannes Stallkamp', 'johannes stallkamp')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>{jstallkamp,ekenel,stiefel}@ira.uka.de +</td></tr><tr><td>8cc07ae9510854ec6e79190cc150f9f1fe98a238</td><td>Article +<br/>Using Deep Learning to Challenge Safety Standard +<br/>for Highly Autonomous Machines in Agriculture +<br/><b>Aarhus University, Finlandsgade 22 8200 Aarhus N, Denmark</b><br/>† These authors contributed equally to this work. +<br/>Academic Editors: Francisco Rovira-Más and Gonzalo Pajares Martinsanz +<br/>Received: 18 December 2015; Accepted: 2 February 2016; Published: 15 February 2016 +</td><td>('32688812', 'Kim Arild Steen', 'kim arild steen')<br/>('2139204', 'Peter Christiansen', 'peter christiansen')<br/>('2550309', 'Henrik Karstoft', 'henrik karstoft')</td><td>pech@eng.au.dk (P.C.); hka@eng.au.dk (H.K.); rnj@eng.au.dk (R.N.J.) +<br/>* Correspondence: kim.steen@eng.au.dk; Tel.: +45-3116-8628 +</td></tr><tr><td>8509abbde2f4b42dc26a45cafddcccb2d370712f</td><td>Improving precision and recall of face recognition in SIPP with combination of +<br/>modified mean search and LSH +<br/>Xihua.Li +</td><td></td><td>lixihua9@126.com +</td></tr><tr><td>855bfc17e90ec1b240efba9100fb760c068a8efa</td><td></td><td></td><td></td></tr><tr><td>858ddff549ae0a3094c747fb1f26aa72821374ec</td><td>Survey on RGB, 3D, Thermal, and Multimodal +<br/>Approaches for Facial Expression Recognition: +<br/>History, Trends, and Affect-related Applications +</td><td>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')<br/>('7855312', 'Sergio Escalera', 'sergio escalera')</td><td></td></tr><tr><td>85041e48b51a2c498f22850ce7228df4e2263372</td><td>Subspace Regression: Predicting +<br/>a Subspace from One Sample +<br/><b>Robotics Institute, Carnegie Mellon University</b><br/>‡ Electrical & Controls Integration Lab, General Motors R&D +</td><td>('34299925', 'Minyoung Kim', 'minyoung kim')</td><td></td></tr><tr><td>85fd2bda5eb3afe68a5a78c30297064aec1361f6</td><td>702003 PSSXXX10.1177/0956797617702003Carr et al.Are You Smiling, or Have I Seen You Before? +<br/>research-article2017 +<br/>Research Article +<br/>Are You Smiling, or Have I Seen You +<br/>Before? Familiarity Makes Faces Look +<br/>Happier +<br/>2017, Vol. 28(8) 1087 –1102 +<br/>© The Author(s) 2017 +<br/>Reprints and permissions: +<br/>sagepub.com/journalsPermissions.nav +<br/>DOI: 10.1177/0956797617702003 +<br/>https://doi.org/10.1177/0956797617702003 +<br/>www.psychologicalscience.org/PS +<br/><b>Columbia Business School, University of California, San Diego</b><br/><b>Behavioural Science Group, Warwick Business School, University of Warwick; and 4Faculty of Psychology</b><br/><b>SWPS University of Social Sciences and Humanities</b></td><td>('5907729', 'Evan W. Carr', 'evan w. carr')<br/>('3122131', 'Piotr Winkielman', 'piotr winkielman')</td><td></td></tr><tr><td>857ad04fca2740b016f0066b152bd1fa1171483f</td><td>Sample Images can be Independently Restored from +<br/> Face Recognition Templates +<br/><b>School of Information Technology and Engineering, University of Ottawa, Ontario, Canada</b><br/>are being piloted or implemented at airports, for +<br/>government identification systems such as passports +<br/>and drivers licenses, and in surveillance applications. +<br/>In this paper, we consider the identifiability of stored +<br/>biometric +<br/>information, and +<br/>for +<br/>biometric privacy and security. +<br/>implications +<br/>its +<br/>Biometric authentication is typically performed by +<br/>a sophisticated software application, which manages +<br/>the user interface and database, and interacts with a +<br/>vendor specific, proprietary biometric algorithm. +<br/>Algorithms undertake the following processing steps: +<br/>1) acquisition of a biometric sample image, 2) +<br/>conversion of the sample image to a biometric +<br/>template, 3) comparison of the new (or "live") +<br/>template to previously stored templates, to calculate a +<br/>match score. High match scores indicate a likelihood +<br/>that the corresponding images are from the same +<br/>individual. The biometric template is a (typically +<br/>vendor specific) compact digital representation of the +<br/>essential features of the sample image. Biometric +<br/>algorithm vendors have uniformly claimed that it is +<br/>impossible or infeasible to recreate the image from the +<br/>template. [2, 3, 4, 7] These claims are supported by: 1) +<br/>the template records features (such as fingerprint +<br/>minutiae) and not image primitives, 2) templates are +<br/>typically calculated using only a small portion of the +<br/>image, 3) templates are small − a few hundred bytes − +<br/>much smaller than the sample image, and 4) the +<br/>proprietary nature of +<br/>the storage format makes +<br/>templates infeasible to "hack". For these reasons, +<br/>biometric templates are considered to be effectively +<br/>non-identifiable data, much like a password hash [7]. +<br/>In fact, these arguments are not valid: this paper +<br/>demonstrates a simple algorithm to recreate sample +<br/>images from templates using only match score results. +<br/>2. METHODS +<br/>A software application (figure 1) was designed with +<br/>the goal of recreating a face image of a specific person +<br/>in a face recognition database. The application has +<br/>local access to a database of face images, and has +<br/>network access to a Face Recognition Server (FRS) +</td><td>('2478519', 'Andy Adler', 'andy adler')</td><td>aadler@uottawa.ca +</td></tr><tr><td>858901405086056361f8f1839c2f3d65fc86a748</td><td>ON TENSOR TUCKER DECOMPOSITION: THE CASE FOR AN +<br/>ADJUSTABLE CORE SIZE +</td><td>('2424633', 'BILIAN CHEN', 'bilian chen')<br/>('1792785', 'ZHENING LI', 'zhening li')<br/>('1789588', 'SHUZHONG ZHANG', 'shuzhong zhang')</td><td></td></tr><tr><td>85188c77f3b2de3a45f7d4f709b6ea79e36bd0d9</td><td>Author manuscript, published in "Workshop on Faces in 'Real-Life' Images: Detection, Alignment, and Recognition, Marseille : +<br/>France (2008)" +</td><td></td><td></td></tr><tr><td>858b51a8a8aa082732e9c7fbbd1ea9df9c76b013</td><td>Can Computer Vision Problems Benefit from +<br/>Structured Hierarchical Classification? +<br/>Sandor Szedmak2 +<br/><b>INTELSIG, Monte ore Institute, University of Li`ege, Belgium</b><br/><b>Intelligent and Interactive Systems, Institute of Computer Science, University of</b><br/>Innsbruck, Austria +</td><td>('3104165', 'Thomas Hoyoux', 'thomas hoyoux')<br/>('1772389', 'Justus H. Piater', 'justus h. piater')</td><td></td></tr><tr><td>856317f27248cdb20226eaae599e46de628fb696</td><td>A Method Based on Convex Cone Model for +<br/>Image-Set Classification with CNN Features +<br/><b>Graduate School of Systems and Information Engineering, University of Tsukuba</b><br/>1-1-1 Tennodai, Tsukuba, Ibaraki, 305-8573, Japan +</td><td>('46230115', 'Naoya Sogi', 'naoya sogi')<br/>('2334316', 'Taku Nakayama', 'taku nakayama')<br/>('1770128', 'Kazuhiro Fukui', 'kazuhiro fukui')</td><td>Email: {sogi, nakayama}@cvlab.cs.tsukuba.ac.jp, kfukui@cs.tsukuba.ac.jp +</td></tr><tr><td>8518b501425f2975ea6dcbf1e693d41e73d0b0af</td><td>Relative Hidden Markov Models for Evaluating Motion Skills +<br/>Computer Science and Engineering +<br/>Arizona State Univerisity, Tempe, AZ 85281 +</td><td>('1689161', 'Qiang Zhang', 'qiang zhang')<br/>('2913552', 'Baoxin Li', 'baoxin li')</td><td>qzhang53,baoxin.li@asu.edu +</td></tr><tr><td>855184c789bca7a56bb223089516d1358823db0b</td><td>Automatic Procedure to Fix Closed-Eyes Image +<br/><b>University of California, Berkeley</b><br/>Figure 1: Pipeline to Fix Closed-Eyes Image +</td><td>('31781046', 'Hung Vu', 'hung vu')</td><td></td></tr><tr><td>853bd61bc48a431b9b1c7cab10c603830c488e39</td><td>Learning Face Representation from Scratch +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences (CASIA</b></td><td>('1716143', 'Dong Yi', 'dong yi')<br/>('1718623', 'Zhen Lei', 'zhen lei')<br/>('40397682', 'Shengcai Liao', 'shengcai liao')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>dong.yi, zlei, scliao, szli@nlpr.ia.ac.cn +</td></tr><tr><td>85639cefb8f8deab7017ce92717674d6178d43cc</td><td>Automatic Analysis of Spontaneous Facial Behavior: +<br/>A Final Project Report +<br/>(UCSD MPLab TR 2001.08, October 31 2001) +<br/><b>cid:1)Institute for Neural Computation</b><br/>(cid:2)Department of Cognitive Science +<br/><b>University of California, San Diego</b><br/><b>cid:3)The Salk Institute and Howard Hughes Medical Institute</b></td><td>('2218905', 'Marian S. Bartlett', 'marian s. bartlett')<br/>('33937541', 'Bjorn Braathen', 'bjorn braathen')<br/>('2039025', 'Ian Fasel', 'ian fasel')<br/>('1714528', 'Terrence J. Sejnowski', 'terrence j. sejnowski')<br/>('1741200', 'Javier R. Movellan', 'javier r. movellan')</td><td></td></tr><tr><td>854dbb4a0048007a49df84e3f56124d387588d99</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 13, NO. 9, SEPTEMBER 2014 +<br/>Spatial-Temporal Recurrent Neural Network for +<br/>Emotion Recognition +</td><td>('38144094', 'Tong Zhang', 'tong zhang')<br/>('40608983', 'Wenming Zheng', 'wenming zheng')<br/>('10338111', 'Zhen Cui', 'zhen cui')<br/>('2378869', 'Yuan Zong', 'yuan zong')<br/>('1678662', 'Yang Li', 'yang li')</td><td></td></tr><tr><td>85674b1b6007634f362cbe9b921912b697c0a32c</td><td>Optimizing Facial Landmark Detection by +<br/>Facial Attribute Learning +<br/><b>The Chinese University of Hong Kong</b></td><td>('3152448', 'Zhanpeng Zhang', 'zhanpeng zhang')<br/>('1693209', 'Ping Luo', 'ping luo')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td></td></tr><tr><td>1d21e5beef23eecff6fff7d4edc16247f0fd984a</td><td>Face Recognition from Video using the Generic +<br/>Shape-Illumination Manifold +<br/>Department of Engineering +<br/><b>University of Cambridge</b><br/>Cambridge, CB2 1PZ, UK +</td><td>('1745672', 'Roberto Cipolla', 'roberto cipolla')</td><td>{oa214,cipolla}@eng.cam.ac.uk +</td></tr><tr><td>1dbbec4ad8429788e16e9f3a79a80549a0d7ac7b</td><td></td><td></td><td></td></tr><tr><td>1d7ecdcb63b20efb68bcc6fd99b1c24aa6508de9</td><td>1860 +<br/>The Hidden Sides of Names—Face Modeling +<br/>with First Name Attributes +</td><td>('2896700', 'Huizhong Chen', 'huizhong chen')<br/>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')<br/>('1739786', 'Bernd Girod', 'bernd girod')</td><td></td></tr><tr><td>1d846934503e2bd7b8ea63b2eafe00e29507f06a</td><td></td><td></td><td></td></tr><tr><td>1d19c6857e798943cd0ecd110a7a0d514c671fec</td><td>Do Deep Neural Networks Learn Facial Action Units +<br/>When Doing Expression Recognition? +<br/><b>Beckman Institute for Advanced Science and Technology</b><br/><b>University of Illinois at Urbana-Champaign</b></td><td>('1911177', 'Pooya Khorrami', 'pooya khorrami')<br/>('40470211', 'Tom Le Paine', 'tom le paine')<br/>('1739208', 'Thomas S. Huang', 'thomas s. huang')</td><td>{pkhorra2,paine1,t-huang1}@illinois.edu +</td></tr><tr><td>1d1a7ef193b958f9074f4f236060a5f5e7642fc1</td><td>Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'13 \ +<br/>675 +<br/>Ensemble of Patterns of Oriented Edge Magnitudes +<br/>Descriptors For Face Recognition +<br/><b>Computer Information Systems, Missouri State University, 901 S. National, Springfield, MO 65804, USA</b><br/>faces; and 3) face tagging, which is a particular case of face +<br/>identification. +</td><td>('1804258', 'Loris Nanni', 'loris nanni')<br/>('1707759', 'Alessandra Lumini', 'alessandra lumini')<br/>('2292370', 'Sheryl Brahnam', 'sheryl brahnam')</td><td>*DEI, University o f Padua, viale Gradenigo 6, Padua, Italy, {loris.nanni, mauro.migliardi}@unipd.it; +<br/>2DEI, Universita di Bologna, Via Venezia 52, 47521 Cesena, Italy, alessandra.lumini@ unibo.it; +<br/>sbrahnam@missouristate.edu +</td></tr><tr><td>1d696a1beb42515ab16f3a9f6f72584a41492a03</td><td>Deeply learned face representations are sparse, selective, and robust +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('1681656', 'Yi Sun', 'yi sun')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>sy011@ie.cuhk.edu.hk +<br/>xgwang@ee.cuhk.edu.hk +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>1d1caaa2312390260f7d20ad5f1736099818d358</td><td>Resource-Allocating Codebook for Patch-based Face Recognition +<br/>School of Electronics and Computer Science +<br/><b>University of Southampton, SO17 1BJ, UK</b></td><td>('34672932', 'Amirthalingam Ramanan', 'amirthalingam ramanan')<br/>('1697360', 'Mahesan Niranjan', 'mahesan niranjan')</td><td>{ar07r,mn}@ecs.soton.ac.uk +</td></tr><tr><td>1dc241ee162db246882f366644171c11f7aed96d</td><td>Deep Action- and Context-Aware Sequence Learning for Activity Recognition +<br/>and Anticipation +<br/><b>Australian National University, 2Smart Vision Systems, CSIRO, 3CVLab, EPFL</b></td><td>('35441838', 'Fatemehsadat Saleh', 'fatemehsadat saleh')<br/>('1688071', 'Basura Fernando', 'basura fernando')<br/>('2862871', 'Mathieu Salzmann', 'mathieu salzmann')<br/>('2370776', 'Lars Petersson', 'lars petersson')<br/>('34234277', 'Lars Andersson', 'lars andersson')</td><td>firstname.lastname@data61.csiro.au, basura.fernando@anu.edu.au, mathieu.salzmann@epfl.ch +</td></tr><tr><td>1d0128b9f96f4c11c034d41581f23eb4b4dd7780</td><td>Automatic Construction Of Robust Spherical Harmonic Subspaces +<br/><b>Imperial College London</b></td><td>('2796644', 'Patrick Snape', 'patrick snape')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')</td><td>{p.snape,i.panagakis,s.zafeiriou}@imperial.ac.uk +</td></tr><tr><td>1d3dd9aba79a53390317ec1e0b7cd742cba43132</td><td>A Maximum Entropy Feature Descriptor for Age Invariant Face Recognition +<br/>(cid:31) +<br/>1Shenzhen Key Lab of Computer Vision and Pattern Recognition +<br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, China</b><br/><b>Centre for Quantum Computation and Intelligent Systems, Faculty of Engineering and IT, University of</b><br/>Technology, Sydney, NSW 2007, Australia +<br/><b>the Chinese University of Hong Kong</b><br/>4Media Lab, Huawei Technologies Co. Ltd., China +<br/><b>Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences</b></td><td>('2856494', 'Dihong Gong', 'dihong gong')<br/>('7137861', 'Jianzhuang Liu', 'jianzhuang liu')<br/>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('1720243', 'Xuelong Li', 'xuelong li')</td><td>dh.gong@siat.ac.cn +<br/>zhifeng.li@siat.ac.cn +<br/>dacheng.tao@uts.edu.au +<br/>liu.jianzhuang@huawei.com +<br/>xuelong_li@opt.ac.cn +</td></tr><tr><td>1d0dd20b9220d5c2e697888e23a8d9163c7c814b</td><td>NEGREL ET AL.: BOOSTED METRIC LEARNING FOR FACE RETRIEVAL +<br/>Boosted Metric Learning for Efficient +<br/>Identity-Based Face Retrieval +<br/>Frederic Jurie +<br/>GREYC, CNRS UMR 6072, ENSICAEN +<br/>Université de Caen Basse-Normandie +<br/>France +</td><td>('2838835', 'Romain Negrel', 'romain negrel')<br/>('2504258', 'Alexis Lechervy', 'alexis lechervy')</td><td>romain.negrel@unicaen.fr +<br/>alexis.lechervy@unicaen.fr +<br/>frederic.jurie@unicaen.fr +</td></tr><tr><td>1d5aad4f7fae6d414ffb212cec1f7ac876de48bf</td><td>Face Retriever: Pre-filtering the Gallery via Deep Neural Net +<br/>Department of Computer Science and Engineering +<br/><b>Michigan State University, East Lansing, MI 48824, U.S.A</b></td><td>('7496032', 'Dayong Wang', 'dayong wang')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td>{dywang, jain}@msu.edu +</td></tr><tr><td>1db23a0547700ca233aef9cfae2081cd8c5a04d7</td><td>www.ijecs.in +<br/>International Journal Of Engineering And Computer Science ISSN:2319-7242 +<br/>Volume 4 Issue 5 May 2015, Page No. 11945-11951 +<br/>Comparative study and evaluation of various data classification +<br/>techniques in data mining +<br/>1Research scholar +<br/>Department of computer science +<br/><b>Raipur institute of technology</b><br/>Raipur, India +<br/> 2Asst. professor +<br/> Department of computer science +<br/><b>Raipur institute of technology</b><br/> Raipur, India +</td><td>('1977125', 'Vivek Verma', 'vivek verma')</td><td>E-mail: vivekverma.exe@gmail.com +</td></tr><tr><td>1d776bfe627f1a051099997114ba04678c45f0f5</td><td>Deployment of Customized Deep Learning based +<br/>Video Analytics On Surveillance Cameras +<br/>AitoeLabs (www.aitoelabs.com) +</td><td>('46175439', 'Pratik Dubal', 'pratik dubal')<br/>('22549601', 'Rohan Mahadev', 'rohan mahadev')<br/>('9745898', 'Suraj Kothawade', 'suraj kothawade')<br/>('46208440', 'Kunal Dargan', 'kunal dargan')</td><td></td></tr><tr><td>1d97735bb0f0434dde552a96e1844b064af08f62</td><td>Weber Binary Pattern and Weber Ternary Pattern +<br/>for Illumination-Robust Face Recognition +<br/><b>Tsinghua University, China</b><br/>Shenzhen Key Laboratory of Information Science and Technology, Guangdong, China +</td><td>('35160104', 'Zuodong Yang', 'zuodong yang')<br/>('2312541', 'Yinyan Jiang', 'yinyan jiang')<br/>('40398990', 'Yong Wu', 'yong wu')<br/>('2265693', 'Zongqing Lu', 'zongqing lu')<br/>('1718891', 'Weifeng Li', 'weifeng li')<br/>('2883861', 'Qingmin Liao', 'qingmin liao')</td><td>(cid:3) E-mail: yangzd13@mails.tsinghua.edu.cn +<br/>y E-mail: Li.Weifeng@sz.tsinghua.edu.cn +</td></tr><tr><td>1d3e01d5e2721dcfafe5a3b39c54ee1c980350bb</td><td></td><td></td><td></td></tr><tr><td>1dff919e51c262c22630955972968f38ba385d8a</td><td>Toward an Affect-Sensitive Multimodal +<br/>Human–Computer Interaction +<br/>Invited Paper +<br/>The ability to recognize affective states of a person we are com- +<br/>municating with is the core of emotional intelligence. Emotional +<br/>intelligenceisa facet of human intelligence thathas been argued to be +<br/>indispensable and perhaps the most important for successful inter- +<br/>personal social interaction. This paper argues that next-generation +<br/>human–computer interaction (HCI) designs need to include the +<br/>essence of emotional intelligence—the ability to recognize a user’s +<br/>affective states—in order to become more human-like, more effec- +<br/>tive, and more efficient. Affective arousal modulates all nonverbal +<br/>communicative cues (facial expressions, body movements, and vocal +<br/>and physiological reactions). In a face-to-face interaction, humans +<br/>detect and interpret those interactive signals of their communicator +<br/>with little or no effort. Yet design and development of an automated +<br/>system that accomplishes these tasks is rather difficult. This paper +<br/>surveys the past work in solving these problems by a computer +<br/>and provides a set of recommendations for developing the first +<br/>part of an intelligent multimodal HCI—an automatic personalized +<br/>analyzer of a user’s nonverbal affective feedback. +<br/>Keywords—Affective computing, affective states, automatic +<br/>analysis of nonverbal communicative cues, human–computer +<br/>interaction (HCI), multimodal human–computer +<br/>interaction, +<br/>personalized human–computer interaction. +<br/>I. INTRODUCTION +<br/>The exploration of how we as human beings react to the +<br/>world and interact with it and each other remains one of +<br/>the greatest scientific challenges. Perceiving, learning, and +<br/>adapting to the world around us are commonly labeled as +<br/>“intelligent” behavior. But what does it mean being intelli- +<br/>gent? Is IQ a good measure of human intelligence and the +<br/>best predictor of somebody’s success in life? There is now +<br/>growing research in the fields of neuroscience, psychology, +<br/>and cognitive science which argues that our common view of +<br/>intelligence is too narrow, ignoring a crucial range of abilities +<br/>Manuscript received October 25, 2002; revised March 5, 2003. The work +<br/>of M. Pantic was supported by the Netherlands Organization for Scientific +<br/>Research (NWO) Grant EW-639.021.202. +<br/><b>The authors are with the Delft University of Technology, Data and Knowl</b><br/>edge Systems Group, Mediamatics Department, 2600 AJ Delft, The Nether- +<br/>Digital Object Identifier 10.1109/JPROC.2003.817122 +<br/>that matter immensely to how we do in life. This range +<br/>of abilities is called emotional intelligence [44], [96] and +<br/>includes the ability to have, express, and recognize affective +<br/>states, coupled with the ability to regulate them, employ them +<br/>for constructive purpose, and skillfully handle the affective +<br/>arousal of others. The skills of emotional intelligence have +<br/>been argued to be a better predictor than IQ for measuring +<br/>aspects of success in life [44], especially in interpersonal +<br/>communication, and learning and adapting to what +<br/>is +<br/>important [10], [96]. +<br/>When it comes to the world of computers, not all of them +<br/>will need emotional skills and probably none will need all +<br/>of the skills that humans need. Yet there are situations where +<br/>the man–machine interaction could be improved by having +<br/>machines capable of adapting to their users and where the in- +<br/>formation about how, when, and how important it is to adapt +<br/>involves information on the user’s affective state. In addition, +<br/>it seems that people regard computers as social agents with +<br/>whom “face-to-(inter)face” interaction may be most easy and +<br/>serviceable [11], [75], [90], [101], [110]. Human–computer +<br/>interaction (HCI) systems capable of sensing and responding +<br/>appropriately to the user’s affective feedback are, therefore, +<br/>likely to be perceived as more natural [73], more efficacious +<br/>and persuasive [93], and more trustworthy [14], [78]. +<br/>These findings, together with recent advances in sensing, +<br/>tracking, analyzing, and animating human nonverbal com- +<br/>municative signals, have produced a surge of interest in +<br/>affective computing by researchers of advanced HCI. This +<br/>intriguing new field focuses on computational modeling of +<br/>human perception of affective states, synthesis/animation of +<br/>affective expressions, and design of affect-sensitive HCI. +<br/>Indeed, the first step toward an intelligent HCI having the +<br/>abilities to sense and respond appropriately to the user’s af- +<br/>fective feedback is to detect and interpret affective states +<br/>shown by the user in an automatic way. This paper focuses +<br/>further on surveying the past work done on solving these +<br/>problems and providing an advanced HCI with one of the +<br/>key skills of emotional intelligence: the ability to recognize +<br/>the user’s nonverbal affective feedback. +<br/>0018-9219/03$17.00 © 2003 IEEE +<br/>1370 +<br/>PROCEEDINGS OF THE IEEE, VOL. 91, NO. 9, SEPTEMBER 2003 +</td><td>('1694605', 'MAJA PANTIC', 'maja pantic')</td><td>lands (e-mail: M.Pantic@cs.tudelft.nl; L.J.M.Rothkrantz@cs.tudelft.nl). +</td></tr><tr><td>1de8f38c35f14a27831130060810cf9471a62b45</td><td>Int J Comput Vis +<br/>DOI 10.1007/s11263-017-0989-7 +<br/>A Branch-and-Bound Framework for Unsupervised Common +<br/>Event Discovery +<br/>Received: 3 June 2016 / Accepted: 12 January 2017 +<br/>© Springer Science+Business Media New York 2017 +</td><td>('39336289', 'Wen-Sheng Chu', 'wen-sheng chu')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')<br/>('1874236', 'Daniel S. Messinger', 'daniel s. messinger')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td></td></tr><tr><td>1da83903c8d476c64c14d6851c85060411830129</td><td>Iterated Support Vector Machines for Distance +<br/>Metric Learning +</td><td>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('6292353', 'Faqiang Wang', 'faqiang wang')<br/>('1698371', 'David Zhang', 'david zhang')<br/>('1737218', 'Liang Lin', 'liang lin')<br/>('2224875', 'Yuchi Huang', 'yuchi huang')<br/>('1803714', 'Deyu Meng', 'deyu meng')<br/>('36685537', 'Lei Zhang', 'lei zhang')</td><td></td></tr><tr><td>1d6068631a379adbcff5860ca2311b790df3a70f</td><td></td><td></td><td></td></tr><tr><td>1dacc2f4890431d867a038fd81c111d639cf4d7e</td><td>2016, Vol. 125, No. 2, 310 –321 +<br/>0021-843X/16/$12.00 +<br/>© 2016 American Psychological Association +<br/>http://dx.doi.org/10.1037/abn0000139 +<br/>Using Social Outcomes to Inform Decision-Making in Schizophrenia: +<br/>Relationships With Symptoms and Functioning +<br/>Timothy R. Campellone, Aaron J. Fisher, and Ann M. Kring +<br/><b>University of California, Berkeley</b><br/>The outcomes of the decisions we make can be used to inform subsequent choices and behavior. We +<br/>investigated whether and how people with and without schizophrenia use positive and negative social +<br/>outcomes and emotional displays to inform decisions to place trust in social partners. We also investi- +<br/>gated the impact of reversals in social partners’ behavior on decisions to trust. Thirty-two people with +<br/>schizophrenia and 29 control participants completed a task in which they decided how much trust to place +<br/>in social partners’ showing either a dynamic emotional (smiling, scowling) or neutral display. Interac- +<br/>tions were predetermined to result in positive (trust reciprocated) or negative (trust abused) outcomes, +<br/>and we modeled changes in trust decisions over the course of repeated interactions. Compared to +<br/>controls, people with schizophrenia were less sensitive to positive social outcomes in that they placed less +<br/>trust in trustworthy social partners during initial interactions. By contrast, people with schizophrenia were +<br/>more sensitive to negative social outcomes during initial interactions with untrustworthy social partners, +<br/>placing less trust in these partners compared to controls. People with schizophrenia did not differ from +<br/>controls in detecting social partner behavior reversals from trustworthy to untrustworthy; however, they +<br/>had difficulties detecting reversals from untrustworthy to trustworthy. Importantly, decisions to trust +<br/>were associated with real-world social functioning. We discuss the implications of these findings for +<br/>understanding social engagement among people with schizophrenia and the development of psychosocial +<br/>interventions for social functioning. +<br/>General Scientific Summary +<br/>People with schizophrenia can have difficulties using decision outcomes to guide subsequent +<br/>decision-making and behavior. This study extends previous work by showing that people with +<br/>schizophrenia also have difficulties using social interaction outcomes to guide subsequent social +<br/>decision-making and behavior. These findings have implications for understanding decreased social +<br/>networks common among people with schizophrenia. +<br/>Keywords: schizophrenia, decision-making, social interactions, trust +<br/>Decision-making is an important part of daily life, with the +<br/>outcomes of decisions influencing subsequent choices and deci- +<br/>sions. While prior research has shown that people with schizo- +<br/>phrenia have difficulty using monetary outcomes to guide subse- +<br/>quent decisions (Heerey & Gold, 2007; Barch & Dowd, 2010), we +<br/>know considerably less about whether people with schizophrenia +<br/>have difficulty using social outcomes to inform decision-making +<br/>in the context of social interactions. We investigated the extent to +<br/>Timothy R. Campellone, Aaron J. Fisher, and Ann M. Kring, Depart- +<br/><b>ment of Psychology, University of California, Berkeley</b><br/><b>Funding was provided by the U.S. National Institutes of Mental</b><br/>Health (Grant 5T32MH089919 to Timothy R. Campellone and Grant +<br/>1R01MH082890 to Ann M. Kring). We are grateful to Janelle Painter, +<br/>Erin Moran, and Jasmine Mote for their help in collecting this data. We are +<br/>also grateful to Stephen Hinshaw for reading a previous version of this +<br/>article. We would also like to thank all the participants in this study. +<br/>Correspondence concerning this article should be addressed to Timothy +<br/><b>R. Campellone, 3210 Tolman Hall, University of California, Berkeley</b><br/>310 +<br/>which people with schizophrenia use social outcomes to inform +<br/>decision-making, and how this is related to motivation/pleasure +<br/>negative symptoms and psychosocial functioning. Because social +<br/>interactions often involve emotion, we also examined whether and +<br/>how people with schizophrenia use social partners’ emotional +<br/>displays to guide learning from social outcomes and inform sub- +<br/>sequent decision-making. +<br/>Monetary Decision-Making and Reversal Learning +<br/>in Schizophrenia +<br/>Studies using reward-learning paradigms with monetary out- +<br/>comes have consistently shown that compared to controls, people +<br/>with schizophrenia have difficulty using positive outcomes to +<br/>inform decision-making (Strauss et al., 2011; Gold et al., 2012). +<br/>These difficulties are associated with poorer functioning (Somlai, +<br/>Moustafa, Kéri, Myers, & Gluck, 2011) as well as greater moti- +<br/>vation/pleasure negative symptoms (Strauss et al., 2011; Gold et +<br/>al., 2012), which are part of the two-factor solution of negative +<br/>symptoms and refer to diminished engagement in and/or pleasure +<br/>derived from social, vocational, and recreational life domains +<br/>(Kring, Gur, Blanchard, Horan, & Reise, 2013). By contrast, +</td><td></td><td>Berkeley, CA 94720-1690. E-mail: tcampellone@berkeley.edu +</td></tr><tr><td>1de690714f143a8eb0d6be35d98390257a3f4a47</td><td>Face Detection Using Spectral Histograms and SVMs +<br/><b>The Florida State University</b><br/>Tallahassee, FL 32306 +</td><td>('3209925', 'Christopher A. Waring', 'christopher a. waring')<br/>('1800002', 'Xiuwen Liu', 'xiuwen liu')</td><td>chwaring@cs.fsu.edu liux@cs.fsu.edu +</td></tr><tr><td>1d7df3df839a6aa8f5392310d46b2a89080a3c25</td><td>Large-Margin Softmax Loss for Convolutional Neural Networks +<br/>Meng Yang4 +<br/><b>School of ECE, Peking University 2School of EIE, South China University of Technology</b><br/><b>Carnegie Mellon University 4College of CS and SE, Shenzhen University</b></td><td>('36326884', 'Weiyang Liu', 'weiyang liu')<br/>('2512949', 'Yandong Wen', 'yandong wen')<br/>('1751019', 'Zhiding Yu', 'zhiding yu')</td><td>WYLIU@PKU.EDU.CN +<br/>WEN.YANDONG@MAIL.SCUT.EDU.CN +<br/>YZHIDING@ANDREW.CMU.EDU +<br/>YANG.MENG@SZU.EDU.CN +</td></tr><tr><td>1d6c09019149be2dc84b0c067595f782a5d17316</td><td>Encoding Video and Label Priors for Multi-label Video Classification +<br/>on YouTube-8M dataset +<br/><b>Seoul National University</b><br/><b>Seoul National University</b><br/><b>Seoul National University</b><br/>SK Telecom Video Tech. Lab +<br/><b>Seoul National University</b></td><td>('19255603', 'Seil Na', 'seil na')<br/>('7877122', 'Youngjae Yu', 'youngjae yu')<br/>('1693291', 'Sangho Lee', 'sangho lee')<br/>('2077253', 'Jisung Kim', 'jisung kim')<br/>('1743920', 'Gunhee Kim', 'gunhee kim')</td><td>seil.na@vision.snu.ac.kr +<br/>yj.yu@vision.snu.ac.kr +<br/>sangho.lee@vision.snu.ac.kr +<br/>joyful.kim@sk.com +<br/>gunhee@snu.ac.kr +</td></tr><tr><td>1d58d83ee4f57351b6f3624ac7e727c944c0eb8d</td><td>Enhanced Local Texture +<br/>Feature Sets for Face +<br/>Recognition under Difficult +<br/>Lighting Conditions +<br/>INRIA & Laboratoire Jean +<br/>Kuntzmann, +<br/>655 avenue de l'Europe, Montbonnot 38330, France +</td><td>('2248421', 'Xiaoyang Tan', 'xiaoyang tan')<br/>('1756114', 'Bill Triggs', 'bill triggs')</td><td></td></tr><tr><td>1d729693a888a460ee855040f62bdde39ae273af</td><td>Photorealistic Face de-Identification by Aggregating +<br/>Donors’ Face Components +<br/>To cite this version: +<br/>gating Donors’ Face Components. Asian Conference on Computer Vision, Nov 2014, Singapore. +<br/>pp.1-16, 2014. <hal-01070658> +<br/>HAL Id: hal-01070658 +<br/>https://hal.archives-ouvertes.fr/hal-01070658 +<br/>Submitted on 2 Oct 2014 +<br/>HAL is a multi-disciplinary open access +<br/>archive for the deposit and dissemination of sci- +<br/>entific research documents, whether they are pub- +<br/>lished or not. The documents may come from +<br/>teaching and research institutions in France or +<br/><b>abroad, or from public or private research centers</b><br/>L’archive ouverte pluridisciplinaire HAL, est +<br/>destin´ee au d´epˆot et `a la diffusion de documents +<br/>scientifiques de niveau recherche, publi´es ou non, +<br/>´emanant des ´etablissements d’enseignement et de +<br/>recherche fran¸cais ou ´etrangers, des laboratoires +<br/>publics ou priv´es. +</td><td>('3095534', 'Saleh Mosaddegh', 'saleh mosaddegh')<br/>('3095534', 'Saleh Mosaddegh', 'saleh mosaddegh')</td><td></td></tr><tr><td>1d4c25f9f8f08f5a756d6f472778ab54a7e6129d</td><td>International Journal of Science and Research (IJSR) +<br/>ISSN (Online): 2319-7064 +<br/>Index Copernicus Value (2014): 6.14 | Impact Factor (2014): 4.438 +<br/>An Innovative Mean Approach for Plastic Surgery +<br/>Face Recognition +<br/>1 Student of M.E., Department of Electronics & Telecommunication Engineering, +<br/><b>P. R. Patil College of Engineering, Amravati Maharashtra India</b><br/>2 Assistant Professor, Department of Electronics & Telecommunication Engineering, +<br/><b>P. R. Patil College of Engineering, Amravati Maharashtra India</b></td><td>('2936550', 'Umesh W. Hore', 'umesh w. hore')</td><td></td></tr><tr><td>71b376dbfa43a62d19ae614c87dd0b5f1312c966</td><td>The Temporal Connection Between Smiles and Blinks +</td><td>('2048839', 'Laura C. Trutoiu', 'laura c. trutoiu')<br/>('1788773', 'Jessica K. Hodgins', 'jessica k. hodgins')<br/>('1737918', 'Jeffrey F. Cohn', 'jeffrey f. cohn')</td><td></td></tr><tr><td>71b07c537a9e188b850192131bfe31ef206a39a0</td><td>Image and Vision Computing 47 (2016) 3–18 +<br/>Contents lists available at ScienceDirect +<br/>Image and Vision Computing +<br/>j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / i m a v i s +<br/>300 Faces In-The-Wild Challenge: database and results夽,夽夽 +<br/><b>aImperial College London, London, UK</b><br/><b>bUniversity of Nottingham, School of Computer Science, Nottingham, UK</b><br/><b>cFaculty of Electrical Engineering, Mathematics, and Computer Science, University of Twente, The Netherlands</b><br/>A R T I C L E +<br/>I N F O +<br/>A B S T R A C T +<br/>Article history: +<br/>Received 19 March 2015 +<br/>Received in revised form 2 October 2015 +<br/>Accepted 4 January 2016 +<br/>Available online 25 January 2016 +<br/>Keywords: +<br/>Facial landmark localization +<br/>Challenge +<br/>Semi-automatic annotation tool +<br/>Facial database +<br/>Computer Vision has recently witnessed great research advance towards automatic facial points detection. +<br/>Numerous methodologies have been proposed during the last few years that achieve accurate and efficient +<br/>performance. However, fair comparison between these methodologies is infeasible mainly due to two issues. +<br/>(a) Most existing databases, captured under both constrained and unconstrained (in-the-wild) conditions +<br/>have been annotated using different mark-ups and, in most cases, the accuracy of the annotations is low. (b) +<br/>Most published works report experimental results using different training/testing sets, different error met- +<br/>rics and, of course, landmark points with semantically different locations. In this paper, we aim to overcome +<br/>the aforementioned problems by (a) proposing a semi-automatic annotation technique that was employed +<br/>to re-annotate most existing facial databases under a unified protocol, and (b) presenting the 300 Faces In- +<br/>The-Wild Challenge (300-W), the first facial landmark localization challenge that was organized twice, in +<br/>2013 and 2015. To the best of our knowledge, this is the first effort towards a unified annotation scheme +<br/>of massive databases and a fair experimental comparison of existing facial landmark localization systems. +<br/>The images and annotations of the new testing database that was used in the 300-W challenge are available +<br/>from http://ibug.doc.ic.ac.uk/resources/300-W_IMAVIS/. +<br/>© 2016 Elsevier B.V. All rights reserved. +<br/>1. Introduction +<br/>During the last decades we notice a wealth of scientific research +<br/>in computer vision for the problem of facial landmark points localiza- +<br/>tion using visual deformable models. The main reason behind this are +<br/>the countless applications that the problem has in human-computer +<br/>interaction and facial expression recognition. Numerous methodolo- +<br/>gies have been proposed that are shown to achieve great accuracy +<br/>and efficiency. They can be roughly divided into two categories: +<br/>generative and discriminative. The generative techniques, which aim +<br/>to find the parameters that maximize the probability of the test +<br/><b>image being generated by the model, include Active Appearance</b><br/>Models (AAMs) [1,2], their improved extensions [3–10] and Pictorial +<br/>夽 The contribution of the first two authors on writing this paper is equal, with +<br/>various steps needed to run 300-W successfully including data annotation, annotation +<br/>tool development, and running the experiments. +<br/>夽夽 This paper has been recommended for acceptance by Richard Bowden, PhD. +<br/>* Corresponding author. +<br/>http://dx.doi.org/10.1016/j.imavis.2016.01.002 +<br/>0262-8856/© 2016 Elsevier B.V. All rights reserved. +<br/>Structures [11–13]. The discriminative techniques can be further +<br/>divided to those that use discriminative response map functions, +<br/>such as Active Shape Models (ASMs) [14], Constrained Local Models +<br/>(CLMs) [15–17] and Deformable Part Models (DPMs) [18], those that +<br/>learn a cascade of regression functions, such as Supervised Descent +<br/>Method (SDM) [19] and others [20–22], and, finally, those that +<br/>employ random forests [23,24]. +<br/>Arguably, the main reason why many researchers of the field +<br/>focus on the problem of face alignment is the plethora of publicly +<br/>available annotated facial databases. These databases can be sepa- +<br/>rated in two major categories: (a) those captured under controlled +<br/>conditions, e.g. Multi-PIE [25], XM2VTS [26], FRGC-V2 [27], and +<br/>AR [28], and (b) those captured under totally unconstrained condi- +<br/>tions (in-the-wild), e.g. LFPW [29], HELEN [30], AFW[18], AFLW[31], +<br/><b>and IBUG [32]. All of them cover large variations, including different</b><br/>subjects, poses, illumination conditions, expressions and occlusions. +<br/>However, for most of them, the provided annotations appear to have +<br/>several limitations. Specifically: +<br/>• The majority of them provide annotations for a relatively small +<br/>subset of images. +</td><td>('3320415', 'Christos Sagonas', 'christos sagonas')<br/>('2788012', 'Epameinondas Antonakos', 'epameinondas antonakos')<br/>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')<br/>('3320415', 'Christos Sagonas', 'christos sagonas')</td><td>E-mail address: c.sagonas@imperial.ac.uk (C. Sagonas). +</td></tr><tr><td>71fd29c2ae9cc9e4f959268674b6b563c06d9480</td><td>End-to-end 3D shape inverse rendering of different classes +<br/>of objects from a single input image +<br/>1Computer Science and Engineering and Information Technology, Shiraz +<br/><b>university, Shiraz, Iran</b><br/>November 17, 2017 +</td><td>('34649340', 'Shima Kamyab', 'shima kamyab')<br/>('2014752', 'Zohreh Azimifar', 'zohreh azimifar')</td><td></td></tr><tr><td>7142ac9e4d5498037aeb0f459f278fd28dae8048</td><td>Semi-Supervised Learning for Optical Flow +<br/>with Generative Adversarial Networks +<br/><b>University of California, Merced</b><br/>2Virginia Tech +<br/>3Nvidia Research +</td><td>('2268189', 'Wei-Sheng Lai', 'wei-sheng lai')<br/>('3068086', 'Jia-Bin Huang', 'jia-bin huang')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td>1{wlai24|mhyang}@ucmerced.edu +<br/>2jbhuang@vt.edu +</td></tr><tr><td>71f36c8e17a5c080fab31fce1ffea9551fc49e47</td><td>Predicting Failures of Vision Systems +<br/>1Virginia Tech +<br/>2Univ. of Texas at Austin +<br/>3Univ. of Washington +<br/><b>Carnegie Mellon University</b></td><td>('40409467', 'Peng Zhang', 'peng zhang')<br/>('2537394', 'Jiuling Wang', 'jiuling wang')</td><td>1{zhangp, parikh}@vt.edu +<br/>2jiuling@utexas.edu +<br/>3ali@cs.uw.edu +<br/>4hebert@ri.cmu.edu +</td></tr><tr><td>7117ed0be436c0291bc6fb6ea6db18de74e2464a</td><td>Under review as a conference paper at ICLR 2017 +<br/>WARPED CONVOLUTIONS: EFFICIENT INVARIANCE TO +<br/>SPATIAL TRANSFORMATIONS +<br/>Visual Geometry Group +<br/><b>University of Oxford</b></td><td>('36478254', 'João F. Henriques', 'joão f. henriques')</td><td>{joao,vedaldi}@robots.ox.ac.uk +</td></tr><tr><td>71e6a46b32a8163c9eda69e1badcee6348f1f56a</td><td>Visually Interpreting Names as Demographic Attributes +<br/>by Exploiting Click-Through Data +<br/><b>National Taiwan University, Taipei, Taiwan</b><br/><b>FX Palo Alto Laboratory, Inc., California, USA</b></td><td>('35081710', 'Yan-Ying Chen', 'yan-ying chen')<br/>('1692811', 'Yin-Hsi Kuo', 'yin-hsi kuo')<br/>('2580465', 'Chun-Che Wu', 'chun-che wu')<br/>('1716836', 'Winston H. Hsu', 'winston h. hsu')</td><td>{yanying,kuonini,kenwu0721}@gmail.com, whsu@ntu.edu.tw +</td></tr><tr><td>713594c18978b965be87651bb553c28f8501df0a</td><td>Fast Proximal Linearized Alternating Direction Method of Multiplier with +<br/>Parallel Splitting +<br/><b>National University of Singapore</b><br/><b>Key Laboratory of Machine Perception (MOE), School of EECS, Peking University</b><br/><b>Cooperative Medianet Innovation Center, Shanghai Jiaotong University</b></td><td>('33224509', 'Canyi Lu', 'canyi lu')<br/>('1775194', 'Huan Li', 'huan li')<br/>('33383055', 'Zhouchen Lin', 'zhouchen lin')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')</td><td>canyilu@gmail.com, lihuan ss@126.com, zlin@pku.edu.cn, eleyans@nus.edu.sg +</td></tr><tr><td>718824256b4461d62d192ab9399cfc477d3660b4</td><td>Selecting Training Data for Cross-Corpus Speech Emotion Recognition: +<br/>Prototypicality vs. Generalization +<br/><b>Institute for Human-Machine Communication, Technische Universit at M unchen, Germany</b></td><td>('30512170', 'Zixing Zhang', 'zixing zhang')<br/>('1740602', 'Felix Weninger', 'felix weninger')<br/>('1705843', 'Gerhard Rigoll', 'gerhard rigoll')</td><td>{schuller|zixing.zhang|weninger|rigoll}@tum.de +</td></tr><tr><td>718d3137adba9e3078fa1f698020b666449f3336</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 8, No. 10, 2017 +<br/>Accuracy Based Feature Ranking Metric for +<br/>Multi-Label Text Classification +<br/><b>Al-Khwarizmi Institute of Computer Science</b><br/><b>University of Engineering and Technology</b><br/>Department of Computer +<br/>Science, +<br/>Department of Computer +<br/>Science, +<br/>Lahore, Pakistan +<br/><b>University of Gujrat, Pakistan</b><br/><b>University of Gujrat, Pakistan</b></td><td>('35637737', 'Muhammad Nabeel Asim', 'muhammad nabeel asim')<br/>('3245405', 'Abdur Rehman', 'abdur rehman')<br/>('1981732', 'Umar Shoaib', 'umar shoaib')</td><td></td></tr><tr><td>714d487571ca0d676bad75c8fa622d6f50df953b</td><td>eBear: An Expressive Bear-Like Robot +</td><td>('49470290', 'Xiao Zhang', 'xiao zhang')<br/>('2314025', 'Ali Mollahosseini', 'ali mollahosseini')<br/>('29764067', 'Evan Boucher', 'evan boucher')<br/>('1783240', 'Richard M. Voyles', 'richard m. voyles')</td><td></td></tr><tr><td>716d6c2eb8a0d8089baf2087ce9fcd668cd0d4c0</td><td>SMITH & DYER: 3D FACIAL LANDMARK ESTIMATION +<br/>Pose-Robust 3D Facial Landmark Estimation +<br/>from a Single 2D Image +<br/>http://www.cs.wisc.edu/~bmsmith +<br/>http://www.cs.wisc.edu/~dyer +<br/>Department of Computer Sciences +<br/><b>University of Wisconsin-Madison</b><br/>Madison, WI USA +</td><td>('2721523', 'Brandon M. Smith', 'brandon m. smith')<br/>('1724754', 'Charles R. Dyer', 'charles r. dyer')</td><td></td></tr><tr><td>7143518f847b0ec57a0ff80e0304c89d7e924d9a</td><td>Speeding-up Age Estimation in Intelligent +<br/>Demographics System via Network Optimization +<br/><b>School of Computer and Information, Hefei University of Technology, Hefei, China</b><br/><b>School of Computer Science and Engineering, Nanyang Technological University, Singapore</b></td><td>('49941674', 'Zhenzhen Hu', 'zhenzhen hu')<br/>('7739626', 'Peng Sun', 'peng sun')<br/>('40096128', 'Yonggang Wen', 'yonggang wen')</td><td>huzhen.ice@gmail.com, {sunp0003, ygwen}@ntu.edu.sg +</td></tr><tr><td>710011644006c18291ad512456b7580095d628a2</td><td>Learning Residual Images for Face Attribute Manipulation +<br/>Fujitsu Research & Development Center, Beijing, China. +</td><td>('48157627', 'Wei Shen', 'wei shen')<br/>('2113095', 'Rujie Liu', 'rujie liu')</td><td>{shenwei, rjliu}@cn.fujitsu.com +</td></tr><tr><td>713db3874b77212492d75fb100a345949f3d3235</td><td>Deep Semantic Face Deblurring +<br/><b>Beijing Institute of Technology</b><br/><b>University of California, Merced</b><br/>3Nvidia +<br/>4Google Cloud +<br/>https://sites.google.com/site/ziyishenmi/cvpr18_face_deblur +</td><td>('2182388', 'Ziyi Shen', 'ziyi shen')<br/>('2268189', 'Wei-Sheng Lai', 'wei-sheng lai')<br/>('39001620', 'Tingfa Xu', 'tingfa xu')<br/>('1690538', 'Jan Kautz', 'jan kautz')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>715b69575dadd7804b4f8ccb419a3ad8b7b7ca89</td><td>1 +<br/>Testing separability and independence of perceptual +<br/>dimensions with general recognition theory: A tutorial and +<br/>new R package (grtools)1 +<br/><b>Florida International University</b><br/><b>University of California, Santa Barbara</b><br/><b>Florida International University</b><br/><b>University of California, Santa Barbara</b></td><td>('2850756', 'Fabian A. Soto', 'fabian a. soto')<br/>('33897174', 'Johnny Fonseca', 'johnny fonseca')<br/>('5854837', 'F. Gregory Ashby', 'f. gregory ashby')</td><td></td></tr><tr><td>71e56f2aebeb3c4bb3687b104815e09bb4364102</td><td>Video Co-segmentation for Meaningful Action Extraction +<br/><b>National University of Singapore, Singapore</b><br/><b>National University of Singapore Research Institute, Suzhou, China</b></td><td>('3036190', 'Jiaming Guo', 'jiaming guo')<br/>('3119455', 'Zhuwen Li', 'zhuwen li')<br/>('1809333', 'Steven Zhiying Zhou', 'steven zhiying zhou')</td><td>{guo.jiaming, lizhuwen, eleclf, elezzy}@nus.edu.sg +</td></tr><tr><td>711bb5f63139ee7a9b9aef21533f959671a7d80e</td><td><b>Helsinki University of Technology Laboratory of Computational Engineering Publications</b><br/>Teknillisen korkeakoulun Laskennallisen tekniikan laboratorion julkaisuja +<br/>Espoo 2007 +<br/>REPORT B68 +<br/>OBJECTS EXTRACTION AND RECOGNITION FOR +<br/>CAMERA-BASED INTERACTION: HEURISTIC AND +<br/>STATISTICAL APPROACHES +<br/>TEKNILLINEN KORKEAKOULU +<br/>TEKNILLINEN KORKEAKOULU +<br/>TEKNISKA HÖGSKOLAN +<br/>TEKNISKA HÖGSKOLAN +<br/><b>HELSINKI UNIVERSITY OF TECHNOLOGY</b><br/><b>HELSINKI UNIVERSITY OF TECHNOLOGY</b><br/>TECHNISCHE UNIVERSITÄT HELSINKI +<br/>TECHNISCHE UNIVERSITÄT HELSINKI +<br/>UNIVERSITE DE TECHNOLOGIE D'HELSINKI +<br/>UNIVERSITE DE TECHNOLOGIE D'HELSINKI +</td><td>('37522511', 'Hao Wang', 'hao wang')</td><td></td></tr><tr><td>76fd801981fd69ff1b18319c450cb80c4bc78959</td><td>Proceedings of the 11th International Conference on Computational Semantics, pages 76–81, +<br/>London, UK, April 15-17 2015. c(cid:13)2015 Association for Computational Linguistics +<br/>76 +</td><td></td><td></td></tr><tr><td>76dc11b2f141314343d1601635f721fdeef86fdb</td><td>Weighted Decoding ECOC for Facial +<br/>Action Unit Classification +</td><td>('1732556', 'Terry Windeatt', 'terry windeatt')</td><td></td></tr><tr><td>76673de6d81bedd6b6be68953858c5f1aa467e61</td><td>Discovering a Lexicon of Parts and Attributes +<br/><b>Toyota Technological Institute at Chicago</b><br/>Chicago, IL 60637, USA +</td><td>('35208858', 'Subhransu Maji', 'subhransu maji')</td><td>smaji@ttic.edu +</td></tr><tr><td>76cd5e43df44e389483f23cb578a9015d1483d70</td><td>BORGHI ET AL.: FACE VERIFICATION FROM DEPTH +<br/>Face Verification from Depth using +<br/>Privileged Information +<br/>Department of Engineering +<br/>"Enzo Ferrari" +<br/><b>University of Modena and Reggio</b><br/>Emilia +<br/>Modena, Italy +</td><td>('12010968', 'Guido Borghi', 'guido borghi')<br/>('2035969', 'Stefano Pini', 'stefano pini')<br/>('32044032', 'Filippo Grazioli', 'filippo grazioli')<br/>('1723285', 'Roberto Vezzani', 'roberto vezzani')<br/>('1741922', 'Rita Cucchiara', 'rita cucchiara')</td><td>guido.borghi@unimore.it +<br/>stefano.pini@unimore.it +<br/>filippo.grazioli@unimore.it +<br/>roberto.vezzani@unimore.it +<br/>rita.cucchiara@unimore.it +</td></tr><tr><td>7643861bb492bf303b25d0306462f8fb7dc29878</td><td>Speeding up 2D-Warping for Pose-Invariant Face Recognition +<br/><b>Human Language Technology and Pattern Recognition Group, RWTH Aachen University, Germany</b></td><td>('1804963', 'Harald Hanselmann', 'harald hanselmann')<br/>('1685956', 'Hermann Ney', 'hermann ney')</td><td>surname@cs.rwth-aachen.de +</td></tr><tr><td>760a712f570f7a618d9385c0cee7e4d0d6a78ed2</td><td></td><td></td><td></td></tr><tr><td>76b11c281ac47fe6d95e124673a408ee9eb568e3</td><td>International Journal of Latest Engineering and Management Research (IJLEMR) +<br/>ISSN: 2455-4847 +<br/>www.ijlemr.com || Volume 02 - Issue 03 || March 2017 || PP. 59-71 +<br/>REAL-TIME MULTI VIEW FACE DETECTION AND POSE +<br/>ESTIMATION +<br/><b>U. G STUDENTS, DEPT OF CSE, ALPHA COLLEGE OF ENGINEERING, CHENNAI</b><br/><b>ALPHA COLLEGE OF ENGINEERING, CHENNAI</b></td><td></td><td></td></tr><tr><td>76ce3d35d9370f0e2e27cfd29ea0941f1462895f</td><td>Hindawi Publishing Corporation +<br/>e Scientific World Journal +<br/>Volume 2014, Article ID 528080, 13 pages +<br/>http://dx.doi.org/10.1155/2014/528080 +<br/>Research Article +<br/>Efficient Parallel Implementation of Active Appearance +<br/>Model Fitting Algorithm on GPU +<br/><b>School of Computer Science and Technology, Tianjin University, Tianjin 300072, China</b><br/><b>College of Computer and Information Engineering, Tianjin Normal University, Tianjin 300387, China</b><br/>Received 25 August 2013; Accepted 19 January 2014; Published 2 March 2014 +<br/>Academic Editors: I. Lanese and G. Wei +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>The active appearance model (AAM) is one of the most powerful model-based object detecting and tracking methods which +<br/>has been widely used in various situations. However, the high-dimensional texture representation causes very time-consuming +<br/>computations, which makes the AAM difficult to apply to real-time systems. The emergence of modern graphics processing +<br/>units (GPUs) that feature a many-core, fine-grained parallel architecture provides new and promising solutions to overcome the +<br/>computational challenge. In this paper, we propose an efficient parallel implementation of the AAM fitting algorithm on GPUs. +<br/>Our design idea is fine grain parallelism in which we distribute the texture data of the AAM, in pixels, to thousands of parallel GPU +<br/>threads for processing, which makes the algorithm fit better into the GPU architecture. We implement our algorithm using the +<br/>compute unified device architecture (CUDA) on the Nvidia’s GTX 650 GPU, which has the latest Kepler architecture. To compare +<br/>the performance of our algorithm with different data sizes, we built sixteen face AAM models of different dimensional textures. +<br/>The experiment results show that our parallel AAM fitting algorithm can achieve real-time performance for videos even on very +<br/>high-dimensional textures. +<br/>1. Introduction +<br/>Detecting and tracking moving deformable objects in a video +<br/>sequence is a complex and difficult task and has been a +<br/>very important part of many applications, such as human +<br/>computer interaction [1], automated surveillance [2], and +<br/>emotion recognition [3]. This task allows us to determine the +<br/>state of objects and helps us analyze their behaviors. +<br/>The active appearance model (AAM) [4], first proposed +<br/>by Cootes et al. [5], is one of the most powerful model-based +<br/>object detecting and tracking algorithms. It is a nonlinear, +<br/>generative, and parametric model and can be traced back +<br/>to the active contour model (or “snakes,” [6]) and the active +<br/>shape model (ASM) [7]. Particularly, the AAM decouples and +<br/>models the shape and the texture of the deformable object +<br/>to generate a variety of instant photos realistically. Therefore, +<br/>the AAM has been widely used in various situations [8–10]. +<br/>The most frequent application of AAMs to date has been face +<br/>modeling and tracking [11]. +<br/>Although the AAM possesses powerful modeling and +<br/>efficient fitting ability, the high computational complexity +<br/>caused by the high-dimensional texture representation limits +<br/>its application in many conditions, for example, real-time +<br/>systems. To make the AAM more applicable to practical +<br/>applications, additional effort must be spent to accelerate the +<br/>computation of the AAM. Therefore, several improvements +<br/>are proposed to achieve this aim. Some methods are proposed +<br/>to reduce the dimension of the texture, such as the Haar +<br/>wavelet [12], the wedgelet-based regression tree [13], and +<br/>the local sampling [14]. However, these methods improve +<br/>efficiency at the expense of decreasing accuracy or losing +<br/>detail information. From another perspective, researchers +<br/>[15, 16] suggest reformulating the AAM in an analytic way to +<br/>speed up the model fitting. A famous method is the inverse +<br/>compositional image alignment (ICIA) [17] algorithm that +<br/>avoids updating texture parameters every frame and is a very +<br/>fast-fitting algorithm for the AAM. However, the limitation +<br/>of this algorithm is that it cannot be applied to the AAMs +</td><td>('1762397', 'Jinwei Wang', 'jinwei wang')<br/>('2518530', 'Xirong Ma', 'xirong ma')<br/>('34854285', 'Yuanping Zhu', 'yuanping zhu')<br/>('35900806', 'Jizhou Sun', 'jizhou sun')<br/>('1762397', 'Jinwei Wang', 'jinwei wang')</td><td>Correspondence should be addressed to Jinwei Wang; wangjinwei@tju.edu.cn +</td></tr><tr><td>76b9fe32d763e9abd75b427df413706c4170b95c</td><td></td><td></td><td></td></tr><tr><td>768c332650a44dee02f3d1d2be1debfa90a3946c</td><td>Bayesian Face Recognition Using Support Vector Machine and Face Clustering +<br/>Department of Information Engineering +<br/><b>The Chinese University of Hong Kong</b><br/>Shatin, Hong Kong +</td><td>('1911510', 'Zhifeng Li', 'zhifeng li')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{zli0, xtang}@ie.cuhk.edu.hk +</td></tr><tr><td>769461ff717d987482b28b32b1e2a6e46570e3ff</td><td>MIC-TJU in MediaEval 2017 Emotional Impact of Movies Task +<br/><b>Gannan Normal University, Ganzhou 341000, China</b><br/><b>Tongji University, Shanghai 201804, China</b></td><td>('40290178', 'Yun Yi', 'yun yi')<br/>('2774427', 'Hanli Wang', 'hanli wang')<br/>('28933059', 'Jiangchuan Wei', 'jiangchuan wei')</td><td></td></tr><tr><td>76d9f5623d3a478677d3f519c6e061813e58e833</td><td>FAST ALGORITHMS FOR THE GENERALIZED FOLEY-SAMMON +<br/>DISCRIMINANT ANALYSIS +</td><td>('35789819', 'Lei-Hong Zhang', 'lei-hong zhang')<br/>('14372428', 'Li-Zhi Liao', 'li-zhi liao')<br/>('1678715', 'Michael K. Ng', 'michael k. ng')</td><td></td></tr><tr><td>76e2d7621019bd45a5851740bd2742afdcf62837</td><td>Article +<br/>Real-Time Detection and Measurement of Eye +<br/>Features from Color Images +<br/><b>Technical University of Cluj Napoca, 28 Memorandumului Street</b><br/><b>Babes Bolyai University, 58-60 Teodor Mihali, C333, Cluj Napoca</b><br/>Academic Editors: Changzhi Li, Roberto Gómez-García and José-María Muñoz-Ferreras +<br/>Received: 28 April 2016; Accepted: 14 July 2016; Published: 16 July 2016 +</td><td>('31630857', 'Diana Borza', 'diana borza')<br/>('1821352', 'Adrian Sergiu Darabant', 'adrian sergiu darabant')<br/>('3331727', 'Radu Danescu', 'radu danescu')</td><td>Cluj Napoca 400114, Romania; borza_diana@yahoo.com +<br/>Romania; adrian.darabant@tvarita.ro +<br/>* Correspondence: Radu.Danescu@cs.utcluj.ro; Tel.: +40-740-502-223 +</td></tr><tr><td>765b2cb322646c52e20417c3b44b81f89860ff71</td><td>PoseShop: Human Image Database +<br/>Construction and Personalized +<br/>Content Synthesis +</td><td>('29889388', 'Tao Chen', 'tao chen')<br/>('37291674', 'Ping Tan', 'ping tan')<br/>('1678872', 'Li-Qian Ma', 'li-qian ma')<br/>('37535930', 'Ming-Ming Cheng', 'ming-ming cheng')<br/>('2947946', 'Ariel Shamir', 'ariel shamir')<br/>('1686809', 'Shi-Min Hu', 'shi-min hu')</td><td></td></tr><tr><td>7644d90efef157e61fe4d773d8a3b0bad5feccec</td><td></td><td></td><td></td></tr><tr><td>763158cef9d1e4041f24fce4cf9d6a3b7a7f08ff</td><td>Hierarchical Modeling and +<br/>Applications to Recognition Tasks +<br/>Thesis submitted for the degree of +<br/>”Doctor of Philosophy” +<br/>by +<br/><b>Submitted to the Senate of the Hebrew University</b><br/>August / 2013 +</td><td>('39161025', 'Alon Zweig', 'alon zweig')</td><td></td></tr><tr><td>764882e6779fbee29c3d87e00302befc52d2ea8d</td><td>Deep Approximately Orthogonal Nonnegative +<br/>Matrix Factorization for Clustering +<br/>School of Automation +<br/>School of Automation +<br/>School of Automation +<br/><b>Guangdong University of Technology</b><br/><b>Guangdong University of Technology</b><br/><b>Guangdong University of Technology</b><br/>Guangzhou, China +<br/>Guangzhou, China +<br/>Guangzhou, China +</td><td>('30185240', 'Yuning Qiu', 'yuning qiu')<br/>('1764724', 'Guoxu Zhou', 'guoxu zhou')<br/>('2454506', 'Kan Xie', 'kan xie')</td><td>yn.qiu@foxmail.com +<br/>guoxu.zhou@qq.com +<br/>kanxiegdut@gmail.com +</td></tr><tr><td>76d939f73a327bf1087d91daa6a7824681d76ea1</td><td>A Thermal Facial Emotion Database +<br/>and Its Analysis +<br/><b>Japan Advanced Institute of Science and Technology</b><br/>1-1 Asahidai, Nomi, Ishikawa, Japan +<br/><b>University of Science, Ho Chi Minh city</b><br/>227 Nguyen Van Cu, Ho Chi Minh city, Vietnam +</td><td>('2319415', 'Hung Nguyen', 'hung nguyen')<br/>('1791753', 'Kazunori Kotani', 'kazunori kotani')<br/>('1753878', 'Fan Chen', 'fan chen')</td><td>{nvhung,ikko,chen-fan}@jaist.ac.jp +<br/>lhbac@hcmuns.edu.vn +</td></tr><tr><td>760ba44792a383acd9ca8bef45765d11c55b48d4</td><td>~ +<br/>I . +<br/>INTRODUCTION AND BACKGROUND +<br/>The purpose of this article is to introduce the +<br/>reader to the basic principles of classification with +<br/>class-specific features. It is written both for readers +<br/>interested in only the basic concepts as well as those +<br/>interested in getting started in applying the method. +<br/>For in-depth coverage, the reader is referred to a more +<br/>detailed article [l]. +<br/>Class-Specific Classifier: +<br/>Avoiding the Curse of +<br/>Dimensionality +<br/>PAUL M. BAGGENSTOSS, Member. lEEE +<br/>US. Naval Undersea Warfare Center +<br/>This article describes a new probabilistic method called the +<br/>“class-specific method” (CSM). CSM has the potential to avoid +<br/>the “curse of dimensionality” which plagues most clmiiiers +<br/>which attempt to determine the decision boundaries in a +<br/>highdimensional featue space. In contrast, in CSM, it is possible +<br/>to build classifiers without a ” n o n feature space. Separate +<br/>Law-dimensional features seta may be de6ned for each class, while +<br/>the decision funetions are projected back to the common raw data +<br/>space. CSM eflectively extends the classical classification theory +<br/>to handle multiple feature spaw.. It is completely general, and +<br/>requires no s i m p l i n g assumption such as Gaussianity or that +<br/>data lies in linear subspaces. +<br/>Manuscript received September 26, 2W2; revised February 12, +<br/>2003. +<br/>This work was supported by the Office of Naval Research. +<br/>Author’s address: US. Naval Undersea Warfare Center, Newport +<br/>Classification is the process of assigning data +<br/>to one of a set of pre-determined class labels [2]. +<br/>Classification is a fundamental problem that has +<br/>to be solved if machines are to approximate the +<br/>human functions of recognizing sounds, images, or +<br/>other sensory inputs. This is why classification is so +<br/>important for automation in today’s commercial and +<br/>military arenas. +<br/>Many of us have first-hand knowledge of +<br/>successful automated recognition systems from +<br/>cameras that recognize faces in airports to computers +<br/>that can scan and read printed and handwritten text, +<br/>or systems that can recognize human speech. These +<br/>systems are becoming more and more reliable and +<br/>accurate. Given reasonably clean input data, the +<br/>performance is often quite good if not perfect. But +<br/>many of these systems fail in applications where +<br/>clean, uncorrupted data is not available or if the +<br/>problem is complicated by variability of conditions +<br/>or by proliferation of inputs from unknown sources. +<br/>In military environments, the targets to he recognized +<br/>are often uncooperative and hidden in clutter and +<br/>interference. In short, military uses of such systems +<br/>still fall far short of what a well-trained alert human +<br/>operator can achieve. +<br/>We are often perplexed by the wide gap of +<br/>as a car door slamming. From +<br/>performance between humans and automated systems. +<br/>Allow a human listener to hear two or three examples +<br/>of a sound-such +<br/>these few examples, the human can recognize +<br/>the sound again and not confuse it with similar +<br/>interfering sounds. But try the same experiment with +<br/>general-purpose classifiers using neural networks +<br/>and the story is quite different. Depending on the +<br/>problem, the automated system may require hundreds, +<br/>thousands, even millions of examples for training +<br/>before it becomes both robust and reliable. +<br/>Why? The answer lies in what is known as the +<br/>“curse of dimensionality.” General-purpose classifiers +<br/>need to extract a large number of measurements, +<br/>or features, from the data to account for all the +<br/>different possibilities of data types. The large +<br/>collection of features form a high-dimensional space +<br/>that the classifier has to sub-divide into decision +<br/>boundaries. It is well-known that the complexity of +<br/>a high-dimensional space increases exponentially +<br/>with the number of measurements [31-and +<br/>so does +<br/>the difficulty of finding the hest decision boundaries +<br/>from a fixed amount of training data. Unless a lot +<br/>EEE A&E SYSTEMS MAGAZINE VOL. 19, NO. 1 JANUARY 2004 PART 2: TUTORIALS-BAGGENSTOSS +<br/>37 +</td><td></td><td>RI, 02841, E-mail: (p.m.baggenstoss@ieee.arg). +</td></tr><tr><td>766728bac030b169fcbc2fbafe24c6e22a58ef3c</td><td>A survey of deep facial landmark detection +<br/>Yongzhe Yan1,2 +<br/>Thierry Chateau1 +<br/>1 Université Clermont Auvergne, France +<br/>2 Wisimage, France +<br/>3 Université de Lyon, CNRS, INSA Lyon, LIRIS, UMR5205, Lyon, France +<br/>Résumé +<br/>La détection de landmarks joue un rôle crucial dans de +<br/>nombreuses applications d’analyse du visage comme la +<br/>reconnaissance de l’identité, des expressions, l’animation +<br/>d’avatar, la reconstruction 3D du visage, ainsi que pour +<br/>les applications de réalité augmentée comme la pose de +<br/>masque ou de maquillage virtuel. L’avènement de l’ap- +<br/>prentissage profond a permis des progrès très importants +<br/>dans ce domaine, y compris sur les corpus non contraints +<br/>(in-the-wild). Nous présentons ici un état de l’art cen- +<br/>tré sur la détection 2D dans une image fixe, et les mé- +<br/>thodes spécifiques pour la vidéo. Nous présentons ensuite +<br/>les corpus existants pour ces trois tâches, ainsi que les mé- +<br/>triques d’évaluations associées. Nous exposons finalement +<br/>quelques résultats, ainsi que quelques pistes de recherche. +<br/>Mots Clef +<br/>Détection de landmark facial, Alignement de visage, Deep +<br/>learning +</td><td>('3015472', 'Xavier Naturel', 'xavier naturel')<br/>('50493659', 'Christophe Garcia', 'christophe garcia')<br/>('48601809', 'Christophe Blanc', 'christophe blanc')<br/>('1762557', 'Stefan Duffner', 'stefan duffner')</td><td>yongzhe.yan@etu.uca.fr +</td></tr><tr><td>7697295ee6fc817296bed816ac5cae97644c2d5b</td><td>Detecting and Recognizing Human-Object Interactions +<br/>Facebook AI Research (FAIR) +</td><td>('2082991', 'Georgia Gkioxari', 'georgia gkioxari')<br/>('39353098', 'Kaiming He', 'kaiming he')</td><td></td></tr><tr><td>7636f94ddce79f3dea375c56fbdaaa0f4d9854aa</td><td>Appl. Math. Inf. Sci. 6 No. 2S pp. 403S-408S (2012) +<br/> An International Journal +<br/>© 2012 NSP +<br/>Applied Mathematics & Information Sciences +<br/>Robust Facial Expression Recognition Using +<br/>a Smartphone Working against Illumination Variation +<br/> Natural Sciences Publishing Cor. +<br/><b>Sejong University, 98 Kunja-Dong, Kwangjin-Gu, Seoul, Korea</b><br/>Received June 22, 2010; Revised March 21, 2011; Accepted 11 June 2011 +<br/>Published online: 1 January 2012 +</td><td>('2413560', 'Kyoung-Sic Cho', 'kyoung-sic cho')<br/>('9270794', 'In-Ho Choi', 'in-ho choi')<br/>('2706430', 'Yong-Guk Kim', 'yong-guk kim')</td><td> @ 2012 NSP +<br/>Corresponding author: Email: ykim@sejong.ac.kr +</td></tr><tr><td>1c80bc91c74d4984e6422e7b0856cf3cf28df1fb</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Hierarchical Adaptive Structural SVM for Domain Adaptation +<br/>Received: date / Accepted: date +</td><td>('2470198', 'Jiaolong Xu', 'jiaolong xu')</td><td></td></tr><tr><td>1ce3a91214c94ed05f15343490981ec7cc810016</td><td>Exploring Photobios +<br/><b>University of Washington</b><br/>2Adobe Systems† +<br/>3Google Inc. +</td><td>('2419955', 'Ira Kemelmacher-Shlizerman', 'ira kemelmacher-shlizerman')<br/>('2177801', 'Eli Shechtman', 'eli shechtman')<br/>('9748713', 'Rahul Garg', 'rahul garg')<br/>('1679223', 'Steven M. Seitz', 'steven m. seitz')</td><td></td></tr><tr><td>1c9efb6c895917174ac6ccc3bae191152f90c625</td><td>Unifying Identification and Context Learning for Person Recognition +<br/><b>CUHK-SenseTime Joint Lab, The Chinese University of Hong Kong</b></td><td>('39360892', 'Qingqiu Huang', 'qingqiu huang')<br/>('50446092', 'Yu Xiong', 'yu xiong')<br/>('1807606', 'Dahua Lin', 'dahua lin')</td><td>{hq016, xy017, dhlin}@ie.cuhk.edu.hk +</td></tr><tr><td>1c2724243b27a18a2302f12dea79d9a1d4460e35</td><td>Fisher+Kernel Criterion for Discriminant Analysis* +<br/><b>National Laboratory on Machine Perception, Peking University, Beijing, P.R. China</b><br/><b>the Chinese University of Hong Kong, Shatin, Hong Kong</b><br/>3 MOE-Microsoft Key Laboratory of Multimedia Computing and Communication & Department of EEIS, +<br/><b>University of Science and Technology of China, Hefei, Anhui, P. R. China</b><br/>4Microsoft Research Asia, Beijing, P.R. China +<br/> +</td><td>('1718245', 'Shu Yang', 'shu yang')<br/>('1698982', 'Shuicheng Yan', 'shuicheng yan')<br/>('38188040', 'Dong Xu', 'dong xu')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')<br/>('1720735', 'Chao Zhang', 'chao zhang')</td><td>Contact: yangshu@cis.pku.edu.cn +</td></tr><tr><td>1ca8c09abb73a02519d8db77e4fe107acfc589b6</td><td>Automatic Understanding of Image and Video Advertisements +<br/><b>University of Pittsburgh</b><br/>IEEE 2017 Conference on +<br/>Computer Vision and Pattern +<br/>Recognition +<br/>Introduction +<br/>Dataset Overview +<br/>Answering Questions about Ads +<br/>• Advertisements implicitly persuade viewers to take certain actions. +<br/>• Understanding ads requires more than recognizing physical content. +<br/>Recognized Concepts (Clarifai): +<br/>Car, Street, Transportation System, Traffic, Road, City, +<br/>Pavement, Crossing, … +<br/>Image Caption (Vinyals et al.): +<br/>A red car driving down a street next to a traffic light. +<br/>True Meaning in Advertisement: +<br/>Automobile drivers should be cautious to avoid crashing +<br/>into cyclists as they share the road. +<br/>• We propose the novel problem of automatic advertisement +<br/>understanding, and provide two datasets with rich annotations. +<br/>• We analyze the common persuasive strategies: symbolism, atypical +<br/>objects, physical processes, cultural knowledge, surprise/shock, etc. +<br/>• We present baseline experiment results for several prediction tasks. +<br/>Dataset Collection +<br/>• 38 topics including commercials and public service announcements +<br/>• 30 sentiments indicating how ads emotionally impress viewers +<br/>• Questions and answers revealing the messages behind the visual ads +<br/>I should stop smoking because my +<br/>lungs are extremely sensitive and +<br/>could go up in smoke. +<br/>I should buy this candy because it +<br/>is unique and rises above the rest, +<br/>like the Swiss Alps. +<br/>• Our dataset contains 64,832 image ads and 3,477 video ads, each +<br/>annotated by 3-5 human workers from Amazon Mechanical Turk. +<br/>Symbolism Detection +<br/>Image +<br/>Video +<br/>Topic +<br/>Symbol +<br/>Topic +<br/>Fun/Exciting +<br/>204,340 +<br/>64,131 +<br/>17,345 +<br/>17,374 +<br/>Sentiment +<br/>Strategy +<br/>Sentiment +<br/>English? +<br/>102,340 +<br/>20,000 +<br/>17,345 +<br/>15,380 +<br/>Q + A Pairs +<br/>Slogan +<br/>Q + A Pairs +<br/>Effectiveness +<br/>202,090 +<br/>11,130 +<br/>17,345 +<br/>16,721 +</td><td>('1996796', 'Zaeem Hussain', 'zaeem hussain')<br/>('2365530', 'Mingda Zhang', 'mingda zhang')<br/>('3186356', 'Xiaozhong Zhang', 'xiaozhong zhang')<br/>('9085797', 'Keren Ye', 'keren ye')<br/>('40540691', 'Christopher Thomas', 'christopher thomas')<br/>('6004292', 'Zuha Agha', 'zuha agha')<br/>('34493995', 'Nathan Ong', 'nathan ong')<br/>('1770205', 'Adriana Kovashka', 'adriana kovashka')</td><td></td></tr><tr><td>1cfe3533759bf95be1fce8ce1d1aa2aeb5bfb4cc</td><td>Recognition of Facial Gestures based on Support +<br/>Vector Machines +<br/><b>Faculty of Informatics, University of Debrecen, Hungary</b><br/>H-4010 Debrecen P.O.Box 12. +</td><td>('47547897', 'Attila Fazekas', 'attila fazekas')</td><td>Attila.Fazekas@inf.unideb.hu +</td></tr><tr><td>1ce4587e27e2cf8ba5947d3be7a37b4d1317fbee</td><td>Deep fusion of visual signatures +<br/>for client-server facial analysis +<br/>Normandie Univ, UNICAEN, +<br/>ENSICAEN, CNRS, GREYC +<br/>Computer Sc. & Engg. +<br/>IIT Kanpur, India +<br/>Frederic Jurie +<br/>Normandie Univ, UNICAEN, +<br/>ENSICAEN, CNRS, GREYC +<br/>Facial analysis is a key technology for enabling human- +<br/>machine interaction. +<br/>In this context, we present a client- +<br/>server framework, where a client transmits the signature of +<br/>a face to be analyzed to the server, and, in return, the server +<br/>sends back various information describing the face e.g. is the +<br/>person male or female, is she/he bald, does he have a mus- +<br/>tache, etc. We assume that a client can compute one (or a +<br/>combination) of visual features; from very simple and effi- +<br/>cient features, like Local Binary Patterns, to more complex +<br/>and computationally heavy, like Fisher Vectors and CNN +<br/>based, depending on the computing resources available. The +<br/>challenge addressed in this paper is to design a common uni- +<br/>versal representation such that a single merged signature is +<br/>transmitted to the server, whatever be the type and num- +<br/>ber of features computed by the client, ensuring nonetheless +<br/>an optimal performance. Our solution is based on learn- +<br/>ing of a common optimal subspace for aligning the different +<br/>face features and merging them into a universal signature. +<br/>We have validated the proposed method on the challenging +<br/>CelebA dataset, on which our method outperforms existing +<br/>state-of-art methods when rich representation is available at +<br/>test time, while giving competitive performance when only +<br/>simple signatures (like LBP) are available at test time due +<br/>to resource constraints on the client. +<br/>1. +<br/>INTRODUCTION +<br/>We propose a novel method in a heterogeneous server- +<br/>client framework for the challenging and important task of +<br/>analyzing images of faces. Facial analysis is a key ingredient +<br/>for assistive computer vision and human-machine interaction +<br/>methods, and systems and incorporating high-performing +<br/>methods in daily life devices is a challenging task. The ob- +<br/>jective of the present paper is to develop state-of-the-art +<br/>technologies for recognizing facial expressions and facial at- +<br/>tributes on mobile and low cost devices. Depending on their +<br/>computing resources, the clients (i.e. the devices on which +<br/>the face image is taken) are capable of computing different +<br/>types of face signatures, from the simplest ones (e.g. LPB) +<br/>to the most complex ones (e.g. very deep CNN features), and +<br/>should be able to eventually combine them into a single rich +<br/>signature. Moreover, it is convenient if the face analyzer, +<br/>which might require significant computing resources, is im- +<br/>plemented on a server receiving face signatures and comput- +<br/>ing facial expressions and attributes from these signatures. +<br/>Keeping the computation of the signatures on the client is +<br/>safer in terms of privacy, as the original images are not trans- +<br/>mitted, and keeping the analysis part on the server is also +<br/>beneficial for easy model upgrades in the future. To limit +<br/>the transmission costs, the signatures have to be made as +<br/>compact as possible. +<br/>In summary, the technology needed +<br/>for this scenario has to be able to merge the different avail- +<br/>able features – the number of features available at test time +<br/>is not known in advance but is dependent on the computing +<br/>resources available on the client – producing a unique rich +<br/>and compact signature of the face, which can be transmitted +<br/>and analyzed by a server. Ideally, we would like the univer- +<br/>sal signature to have the following properties: when all the +<br/>features are available, we would like the performance of the +<br/>signature to be better than the one of a system specifically +<br/>optimized for any single type of feature. +<br/>In addition, we +<br/>would like to have reasonable performance when only one +<br/>type of feature is available at test time. +<br/>For developing such a system, we propose a hybrid deep +<br/>neural network and give a method to carefully fine-tune the +<br/>network parameters while learning with all or a subset of +<br/>features available. Thus, the proposed network can process a +<br/>number of wide ranges of feature types such as hand-crafted +<br/>LBP and FV, or even CNN features which are learned end- +<br/>to-end. +<br/>While CNNs have been quite successful in computer vi- +<br/>sion [1], representing images with CNN features is relatively +<br/>time consuming, much more than some simple hand-crafted +<br/>features such as LBP. Thus, the use of CNN in real-time ap- +<br/>plications is still not feasible. In addition, the use of robust +<br/>hand-crafted features such as FV in hybrid architectures can +<br/>give performance comparable to Deep CNN features [2]. The +<br/>main advantage of learning hybrid architectures is to avoid +<br/>having large numbers of convolutional and pooling layers. +<br/>Again from [2], we can also observe that hybrid architec- +<br/>tures improve the performance of hand-crafted features e.g. +<br/>FVs. Therefore, hybrid architectures are useful for the cases +<br/>where only hand-crafted features, and not the original im- +<br/>ages, are available during training and testing time. This +<br/>scenario is useful when it is not possible to share training +<br/>images due to copyright or privacy issues. +<br/>Hybrid networks are particularly adapted to our client- +</td><td>('2078892', 'Binod Bhattarai', 'binod bhattarai')<br/>('2515597', 'Gaurav Sharma', 'gaurav sharma')</td><td>binod.bhattarai@unicaen.fr +<br/>grv@cse.iitk.ac.in +<br/>frederic.jurie@unicaen.fr +</td></tr><tr><td>1c30bb689a40a895bd089e55e0cad746e343d1e2</td><td>Learning Spatiotemporal Features with 3D Convolutional Networks +<br/><b>Facebook AI Research, 2Dartmouth College</b></td><td>('1687325', 'Du Tran', 'du tran')<br/>('2276554', 'Rob Fergus', 'rob fergus')<br/>('1732879', 'Lorenzo Torresani', 'lorenzo torresani')<br/>('2210374', 'Manohar Paluri', 'manohar paluri')</td><td>{dutran,lorenzo}@cs.dartmouth.edu +<br/>{lubomir,robfergus,mano}@fb.com +</td></tr><tr><td>1c4ceae745fe812d8251fda7aad03210448ae25e</td><td>EURASIP Journal on Applied Signal Processing 2004:4, 522–529 +<br/>c(cid:1) 2004 Hindawi Publishing Corporation +<br/>Optimization of Color Conversion for Face Recognition +<br/><b>Virginia Polytechnic Institute and State University</b><br/>Blacksburg, VA 24061-0111, USA +<br/><b>Seattle Paci c University, Seattle, WA 98119-1957, USA</b><br/><b>Virginia Polytechnic Institute and State University</b><br/>Blacksburg, VA 24061-0111, USA +<br/>Received 5 November 2002; Revised 16 October 2003 +<br/>This paper concerns the conversion of color images to monochromatic form for the purpose of human face recognition. Many +<br/>face recognition systems operate using monochromatic information alone even when color images are available. In such cases, +<br/>simple color transformations are commonly used that are not optimal for the face recognition task. We present a framework +<br/>for selecting the transformation from face imagery using one of three methods: Karhunen-Lo`eve analysis, linear regression of +<br/>color distribution, and a genetic algorithm. Experimental results are presented for both the well-known eigenface method and for +<br/>extraction of Gabor-based face features to demonstrate the potential for improved overall system performance. Using a database +<br/>of 280 images, our experiments using these methods resulted in performance improvements of approximately 4% to 14%. +<br/>Keywords and phrases: face recognition, color image analysis, color conversion, Karhunen-Lo`eve analysis. +<br/>1. +<br/>INTRODUCTION +<br/>Most single-view face recognition systems operate using in- +<br/>tensity (monochromatic) information alone. This is true +<br/>even for systems that accept color imagery as input. The +<br/>reason for this is not +<br/>that multispectral data is lack- +<br/>ing in information content, but often because of practical +<br/>considerations—difficulties associated with illumination and +<br/>color balancing, for example, as well as compatibility with +<br/>legacy systems. Associated with this is a lack of color image +<br/>databases with which to develop and test new algorithms. Al- +<br/>though work is in progress that will eventually aid in color- +<br/>based tasks (e.g., through color constancy [1]), those efforts +<br/>are still in the research stage. +<br/>When color information is present, most of today’s face +<br/>recognition systems convert the image to monochromatic +<br/>form using simple transformations. For example, a common +<br/>mapping [2, 3] produces an intensity value Ii by taking the +<br/>average of red, green, and blue (RGB) values (Ir, Ig, and Ib, +<br/>resp.): +<br/>Ii(x, y) = Ir(x, y) + Ig(x, y) + Ib(x, y) +<br/>(1) +<br/>The resulting image is then used for feature extraction and +<br/>analysis. +<br/>We argue that more effective system performance is pos- +<br/>sible if a color transformation is chosen that better matches +<br/>the task at hand. For example, the mapping in (1) implic- +<br/>itly assumes a uniform distribution of color values over the +<br/>entire color space. For a task such as face recognition, color +<br/>values tend to be more tightly confined to a small portion of +<br/>the color space, and it is possible to exploit this narrow con- +<br/>centration during color conversion. If the transformation is +<br/>selected based on the expected color distribution, then it is +<br/>reasonable to expect improved recognition accuracies. +<br/>This paper presents a task-oriented approach for select- +<br/>ing the color-to-grayscale image transformation. Our in- +<br/>tended application is face recognition, although the frame- +<br/>work that we present is applicable to other problem domains. +<br/>We assume that frontal color views of the human face +<br/>are available, and we develop a method for selecting alter- +<br/>nate weightings of the separate color values in computing a +<br/>single monochromatic value. Given the rich color content +<br/>of the human face, it is desirable to maximize the use of +<br/>this content even when full-color computation and match- +<br/>ing is not used. As an illustration of this framework, we +<br/>have used the Karhunen-Lo`eve (KL) transformation (also +<br/>known as principal components analysis) of observed distri- +<br/>butions in the color space to determine the improved map- +<br/>ping. +</td><td>('1719681', 'Creed F. Jones', 'creed f. jones')<br/>('1731164', 'A. Lynn Abbott', 'a. lynn abbott')</td><td>Email: crjones4@vt.edu +<br/>Email: abbott@vt.edu +</td></tr><tr><td>1c3073b57000f9b6dbf1c5681c52d17c55d60fd7</td><td>THÈSEprésentéepourl’obtentiondutitredeDOCTEURDEL’ÉCOLENATIONALEDESPONTSETCHAUSSÉESSpécialité:InformatiqueparCharlotteGHYSAnalyse,Reconstruction3D,&AnimationduVisageAnalysis,3DReconstruction,&AnimationofFacesSoutenancele19mai2010devantlejurycomposéde:Rapporteurs:MajaPANTICDimitrisSAMARASExaminateurs:MichelBARLAUDRenaudKERIVENDirectiondethèse:NikosPARAGIOSBénédicteBASCLE </td><td></td><td></td></tr><tr><td>1cee993dc42626caf5dbc26c0a7790ca6571d01a</td><td>Optimal Illumination for Image and Video Relighting +<br/>Shree K.Nayar +<br/>Peter N.Belhumeur +<br/><b>Columbia University</b><br/>It has been shown in the literature that image-based relighting of +<br/>scenes with unknown geometry can be achieved through linear +<br/>combinations of a set of pre-acquired reference images. Since the +<br/>placement and brightness of the light sources can be controlled, it +<br/>is natural to ask: what is the optimal way to illuminate the scene to +<br/>reduce the number of reference images that are needed? +<br/>In this work we show that the best way to light the scene (i.e., the +<br/>way that minimizes the number of reference images) is not using +<br/>a sequence of single, compact light sources as is most commonly +<br/>done, but rather to use a sequence of lighting patterns as given by an +<br/>object-dependent lighting basis. While this lighting basis, which we +<br/>call the optimal lighting basis (OLB), depends on camera and scene +<br/>properties, we show that it can be determined as a simple calibration +<br/>procedure before acquisition, through the SVD decomposition of +<br/>the images of the object lighted by single light sources (Fig. 1). +<br/>of basis images used, and for a set of four experiments (relighting +<br/>of a sphere, a face, a buddha statue, and a dragon). For any given +<br/>number of optimal lighting basis images, the corresponding num- +<br/>ber of images of any other lighting basis that are needed to achieve +<br/>the same reconstruction error equals the gain value. For instance, in +<br/>the ‘buddha’ experiment instead of 6 optimal basis images, we will +<br/>need to use 6× 1.8 ≈ 11 SHLB images, 6× 1.5 ≈ 9 FLB images or +<br/>6× 2.3 ≈ 14 HaLB images. +<br/>Figure 1: Computing the optimal lighting basis using SVD. First row: Images of the +<br/>object illuminated by a single light source in different positions. Second row: Lighting +<br/>patterns from the optimal lighting basis, containing both positive values, shown in +<br/>grey, and negative values, shown in blue. Third row: Offset and scaling of the optimal +<br/>lighting basis in order to make all its values positive. +<br/>We demonstrate with experiments on real and synthetic data that +<br/>the optimal lighting basis significantly reduces the number of refer- +<br/>ence images that are needed to achieve a desired level of accuracy +<br/>in the relit images. In particular, we show that the scene-dependent +<br/>optimal lighting basis (OBL) performs much better than the Fourier +<br/>lighting basis (FLB), Haar lighting basis (HaLB) and spherical har- +<br/>monic lighting basis (SHLB). +<br/>In Fig. 2 we show some reconstructed images of synthetic objects +<br/>which have been illuminated by SHLB and OLB. Observe how +<br/>when we reconstruct from images illuminated by OLB, the error is +<br/>significantly smaller. In Fig. 3 we plot the gains of the optimal light- +<br/>ing basis with respect the other basis, as a function of the number +<br/>Figure 3: Gains of the OLB with respect all the other lighting basis, (for a set of 4 +<br/>experiments), plotted as a function of the number of basis images used. +<br/>This reduction in the number of needed images is particularly criti- +<br/>cal in the problem of relighting in video, as corresponding points on +<br/>moving objects must be aligned from frame to frame during each +<br/>cycle of the lighting basis. We show, however, that the efficiencies +<br/>gained by the optimal lighting basis makes relighting in video pos- +<br/>sible using only a simple optical flow alignment. Furthermore, in +<br/>our experiments we verify that although the optimal lighting basis +<br/>is computed for an initial orientation of the object, the reconstruc- +<br/>tion error does not increase noticeably as the object changes its pose +<br/>along the video sequence. +<br/>We have performed several relighting experiments on real video se- +<br/>quences of moving objects, moving faces, and scenes containing +<br/>both. In each case, although a single video clip was captured, we +<br/>are able to relight again and again, controlling the lighting direc- +<br/>tion, extent, and color. Fig. 4 shows some frames of one of these +<br/>sequences. +<br/>Ground Truth +<br/>FLB 16 basis OLB 16 basis +<br/>Error FLB +<br/>Error OLB +<br/>SHLB 3 basis OLB 3 basis +<br/>Ground Truth +<br/>Figure 2: Examples of reconstructed images and reconstruction errors, for different +<br/>lighting basis. Note that OLB performs much better. +<br/>Error SHLB +<br/>Error OLB +<br/>Figure 4: Two frames of a video sequence, illuminated with the optimal lighting +<br/>basis (first row), and relighted with a point light source (second row) and with an +<br/>environmental light (third row). +</td><td>('1994318', 'Francesc Moreno-Noguer', 'francesc moreno-noguer')</td><td></td></tr><tr><td>1c147261f5ab1b8ee0a54021a3168fa191096df8</td><td>Journal of Information Security, 2016, 7, 141-151 +<br/>Published Online April 2016 in SciRes. http://www.scirp.org/journal/jis +<br/>http://dx.doi.org/10.4236/jis.2016.73010 +<br/>Face Recognition across Time Lapse Using +<br/>Convolutional Neural Networks +<br/><b>George Mason University, Fairfax, VA, USA</b><br/>Received 12 February 2016; accepted 8 April 2016; published 11 April 2016 +<br/>Copyright © 2016 by authors and Scientific Research Publishing Inc. +<br/>This work is licensed under the Creative Commons Attribution International License (CC BY). +<br/>http://creativecommons.org/licenses/by/4.0/ +<br/> +<br/> +</td><td>('2710867', 'Hachim El Khiyari', 'hachim el khiyari')<br/>('1781577', 'Harry Wechsler', 'harry wechsler')</td><td></td></tr><tr><td>1c17450c4d616e1e1eece248c42eba4f87de9e0d</td><td>YANG, LIN, CHANG, CHEN: AUTOMATIC AGE ESTIMATION VIA DEEP RANKING +<br/>Automatic Age Estimation from Face Images +<br/>via Deep Ranking +<br/><b>Research Center for Information</b><br/>Technology Innovation +<br/>Academia Sinica +<br/>Taipei, Taiwan +<br/><b>Institute of Information Science</b><br/>Academia Sinica +<br/>Taipei, Taiwan +</td><td>('35436145', 'Huei-Fang Yang', 'huei-fang yang')<br/>('36181124', 'Bo-Yao Lin', 'bo-yao lin')<br/>('34692779', 'Kuang-Yu Chang', 'kuang-yu chang')<br/>('1720473', 'Chu-Song Chen', 'chu-song chen')</td><td>hfyang@citi.sinica.edu.tw +<br/>boyaolin@iis.sinica.edu.tw +<br/>kuangyu@iis.sinica.edu.tw +<br/>song@iis.sinica.edu.tw +</td></tr><tr><td>1c93b48abdd3ef1021599095a1a5ab5e0e020dd5</td><td>JOURNAL OF LATEX CLASS FILES, VOL. *, NO. *, JANUARY 2009 +<br/>A Compositional and Dynamic Model for Face Aging +</td><td>('3133970', 'Song-Chun Zhu', 'song-chun zhu')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td></td></tr><tr><td>1c41965c5e1f97b1504c1bdde8037b5e0417da5e</td><td>Interaction-aware Spatio-temporal Pyramid +<br/>Attention Networks for Action Classification +<br/><b>University of Chinese Academy of Sciences</b><br/>2 CAS Center for Excellence in Brain Science and Intelligence Technology, National +<br/><b>Laboratory of Pattern Recognition, Institute of Automation, CAS</b><br/>3 Meitu, 4 National Computer network Emergency Response technical +<br/>Team/Coordination Center of China +</td><td>('1807325', 'Yang Du', 'yang du')<br/>('2034987', 'Chunfeng Yuan', 'chunfeng yuan')<br/>('46708348', 'Bing Li', 'bing li')<br/>('40027215', 'Lili Zhao', 'lili zhao')<br/>('2082374', 'Yangxi Li', 'yangxi li')<br/>('40506509', 'Weiming Hu', 'weiming hu')</td><td>duyang2014@ia.ac.cn,{cfyuan,bli,wmhu}@nlpr.ia.ac.cn, +<br/>lili.zhao@meitu.com, liyangxi@outlook.com +</td></tr><tr><td>1cbd3f96524ca2258fd2d5c504c7ea8da7fb1d16</td><td>Fusion of audio-visual features using hierarchical classifier systems for +<br/>the recognition of affective states and the state of depression +<br/><b>Institute of Neural Information Processing, Ulm University, Ulm, Germany</b><br/>Keywords: +<br/>Emotion Recognition, Multiple Classifier Systems, Affective Computing, Information Fusion +</td><td>('1860319', 'Michael Glodek', 'michael glodek')<br/>('3243891', 'Sascha Meudt', 'sascha meudt')<br/>('1685857', 'Friedhelm Schwenker', 'friedhelm schwenker')</td><td>firstname.lastname@uni-ulm.de +</td></tr><tr><td>1cad5d682393ffbb00fd26231532d36132582bb4</td><td>Spatio-Temporal Action Detection with +<br/>Cascade Proposal and Location Anticipation +<br/><b>Institute for Robotics and Intelligent</b><br/>Systems +<br/><b>University of Southern California</b><br/>Los Angeles, CA, USA +</td><td>('3469030', 'Zhenheng Yang', 'zhenheng yang')<br/>('3029956', 'Jiyang Gao', 'jiyang gao')<br/>('27735100', 'Ram Nevatia', 'ram nevatia')<br/>('3469030', 'Zhenheng Yang', 'zhenheng yang')<br/>('3029956', 'Jiyang Gao', 'jiyang gao')<br/>('27735100', 'Ram Nevatia', 'ram nevatia')</td><td>zhenheny@usc.edu +<br/>jiyangga@usc.edu +<br/>nevatia@usc.edu +</td></tr><tr><td>1c1a98df3d0d5e2034ea723994bdc85af45934db</td><td>Guided Unsupervised Learning of Mode Specific Models for Facial Point +<br/>Detection in the Wild +<br/><b>School of Computer Science, The University of Nottingham</b></td><td>('2736086', 'Shashank Jaiswal', 'shashank jaiswal')<br/>('2449665', 'Timur R. Almaev', 'timur r. almaev')</td><td>{psxsj3,psxta4,michel.valstar}@nottingham.ac.uk +</td></tr><tr><td>1ca815327e62c70f4ee619a836e05183ef629567</td><td>Global Supervised Descent Method +<br/><b>Carnegie Mellon University, Pittsburgh PA</b></td><td>('3182065', 'Xuehan Xiong', 'xuehan xiong')<br/>('1707876', 'Fernando De la Torre', 'fernando de la torre')</td><td>{xxiong,ftorre}@andrew.cmu.edu +</td></tr><tr><td>1c6be6874e150898d9db984dd546e9e85c85724e</td><td></td><td></td><td></td></tr><tr><td>1c65f3b3c70e1ea89114f955624d7adab620a013</td><td></td><td></td><td></td></tr><tr><td>1c530de1a94ac70bf9086e39af1712ea8d2d2781</td><td>Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-16) +<br/>Sparsity Conditional Energy Label +<br/>Distribution Learning for Age Estimation +<br/>Key Lab of Computer Network and Information Integration (Ministry of Education) +<br/><b>School of Computer Science and Engineering, Southeast University, Nanjing 211189, China</b></td><td>('2442058', 'Xu Yang', 'xu yang')<br/>('1735299', 'Xin Geng', 'xin geng')<br/>('1725992', 'Deyu Zhou', 'deyu zhou')</td><td>{x.yang,xgeng,d.zhou}@seu.edu.cn +</td></tr><tr><td>1c6e22516ceb5c97c3caf07a9bd5df357988ceda</td><td></td><td></td><td></td></tr><tr><td>82f8652c2059187b944ce65e87bacb6b765521f6</td><td>Discriminative Object Categorization with +<br/>External Semantic Knowledge +<br/>Dissertation Proposal +<br/>by +<br/>Department of Computer Science +<br/><b>University of Texas at Austin</b><br/>Committee: +<br/>Prof. Kristen Grauman (Advisor) +<br/>Prof. Fei Sha +<br/>Prof. J. K. Aggarwal +</td><td>('35788904', 'Sung Ju Hwang', 'sung ju hwang')<br/>('1797655', 'Raymond Mooney', 'raymond mooney')<br/>('2302443', 'Pradeep Ravikumar', 'pradeep ravikumar')</td><td></td></tr><tr><td>82bef8481207de9970c4dc8b1d0e17dced706352</td><td></td><td></td><td></td></tr><tr><td>825f56ff489cdd3bcc41e76426d0070754eab1a8</td><td>Making Convolutional Networks Recurrent for Visual Sequence Learning +<br/>NVIDIA +</td><td>('40058797', 'Xiaodong Yang', 'xiaodong yang')</td><td>{xiaodongy,pmolchanov,jkautz}@nvidia.com +</td></tr><tr><td>82d2af2ffa106160a183371946e466021876870d</td><td>A Novel Space-Time Representation on the Positive Semidefinite Cone +<br/>for Facial Expression Recognition +<br/>1IMT Lille Douai, Univ. Lille, CNRS, UMR 9189 – CRIStAL – +<br/>Centre de Recherche en Informatique Signal et Automatique de Lille, F-59000 Lille, France +<br/>2Univ. Lille, CNRS, UMR 8524, Laboratoire Paul Painlev´e, F-59000 Lille, France. +</td><td>('37809060', 'Anis Kacem', 'anis kacem')<br/>('2909056', 'Mohamed Daoudi', 'mohamed daoudi')<br/>('2125606', 'Boulbaba Ben Amor', 'boulbaba ben amor')</td><td></td></tr><tr><td>824d1db06e1c25f7681e46199fd02cb5fc343784</td><td>Representing Relative Visual Attributes +<br/>with a Reference-Point-Based Decision Model +<br/>Marc T. Law +<br/><b>University of Toronto</b><br/><b>Shanghai Jiao Tong University</b><br/><b>University of Michigan-Shanghai Jiao Tong University Joint Institute</b></td><td>('38481975', 'Paul Weng', 'paul weng')</td><td></td></tr><tr><td>82ccd62f70e669ec770daf11d9611cab0a13047e</td><td>Sparse Variation Pattern for Texture Classification +<br/>Electrical Engineering Department +<br/>Computer Science and Software Engineering +<br/>Electrical Engineering Department +<br/><b>Tafresh University</b><br/>Tafresh, Iran +<br/><b>The University of Western Australia</b><br/><b>Central Tehran Branch, Azad University</b><br/>WA 6009, Australia +<br/>Tehran, Iran +</td><td>('2014145', 'Mohammad Tavakolian', 'mohammad tavakolian')<br/>('3046235', 'Farshid Hajati', 'farshid hajati')<br/>('1747500', 'Ajmal S. Mian', 'ajmal s. mian')<br/>('2997971', 'Soheila Gheisari', 'soheila gheisari')</td><td>m tavakolian,hajati@tafreshu.ac.ir +<br/>ajmal.mian@uwa.edu.au +<br/>gheisari.s@iauctb.ac.ir +</td></tr><tr><td>82eff71af91df2ca18aebb7f1153a7aed16ae7cc</td><td>MSU-AVIS dataset: +<br/>Fusing Face and Voice Modalities for Biometric +<br/>Recognition in Indoor Surveillance Videos +<br/><b>Michigan State University, USA</b><br/><b>Yarmouk University, Jordan</b></td><td>('39617163', 'Anurag Chowdhury', 'anurag chowdhury')<br/>('2447931', 'Yousef Atoum', 'yousef atoum')<br/>('1849929', 'Luan Tran', 'luan tran')<br/>('49543771', 'Xiaoming Liu', 'xiaoming liu')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td></td></tr><tr><td>82c303cf4852ad18116a2eea31e2291325bc19c3</td><td>Journal of Image and Graphics, Volume 2, No.1, June, 2014 +<br/>Fusion Based FastICA Method: Facial Expression +<br/>Recognition +<br/><b>Computer Science, Engineering and Mathematics School, Flinders University, Australia</b></td><td>('3105876', 'Humayra B. Ali', 'humayra b. ali')<br/>('1739260', 'David M W Powers', 'david m w powers')</td><td>Email: {ali0041, david.powers}@flinders.edu.au +</td></tr><tr><td>8210fd10ef1de44265632589f8fc28bc439a57e6</td><td>Single Sample Face Recognition via Learning Deep +<br/>Supervised Auto-Encoders +<br/>Shenghua Gao, Yuting Zhang, Kui Jia, Jiwen Lu, Yingying Zhang +</td><td></td><td></td></tr><tr><td>82a4a35b2bae3e5c51f4d24ea5908c52973bd5be</td><td>Real-time emotion recognition for gaming using +<br/>deep convolutional network features +<br/>S´ebastien Ouellet +</td><td></td><td></td></tr><tr><td>82a610a59c210ff77cfdde7fd10c98067bd142da</td><td>UC San Diego +<br/>UC San Diego Electronic Theses and Dissertations +<br/>Title +<br/>Human attention and intent analysis using robust visual cues in a Bayesian framework +<br/>Permalink +<br/>https://escholarship.org/uc/item/1cb8d7vw +<br/>Author +<br/>McCall, Joel Curtis +<br/>Publication Date +<br/>2006-01-01 +<br/>Peer reviewed|Thesis/dissertation +<br/>eScholarship.org +<br/>Powered by the California Digital Library +<br/><b>University of California</b></td><td></td><td></td></tr><tr><td>829f390b3f8ad5856e7ba5ae8568f10cee0c7e6a</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 57– No.20, November 2012 +<br/>A Robust Rotation Invariant Multiview Face Detection in +<br/>Erratic Illumination Condition +<br/>G.Nirmala Priya +<br/>Associate Professor, Department of ECE +<br/><b>Sona College of Technology</b></td><td>('48201570', 'Salem', 'salem')</td><td></td></tr><tr><td>82f4e8f053d20be64d9318529af9fadd2e3547ef</td><td>Technical Report: +<br/>Multibiometric Cryptosystems +</td><td>('2743820', 'Abhishek Nagar', 'abhishek nagar')<br/>('34633765', 'Karthik Nandakumar', 'karthik nandakumar')<br/>('40437942', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>82b43bc9213230af9db17322301cbdf81e2ce8cc</td><td>Attention-Set based Metric Learning for Video Face Recognition +<br/>Center for Research on Intelligent Perception and Computing, +<br/><b>Institute of Automation, Chinese Academy of Sciences</b></td><td>('33079499', 'Yibo Hu', 'yibo hu')<br/>('33680526', 'Xiang Wu', 'xiang wu')<br/>('1705643', 'Ran He', 'ran he')</td><td>yibo.hu@cripac.ia.ac.cn, alfredxiangwu@gmail.com, rhe@nlpr.ia.ac.cn +</td></tr><tr><td>82d781b7b6b7c8c992e0cb13f7ec3989c8eafb3d</td><td>141 +<br/>REFERENCES +<br/>1. +<br/>2. +<br/>3. +<br/>4. +<br/>5. +<br/>6. +<br/>7. +<br/>8. +<br/>9. +<br/>Adler A., Youmaran R. and Loyka S., “Towards a Measure of +<br/>Biometric Information”, Canadian Conference on Electrical and +<br/>Computer Engineering, pp. 210-213, 2006. +<br/>Military Academy, West Point, New York, pp. 452-458, 2005. +<br/>Security and Trust, St. Andrews, New Brunswick, Canada, pp. 1-8, +<br/>2005. +<br/>Structural Model for Biometric Sketch Recognition”, Proceedings of +<br/>DAGM, Magdeburg, Germany, Vol. 2781, pp. 187-195, 2003. +<br/>of Security”, The First UAE International Conference on Biological +<br/>and Medical Physics, pp. 1-4, 2005. +<br/>Avraam Kasapis., “MLPs and Pose, Expression Classification”, +<br/>Proceedings of UNiS Report, pp. 1-87, 2003. +<br/>Detection for Storage Area Networks (SANs)”, Proceedings of 22nd +<br/>IEEE / 13th NASA Goddard Conference on Mass Storage Systems and +<br/>Technologies, pp. 118-127, 2005. +<br/>Black M.J. and Yacoob Y., “Recognizing Facial Expressions in Image +<br/>Sequences using Local Parameterized Models of Image Motion”, Int. +<br/>Journal Computer Vision, Vol. 25, No. 1, pp. 23-48, 1997. +<br/>10. +<br/>Recognition using a State-Based Model of Spatially-Localized Facial +</td><td>('1689298', 'Ahmed', 'ahmed')<br/>('1689298', 'Ahmed', 'ahmed')<br/>('29977973', 'Angle', 'angle')<br/>('20765969', 'Bolle', 'bolle')<br/>('16848439', 'Bourel', 'bourel')</td><td></td></tr><tr><td>82417d8ec8ac6406f2d55774a35af2a1b3f4b66e</td><td>Some faces are more equal than others: +<br/>Hierarchical organization for accurate and +<br/>efficient large-scale identity-based face retrieval +<br/>GREYC, CNRS UMR 6072, Universit´e de Caen Basse-Normandie, France1 +<br/>Technicolor, Rennes, France2 +</td><td>('48467774', 'Binod Bhattarai', 'binod bhattarai')<br/>('2515597', 'Gaurav Sharma', 'gaurav sharma')</td><td></td></tr><tr><td>82e66c4832386cafcec16b92ac88088ffd1a1bc9</td><td>OpenFace: A general-purpose face recognition +<br/>library with mobile applications +<br/>June 2016 +<br/>CMU-CS-16-118 +<br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA 15213 +<br/><b>Poznan University of Technology</b></td><td>('1773498', 'Brandon Amos', 'brandon amos')<br/>('1747303', 'Mahadev Satyanarayanan', 'mahadev satyanarayanan')</td><td></td></tr><tr><td>82eb267b8e86be0b444e841b4b4ed4814b6f1942</td><td>Single Image 3D Interpreter Network +<br/><b>Massachusetts Institute of Technology</b><br/><b>Stanford University</b><br/>3Facebook AI Research +<br/>4Google Research +</td><td>('3045089', 'Jiajun Wu', 'jiajun wu')<br/>('3222730', 'Tianfan Xue', 'tianfan xue')<br/>('35198686', 'Joseph J. Lim', 'joseph j. lim')<br/>('39402399', 'Yuandong Tian', 'yuandong tian')<br/>('1763295', 'Joshua B. Tenenbaum', 'joshua b. tenenbaum')<br/>('1690178', 'Antonio Torralba', 'antonio torralba')<br/>('1768236', 'William T. Freeman', 'william t. freeman')</td><td></td></tr><tr><td>826c66bd182b54fea3617192a242de1e4f16d020</td><td>978-1-5090-4117-6/17/$31.00 ©2017 IEEE +<br/>1602 +<br/>ICASSP 2017 +</td><td></td><td></td></tr><tr><td>499f1d647d938235e9186d968b7bb2ab20f2726d</td><td>Face Recognition via Archetype Hull Ranking +<br/><b>The Chinese University of Hong Kong, Hong Kong</b><br/><b>IBM T. J. Watson Research Center, Yorktown Heights, NY, USA</b></td><td>('3331521', 'Yuanjun Xiong', 'yuanjun xiong')</td><td>{yjxiong,xtang}@ie.cuhk.edu.hk +<br/>weiliu@us.ibm.com +<br/>zhaodeli@gmail.com +</td></tr><tr><td>4919663c62174a9bc0cc7f60da8f96974b397ad2</td><td>HUMAN AGE ESTIMATION USING ENHANCED BIO-INSPIRED FEATURES (EBIF) +<br/><b>Faculty of Computers and Information, Cairo University, Cairo, Egypt</b></td><td>('3144122', 'Motaz El-Saban', 'motaz el-saban')</td><td>{mohamed.y.eldib,motaz.elsaban}@gmail.com +</td></tr><tr><td>49f70f707c2e030fe16059635df85c7625b5dc7e</td><td>www.ietdl.org +<br/>Received on 29th May 2014 +<br/>Revised on 29th August 2014 +<br/>Accepted on 23rd September 2014 +<br/>doi: 10.1049/iet-bmt.2014.0033 +<br/>ISSN 2047-4938 +<br/>Face recognition under illumination variations based +<br/>on eight local directional patterns +<br/><b>Utah State University, Logan, UT 84322-4205, USA</b></td><td>('2147212', 'Mohammad Reza Faraji', 'mohammad reza faraji')<br/>('1725739', 'Xiaojun Qi', 'xiaojun qi')</td><td>E-mail: Mohammadreza.Faraji@aggiemail.usu.edu +</td></tr><tr><td>4967b0acc50995aa4b28e576c404dc85fefb0601</td><td> Vol. 4, No. 1 Jan 2013 ISSN 2079-8407 +<br/>Journal of Emerging Trends in Computing and Information Sciences +<br/>©2009-2013 CIS Journal. All rights reserved. +<br/>An Automatic Face Detection and Gender Classification from +<br/>http://www.cisjournal.org +<br/> Color Images using Support Vector Machine +<br/>1, 2, 3 Department of Electrical & Electronic Engineering, International +<br/><b>University of Business Agriculture and Technology, Dhaka-1230, Bangladesh</b><br/> +</td><td>('2832495', 'Md. Hafizur Rahman', 'md. hafizur rahman')<br/>('2226529', 'Suman Chowdhury', 'suman chowdhury')<br/>('36231591', 'Md. Abul Bashar', 'md. abul bashar')</td><td></td></tr><tr><td>49820ae612b3c0590a8a78a725f4f378cb605cd1</td><td>Evaluation of Smile Detection Methods with +<br/>Images in Real-world Scenarios +<br/><b>Beijing University of Posts and Telecommunications, Beijing, China</b></td><td>('22550265', 'Zhoucong Cui', 'zhoucong cui')<br/>('1678529', 'Shuo Zhang', 'shuo zhang')<br/>('23224233', 'Jiani Hu', 'jiani hu')<br/>('1774956', 'Weihong Deng', 'weihong deng')</td><td></td></tr><tr><td>4972aadcce369a8c0029e6dc2f288dfd0241e144</td><td>Multi-target Unsupervised Domain Adaptation +<br/>without Exactly Shared Categories +</td><td>('2076460', 'Huanhuan Yu', 'huanhuan yu')<br/>('27096523', 'Menglei Hu', 'menglei hu')<br/>('1680768', 'Songcan Chen', 'songcan chen')</td><td></td></tr><tr><td>49dd4b359f8014e85ed7c106e7848049f852a304</td><td></td><td></td><td></td></tr><tr><td>49e975a4c60d99bcc42c921d73f8d89ec7130916</td><td>Human and computer recognition of facial expressions of emotion +<br/>J.M. Susskind a, G. Littlewort b, M.S. Bartlett b, J. Movellan b, A.K. Anderson a,c,∗ +<br/><b>b Machine Perception Laboratory, Institute of Neural Computation, University of California, San Diego, United States</b><br/><b>c Rotman Research Institute, Baycrest Centre for Geriatric Care, Toronto, Ont. M6A 2E1, Canada</b><br/><b>University of Toronto, Canada</b><br/>Available online 12 June 2006 +</td><td></td><td></td></tr><tr><td>49e85869fa2cbb31e2fd761951d0cdfa741d95f3</td><td>253 +<br/>Adaptive Manifold Learning +</td><td>('2923061', 'Zhenyue Zhang', 'zhenyue zhang')<br/>('1697912', 'Jing Wang', 'jing wang')<br/>('1750350', 'Hongyuan Zha', 'hongyuan zha')</td><td></td></tr><tr><td>49659fb64b1d47fdd569e41a8a6da6aa76612903</td><td></td><td></td><td></td></tr><tr><td>490a217a4e9a30563f3a4442a7d04f0ea34442c8</td><td>International Journal on Soft Computing, Artificial Intelligence and Applications (IJSCAI), Vol.2, No.4, August 2013 +<br/>An SOM-based Automatic Facial Expression +<br/>Recognition System +<br/>Hsieh1, andPa-Chun Wang2 +<br/>1Department of Computer Science &InformationEngineering,National Central +<br/><b>University, Taiwan, R.O.C</b><br/>2Cathay General Hospital, Taiwan, R.O.C. +</td><td>('1720774', 'Mu-Chun Su', 'mu-chun su')<br/>('4226881', 'Chun-Kai Yang', 'chun-kai yang')<br/>('40179526', 'Shih-Chieh Lin', 'shih-chieh lin')</td><td>E-mail: muchun@csie.ncu.edu.tw +</td></tr><tr><td>49a7949fabcdf01bbae1c2eb38946ee99f491857</td><td>A CONCATENATING FRAMEWORK OF SHORTCUT +<br/>CONVOLUTIONAL NEURAL NETWORKS +</td><td></td><td>Yujian Li (liyujian@bjut.edu.cn), Ting Zhang, Zhaoying Liu, Haihe Hu +</td></tr><tr><td>4934d44aa89b6d871eb6709dd1d1eebf16f3aaf1</td><td>A Deep Sum-Product Architecture for Robust Facial Attributes Analysis +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('1693209', 'Ping Luo', 'ping luo')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>pluo.lhi@gmail.com +<br/>xgwang@ee.cuhk.edu.hk +<br/>xtang@ie.cuhk.edu.hk +</td></tr><tr><td>499343a2fd9421dca608d206e25e53be84489f44</td><td>Anil Kumar.C, et.al, International Journal of Technology and Engineering Science [IJTES]TM +<br/> +<br/>Volume 1[9], pp: 1371-1375, December 2013 +<br/>Face Recognition with Name Using Local Weber‟s +<br/>Law Descriptor +<br/>1C.Anil kumar,2A.Rajani,3I.Suneetha +<br/>1M.Tech Student,2Assistant Professor,3Associate Professor +<br/><b>Annamacharya Institute of Technology and Sciences, Tirupati, India</b><br/>on FERET +</td><td></td><td>1Anilyadav.kumar7@gmail.com,2rajanirevanth446@gmail.com,3iralasuneetha.aits@gmail.com +</td></tr><tr><td>498fd231d7983433dac37f3c97fb1eafcf065268</td><td>LINEAR DISENTANGLED REPRESENTATION LEARNING FOR FACIAL ACTIONS +<br/>1Dept. of Computer Science +<br/>2Dept. of Electrical & Computer Engineering +<br/><b>Johns Hopkins University, 3400 N. Charles Street, Baltimore, MD 21218, USA</b><br/>Fig. 1. The separability of the neutral face yn and expression +<br/>component ye. We find yn is better for identity recognition +<br/>than y and ye is better for expression recognition than y. +</td><td>('40031188', 'Xiang Xiang', 'xiang xiang')<br/>('1709073', 'Trac D. Tran', 'trac d. tran')</td><td></td></tr><tr><td>49e1aa3ecda55465641b2c2acc6583b32f3f1fc6</td><td>International Journal of Emerging Technology and Advanced Engineering +<br/>Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 5, May 2012) +<br/>Support Vector Machine for age classification +<br/>1Assistant Professor, CSE, RSR RCET, Kohka Bhilai +<br/>2,3 Sr. Assistant Professor, CSE, SSCET, Junwani Bhilai +</td><td>('6552360', 'Sangeeta Agrawal', 'sangeeta agrawal')<br/>('40618181', 'Rohit Raja', 'rohit raja')<br/>('40323262', 'Sonu Agrawal', 'sonu agrawal')</td><td>1agrawal.sans@gmail.com +<br/>2rohitraja4u@gmail.com +<br/>3agrawalsonu@gmail.com +</td></tr><tr><td>499f2b005e960a145619305814a4e9aa6a1bba6a</td><td>Robust human face recognition based on locality preserving +<br/>sparse overcomplete block approximation +<br/><b>University of Geneva</b><br/>7 Route de Drize, Geneva, Switzerland +</td><td>('36133844', 'Dimche Kostadinov', 'dimche kostadinov')<br/>('8995309', 'Sviatoslav Voloshynovskiy', 'sviatoslav voloshynovskiy')<br/>('1682792', 'Sohrab Ferdowsi', 'sohrab ferdowsi')</td><td></td></tr><tr><td>497bf2df484906e5430aa3045cf04a40c9225f94</td><td>Sensors 2013, 13, 16682-16713; doi:10.3390/s131216682 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>Hierarchical Recognition Scheme for Human Facial Expression +<br/>Recognition Systems +<br/><b>UC Lab, Kyung Hee University, Yongin-Si 446-701, Korea</b><br/><b>Division of Information and Computer Engineering, Ajou University, Suwon 443-749, Korea</b><br/>Tel.: +82-31-201-2514. +<br/>Received: 28 October 2013; in revised form: 30 November 2013 / Accepted: 2 December 2013 / +<br/>Published: 5 December 2013 +</td><td>('1711083', 'Muhammad Hameed Siddiqi', 'muhammad hameed siddiqi')<br/>('1700806', 'Sungyoung Lee', 'sungyoung lee')<br/>('1750915', 'Young-Koo Lee', 'young-koo lee')<br/>('1714762', 'Adil Mehmood Khan', 'adil mehmood khan')<br/>('34601872', 'Phan Tran Ho Truc', 'phan tran ho truc')</td><td>E-Mails: siddiqi@oslab.khu.ac.kr (M.H.S.); sylee@oslab.khu.ac.kr (S.L.); yklee@khu.ac.kr (Y.-K.L.) +<br/>E-Mail: amtareen@ajou.ac.kr +<br/>* Author to whom correspondence should be addressed; E-Mail: pthtruc@oslab.khu.ac.kr; +</td></tr><tr><td>492f41e800c52614c5519f830e72561db205e86c</td><td>A Deep Regression Architecture with Two-Stage Re-initialization for +<br/>High Performance Facial Landmark Detection +<br/>Jiangjing Lv1 +<br/><b>Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences</b><br/><b>University of Chinese Academy of Sciences</b><br/><b>Institute of Automation, Chinese Academy of Sciences</b></td><td>('3492237', 'Xiaohu Shao', 'xiaohu shao')<br/>('1757173', 'Junliang Xing', 'junliang xing')<br/>('2095535', 'Cheng Cheng', 'cheng cheng')<br/>('39959302', 'Xi Zhou', 'xi zhou')</td><td>{lvjiangjing,shaoxiaohu,chengcheng,zhouxi}@cigit.ac.cn +<br/>jlxing@nlpr.ia.ac.cn +</td></tr><tr><td>49df381ea2a1e7f4059346311f1f9f45dd997164</td><td>2018 +<br/>On the Use of Client-Specific Information for Face +<br/>Presentation Attack Detection Based on Anomaly +<br/>Detection +</td><td>('1690611', 'Shervin Rahimzadeh Arashloo', 'shervin rahimzadeh arashloo')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td></td></tr><tr><td>493ec9e567c5587c4cbeb5f08ca47408ca2d6571</td><td>You et al. Complex Adapt Syst Model (2016) 4:22 +<br/>DOI 10.1186/s40294‑016‑0034‑7 +<br/>RESEARCH +<br/>Combining graph embedding +<br/>and sparse regression with structure low‑rank +<br/>representation for semi‑supervised learning +<br/>Open Access +<br/>*Correspondence: +<br/>1 School of IoT Engineering, +<br/><b>Jiangnan University, Wuxi</b><br/>China +<br/>Full list of author information +<br/>is available at the end of the +<br/>article +</td><td>('1766488', 'Vasile Palade', 'vasile palade')</td><td>youcongzhe@gmail.com +</td></tr><tr><td>49570b41bd9574bd9c600e24b269d945c645b7bd</td><td>A Framework for Performance Evaluation +<br/>of Face Recognition Algorithms +<br/><b>Visual Computing and Communications Lab, Arizona State University</b></td><td>('40401270', 'John A. Black', 'john a. black')<br/>('1743991', 'Sethuraman Panchanathan', 'sethuraman panchanathan')</td><td></td></tr><tr><td>496074fcbeefd88664b7bd945012ca22615d812e</td><td>Review +<br/>Driver Distraction Using Visual-Based Sensors +<br/>and Algorithms +<br/>1 Grupo TSK, Technological Scientific Park of Gijón, 33203 Gijón, Asturias, Spain; +<br/><b>University of Oviedo, Campus de Viesques, 33204 Gij n</b><br/>Academic Editor: Gonzalo Pajares Martinsanz +<br/>Received: 14 July 2016; Accepted: 24 October 2016; Published: 28 October 2016 +</td><td>('8306548', 'Rubén Usamentiaga', 'rubén usamentiaga')<br/>('27666409', 'Juan Luis Carús', 'juan luis carús')</td><td>juanluis.carus@grupotsk.com +<br/>Asturias, Spain; rusamentiaga@uniovi.es (R.U.); rcasado@lsi.uniovi.es (R.C.) +<br/>* Corrospondence: alberto.fernandez@grupotsk.com; Tel.: +34-984-29-12-12; Fax: +34-984-39-06-12 +</td></tr><tr><td>40205181ed1406a6f101c5e38c5b4b9b583d06bc</td><td>Using Context to Recognize People in Consumer Images +</td><td>('39460815', 'Andrew C. Gallagher', 'andrew c. gallagher')<br/>('1746230', 'Tsuhan Chen', 'tsuhan chen')</td><td></td></tr><tr><td>40dab43abef32deaf875c2652133ea1e2c089223</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Facial Communicative Signals +<br/>Valence Recognition in Task-Oriented Human-Robot Interaction +<br/>Received: date / Accepted: date +</td><td>('33734208', 'Christian Lang', 'christian lang')</td><td></td></tr><tr><td>40b0fced8bc45f548ca7f79922e62478d2043220</td><td>Do Convnets Learn Correspondence? +<br/><b>University of California Berkeley</b></td><td>('1753210', 'Trevor Darrell', 'trevor darrell')<br/>('34703740', 'Jonathan Long', 'jonathan long')<br/>('40565777', 'Ning Zhang', 'ning zhang')</td><td>{jonlong, nzhang, trevor}@cs.berkeley.edu +</td></tr><tr><td>405b43f4a52f70336ac1db36d5fa654600e9e643</td><td>What can we learn about CNNs from a large scale controlled object dataset? +<br/>UWM +<br/>AUT +<br/>USC +</td><td>('3177797', 'Ali Borji', 'ali borji')<br/>('2391309', 'Saeed Izadi', 'saeed izadi')<br/>('7326223', 'Laurent Itti', 'laurent itti')</td><td>borji@uwm.edu +<br/>sizadi@aut.ac.ir +<br/>itti@usc.edu +</td></tr><tr><td>40b86ce698be51e36884edcc8937998979cd02ec</td><td>Yüz ve İsim İlişkisi kullanarak Haberlerdeki Kişilerin Bulunması +<br/>Finding Faces in News Photos Using Both Face and Name Information +<br/>Derya Ozkan, Pınar Duygulu +<br/>Bilgisayar Mühendisliği Bölümü, Bilkent Üniversitesi, 06800, Ankara +<br/>Özetçe +<br/>Bu çalışmada, haber fotoğraflarından oluşan geniş veri +<br/>kümelerinde kişilerin sorgulanmasını sağlayan bir yöntem +<br/>sunulmuştur. Yöntem isim ve yüzlerin ilişkilendirilmesine +<br/>dayanmaktadır. Haber başlığında kişinin ismi geçiyor ise +<br/>fotoğrafta da o kişinin yüzünün bulunacağı varsayımıyla, ilk +<br/>olarak sorgulanan isim ile ilişkilendirilmiş, fotoğraflardaki +<br/>tüm yüzler seçilir. Bu yüzler arasında sorgu kişisine ait farklı +<br/>koşul, poz ve zamanlarda çekilmiş pek çok resmin yanında, +<br/>haberde ismi geçen başka kişilere ait yüzler ya da kullanılan +<br/>yüz bulma yönteminin hatasından kaynaklanan yüz olmayan +<br/>resimler de bulunabilir. Yine de, çoğu zaman, sorgu kişisine +<br/>ait resimler daha çok olup, bu resimler birbirine diğerlerine +<br/>olduğundan daha çok benzeyeceklerdir. Bu nedenle, yüzler +<br/>arasındaki benzerlikler çizgesel olarak betimlendiğinde , +<br/>birbirine en çok benzeyen yüzler bu çizgede en yoğun bileşen +<br/>olacaktır. Bu çalışmada, sorgu ismiyle ilişkilendirilmiş, +<br/>yüzler arasında birbirine en çok benzeyen alt kümeyi bulan, +<br/>çizgeye dayalı bir yöntem sunulmaktadır. +</td><td></td><td>deryao@cs.bilkent.edu.tr, duygulu@cs.bilkent.edu.tr +</td></tr><tr><td>40a74eea514b389b480d6fe8b359cb6ad31b644a</td><td>Discrete Deep Feature Extraction: A Theory and New Architectures +<br/>Aleksandar Stani´c1 +<br/>Helmut B¨olcskei1 +<br/>1Dept. IT & EE, ETH Zurich, Switzerland +<br/><b>University of Vienna, Austria</b></td><td>('2076040', 'Thomas Wiatowski', 'thomas wiatowski')<br/>('2208878', 'Michael Tschannen', 'michael tschannen')<br/>('1690644', 'Philipp Grohs', 'philipp grohs')</td><td></td></tr><tr><td>403a108dec92363fd1f465340bd54dbfe65af870</td><td>describing images with statistics of local non-binarized pixel patterns +<br/>Local Higher-Order Statistics (LHS) +<br/>aGREYC CNRS UMR 6072, Universit´e de Caen Basse-Normandie, France +<br/><b>bMax Planck Institute for Informatics, Germany</b></td><td>('2515597', 'Gaurav Sharma', 'gaurav sharma')</td><td></td></tr><tr><td>40ee38d7ff2871761663d8634c3a4970ed1dc058</td><td>Three-Dimensional Face Recognition: A Fishersurface +<br/>Approach +<br/><b>The University of York, United Kingdom</b></td><td>('2023950', 'Thomas Heseltine', 'thomas heseltine')<br/>('1737428', 'Nick Pears', 'nick pears')<br/>('2405628', 'Jim Austin', 'jim austin')</td><td></td></tr><tr><td>402f6db00251a15d1d92507887b17e1c50feebca</td><td>3D Facial Action Units Recognition for Emotional +<br/>Expression +<br/>1Department of Information Technology and Communication, Politeknik Kuching, Sarawak, Malaysia +<br/>2Faculty of Computer Science and Information Technology, Universiti Malaysia Sarawak, Kota Samarahan, Sarawak, Malaysia +<br/>The muscular activities caused the activation of certain AUs for every facial expression at the certain duration of time +<br/>throughout the facial expression. This paper presents the methods to recognise facial Action Unit (AU) using facial distance +<br/>of the facial features which activates the muscles. The seven facial action units involved are AU1, AU4, AU6, AU12, AU15, +<br/>AU17 and AU25 that characterises happy and sad expression. The recognition is performed on each AU according to rules +<br/>defined based on the distance of each facial points. The facial distances chosen are extracted from twelve facial features. +<br/>Then the facial distances are trained using Support Vector Machine (SVM) and Neural Network (NN). Classification result +<br/>using SVM is presented with several different SVM kernels while result using NN is presented for each training, validation +<br/>and testing phase. +<br/>Keywords: Facial action units recognition, 3D AU recognition, facial expression +<br/> +</td><td>('2801456', 'Hamimah Ujir', 'hamimah ujir')<br/>('3310557', 'Jacey-Lynn Minoi', 'jacey-lynn minoi')</td><td></td></tr><tr><td>404042a1dcfde338cf24bc2742c57c0fb1f48359</td><td>中国图象图形学报 vol.8, no.8, pp.849-859, 2003. +<br/>脸部特征定位方法综述1 +<br/>林维训 潘纲 吴朝晖 潘云鹤 +<br/>（浙江大学计算机系 310027） +<br/>摘 要 脸部特征定位是人脸分析技术的一个重要组成部分，其目标是在图像或图像序列中的指定 +<br/>区域内搜索人脸特征（如眼、鼻、嘴、耳等）的位置。它可广泛应用于人脸检测和定位、人脸识别、 +<br/>姿态识别、表情识别、头部像压缩及重构、脸部动画等领域。近年来该领域的研究有了较大的发展， +<br/>为了让相关领域内的理论研究和开发人员对目前的进展有一个全面的了解，本文将近年来提出的脸 +<br/>部特征定位方法根据其所依据的基本信息类型分为基于先验知识、几何形状、色彩、外观和关联信 +<br/>息等五类并分别作了介绍，对各类方法的性能作了一些比较和讨论，对未来的发展作了展望。 +<br/>关键词 脸部特征定位 脸部特征提取 +<br/>中图法分类号：TP391.41 +<br/>A Survey on Facial Features Localization +<br/><b>College of Computer Science, Zhejiang University</b></td><td></td><td></td></tr><tr><td>4015e8195db6edb0ef8520709ca9cb2c46f29be7</td><td><b>UNIVERSITY OF TARTU</b><br/>FACULTY OF MATHEMATICS AND COMPUTER SCIENCE +<br/><b>Institute of Computer Science</b><br/>Computer Science Curriculum +<br/>Smile Detector Based on the Motion of +<br/>Face Reference Points +<br/>Bachelor’s Thesis (6 ECTS) +<br/>Supervisor: Gholamreza Anbarjafari, PhD +<br/>Tartu 2014 +</td><td>('3168586', 'Andres Traumann', 'andres traumann')</td><td></td></tr><tr><td>407bb798ab153bf6156ba2956f8cf93256b6910a</td><td>Fisher Pruning of Deep Nets for Facial Trait +<br/>Classification +<br/><b>McGill University</b><br/><b>University Street, Montreal, QC H3A 0E9, Canada</b></td><td>('1992537', 'Qing Tian', 'qing tian')<br/>('1699104', 'Tal Arbel', 'tal arbel')<br/>('1713608', 'James J. Clark', 'james j. clark')</td><td></td></tr><tr><td>40fb4e8932fb6a8fef0dddfdda57a3e142c3e823</td><td>A Mixed Generative-Discriminative Framework for Pedestrian Classification +<br/>Dariu M. Gavrila2,3 +<br/>1 Image & Pattern Analysis Group, Dept. of Math. and Comp. Sc., Univ. of Heidelberg, Germany +<br/>2 Environment Perception, Group Research, Daimler AG, Ulm, Germany +<br/>3 Intelligent Systems Lab, Faculty of Science, Univ. of Amsterdam, The Netherlands +</td><td>('1765022', 'Markus Enzweiler', 'markus enzweiler')</td><td>{uni-heidelberg.enzweiler,dariu.gavrila}@daimler.com +</td></tr><tr><td>40dd2b9aace337467c6e1e269d0cb813442313d7</td><td>This thesis has been submitted in fulfilment of the requirements for a postgraduate degree +<br/><b>e.g. PhD, MPhil, DClinPsychol) at the University of Edinburgh. Please note the following</b><br/>terms and conditions of use: +<br/>This work is protected by copyright and other intellectual property rights, which are +<br/>retained by the thesis author, unless otherwise stated. +<br/>A copy can be downloaded for personal non-commercial research or study, without +<br/>prior permission or charge. +<br/>This thesis cannot be reproduced or quoted extensively from without first obtaining +<br/>permission in writing from the author. +<br/>The content must not be changed in any way or sold commercially in any format or +<br/>medium without the formal permission of the author. +<br/>When referring to this work, full bibliographic details including the author, title, +<br/>awarding institution and date of the thesis must be given. +</td><td></td><td></td></tr><tr><td>407de9da58871cae7a6ded2f3a6162b9dc371f38</td><td>TraMNet - Transition Matrix Network for +<br/>Efficient Action Tube Proposals +<br/><b>Oxford Brookes University, UK</b></td><td>('1931660', 'Gurkirt Singh', 'gurkirt singh')<br/>('49348905', 'Suman Saha', 'suman saha')<br/>('1754181', 'Fabio Cuzzolin', 'fabio cuzzolin')</td><td>gurkirt.singh-2015@brookes.ac.uk +</td></tr><tr><td>405526dfc79de98f5bf3c97bf4aa9a287700f15d</td><td>MegaFace: A Million Faces for Recognition at Scale +<br/>D. Miller +<br/>E. Brossard +<br/>S. Seitz +<br/>Dept. of Computer Science and Engineering +<br/><b>University of Washington</b><br/>I. Kemelmacher-Shlizerman +<br/>Figure 1: We evaluate how recognition performs with increasing numbers of faces in the database: (a) shows rank-1 iden- +<br/>tification rates, and (b) rank-10. Recognition rates drop once the number of distractors increases. We also present first +<br/>large-scale human recognition results (up to 10K distractors). Interestingly, Google’s deep learning based FaceNet is more +<br/>robust at scale than humans. See http://megaface.cs.washington.edu to participate in the challenge. +</td><td></td><td></td></tr><tr><td>40cd062438c280c76110e7a3a0b2cf5ef675052c</td><td></td><td></td><td></td></tr><tr><td>40b7e590dfd1cdfa1e0276e9ca592e02c1bd2b5b</td><td>Beyond Trade-off: Accelerate FCN-based Face Detector with Higher Accuracy +<br/><b>Beihang University, 2The Chinese University of Hong Kong, 3Sensetime Group Limited</b></td><td>('12920342', 'Guanglu Song', 'guanglu song')<br/>('1715752', 'Yu Liu', 'yu liu')<br/>('40452812', 'Ming Jiang', 'ming jiang')<br/>('33598672', 'Yujie Wang', 'yujie wang')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('2858789', 'Biao Leng', 'biao leng')</td><td>{guanglusong,jiangming1406,yujiewang,lengbiao}@buaa.edu.cn, +<br/>yuliu@ee.cuhk.edu.hk, yanjunjie@sensetime.com +</td></tr><tr><td>40a5b32e261dc5ccc1b5df5d5338b7d3fe10370d</td><td>Feedback-Controlled Sequential Lasso Screening +<br/>Department of Electrical Engineering +<br/><b>Princeton University</b></td><td>('1719525', 'Yun Wang', 'yun wang')<br/>('1734498', 'Xu Chen', 'xu chen')<br/>('1693135', 'Peter J. Ramadge', 'peter j. ramadge')</td><td></td></tr><tr><td>40a1935753cf91f29ffe25f6c9dde2dc49bf2a3a</td><td>80 +</td><td></td><td></td></tr><tr><td>40a9f3d73c622cceee5e3d6ca8faa56ed6ebef60</td><td>AUTOMATIC LIP TRACKING AND ACTION UNITS CLASSIFICATION USING +<br/>TWO-STEP ACTIVE CONTOURS AND PROBABILISTIC NEURAL NETWORKS +<br/>Faculty of Electrical and +<br/>Computer Engineering +<br/><b>University of Tabriz, Tabriz, Iran</b><br/>WonSook LEE +<br/>School of Information Technology +<br/>and Engineering (SITE) +<br/>Faculty of Engineering, +<br/><b>University of Ottawa, Canada</b><br/>Faculty of Electrical and +<br/>Computer Engineering +<br/><b>University of Tabriz, Tabriz, Iran</b><br/> +<br/> +</td><td>('3210269', 'Hadi Seyedarabi', 'hadi seyedarabi')<br/>('2488201', 'Ali Aghagolzadeh', 'ali aghagolzadeh')</td><td>email: hadis@discover.uottawa.ca +<br/>email: wslee@uottawa.ca +<br/>email: aghagol@tabrizu.ac.ir +</td></tr><tr><td>40a34d4eea5e32dfbcef420ffe2ce7c1ee0f23cd</td><td>Bridging Heterogeneous Domains With Parallel Transport For Vision and +<br/>Multimedia Applications +<br/>Dept. of Video and Multimedia Technologies Research +<br/>AT&T Labs-Research +<br/>San Francisco, CA 94108 +</td><td>('33692583', 'Raghuraman Gopalan', 'raghuraman gopalan')</td><td></td></tr><tr><td>40389b941a6901c190fb74e95dc170166fd7639d</td><td>Automatic Facial Expression Recognition +<br/>Emotient +<br/>http://emotient.com +<br/>February 12, 2014 +<br/>Imago animi vultus est, indices oculi. (Cicero) +<br/>Introduction +<br/>The face is innervated by two different brain systems that compete for control of its muscles: +<br/>a cortical brain system related to voluntary and controllable behavior, and a sub-cortical +<br/>system responsible for involuntary expressions. The interplay between these two systems +<br/>generates a wealth of information that humans constantly use to read the emotions, inten- +<br/>tions, and interests [25] of others. +<br/>Given the critical role that facial expressions play in our daily life, technologies that can +<br/>interpret and respond to facial expressions automatically are likely to find a wide range of +<br/>applications. For example, in pharmacology, the effect of new anti-depression drugs could +<br/>be assessed more accurately based on daily records of the patients’ facial expressions than +<br/>asking the patients to fill out a questionnaire, as it is currently done [7]. Facial expression +<br/>recognition may enable a new generation of teaching systems to adapt to the expression +<br/>of their students in the way good teachers do [61]. Expression recognition could be used +<br/>to assess the fatigue of drivers and air-pilots [58, 59]. Daily-life robots with automatic +<br/>expression recognition will be able to assess the states and intentions of humans and respond +<br/>accordingly [41]. Smart phones with expression analysis may help people to prepare for +<br/>important meetings and job interviews. +<br/>Thanks to the introduction of machine learning methods, recent years have seen great +<br/>progress in the field of automatic facial expression recognition. Commercial real-time ex- +<br/>pression recognition systems are starting to be used in consumer applications, e.g., smile +<br/>detectors embedded in digital cameras [62]. Nonetheless, considerable progress has yet to be +<br/>made: Methods for face detection and tracking (the first step of automated face analysis) +<br/>work well for frontal views of adult Caucasian and Asian faces [50], but their performance +</td><td>('1775637', 'Jacob Whitehill', 'jacob whitehill')<br/>('40648952', 'Marian Stewart', 'marian stewart')<br/>('1741200', 'Javier R. Movellan', 'javier r. movellan')</td><td></td></tr><tr><td>40e1743332523b2ab5614bae5e10f7a7799161f4</td><td>Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural +<br/>Networks +<br/><b>Centre for Vision, Speech and Signal Processing, University of Surrey, Guildford GU2 7XH, UK</b><br/><b>School of IoT Engineering, Jiangnan University, Wuxi 214122, China</b></td><td>('2976854', 'Zhen-Hua Feng', 'zhen-hua feng')<br/>('1748684', 'Josef Kittler', 'josef kittler')</td><td>{z.feng, j.kittler, m.a.rana}@surrey.ac.uk, patrikhuber@gmail.com, wu xiaojun@jiangnan.edu.cn +</td></tr><tr><td>40c8cffd5aac68f59324733416b6b2959cb668fd</td><td>Pooling Facial Segments to Face: The Shallow and Deep Ends +<br/>Department of Electrical and Computer Engineering and the Center for Automation Research, +<br/><b>UMIACS, University of Maryland, College Park, MD</b></td><td>('3152615', 'Upal Mahbub', 'upal mahbub')<br/>('40599829', 'Sayantan Sarkar', 'sayantan sarkar')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>{umahbub, ssarkar2, rama}@umiacs.umd.edu +</td></tr><tr><td>40273657e6919455373455bd9a5355bb46a7d614</td><td>Anonymizing k-Facial Attributes via Adversarial Perturbations +<br/>1 IIIT Delhi, New Delhi, India +<br/>2 Ministry of Electronics and Information Technology, New Delhi, India +</td><td>('24380882', 'Saheb Chhabra', 'saheb chhabra')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('50046315', 'Gaurav Gupta', 'gaurav gupta')</td><td>{sahebc, rsingh, mayank@iiitd.ac.in}, gauravg@gov.in +</td></tr><tr><td>40b10e330a5511a6a45f42c8b86da222504c717f</td><td>Implementing the Viola-Jones +<br/>Face Detection Algorithm +<br/>Kongens Lyngby 2008 +<br/>IMM-M.Sc.-2008-93 +</td><td>('24007383', 'Ole Helvig Jensen', 'ole helvig jensen')</td><td></td></tr><tr><td>40bb090a4e303f11168dce33ed992f51afe02ff7</td><td>Marginal Loss for Deep Face Recognition +<br/><b>Imperial College London</b><br/><b>Imperial College London</b><br/><b>Imperial College London</b><br/>UK +<br/>UK +<br/>UK +</td><td>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('2321938', 'Yuxiang Zhou', 'yuxiang zhou')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>j.deng16@imperial.ac.uk +<br/>yuxiang.zhou10@imperial.ac.uk +<br/>s.zafeiriou@imperial.ac.uk +</td></tr><tr><td>40ca925befa1f7e039f0cd40d57dbef6007b4416</td><td>Sampling Matters in Deep Embedding Learning +<br/>UT Austin +<br/>A9/Amazon +<br/>Amazon +<br/>Philipp Kr¨ahenb¨uhl +<br/>UT Austin +</td><td>('2978413', 'Chao-Yuan Wu', 'chao-yuan wu')<br/>('1758550', 'R. Manmatha', 'r. manmatha')<br/>('1691629', 'Alexander J. Smola', 'alexander j. smola')</td><td>cywu@cs.utexas.edu +<br/>manmatha@a9.com +<br/>smola@amazon.com +<br/>philkr@cs.utexas.edu +</td></tr><tr><td>4042bbb4e74e0934f4afbedbe92dd3e37336b2f4</td><td></td><td></td><td></td></tr><tr><td>4026dc62475d2ff2876557fc2b0445be898cd380</td><td>An Affective User Interface Based on Facial Expression +<br/>Recognition and Eye-Gaze Tracking +<br/><b>School of Computer Engineering, Sejong University, Seoul, Korea</b></td><td>('7236280', 'Soo-Mi Choi', 'soo-mi choi')<br/>('2706430', 'Yong-Guk Kim', 'yong-guk kim')</td><td>{smchoi,ykim}@sejong.ac.kr +</td></tr><tr><td>40f127fa4459a69a9a21884ee93d286e99b54c5f</td><td>Optimizing Apparent Display Resolution +<br/>Enhancement for Arbitrary Videos +</td><td>('2267017', 'Michael Stengel', 'michael stengel')<br/>('1701306', 'Martin Eisemann', 'martin eisemann')<br/>('34751565', 'Stephan Wenger', 'stephan wenger')<br/>('2765149', 'Benjamin Hell', 'benjamin hell')</td><td></td></tr><tr><td>401e6b9ada571603b67377b336786801f5b54eee</td><td>Active Image Clustering: Seeking Constraints from +<br/>Humans to Complement Algorithms +<br/>November 22, 2011 +</td><td></td><td></td></tr><tr><td>406431d2286a50205a71f04e0b311ba858fc7b6c</td><td>3D FACIAL EXPRESSION CLASSIFICATION USING +<br/>A STATISTICAL MODEL OF SURFACE NORMALS +<br/>AND A MODULAR APPROACH +<br/>A thesis submitted to +<br/><b>University of Birmingham</b><br/>for the degree of +<br/>DOCTOR OF PHILOSOPHY +<br/>School of Electronic, Electrical & Computer Engineering +<br/><b>University of Birmingham</b><br/>August 2012 +</td><td>('2801456', 'Hamimah Ujir', 'hamimah ujir')</td><td></td></tr><tr><td>40217a8c60e0a7d1735d4f631171aa6ed146e719</td><td>Part-Pair Representation for Part Localization +<br/><b>Columbia University</b></td><td>('2454675', 'Jiongxin Liu', 'jiongxin liu')<br/>('3173493', 'Yinxiao Li', 'yinxiao li')<br/>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')</td><td>{liujx09, yli, belhumeur}@cs.columbia.edu +</td></tr><tr><td>2e20ed644e7d6e04dd7ab70084f1bf28f93f75e9</td><td></td><td></td><td></td></tr><tr><td>2e8e6b835e5a8f55f3b0bdd7a1ff765a0b7e1b87</td><td>International Journal of Computer Vision manuscript No. +<br/>(will be inserted by the editor) +<br/>Pointly-Supervised Action Localization +<br/>Received: date / Accepted: date +</td><td>('2606260', 'Pascal Mettes', 'pascal mettes')</td><td></td></tr><tr><td>2eb37a3f362cffdcf5882a94a20a1212dfed25d9</td><td>4 +<br/>Local Feature Based Face Recognition +<br/>R.I.T., Rajaramnagar and S.G.G.S. COE &T, Nanded +<br/>India +<br/>1. Introduction +<br/>A reliable automatic face recognition (AFR) system is a need of time because in today's +<br/>networked world, maintaining the security of private information or physical property is +<br/>becoming increasingly important and difficult as well. Most of the time criminals have been +<br/>taking the advantage of fundamental flaws in the conventional access control systems i.e. +<br/>the systems operating on credit card, ATM etc. do not grant access by "who we are", but by +<br/>"what we have”. The biometric based access control systems have a potential to overcome +<br/>most of the deficiencies of conventional access control systems and has been gaining the +<br/>importance in recent years. These systems can be designed with biometric traits such as +<br/>fingerprint, face, iris, signature, hand geometry etc. But comparison of different biometric +<br/>traits shows that face is very attractive biometric because of its non-intrusiveness and social +<br/>acceptability. It provides automated methods of verifying or recognizing the identity of a +<br/>living person based on its facial characteristics. +<br/>In last decade, major advances occurred in face recognition, with many systems capable of +<br/>achieving recognition rates greater than 90%. However real-world scenarios remain a +<br/>challenge, because face acquisition process can undergo to a wide range of variations. Hence +<br/>the AFR can be thought as a very complex object recognition problem, where the object to be +<br/>recognized is the face. This problem becomes even more difficult because the search is done +<br/>among objects belonging to the same class and very few images of each class are available to +<br/>train the system. Moreover different problems arise when images are acquired under +<br/>uncontrolled conditions such as illumination variations, pose changes, occlusion, person +<br/>appearance at different ages, expression changes and face deformations. The numbers of +<br/>approaches has been proposed by various researchers to deal with these problems but still +<br/>reported results cannot suffice the need of the reliable AFR system in presence of all facial +<br/>image variations. A recent survey paper (Abate et al., 2007) states that the sensibility of the +<br/>AFR systems to illumination and pose variations are the main problems researchers have +<br/>been facing up till. +<br/>2. Face recognition methods +<br/>The existing face recognition methods can be divided into two categories: holistic matching +<br/>methods and local matching methods.The holistic matching methods use complete face +<br/>region as a input to face recognition system and constructs a lower dimensional subspace +<br/>using principal component analysis (PCA) (Turk & Pentland, 1991), linear discriminant +<br/>www.intechopen.com +</td><td>('2321206', 'Sanjay A. Pardeshi', 'sanjay a. pardeshi')<br/>('3092481', 'Sanjay N. Talbar', 'sanjay n. talbar')</td><td></td></tr><tr><td>2e0addeffba4be98a6ad0460453fbab52616b139</td><td>Face View Synthesis +<br/>Using A Single Image +<br/>Thesis Proposal +<br/>May 2006 +<br/>Committee Members +<br/>Henry Schneiderman (Chair) +<br/>Alexei (Alyosha) Efros +<br/><b>Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, Pennsylvania 15213 +<br/><b>c(cid:13) Carnegie Mellon University</b></td><td>('2989714', 'Jiang Ni', 'jiang ni')<br/>('1709305', 'Martial Hebert', 'martial hebert')<br/>('38998440', 'David Kriegman', 'david kriegman')</td><td></td></tr><tr><td>2e5cfa97f3ecc10ae8f54c1862433285281e6a7c</td><td></td><td></td><td></td></tr><tr><td>2e091b311ac48c18aaedbb5117e94213f1dbb529</td><td>Collaborative Facial Landmark Localization +<br/>for Transferring Annotations Across Datasets +<br/><b>University of Wisconsin Madison</b><br/>http://www.cs.wisc.edu/~lizhang/projects/collab-face-landmarks/ +</td><td>('1893050', 'Brandon M. Smith', 'brandon m. smith')<br/>('40396555', 'Li Zhang', 'li zhang')</td><td></td></tr><tr><td>2e1415a814ae9abace5550e4893e13bd988c7ba1</td><td>International Journal of Engineering Trends and Technology (IJETT) – Volume 21 Number 3 – March 2015 +<br/>Dictionary Based Face Recognition in Video Using +<br/>Fuzzy Clustering and Fusion +<br/>#1IInd year M.E. Student, #2Assistant Professor +<br/><b>Dhanalakshmi Srinivasan College of Engineering</b><br/>Coimbatore,Tamilnadu,India. +<br/><b>Anna University</b></td><td></td><td></td></tr><tr><td>2e0e056ed5927a4dc6e5c633715beb762628aeb0</td><td></td><td></td><td></td></tr><tr><td>2e8a0cc071017845ee6f67bd0633b8167a47abed</td><td>Spatio-Temporal Covariance Descriptors for Action and Gesture Recognition +<br/>NICTA, PO Box 6020, St Lucia, QLD 4067, Australia ∗ +<br/><b>University of Queensland, School of ITEE, QLD 4072, Australia</b></td><td>('2706642', 'Andres Sanin', 'andres sanin')<br/>('1781182', 'Conrad Sanderson', 'conrad sanderson')<br/>('2270092', 'Brian C. Lovell', 'brian c. lovell')</td><td></td></tr><tr><td>2e68190ebda2db8fb690e378fa213319ca915cf8</td><td>Generating Videos with Scene Dynamics +<br/>MIT +<br/>UMBC +<br/>MIT +</td><td>('1856025', 'Carl Vondrick', 'carl vondrick')<br/>('2367683', 'Hamed Pirsiavash', 'hamed pirsiavash')<br/>('1690178', 'Antonio Torralba', 'antonio torralba')</td><td>vondrick@mit.edu +<br/>hpirsiav@umbc.edu +<br/>torralba@mit.edu +</td></tr><tr><td>2e0d56794379c436b2d1be63e71a215dd67eb2ca</td><td>Improving precision and recall of face recognition in SIPP with combination of +<br/>modified mean search and LSH +<br/>Xihua.Li +</td><td></td><td>lixihua9@126.com +</td></tr><tr><td>2ee8900bbde5d3c81b7ed4725710ed46cc7e91cd</td><td></td><td></td><td></td></tr><tr><td>2e475f1d496456831599ce86d8bbbdada8ee57ed</td><td>Groupsourcing: Team Competition Designs for +<br/>Crowdsourcing +<br/><b>L3S Research Center, Hannover, Germany</b></td><td>('2993225', 'Markus Rokicki', 'markus rokicki')<br/>('2553718', 'Sergej Zerr', 'sergej zerr')<br/>('1745880', 'Stefan Siersdorfer', 'stefan siersdorfer')</td><td>{rokicki,siersdorfer,zerr}@L3S.de +</td></tr><tr><td>2ef51b57c4a3743ac33e47e0dc6a40b0afcdd522</td><td>Leveraging Billions of Faces to Overcome +<br/>Performance Barriers in Unconstrained Face +<br/>Recognition +<br/>face.com +</td><td>('2188620', 'Yaniv Taigman', 'yaniv taigman')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td>{yaniv, wolf}@face.com +</td></tr><tr><td>2e231f1e7e641dd3619bec59e14d02e91360ac01</td><td>FUSION NETWORK FOR FACE-BASED AGE ESTIMATION +<br/><b>The University of Warwick, Coventry, UK</b><br/><b>School of Management, University of Bath, Bath, UK</b><br/><b>School of Computing and Mathematics, Charles Sturt University, Wagga Wagga, Australia</b></td><td>('1750506', 'Haoyi Wang', 'haoyi wang')<br/>('40655450', 'Xingjie Wei', 'xingjie wei')<br/>('1901920', 'Victor Sanchez', 'victor sanchez')<br/>('1799504', 'Chang-Tsun Li', 'chang-tsun li')</td><td>{h.wang.16, vsanchez, C-T.Li}@warwick.ac.uk, x.wei@bath.ac.uk +</td></tr><tr><td>2e6cfeba49d327de21ae3186532e56cadeb57c02</td><td>Real Time Eye Gaze Tracking with 3D Deformable Eye-Face Model +<br/><b>Rensselaer Polytechnic Institute</b><br/>110 8th Street, Troy, NY, USA +</td><td>('1771700', 'Kang Wang', 'kang wang')<br/>('1726583', 'Qiang Ji', 'qiang ji')</td><td>{wangk10, jiq}@rpi.edu +</td></tr><tr><td>2ee817981e02c4709d65870c140665ed25b005cc</td><td>Sparse Representations and Random Projections for +<br/>Robust and Cancelable Biometrics +<br/>(Invited Paper) +<br/>Center for Automation Research +<br/><b>University of Maryland</b><br/><b>College Park, MD 20742 USA</b><br/><b>DAP - University of Sassari</b><br/>piazza Duomo, 6 +<br/>Alghero 07041 Italy +<br/>robust and secure physiological biometrics recognition such +<br/>as face and iris [6], [7], [9], [1]. In this paper, we categorize +<br/>approaches to biometrics based on sparse representations. +</td><td>('1741177', 'Vishal M. Patel', 'vishal m. patel')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')<br/>('1725688', 'Massimo Tistarelli', 'massimo tistarelli')</td><td>{pvishalm,rama}@umiacs.umd.edu +<br/>tista@uniss.it +</td></tr><tr><td>2e98329fdec27d4b3b9b894687e7d1352d828b1d</td><td>Using Affect Awareness to Modulate Task Experience: +<br/>A Study Amongst Pre-Elementary School Kids +<br/><b>Carnegie Mellon University</b><br/>5000 Forbes Avenue, +<br/>Pittsburgh, PA 15213 +</td><td>('29120285', 'Vivek Pai', 'vivek pai')<br/>('1760345', 'Raja Sooriamurthi', 'raja sooriamurthi')</td><td></td></tr><tr><td>2e19371a2d797ab9929b99c80d80f01a1fbf9479</td><td></td><td></td><td></td></tr><tr><td>2ed4973984b254be5cba3129371506275fe8a8eb</td><td> +<br/>THE EFFECTS OF MOOD ON +<br/>EMOTION RECOGNITION AND +<br/>ITS RELATIONSHIP WITH THE +<br/>GLOBAL VS LOCAL +<br/>INFORMATION PROCESSING +<br/>STYLES +<br/>BASIC RESEARCH PROGRAM +<br/>WORKING PAPERS +<br/>SERIES: PSYCHOLOGY +<br/>WP BRP 60/PSY/2016 +<br/>This Working Paper is an output of a research project implemented at the National Research +<br/><b>University Higher School of Economics (HSE). Any opinions or claims contained in this</b><br/>Working Paper do not necessarily reflect the views of HSE +<br/> +</td><td>('15615673', 'Victoria Ovsyannikova', 'victoria ovsyannikova')</td><td></td></tr><tr><td>2e9c780ee8145f29bd1a000585dd99b14d1f5894</td><td>Simultaneous Adversarial Training - Learn from +<br/>Others’ Mistakes +<br/><b>Lite-On Singapore Pte. Ltd, 2Imperial College London</b></td><td>('9949538', 'Zukang Liao', 'zukang liao')</td><td></td></tr><tr><td>2ebc35d196cd975e1ccbc8e98694f20d7f52faf3</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. +<br/>IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +<br/>Towards Wide-angle Micro Vision Sensors +</td><td>('2724462', 'Sanjeev J. Koppal', 'sanjeev j. koppal')<br/>('2407724', 'Ioannis Gkioulekas', 'ioannis gkioulekas')<br/>('2140759', 'Kenneth B. Crozier', 'kenneth b. crozier')</td><td></td></tr><tr><td>2e3d081c8f0e10f138314c4d2c11064a981c1327</td><td></td><td></td><td></td></tr><tr><td>2e86402b354516d0a8392f75430156d629ca6281</td><td></td><td></td><td></td></tr><tr><td>2ea78e128bec30fb1a623c55ad5d55bb99190bd2</td><td>Residual vs. Inception vs. Classical Networks for +<br/>Low-Resolution Face Recognition +<br/><b>Vision and Fusion Lab, Karlsruhe Institute of Technology KIT, Karlsruhe, Germany</b><br/>2Fraunhofer IOSB, Karlsruhe, Germany +<br/>{christian.herrmann,dieter.willersinn, +</td><td>('37646107', 'Christian Herrmann', 'christian herrmann')<br/>('1783486', 'Dieter Willersinn', 'dieter willersinn')</td><td>juergen.beyerer}@iosb.fraunhofer.de +</td></tr><tr><td>2e8eb9dc07deb5142a99bc861e0b6295574d1fbd</td><td>Analysis by Synthesis: 3D Object Recognition by Object Reconstruction +<br/><b>University of California, Irvine</b><br/><b>University of California, Irvine</b></td><td>('1888731', 'Mohsen Hejrati', 'mohsen hejrati')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')</td><td>shejrati@ics.uci.edu +<br/>dramanan@ics.uci.edu +</td></tr><tr><td>2e0f5e72ad893b049f971bc99b67ebf254e194f7</td><td>Apparel Classification with Style +<br/>1ETH Z¨urich, Switzerland 2Microsoft, Austria 3Kooaba AG, Switzerland +<br/>4KU Leuven, Belgium +</td><td>('1696393', 'Lukas Bossard', 'lukas bossard')<br/>('1727791', 'Matthias Dantone', 'matthias dantone')<br/>('1695579', 'Christian Leistner', 'christian leistner')<br/>('1793359', 'Christian Wengert', 'christian wengert')<br/>('1726249', 'Till Quack', 'till quack')<br/>('1681236', 'Luc Van Gool', 'luc van gool')</td><td></td></tr><tr><td>2e3c893ac11e1a566971f64ae30ac4a1f36f5bb5</td><td>Simultaneous Object Detection and Ranking with +<br/>Weak Supervision +<br/>Department of Engineering Science +<br/><b>University of Oxford</b><br/>United Kingdom +</td><td>('1758219', 'Matthew B. Blaschko', 'matthew b. blaschko')<br/>('1687524', 'Andrea Vedaldi', 'andrea vedaldi')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>2ed3ce5cf9e262bcc48a6bd998e7fb70cf8a971c</td><td>Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 26 January 2017 doi:10.20944/preprints201701.0120.v1 +<br/>Peer-reviewed version available at Sensors 2017, 17, 275; doi:10.3390/s17020275 +<br/>Article +<br/>Active AU Based Patch Weighting for Facial +<br/>Expression Recognition +<br/><b>School of Computer Science and Software Engineering, Shenzhen University, Nanhai Ave 3688, Shenzhen</b><br/>Guangdong 518060, China +</td><td>('34181727', 'Weicheng Xie', 'weicheng xie')<br/>('1687690', 'LinLin Shen', 'linlin shen')<br/>('5828998', 'Meng Yang', 'meng yang')<br/>('5383601', 'Zhihui Lai', 'zhihui lai')</td><td>* Correspondence: llshen@szu.edu.cn; Tel.: +86-755-8693-5089 +</td></tr><tr><td>2edc6df161f6aadbef9c12408bdb367e72c3c967</td><td>Improved Spatiotemporal Local Monogenic Binary Pattern +<br/>for Emotion Recognition in The Wild +<br/>Center for Machine Vision +<br/>Research +<br/>Department of Computer +<br/>Science and Engineering +<br/><b>University of Oulu, Finland</b><br/>Center for Machine Vision +<br/>Research +<br/>Department of Computer +<br/>Science and Engineering +<br/><b>University of Oulu, Finland</b><br/>Center for Machine Vision +<br/>Research +<br/>Department of Computer +<br/>Science and Engineering +<br/><b>University of Oulu, Finland</b><br/>Center for Machine Vision +<br/>Research +<br/>Department of Computer +<br/>Science and Engineering +<br/><b>University of Oulu, Finland</b><br/>Matti Pietikänen +<br/>Center for Machine Vision +<br/>Research +<br/>Department of Computer +<br/>Science and Engineering +<br/><b>University of Oulu, Finland</b></td><td>('18780812', 'Xiaohua Huang', 'xiaohua huang')<br/>('2512942', 'Qiuhai He', 'qiuhai he')<br/>('1836646', 'Xiaopeng Hong', 'xiaopeng hong')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')</td><td>huang.xiaohua@ee.oulu.fi +<br/>qiuhai.he@ee.oulu.fi +<br/>xhong@ee.oulu.fi +<br/>gyzhao@ee.oulu.fi +<br/>mkp@ee.oulu.fi +</td></tr><tr><td>2ec7d6a04c8c72cc194d7eab7456f73dfa501c8c</td><td>International Journal of Scientific Research and Management Studies (IJSRMS) +<br/>ISSN: 2349-3771 +<br/> +<br/>Volume 3 Issue 4, pg: 164-169 +<br/>A REVIEW ON TEXTURE BASED EMOTION RECOGNITION +<br/>FROM FACIAL EXPRESSION +<br/>1U.G. Scholars, 2Assistant Professor, +<br/>Dept. of E & C Engg., MIT Moradabad, Ram Ganga Vihar, Phase II, Moradabad, India. +</td><td>('5255436', 'Shubham Kashyap', 'shubham kashyap')<br/>('2036732', 'Pankaj Pandey', 'pankaj pandey')<br/>('36216996', 'Prashant Kumar', 'prashant kumar')</td><td></td></tr><tr><td>2eb9f1dbea71bdc57821dedbb587ff04f3a25f07</td><td>Face for Ambient Interface +<br/><b>Imperial College, 180 Queens Gate</b><br/>London SW7 2AZ, U.K. +</td><td>('1694605', 'Maja Pantic', 'maja pantic')</td><td>m.pantic@imperial.ac.uk +</td></tr><tr><td>2e1fd8d57425b727fd850d7710d38194fa6e2654</td><td>Learning Structured Appearance Models +<br/>from Captioned Images of Cluttered Scenes ∗ +<br/><b>University of Toronto</b><br/><b>Bielefeld University</b></td><td>('37894231', 'Michael Jamieson', 'michael jamieson')<br/>('1724954', 'Sven Wachsmuth', 'sven wachsmuth')</td><td>{jamieson, afsaneh, sven, suzanne}@cs.toronto.edu +<br/>swachsmu@techfak.uni-bielefeld.de +</td></tr><tr><td>2e1b1969ded4d63b69a5ec854350c0f74dc4de36</td><td></td><td></td><td></td></tr><tr><td>2e832d5657bf9e5678fd45b118fc74db07dac9da</td><td>Running head: RECOGNITION OF FACIAL EXPRESSIONS OF EMOTION +<br/>1 +<br/>Recognition of Facial Expressions of Emotion: The Effects of Anxiety, Depression, and Fear of Negative +<br/>Evaluation +<br/>Rachel Merchak +<br/><b>Wittenberg University</b><br/><b>Rachel Merchak, Wittenberg University</b><br/>Author Note +<br/>This research was conducted in collaboration with Dr. Stephanie Little, Psychology Department, +<br/><b>Wittenberg University, and Dr. Michael Anes, Wittenberg University</b><br/>Correspondence concerning this article should be addressed to Rachel Merchak, 10063 Fox +<br/>Chase Drive, Loveland, OH 45140. +</td><td></td><td>E‐mail: merchakr@wittenberg.edu +</td></tr><tr><td>2be0ab87dc8f4005c37c523f712dd033c0685827</td><td>RELAXED LOCAL TERNARY PATTERN FOR FACE RECOGNITION +<br/>BeingThere Centre +<br/><b>Institute of Media Innovation</b><br/><b>Nanyang Technological University</b><br/>50 Nanyang Drive, Singapore 637553. +<br/>School of Electrical & Electronics Engineering +<br/><b>Nanyang Technological University</b><br/>50 Nanyang Avenue, Singapore 639798 +</td><td>('1690809', 'Jianfeng Ren', 'jianfeng ren')<br/>('3307580', 'Xudong Jiang', 'xudong jiang')<br/>('34316743', 'Junsong Yuan', 'junsong yuan')</td><td></td></tr><tr><td>2bb2ba7c96d40e269fc6a2d5384c739ff9fa16eb</td><td>Image-based recommendations on styles and substitutes +<br/>Julian McAuley +<br/>UC San Diego +<br/><b>University of Adelaide</b><br/>Qinfeng (‘Javen’) Shi +<br/><b>University of Adelaide</b></td><td>('2110208', 'Christopher Targett', 'christopher targett')</td><td>jmcauley@ucsd.edu +<br/>christopher.targett@student.adelaide.edu.au +<br/>javen.shi@adelaide.edu.au +</td></tr><tr><td>2b339ece73e3787f445c5b92078e8f82c9b1c522</td><td>Human Re-identification in Crowd Videos Using +<br/>Personal, Social and Environmental Constraints +<br/><b>University of Central Florida, Orlando, USA</b><br/>Center for Research in Computer Vision, +</td><td>('2963501', 'Shayan Modiri Assari', 'shayan modiri assari')<br/>('1803711', 'Haroon Idrees', 'haroon idrees')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>{smodiri,haroon,shah}@cs.ucf.edu +</td></tr><tr><td>2b4d092d70efc13790d0c737c916b89952d4d8c7</td><td>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 32, XXXX-XXXX (2016) +<br/>Robust Facial Expression Recognition using Local Haar +<br/>Mean Binary Pattern +<br/>1,2 Department of Computer Engineering +<br/><b>Charotar University of Science and Technology, Changa, India</b><br/><b>Gujarat Technological University, V.V.Nagar, India</b><br/>In this paper, we propose a hybrid statistical feature extractor, Local Haar Mean Bina- +<br/>ry Pattern (LHMBP). It extracts level-1 haar approximation coefficients and computes Local +<br/>Mean Binary Pattern (LMBP) of it. LMBP code of pixel is obtained by weighting the +<br/>thresholded neighbor value of 3 3 patch on its mean. LHMBP produces highly discrimina- +<br/>tive code compared to other state of the art methods. To localize appearance features, ap- +<br/>proximation subband is divided into M N regions. LHMBP feature descriptor is derived +<br/>by concatenating LMBP distribution of each region. We also propose a novel template +<br/>matching strategy called Histogram Normalized Absolute Difference (HNAD) for histogram +<br/>based feature comparison. Experiments prove the superiority of HNAD over well-known +<br/>template matching techniques such as L2 norm and Chi-Square. We also investigated +<br/>LHMBP for expression recognition in low resolution. The performance of the proposed ap- +<br/>proach is tested on well-known CK, JAFFE, and SFEW facial expression datasets in diverse +<br/>situations. +<br/>Keywords: affective computing, appearance based feature, local binary pattern, Gabor filter, +<br/>support vector machine. +<br/>1. INTRODUCTION +<br/>Facial Expression Recognition (FER) is a classical problem of pattern recognition +<br/>and machine learning. It plays a vital role in social communication and in conveying +<br/>emotions [1]. In the earlier development stage, the scope of FER was confined to psy- +<br/>chological studies only, but nowadays it covers a broad range of applications including +<br/>human-computer interfaces (HCI), industrial automation, surveillance systems, senti- +<br/>ment identification, etc. Precise recognition of facial expressions can become a driving +<br/>force for the future automation interfaces like car driving, robotics, driver alert systems, +<br/>etc. +<br/>According to input, expression recognition systems can be classified as static or +<br/>dynamic. In static approaches, features are computed from given still image only. +<br/>Whereas in dynamic approaches, temporal relationships between features over the +<br/>image sequence is extracted. Temporal relationships play a major role in expression +<br/>recognition from an image sequence. In last decade, many video-based methods have +<br/>been studied [2]. Research is also focused on detecting micro-expressions [2], [3], [4], +<br/>[5], recognition of spontaneous expressions [6], analysis of multi-views or profile views +<br/>[7] and fusion of geometric and appearance features [8], [9], [10]. Nowadays, deep +<br/>1249 +</td><td>('9318822', 'MAHESH GOYANI', 'mahesh goyani')<br/>('11384332', 'NARENDRA PATEL', 'narendra patel')</td><td>E-mail: mgoyani@gmail.com, nmpatel@bvmengineerring.ac.in +</td></tr><tr><td>2bb53e66aa9417b6560e588b6235e7b8ebbc294c</td><td>SEMANTIC EMBEDDING SPACE FOR ZERO-SHOT ACTION RECOGNITION +<br/><b>School of EECS, Queen Mary University of London, London, UK</b></td><td>('47158489', 'Xun Xu', 'xun xu')<br/>('2073354', 'Shaogang Gong', 'shaogang gong')</td><td></td></tr><tr><td>2b0ff4b82bac85c4f980c40b3dc4fde05d3cc23f</td><td>An Effective Approach for Facial Expression Recognition with Local Binary +<br/>Pattern and Support Vector Machine +</td><td>('20656805', 'Thi Nhan', 'thi nhan')<br/>('9872793', 'Il Choi', 'il choi')</td><td>*1School of Media, Soongsil University, ctnhen@yahoo.com +<br/>2School of Media, Soongsil University, an_tth@yahoo.com +<br/>3School of Media, Soongsil University, hic@ssu.ac.kr +</td></tr><tr><td>2b3ceb40dced78a824cf67054959e250aeaa573b</td><td></td><td></td><td></td></tr><tr><td>2be8e06bc3a4662d0e4f5bcfea45631b8beca4d0</td><td>Watch and Learn: Semi-Supervised Learning of Object Detectors From Videos +<br/><b>Robotics Institute, Carnegie Mellon University</b><br/>The availability of large labeled image datasets [1, 2] has been one of the +<br/>key factors for advances in recognition. These datasets have not only helped +<br/>boost performance, but have also fostered the development of new tech- +<br/>niques. However, compared to images, videos seem like a more natural +<br/>source of training data because of the additional temporal continuity they +<br/>offer for both learning and labeling. The available video datasets lack the +<br/>richness and variety of annotations offered by benchmark image datasets. +<br/>It also seems unlikely that human per-image labeling will scale to the web- +<br/>scale video data without using temporal constraints. In this paper, we show +<br/>how to exploit the temporal information provided by videos to enable semi- +<br/>supervised learning. +<br/>We present a scalable framework that discovers and localizes multiple ob- +<br/>jects in video using semi-supervised learning (see Figure 1). It tackles this +<br/>challenging problem in long video (a million frames in our experiments) +<br/>starting from only a few labeled examples. +<br/>In addition, we present our +<br/>algorithm in a realistic setting of sparse labels [3], i.e., in the few initial +<br/>“labeled” frames, not all objects are annotated. This setting relaxes the as- +<br/>sumption that in a given frame all object instances have been exhaustively +<br/>annotated. It also implies that we do not know if any unannotated region +<br/>in the frame is an instance of the object category or the background, and +<br/>thus cannot use any region from our input as negative data. While much of +<br/>the past work has ignored this type of sparse labeling and lack of explicit +<br/>negatives, we show ways to overcome this handicap. +<br/>Contributions: Our semi-supervised learning (SSL) framework localizes +<br/>multiple unknown objects in videos. Starting from sparsely labeled objects, +<br/>it iteratively labels new training examples in the videos. Our key contribu- +<br/>tions are: 1) We tackle the SSL problem for discovering multiple objects in +<br/>sparsely labeled videos; 2) We present an approach to constrain SSL [6] by +<br/>combining multiple weak cues in videos and exploiting decorrelated errors +<br/>by modeling data in multiple feature spaces. We demonstrate its effective- +<br/>ness as compared to traditional tracking-by-detection approaches. 3) Given +<br/>the redundancy in video data, we need a method that can automatically de- +<br/>termine the relevance of training examples to the target detection task. We +<br/>present a way to include relevance and diversity of the training examples in +<br/>each iteration of the SSL, leading to scalable incremental learning. +<br/>Our algorithm starts with a few sparsely annotated video frames (L) and +<br/>iteratively discovers new instances in the large unlabeled set of videos (U ). +<br/>Simply put, we first train detectors on annotated objects, followed by de- +<br/>tection on input videos. We determine good detections (removing confident +<br/>false positives) which serve as starting points for short-term tracking. The +<br/>short-term tracking aims to label unseen examples reliably. Amongst these +<br/>newly labeled examples, we identify diverse examples which are used to +<br/>update the detector without re-training from scratch. We iteratively repeat +<br/>this process to label new examples. We now describe our algorithm. +<br/>Sparse Annotations (lack of explicit negatives): We start with a few sparsely +<br/>annotated frames in a random subset of U . Sparse labeling implies that un- +<br/>like standard tracking-by-detection approaches, we cannot sample negatives +<br/>from the vicinity of labeled positives. We use random images from the in- +<br/>ternet as negative data for training object detectors on these sparse labels. +<br/>We use these detectors to detect objects on a subset of the video, e.g., every +<br/>30 frames. Training on a few positives without domain negatives results in +<br/>high confidence false positives. Removing such false positives is important +<br/>because if we track them, we will add many more bad training examples, +<br/>thus degrading the detector’s performance over iterations. +<br/>Temporally consistent detections: We first remove detections that are tem- +<br/>porally inconsistent using a smoothness prior on the motion of detections. +<br/>Decorrelated errors: To remove high confidence false positives, we rely +<br/>on the principle of decorrelated errors (similar to multi-view SSL [5]). The +<br/>intuition is that the detector makes mistakes that are related to its feature +</td><td>('1806773', 'Ishan Misra', 'ishan misra')<br/>('1781242', 'Abhinav Shrivastava', 'abhinav shrivastava')<br/>('1709305', 'Martial Hebert', 'martial hebert')</td><td></td></tr><tr><td>2bcec23ac1486f4106a3aa588b6589e9299aba70</td><td>An Uncertain Future: Forecasting from Static +<br/>Images using Variational Autoencoders +<br/><b>The Robotics Institute, Carnegie Mellon University</b></td><td>('14192361', 'Jacob Walker', 'jacob walker')<br/>('2786693', 'Carl Doersch', 'carl doersch')<br/>('1737809', 'Abhinav Gupta', 'abhinav gupta')<br/>('1709305', 'Martial Hebert', 'martial hebert')</td><td></td></tr><tr><td>2b773fe8f0246536c9c40671dfa307e98bf365ad</td><td>Hindawi Publishing Corporation +<br/>Computational and Mathematical Methods in Medicine +<br/>Volume 2013, Article ID 106867, 14 pages +<br/>http://dx.doi.org/10.1155/2013/106867 +<br/>Research Article +<br/>Fast Discriminative Stochastic Neighbor Embedding Analysis +<br/><b>School of Computer Science and Technology, Zhejiang University of Technology, Hangzhou 310023, China</b><br/>Received 9 February 2013; Accepted 22 March 2013 +<br/>Academic Editor: Carlo Cattani +<br/>which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. +<br/>Feature is important for many applications in biomedical signal analysis and living system analysis. A fast discriminative stochastic +<br/>neighbor embedding analysis (FDSNE) method for feature extraction is proposed in this paper by improving the existing DSNE +<br/>method. The proposed algorithm adopts an alternative probability distribution model constructed based on its K-nearest neighbors +<br/>from the interclass and intraclass samples. Furthermore, FDSNE is extended to nonlinear scenarios using the kernel trick and +<br/>then kernel-based methods, that is, KFDSNE1 and KFDSNE2. FDSNE, KFDSNE1, and KFDSNE2 are evaluated in three aspects: +<br/>visualization, recognition, and elapsed time. Experimental results on several datasets show that, compared with DSNE and MSNP, +<br/>the proposed algorithm not only significantly enhances the computational efficiency but also obtains higher classification accuracy. +<br/>1. Introduction +<br/>In recent years, dimensional reduction which can reduce the +<br/>curse of dimensionality [1] and remove irrelevant attributes in +<br/>high-dimensional space plays an increasingly important role +<br/>in many areas. It promotes the classification, visualization, +<br/>and compression of the high dimensional data. In machine +<br/>learning, dimension reduction is used to reduce the dimen- +<br/>sion by mapping the samples from the high-dimensional +<br/>space to the low-dimensional space. There are many purposes +<br/>of studying it: firstly, to reduce the amount of storage, sec- +<br/>ondly, to remove the influence of noise, thirdly, to understand +<br/>data distribution easily, and last but not least, to achieve good +<br/>results in classification or clustering. +<br/>Currently, many dimensional reduction methods have +<br/>been proposed, and they can be classified variously from dif- +<br/>ferent perspectives. Based on the nature of the input data, +<br/>they are broadly categorized into two classes: linear subspace +<br/>methods which try to find a linear subspace as feature space +<br/>so as to preserve certain kind of characteristics of observed +<br/>data, and nonlinear approaches such as kernel-based tech- +<br/>niques and geometry-based techniques; from the class labels’ +<br/>perspective, they are divided into supervised learning and +<br/>unsupervised learning; furthermore, the purpose of the for- +<br/>mer is to maximize the recognition rate between classes while +<br/>the latter is for making the minimum of information loss. In +<br/>addition, judging whether samples utilize local information +<br/>or global information, we divide them into local method and +<br/>global method. +<br/>We briefly introduce several existing dimensional reduc- +<br/>tion techniques. In the main linear techniques, principal +<br/>component analysis (PCA) [2] aims at maximizing the vari- +<br/>ance of the samples in the low-dimensional representation +<br/>with a linear mapping matrix. It is global and unsupervised. +<br/>Different from PCA, linear discriminant analysis (LDA) [3] +<br/>learns a linear projection with the assistance of class labels. +<br/>It computes the linear transformation by maximizing the +<br/>amount of interclass variance relative to the amount of intra- +<br/>class variance. Based on LDA, marginal fisher analysis (MFA) +<br/>[4], local fisher discriminant analysis (LFDA) [5], and max- +<br/>min distance analysis (MMDA) [6] are proposed. All of the +<br/>three are linear supervised dimensional reduction methods. +<br/>MFA utilizes the intrinsic graph to characterize the intraclass +<br/>compactness and uses meanwhile the penalty graph to char- +<br/>acterize interclass separability. LFDA introduces the locality +<br/>to the LFD algorithm and is particularly useful for samples +<br/>consisting of intraclass separate clusters. MMDA considers +<br/>maximizing the minimum pairwise samples of interclass. +<br/>To deal with nonlinear structural data, which can often be +<br/>found in biomedical applications [7–10], a number of nonlin- +<br/>ear approaches have been developed for dimensional reduc- +<br/>tion. Among these kernel-based techniques and geometry- +<br/>based techniques are two hot issues. Kernel-based techniques +</td><td>('1807755', 'Jianwei Zheng', 'jianwei zheng')<br/>('1767635', 'Hong Qiu', 'hong qiu')<br/>('2587047', 'Xinli Xu', 'xinli xu')<br/>('7634945', 'Wanliang Wang', 'wanliang wang')<br/>('1802128', 'Qiongfang Huang', 'qiongfang huang')<br/>('1807755', 'Jianwei Zheng', 'jianwei zheng')</td><td>Correspondence should be addressed to Jianwei Zheng; zjw@zjut.edu.cn +</td></tr><tr><td>2bab44d3a4c5ca79fb8f87abfef4456d326a0445</td><td>Player Identification in Soccer Videos +<br/><b>Dipartimento di Sistemi e Informatica, University of Florence</b><br/>Via S. Marta, 3 - 50139 Florence, Italy +</td><td>('1801509', 'Marco Bertini', 'marco bertini')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')<br/>('2308851', 'Walter Nunziati', 'walter nunziati')</td><td>bertini@dsi.unifi.it, delbimbo@dsi.unifi.it, nunziati@dsi.unifi.it +</td></tr><tr><td>2b0102d77d3d3f9bc55420d862075934f5c85bec</td><td>Slicing Convolutional Neural Network for Crowd Video Understanding +<br/><b>The Chinese University of Hong Kong</b><br/><b>The Chinese University of Hong Kong</b></td><td>('2205438', 'Jing Shao', 'jing shao')</td><td>jshao@ee.cuhk.edu.hk, ccloy@ie.cuhk.edu.hk, kkang@ee.cuhk.edu.hk, xgwang@ee.cuhk.edu.hk +</td></tr><tr><td>2b435ee691718d0b55d057d9be4c3dbb8a81526e</td><td>DREUW ET AL.: SURF-FACE RECOGNITION +<br/>SURF-Face: Face Recognition Under +<br/>Viewpoint Consistency Constraints +<br/>Human Language Technology and +<br/>Pattern Recognition +<br/><b>RWTH Aachen University</b><br/>Aachen, Germany +</td><td>('1967060', 'Philippe Dreuw', 'philippe dreuw')<br/>('2044128', 'Pascal Steingrube', 'pascal steingrube')<br/>('1804963', 'Harald Hanselmann', 'harald hanselmann')<br/>('1685956', 'Hermann Ney', 'hermann ney')</td><td>dreuw@cs.rwth-aachen.de +<br/>steingrube@cs.rwth-aachen.de +<br/>hanselmann@cs.rwth-aachen.de +<br/>ney@cs.rwth-aachen.de +</td></tr><tr><td>2b1327a51412646fcf96aa16329f6f74b42aba89</td><td>Under review as a conference paper at ICLR 2016 +<br/>IMPROVING PERFORMANCE OF RECURRENT NEURAL +<br/>NETWORK WITH RELU NONLINEARITY +<br/>Qualcomm Research +<br/>San Diego, CA 92121, USA +</td><td>('2390504', 'Sachin S. Talathi', 'sachin s. talathi')</td><td>{stalathi,avartak}@qti.qualcomm.com +</td></tr><tr><td>2b5cb5466eecb131f06a8100dcaf0c7a0e30d391</td><td>A Comparative Study of Active Appearance Model +<br/>Annotation Schemes for the Face +<br/>Face Aging Group +<br/>UNCW, USA +<br/>Face Aging Group +<br/>UNCW, USA +<br/>Face Aging Group +<br/>UNCW, USA +</td><td>('2401418', 'Amrutha Sethuram', 'amrutha sethuram')<br/>('1710348', 'Karl Ricanek', 'karl ricanek')<br/>('37804931', 'Eric Patterson', 'eric patterson')</td><td>sethurama@uncw.edu +<br/>ricanekk@uncw.edu +<br/>pattersone@uncw.edu +</td></tr><tr><td>2b64a8c1f584389b611198d47a750f5d74234426</td><td>Deblurring Face Images with Exemplars +<br/><b>Dalian University of Technology, Dalian, China</b><br/><b>University of California, Merced, USA</b></td><td>('1786024', 'Zhe Hu', 'zhe hu')<br/>('4642456', 'Zhixun Su', 'zhixun su')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td></td></tr><tr><td>2b632f090c09435d089ff76220fd31fd314838ae</td><td>Early Adaptation of Deep Priors in Age Prediction from Face Images +<br/>Computer Vision Lab +<br/>D-ITET, ETH Zurich +<br/>Computer Vision Lab +<br/>D-ITET, ETH Zurich +<br/>CVL, D-ITET, ETH Zurich +<br/>Merantix GmbH +</td><td>('35647143', 'Mahdi Hajibabaei', 'mahdi hajibabaei')<br/>('5328844', 'Anna Volokitin', 'anna volokitin')<br/>('1732855', 'Radu Timofte', 'radu timofte')</td><td>hmahdi@student.ethz.ch +<br/>voanna@vision.ee.ethz.ch +<br/>timofter@vision.ee.ethz.ch +</td></tr><tr><td>2b10a07c35c453144f22e8c539bf9a23695e85fc</td><td>Standardization of Face Image Sample Quality(cid:63) +<br/><b>University of Science and Technology of China</b><br/>Hefei 230026, China +<br/>2Center for Biometrics and Security Research & +<br/>National Laboratory of Pattern Recognition +<br/><b>Institute of Automation, Chinese Academy of Sciences, Beijing 100080, China</b><br/>http://www.cbsr.ia.ac.cn +</td><td>('39609587', 'Xiufeng Gao', 'xiufeng gao')<br/>('34679741', 'Stan Z. Li', 'stan z. li')<br/>('3168566', 'Rong Liu', 'rong liu')<br/>('2777824', 'Peiren Zhang', 'peiren zhang')</td><td></td></tr><tr><td>2b84630680e2c906f8d7ac528e2eb32c99ef203a</td><td>We are not All Equal: Personalizing Models for +<br/>Facial Expression Analysis +<br/>with Transductive Parameter Transfer +<br/><b>DISI, University of Trento, Italy</b><br/><b>DIEI, University of Perugia, Italy</b><br/>3 Fondazione Bruno Kessler (FBK), Italy +</td><td>('1716310', 'Enver Sangineto', 'enver sangineto')<br/>('2933565', 'Gloria Zen', 'gloria zen')<br/>('40811261', 'Elisa Ricci', 'elisa ricci')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')</td><td></td></tr><tr><td>2b507f659b341ed0f23106446de8e4322f4a3f7e</td><td>Deep Identity-aware Transfer of Facial Attributes +<br/><b>The Hong Kong Polytechnic University 2Harbin Institute of Technology</b></td><td>('1701799', 'Mu Li', 'mu li')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('1698371', 'David Zhang', 'david zhang')</td><td>csmuli@comp.polyu.edu.hk cswmzuo@gmail.com csdzhang@comp.polyu.edu.hk +</td></tr><tr><td>2b7ef95822a4d577021df16607bf7b4a4514eb4b</td><td>Emergence of Object-Selective Features in +<br/>Unsupervised Feature Learning +<br/>Computer Science Department +<br/><b>Stanford University</b><br/>Stanford, CA 94305 +</td><td>('5574038', 'Adam Coates', 'adam coates')<br/>('2354728', 'Andrej Karpathy', 'andrej karpathy')<br/>('1701538', 'Andrew Y. Ng', 'andrew y. ng')</td><td>{acoates,karpathy,ang}@cs.stanford.edu +</td></tr><tr><td>2b8dfbd7cae8f412c6c943ab48c795514d53c4a7</td><td>529 +<br/>2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +<br/>978-1-4799-2893-4/14/$31.00 ©2014 IEEE +<br/>RECOGNITION +<br/>1. INTRODUCTION +<br/>(d1,d2)∈[0;d]2 +<br/>d1+d2≤d +</td><td></td><td>e-mail: firstname.lastname@technicolor.com +<br/>e-mail: firstname.lastname@univ-poitiers.fr +</td></tr><tr><td>2b869d5551b10f13bf6fcdb8d13f0aa4d1f59fc4</td><td>Ring loss: Convex Feature Normalization for Face Recognition +<br/>Department of Electrical and Computer Engineering +<br/><b>Carnegie Mellon University</b></td><td>('3049981', 'Yutong Zheng', 'yutong zheng')<br/>('2628116', 'Dipan K. Pal', 'dipan k. pal')<br/>('1794486', 'Marios Savvides', 'marios savvides')</td><td>{yutongzh, dipanp, marioss}@andrew.cmu.edu +</td></tr><tr><td>2bae810500388dd595f4ebe992c36e1443b048d2</td><td>International Journal of Bioelectromagnetism +<br/>Vol. 18, No. 1, pp. 13 - 18, 2016 +<br/>www.ijbem.org +<br/>Analysis of Facial Expression Recognition +<br/>by Event-related Potentials +<br/> Department of Information and Computer Engineering, +<br/><b>National Institute of Technology, Toyota College, Japan</b><br/><b>Toyota College, 2-1 Eisei, Toyota-shi, Aichi, 471-8525 Japan</b></td><td>('2179262', 'Taichi Hayasaka', 'taichi hayasaka')<br/>('2179262', 'Taichi Hayasaka', 'taichi hayasaka')</td><td>E-mail: hayasaka@toyota-ct.ac.jp, phone +81 565 36 5861, fax +81 565 36 5926 +</td></tr><tr><td>2b42f83a720bd4156113ba5350add2df2673daf0</td><td>Action Recognition and Detection by Combining +<br/>Motion and Appearance Features +<br/><b>The Chinese University of Hong Kong</b><br/><b>Shenzhen Key Lab of CVPR, Shenzhen Institutes of Advanced Technology</b><br/>Chinese Academy of Sciences, Shenzhen, China +</td><td>('33345248', 'Limin Wang', 'limin wang')<br/>('39843569', 'Yu Qiao', 'yu qiao')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>07wanglimin@gmail.com, yu.qiao@siat.ac.cn, xtang@ie.cuhk.edu.hk +</td></tr><tr><td>2bbbbe1873ad2800954058c749a00f30fe61ab17</td><td> +<br/> ISSN(Online): 2320-9801 +<br/> ISSN (Print): 2320-9798 +<br/>International Journal of Innovative Research in Computer and Communication Engineering +<br/>(An ISO 3297: 2007 Certified Organization) +<br/>Vol.2, Special Issue 1, March 2014 +<br/>Proceedings of International Conference On Global Innovations In Computing Technology (ICGICT’14) +<br/>Organized by +<br/>Department of CSE, JayShriram Group of Institutions, Tirupur, Tamilnadu, India on 6th & 7th March 2014 +<br/>Face Verification across Ages Using Self +<br/>Organizing Map +<br/>B.Mahalakshmi1, K.Duraiswamy2, P.Gnanasuganya3, P.Aruldhevi4, R.Sundarapandiyan5 +<br/><b>K.S.Rangasamy College of Technology, Namakkal, TamilNadu, India</b><br/><b>Dean, K.S.Rangasamy College of Technology, Namakkal, TamilNadu, India</b><br/><b>B.E, K.S.Rangasamy College of Technology, Namakkal, TamilNadu, India</b></td><td></td><td></td></tr><tr><td>2baec98c19804bf19b480a9a0aa814078e28bb3d</td><td></td><td></td><td></td></tr><tr><td>47fdbd64edd7d348713253cf362a9c21f98e4296</td><td>FACIAL POINT DETECTION BASED ON A CONVOLUTIONAL NEURAL NETWORK WITH +<br/>OPTIMAL MINI-BATCH PROCEDURE +<br/><b>Chubu University</b><br/>1200, Matsumoto-cho, Kasugai, AICHI +</td><td>('2488607', 'Masatoshi Kimura', 'masatoshi kimura')<br/>('35008538', 'Yuji Yamauchi', 'yuji yamauchi')</td><td></td></tr><tr><td>47382cb7f501188a81bb2e10cfd7aed20285f376</td><td>Articulated Pose Estimation Using Hierarchical Exemplar-Based Models +<br/><b>Columbia University in the City of New York</b></td><td>('2454675', 'Jiongxin Liu', 'jiongxin liu')<br/>('3173493', 'Yinxiao Li', 'yinxiao li')</td><td>{liujx09, yli, allen, belhumeur}@cs.columbia.edu +</td></tr><tr><td>473366f025c4a6e0783e6174ca914f9cb328fe70</td><td>Review of +<br/>Action Recognition and Detection +<br/>Methods +<br/>Department of Electrical Engineering and Computer Science +<br/><b>York University</b><br/>Toronto, Ontario +<br/>Canada +</td><td>('1709096', 'Richard P. Wildes', 'richard p. wildes')</td><td></td></tr><tr><td>477236563c6a6c6db922045453b74d3f9535bfa1</td><td>International Journal of Science and Research (IJSR) +<br/>ISSN (Online): 2319-7064 +<br/>Index Copernicus Value (2013): 6.14 | Impact Factor (2014): 5.611 +<br/>Attribute Based Image Search Re-Ranking +<br/>Snehal S Patil1, Ajay Dani2 +<br/><b>Master of Computer Engg, Savitribai Phule Pune University, G. H. Raisoni Collage of Engg and Technology, Wagholi, Pune</b><br/><b>G. H .Raisoni Collage of Engg and Technology, Wagholi, Pune</b><br/>integrating +<br/>images by +</td><td></td><td></td></tr><tr><td>4793f11fbca4a7dba898b9fff68f70d868e2497c</td><td>Kinship Verification through Transfer Learning +<br/>Siyu Xia∗ +<br/>CSE, SUNY at Buffalo, USA +<br/><b>and Southeast University, China</b><br/>CSE +<br/>CSE +<br/>SUNY at Buffalo, USA +<br/>SUNY at Buffalo, USA +</td><td>('2025056', 'Ming Shao', 'ming shao')<br/>('1708679', 'Yun Fu', 'yun fu')</td><td>xsy@seu.edu.cn +<br/>mingshao@buffalo.edu +<br/>yunfu@buffalo.edu +</td></tr><tr><td>470dbd3238b857f349ebf0efab0d2d6e9779073a</td><td>Unsupervised Simultaneous Orthogonal Basis Clustering Feature Selection +<br/>School of Electrical Engineering, KAIST, South Korea +<br/>In this paper, we propose a novel unsupervised feature selection method: Si- +<br/>multaneous Orthogonal basis Clustering Feature Selection (SOCFS). To per- +<br/>form feature selection on unlabeled data effectively, a regularized regression- +<br/>based formulation with a new type of target matrix is designed. The target +<br/>matrix captures latent cluster centers of the projected data points by per- +<br/>forming the orthogonal basis clustering, and then guides the projection ma- +<br/>trix to select discriminative features. Unlike the recent unsupervised feature +<br/>selection methods, SOCFS does not explicitly use the pre-computed local +<br/>structure information for data points represented as additional terms of their +<br/>objective functions, but directly computes latent cluster information by the +<br/>target matrix conducting orthogonal basis clustering in a single unified term +<br/>of the proposed objective function. +<br/>Since the target matrix is put in a single unified term for regression of +<br/>the proposed objective function, feature selection and clustering are simul- +<br/>taneously performed. In this way, the projection matrix for feature selection +<br/>is more properly computed by the estimated latent cluster centers of the +<br/>projected data points. To the best of our knowledge, this is the first valid +<br/>formulation to consider feature selection and clustering together in a sin- +<br/>gle unified term of the objective function. The proposed objective function +<br/>has fewer parameters to tune and does not require complicated optimization +<br/>tools so just a simple optimization algorithm is sufficient. Substantial ex- +<br/>periments are performed on several publicly available real world datasets, +<br/>which shows that SOCFS outperforms various unsupervised feature selec- +<br/>tion methods and that latent cluster information by the target matrix is ef- +<br/>fective for regularized regression-based feature selection. +<br/>Problem Formulation: Given training data, let X = [x1, . . . ,xn] ∈ Rd×n +<br/>denote the data matrix with n instances where dimension is d and T = +<br/>[t1, . . . ,tn] ∈ Rm×n denote the corresponding target matrix where dimension +<br/>is m. We start from the regularized regression-based formulation to select +<br/>maximum r features is minW (cid:107)WT X− T(cid:107)2 +<br/>s.t. (cid:107)W(cid:107)2,0 ≤ r. To exploit +<br/>such formulation on unlabeled data more effectively, it is crucial for the tar- +<br/>get matrix T to have discriminative destinations for projected clusters. To +<br/>this end, a new type of target matrix T is proposed to conduct clustering di- +<br/>rectly on the projected data points WT X. We allow extra degrees of freedom +<br/>to T by decomposing it into two other matrices B ∈ Rm×c and E ∈ Rn×c as +<br/>T = BET with additional constraints as +<br/>(1) +<br/>F + λ(cid:107)W(cid:107)2,1 +<br/>(cid:107)WT X− BET(cid:107)2 +<br/>s.t. BT B = I, ET E = I, E ≥ 0, +<br/>min +<br/>W,B,E +<br/>where λ > 0 is a weighting parameter for the relaxed regularizer (cid:107)W(cid:107)2,1 +<br/>that induces row sparsity of the projection matrix W. The meanings of the +<br/>constraints BT B = I,ET E = I,E ≥ 0 are as follows: 1) the orthogonal con- +<br/>straint of B lets each column of B be independent; 2) the orthogonal and +<br/>the nonnegative constraint of E make each row of E has only one non-zero +<br/>element [2]. From 1) and 2), we can clearly interpret B as the basis matrix, +<br/>which has orthogonality and E as the encoding matrix, where the non-zero +<br/>element of each column of ET selects one column in B. +<br/>While optimizing problem (1), T = BET acts like clustering of projected +<br/>data points WT X with orthogonal basis B and encoder E, so T can estimate +<br/>latent cluster centers of the WT X. Then, W successively projects X close +<br/>to corresponding latent cluster centers, which are estimated by T. Note that +<br/>the orthogonal constraint of B makes each projected cluster in WT X be sep- +<br/>arated (independent of each other), and it helps W to be a better projection +<br/>matrix for selecting more discriminative features. If the clustering is directly +<br/>performed on X not on WT X, the orthogonal constraint of B extremely re- +<br/>stricts the degree of freedom of B. However, since features are selected by +<br/>W and the clustering is carried out on WT X in our formulation, so the or- +<br/>thogonal constraint of B is highly reasonable. A schematic illustration of +<br/>the proposed method is shown in Figure 1. +</td><td>('2086576', 'Dongyoon Han', 'dongyoon han')<br/>('1769295', 'Junmo Kim', 'junmo kim')</td><td></td></tr><tr><td>473031328c58b7461753e81251379331467f7a69</td><td>Exploring Fisher Vector and Deep Networks for Action Spotting +<br/><b>Shenzhen key lab of Comp. Vis. and Pat. Rec., Shenzhen Institutes of Advanced Technology, CAS, China</b><br/><b>The Chinese University of Hong Kong</b></td><td>('1915826', 'Zhe Wang', 'zhe wang')<br/>('33345248', 'Limin Wang', 'limin wang')<br/>('35031371', 'Wenbin Du', 'wenbin du')<br/>('33427555', 'Yu Qiao', 'yu qiao')</td><td>buptwangzhe2012@gmail.com, 07wanglimin@gmail.com, wb.du@siat.ac.cn, yu.qiao@siat.ac.cn +</td></tr><tr><td>47638197d83a8f8174cdddc44a2c7101fa8301b7</td><td>Object-Centric Anomaly Detection by Attribute-Based Reasoning +<br/><b>Rutgers University</b><br/><b>University of Washington</b><br/>Ahmed Elgammal +<br/><b>Rutgers University</b></td><td>('3139794', 'Babak Saleh', 'babak saleh')<br/>('2270286', 'Ali Farhadi', 'ali farhadi')</td><td>babaks@cs.rutgers.edu +<br/>ali@cs.uw.edu +<br/>elgammal@cs.rutgers.edu +</td></tr><tr><td>47541d04ec24662c0be438531527323d983e958e</td><td>Affective Information Processing +</td><td></td><td></td></tr><tr><td>476f177b026830f7b31e94bdb23b7a415578f9a4</td><td>INTRA-CLASS MULTI-OUTPUT REGRESSION BASED SUBSPACE ANALYSIS +<br/><b>University of California Santa Barbara</b><br/><b>University of California Santa Barbara</b></td><td>('32919393', 'Swapna Joshi', 'swapna joshi')</td><td>(cid:63){karthikeyan,swapna,manj}@ece.ucsb.edu +<br/>†{grafton}@psych.ucsb.edu +</td></tr><tr><td>474b461cd12c6d1a2fbd67184362631681defa9e</td><td>2014 IEEE International +<br/>Conference on Systems, Man +<br/>and Cybernetics +<br/>(SMC 2014) +<br/>San Diego, California, USA +<br/>5-8 October 2014 +<br/>Pages 1-789 +<br/>IEEE Catalog Number: +<br/>ISBN: +<br/>CFP14SMC-POD +<br/>978-1-4799-3841-4 +<br/>1/5 +</td><td></td><td></td></tr><tr><td>472ba8dd4ec72b34e85e733bccebb115811fd726</td><td>Cosine Similarity Metric Learning +<br/>for Face Veri cation +<br/><b>School of Computer Science, University of Nottingham</b><br/>Jubilee Campus, Wollaton Road, Nottingham, NG8 1BB, UK +<br/>http://www.nottingham.ac.uk/cs/ +</td><td>('2243665', 'Hieu V. Nguyen', 'hieu v. nguyen')<br/>('1735386', 'Li Bai', 'li bai')</td><td>{vhn,bai}@cs.nott.ac.uk +</td></tr><tr><td>47ca2df3d657d7938d7253bed673505a6a819661</td><td><b>UNIVERSITY OF CALIFORNIA</b><br/>Santa Barbara +<br/>Facial Expression Analysis on Manifolds +<br/>A Dissertation submitted in partial satisfaction of the +<br/>requirements for the degree Doctor of Philosophy +<br/>in Computer Science +<br/>by +<br/>Committee in charge: +<br/>Professor B.S. Manjunath +<br/>September 2006 +</td><td>('13303219', 'Ya Chang', 'ya chang')<br/>('1752714', 'Matthew Turk', 'matthew turk')<br/>('1706938', 'Yuan-Fang Wang', 'yuan-fang wang')<br/>('2875421', 'Andy Beall', 'andy beall')</td><td></td></tr><tr><td>47d4838087a7ac2b995f3c5eba02ecdd2c28ba14</td><td>JOURNAL OF IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, VOL. XX, NO. X, XXX 2017 +<br/>Automatic Recognition of Facial Displays of +<br/>Unfelt Emotions +<br/>Escalera, Xavier Bar´o, Sylwia Hyniewska, Member, IEEE, J¨uri Allik, +</td><td>('38370357', 'Kaustubh Kulkarni', 'kaustubh kulkarni')<br/>('22197083', 'Ciprian Adrian Corneanu', 'ciprian adrian corneanu')<br/>('22211769', 'Ikechukwu Ofodile', 'ikechukwu ofodile')<br/>('3087532', 'Gholamreza Anbarjafari', 'gholamreza anbarjafari')</td><td></td></tr><tr><td>47eba2f95679e106e463e8296c1f61f6ddfe815b</td><td>Deep Co-occurrence Feature Learning for Visual Object Recognition +<br/><b>Research Center for Information Technology Innovation, Academia Sinica</b><br/><b>National Taiwan University</b><br/><b>Graduate Institute of Electronics Engineering, National Taiwan University</b><br/><b>Smart Network System Institute, Institute for Information Industry</b></td><td>('28990604', 'Ya-Fang Shih', 'ya-fang shih')<br/>('28982867', 'Yang-Ming Yeh', 'yang-ming yeh')<br/>('1744044', 'Yen-Yu Lin', 'yen-yu lin')<br/>('34779427', 'Ming-Fang Weng', 'ming-fang weng')<br/>('2712675', 'Yi-Chang Lu', 'yi-chang lu')<br/>('37761361', 'Yung-Yu Chuang', 'yung-yu chuang')</td><td></td></tr><tr><td>47a2727bd60e43f3253247b6d6f63faf2b67c54b</td><td>Semi-supervised Vocabulary-informed Learning +<br/>Disney Research +</td><td>('35782003', 'Yanwei Fu', 'yanwei fu')<br/>('14517812', 'Leonid Sigal', 'leonid sigal')</td><td>y.fu@qmul.ac.uk, lsigal@disneyresearch.com +</td></tr><tr><td>47d3b923730746bfaabaab29a35634c5f72c3f04</td><td>ISSN : 2248-9622, Vol. 7, Issue 7, ( Part -3) July 2017, pp.30-38 +<br/>RESEARCH ARTICLE +<br/> OPEN ACCESS +<br/>Real-Time Facial Expression Recognition App Development on +<br/>Smart Phones +<br/><b>Florida Institute Of Technology, Melbourne Fl</b><br/>USA +</td><td>('7155812', 'Humaid Alshamsi', 'humaid alshamsi')<br/>('7155812', 'Humaid Alshamsi', 'humaid alshamsi')</td><td></td></tr><tr><td>47190d213caef85e8b9dd0d271dbadc29ed0a953</td><td>The Devil of Face Recognition is in the Noise +<br/>1 SenseTime Research +<br/><b>University of California San Diego</b><br/><b>Nanyang Technological University</b></td><td>('1682816', 'Fei Wang', 'fei wang')<br/>('3203648', 'Liren Chen', 'liren chen')<br/>('46651787', 'Cheng Li', 'cheng li')<br/>('1937119', 'Shiyao Huang', 'shiyao huang')<br/>('47557603', 'Yanjie Chen', 'yanjie chen')<br/>('49215552', 'Chen Qian', 'chen qian')<br/>('1717179', 'Chen Change Loy', 'chen change loy')</td><td>{wangfei, chengli, huangshiyao, chenyanjie, qianchen}@sensetime.com, +<br/>lic002@eng.ucsd.edu, ccloy@ieee.org +</td></tr><tr><td>47e3029a3d4cf0a9b0e96252c3dc1f646e750b14</td><td>International Conference on Computer Systems and Technologies - CompSysTech’07 +<br/>Facial Expression Recognition in still pictures and videos using Active +<br/>Appearance Models. A comparison approach. +<br/>Drago(cid:1) Datcu +<br/>Léon Rothkrantz +</td><td></td><td></td></tr><tr><td>475e16577be1bfc0dd1f74f67bb651abd6d63524</td><td>DAiSEE: Towards User Engagement Recognition in the Wild +<br/>Microsoft +<br/>Vineeth N Balasubramanian +<br/>Indian Institution of Technology Hyderabad +</td><td>('38330340', 'Abhay Gupta', 'abhay gupta')</td><td>abhgup@microsoft.com +<br/>vineethnb@iith.ac.in +</td></tr><tr><td>471befc1b5167fcfbf5280aa7f908eff0489c72b</td><td>570 +<br/>Class-Specific Kernel-Discriminant +<br/>Analysis for Face Verification +<br/>class problems ( +</td><td>('2123731', 'Georgios Goudelis', 'georgios goudelis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td></td></tr><tr><td>47f8b3b3f249830b6e17888df4810f3d189daac1</td><td></td><td></td><td></td></tr><tr><td>47e8db3d9adb79a87c8c02b88f432f911eb45dc5</td><td>MAGMA: Multi-level accelerated gradient mirror descent algorithm for +<br/>large-scale convex composite minimization +<br/>July 15, 2016 +</td><td>('39984225', 'Vahan Hovhannisyan', 'vahan hovhannisyan')<br/>('3112745', 'Panos Parpas', 'panos parpas')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td></td></tr><tr><td>47aeb3b82f54b5ae8142b4bdda7b614433e69b9a</td><td></td><td></td><td></td></tr><tr><td>47dabb566f2bdd6b3e4fa7efc941824d8b923a13</td><td>Probabilistic Temporal Head Pose Estimation +<br/>Using a Hierarchical Graphical Model +<br/><b>Centre for Intelligent Machines, McGill University, Montreal, Canada</b></td><td>('2515930', 'Meltem Demirkus', 'meltem demirkus')<br/>('1724729', 'Doina Precup', 'doina precup')<br/>('1713608', 'James J. Clark', 'james j. clark')<br/>('1699104', 'Tal Arbel', 'tal arbel')</td><td></td></tr><tr><td>47f5f740e225281c02c8a2ae809be201458a854f</td><td>Simultaneous Unsupervised Learning of Disparate Clusterings +<br/><b>University of Texas, Austin, TX 78712-1188, USA</b><br/>Received 14 April 2008; accepted 05 May 2008 +<br/>DOI:10.1002/sam.10007 +<br/>Published online 3 November 2008 in Wiley InterScience (www.interscience.wiley.com). +</td><td>('3164102', 'Prateek Jain', 'prateek jain')<br/>('1751621', 'Raghu Meka', 'raghu meka')<br/>('1783667', 'Inderjit S. Dhillon', 'inderjit s. dhillon')</td><td></td></tr><tr><td>47bf7a8779c68009ea56a7c20e455ccdf0e3a8fa</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 83 – No 5, December 2013 +<br/>Automatic Face Recognition System using Pattern +<br/>Recognition Techniques: A Survey +<br/>Department of Computer Science Department of Computer Science +<br/><b>Assam University, Silchar-788011 Assam University, Silchar</b></td><td>('37792796', 'Ningthoujam Sunita Devi', 'ningthoujam sunita devi')</td><td></td></tr><tr><td>47b508abdaa5661fe14c13e8eb21935b8940126b</td><td> Volume 4, Issue 12, December 2014 ISSN: 2277 128X +<br/>International Journal of Advanced Research in +<br/> Computer Science and Software Engineering +<br/> Research Paper +<br/> Available online at: www.ijarcsse.com +<br/>An Efficient Method for Feature Extraction of Face +<br/>Recognition Using PCA +<br/>(M.Tech. Student) +<br/>Computer Science & Engineering +<br/><b>Iftm University, Moradabad-244001 U.P</b></td><td>('9247488', 'Tara Prasad Singh', 'tara prasad singh')</td><td></td></tr><tr><td>477811ff147f99b21e3c28309abff1304106dbbe</td><td></td><td></td><td></td></tr><tr><td>47e14fdc6685f0b3800f709c32e005068dfc8d47</td><td></td><td></td><td></td></tr><tr><td>473cbc5ec2609175041e1410bc6602b187d03b23</td><td>Semantic Audio-Visual Data Fusion for Automatic Emotion Recognition +<br/>Man-Machine Interaction Group +<br/><b>Delft University of Technology</b><br/>2628 CD, Delft, +<br/>The Netherlands +<br/>KEYWORDS +<br/>Data fusion, automatic emotion recognition, speech analysis, +<br/>face detection, facial feature extraction, facial characteristic +<br/>point extraction, Active Appearance Models, Support Vector +<br/>Machines. +</td><td>('2866326', 'Dragos Datcu', 'dragos datcu')</td><td>E-mail: {D.Datcu ; L.J.M.Rothkrantz}@tudelft.nl +</td></tr><tr><td>782188821963304fb78791e01665590f0cd869e8</td><td></td><td></td><td></td></tr><tr><td>78a4cabf0afc94da123e299df5b32550cd638939</td><td></td><td></td><td></td></tr><tr><td>78f08cc9f845dc112f892a67e279a8366663e26d</td><td>TECHNISCHE UNIVERSIT ¨AT M ¨UNCHEN +<br/>Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +<br/>Semi-Autonomous Data Enrichment and +<br/>Optimisation for Intelligent Speech Analysis +<br/>Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +<br/>der Technischen Universit¨at M¨unchen zur Erlangung des akademischen Grades eines +<br/>Doktor-Ingenieurs (Dr.-Ing.) +<br/>genehmigten Dissertation. +<br/>Vorsitzender: +<br/>Univ.-Prof. Dr.-Ing. habil. Dr. h.c. Alexander W. Koch +<br/>Pr¨ufer der Dissertation: +<br/>1. +<br/>Univ.-Prof. Dr.-Ing. habil. Bj¨orn W. Schuller, +<br/>Universit¨at Passau +<br/>2. Univ.-Prof. Gordon Cheng, Ph.D. +<br/>Die Dissertation wurde am 30.09.2014 bei der Technischen Universit¨at M¨unchen einge- +<br/>reicht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik am 07.04.2015 +<br/>angenommen. +</td><td>('1742291', 'Zixing Zhang', 'zixing zhang')</td><td></td></tr><tr><td>78d645d5b426247e9c8f359694080186681f57db</td><td>Gender Classification by LUT Based Boosting +<br/>of Overlapping Block Patterns +<br/><b>Tampere University of Technology, Tampere, Finland</b><br/><b>Idiap Research Institute, Martigny, Switzerland</b></td><td>('3350574', 'Rakesh Mehta', 'rakesh mehta')</td><td>rakesh.mehta@tut.fi +<br/>{manuel.guenther,marcel}@idiap.ch +</td></tr><tr><td>7862d40da0d4e33cd6f5c71bbdb47377e4c6b95a</td><td>Demography-based Facial Retouching Detection using Subclass Supervised +<br/>Sparse Autoencoder +<br/><b>University of Notre Dame, 2IIIT-Delhi</b></td><td>('5014060', 'Aparna Bharati', 'aparna bharati')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('1799014', 'Kevin W. Bowyer', 'kevin w. bowyer')<br/>('1743927', 'Xin Tong', 'xin tong')</td><td>1{abharati, kwb, xtong1}@nd.edu, 2{mayank, rsingh}@iiitd.ac.in +</td></tr><tr><td>783f3fccde99931bb900dce91357a6268afecc52</td><td>Hindawi Publishing Corporation +<br/>EURASIP Journal on Image and Video Processing +<br/>Volume 2009, Article ID 945717, 14 pages +<br/>doi:10.1155/2009/945717 +<br/>Research Article +<br/>Adapted Active Appearance Models +<br/>1 SUP ´ELEC/IETR, Avenue de la Boulaie, 35511 Cesson-S´evign´e, France +<br/>2 Orange Labs—TECH/IRIS, 4 rue du clos courtel, 35 512 Cesson S´evign´e, France +<br/>Received 5 January 2009; Revised 2 September 2009; Accepted 20 October 2009 +<br/>Recommended by Kenneth M. Lam +<br/>Active Appearance Models (AAMs) are able to align efficiently known faces under duress, when face pose and illumination are +<br/>controlled. We propose Adapted Active Appearance Models to align unknown faces in unknown poses and illuminations. Our +<br/>proposal is based on the one hand on a specific transformation of the active model texture in an oriented map, which changes the +<br/>AAM normalization process; on the other hand on the research made in a set of different precomputed models related to the most +<br/>adapted AAM for an unknown face. Tests on public and private databases show the interest of our approach. It becomes possible +<br/>to align unknown faces in real-time situations, in which light and pose are not controlled. +<br/>Copyright © 2009 Renaud S´eguier et al. This is an open access article distributed under the Creative Commons Attribution +<br/>License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly +<br/>cited. +<br/>1. Introduction +<br/>All applications related to face analysis and synthesis (Man- +<br/>Machine Interaction, compression in video communication, +<br/>augmented reality) need to detect and then to align the user’s +<br/>face. This latest process consists in the precise localization of +<br/>the eyes, nose, and mouth gravity center. Face detection can +<br/>now be realized in real time and in a rather efficient manner +<br/>[1, 2]; the technical bottleneck lies now in the face alignment +<br/>when it is done in real conditions, which is precisely the +<br/>object of this paper. +<br/>Since such Active Appearance Models (AAMs) as those +<br/>described in [3] exist, it is therefore possible to align faces +<br/>in real time. The AAMs exploit a set of face examples in +<br/>order to extract a statistical model. To align an unknown +<br/>face in new image, the models parameters must be tuned, in +<br/>order to match the analyzed face features in the best possible +<br/>way. There is no difficulty to align a face featuring the same +<br/>characteristics (same morphology, illumination, and pose) +<br/>as those constituting the example data set. Unfortunately, +<br/>AAMs are less outstanding when illumination, pose, and +<br/>face type changes. We suggest in this paper a robust Active +<br/>Appearance Model allowing a real-time implementation. In +<br/>the next section, we will survey the different techniques, +<br/>which aim to increase the AAM robustness. We will see +<br/>that none of them address at the same time the three types +<br/>of robustness, we are interested in pose, illumination, and +<br/>identity. It must be pointed out that we do not consider the +<br/>robustness against occlusion as [4] does, for example, when +<br/>a person moves his hand around the face. +<br/>After a quick introduction of the Active Appearance +<br/>Models and their limitations (Section 3), we will present our +<br/>two main contributions in Section 4.1 in order to improve +<br/>AAM robustness in illumination, pose, and identity. Exper- +<br/>iments will be conducted and discussed in Section 5 before +<br/>drawing a conclusion, suggesting new research directions in +<br/>the last section. +<br/>2. State of the Art +<br/>We propose to classify the methods which lead to an increase +<br/>of the AAM robustness as follows. The specific types of +<br/>dedicated robustness are in italic. +<br/>(i) Preprocess +<br/>(1) Invariant features (illumination) +<br/>(2) Canonical representation (illumination) +<br/>(ii) Parameter space extension +<br/>(1) Light modeling (illumination) +<br/>(2) 3D modeling (pose) +</td><td>('3353560', 'Sylvain Le Gallou', 'sylvain le gallou')<br/>('40427923', 'Gaspard Breton', 'gaspard breton')<br/>('34798028', 'Christophe Garcia', 'christophe garcia')</td><td>Correspondence should be addressed to Renaud S´eguier, renaud.seguier@supelec.fr +</td></tr><tr><td>7897c8a9361b427f7b07249d21eb9315db189496</td><td></td><td></td><td></td></tr><tr><td>7859667ed6c05a467dfc8a322ecd0f5e2337db56</td><td>Web-Scale Transfer Learning for Unconstrained 1:N Face Identification +<br/>Facebook AI Research +<br/>Menlo Park, CA 94025, USA +<br/><b>Tel Aviv University</b><br/>Tel Aviv, Israel +</td><td>('2188620', 'Yaniv Taigman', 'yaniv taigman')<br/>('32447229', 'Ming Yang', 'ming yang')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td>{yaniv, mingyang, ranzato}@fb.com +<br/>wolf@cs.tau.ac.il +</td></tr><tr><td>78c1ad33772237bf138084220d1ffab800e1200d</td><td><b>State Key Laboratory of Software Development Environment, Beihang University, P.R.China</b><br/><b>University of Michigan, Ann Arbor</b></td><td>('48545182', 'Lei Huang', 'lei huang')<br/>('8342699', 'Jia Deng', 'jia deng')</td><td></td></tr><tr><td>78436256ff8f2e448b28e854ebec5e8d8306cf21</td><td>Measuring and Understanding Sensory Representations within +<br/>Deep Networks Using a Numerical Optimization Framework +<br/><b>Harvard University, Cambridge, MA</b><br/>USA +<br/><b>Center for Brain Science, Harvard University, Cambridge, MA, USA</b><br/><b>Harvard University, Cambridge, MA, USA</b></td><td>('1739108', 'Chuan-Yung Tsai', 'chuan-yung tsai')<br/>('2042941', 'David D. Cox', 'david d. cox')</td><td>∗ E-mail: davidcox@fas.harvard.edu +</td></tr><tr><td>78f438ed17f08bfe71dfb205ac447ce0561250c6</td><td></td><td></td><td></td></tr><tr><td>78f79c83b50ff94d3e922bed392737b47f93aa06</td><td>The Computer Expression Recognition Toolbox (CERT) +<br/>Mark Frank3, Javier Movellan1, and Marian Bartlett1 +<br/><b>Machine Perception Laboratory, University of California, San Diego</b><br/><b>University of Arizona</b><br/><b>University of Buffalo</b></td><td>('2724380', 'Gwen Littlewort', 'gwen littlewort')<br/>('1775637', 'Jacob Whitehill', 'jacob whitehill')<br/>('4072965', 'Tingfan Wu', 'tingfan wu')</td><td>{gwen, jake, ting, movellan}@mplab.ucsd.edu, +<br/>ianfasel@cs.arizona.edu, mfrank83@buffalo.edu, marni@salk.edu +</td></tr><tr><td>78fede85d6595e7a0939095821121f8bfae05da6</td><td>KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS VOL. 9, NO. 2, Feb. 2015 742 +<br/>Copyright © 2015 KSII +<br/>Discriminant Metric Learning Approach for +<br/>Face Verification +<br/>1 Department of Computer Science and Information Engineering +<br/><b>National Kaohsiung University of Applied Sciences, Kaohsiung, Kaohsiung, Taiwan, ROC</b><br/>2 Department of Computer Science and Information Engineering +<br/><b>National Cheng Kung University, Tainan, Taiwan, ROC</b><br/>Received September 3, 2014; revised November 12, 2014; accepted December 13, 2014; +<br/> published February 28, 2015 +</td><td>('37284667', 'Ju-Chin Chen', 'ju-chin chen')<br/>('36612683', 'Pei-Hsun Wu', 'pei-hsun wu')<br/>('3461535', 'Jenn-Jier James Lien', 'jenn-jier james lien')<br/>('37284667', 'Ju-Chin Chen', 'ju-chin chen')</td><td>[e-mail: jc.chen@cc.kuas.edu.tw] +<br/>[e-mail: jjlien@csie.ncku.edu.tw] +</td></tr><tr><td>78598e7005f7c96d64cc47ff47e6f13ae52245b8</td><td>Hand2Face: Automatic Synthesis and Recognition of Hand Over Face Occlusions +<br/>Synthetic Reality Lab +<br/>Department of Computer Science +<br/><b>University of Central Florida</b><br/>Orlando, Florida +<br/>Synthetic Reality Lab +<br/>Department of Computer Science +<br/><b>University of Central Florida</b><br/>Orlando, Florida +<br/>Tadas Baltruˇsaitis +<br/><b>Language Technology Institute</b><br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA +<br/><b>Language Technology Institute</b><br/>School of Computer Science +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA +</td><td>('2974242', 'Behnaz Nojavanasghari', 'behnaz nojavanasghari')<br/>('32827434', 'Charles E. Hughes', 'charles e. hughes')<br/>('1767184', 'Louis-Philippe Morency', 'louis-philippe morency')</td><td>Email: behnaz@eecs.ucf.edu +<br/>Email: ceh@cs.ucf.edu +<br/>Email: tbaltrus@cs.cmu.edu +<br/>Email: morency@cs.cmu.edu +</td></tr><tr><td>7862f646d640cbf9f88e5ba94a7d642e2a552ec9</td><td>Being John Malkovich +<br/><b>University of Washington</b><br/>2 Adobe Systems +<br/>3 Google Inc. +</td><td>('2419955', 'Ira Kemelmacher-Shlizerman', 'ira kemelmacher-shlizerman')<br/>('40416141', 'Aditya Sankar', 'aditya sankar')<br/>('2177801', 'Eli Shechtman', 'eli shechtman')<br/>('1679223', 'Steven M. Seitz', 'steven m. seitz')</td><td>{kemelmi,aditya,seitz}@cs.washington.edu +<br/>elishe@adobe.com +</td></tr><tr><td>78a11b7d2d7e1b19d92d2afd51bd3624eca86c3c</td><td></td><td></td><td></td></tr><tr><td>78a4eb59ec98994bebcf3a5edf9e1d34970c45f6</td><td>Conveying Shape and Features with Image-Based Relighting +<br/><b>Stanford University</b><br/><b>Stanford University</b><br/><b>Stanford University</b><br/><b>Stanford University</b><br/>Microsoft Research +<br/><b>Stanford University</b></td><td>('36475465', 'David Akers', 'david akers')<br/>('1967534', 'Frank Losasso', 'frank losasso')<br/>('37133509', 'John Rick', 'john rick')<br/>('2367620', 'Jeff Klingner', 'jeff klingner')<br/>('1820412', 'Maneesh Agrawala', 'maneesh agrawala')<br/>('1689128', 'Pat Hanrahan', 'pat hanrahan')</td><td></td></tr><tr><td>781c2553c4ed2a3147bbf78ad57ef9d0aeb6c7ed</td><td>Int J Comput Vis +<br/>DOI 10.1007/s11263-017-1023-9 +<br/>Tubelets: Unsupervised Action Proposals from Spatiotemporal +<br/>Super-Voxels +<br/>Cees G. M. Snoek1 +<br/>Received: 25 June 2016 / Accepted: 18 May 2017 +<br/>© The Author(s) 2017. This article is an open access publication +</td><td>('40027484', 'Mihir Jain', 'mihir jain')<br/>('1681054', 'Hervé Jégou', 'hervé jégou')</td><td></td></tr><tr><td>78174c2be084e67f48f3e8ea5cb6c9968615a42c</td><td>Periocular Recognition Using CNN Features +<br/>Off-the-Shelf +<br/><b>School of Information Technology (ITE), Halmstad University, Box 823, 30118 Halmstad, Sweden</b></td><td>('51446244', 'Kevin Hernandez-Diaz', 'kevin hernandez-diaz')<br/>('2847751', 'Fernando Alonso-Fernandez', 'fernando alonso-fernandez')<br/>('5058247', 'Josef Bigun', 'josef bigun')</td><td>Email: kevin.hernandez-diaz@hh.se, feralo@hh.se, josef.bigun@hh.se +</td></tr><tr><td>78df7d3fdd5c32f037fb5cc2a7c104ac1743d74e</td><td>TEMPORAL PYRAMID POOLING CNN FOR ACTION RECOGNITION +<br/>Temporal Pyramid Pooling Based Convolutional +<br/>Neural Network for Action Recognition +</td><td>('40378631', 'Peng Wang', 'peng wang')<br/>('2572430', 'Yuanzhouhan Cao', 'yuanzhouhan cao')<br/>('40529029', 'Chunhua Shen', 'chunhua shen')<br/>('2161037', 'Lingqiao Liu', 'lingqiao liu')<br/>('1724393', 'Heng Tao Shen', 'heng tao shen')</td><td></td></tr><tr><td>780557daaa39a445b24c41f637d5fc9b216a0621</td><td>Large Video Event Ontology Browsing, Search and +<br/>Tagging (EventNet Demo) +<br/><b>Columbia University, New York, NY 10027, USA</b></td><td>('2368325', 'Hongliang Xu', 'hongliang xu')<br/>('35984288', 'Guangnan Ye', 'guangnan ye')<br/>('2664705', 'Yitong Li', 'yitong li')<br/>('40313086', 'Dong Liu', 'dong liu')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>{hx2168, gy2179, yl3029, dl2713, sc250}@columbia.edu +</td></tr><tr><td>78fdf2b98cf6380623b0e20b0005a452e736181e</td><td></td><td></td><td></td></tr><tr><td>788a7b59ea72e23ef4f86dc9abb4450efefeca41</td><td></td><td></td><td></td></tr><tr><td>787c1bb6d1f2341c5909a0d6d7314bced96f4681</td><td>Face Detection and Verification in Unconstrained +<br/>Videos: Challenges, Detection, and Benchmark +<br/>Evaluation +<br/>IIIT-D-MTech-CS-GEN-13-106 +<br/>July 16, 2015 +<br/><b>Indraprastha Institute of Information Technology, Delhi</b><br/>Thesis Advisors +<br/>Dr. Mayank Vatsa +<br/>Dr. Richa Singh +<br/>Submitted in partial fulfillment of the requirements +<br/>for the Degree of M.Tech. in Computer Science +<br/>c(cid:13) Shah, 2015 +<br/>Keywords: face recognition, face detection, face verification +</td><td>('25087736', 'Mahek Shah', 'mahek shah')</td><td></td></tr><tr><td>7808937b46acad36e43c30ae4e9f3fd57462853d</td><td>Describing People: A Poselet-Based Approach to Attribute Classification ∗ +<br/>1EECS, U.C. Berkeley, Berkeley, CA 94720 +<br/><b>Adobe Systems, Inc., 345 Park Ave, San Jose, CA</b></td><td>('35208858', 'Subhransu Maji', 'subhransu maji')<br/>('1689212', 'Jitendra Malik', 'jitendra malik')</td><td>{lbourdev,smaji,malik}@eecs.berkeley.edu +</td></tr><tr><td>8b2c090d9007e147b8c660f9282f357336358061</td><td><b>Lake Forest College</b><br/><b>Lake Forest College Publications</b><br/>Senior Theses +<br/>4-23-2018 +<br/>Student Publications +<br/>Emotion Classification based on Expressions and +<br/>Body Language using Convolutional Neural +<br/>Networks +<br/>Follow this and additional works at: https://publications.lakeforest.edu/seniortheses +<br/>Part of the Neuroscience and Neurobiology Commons +<br/>Recommended Citation +<br/>Tanveer, Aasimah S., "Emotion Classification based on Expressions and Body Language using Convolutional Neural Networks" +<br/>(2018). Senior Theses. +<br/><b>This Thesis is brought to you for free and open access by the Student Publications at Lake Forest College Publications. It has been accepted for</b><br/><b>inclusion in Senior Theses by an authorized administrator of Lake Forest College Publications. For more information, please contact</b></td><td></td><td>Lake Forest College, tanveeras@lakeforest.edu +<br/>levinson@lakeforest.edu. +</td></tr><tr><td>8ba67f45fbb1ce47a90df38f21834db37c840079</td><td>People Search and Activity Mining in Large-Scale +<br/>Community-Contributed Photos +<br/><b>National Taiwan University, Taipei, Taiwan</b><br/>Winston H. Hsu, Hong-Yuan Mark Liao +<br/>Advised by +</td><td>('35081710', 'Yan-Ying Chen', 'yan-ying chen')</td><td>yanying@cmlab.csie.ntu.edu.tw +</td></tr><tr><td>8b547b87fd95c8ff6a74f89a2b072b60ec0a3351</td><td>Initial Perceptions of a Casual Game to Crowdsource +<br/>Facial Expressions in the Wild +<br/><b>Games Studio, Faculty of Engineering and IT, University of Technology, Sydney</b></td><td>('1733360', 'Chek Tien Tan', 'chek tien tan')<br/>('2117735', 'Hemanta Sapkota', 'hemanta sapkota')<br/>('2823535', 'Daniel Rosser', 'daniel rosser')<br/>('3141633', 'Yusuf Pisan', 'yusuf pisan')</td><td>chek@gamesstudio.org +<br/>hemanta.sapkota@student.uts.edu.au +<br/>daniel.j.rosser@gmail.com +<br/>yusuf.pisan@gamesstudio.org +</td></tr><tr><td>8bed7ff2f75d956652320270eaf331e1f73efb35</td><td>Emotion Recognition in the Wild using +<br/>Deep Neural Networks and Bayesian Classifiers +<br/>Elena Ba(cid:138)ini S¨onmez +<br/><b>University of Calabria - DeMACS</b><br/>Via Pietro Bucci +<br/>Rende (CS), Italy +<br/><b>Plymouth University - CRNS</b><br/>Portland Square PL4 8AA +<br/>Plymouth, United Kingdom +<br/>ac.uk +<br/><b>Istanbul Bilgi University - DCE</b><br/>Eski Silahtaraa Elektrik Santral Kazm +<br/>Karabekir Cad. No: 2/13 34060 Eyp +<br/>Istanbul, Turkey +<br/><b>University of Calabria - DeMACS</b><br/>Via Pietro Bucci +<br/>Rende (CS), Italy +<br/><b>Plymouth University - CRNS</b><br/>Portland Square PL4 8AA +<br/>Plymouth, United Kingdom +</td><td>('32751441', 'Luca Surace', 'luca surace')<br/>('3366919', 'Massimiliano Patacchiola', 'massimiliano patacchiola')<br/>('3205804', 'William Spataro', 'william spataro')<br/>('1692929', 'Angelo Cangelosi', 'angelo cangelosi')</td><td>lucasurace11@gmail.com +<br/>massimiliano.patacchiola@plymouth. +<br/>ebsonmez@bilgi.edu.tr +<br/>william.spataro@unical.it +<br/>angelo.cangelosi@plymouth.ac.uk +</td></tr><tr><td>8b7191a2b8ab3ba97423b979da6ffc39cb53f46b</td><td>Search Pruning in Video Surveillance Systems: Efficiency-Reliability Tradeoff +<br/>EURECOM +<br/>Sophia Antipolis, France +</td><td>('3299530', 'Antitza Dantcheva', 'antitza dantcheva')<br/>('1688531', 'Petros Elia', 'petros elia')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td>{Antitza.Dantcheva, Arun.Singh, Petros.Elia, Jean-Luc.Dugelay}@eurecom.fr +</td></tr><tr><td>8bf57dc0dd45ed969ad9690033d44af24fd18e05</td><td>Subject-Independent Emotion Recognition from Facial Expressions +<br/>using a Gabor Feature RBF Neural Classifier Trained with Virtual +<br/>Samples Generated by Concurrent Self-Organizing Maps +<br/>VICTOR-EMIL NEAGOE, ADRIAN-DUMITRU CIOTEC +<br/>Depart. Electronics, Telecommunications & Information Technology +<br/><b>Polytechnic University of Bucharest</b><br/>Splaiul Independentei No. 313, Sector 6, Bucharest, +<br/>ROMANIA +</td><td></td><td>victoremil@gmail.com, adryyandc@yahoo.com +</td></tr><tr><td>8bf243817112ac0aa1348b40a065bb0b735cdb9c</td><td>LEARNING A REPRESSION NETWORK FOR PRECISE VEHICLE SEARCH +<br/><b>Institute of Digital Media</b><br/><b>School of Electrical Engineering and Computer Science, Peking University</b><br/>No.5 Yiheyuan Road, 100871, Beijing, China +</td><td>('17872416', 'Qiantong Xu', 'qiantong xu')<br/>('13318784', 'Ke Yan', 'ke yan')<br/>('1705972', 'Yonghong Tian', 'yonghong tian')</td><td>{xuqiantong, keyan, yhtian}@pku.edu.cn +</td></tr><tr><td>8bfada57140aa1aa22a575e960c2a71140083293</td><td>Can we match Ultraviolet Face Images against their Visible +<br/>Counterparts? +<br/><b>aMILab, LCSEE, West Virginia University, Morgantown, West Virginia, USA</b></td><td>('33240042', 'Neeru Narang', 'neeru narang')<br/>('1731727', 'Thirimachos Bourlai', 'thirimachos bourlai')<br/>('1678573', 'Lawrence A. Hornak', 'lawrence a. hornak')<br/>('11898042', 'Paul D. Coverdell', 'paul d. coverdell')</td><td></td></tr><tr><td>8b8728edc536020bc4871dc66b26a191f6658f7c</td><td></td><td></td><td></td></tr><tr><td>8befcd91c24038e5c26df0238d26e2311b21719a</td><td>A Joint Sequence Fusion Model for Video +<br/>Question Answering and Retrieval +<br/>Department of Computer Science and Engineering, +<br/><b>Seoul National University, Seoul, Korea</b><br/>http://vision.snu.ac.kr/projects/jsfusion/ +</td><td>('7877122', 'Youngjae Yu', 'youngjae yu')<br/>('2175130', 'Jongseok Kim', 'jongseok kim')</td><td>{yj.yu,js.kim}@vision.snu.ac.kr, gunhee@snu.ac.kr +</td></tr><tr><td>8bbbdff11e88327816cad3c565f4ab1bb3ee20db</td><td>Automatic Semantic Face Recognition +<br/><b>University of Southampton</b><br/>Southampton, United Kingdom +</td><td>('19249411', 'Nawaf Yousef Almudhahka', 'nawaf yousef almudhahka')<br/>('1727698', 'Mark S. Nixon', 'mark s. nixon')<br/>('31534955', 'Jonathon S. Hare', 'jonathon s. hare')</td><td>{nya1g14,msn,jsh2}@ecs.soton.ac.uk +</td></tr><tr><td>8bdf6f03bde08c424c214188b35be8b2dec7cdea</td><td>Exploiting Unintended Feature Leakage in Collaborative Learning∗ +<br/>UCL +<br/><b>Cornell University</b><br/><b>UCL and Alan Turing Institute</b><br/>Cornell Tech +</td><td>('2008164', 'Luca Melis', 'luca melis')<br/>('3469125', 'Congzheng Song', 'congzheng song')<br/>('1728207', 'Emiliano De Cristofaro', 'emiliano de cristofaro')<br/>('1723945', 'Vitaly Shmatikov', 'vitaly shmatikov')</td><td>luca.melis.14@alumni.ucl.ac.uk +<br/>cs2296@cornell.edu +<br/>e.decristofaro@ucl.ac.uk +<br/>shmat@cs.cornell.edu +</td></tr><tr><td>8b744786137cf6be766778344d9f13abf4ec0683</td><td>978-1-4799-9988-0/16/$31.00 ©2016 IEEE +<br/>2697 +<br/>ICASSP 2016 +</td><td></td><td></td></tr><tr><td>8b10383ef569ea0029a2c4a60cc2d8c87391b4db</td><td>ZHOU,MILLERANDZHANG:AGECLASSIFICATIONUSINGRADONTRANSFORM... +<br/>Age classification using Radon transform +<br/>and entropy based scaling SVM +<br/>Paul Miller1 +<br/><b>The Institute of Electronics</b><br/>Communications +<br/>and Information Technology +<br/><b>Queen s University Belfast</b><br/>2 School of Computing +<br/><b>University of Dundee</b><br/>United Kingdom +</td><td>('2040772', 'Huiyu Zhou', 'huiyu zhou')<br/>('1744844', 'Jianguo Zhang', 'jianguo zhang')</td><td>h.zhou@ecit.qub.ac.uk +<br/>p.miller@ecit.qub.ac.uk +<br/>jgzhang@computing.dundee.ac.uk +</td></tr><tr><td>8b30259a8ab07394d4dac971f3d3bd633beac811</td><td>Representing Sets of Instances for Visual Recognition +<br/>1 National Key Laboratory for Novel Software Technology +<br/><b>Nanjing University, China</b><br/>2 Minieye, Youjia Innovation LLC, China +</td><td>('1808816', 'Jianxin Wu', 'jianxin wu')<br/>('2226422', 'Bin-Bin Gao', 'bin-bin gao')<br/>('15527784', 'Guoqing Liu', 'guoqing liu')</td><td>∗ wujx2001@nju.edu.cn, gaobb@lamda.nju.edu.cn +<br/>guoqing@minieye.cc +</td></tr><tr><td>8b61fdc47b5eeae6bc0a52523f519eaeaadbc8c8</td><td>HU, LIU, LI, LIU: TEMPORAL PERCEPTIVE NETWORK FOR ACTION RECOGNITION +<br/>Temporal Perceptive Network for +<br/>Skeleton-Based Action Recognition +<br/><b>Institute of Computer Science and</b><br/>Technology +<br/><b>Peking University</b><br/>Beijing, China +<br/>Sijie Song +</td><td>('9956463', 'Yueyu Hu', 'yueyu hu')<br/>('49046516', 'Chunhui Liu', 'chunhui liu')<br/>('3128506', 'Yanghao Li', 'yanghao li')<br/>('41127426', 'Jiaying Liu', 'jiaying liu')</td><td>huyy@pku.edu.cn +<br/>liuchunhui@pku.edu.cn +<br/>lyttonhao@pku.edu.cn +<br/>ssj940920@pku.edu.cn +<br/>liujiaying@pku.edu.cn +</td></tr><tr><td>8b19efa16a9e73125ab973429eb769d0ad5a8208</td><td>SCAR: Dynamic adaptation for person detection and +<br/>persistence analysis in unconstrained videos +<br/>Department of Computer Science +<br/><b>Stevens Institute of Technology</b><br/>Hoboken, NJ 07030, USA +</td><td>('2789357', 'George Kamberov', 'george kamberov')<br/>('3219999', 'Matt Burlick', 'matt burlick')<br/>('2283008', 'Lazaros Karydas', 'lazaros karydas')<br/>('3228177', 'Olga Koteoglou', 'olga koteoglou')</td><td>gkambero,mburlick,lkarydas,okoteogl@stevens.edu (cid:63) +</td></tr><tr><td>8b6fded4d08bf0b7c56966b60562ee096af1f0c4</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 59– No.3, December 2012 +<br/>A Neural Network based Facial Expression Recognition +<br/>using Fisherface +<br/>Department of Mathematics +<br/><b>Semarang State University</b><br/>Semarang, 50229, Indonesia +<br/> +</td><td>('39807349', 'Zaenal Abidin', 'zaenal abidin')</td><td></td></tr><tr><td>8bf647fed40bdc9e35560021636dfb892a46720e</td><td>Learning to Hash-tag Videos with Tag2Vec +<br/>CVIT, KCIS, IIIT Hyderabad, India +<br/>P J Narayanan +<br/>http://cvit.iiit.ac.in/research/projects/tag2vec +<br/>Figure 1. Learning a direct mapping from videos to hash-tags : sample frames from short video clips with user-given hash-tags +<br/>(left); a sample frame from a query video and hash-tags suggested by our system for this query (right). +</td><td>('2461059', 'Aditya Singh', 'aditya singh')<br/>('3448416', 'Saurabh Saini', 'saurabh saini')<br/>('1962817', 'Rajvi Shah', 'rajvi shah')</td><td>{(aditya.singh,saurabh.saini,rajvi.shah)@research.,pjn@}iiit.ac.in +</td></tr><tr><td>8b2704a5218a6ef70e553eaf0a463bd55129b69d</td><td>Sensors 2013, 13, 7714-7734; doi:10.3390/s130607714 +<br/>OPEN ACCESS +<br/>sensors +<br/>ISSN 1424-8220 +<br/>www.mdpi.com/journal/sensors +<br/>Article +<br/>Geometric Feature-Based Facial Expression Recognition in +<br/>Image Sequences Using Multi-Class AdaBoost and Support +<br/>Vector Machines +<br/><b>Division of Computer Engineering, Chonbuk National University, Jeonju-si, Jeollabuk-do</b><br/>Tel.: +82-63-270-2406; Fax: +82-63-270-2394. +<br/>Received: 3 May 2013; in revised form: 29 May 2013 / Accepted: 3 June 2013 / +<br/>Published: 14 June 2013 +</td><td>('32322842', 'Deepak Ghimire', 'deepak ghimire')<br/>('2034182', 'Joonwhoan Lee', 'joonwhoan lee')</td><td>Korea; E-Mail: deep@jbnu.ac.kr +<br/>* Author to whom correspondence should be addressed; E-Mail: chlee@jbnu.ac.kr; +</td></tr><tr><td>8bb21b1f8d6952d77cae95b4e0b8964c9e0201b0</td><td>Methoden +<br/>at 11/2013 +<br/>(cid:2)(cid:2)(cid:2) +<br/>Multimodale Interaktion +<br/>auf einer sozialen Roboterplattform +<br/>Multimodal Interaction on a Social Robotic Platform +<br/>Zusammenfassung Dieser Beitrag beschreibt die multimo- +<br/>dalen Interaktionsmöglichkeiten mit der Forschungsroboter- +<br/>plattform ELIAS. Zunächst wird ein Überblick über die Ro- +<br/>boterplattform sowie die entwickelten Verarbeitungskompo- +<br/>nenten gegeben, die Einteilung dieser Komponenten erfolgt +<br/>nach dem Konzept von wahrnehmenden und agierenden Mo- +<br/>dalitäten. Anschließend wird das Zusammenspiel der Kom- +<br/>ponenten in einem multimodalen Spieleszenario näher be- +<br/>trachtet. (cid:2)(cid:2)(cid:2) Summary +<br/>This paper presents the mul- +<br/>timodal +<br/>interaction capabilities of the robotic research plat- +<br/>form ELIAS. An overview of the robotic platform as well +<br/>as the developed processing components is presented, the +<br/>classification of the components follows the concept of sen- +<br/>sing and acting modalities. Finally, +<br/>the interplay between +<br/>those components within a multimodal gaming scenario is +<br/>described. +<br/>Schlagwörter Mensch-Roboter-Interaktion, Multimodalität, Gesten, Blick (cid:2)(cid:2)(cid:2) Keywords Human-robot interaction, +<br/>multimodal, gestures, gaze +<br/>1 Einleitung +<br/>Eine intuitive und natürliche Bedienbarkeit der zuneh- +<br/>mend komplexeren Technik wird für den Menschen +<br/>immer wichtiger, da im heutigen Alltag eine Vielzahl an +<br/>technischen Geräten mit wachsendem Funktionsumfang +<br/>anzutreffen ist. Unterschiedliche Aktivitäten in der For- +<br/>schungsgemeinschaft haben sich schon seit längerer Zeit +<br/>mit verbalen sowie nonverbalen Kommunikationsformen +<br/>(bspw. Emotions- und Gestenerkennung) in der Mensch- +<br/>Maschine-Interaktion beschäftigt. Gerade in der jüngeren +<br/>Zeit trugen auf diesem Forschungsfeld unterschiedliche +<br/>Innovationen (bspw. Touchscreen, Gestensteuerung im +<br/>Fernseher) dazu bei, dass intuitive und natürliche Bedien- +<br/>konzepte mehr und mehr im Alltag Verwendung finden. +<br/>Auch Möglichkeiten zur Sprach- und Gestensteuerung +<br/>von Konsolen und Mobiltelefonen finden heute vermehr- +<br/>ten Einsatz in der Gerätebedienung. Diese natürlicheren +<br/>und multimodalen Benutzerschnittstellen sind dem Nut- +<br/>zer schnell zugänglich und erlauben eine intuitivere +<br/>Interaktion mit komplexen technischen Geräten. +<br/>Auch für Robotersysteme bietet sich eine multimodale +<br/>Interaktion an, um die Benutzung und den Zugang zu +<br/>den Funktionalitäten zu vereinfachen. Der Mensch soll +<br/>in seiner Kommunikation idealerweise vollkommene Ent- +<br/>scheidungsfreiheit bei der Wahl der Modalitäten haben, +<br/>um sein gewünschtes Ziel zu erreichen. Dafür werden +<br/>in diesem Beitrag die wahrnehmenden und agieren- +<br/>den Modalitäten einer, rein auf Kommunikationsaspekte +<br/>reduzierten, Forschungsroboterplattform beispielhaft in +<br/>einer Spieleanwendung untersucht. +<br/>1.1 Struktur des Beitrags +<br/>In diesem Beitrag wird zunächst ein kurzer Über- +<br/>blick über die multimodale Interaktion im Allgemeinen +<br/>gegeben, hierbei erfolgt eine Betrachtung nach wahr- +<br/>nehmenden und agierenden Modalitäten. Im nächsten +<br/>Abschnitt werden Arbeiten vorgestellt, die sich auch mit +<br/>multimodalen Robotersystemen beschäftigen. Im darauf +<br/>folgenden Abschnitt wird die Roboterplattform ELIAS +<br/>mit den wahrnehmenden, verarbeitenden und agierenden +<br/>at – Automatisierungstechnik 61 (2013) 11 / DOI 10.1515/auto.2013.1062 © Oldenbourg Wissenschaftsverlag +<br/> - 10.1515/auto.2013.1062 +<br/>Downloaded from De Gruyter Online at 09/27/2016 10:08:34PM +<br/>via Technische Universität München +<br/>737 +</td><td>('35116429', 'Jürgen Blume', 'jürgen blume')<br/>('1682283', 'Tobias Rehrl', 'tobias rehrl')<br/>('1705843', 'Gerhard Rigoll', 'gerhard rigoll')</td><td>Korrespondenzautor: blume@tum.de +</td></tr><tr><td>8b1db0894a23c4d6535b5adf28692f795559be90</td><td>Biometric and Surveillance Technology for Human and Activity Identification X, edited by Ioannis Kakadiaris, +<br/>Walter J. Scheirer, Laurence G. Hassebrook, Proc. of SPIE Vol. 8712, 87120Q · © 2013 SPIE +<br/>CCC code: 0277-786X/13/$18 · doi: 10.1117/12.2018974 +<br/>Proc. of SPIE Vol. 8712 87120Q-1 +</td><td></td><td></td></tr><tr><td>8b2e3805b37c18618b74b243e7a6098018556559</td><td>Workshop track - ICLR 2018 +<br/>IMPROVING VARIATIONAL AUTOENCODER WITH DEEP +<br/><b>University of Nottingham, Nottingham, UK</b><br/><b>Shenzhen University, Shenzhen, China</b></td><td>('3468964', 'Xianxu Hou', 'xianxu hou')<br/>('1698461', 'Guoping Qiu', 'guoping qiu')</td><td>xianxu.hou@nottingham.edu.cn +<br/>guoping.qiu@nottingham.ac.uk +</td></tr><tr><td>8b74252625c91375f55cbdd2e6415e752a281d10</td><td>Using Convolutional 3D Neural Networks for +<br/>User-Independent Continuous Gesture Recognition +<br/>Necati Cihan Camgoz, Simon Hadfield +<br/><b>University of Surrey</b><br/>Guildford, GU2 7XH, UK +<br/>Human Technology & Pattern Recognition +<br/><b>RWTH Aachen University, Germany</b><br/><b>University of Surrey</b><br/>Guildford, GU2 7XH, UK +</td><td>('2309364', 'Oscar Koller', 'oscar koller')<br/>('1695195', 'Richard Bowden', 'richard bowden')</td><td>{n.camgoz, s.hadfield}@surrey.ac.uk +<br/>koller@cs.rwth-aachen.de +<br/>r.bowden@surrey.ac.uk +</td></tr><tr><td>8b38124ff02a9cf8ad00de5521a7f8a9fa4d7259</td><td>Real-time 3D Face Fitting and Texture Fusion +<br/>on In-the-wild Videos +<br/>Centre for Vision, Speech and Signal Processing +<br/>Image Understanding and Interactive Robotics +<br/><b>University of Surrey</b><br/>Guildford, GU2 7XH, United Kingdom +<br/>Contact: http://www.patrikhuber.ch +<br/><b>Reutlingen University</b><br/>D-72762 Reutlingen, Germany +</td><td>('39976184', 'Patrik Huber', 'patrik huber')<br/>('49759031', 'William Christmas', 'william christmas')<br/>('1748684', 'Josef Kittler', 'josef kittler')<br/>('49330989', 'Philipp Kopp', 'philipp kopp')</td><td></td></tr><tr><td>134f1cee8408cca648d8b4ca44b38b0a7023af71</td><td>Partially Shared Multi-Task Convolutional Neural Network with Local +<br/>Constraint for Face Attribute Learning +<br/><b>College of Information Science and Electronic Engineering</b><br/><b>Zhejiang University, China</b></td><td>('41021477', 'Jiajiong Cao', 'jiajiong cao')<br/>('2367491', 'Yingming Li', 'yingming li')<br/>('1720488', 'Zhongfei Zhang', 'zhongfei zhang')</td><td>{jiajiong, yingming, zhongfei}@zju.edu.cn +</td></tr><tr><td>13719bbb4bb8bbe0cbcdad009243a926d93be433</td><td>Deep LDA-Pruned Nets for Efficient Facial Gender Classification +<br/><b>McGill University</b><br/><b>University Street, Montral, QC H3A 0E9, Canada</b></td><td>('1992537', 'Qing Tian', 'qing tian')<br/>('1699104', 'Tal Arbel', 'tal arbel')<br/>('1713608', 'James J. Clark', 'james j. clark')</td><td>{qtian,arbel,clark}@cim.mcgill.ca +</td></tr><tr><td>134db6ca13f808a848321d3998e4fe4cdc52fbc2</td><td>IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 36, NO. 2, APRIL 2006 +<br/>433 +<br/>Dynamics of Facial Expression: Recognition of +<br/>Facial Actions and Their Temporal Segments +<br/>From Face Profile Image Sequences +</td><td>('1694605', 'Maja Pantic', 'maja pantic')<br/>('1744405', 'Ioannis Patras', 'ioannis patras')</td><td></td></tr><tr><td>133dd0f23e52c4e7bf254e8849ac6f8b17fcd22d</td><td>This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI +<br/>Active Clustering with Model-Based +<br/>Uncertainty Reduction +</td><td>('2228109', 'Caiming Xiong', 'caiming xiong')<br/>('34187462', 'David M. Johnson', 'david m. johnson')<br/>('3587688', 'Jason J. Corso', 'jason j. corso')</td><td></td></tr><tr><td>1329206dbdb0a2b9e23102e1340c17bd2b2adcf5</td><td>Part-based R-CNNs for +<br/>Fine-grained Category Detection +<br/><b>University of California, Berkeley</b></td><td>('40565777', 'Ning Zhang', 'ning zhang')<br/>('1753210', 'Trevor Darrell', 'trevor darrell')</td><td>{nzhang,jdonahue,rbg,trevor}@eecs.berkeley.edu +</td></tr><tr><td>1369e9f174760ea592a94177dbcab9ed29be1649</td><td>Geometrical Facial Modeling for Emotion Recognition +</td><td>('3250085', 'Giampaolo L. Libralon', 'giampaolo l. libralon')</td><td></td></tr><tr><td>133900a0e7450979c9491951a5f1c2a403a180f0</td><td>JOURNAL OF LATEX CLASS FILES +<br/>Social Grouping for Multi-target Tracking and +<br/>Head Pose Estimation in Video +</td><td>('12561781', 'Zhen Qin', 'zhen qin')<br/>('3564227', 'Christian R. Shelton', 'christian r. shelton')</td><td></td></tr><tr><td>13bda03fc8984d5943ed8d02e49a779d27c84114</td><td>Efficient Object Detection Using Cascades of Nearest Convex Model Classifiers +<br/><b>Eskisehir Osmangazi University</b><br/>Laboratoire Jean Kuntzmann +<br/>Meselik Kampusu, 26480, Eskisehir Turkey +<br/>B.P. 53, 38041 Grenoble Cedex 9, France +</td><td>('2277308', 'Hakan Cevikalp', 'hakan cevikalp')<br/>('1756114', 'Bill Triggs', 'bill triggs')</td><td>hakan.cevikalp@gmail.com +<br/>Bill.Triggs@imag.fr +</td></tr><tr><td>13db9466d2ddf3c30b0fd66db8bfe6289e880802</td><td>I.J. Image, Graphics and Signal Processing, 2017, 1, 27-32 +<br/>Published Online January 2017 in MECS (http://www.mecs-press.org/) +<br/>DOI: 10.5815/ijigsp.2017.01.04 +<br/>Transfer Subspace Learning Model for Face +<br/>Recognition at a Distance +<br/>MIT, Pune ,India +<br/>AISSM’S IOT,India +<br/><b>College of Engineering Pune, India</b><br/>learning algorithms work +</td><td>('3335915', 'Alwin Anuse', 'alwin anuse')<br/>('32032353', 'Vibha Vyas', 'vibha vyas')</td><td>Email: alwin.anuse@mitpune.edu.in +<br/>Email: deshmukhnilima@gmail.com +<br/>Email: vsv.extc@coep.ac.in +</td></tr><tr><td>13a994d489c15d440c1238fc1ac37dad06dd928c</td><td>Learning Discriminant Face Descriptor for Face +<br/>Recognition +<br/>Center for Biometrics and Security Research & National Laboratory of Pattern +<br/><b>Recognition, Institute of Automation, Chinese Academy of Sciences</b></td><td>('1718623', 'Zhen Lei', 'zhen lei')<br/>('34679741', 'Stan Z. Li', 'stan z. li')</td><td>fzlei,szlig@nlpr.ia.ac.cn +</td></tr><tr><td>131178dad3c056458e0400bed7ee1a36de1b2918</td><td>Visual Reranking through Weakly Supervised Multi-Graph Learning +<br/><b>Xidian University, Xi an, China</b><br/><b>Xiamen University, Xiamen, China</b><br/><b>IBM Watson Research Center, Armonk, NY, USA</b><br/><b>University of Technology, Sydney, Australia</b></td><td>('1715156', 'Cheng Deng', 'cheng deng')<br/>('1725599', 'Rongrong Ji', 'rongrong ji')<br/>('39059457', 'Wei Liu', 'wei liu')<br/>('1692693', 'Dacheng Tao', 'dacheng tao')<br/>('10699750', 'Xinbo Gao', 'xinbo gao')</td><td>{chdeng.xd, jirongrong, wliu.cu, dacheng.tao, xbgao.xidian}@gmail.com +</td></tr><tr><td>13141284f1a7e1fe255f5c2b22c09e32f0a4d465</td><td>Object Tracking by +<br/>Oversampling Local Features +</td><td>('2619131', 'Federico Pernici', 'federico pernici')<br/>('8196487', 'Alberto Del Bimbo', 'alberto del bimbo')</td><td></td></tr><tr><td>132527383890565d18f1b7ad50d76dfad2f14972</td><td>JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 22, 1033-1046 (2006) +<br/>Facial Expression Classification Using PCA and +<br/>Hierarchical Radial Basis Function Network* +<br/>Department of Computer Science and Information Engineering +<br/><b>National Taipei University</b><br/>Sanshia, 237 Taiwan +<br/>Intelligent human-computer interaction (HCI) integrates versatile tools such as per- +<br/>ceptual recognition, machine learning, affective computing, and emotion cognition to +<br/>enhance the ways humans interact with computers. Facial expression analysis is one of +<br/>the essential medium of behavior interpretation and emotion modeling. In this paper, we +<br/>modify and develop a reconstruction method utilizing Principal Component Analysis +<br/>(PCA) to perform facial expression recognition. A framework of hierarchical radial basis +<br/>function network (HRBFN) is further proposed to classify facial expressions based on +<br/>local features extraction by PCA technique from lips and eyes images. It decomposes the +<br/>acquired data into a small set of characteristic features. The objective of this research is +<br/>to develop a more efficient approach to discriminate between seven prototypic facial ex- +<br/>pressions, such as neutral, smile, anger, surprise, fear, disgust, and sadness. A construc- +<br/>tive procedure is detailed and the system performance is evaluated on a public database +<br/>“Japanese Females Facial Expression (JAFFE).” We conclude that local images of lips +<br/>and eyes can be treated as cues for facial expression. As anticipated, the experimental +<br/>results demonstrate the potential capabilities of the proposed approach. +<br/>Keywords: intelligent human-computer interaction, facial expression classification, hier- +<br/>archical radial basis function network, principal component analysis, local features +<br/>1. INTRODUCTION +<br/>The intelligent human-computer interaction (HCI) technologies play important roles +<br/>in the development of advanced and ambient communication/computation. In contrast to +<br/>the conventional mechanisms of passive manipulation, intelligent HCI integrates versa- +<br/>tile tools such as perceptual recognition, machine learning, affective computing, and +<br/>emotion cognition to enhance the ways humans interact with computers. Migrating from +<br/>W4 (what, where, when, who) to W5+ (what, where, when, who, why, how), novel intel- +<br/>ligent interface design has placed emphasis on both apparent and internal behavior of +<br/>users [1]. Nonverbal information such as facial expression, posture, gesture, and eye gaze +<br/>is suitable for behavior interpretation. Facial data analysis is one of the essential medium +<br/>of perceptual processing and emotion modeling. +<br/>Received August 16, 2005; accepted January 17, 2006. +<br/>Communicated by Jhing-Fa Wang, Pau-Choo Chung and Mark Billinghurst. +<br/>* This work was supported in part by the National Science Council of Taiwan, R.O.C., under grants No. NSC +<br/>88-2213-E216-010 and No. NSC 89-2213-E216-016. +<br/>* The preliminary content of this paper has been presented in “International Conference on Neural Information +<br/>Processing,” Perth, Australia, November 1999. Acknowledgement also due to Mr. Der-Chen Pan at the Na- +<br/><b>tional Taipei University for his help in performing simulations. The author would like to thank Mr. Ming</b><br/>Shon Chen at Ulead System Inc., Taipei, Taiwan, for his early work and assistance in this research. +<br/>1033 +</td><td>('39548632', 'Daw-Tung Lin', 'daw-tung lin')</td><td></td></tr><tr><td>13604bbdb6f04a71dea4bd093794e46730b0a488</td><td>Robust Loss Functions under Label Noise for +<br/>Deep Neural Networks +<br/>Microsoft, Bangalore +<br/><b>Indian Institute of Science, Bangalore</b><br/><b>Indian Institute of Science, Bangalore</b></td><td>('3201314', 'Aritra Ghosh', 'aritra ghosh')<br/>('47602083', 'Himanshu Kumar', 'himanshu kumar')<br/>('1711348', 'P. S. Sastry', 'p. s. sastry')</td><td>arghosh@microsoft.com +<br/>himanshukr@ee.iisc.ernet.in +<br/>sastry@ee.iisc.ernet.in +</td></tr><tr><td>1394ca71fc52db972366602a6643dc3e65ee8726</td><td>See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/308407783 +<br/>EmoReact: A Multimodal Approach and Dataset +<br/>for Recognizing Emotional Responses in Children +<br/>Conference Paper · November 2016 +<br/>DOI: 10.1145/2993148.2993168 +<br/>CITATIONS +<br/>READS +<br/>95 +<br/>4 authors, including: +<br/>Behnaz Nojavanasghari +<br/><b>University of Central Florida</b><br/>4 PUBLICATIONS 20 CITATIONS +<br/>Tadas Baltrusaitis +<br/><b>Carnegie Mellon University</b><br/>30 PUBLICATIONS 247 CITATIONS +<br/>SEE PROFILE +<br/>SEE PROFILE +<br/>Charles E. Hughes +<br/><b>University of Central Florida</b><br/>185 PUBLICATIONS 1,248 CITATIONS +<br/>SEE PROFILE +<br/>All in-text references underlined in blue are linked to publications on ResearchGate, +<br/>letting you access and read them immediately. +<br/>Available from: Behnaz Nojavanasghari +<br/>Retrieved on: 13 October 2016 +</td><td></td><td></td></tr><tr><td>137aa2f891d474fce1e7a1d1e9b3aefe21e22b34</td><td>Is the Eye Region More Reliable Than the Face? A Preliminary Study of +<br/>Face-based Recognition on a Transgender Dataset +<br/><b>Institute of Interdisciplinary Studies in Identity Sciences (IISIS</b><br/><b>University of North Carolina Wilmington</b></td><td>('1805620', 'Gayathri Mahalingam', 'gayathri mahalingam')<br/>('3275890', 'Karl Ricanek', 'karl ricanek')</td><td>{mahalingamg, ricanekk}@uncw.edu +</td></tr><tr><td>13b1b18b9cfa6c8c44addb9a81fe10b0e89db32a</td><td>A Hierarchical Deep Temporal Model for +<br/>Group Activity Recognition +<br/>by +<br/><b>B. Tech., Indian Institute of Technology Jodhpur</b><br/>Thesis Submitted in Partial Fulfillment of the +<br/>Requirements for the Degree of +<br/>Master of Science +<br/>in the +<br/>School of Computing Science +<br/>Faculty of Applied Sciences +<br/><b>SIMON FRASER UNIVERSITY</b><br/>Spring 2016 +<br/>However, in accordance with the Copyright Act of Canada, this work may be +<br/>reproduced without authorization under the conditions for “Fair Dealing.” +<br/>Therefore, limited reproduction of this work for the purposes of private study, +<br/>research, education, satire, parody, criticism, review and news reporting is likely +<br/>All rights reserved. +<br/>to be in accordance with the law, particularly if cited appropriately. +</td><td>('2716937', 'Srikanth Muralidharan', 'srikanth muralidharan')<br/>('2716937', 'Srikanth Muralidharan', 'srikanth muralidharan')</td><td></td></tr><tr><td>1329bcac5ebd0b08ce33ae1af384bd3e7a0deaca</td><td>Dataset Issues in Object Recognition +<br/>J. Ponce1,2, T.L. Berg3, M. Everingham4, D.A. Forsyth1, M. Hebert5, +<br/>S. Lazebnik1, M. Marszalek6, C. Schmid6, B.C. Russell7, A. Torralba7, +<br/>C.K.I. Williams8, J. Zhang6, and A. Zisserman4 +<br/><b>University of Illinois at Urbana-Champaign, USA</b><br/>2 Ecole Normale Sup´erieure, Paris, France +<br/><b>University of California at Berkeley, USA</b><br/><b>Oxford University, UK</b><br/><b>Carnegie Mellon University, Pittsburgh, USA</b><br/>6 INRIA Rhˆone-Alpes, Grenoble, France +<br/>7 MIT, Cambridge, USA +<br/><b>University of Edinburgh, Edinburgh, UK</b></td><td></td><td></td></tr><tr><td>133da0d8c7719a219537f4a11c915bf74c320da7</td><td>International Journal of Computer Applications (0975 – 8887) +<br/>Volume 123 – No.4, August 2015 +<br/>A Novel Method for 3D Image Segmentation with Fusion +<br/>of Two Images using Color K-means Algorithm +<br/>Dept. of CSE +<br/>ITM Universe +<br/>Gwalior +<br/>Dept. of CSE +<br/>ITM Universe +<br/>Gwalior +<br/>two +</td><td></td><td></td></tr><tr><td>13c250fb740cb5616aeb474869db6ab11560e2a6</td><td>LEARNING LANGUAGE–VISION CORRESPONDENCES +<br/>by +<br/>A thesis submitted in conformity with the requirements +<br/>for the degree of Doctor of Philosophy +<br/>Graduate Department of Computer Science +<br/><b>University of Toronto</b></td><td>('38986168', 'Michael Jamieson', 'michael jamieson')<br/>('38986168', 'Michael Jamieson', 'michael jamieson')</td><td></td></tr><tr><td>13940d0cc90dbf854a58f92d533ce7053aac024a</td><td><b>Boston University</b><br/>OpenBU +<br/>Theses & Dissertations +<br/>http://open.bu.edu +<br/><b>Boston University Theses and Dissertations</b><br/>2015 +<br/>Local learning by partitioning +<br/>http://hdl.handle.net/2144/15204 +<br/><b>Boston University</b></td><td>('2870611', 'Wang', 'wang')<br/>('17099457', 'Joseph', 'joseph')</td><td></td></tr><tr><td>133f01aec1534604d184d56de866a4bd531dac87</td><td>Effective Unconstrained Face Recognition by +<br/>Combining Multiple Descriptors and Learned +<br/>Background Statistics +</td><td>('1776343', 'Lior Wolf', 'lior wolf')<br/>('1756099', 'Tal Hassner', 'tal hassner')<br/>('2188620', 'Yaniv Taigman', 'yaniv taigman')</td><td></td></tr><tr><td>131bfa2ae6a04fd3b921ccb82b1c3f18a400a9c1</td><td>Elastic Graph Matching versus Linear Subspace +<br/>Methods for Frontal Face Verification +<br/>Dept. of Informatics +<br/><b>Aristotle University of Thessaloniki, Box 451, 54124 Thessaloniki, Greece</b><br/>Tel: +30-2310-996361, Fax: +30-2310-998453 +</td><td>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1737071', 'Anastasios Tefas', 'anastasios tefas')<br/>('1698588', 'Ioannis Pitas', 'ioannis pitas')</td><td>E-mail: pitas@zeus.csd.auth.gr +</td></tr><tr><td>13841d54c55bd74964d877b4b517fa94650d9b65</td><td>Generalised Ambient Reflection Models for Lambertian and +<br/>Phong Surfaces +<br/>Author +<br/>Zhang, Paul, Gao, Yongsheng +<br/>Published +<br/>2009 +<br/>Conference Title +<br/>Proceedings of the 2009 IEEE International Conference on Image Processing (ICIP 2009) +<br/>DOI +<br/>https://doi.org/10.1109/ICIP.2009.5413812 +<br/>Copyright Statement +<br/>© 2009 IEEE. Personal use of this material is permitted. However, permission to reprint/ +<br/>republish this material for advertising or promotional purposes or for creating new collective +<br/>works for resale or redistribution to servers or lists, or to reuse any copyrighted component of +<br/>this work in other works must be obtained from the IEEE. +<br/>Downloaded from +<br/>http://hdl.handle.net/10072/30001 +<br/>Griffith Research Online +<br/>https://research-repository.griffith.edu.au +</td><td></td><td></td></tr><tr><td>1389ba6c3ff34cdf452ede130c738f37dca7e8cb</td><td>A Convolution Tree with Deconvolution Branches: Exploiting Geometric +<br/>Relationships for Single Shot Keypoint Detection +<br/>Department of Electrical and Computer Engineering, CFAR and UMIACS +<br/><b>University of Maryland-College Park, USA</b></td><td>('40080979', 'Amit Kumar', 'amit kumar')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td>akumar14@umiacs.umd.edu, rama@umiacs.umd.edu +</td></tr><tr><td>131e395c94999c55c53afead65d81be61cd349a4</td><td></td><td></td><td></td></tr><tr><td>1384a83e557b96883a6bffdb8433517ec52d0bea</td><td></td><td></td><td></td></tr><tr><td>13fd0a4d06f30a665fc0f6938cea6572f3b496f7</td><td></td><td></td><td></td></tr><tr><td>132f88626f6760d769c95984212ed0915790b625</td><td>UC Irvine +<br/>UC Irvine Electronic Theses and Dissertations +<br/>Title +<br/>Exploring Entity Resolution for Multimedia Person Identification +<br/>Permalink +<br/>https://escholarship.org/uc/item/9t59f756 +<br/>Author +<br/>Zhang, Liyan +<br/>Publication Date +<br/>2014-01-01 +<br/>Peer reviewed|Thesis/dissertation +<br/>eScholarship.org +<br/>Powered by the California Digital Library +<br/><b>University of California</b></td><td></td><td></td></tr><tr><td>13aef395f426ca8bd93640c9c3f848398b189874</td><td>Image Preprocessing and Complete 2DPCA with Feature +<br/>Extraction for Gender Recognition +<br/>NSF REU 2017: Statistical Learning and Data Mining +<br/><b>University of North Carolina Wilmington</b></td><td></td><td></td></tr><tr><td>13f6ab2f245b4a871720b95045c41a4204626814</td><td>RESEARCH ARTICLE +<br/>Cortex commands the performance of +<br/>skilled movement +<br/><b>Janelia Research Campus, Howard Hughes Medical Institute, Ashburn, United</b><br/>States +</td><td>('13837962', 'Jian-Zhong Guo', 'jian-zhong guo')<br/>('35466277', 'Austin R Graves', 'austin r graves')<br/>('31262308', 'Wendy W Guo', 'wendy w guo')<br/>('12009815', 'Jihong Zheng', 'jihong zheng')<br/>('3031589', 'Allen Lee', 'allen lee')<br/>('38033405', 'Nuo Li', 'nuo li')<br/>('40634144', 'John J Macklin', 'john j macklin')<br/>('34447371', 'James W Phillips', 'james w phillips')<br/>('1875164', 'Brett D Mensh', 'brett d mensh')<br/>('2424812', 'Kristin Branson', 'kristin branson')<br/>('5832202', 'Adam W Hantman', 'adam w hantman')</td><td></td></tr><tr><td>13be4f13dac6c9a93f969f823c4b8c88f607a8c4</td><td>Families in the Wild (FIW): Large-Scale Kinship Image +<br/>Database and Benchmarks +<br/>Dept. of Electrical and Computer Engineering +<br/><b>Northeastern University</b><br/>Boston, MA, USA +</td><td>('14802538', 'Joseph P. Robinson', 'joseph p. robinson')<br/>('2025056', 'Ming Shao', 'ming shao')<br/>('1746738', 'Yue Wu', 'yue wu')<br/>('1708679', 'Yun Fu', 'yun fu')</td><td>{jrobins1, mingshao, yuewu, yunfu}@ece.neu.edu +</td></tr><tr><td>13afc4f8d08f766479577db2083f9632544c7ea6</td><td>Multiple Kernel Learning for +<br/>Emotion Recognition in the Wild +<br/>Machine Perception Laboratory +<br/>UCSD +<br/>EmotiW Challenge, ICMI, 2013 +<br/>1 +</td><td>('39707211', 'Karan Sikka', 'karan sikka')<br/>('1963167', 'Karmen Dykstra', 'karmen dykstra')<br/>('1924458', 'Suchitra Sathyanarayana', 'suchitra sathyanarayana')<br/>('2724380', 'Gwen Littlewort', 'gwen littlewort')</td><td></td></tr><tr><td>13188a88bbf83a18dd4964e3f89d0bc0a4d3a0bd</td><td><b>HOD, St. Joseph College of Information Technology, Songea, Tanzania</b></td><td></td><td></td></tr><tr><td>13d9da779138af990d761ef84556e3e5c1e0eb94</td><td>Int J Comput Vis (2008) 77: 3–24 +<br/>DOI 10.1007/s11263-007-0093-5 +<br/>Learning to Locate Informative Features for Visual Identification +<br/>Received: 18 August 2005 / Accepted: 11 September 2007 / Published online: 9 November 2007 +<br/>© Springer Science+Business Media, LLC 2007 +</td><td>('3236352', 'Andras Ferencz', 'andras ferencz')<br/>('1689212', 'Jitendra Malik', 'jitendra malik')</td><td></td></tr><tr><td>1316296fae6485c1510f00b1b57fb171b9320ac2</td><td>FaceID-GAN: Learning a Symmetry Three-Player GAN +<br/>for Identity-Preserving Face Synthesis +<br/><b>CUHK - SenseTime Joint Lab, The Chinese University of Hong Kong</b><br/>2SenseTime Research +<br/><b>Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences</b></td><td>('8035201', 'Yujun Shen', 'yujun shen')<br/>('47571885', 'Ping Luo', 'ping luo')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>{sy116, pluo, xtang}@ie.cuhk.edu.hk, yanjunjie@sensetime.com, xgwang@ee.cuhk.edu.hk +</td></tr><tr><td>7f57e9939560562727344c1c987416285ef76cda</td><td>Accessorize to a Crime: Real and Stealthy Attacks on +<br/>State-of-the-Art Face Recognition +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA, USA +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA, USA +<br/><b>Carnegie Mellon University</b><br/>Pittsburgh, PA, USA +<br/><b>University of North Carolina</b><br/>Chapel Hill, NC, USA +</td><td>('36301492', 'Mahmood Sharif', 'mahmood sharif')<br/>('38572260', 'Lujo Bauer', 'lujo bauer')<br/>('38181360', 'Sruti Bhagavatula', 'sruti bhagavatula')<br/>('1746214', 'Michael K. Reiter', 'michael k. reiter')</td><td>mahmoods@cmu.edu +<br/>lbauer@cmu.edu +<br/>srutib@cmu.edu +<br/>reiter@cs.unc.edu +</td></tr><tr><td>7fc5b6130e9d474dfb49d9612b6aa0297d481c8e</td><td>Dimensionality Reduction on Grassmannian via Riemannian +<br/>Optimization: +<br/>A Generalized Perspective +<br/><b>Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang 110016, China</b><br/><b>University of Chinese Academy of Sciences, Beijing, 100049, China</b><br/>3Key Laboratory of Optical-Electronics Information Processing +<br/>November 20, 2017 +</td><td>('1803285', 'Tianci Liu', 'tianci liu')<br/>('2172914', 'Zelin Shi', 'zelin shi')<br/>('2556853', 'Yunpeng Liu', 'yunpeng liu')</td><td></td></tr><tr><td>7f511a6a2b38a26f077a5aec4baf5dffc981d881</td><td>LOW-LATENCY HUMAN ACTION RECOGNITION WITH WEIGHTED MULTI-REGION +<br/>CONVOLUTIONAL NEURAL NETWORK +<br/><b>cid:63)University of Science and Technology of China, Hefei, Anhui, China</b><br/>†HERE Technologies, Chicago, Illinois, USA +</td><td>('49417387', 'Yunfeng Wang', 'yunfeng wang')<br/>('38272296', 'Wengang Zhou', 'wengang zhou')<br/>('46324995', 'Qilin Zhang', 'qilin zhang')<br/>('49897466', 'Xiaotian Zhu', 'xiaotian zhu')<br/>('7179232', 'Houqiang Li', 'houqiang li')</td><td></td></tr><tr><td>7f21a7441c6ded38008c1fd0b91bdd54425d3f80</td><td>Real Time System for Facial Analysis +<br/><b>Tampere University of Technology, Finland</b><br/>I. +<br/>INTRODUCTION +<br/>Most signal or image processing algorithms should be +<br/>designed with real-time execution in mind. Most use cases +<br/>compute on an embedded platform while receiving streaming +<br/>data as a constant data flow. In machine learning, however, the +<br/>real time deployment and streaming data processing are less +<br/>often a design criterion. Instead, the bulk of machine learning is +<br/>executed offline on the cloud without any real time restrictions. +<br/>However, the real time use is rapidly becoming more important +<br/>as deep learning systems are appearing into, for example, +<br/>autonomous vehicles and working machines. +<br/>In this work, we describe the functionality of our demo +<br/>system integrating a number of common real time machine +<br/>learning systems together. The demo system consists of a +<br/>screen, webcam and a computer, and it estimates the age, +<br/>gender and facial expression of all faces seen by the webcam. +<br/>A picture of the system in use is shown in Figure 1. There is +<br/>also a Youtube video at https://youtu.be/Kfe5hKNwrCU and +<br/>the code is freely available at https://github.com/mahehu/TUT- +<br/>live-age-estimator. +<br/>Apart from serving as an illustrative example of modern +<br/>human level machine learning for the general public, the +<br/>system also highlights several aspects that are common in real +<br/>time machine learning systems. First, the subtasks needed to +<br/>achieve the three recognition results represent a wide variety of +<br/>machine learning problems: (1) object detection is used to find +<br/>the faces, (2) age estimation represents a regression problem +<br/>with a real-valued target output (3) gender prediction is a +<br/>binary classification problem, and (4) facial expression +<br/>prediction is a multi-class classification problem. Moreover, all +<br/>these tasks should operate in unison, such that each task will +<br/>receive enough resources from a limited pool. +<br/>In the remainder of this paper, we first describe the system +<br/>level multithreaded architecture for real time processing in +<br/>Section II. This is followed by detailed discussion individual +<br/>components of the system in Section III. Next, we report +<br/>experimental results on the accuracy of each individual +<br/>recognition component in Section IV, and finally, discuss the +<br/>benefits of demonstrating the potential of modern machine +<br/>learning to both general public and experts in the field. +<br/>II. SYSTEM LEVEL FUNCTIONALITY +<br/>The challenge in real-time operation is that there are +<br/>numerous components in the system, and each uses different +<br/>amount of execution time. The system should be designed +<br/>such that the operation appears smooth, which means that the +<br/>most visible tasks should be fast and have the priority in +<br/>scheduling. +<br/>Figure 1. Demo system recognizes the age, gender and facial +<br/>expression in real time. +<br/>The system is running in threads, as illustrated in Figure 2. +<br/>The whole system is controlled by the upper level controller +<br/>and visualization thread, which owns and starts the sub- +<br/>threads dedicated for individual tasks. The main thread holds +<br/>all data and executes the visualization loop showing the +<br/>recognition results to the user at 25 frames per second. +<br/>The recognition process starts from the grabber thread, +<br/>which is connected to a webcam. The thread requests video +<br/>frames from camera for feeding them into a FIFO buffer +<br/>located inside the controller thread. At grab time, each frame is +<br/>wrapped inside a class object, which holds the necessary meta +<br/>data related to each frame. More specifically, each frame is +<br/>linked with a timestamp and a flag indicating whether the face +<br/>detection has already been executed and +<br/>locations +<br/>(bounding boxes) of all found faces in the scene. +<br/>the +<br/>The actual face analysis consists of two parts: face +<br/>detection and face analysis. The detection is executed in the +<br/>detection thread, which operates asynchronously, requesting +<br/>new non-processed frames from the controller thread. After +<br/>face detection, the locations of found faces are sent to the +<br/>controller thread, which then matches each new face with all +<br/>face objects from the previous frames using straightforward +<br/>centroid tracking. Tracking allows us to average the estimates +<br/>for each face over a number of recent frames. +<br/>The detection thread operates on the average faster than the +<br/>frame rate, but sometimes there are delays due to high load on +<br/>the other threads. Therefore, the controller thread holds a +<br/>buffer of the most recent frames, in order to increase the +<br/>flexibility of processing time. +<br/>The recognition thread is responsible for assessing the age, +<br/>gender and facial expression of each face crop found from the +<br/>image. The thread operates also in an asynchronous mode, +<br/>requesting new non-processed (but face-detected) frames from +</td><td>('51232696', 'Janne Tommola', 'janne tommola')<br/>('51149972', 'Pedram Ghazi', 'pedram ghazi')<br/>('51131997', 'Bishwo Adhikari', 'bishwo adhikari')<br/>('1847889', 'Heikki Huttunen', 'heikki huttunen')</td><td></td></tr><tr><td>7fce5769a7d9c69248178989a99d1231daa4fce9</td><td>(IJACSA) International Journal of Advanced Computer Science and Applications, +<br/>Vol. 7, No. 5, 2016 +<br/>Towards Face Recognition Using Eigenface +<br/>Department of Computer Engineering +<br/><b>King Faisal University</b><br/>Hofuf, Al-Ahsa 31982, Saudi Arabia +</td><td>('39604645', 'Md. Al-Amin Bhuiyan', 'md. al-amin bhuiyan')</td><td></td></tr><tr><td>7fa2605676c589a7d1a90d759f8d7832940118b5</td><td>A New Approach to Clothing Classification using Mid-Level Layers +<br/>Department of Electrical and Computer Engineering +<br/><b>Clemson University, Clemson, SC</b></td><td>('2181472', 'Bryan Willimon', 'bryan willimon')</td><td>{rwillim,iwalker,stb}@clemson.edu +</td></tr><tr><td>7ff42ee09c9b1a508080837a3dc2ea780a1a839b</td><td>Data Fusion for Real-time Multimodal Emotion Recognition through Webcams +<br/>and Microphones in E-Learning +<br/><b>Welten Institute, Research Centre for Learning, Teaching and Technology, Faculty of</b><br/><b>Psychology and Educational Sciences, Open University of the Netherlands, Valkenburgerweg</b><br/>177, 6419 AT Heerlen, The Netherlands +</td><td>('2565070', 'Kiavash Bahreini', 'kiavash bahreini')<br/>('1717772', 'Rob Nadolski', 'rob nadolski')<br/>('3235367', 'Wim Westera', 'wim westera')</td><td>{kiavash.bahreini, rob.nadolski, wim.westera}@ou.nl +</td></tr><tr><td>7fb5006b6522436ece5bedf509e79bdb7b79c9a7</td><td>Multi-Task Convolutional Neural Network for Face Recognition +<br/>Department of Computer Science and Engineering +<br/><b>Michigan State University, East Lansing MI</b></td><td>('2399004', 'Xi Yin', 'xi yin')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td>{yinxi1,liuxm}@msu.edu +</td></tr><tr><td>7f533bd8f32525e2934a66a5b57d9143d7a89ee1</td><td>Audio-Visual Identity Grounding for Enabling Cross Media Search +<br/>Paper ID 22 +</td><td>('1950685', 'Kevin Brady', 'kevin brady')</td><td></td></tr><tr><td>7f44f8a5fd48b2d70cc2f344b4d1e7095f4f1fe5</td><td>Int J Comput Vis (2016) 119:60–75 +<br/>DOI 10.1007/s11263-015-0839-4 +<br/>Sparse Output Coding for Scalable Visual Recognition +<br/>Received: 15 May 2013 / Accepted: 16 June 2015 / Published online: 26 June 2015 +<br/>© Springer Science+Business Media New York 2015 +</td><td>('1729034', 'Bin Zhao', 'bin zhao')</td><td></td></tr><tr><td>7f4bc8883c3b9872408cc391bcd294017848d0cf</td><td> +<br/> +<br/>Computer +<br/>Sciences +<br/>Department +<br/>The Multimodal Focused Attribute Model: A Nonparametric +<br/>Bayesian Approach to Simultaneous Object Classification and +<br/>Attribute Discovery +<br/>Technical Report #1697 +<br/>January 2012 +<br/> +</td><td>('6256616', 'Jake Rosin', 'jake rosin')<br/>('1724754', 'Charles R. Dyer', 'charles r. dyer')<br/>('1832364', 'Xiaojin Zhu', 'xiaojin zhu')</td><td></td></tr><tr><td>7f6061c83dc36633911e4d726a497cdc1f31e58a</td><td>YouTube-8M: A Large-Scale Video Classification +<br/>Benchmark +<br/>Paul Natsev +<br/>Google Research +</td><td>('2461984', 'Sami Abu-El-Haija', 'sami abu-el-haija')<br/>('1805076', 'George Toderici', 'george toderici')<br/>('32575647', 'Nisarg Kothari', 'nisarg kothari')<br/>('2119006', 'Joonseok Lee', 'joonseok lee')<br/>('2758088', 'Balakrishnan Varadarajan', 'balakrishnan varadarajan')<br/>('2259154', 'Sudheendra Vijayanarasimhan', 'sudheendra vijayanarasimhan')</td><td>haija@google.com +<br/>gtoderici@google.com +<br/>ndk@google.com +<br/>joonseok@google.com +<br/>natsev@google.com +<br/>balakrishnanv@google.com +<br/>svnaras@google.com +</td></tr><tr><td>7fa3d4be12e692a47b991c0b3d3eba3a31de4d05</td><td>Efficient Online Spatio-Temporal Filtering +<br/>for Video Event Detection +<br/>1 Department of Computer Science and Engineering, +<br/><b>Shanghai Jiao Tong University, Shanghai 200240, China</b><br/>2 School of Electrical and Electronic Engineering, +<br/><b>Nanyang Technological University, Singapore 639798, Singapore</b><br/>3 Computer Science and Engineering Division, +<br/><b>University of Michigan</b><br/>Ann Arbor, MI 48105, USA +</td><td>('3084614', 'Xinchen Yan', 'xinchen yan')<br/>('34316743', 'Junsong Yuan', 'junsong yuan')<br/>('2574445', 'Hui Liang', 'hui liang')</td><td>skywalkeryxc@gmail.com +<br/>jsyuan@ntu.edu.sg, hliang1@e.ntu.edu.sg +</td></tr><tr><td>7f445191fa0475ff0113577d95502a96dc702ef9</td><td>Towards an Unequivocal Representation of Actions +<br/><b>University of Bristol</b><br/><b>University of Bristol</b><br/><b>University of Bristol</b></td><td>('2052236', 'Michael Wray', 'michael wray')<br/>('3420479', 'Davide Moltisanti', 'davide moltisanti')<br/>('1728459', 'Dima Damen', 'dima damen')</td><td>firstname.surname@bristol.ac.uk +</td></tr><tr><td>7f82f8a416170e259b217186c9e38a9b05cb3eb4</td><td>Multi-Attribute Robust Component Analysis for Facial UV Maps +<br/><b>Imperial College London, London, UK</b><br/><b>Middlesex University London, London, UK</b><br/><b>Goldsmiths, University of London, London, UK</b></td><td>('24278037', 'Stylianos Moschoglou', 'stylianos moschoglou')<br/>('31243357', 'Evangelos Ververas', 'evangelos ververas')<br/>('1780393', 'Yannis Panagakis', 'yannis panagakis')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')</td><td>{s.moschoglou, e.ververas16, i.panagakis, s.zafeiriou}@imperial.ac.uk, m.nicolaou@gold.ac.uk +</td></tr><tr><td>7f36dd9ead29649ed389306790faf3b390dc0aa2</td><td>MOVEMENT DIFFERENCES BETWEEN DELIBERATE +<br/>AND SPONTANEOUS FACIAL EXPRESSIONS: +<br/>ZYGOMATICUS MAJOR ACTION IN SMILING +</td><td>('2059653', 'Zara Ambadar', 'zara ambadar')</td><td></td></tr><tr><td>7f6cd03e3b7b63fca7170e317b3bb072ec9889e0</td><td>A Face Recognition Signature Combining Patch-based +<br/>Features with Soft Facial Attributes +<br/>L. Zhang, P. Dou, I.A. Kakadiaris +<br/>Computational Biomedicine Lab, 4849 Calhoun Rd, Rm 373, Houston, TX 77204 +</td><td></td><td></td></tr><tr><td>7fab17ef7e25626643f1d55257a3e13348e435bd</td><td>Age Progression/Regression by Conditional Adversarial Autoencoder +<br/><b>The University of Tennessee, Knoxville, TN, USA</b></td><td>('1786391', 'Zhifei Zhang', 'zhifei zhang')<br/>('46970616', 'Yang Song', 'yang song')<br/>('1698645', 'Hairong Qi', 'hairong qi')</td><td>{zzhang61, ysong18, hqi}@utk.edu +</td></tr><tr><td>7f6599e674a33ed64549cd512ad75bdbd28c7f6c</td><td>Kernel Alignment Inspired +<br/>Linear Discriminant Analysis +<br/>Department of Computer Science and Engineering, +<br/><b>University of Texas at Arlington, TX, USA</b></td><td>('1747268', 'Shuai Zheng', 'shuai zheng')</td><td>zhengs123@gmail.com, chqding@uta.edu +</td></tr><tr><td>7f9260c00a86a0d53df14469f1fa10e318ee2a3c</td><td>HOW IRIS RECOGNITION WORKS +<br/><b>University of Cambridge, The Computer Laboratory, Cambridge CB3 0FD, U.K</b></td><td>('1781325', 'John Daugman', 'john daugman')</td><td></td></tr><tr><td>7f97a36a5a634c30de5a8e8b2d1c812ca9f971ae</td><td>Incremental Classifier Learning with Generative Adversarial Networks +<br/><b>Northeastern University 2Microsoft Research 3City University of New York</b></td><td>('1746738', 'Yue Wu', 'yue wu')<br/>('1691128', 'Zicheng Liu', 'zicheng liu')</td><td>{yuewu,yunfu}@ece.neu.edu, yye@gradcenter.cuny.edu +<br/>{yiche,lijuanw,zliu,yandong.guo,zhang}@microsoft.com +</td></tr><tr><td>7f2a4cd506fe84dee26c0fb41848cb219305173f</td><td>International Journal of Hybrid Information Technology +<br/>Vol.8, No.2 (2015), pp.109-120 +<br/>http://dx.doi.org/10.14257/ijhit.2015.8.2.10 +<br/>Face Detection and Pose Estimation Based on Evaluating Facial +<br/>Feature Selection +<br/><b>School of Information Science and Engineering, Central South University, Changsha</b><br/>410083, China +<br/><b>Huazhong University of</b><br/>Science and Technology, Wuhan, China +<br/><b>Collage of Sciences, Baghdad University, Iraq</b></td><td>('2759156', 'Hiyam Hatem', 'hiyam hatem')<br/>('2742321', 'Mohammed Lutf', 'mohammed lutf')<br/>('2462860', 'Jumana Waleed', 'jumana waleed')</td><td>hiamhatim2005@yahoo.com, bjzou@vip.163.com, aed.m.muttasher@gmail.com, +<br/>jumana_waleed@yahoo.com, mohammed.lutf@gmail.com1 +</td></tr><tr><td>7fd700f4a010d765c506841de9884df394c1de1c</td><td>Correlational Spectral Clustering +<br/><b>Max Planck Institute for Biological Cybernetics</b><br/>72076 T¨ubingen, Germany +</td><td>('1758219', 'Matthew B. Blaschko', 'matthew b. blaschko')<br/>('1787591', 'Christoph H. Lampert', 'christoph h. lampert')</td><td>{blaschko,chl}@tuebingen.mpg.de +</td></tr><tr><td>7f59657c883f77dc26393c2f9ed3d19bdf51137b</td><td><b>University of Wollongong</b><br/>Research Online +<br/>Faculty of Informatics - Papers (Archive) +<br/>Faculty of Engineering and Information Sciences +<br/>2006 +<br/>Facial expression recognition for multiplayer online +<br/>games +<br/>Publication Details +<br/>Zhan, C., Li, W., Ogunbona, P. O. & Safaei, F. (2006). Facial expression recognition for multiplayer online games. Joint International +<br/><b>Conference on CyberGames and Interactive Entertainment (pp. 52-58). Western Australia: Murdoch university</b><br/>Research Online is the open access institutional repository for the +<br/><b>University of Wollongong. For further information contact the UOW</b></td><td>('3283367', 'Ce Zhan', 'ce zhan')<br/>('1685696', 'Wanqing Li', 'wanqing li')<br/>('1719314', 'Philip O. Ogunbona', 'philip o. ogunbona')<br/>('1803733', 'Farzad Safaei', 'farzad safaei')</td><td>University of Wollongong, czhan@uow.edu.au +<br/>University of Wollongong, wanqing@uow.edu.au +<br/>University of Wollongong, philipo@uow.edu.au +<br/>University of Wollongong, farzad@uow.edu.au +<br/>Library: research-pubs@uow.edu.au +</td></tr><tr><td>7f23a4bb0c777dd72cca7665a5f370ac7980217e</td><td>Improving Person Re-identification by Attribute and Identity Learning +<br/><b>University of Technology Sydney</b></td><td>('9919679', 'Yutian Lin', 'yutian lin')<br/>('14904242', 'Liang Zheng', 'liang zheng')<br/>('7435343', 'Zhedong Zheng', 'zhedong zheng')<br/>('1887625', 'Yu Wu', 'yu wu')<br/>('1698559', 'Yi Yang', 'yi yang')</td><td>yutianlin477,liangzheng06,zdzheng12,wu08yu,yee.i.yang@gmail.com +</td></tr><tr><td>7f268f29d2c8f58cea4946536f5e2325777fa8fa</td><td>Facial Emotion Recognition in Curvelet Domain +<br/><b>Indian Institute of Informaiton Technology, Allahabad, India</b><br/>Allahabad, India - 211012 +</td><td>('35077572', 'Gyanendra K Verma', 'gyanendra k verma')<br/>('30102998', 'Bhupesh Kumar Singh', 'bhupesh kumar singh')</td><td>gyanendra@iiita.ac.in , rs65@iiita.ac.in +</td></tr><tr><td>7fc3442c8b4c96300ad3e860ee0310edb086de94</td><td>Similarity Scores based on Background Samples +<br/><b>The School of Computer Science, Tel-Aviv University, Israel</b><br/><b>Computer Science Division, The Open University of Israel, Israel</b><br/>3 face.com +</td><td>('1776343', 'Lior Wolf', 'lior wolf')<br/>('1756099', 'Tal Hassner', 'tal hassner')<br/>('2188620', 'Yaniv Taigman', 'yaniv taigman')</td><td></td></tr><tr><td>7f3a73babe733520112c0199ff8d26ddfc7038a0</td><td></td><td></td><td></td></tr><tr><td>7f8d44e7fd2605d580683e47bb185de7f9ea9e28</td><td>Predicting Personal Traits from Facial Images using Convolutional Neural +<br/>Networks Augmented with Facial Landmark Information +<br/><b>The Hebrew University of Jerusalem, Israel</b><br/>2Microsoft Research, Cambridge, United Kingdom +<br/><b>Machine Intelligence Lab (MIL), Cambridge University</b></td><td>('2291654', 'Yoad Lewenberg', 'yoad lewenberg')<br/>('1698412', 'Yoram Bachrach', 'yoram bachrach')<br/>('1808862', 'Sukrit Shankar', 'sukrit shankar')<br/>('1716777', 'Antonio Criminisi', 'antonio criminisi')</td><td>yoadlew@cs.huji.ac.il +<br/>yobach@microsoft.com +<br/>ss965@cam.ac.uk +<br/>antcrim@microsoft.com +</td></tr><tr><td>7f1f3d7b1a4e7fc895b77cb23b1119a6f13e4d3a</td><td>Proc. of IEEE International +<br/>Symposium on Computational +<br/>Intelligence in Robotics and +<br/>Automation (CIRA), July.16-20, +<br/>2003, Kobe Japan, pp. 954-959 +<br/>Multi-Subregion Based Probabilistic Approach Toward +<br/>Pose-Invariant Face Recognition +<br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA</b><br/>2SANYO Electric Co., Ltd., Osaka, Japan 573-8534 +</td><td>('1733113', 'Takeo Kanade', 'takeo kanade')<br/>('3151943', 'Akihiko Yamada', 'akihiko yamada')</td><td>E-mail: tk@ri.cmu.edu, aki-yamada@rd.sanyo.co.jp, +</td></tr><tr><td>7fcfd72ba6bc14bbb90b31fe14c2c77a8b220ab2</td><td>Robust FEC-CNN: A High Accuracy Facial Landmark Detection System +<br/>1 Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), +<br/><b>Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b><br/>3 CAS Center for Excellence in Brain Science and Intelligence Technology +</td><td>('3469114', 'Zhenliang He', 'zhenliang he')<br/>('1698586', 'Jie Zhang', 'jie zhang')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{zhenliang.he,jie.zhang,meina.kan,shiguang.shan,xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>7f205b9fca7e66ac80758c4d6caabe148deb8581</td><td>Page 1 of 47 +<br/>Computing Surveys +<br/>A Survey on Mobile Social Signal Processing +<br/>Understanding human behaviour in an automatic but non-intrusive manner is an important area for various applications. This requires the +<br/>collaboration of information technology with human sciences to transfer existing knowledge of human behaviour into self-acting tools. These +<br/>tools will reduce human error that is introduced by current obtrusive methods such as questionnaires. To achieve unobtrusiveness, we focus on +<br/>exploiting the pervasive and ubiquitous character of mobile devices. +<br/>In this article, a survey of existing techniques for extracting social behaviour through mobile devices is provided. Initially we expose the +<br/>terminology used in the area and introduce a concrete architecture for social signal processing applications on mobile phones, constituted by +<br/>sensing, social interaction detection, behavioural cues extraction, social signal inference and social behaviour understanding. Furthermore, we +<br/>present state-of-the-art techniques applied to each stage of the process. Finally, potential applications are shown while arguing about the main +<br/>challenges of the area. +<br/>Categories and Subject Descriptors: General and reference [Document Types]: Surveys and Overviews; Human-centered computing [Collab- +<br/>orative and social computing, Ubiquitous and mobile computing] +<br/>General Terms: Design, Theory, Human Factors, Performance +<br/>Additional Key Words and Phrases: Social Signal Processing, mobile phones, social behaviour +<br/>ACM Reference Format: +<br/>Processing. ACM V, N, Article A (January YYYY), 35 pages. +<br/>DOI:http://dx.doi.org/10.1145/0000000.0000000 +<br/>1. INTRODUCTION +<br/>Human behaviour understanding has received a great deal of interest since the beginning of the previous century. +<br/>People initially conducted research on the way animals behave when they are surrounded by creatures of the same +<br/>species. Acquiring basic underlying knowledge of animal relations led to extending this information to humans +<br/>in order to understand social behaviour, social relations etc. Initial experiments were conducted by empirically +<br/>observing people and retrieving feedback from them. These methods gave rise to well-established psychological +<br/>approaches for understanding human behaviour, such as surveys, questionnaires, camera recordings and human +<br/>observers. Nevertheless, these methods introduce several limitations including various sources of error. Complet- +<br/>ing surveys and questionnaires induces partiality, unconcern etc. [Groves 2004], human error [Reason 1990], and +<br/>additional restrictions in scalability of the experiments. Accumulating these research problems leads to a common +<br/>challenge, the lack of automation in an unobtrusive manner. +<br/>An area that has focussed on detecting social behaviour automatically and has received a great amount of at- +<br/>tention is Social Signal Processing (SSP). The main target of the field is to model, analyse and synthesise human +<br/>behaviour with limited user intervention. To achieve these targets, researchers presented three key terms which +</td><td>('23537960', 'NIKLAS PALAGHIAS', 'niklas palaghias')<br/>('3339833', 'SEYED AMIR HOSEINITABATABAEI', 'seyed amir hoseinitabatabaei')<br/>('2082222', 'MICHELE NATI', 'michele nati')<br/>('1929850', 'ALEXANDER GLUHAK', 'alexander gluhak')<br/>('1693389', 'KLAUS MOESSNER', 'klaus moessner')<br/>('23537960', 'NIKLAS PALAGHIAS', 'niklas palaghias')<br/>('3339833', 'SEYED AMIR HOSEINITABATABAEI', 'seyed amir hoseinitabatabaei')<br/>('2082222', 'MICHELE NATI', 'michele nati')<br/>('1929850', 'ALEXANDER GLUHAK', 'alexander gluhak')<br/>('1693389', 'KLAUS MOESSNER', 'klaus moessner')</td><td></td></tr><tr><td>7fc76446d2b11fc0479df6e285723ceb4244d4ef</td><td>JRPIT 42.1.QXP:Layout 1 12/03/10 2:11 PM Page 3 +<br/>Laplacian MinMax Discriminant Projection and its +<br/>Applications +<br/><b>Zhejiang Normal University, Jinhua, China</b><br/>Jie Yang +<br/><b>Institute of Image Processing and Pattern Recognition, Shanghai Jiao Tong University, Shanghai, China</b><br/>A new algorithm, Laplacian MinMax Discriminant Projection (LMMDP), is proposed in this paper +<br/>for supervised dimensionality reduction. LMMDP aims at learning a discriminant linear +<br/>transformation. Specifically, we define the within-class scatter and the between-class scatter using +<br/>similarities which are based on pairwise distances in sample space. After the transformation, the +<br/>considered pairwise samples within the same class are as close as possible, while those between +<br/>classes are as far as possible. The structural information of classes is contained in the within-class +<br/>and the between-class Laplacian matrices. Therefore, the discriminant projection subspace can be +<br/>derived by controlling the structural evolution of Laplacian matrices. The performance on several +<br/>data sets demonstrates the competence of the proposed algorithm. +<br/>ACM Classification: I.5 +<br/>Keywords: Manifold Learning; Dimensionality Reduction; Supervised Learning; Discriminant +<br/>Analysis +<br/>1. INTRODUCTION +<br/>Dimensionality reduction has attracted tremendous attention in the pattern recognition community +<br/>over the past few decades and many new algorithms have been developed. Among these algorithms, +<br/>linear dimensionality reduction is widely spread for its simplicity and effectiveness. Principal +<br/>component analysis (PCA), as a classic linear method for unsupervised dimensionality reduction, +<br/>aims at learning a kind of subspaces where the maximum covariance of all training samples are +<br/>preserved (Turk,1991). Locality Preserving Projections, as another typical approach for +<br/>unsupervised dimensionality reduction, seeks projections to preserve the local structure of the +<br/>sample space (He, 2005). However, unsupervised learning algorithms cannot properly model the +<br/>underlying structures and characteristics of different classes (Zhao, 2007). Discriminant features are +<br/>often obtained by supervised dimensionality reduction. Linear discriminant analysis (LDA) is one +<br/>of the most popular supervised techniques for classification (Fukunaga, 1990; Belhumeur, 1997). +<br/>LDA aims at learning discriminant subspace where the within-class scatter is minimized and the +<br/>between-class scatter of samples is maximized at the same time. Many improved LDAs up to date +<br/>have demonstrated competitive performance in object classification (Howland, 2004; Liu, 2007; +<br/>Martinez, 2006; Wang and Tang, 2004a; Yang, 2005). +<br/>Copyright© 2010, Australian Computer Society Inc. General permission to republish, but not for profit, all or part of this +<br/>material is granted, provided that the JRPIT copyright notice is given and that reference is made to the publication, to its +<br/>date of issue, and to the fact that reprinting privileges were granted by permission of the Australian Computer Society Inc. +<br/>Manuscript received: 15 April 2008 +<br/>Communicating Editor: Tele Tan +</td><td>('3185576', 'Zhonglong Zheng', 'zhonglong zheng')<br/>('3140483', 'Xueping Chang', 'xueping chang')</td><td>Email: zhonglong@sjtu.org +</td></tr><tr><td>7a9ef21a7f59a47ce53b1dff2dd49a8289bb5098</td><td></td><td></td><td></td></tr><tr><td>7af38f6dcfbe1cd89f2307776bcaa09c54c30a8b</td><td>eaig i C e Vii ad Beyd: +<br/>Deve +<br/>h . Weg +<br/>Deae f C e Sciece +<br/>ichiga Sae Uiveiy +<br/>Ea aig 48824 +<br/>Abac +<br/>Thi chae id ce wha i ca +<br/>aic +<br/>ve +<br/>by h a cgiive deve +<br/>ih i deeied befe he \bih" f he ye. Afe he \bih" i eab +<br/> +<br/>aach i ea +<br/>deve +<br/>way aia +<br/> ea whi +<br/>de deve +</td><td></td><td>weg@c. .ed +</td></tr><tr><td>7a81967598c2c0b3b3771c1af943efb1defd4482</td><td>Do We Need More Training Data? +</td><td>('32542103', 'Xiangxin Zhu', 'xiangxin zhu')</td><td></td></tr><tr><td>7ae0212d6bf8a067b468f2a78054c64ea6a577ce</td><td>Human Face Processing Techniques +<br/>With Application To +<br/>Large Scale Video Indexing +<br/>DOCTOR OF +<br/>PHILOSOPHY +<br/>Department of Informatics, +<br/>School of Multidisciplinary Sciences, +<br/><b>The Graduate University for Advanced Studies (SOKENDAI</b><br/>2006 (School Ｙear) +<br/>September 2006 +</td><td></td><td></td></tr><tr><td>7a9c317734acaf4b9bd8e07dd99221c457b94171</td><td>Lorentzian Discriminant Projection and Its Applications +<br/><b>Dalian University of Technology, Dalian 116024, China</b><br/>2 Microsoft Research Asia, Beijing 100080, China +</td><td>('34469457', 'Risheng Liu', 'risheng liu')<br/>('4642456', 'Zhixun Su', 'zhixun su')<br/>('33383055', 'Zhouchen Lin', 'zhouchen lin')<br/>('40290490', 'Xiaoyu Hou', 'xiaoyu hou')</td><td>zxsu@dlut.edu.cn +</td></tr><tr><td>7a0fb972e524cb9115cae655e24f2ae0cfe448e0</td><td>Facial Expression Classification Using RBF AND Back-Propagation Neural Networks +<br/>R.Q.Feitosa1,2, +<br/>M.M.B.Vellasco1,2, +<br/>D.T.Oliveira1, +<br/>D.V.Andrade1, +<br/>S.A.R.S.Maffra1 +<br/><b>Catholic University of Rio de Janeiro, Brazil</b><br/>Department of Electric Engineering +<br/><b>State University of Rio de Janeiro, Brazil</b><br/>Department of Computer Engineering +</td><td></td><td>e-mail: [raul, marley]@ele.puc -rio.br, tuler@inf.puc-rio.br, [diogo, sam]@tecgraf.puc-rio.br +</td></tr><tr><td>7ad77b6e727795a12fdacd1f328f4f904471233f</td><td>Supervised Local Descriptor Learning +<br/>for Human Action Recognition +</td><td>('34798935', 'Xiantong Zhen', 'xiantong zhen')<br/>('40255667', 'Feng Zheng', 'feng zheng')<br/>('40799321', 'Ling Shao', 'ling shao')<br/>('1720247', 'Xianbin Cao', 'xianbin cao')<br/>('40147776', 'Dan Xu', 'dan xu')</td><td></td></tr><tr><td>7a3d46f32f680144fd2ba261681b43b86b702b85</td><td>Multi-label Learning Based Deep Transfer Neural Network for Facial Attribute +<br/>Classification +<br/><b>School of Information Science and Engineering, Xiamen University, Xiamen 361005, China</b><br/><b>bSchool of Computer and Information Engineering, Xiamen University of Technology, Xiamen 361024, China</b><br/>aFujian Key Laboratory of Sensing and Computing for Smart City, +<br/><b>cSchool of Computer Science, The University of Adelaide, Adelaide, SA 5005, Australia</b></td><td>('41034942', 'Ni Zhuang', 'ni zhuang')<br/>('40461734', 'Yan Yan', 'yan yan')<br/>('47336404', 'Si Chen', 'si chen')<br/>('37414077', 'Hanzi Wang', 'hanzi wang')<br/>('1780381', 'Chunhua Shen', 'chunhua shen')</td><td></td></tr><tr><td>7a97de9460d679efa5a5b4c6f0b0a5ef68b56b3b</td><td></td><td></td><td></td></tr><tr><td>7a7f2403e3cc7207e76475e8f27a501c21320a44</td><td>Emotion Recognition from Multi-Modal Information +<br/>Department of Computer Science and Information Engineering, +<br/><b>National Cheng Kung University, Tainan, Taiwan, R.O.C</b></td><td>('1681512', 'Chung-Hsien Wu', 'chung-hsien wu')<br/>('1709777', 'Jen-Chun Lin', 'jen-chun lin')<br/>('1691390', 'Wen-Li Wei', 'wen-li wei')<br/>('2891156', 'Kuan-Chun Cheng', 'kuan-chun cheng')</td><td>E-mail: chunghsienwu@gmail.com, jenchunlin@gmail.com, lilijinjin@gmail.com, davidcheng817@gmail.com +</td></tr><tr><td>7aafeb9aab48fb2c34bed4b86755ac71e3f00338</td><td>Article +<br/>Real Time 3D Facial Movement Tracking Using a +<br/>Monocular Camera +<br/><b>School of Electronics and Information Engineering, Tongji University, Caoan Road 4800, Shanghai</b><br/><b>Kumamoto University, 2-39-1 Kurokami, Kumamoto shi</b><br/>Academic Editor: Vittorio M. N. Passaro +<br/>Received: 9 May 2016; Accepted: 20 July 2016; Published: 25 July 2016 +</td><td>('2576907', 'Yanchao Dong', 'yanchao dong')<br/>('1715838', 'Yanming Wang', 'yanming wang')<br/>('2721582', 'Jiguang Yue', 'jiguang yue')<br/>('3256415', 'Zhencheng Hu', 'zhencheng hu')</td><td>China; 11wanggyanming@tongji.edu.cn (Y.W.); yuejiguang@tongji.edu.cn (J.Y.) +<br/>Japan; hu@cs.kumamoto-u.ac.jp +<br/>* Correspondence: dongyanchao@tongji.edu.cn; Tel.: +86-21-6958-3806 +</td></tr><tr><td>7a84368ebb1a20cc0882237a4947efc81c56c0c0</td><td>Robust and Efficient Parametric Face Alignment +<br/>†Dept. of Computing, +<br/><b>Imperial College London</b><br/>180 Queen’s Gate +<br/>London SW7 2AZ, U.K. +<br/>∗EEMCS +<br/><b>University of Twente</b><br/>Drienerlolaan 5 +<br/>7522 NB Enschede +<br/>The Netherlands ∗ +</td><td>('2610880', 'Georgios Tzimiropoulos', 'georgios tzimiropoulos')<br/>('1776444', 'Stefanos Zafeiriou', 'stefanos zafeiriou')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td>{gt204,s.zafeiriou,m.pantic}@imperial.ac.uk +</td></tr><tr><td>7aa4c16a8e1481629f16167dea313fe9256abb42</td><td>978-1-5090-4117-6/17/$31.00 ©2017 IEEE +<br/>2981 +<br/>ICASSP 2017 +</td><td></td><td></td></tr><tr><td>7a85b3ab0efb6b6fcb034ce13145156ee9d10598</td><td></td><td></td><td></td></tr><tr><td>7ab930146f4b5946ec59459f8473c700bcc89233</td><td></td><td></td><td></td></tr><tr><td>7a65fc9e78eff3ab6062707deaadde024d2fad40</td><td>A Study on Apparent Age Estimation +<br/><b>West Virginia University, Morgantown WV 26506, USA</b><br/><b>Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS), Institute of</b><br/>Computing Technology, CAS, Beijing, 100190, China +</td><td>('1736182', 'Yu Zhu', 'yu zhu')<br/>('1698571', 'Yan Li', 'yan li')<br/>('2501850', 'Guowang Mu', 'guowang mu')<br/>('1822413', 'Guodong Guo', 'guodong guo')</td><td>yzhu4@mix.wvu.edu, yan.li@vipl.ict.ac.cn, guowang.mu@mail.wvu.edu , +<br/>Guodong.Guo@mail.wvu.edu (corresponding author) +</td></tr><tr><td>7ad7897740e701eae455457ea74ac10f8b307bed</td><td>Random Subspace Two-dimensional LDA for Face Recognition* +</td><td>('29980351', 'Garrett Bingham', 'garrett bingham')</td><td></td></tr><tr><td>7ac9aaafe4d74542832c273acf9d631cb8ea6193</td><td>Deep Micro-Dictionary Learning and Coding Network +<br/><b>University of Trento, Trento, Italy</b><br/>2Department of Electrical Engineering, Hong Kong Polytechnic Unversity, Hong Kong, China +<br/>3Lingxi Artificial Interlligence Co., Ltd, Shen Zhen, China +<br/>4Computer Vision Laboratory, ´Ecole Polytechnique F´ed´erale de Lausanne, Lausanne, Switzerland +<br/><b>University of Oxford, Oxford, UK</b><br/><b>Texas State University, San Marcos, USA</b></td><td>('46666325', 'Hao Tang', 'hao tang')<br/>('49567679', 'Heng Wei', 'heng wei')<br/>('38505394', 'Wei Xiao', 'wei xiao')<br/>('47824598', 'Wei Wang', 'wei wang')<br/>('40147776', 'Dan Xu', 'dan xu')<br/>('1703601', 'Nicu Sebe', 'nicu sebe')</td><td>{hao.tang, niculae.sebe}@unitn.it, 15102924d@connect.polyu.hk, xiaoweithu@163.com +<br/>wei.wang@epfl.ch, danxu@robots.ox.ac.uk, y y34@txstate.edu +</td></tr><tr><td>7a1ce696e260899688cb705f243adf73c679f0d9</td><td>Predicting Missing Demographic Information in +<br/>Biometric Records using Label Propagation +<br/>Techniques +<br/>Department of Computer Science and Engineering +<br/>Department of Computer Science and Engineering +<br/><b>Michigan State University</b><br/>East Lansing, Michigan 48824 +<br/><b>Michigan State University</b><br/>East Lansing, Michigan 48824 +</td><td>('3153117', 'Thomas Swearingen', 'thomas swearingen')<br/>('1698707', 'Arun Ross', 'arun ross')</td><td>Email: swearin3@msu.edu +<br/>Email: rossarun@msu.edu +</td></tr><tr><td>7a7b1352d97913ba7b5d9318d4c3d0d53d6fb697</td><td>Attend and Rectify: a Gated Attention +<br/>Mechanism for Fine-Grained Recovery +<br/>†Computer Vision Center and Universitat Aut`onoma de Barcelona (UAB), +<br/>Campus UAB, 08193 Bellaterra, Catalonia Spain +<br/>‡Visual Tagging Services, Parc de Recerca, Campus UAB +</td><td>('1739551', 'Josep M. Gonfaus', 'josep m. gonfaus')<br/>('7153363', 'Guillem Cucurull', 'guillem cucurull')<br/>('1696387', 'F. Xavier Roca', 'f. xavier roca')</td><td></td></tr><tr><td>7aa062c6c90dba866273f5edd413075b90077b51</td><td>I.J. Information Technology and Computer Science, 2017, 5, 40-51 +<br/>Published Online May 2017 in MECS (http://www.mecs-press.org/) +<br/>DOI: 10.5815/ijitcs.2017.05.06 +<br/>Minimizing Separability: A Comparative Analysis +<br/>of Illumination Compensation Techniques in Face +<br/>Recognition +<br/><b>Baze University, Abuja, Nigeria</b></td><td>('7392398', 'Chollette C. Olisah', 'chollette c. olisah')</td><td>E-mail: chollette.olisah@bazeuniversity.edu.ng +</td></tr><tr><td>7a131fafa7058fb75fdca32d0529bc7cb50429bd</td><td>Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and +<br/>Identity Preserving Frontal View Synthesis +<br/>1National Laboratory of Pattern Recognition, CASIA +<br/>2Center for Research on Intelligent Perception and Computing, CASIA +<br/><b>University of Chinese Academy of Sciences, Beijing, China</b></td><td>('48241673', 'Rui Huang', 'rui huang')<br/>('50202300', 'Shu Zhang', 'shu zhang')<br/>('50290162', 'Tianyu Li', 'tianyu li')<br/>('1705643', 'Ran He', 'ran he')</td><td>huangrui@cmu.edu, tianyu.lizard@gmail.com, {shu.zhang, rhe}@nlpr.ia.ac.cn +</td></tr><tr><td>1451e7b11e66c86104f9391b80d9fb422fb11c01</td><td>IET Signal Processing +<br/>Research Article +<br/>Image privacy protection with secure JPEG +<br/>transmorphing +<br/>ISSN 1751-9675 +<br/>Received on 30th December 2016 +<br/>Revised 13th July 2017 +<br/>Accepted on 11th August 2017 +<br/>doi: 10.1049/iet-spr.2016.0756 +<br/>www.ietdl.org +<br/>1Multimedia Signal Processing Group, Electrical Engineering Department, EPFL, Station 11, Lausanne, Switzerland +</td><td>('1681498', 'Touradj Ebrahimi', 'touradj ebrahimi')</td><td> E-mail: lin.yuan@epfl.ch +</td></tr><tr><td>14761b89152aa1fc280a33ea4d77b723df4e3864</td><td></td><td></td><td></td></tr><tr><td>14b87359f6874ff9b8ee234b18b418e57e75b762</td><td>H. GAO ET AL: FACE ALIGNMENT USING A RANKING MODEL BASED ON RT +<br/>Face Alignment Using a Ranking Model +<br/>based on Regression Trees +<br/>Hazım Kemal Ekenel1,2 +<br/><b>Institute for Anthropomatics</b><br/><b>Karlsruhe Institute of Technology</b><br/>Karlsruhe, Germany +<br/>2 Faculty of Computer and Informatics +<br/><b>Istanbul Technical University</b><br/>Istanbul, Turkey +</td><td>('1697965', 'Hua Gao', 'hua gao')<br/>('1742325', 'Rainer Stiefelhagen', 'rainer stiefelhagen')</td><td>gao@kit.edu +<br/>ekenel@kit.edu +<br/>rainer.stiefelhagen@kit.edu +</td></tr><tr><td>14fdec563788af3202ce71c021dd8b300ae33051</td><td>Social Influence Analysis based on Facial Emotions +<br/>Department of Computer Science and Engineering +<br/><b>Nagoya Institute of Technology, Gokiso, Showa-ku, Nagoya, 466-8555 Japan</b></td><td>('2159044', 'Pankaj Mishra', 'pankaj mishra')<br/>('1679044', 'Takayuki Ito', 'takayuki ito')</td><td>{pankaj.mishra, rafik}@itolab.nitech.ac.jp, +<br/>ito.takayuki@nitech.ac.jp +</td></tr><tr><td>142e5b4492bc83b36191be4445ef0b8b770bf4b0</td><td>Discriminative Analysis of Brain Function +<br/>at Resting-State for Attention-Deficit/Hyperactivity +<br/>Disorder +<br/>Y.F. Wang2, and T. Z. Jiang1 +<br/><b>National Laboratory of Pattern Recognition, Institute of Automation</b><br/>Chinese Academy of Sciences, P.R. China +<br/><b>Institute of Mental Health, Peking University, P.R. China</b></td><td>('2339602', 'M. Liang', 'm. liang')</td><td>czzhu@nlpr.ia.ac.cn +</td></tr><tr><td>14b016c7a87d142f4b9a0e6dc470dcfc073af517</td><td>Modest proposals for improving biometric recognition papers +<br/>NIST, Gaithersburg MD +<br/><b>San Jose State University, San Jose, CA</b></td><td>('2145366', 'James R. Matey', 'james r. matey')<br/>('34958610', 'George W. Quinn', 'george w. quinn')<br/>('2136478', 'Patrick Grother', 'patrick grother')<br/>('2326261', 'Elham Tabassi', 'elham tabassi')<br/>('1707135', 'James L. Wayman', 'james l. wayman')</td><td>POC: james.matey@NIST.gov +<br/>jlwayman@aol.com +</td></tr><tr><td>14b66748d7c8f3752dca23991254fca81b6ee86c</td><td>A. RICHARD, J. GALL: A BOW-EQUIVALENT NEURAL NETWORK +<br/>A BoW-equivalent Recurrent Neural Network +<br/>for Action Recognition +<br/><b>Institute of Computer Science III</b><br/><b>University of Bonn</b><br/>Bonn, Germany +</td><td>('32774629', 'Alexander Richard', 'alexander richard')<br/>('2946643', 'Juergen Gall', 'juergen gall')</td><td>richard@iai.uni-bonn.de +<br/>gall@iai.uni-bonn.de +</td></tr><tr><td>14e8dbc0db89ef722c3c198ae19bde58138e88bf</td><td>HapFACS: an Open Source API/Software to +<br/>Generate FACS-Based Expressions for ECAs +<br/>Animation and for Corpus Generation +<br/>Christine Lisetti +<br/>School of Computing and Information Sciences +<br/>School of Computing and Information Sciences +<br/><b>Florida International University</b><br/>Miami, Florida, USA +<br/><b>Florida International University</b><br/>Miami, Florida, USA +</td><td>('1809087', 'Reza Amini', 'reza amini')</td><td>Email: ramin001@fiu.edu +<br/>Email: lisetti@cis.fiu.edu +</td></tr><tr><td>14fa27234fa2112014eda23da16af606db7f3637</td><td></td><td></td><td></td></tr><tr><td>1459d4d16088379c3748322ab0835f50300d9a38</td><td>Cross-Domain Visual Matching via Generalized +<br/>Similarity Measure and Feature Learning +</td><td>('40461403', 'Liang Lin', 'liang lin')<br/>('2749191', 'Guangrun Wang', 'guangrun wang')<br/>('1724520', 'Wangmeng Zuo', 'wangmeng zuo')<br/>('2340559', 'Xiangchu Feng', 'xiangchu feng')<br/>('40396552', 'Lei Zhang', 'lei zhang')</td><td></td></tr><tr><td>14e949f5754f9e5160e8bfa3f1364dd92c2bb8d6</td><td></td><td></td><td></td></tr><tr><td>146bbf00298ee1caecde3d74e59a2b8773d2c0fc</td><td><b>University of Groningen</b><br/>4D Unconstrained Real-time Face Recognition Using a Commodity Depthh Camera +<br/>Schimbinschi, Florin; Wiering, Marco; Mohan, R.E.; Sheba, J.K. +<br/>Published in: +<br/>7th IEEE Conference on Industrial Electronics and Applications +<br/>IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to +<br/>cite from it. Please check the document version below. +<br/>Document Version +<br/>Final author's version (accepted by publisher, after peer review) +<br/>Publication date: +<br/>2012 +<br/><b>Link to publication in University of Groningen/UMCG research database</b><br/>Citation for published version (APA): +<br/>Schimbinschi, F., Wiering, M., Mohan, R. E., & Sheba, J. K. (2012). 4D Unconstrained Real-time Face +<br/>Recognition Using a Commodity Depthh Camera. In 7th IEEE Conference on Industrial Electronics and +<br/>Applications : ICIEA +<br/>Copyright +<br/>Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the +<br/>author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons). +<br/>Take-down policy +<br/>If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately +<br/>and investigate your claim. +<br/><b>Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the</b><br/>number of authors shown on this cover page is limited to 10 maximum. +<br/>Download date: 03-09-2017 +<br/> </td><td></td><td></td></tr><tr><td>14e9158daf17985ccbb15c9cd31cf457e5551990</td><td>ConvNets with Smooth Adaptive Activation Functions for +<br/>Regression +<br/>Tahsin M. Kurc1,2 +<br/><b>Stony Brook University</b><br/>2Oak Ridge National Laboratory +<br/><b>Stony Brook University Hospital</b></td><td>('2321406', 'Le Hou', 'le hou')<br/>('1686020', 'Dimitris Samaras', 'dimitris samaras')<br/>('1735710', 'Joel H. Saltz', 'joel h. saltz')<br/>('1755448', 'Yi Gao', 'yi gao')</td><td></td></tr><tr><td>14ce7635ff18318e7094417d0f92acbec6669f1c</td><td>DeepFace: Closing the Gap to Human-Level Performance in Face Verification +<br/>Marc’Aurelio Ranzato +<br/>Facebook AI Group +<br/>Menlo Park, CA, USA +<br/><b>Tel Aviv University</b><br/>Tel Aviv, Israel +</td><td>('2188620', 'Yaniv Taigman', 'yaniv taigman')<br/>('2909406', 'Ming Yang', 'ming yang')<br/>('1776343', 'Lior Wolf', 'lior wolf')</td><td>{yaniv, mingyang, ranzato}@fb.com +<br/>wolf@cs.tau.ac.il +</td></tr><tr><td>1450296fb936d666f2f11454cc8f0108e2306741</td><td>Learning to Discover Cross-Domain Relations +<br/>with Generative Adversarial Networks +</td><td>('2509132', 'Taeksoo Kim', 'taeksoo kim')</td><td></td></tr><tr><td>140438a77a771a8fb656b39a78ff488066eb6b50</td><td>Localizing Parts of Faces Using a Consensus of Exemplars +<br/>(cid:63)Kriegman-Belhumeur Vision Technologies∗ +<br/><b>University of Maryland, College Park</b><br/><b>University of California, San Diego</b><br/><b>Columbia University</b></td><td>('1767767', 'Peter N. Belhumeur', 'peter n. belhumeur')<br/>('34734622', 'David W. Jacobs', 'david w. jacobs')<br/>('1765887', 'David J. Kriegman', 'david j. kriegman')<br/>('40631426', 'Neeraj Kumar', 'neeraj kumar')</td><td></td></tr><tr><td>143bee9120bcd7df29a0f2ad6f0f0abfb23977b8</td><td>Shared Gaussian Process Latent Variable Model +<br/>for Multi-view Facial Expression Recognition +<br/><b>Imperial College London, UK</b><br/><b>EEMCS, University of Twente, The Netherlands</b></td><td>('2308430', 'Stefanos Eleftheriadis', 'stefanos eleftheriadis')<br/>('1729713', 'Ognjen Rudovic', 'ognjen rudovic')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>14d72dc9f78d65534c68c3ed57305f14bd4b5753</td><td>Exploiting Multi-Grain Ranking Constraints for Precisely Searching +<br/>Visually-similar Vehicles +<br/>1National Engineering Laboratory for Video Technology, School of EE&CS, +<br/><b>Peking University, Beijing, China</b><br/>2Cooperative Medianet Innovation Center, China +<br/><b>Beijing Institute of Technology, China</b></td><td>('13318784', 'Ke Yan', 'ke yan')<br/>('5765799', 'Yaowei Wang', 'yaowei wang')<br/>('1687907', 'Wei Zeng', 'wei zeng')<br/>('1705972', 'Yonghong Tian', 'yonghong tian')<br/>('34097174', 'Tiejun Huang', 'tiejun huang')</td><td>{keyan, yhtian, weizeng, tjhuang}@pku.edu.cn;yaoweiwang@bit.edu.cn +</td></tr><tr><td>14b162c2581aea1c0ffe84e7e9273ab075820f52</td><td>Training Object Class Detectors from Eye Tracking Data +<br/><b>School of Informatics, University of Edinburgh, UK</b></td><td>('1749373', 'Dim P. Papadopoulos', 'dim p. papadopoulos')<br/>('2505673', 'Frank Keller', 'frank keller')<br/>('1749692', 'Vittorio Ferrari', 'vittorio ferrari')</td><td></td></tr><tr><td>14ff9c89f00dacc8e0c13c94f9fadcd90e4e604d</td><td>Correlation Filter Cascade for Facial Landmark Localization +<br/>Pattern Analysis and Computer Vision Department +<br/>School of Computing +<br/>Istituto Italiano di Tecnologia, Genova, Italy +<br/><b>National University of Singapore, Singapore</b></td><td>('2860592', 'Hamed Kiani Galoogahi', 'hamed kiani galoogahi')<br/>('1715286', 'Terence Sim', 'terence sim')</td><td>hamed.kiani@iit.it +<br/>tsim@comp.nus.edu.sg +</td></tr><tr><td>14fdce01c958043140e3af0a7f274517b235adf3</td><td></td><td></td><td></td></tr><tr><td>14b69626b64106bff20e17cf8681790254d1e81c</td><td>Hybrid Super Vector with Improved Dense Trajectories for Action Recognition +<br/><b>Shenzhen Key Lab of CVPR, Shenzhen Institutes of Advanced Technology, CAS, China</b><br/><b>Southwest Jiaotong University, Chengdu, P.R. China</b><br/><b>The Chinese University of Hong Kong, Hong Kong</b></td><td>('1766837', 'Xiaojiang Peng', 'xiaojiang peng')<br/>('40795365', 'LiMin Wang', 'limin wang')<br/>('2985266', 'Zhuowei Cai', 'zhuowei cai')<br/>('40285012', 'Yu Qiao', 'yu qiao')<br/>('39657084', 'Qiang Peng', 'qiang peng')</td><td>fxiaojiangp,07wanglimin,iamcaizhuoweig@gmail.com, yu.qiao@siat.ac.cn, qpeng@swjtu.edu.cn +</td></tr><tr><td>14070478b8f0d84e5597c3e67c30af91b5c3a917</td><td>Detecting Social Actions of Fruit Flies +<br/><b>California Institute of Technology, Pasadena, California, USA</b><br/><b>Howard Hughes Medical Institute (HHMI</b></td><td>('2948199', 'Eyrun Eyjolfsdottir', 'eyrun eyjolfsdottir')<br/>('3251767', 'Steve Branson', 'steve branson')<br/>('2232848', 'Xavier P. Burgos-Artizzu', 'xavier p. burgos-artizzu')<br/>('2954028', 'Eric D. Hoopfer', 'eric d. hoopfer')<br/>('20299567', 'Jonathan Schor', 'jonathan schor')<br/>('30334638', 'David J. Anderson', 'david j. anderson')<br/>('1690922', 'Pietro Perona', 'pietro perona')</td><td></td></tr><tr><td>14fb3283d4e37760b7dc044a1e2906e3cbf4d23a</td><td>Weak Attributes for Large-Scale Image Retrieval∗ +<br/><b>Columbia University, New York, NY</b></td><td>('1815972', 'Felix X. Yu', 'felix x. yu')<br/>('1725599', 'Rongrong Ji', 'rongrong ji')<br/>('3138710', 'Ming-Hen Tsai', 'ming-hen tsai')<br/>('35984288', 'Guangnan Ye', 'guangnan ye')<br/>('9546964', 'Shih-Fu Chang', 'shih-fu chang')</td><td>y{yuxinnan, rrji, yegn, sfchang}@ee.columbia.edu +<br/>xminghen@cs.columbia.edu +</td></tr><tr><td>14811696e75ce09fd84b75fdd0569c241ae02f12</td><td>Margin-Based Discriminant Dimensionality Reduction for Visual Recognition +<br/><b>Eskisehir Osmangazi University</b><br/>Laboratoire Jean Kuntzmann +<br/>Meselik Kampusu 26480 Eskisehir Turkey +<br/>B.P. 53, 38041 Grenoble Cedex 9, France +<br/>Fr´ed´eric Jurie +<br/><b>University of Caen</b><br/>Universit´e de Caen - F-14032 Caen, France +<br/><b>Rowan University</b><br/>201 Mullica Hill Road, Glassboro NJ USA +</td><td>('2277308', 'Hakan Cevikalp', 'hakan cevikalp')<br/>('1756114', 'Bill Triggs', 'bill triggs')<br/>('1780024', 'Robi Polikar', 'robi polikar')</td><td>Hakan.Cevikalp@gmail.com +<br/>Bill.Triggs@imag.fr +<br/>Frederic.Jurie@unicaen.fr +<br/>polikar@rowan.edu +</td></tr><tr><td>141eab5f7e164e4ef40dd7bc19df9c31bd200c5e</td><td></td><td></td><td></td></tr><tr><td>14e759cb019aaf812d6ac049fde54f40c4ed1468</td><td>Subspace Methods +<br/>Synonyms +<br/>{ Multiple similarity method +<br/>Related Concepts +<br/>{ Principal component analysis (PCA) +<br/>{ Subspace analysis +<br/>{ Dimensionality reduction +<br/>De(cid:12)nition +<br/>Subspace analysis in computer vision is a generic name to describe a general +<br/>framework for comparison and classification of subspaces. A typical approach in +<br/>subspace analysis is the subspace method (SM) that classify an input pattern +<br/>vector into several classes based on the minimum distance or angle between the +<br/>input pattern vector and each class subspace, where a class subspace corresponds +<br/>to the distribution of pattern vectors of the class in high dimensional vector +<br/>space. +<br/>Background +<br/>Comparison and classification of subspaces has been one of the central prob- +<br/>lems in computer vision, where an image set of an object to be classified is +<br/>compactly represented by a subspace in high dimensional vector space. +<br/>The subspace method is one of the most effective classification method in +<br/>subspace analysis, which was developed by two Japanese researchers, Watanabe +<br/>and Iijima around 1970, independently [1, 2]. Watanabe and Iijima named their +<br/>methods the CLAFIC [3] and the multiple similarity method [4], respectively. +<br/>The concept of the subspace method is derived from the observation that pat- +<br/>terns belonging to a class forms a compact cluster in high dimensional vector +<br/>space, where, for example, a w×h pixels image pattern is usually represented as a +<br/>vector in w×h-dimensional vector space. The compact cluster can be represented +<br/>by a subspace, which is generated by using Karhunen-Lo`eve (KL) expansion, also +<br/>known as the principal component analysis (PCA). Note that a subspace is gen- +<br/>erated for each class, unlike the Eigenface Method [5] in which only one subspace +<br/>(called eigenspace) is generated. +<br/>The SM has been known as one of the most useful methods in pattern recog- +<br/>nition field, since its algorithm is very simple and it can handle classification +<br/>of multiple classes. However, its classification performance was not sufficient for +<br/>many applications in practice, because class subspaces are generated indepen- +<br/>dently of each other [1]. There is no reason to assume a priori that each class +</td><td>('1770128', 'Kazuhiro Fukui', 'kazuhiro fukui')</td><td></td></tr><tr><td>1442319de86d171ce9595b20866ec865003e66fc</td><td>Vision-Based Fall Detection with Convolutional +<br/>Neural Networks +<br/><b>DeustoTech - University of Deusto</b><br/>Avenida de las Universidades, 24 - 48007, Bilbao, Spain +<br/>2 Dept. of Computer Science and Artificial Intelligence, Basque +<br/><b>Country University, San Sebastian, Spain</b><br/>P. Manuel Lardizabal, 1 - 20018, San Sebastian, Spain +<br/>3 Ikerbasque, Basque Foundation for Science, Bilbao, Spain +<br/>Maria Diaz de Haro, 3 - 48013 Bilbao, Spain +<br/>4 Donostia International Physics Center (DIPC), San Sebastian, Spain +<br/>P. Manuel Lardizabal, 4 - 20018, San Sebastian, Spain +</td><td>('2481918', 'Gorka Azkune', 'gorka azkune')<br/>('3147227', 'Ignacio Arganda-Carreras', 'ignacio arganda-carreras')</td><td>{adrian.nunez@deusto.es, gorka.azkune@deusto.es, ignacio.arganda@ehu.es} +</td></tr><tr><td>146a7ecc7e34b85276dd0275c337eff6ba6ef8c0</td><td>This is a pre-print of the original paper submitted for review in FG 2017. +<br/>AFFACT - Alignment Free Facial Attribute Classification Technique +<br/>Vision and Security Technology (VAST) Lab, +<br/><b>University of Colorado Colorado Springs</b><br/>∗ authors with equal contribution +</td><td>('2974221', 'Andras Rozsa', 'andras rozsa')<br/>('1760117', 'Terrance E. Boult', 'terrance e. boult')</td><td>{mgunther,arozsa,tboult}@vast.uccs.edu +</td></tr><tr><td>148eb413bede35487198ce7851997bf8721ea2d6</td><td>People Search in Surveillance Videos +<br/>Four Eyes Lab, UCSB +<br/>IBM Research +<br/>IBM Research +<br/>IBM Research +<br/>Four Eyes Lab, UCSB +<br/>INTRODUCTION +<br/>1. +<br/>In traditional surveillance scenarios, users are required to +<br/>watch video footage corresponding to extended periods of +<br/>time in order to find events of interest. However, this pro- +<br/>cess is resource-consuming, and suffers from high costs of +<br/>employing security personnel. The field of intelligent vi- +<br/>sual surveillance [2] seeks to address these issues by applying +<br/>computer vision techniques to automatically detect specific +<br/>events in long video streams. The events can then be pre- +<br/>sented to the user or be indexed into a database to allow +<br/>queries such as “show me the red cars that entered a given +<br/>parking lot from 7pm to 9pm on Monday” or “show me the +<br/>faces of people who left the city’s train station last week.” +<br/>In this work, we are interested in analyzing people, by ex- +<br/>tracting information that can be used to search for them in +<br/>surveillance videos. Current research on this topic focuses +<br/>on approaches based on face recognition, where the goal is +<br/>to establish the identity of a person given an image of a +<br/>face. However, face recognition is still a very challenging +<br/>problem, especially in low resolution images with variations +<br/>in pose and lighting, which is often the case in surveillance +<br/>data. State-of-the-art face recognition systems [1] require +<br/>a fair amount of resolution in order to produce reliable re- +<br/>sults, but in many cases this level of detail is not available +<br/>in surveillance applications. +<br/>We approach the problem in an alternative way, by avoiding +<br/>face recognition and proposing a framework for finding peo- +<br/>ple based on parsing the human body and exploiting part +<br/>attributes. Those include visual attributes such as facial hair +<br/>type (beards, mustaches, absence of facial hair), type of eye- +<br/>wear (sunglasses, eyeglasses, absence of glasses), hair type +<br/>(baldness, hair, wearing a hat), and clothing color. While +<br/>face recognition is still a difficult problem, accurate and ef- +<br/>ficient face detectors1 based on learning approaches [6] are +<br/>available. Those have been demonstrated to work well on +<br/>challenging low-resolution images, with variations in pose +<br/>and lighting. In our method, we employ this technology to +<br/>design detectors for facial attributes from large sets of train- +<br/>ing data. +<br/>1The face detection problem consists of localizing faces in +<br/>images, while face recognition aims to establish the identity +<br/>of a person given an image of a face. Face detection is a +<br/>challenging problem, but it is arguably not as complex as +<br/>face recognition. +<br/>Our technique falls into the category of short term recogni- +<br/>tion methods, taking advantage of features present in brief +<br/>intervals in time, such as clothing color, hairstyle, and makeup, +<br/>which are generally considered an annoyance in face recogni- +<br/>tion methods. There are several applications that naturally +<br/>fit within a short term recognition framework. An example +<br/>is in criminal investigation, when the police are interested in +<br/>locating a suspect. In those cases, eyewitnesses typically fill +<br/>out a suspect description form, where they indicate personal +<br/>traits of the suspect as seen at the moment when the crime +<br/>was committed. Those include facial hair type, hair color, +<br/>clothing type, etc. Based on that description, the police +<br/>manually scan the entire video archive looking for a person +<br/>with similar characteristics. This process is tedious and time +<br/>consuming, and could be drastically accelerated by the use +<br/>of our technique. Another application is on finding missing +<br/>people. Parents looking for their children in an amusement +<br/>park could provide a description including clothing and eye- +<br/>wear type, and videos from multiple cameras in the park +<br/>would then be automatically searched. +</td><td>('2000950', 'Daniel A. Vaquero', 'daniel a. vaquero')<br/>('1723233', 'Rogerio S. Feris', 'rogerio s. feris')<br/>('11081274', 'Lisa Brown', 'lisa brown')<br/>('1690709', 'Arun Hampapur', 'arun hampapur')<br/>('1752714', 'Matthew Turk', 'matthew turk')</td><td>daniel@cs.ucsb.edu +<br/>rsferis@us.ibm.com +<br/>lisabr@us.ibm.com +<br/>arunh@us.ibm.com +<br/>mturk@cs.ucsb.edu +</td></tr><tr><td>1462bc73834e070201acd6e3eaddd23ce3c1a114</td><td>International Journal of Science, Engineering and Technology Research (IJSETR), Volume 3, Issue 4, April 2014 +<br/>FACE AUTHENTICATION /RECOGNITION +<br/>SYSTEM FOR FORENSIC APPLICATION +<br/>USING SKETCH BASED ON THE SIFT +<br/>FEATURES APPROACH +<br/>Department of Electronics Engineering KITS, +<br/><b>RTMNU Nagpur University, India</b><br/> +</td><td></td><td></td></tr><tr><td>14014a1bdeb5d63563b68b52593e3ac1e3ce7312</td><td>ALNAJAR et al.: EXPRESSION-INVARIANT AGE ESTIMATION +<br/>Expression-Invariant Age Estimation +<br/>Jose Alvarez2 +<br/><b>ISLA Lab, Informatics Institute</b><br/><b>University of Amsterdam</b><br/>Amsterdam, The Netherlands +<br/>2 NICTA +<br/>Canberra ACT 2601 +<br/>Australia +</td><td>('1765602', 'Fares Alnajar', 'fares alnajar')<br/>('39793067', 'Zhongyu Lou', 'zhongyu lou')<br/>('1695527', 'Theo Gevers', 'theo gevers')</td><td>F.alnajar@uva.nl +<br/>z.lou@uva.nl +<br/>jose.alvarez@nicta.com.au +<br/>th.gevers@uva.nl +</td></tr><tr><td>1473a233465ea664031d985e10e21de927314c94</td><td></td><td></td><td></td></tr><tr><td>140c95e53c619eac594d70f6369f518adfea12ef</td><td>Pushing the Frontiers of Unconstrained Face Detection and Recognition: IARPA Janus Benchmark A +<br/>The development of accurate and scalable unconstrained face recogni- +<br/>tion algorithms is a long term goal of the biometrics and computer vision +<br/>communities. The term “unconstrained” implies a system can perform suc- +<br/>cessful identifications regardless of face image capture presentation (illumi- +<br/>nation, sensor, compression) or subject conditions (facial pose, expression, +<br/>occlusion). While automatic, as well as human, face identification in certain +<br/>scenarios may forever be elusive, such as when a face is heavily occluded or +<br/>captured at very low resolutions, there still remains a large gap between au- +<br/>tomated systems and human performance on familiar faces. In order to close +<br/>this gap, large annotated sets of imagery are needed that are representative +<br/>of the end goals of unconstrained face recognition. This will help continue +<br/>to push the frontiers of unconstrained face detection and recognition, which +<br/>are the primary goals of the IARPA Janus program. +<br/>The current state of the art in unconstrained face recognition is high +<br/>accuracy (roughly 99% true accept rate at a false accept rate of 1.0%) on +<br/>faces that can be detected with a commodity face detectors, but unknown +<br/>accuracy on other faces. Despite the fact that face detection and recognition +<br/>research generally has advanced somewhat independently, the frontal face +<br/>detector filtering approach used for key in the wild face recognition datasets +<br/>means that progress in face recognition is currently hampered by progress +<br/>in face detection. Hence, a major need exists for a face recognition dataset +<br/>that captures as wide of a range of variations as possible to offer challenges +<br/>to both face detection as well as face recognition. +<br/>In this paper we introduce the IARPA Janus Benchmark A (IJB-A), +<br/>which is publicly available for download. The IJB-A contains images and +<br/>videos from 500 subjects captured from “in the wild” environment. All la- +<br/>belled subjects have been manually localized with bounding boxes for face +<br/>detection, as well as fiducial landmarks for the center of the two eyes (if +<br/>visible) and base of the nose. Manual bounding box annotations for all non- +<br/>labelled subjects (i.e., other persons captured in the imagery) have been cap- +<br/>tured as well. All imagery is Creative Commons licensed, which is a license +<br/>that allows open re-distribution provided proper attribution is made to the +<br/>data creator. The subjects have been intentionally sampled to contain wider +<br/>geographic distribution than previous datasets. Recognition and detection +<br/>protocols are provided which are motivated by operational deployments of +<br/>face recognition systems. An example of images and video from IJB-A can +<br/>be found in Figure 3. +<br/>The IJB-A dataset has the following claimed contributions: (i) The most +<br/>unconstrained database released to date; (ii) The first joint face detection and +<br/>face recognition benchmark dataset collected in the wild; (iii) Meta-data +<br/>providing subject gender and skin color, and occlusion (eyes, mouth/nose, +<br/>and forehead), facial hear, and coarse pose information for each imagery +<br/>instance; (iv) Widest geographic distribution of any public face dataset; (v) +<br/>The first in the wild dataset to contain a mixture of images and videos; (vi) +<br/>Clear authority for re-distribution; (vii) Protocols for identification (search) +<br/>and verification (compare); (viii) Baseline accuracies from off the shelf de- +<br/>tectors and recognition algorithms; and (ix) Protocols for both template and +<br/>model-based face recognition. +<br/>Every subject in the dataset contains at least five images and one video. +<br/>IJB-A consists of a total of 5,712 images and 2,085 videos, with an average +<br/>of 11.4 images and 4.2 videos per subject. +</td><td>('1885566', 'Emma Taborsky', 'emma taborsky')<br/>('1917247', 'Austin Blanton', 'austin blanton')<br/>('39403529', 'Jordan Cheney', 'jordan cheney')<br/>('2040584', 'Kristen Allen', 'kristen allen')<br/>('2136478', 'Patrick Grother', 'patrick grother')<br/>('2578654', 'Alan Mah', 'alan mah')<br/>('6680444', 'Anil K. Jain', 'anil k. jain')</td><td></td></tr><tr><td>14418ae9a6a8de2b428acb2c00064da129632f3e</td><td>Discovering the Spatial Extent of Relative Attributes +<br/><b>University of California Davis</b><br/>Introduction +<br/>Visual attributes are human-nameable object properties that serve as an in- +<br/>termediate representation between low-level image features and high-level +<br/>objects or scenes [3, 4, 5]. They can offer a great gateway for human- +<br/>object interaction. For example, when we want to interact with an unfa- +<br/>miliar object, it is likely that we first infer its attributes from its appear- +<br/>ance (e.g., is it furry or slippery?) and then decide how to interact with +<br/>it. Thus, modelling visual attributes would be valuable for understanding +<br/>human-object interactions. Researchers have developed systems that model +<br/>binary attributes [3, 4, 5]—a property’s presence/absence (e.g., “is furry/not +<br/>furry”)—and relative attributes [6, 8]—a property’s relative strength (e.g., +<br/>“furrier than”). In this work, we focus on relative attributes since they of- +<br/>ten describe object properties better than binary ones [6], especially if the +<br/>property exhibits large appearance variations (see Fig. 1). +<br/>While most existing work use global image representations to model +<br/>attributes (e.g., [5, 6]), recent work demonstrates the effectiveness of using +<br/>localized part-based representations [1, 7, 9]. They show that attributes—be +<br/>it global (“is male”) or local (“smiling”)—can be more accurately learned +<br/>by first bringing the underlying object-parts into correspondence, and then +<br/>modeling the attributes conditioned on those object-parts. To compute such +<br/>correspondences, pre-trained part detectors are used (e.g., faces [7] and peo- +<br/>ple [1, 9]). However, because the part detectors are trained independently of +<br/>the attribute, the learned parts may not necessarily be useful for modeling +<br/>the desired attribute. Furthermore, some objects do not naturally have well- +<br/>defined parts, which means modeling the part-based detector itself becomes +<br/>a challenge. The approach of [2] address these issues by discovering useful +<br/>and localized attributes. However, it requires a human-in-the-loop, which +<br/>limits its scalability. +<br/>So, how can we develop robust visual representations for relative at- +<br/>tributes, without expensive and potentially uninformative pre-trained part +<br/>detectors or humans-in-the-loop? To do so, we will need to automatically +<br/>identify the visual patterns in each image whose appearance correlates with +<br/>attribute strength. +<br/>In this work, we propose a method that automatically +<br/>discovers the spatial extent of relative attributes in images across varying at- +<br/>tribute strengths. The main idea is to leverage the fact that the visual concept +<br/>underlying the attribute undergos a gradual change in appearance across +<br/>the attribute spectrum. In this way, we propose to discover a set of local, +<br/>transitive connections (“visual chains”) that establish correspondences be- +<br/>tween the same object-part, even when its appearance changes drastically +<br/>over long ranges. Given the candidate set of visual chains, we then automat- +<br/>ically select those that together best model the changing appearance of the +<br/>attribute across the attribute spectrum. Importantly, by combining a subset +<br/>of the most-informative discovered visual chains, our approach aims to dis- +<br/>cover the full spatial extent of the attribute, whether it be concentrated on a +<br/>particular object-part or spread across a larger spatial area. +<br/>2 Approach +<br/>Given an image collection S={I1, . . . ,IN} with pairwise ordered and un- +<br/>ordered image-level relative comparisons of an attribute (i.e., in the form of +<br/>Ω(Ii)>Ω(Ij) and Ω(Ii)≈Ω(Ij), where i, j∈{1, . . . ,N} and Ω(Ii) is Ii’s at- +<br/>tribute strength), our goal is to discover the spatial extent of the attribute in +<br/>each image and learn a ranking function that predicts the attribute strength +<br/>for any new image. +<br/>There are three main steps to our approach: (1) initializing a candidate +<br/>set of visual chains; (2) iteratively growing each visual chain along the at- +<br/>tribute spectrum; and (3) ranking the chains according to their relevance to +<br/>the target attribute to create an ensemble image representation. +<br/>Initializing candidate visual chains: A visual attribute can potentially +<br/>exhibit large appearance variations across the attribute spectrum. Take the +<br/>(top) Given pairs of images, each ordered according to rela- +<br/>Figure 1: +<br/>tive attribute strength (e.g., “higher/lower-at-the-heel”), (bottom) our ap- +<br/>proach automatically discovers the attribute’s spatial extent in each image, +<br/>and learns a ranking function that orders the image collection according to +<br/>predicted attribute strength. +<br/>high-at-the-heel attribute as an example: high-heeled shoes have strong +<br/>vertical gradients while flat-heeled shoes have strong horizontal gradients. +<br/>However, the attribute’s appearance will be quite similar in any local region +<br/>of the attribute spectrum. Therefore, we start with multiple short but visu- +<br/>ally homogeneous chains of image regions in a local region of the attribute +<br/>spectrum, and smoothly grow them out to cover the entire spectrum. +<br/>We start by first sorting the images in S in descending order of predicted +<br/>attribute strength—with ˜I1 as the strongest image and ˜IN as the weakest— +<br/>using a linear SVM-ranker trained with global image features. To initialize +<br/>a single chain, we take the top Ninit images and select a set of patches (one +<br/>from each image) whose appearance varies smoothly with its neighbors in +<br/>the chain, by minimizing the following objective function: +<br/>Ninit∑ +<br/>||φ (Pi)− φ (Pi−1)||2, +<br/>i=2 +<br/>min +<br/>C(P) = +<br/>(1) +<br/>where φ (Pi) is the appearance feature of patch Pi in ˜Ii, and P ={P1, . . . ,PNinit} +<br/>is the set of patches in a chain. Candidate patches for each image are densely +<br/>sampled at multiple scales. This objective enforces local smoothness: the +<br/>appearances of the patches in the images with neighboring indices should +<br/>vary smoothly within a chain. Given the objective’s chain structure, we can +<br/>efficiently find its global optimum using Dynamic Programming (DP). +<br/>In the backtracking stage of DP, we obtain a large number of K-best +<br/>solutions. We then perform a chain-level non-maximum-suppression (NMS) +<br/>to remove redundant chains to retain a set of Kinit diverse candidate chains. +<br/>Iteratively growing each visual chain: The initial set of Kinit chains are +<br/>visually homogeneous but cover only a tiny fraction of the attribute spec- +<br/>trum. We next iteratively grow each chain to cover the entire attribute spec- +<br/>trum by training a model that adapts to the attribute’s smoothly changing +<br/>appearance. Specifically, for each chain, we iteratively train a detector and +<br/>in each iteration and use it to grow the chain while simultaneously refining +<br/>it. To grow the chain, we again minimize Eqn. 1 but now with an additional +<br/>term: +<br/>t∗Niter∑ +<br/>t∗Niter∑ +<br/>wT +<br/>t φ (Pi), +<br/>||φ (Pi)− φ (Pi−1)||2 − λ +<br/>i=2 +<br/>i=1 +<br/>min +<br/>C(P) = +<br/>(2) +<br/>where wt is a linear SVM detector learned from the patches in the chain +<br/>from the (t−1)-th iteration, P = {P1, . . . ,Pt∗Niter} is the set of patches in a +<br/>chain, and Niter is the number of images considered in each iteration. As +<br/>before, the first term enforces local smoothness. The second term is the +<br/>detection term: since the ordering of the images in the chain is only a rough +<br/>estimate and thus possibly noisy, wt prevents the inference from drifting in +<br/>the cases where local smoothness does not strictly hold. λ is a constant that +<br/>trades-off the two terms. We use the same DP inference procedure used to +<br/>optimize Eqn. 1. +<br/>Once P is found, we train a new detector with all of its patches as posi- +<br/>tive instances. The negative instances consist of randomly sampled patches +<br/>strongweak,Attribute: “high-at-the-heel”,, </td><td>('2299381', 'Fanyi Xiao', 'fanyi xiao')<br/>('1883898', 'Yong Jae Lee', 'yong jae lee')</td><td></td></tr><tr><td>14ba910c46d659871843b31d5be6cba59843a8b8</td><td>Face Recognition in Movie Trailers via Mean Sequence Sparse +<br/>Representation-based Classification +<br/><b>Center for Research in Computer Vision, University of Central Florida, Orlando, FL</b></td><td>('16131262', 'Enrique G. Ortiz', 'enrique g. ortiz')<br/>('2003981', 'Alan Wright', 'alan wright')<br/>('1745480', 'Mubarak Shah', 'mubarak shah')</td><td>eortiz@cs.ucf.edu, alanwright@knights.ucf.edu, shah@crcv.ucf.edu +</td></tr><tr><td>1467c4ab821c3b340abe05a1b13a19318ebbce98</td><td>Multitask and Transfer Learning for +<br/>Multi-Aspect Data +<br/>Bernardino Romera Paredes +<br/>UCL +<br/>A dissertation submitted in partial fulfillment +<br/>of the requirements for the degree of +<br/><b>Doctor of Philosophy of University College London</b></td><td></td><td></td></tr><tr><td>14318d2b5f2cf731134a6964d8193ad761d86942</td><td>FaceDNA: Intelligent Face Recognition +<br/>System with Intel RealSense 3D Camera +<br/><b>National Taiwan University</b><br/> +</td><td>('1678531', 'Dan Ye', 'dan ye')<br/>('40063567', 'Shih-Wei Liao', 'shih-wei liao')</td><td></td></tr><tr><td>142dcfc3c62b1f30a13f1f49c608be3e62033042</td><td>Adaptive Region Pooling for Object Detection +<br/>UC Merced +<br/>Qualcomm Research, San Diego +<br/>UC Merced +</td><td>('2580349', 'Yi-Hsuan Tsai', 'yi-hsuan tsai')<br/>('1872879', 'Onur C. Hamsici', 'onur c. hamsici')<br/>('1715634', 'Ming-Hsuan Yang', 'ming-hsuan yang')</td><td>ytsai2@ucmerced.edu +<br/>ohamsici@qti.qualcomm.com +<br/>mhyang@ucmerced.edu +</td></tr><tr><td>14c0f9dc9373bea1e27b11fa0594c86c9e632c8d</td><td>Adaptive Exponential Smoothing for Online Filtering of Pixel Prediction Maps +<br/>School of Electrical and Electronic Engineering, +<br/><b>Nanyang Technological University, Singapore</b></td><td>('3064975', 'Kang Dang', 'kang dang')<br/>('1691251', 'Jiong Yang', 'jiong yang')<br/>('34316743', 'Junsong Yuan', 'junsong yuan')</td><td>{dang0025, yang0374}@e.ntu.edu.sg, jsyuan@ntu.edu.sg +</td></tr><tr><td>1439bf9ba7ff97df9a2da6dae4784e68794da184</td><td>LGE-KSVD: Flexible Dictionary Learning for Optimized Sparse +<br/>Representation Classification +<br/>Raymond Ptucha +<br/><b>Rochester Institute of Technology</b><br/>Rochester, NY, USA +</td><td></td><td>rwpeec@rit.edu +</td></tr><tr><td>141768ab49a5a9f5adcf0cf7e43a23471a7e5d82</td><td>Relative Facial Action Unit Detection +<br/>Department of Computing and Software +<br/><b>McMaster University</b><br/>Hamilton, Canada +</td><td>('1736464', 'Mahmoud Khademi', 'mahmoud khademi')</td><td>khademm@mcmaster.ca +</td></tr><tr><td>14e428f2ff3dc5cf96e5742eedb156c1ea12ece1</td><td>Facial Expression Recognition Using Neural Network Trained with Zernike +<br/>Moments +<br/>Dept. Génie-Electrique +<br/>Université M.C.M Souk-Ahras +<br/>Souk-Ahras, Algeria +</td><td>('3112602', 'Mohammed Saaidia', 'mohammed saaidia')</td><td>mohamed.saaidia@univ-soukahras.dz +</td></tr><tr><td>14bca107bb25c4dce89210049bf39ecd55f18568</td><td>X.HUANG:EMOTIONRECOGNITIONFROMFACIALIMAGES +<br/>Emotion recognition from facial images with +<br/>arbitrary views +<br/>Center for Machine Vision Research +<br/>Department of Computer Science and +<br/>Engineering +<br/><b>University of Oulu</b><br/>Oulu, Finland +</td><td>('18780812', 'Xiaohua Huang', 'xiaohua huang')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')<br/>('1714724', 'Matti Pietikäinen', 'matti pietikäinen')</td><td>huang.xiaohua@ee.oulu.fi +<br/>gyzhao@ee.oulu.fi +<br/>mkp@ee.oulu.fi +</td></tr><tr><td>14a5feadd4209d21fa308e7a942967ea7c13b7b6</td><td>978-1-4673-0046-9/12/$26.00 ©2012 IEEE +<br/>1025 +<br/>ICASSP 2012 +</td><td></td><td></td></tr><tr><td>14fee990a372bcc4cb6dc024ab7fc4ecf09dba2b</td><td>Modeling Spatio-Temporal Human Track Structure for Action +<br/>Localization +</td><td>('2926143', 'Anton Osokin', 'anton osokin')</td><td></td></tr><tr><td>14ee4948be56caeb30aa3b94968ce663e7496ce4</td><td>Jang, Y; Gunes, H; Patras, I +<br/>© Copyright 2018 IEEE +<br/>For additional information about this publication click this link. +<br/>http://qmro.qmul.ac.uk/xmlui/handle/123456789/36405 +<br/>Information about this research object was correct at the time of download; we occasionally +<br/>make corrections to records, please therefore check the published record when citing. For +</td><td></td><td>more information contact scholarlycommunications@qmul.ac.uk +</td></tr><tr><td>8ec82da82416bb8da8cdf2140c740e1574eaf84f</td><td>CHUNG AND ZISSERMAN: BMVC AUTHOR GUIDELINES +<br/>Lip Reading in Profile +<br/>http://www.robots.ox.ac.uk/~joon +<br/>http://www.robots.ox.ac.uk/~az +<br/>Visual Geometry Group +<br/>Department of Engineering Science +<br/><b>University of Oxford</b><br/>Oxford, UK +</td><td>('2863890', 'Joon Son Chung', 'joon son chung')<br/>('1688869', 'Andrew Zisserman', 'andrew zisserman')</td><td></td></tr><tr><td>8ee62f7d59aa949b4a943453824e03f4ce19e500</td><td>Robust Head-Pose Estimation Based on +<br/>Partially-Latent Mixture of Linear Regression +<br/>∗INRIA Grenoble Rhˆone-Alpes, Montbonnot Saint-Martin, France +<br/>†INRIA Rennes Bretagne Atlantique, Rennes, France +</td><td>('2188660', 'Vincent Drouard', 'vincent drouard')<br/>('1794229', 'Radu Horaud', 'radu horaud')<br/>('3307172', 'Antoine Deleforge', 'antoine deleforge')<br/>('1690536', 'Georgios Evangelidis', 'georgios evangelidis')</td><td></td></tr><tr><td>8e0ede53dc94a4bfcf1238869bf1113f2a37b667</td><td>Joint Patch and Multi-label Learning for Facial Action Unit Detection +<br/><b>School of Comm. and Info. Engineering, Beijing University of Posts and Telecom., Beijing China</b><br/><b>Robotics Institute, Carnegie Mellon University, Pittsburgh, PA</b><br/><b>University of Pittsburgh, Pittsburgh, PA</b></td><td>('2393320', 'Kaili Zhao', 'kaili zhao')<br/>('1720776', 'Honggang Zhang', 'honggang zhang')</td><td></td></tr><tr><td>8e33183a0ed7141aa4fa9d87ef3be334727c76c0</td><td>– COS429 Written Report, Fall 2017 – +<br/>Robustness of Face Recognition to Image Manipulations +<br/>1. Motivation +<br/>We can often recognize pictures of people we know even if the image has low resolution or obscures +<br/>part of the face, if the camera angle resulted in a distorted image of the subject’s face, or if the +<br/>subject has aged or put on makeup since we last saw them. Although this is a simple recognition task +<br/>for a human, when we think about how we accomplish this task, it seems non-trivial for computer +<br/>algorithms to recognize faces despite visual changes. +<br/>Computer facial recognition is relied upon for many application where accuracy is important. +<br/>Facial recognition systems have applications ranging from airport security and suspect identification +<br/>to personal device authentication and face tagging [7]. In these real-world applications, the system +<br/>must continue to recognize images of a person who looks slightly different due to the passage of +<br/>time, a change in environment, or a difference in clothing. +<br/>Therefore, we are interested in investigating face recognition algorithms and their robustness to +<br/>image changes resulting from realistically plausible manipulations. Furthermore, we are curious +<br/>about whether the impact of image manipulations on computer algorithms’ face recognition ability +<br/>mirrors related insights from neuroscience about humans’ face recognition abilities. +<br/>2. Goal +<br/>In this project, we implement both face recognition algorithms and image manipulations. We then +<br/>analyze the impact of each image manipulation on the recognition accuracy each algorithm, and +<br/>how these influences depend on the accuracy of each algorithm on non-manipulated images. +<br/>3. Background and Related Work +<br/>Researchers have developed a wide variety of face recognition algorithms, such as traditional +<br/>statistical methods such as PCA, more opaque methods such as deep neural networks, and proprietary +<br/>systems used by governments and corporations [1][13][14]. +<br/>Similarly, others have developed image manipulations using principles from linear algebra, such +<br/>as mimicking distortions from lens distortions, as well as using neural networks, such as a system +<br/>for transforming images according to specified characteristics [12][16]. +<br/>Furthermore, researchers in psychology have studied face recognition in humans. A study of +<br/>“super-recognizers” (people with extraordinarily high powers of face recognition) and “developmen- +<br/>tal prosopagnosics” (people with severely impaired face recognition abilities) found that inverting +<br/>images of faces impaired recognition ability more for people with stronger face recognition abilities +<br/>[11]. This could indicate that image manipulations tend to equalize face recognition abilities, and +<br/>we investigate whether this is the case with the manipulations and face recognition algorithms we +<br/>test. +</td><td>('1897270', 'Cathy Chen', 'cathy chen')</td><td></td></tr><tr><td>8e3d0b401dec8818cd0245c540c6bc032f169a1d</td><td>McGan: Mean and Covariance Feature Matching GAN +</td><td>('2211263', 'Youssef Mroueh', 'youssef mroueh')</td><td></td></tr><tr><td>8e3c97e420e0112c043929087d6456d8ab61e95c</td><td>SAFDARNEJAD et al.: ROBUST GLOBAL MOTION COMPENSATION +<br/>Robust Global Motion Compensation in +<br/>Presence of Predominant Foreground +<br/>https://www.msu.edu/~safdarne/ +<br/>http://www.cse.msu.edu/~liuxm/ +<br/>http://www.egr.msu.edu/ndel/profile/lalita-udpa +<br/><b>Michigan State University</b><br/>East Lansing +<br/>Michigan, USA +</td><td>('2941187', 'Seyed Morteza Safdarnejad', 'seyed morteza safdarnejad')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')<br/>('1938832', 'Lalita Udpa', 'lalita udpa')</td><td></td></tr><tr><td>8e0ab1b08964393e4f9f42ca037220fe98aad7ac</td><td>UV-GAN: Adversarial Facial UV Map Completion for Pose-invariant Face +<br/>Recognition +<br/><b>Imperial College London</b></td><td>('3234063', 'Jiankang Deng', 'jiankang deng')<br/>('1902288', 'Shiyang Cheng', 'shiyang cheng')<br/>('4091869', 'Niannan Xue', 'niannan xue')<br/>('47943220', 'Yuxiang Zhou', 'yuxiang zhou')</td><td>j.deng16, shiyang.cheng11,n.xue15,yuxiang.zhou10,s.zafeiriou@imperial.ac.uk +</td></tr><tr><td>8e94ed0d7606408a0833e69c3185d6dcbe22bbbe</td><td>© 2012 IEEE. Personal use of this material is permitted. Permission from IEEE +<br/>must be obtained for all other uses, in any current or future media, including +<br/>reprinting/republishing this material for advertising or promotional purposes, +<br/>creating new collective works, for resale or redistribution to servers or lists, or +<br/>reuse of any copyrighted component of this work in other works. +<br/>Pre-print of article that will appear at WACV 2012. +</td><td></td><td></td></tr><tr><td>8e461978359b056d1b4770508e7a567dbed49776</td><td>LOMo: Latent Ordinal Model for Facial Analysis in Videos +<br/>Marian Bartlett1,∗,‡ +<br/>1UCSD, USA +<br/>2MPI for Informatics, Germany +<br/>3IIT Kanpur, India +</td><td>('39707211', 'Karan Sikka', 'karan sikka')<br/>('39396475', 'Gaurav Sharma', 'gaurav sharma')</td><td></td></tr><tr><td>8e4808e71c9b9f852dc9558d7ef41566639137f3</td><td>Adversarial Generative Nets: Neural Network +<br/>Attacks on State-of-the-Art Face Recognition +<br/><b>Carnegie Mellon University</b><br/><b>University of North Carolina at Chapel Hill</b></td><td>('36301492', 'Mahmood Sharif', 'mahmood sharif')<br/>('38181360', 'Sruti Bhagavatula', 'sruti bhagavatula')<br/>('38572260', 'Lujo Bauer', 'lujo bauer')<br/>('1746214', 'Michael K. Reiter', 'michael k. reiter')</td><td>{mahmoods, srutib, lbauer}@cmu.edu +<br/>reiter@cs.unc.edu +</td></tr><tr><td>8ea30ade85880b94b74b56a9bac013585cb4c34b</td><td>FROM TURBO HIDDEN MARKOV MODELS TO TURBO STATE-SPACE MODELS +<br/>Institut Eur´ecom +<br/>Multimedia Communications Department +<br/>BP 193, 06904 Sophia Antipolis Cedex, France +</td><td>('1723883', 'Florent Perronnin', 'florent perronnin')<br/>('1709849', 'Jean-Luc Dugelay', 'jean-luc dugelay')</td><td>fflorent.perronnin, jean-luc.dugelayg@eurecom.fr +</td></tr><tr><td>8ed32c8fad924736ebc6d99c5c319312ba1fa80b</td><td></td><td></td><td></td></tr><tr><td>8e0ad1ccddc7ec73916eddd2b7bbc0019d8a7958</td><td>Segment-based SVMs for +<br/>Time Series Analysis +<br/>CMU-RI-TR-12-1 +<br/>Submitted in partial fulfillment of the +<br/>requirements for the degree of +<br/>Doctor of Philosophy in Robotics +<br/><b>The Robotics Institute</b><br/><b>Carnegie Mellon University</b><br/>Pittsburgh, Pennsylvania 15213 +<br/>Version: 20 Jan 2012 +<br/>Thesis Committee: +<br/>Fernando De la Torre (chair) +</td><td>('1698158', 'Minh Hoai Nguyen', 'minh hoai nguyen')<br/>('1709305', 'Martial Hebert', 'martial hebert')<br/>('1730156', 'Carlos Guestrin', 'carlos guestrin')<br/>('2038264', 'Frank Dellaert', 'frank dellaert')<br/>('1698158', 'Minh Hoai Nguyen', 'minh hoai nguyen')</td><td></td></tr><tr><td>8e8e3f2e66494b9b6782fb9e3f52aeb8e1b0d125</td><td>in any current or +<br/>future media, +<br/>for all other uses, +<br/> 2012 IEEE. Personal use of this material is permitted. Permission from IEEE must be +<br/>obtained +<br/>including +<br/>reprinting/republishing this material for advertising or promotional purposes, creating +<br/>new collective works, for resale or redistribution to servers or lists, or reuse of any +<br/>copyrighted component of this work in other works. +<br/>Pre-print of article that will appear at BTAS 2012.!! +</td><td></td><td></td></tr><tr><td>8e378ef01171b33c59c17ff5798f30293fe30686</td><td>Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +<br/>der Technischen Universit¨at M¨unchen +<br/>A System for Automatic Face Analysis +<br/>Based on +<br/>Statistical Shape and Texture Models +<br/>Ronald M¨uller +<br/>Vollst¨andiger Abdruck der von der Fakult¨at +<br/>f¨ur Elektrotechnik und Informationstechnik +<br/>der Technischen Universit¨at M¨unchen +<br/>zur Erlangung des akademischen Grades eines +<br/>Doktor-Ingenieurs +<br/>genehmigten Dissertation +<br/>Vorsitzender: Prof. Dr. rer. nat. Bernhard Wolf +<br/>Pr¨ufer der Dissertation: +<br/>1. Prof. Dr.-Ing. habil. Gerhard Rigoll +<br/>2. Prof. Dr.-Ing. habil. Alexander W. Koch +<br/>Die Dissertation wurde am 28.02.2008 bei der Technischen Universit¨at M¨unchen +<br/>eingereicht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik +<br/>am 18.09.2008 angenommen. +</td><td></td><td></td></tr><tr><td>8ed051be31309a71b75e584bc812b71a0344a019</td><td>Class-based feature matching across unrestricted +<br/>transformations +</td><td>('1938475', 'Evgeniy Bart', 'evgeniy bart')<br/>('1743045', 'Shimon Ullman', 'shimon ullman')</td><td></td></tr><tr><td>8e36100cb144685c26e46ad034c524b830b8b2f2</td><td>Modeling Facial Geometry using Compositional VAEs +<br/>1 ´Ecole Polytechnique F´ed´erale de Lausanne +<br/>2Facebook Reality Labs, Pittsburgh +</td><td>('33846296', 'Chenglei Wu', 'chenglei wu')<br/>('14373499', 'Jason Saragih', 'jason saragih')<br/>('1717736', 'Pascal Fua', 'pascal fua')<br/>('1774867', 'Yaser Sheikh', 'yaser sheikh')</td><td>{firstname.lastname}@epfl.ch, {firstname.lastname}@fb.com +</td></tr><tr><td>8ed33184fccde677ec8413ae06f28ea9f2ca70f3</td><td>Multimodal Visual Concept Learning with Weakly Supervised Techniques +<br/><b>School of E.C.E., National Technical University of Athens, Greece</b></td><td>('7311172', 'Giorgos Bouritsas', 'giorgos bouritsas')<br/>('2539459', 'Petros Koutras', 'petros koutras')<br/>('2641229', 'Athanasia Zlatintsi', 'athanasia zlatintsi')<br/>('1750686', 'Petros Maragos', 'petros maragos')</td><td>gbouritsas@gmail.com, {pkoutras, nzlat, maragos}@cs.ntua.gr +</td></tr><tr><td>8ee5b1c9fb0bded3578113c738060290403ed472</td><td>Extending Explicit Shape Regression with +<br/>Mixed Feature Channels and Pose Priors +<br/><b>Karlsruhe Institute of</b><br/>Technology (KIT) +<br/>Karlsruhe, Germany +<br/>Hazım Kemal Ekenel +<br/>´Ecole Polytechnique F´ed´erale +<br/>de Lausanne (EPFL) +<br/>Lausanne, Switzerland +<br/>Istanbul Technical +<br/><b>University (ITU</b><br/>Istanbul, Turkey +</td><td>('39610204', 'Matthias Richter', 'matthias richter')<br/>('1697965', 'Hua Gao', 'hua gao')</td><td>matthias.richter@kit.edu +<br/>hua.gao@epfl.ch +<br/>ekenel@itu.edu.tr +</td></tr><tr><td>8e0becfc5fe3ecdd2ac93fabe34634827b21ef2b</td><td>International Journal of Computer Vision manuscript No. +<br/>(will be inserted by the editor) +<br/>Learning from Longitudinal Face Demonstration - +<br/>Where Tractable Deep Modeling Meets Inverse Reinforcement Learning +<br/>Savvides · Tien D. Bui +<br/>Received: date / Accepted: date +</td><td>('1876581', 'Chi Nhan Duong', 'chi nhan duong')</td><td></td></tr><tr><td>8efda5708bbcf658d4f567e3866e3549fe045bbb</td><td>Pre-trained Deep Convolutional Neural Networks +<br/>for Face Recognition +<br/>Siebert Looije +<br/>S2209276 +<br/>January 2018 +<br/>MSc. Thesis +<br/>Artificial Intelligence +<br/><b>University of Groningen, The Netherlands</b><br/>Supervisors +<br/>Dr. M.A. (Marco) Wiering +<br/>K. (Klaas) Dijkstra, MSc. +<br/><b>ALICE Institute</b><br/><b>University of Groningen</b><br/>Nijenborgh 9, 9747 AG, Groningen, The Netherlands +<br/><b>facultyofmathematicsandnaturalsciencesarti cialintelligence22-09-2016|1ATitleA.UthorRijksuniversiteitGroningenSomeFaculty</b></td><td></td><td></td></tr><tr><td>2227f978f084ebb18cb594c0cfaf124b0df6bf95</td><td>Pillar Networks for action recognition +<br/>B Sengupta +<br/>Cortexica Vision Systems Limited +<br/><b>Imperial College London</b><br/>London, UK +<br/>Y Qian +<br/>Cortexica Vision Systems Limited +<br/>30 Stamford Street SE1 9LQ +<br/>London, UK +</td><td></td><td>b.sengupta@imperial.ac.uk +<br/>yu.qian@cortexica.com +</td></tr><tr><td>225fb9181545f8750061c7693661b62d715dc542</td><td></td><td></td><td></td></tr><tr><td>22043cbd2b70cb8195d8d0500460ddc00ddb1a62</td><td>Separability-Oriented Subclass Discriminant +<br/>Analysis +</td><td>('2986129', 'Huan Wan', 'huan wan')<br/>('27838939', 'Hui Wang', 'hui wang')<br/>('35009947', 'Gongde Guo', 'gongde guo')<br/>('10803956', 'Xin Wei', 'xin wei')</td><td></td></tr><tr><td>22137ce9c01a8fdebf92ef35407a5a5d18730dde</td><td></td><td></td><td></td></tr><tr><td>22e2066acfb795ac4db3f97d2ac176d6ca41836c</td><td>Coarse-to-Fine Auto-Encoder Networks (CFAN) +<br/>for Real-Time Face Alignment +<br/>1 Key Lab of Intelligent Information Processing of Chinese Academy of Sciences +<br/><b>CAS), Institute of Computing Technology, CAS, Beijing 100190, China</b><br/><b>University of Chinese Academy of Sciences, Beijing 100049, China</b></td><td>('1698586', 'Jie Zhang', 'jie zhang')<br/>('1685914', 'Shiguang Shan', 'shiguang shan')<br/>('1693589', 'Meina Kan', 'meina kan')<br/>('1710220', 'Xilin Chen', 'xilin chen')</td><td>{jie.zhang,shiguang.shan,meina.kan,xilin.chen}@vipl.ict.ac.cn +</td></tr><tr><td>22717ad3ad1dfcbb0fd2f866da63abbde9af0b09</td><td>A Learning-based Control Architecture for Socially +<br/>Assistive Robots Providing Cognitive Interventions +<br/>by +<br/>A thesis submitted in conformity with the requirements +<br/>for the degree of Masters of Applied Science +<br/>Mechanical and Industrial Engineering +<br/><b>University of Toronto</b></td><td>('39999379', 'Jeanie Chan', 'jeanie chan')<br/>('39999379', 'Jeanie Chan', 'jeanie chan')</td><td></td></tr><tr><td>2288696b6558b7397bdebe3aed77bedec7b9c0a9</td><td>WU, WANG, YANG, JI: JOINT ATTENTION ON MULTI-LEVEL DEEP FEATURES 1 +<br/>Action Recognition with Joint Attention +<br/>on Multi-Level Deep Features +<br/>Dept of Automation +<br/><b>Tsinghua University</b><br/>Beijing, China +</td><td>('35585536', 'Jialin Wu', 'jialin wu')<br/>('29644358', 'Gu Wang', 'gu wang')<br/>('3432961', 'Wukui Yang', 'wukui yang')<br/>('7807689', 'Xiangyang Ji', 'xiangyang ji')</td><td>wujl13@mails.tsinghua.edu.cn +<br/>wangg12@mails.tsinghua.edu.cn +<br/>yang-wk15@mails.tsinghua.edu.cn +<br/>xyji@mail.tsinghua.edu.cn +</td></tr><tr><td>22264e60f1dfbc7d0b52549d1de560993dd96e46</td><td>UnitBox: An Advanced Object Detection Network +<br/>Thomas Huang1 +<br/><b>University of Illinois at Urbana Champaign</b><br/>2Megvii Inc +</td><td>('3451838', 'Jiahui Yu', 'jiahui yu')<br/>('1691963', 'Yuning Jiang', 'yuning jiang')<br/>('2969311', 'Zhangyang Wang', 'zhangyang wang')<br/>('2695115', 'Zhimin Cao', 'zhimin cao')</td><td>{jyu79, zwang119, t-huang1}@illinois.edu, {jyn, czm}@megvii.com +</td></tr><tr><td>22dada4a7ba85625824489375184ba1c3f7f0c8f</td><td></td><td></td><td></td></tr><tr><td>221252be5d5be3b3e53b3bbbe7a9930d9d8cad69</td><td>ZHU, VONDRICK, RAMANAN, AND FOWLKES: MORE DATA OR BETTER MODELS +<br/>Do We Need More Training Data or Better +<br/>Models for Object Detection? +<br/>1 Computer Science Department +<br/><b>University of California</b><br/>Irvine, CA, USA +<br/>2 CSAIL +<br/><b>Massachusetts Institute of Technology</b><br/>Cambridge, MA, USA +<br/>(Work performed while at UC Irvine) +</td><td>('32542103', 'Xiangxin Zhu', 'xiangxin zhu')<br/>('1856025', 'Carl Vondrick', 'carl vondrick')<br/>('1770537', 'Deva Ramanan', 'deva ramanan')<br/>('3157443', 'Charless C. Fowlkes', 'charless c. fowlkes')</td><td>xzhu@ics.uci.edu +<br/>vondrick@mit.edu +<br/>dramanan@ics.uci.edu +<br/>fowlkes@ics.uci.edu +</td></tr><tr><td>223ec77652c268b98c298327d42aacea8f3ce23f</td><td>TR-CS-11-02 +<br/>Acted Facial Expressions In The Wild +<br/>Database +<br/>September 2011 +<br/>ANU Computer Science Technical Report Series +</td><td>('1735697', 'Abhinav Dhall', 'abhinav dhall')<br/>('1717204', 'Roland Goecke', 'roland goecke')<br/>('27011207', 'Tom Gedeon', 'tom gedeon')</td><td></td></tr><tr><td>22df6b6c87d26f51c0ccf3d4dddad07ce839deb0</td><td>Fast Action Proposals for Human Action Detection and Search +<br/>School of Electrical and Electronic Engineering +<br/><b>Nanyang Technological University, Singapore</b></td><td>('2352391', 'Gang Yu', 'gang yu')<br/>('34316743', 'Junsong Yuan', 'junsong yuan')</td><td>iskicy@gmail.com, jsyuan@ntu.edu.sg +</td></tr><tr><td>228558a2a38a6937e3c7b1775144fea290d65d6c</td><td>Nonparametric Context Modeling of Local Appearance +<br/>for Pose- and Expression-Robust Facial Landmark Localization +<br/><b>University of Wisconsin Madison</b><br/>Zhe Lin2 +<br/>2Adobe Research +<br/>http://www.cs.wisc.edu/~lizhang/projects/face-landmark-localization/ +</td><td>('1893050', 'Brandon M. Smith', 'brandon m. smith')<br/>('1721019', 'Jonathan Brandt', 'jonathan brandt')<br/>('40396555', 'Li Zhang', 'li zhang')</td><td></td></tr><tr><td>22fdd8d65463f520f054bf4f6d2d216b54fc5677</td><td>International Journal of Emerging Technology and Advanced Engineering +<br/>Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 3, Issue 8, August 2013) +<br/>Efficient Small and Capital Handwritten Character +<br/>Recognition with Noise Reduction +<br/><b>IES College of Technology, Bhopal</b></td><td>('1926347', 'Shailendra Tiwari', 'shailendra tiwari')<br/>('2152231', 'Sandeep Kumar', 'sandeep kumar')</td><td></td></tr><tr><td>2251a88fbccb0228d6d846b60ac3eeabe468e0f1</td><td>Matrix-Based Kernel Subspace Methods +<br/>Integrated Data Systems Department +<br/>Siemens Corporate Research +<br/><b>College Road East, Princeton, NJ</b></td><td>('1682187', 'S. Kevin Zhou', 's. kevin zhou')</td><td>Email: {kzhou}@scr.siemens.com +</td></tr><tr><td>22e678d3e915218a7c09af0d1602e73080658bb7</td><td>Adventures in Archiving and Using Three Years of Webcam Images +<br/>Department of Computer Science and Engineering +<br/><b>Washington University, St. Louis, MO, USA</b></td><td>('1990750', 'Nathan Jacobs', 'nathan jacobs')<br/>('39795519', 'Walker Burgin', 'walker burgin')<br/>('1761429', 'Robert Pless', 'robert pless')</td><td>{jacobsn,wsb1,rzs1,dyr1,pless}@cse.wustl.edu +</td></tr><tr><td>2201f187a7483982c2e8e2585ad9907c5e66671d</td><td>Joint Face Alignment and 3D Face Reconstruction +<br/><b>College of Computer Science, Sichuan University, Chengdu, China</b><br/>2 Department of Computer Science and Engineering +<br/><b>Michigan State University, East Lansing, MI, U.S.A</b></td><td>('50207647', 'Feng Liu', 'feng liu')<br/>('39422721', 'Dan Zeng', 'dan zeng')<br/>('7345195', 'Qijun Zhao', 'qijun zhao')<br/>('1759169', 'Xiaoming Liu', 'xiaoming liu')</td><td></td></tr><tr><td>227b18fab568472bf14f9665cedfb95ed33e5fce</td><td>Compositional Dictionaries for Domain Adaptive +<br/>Face Recognition +</td><td>('2077648', 'Qiang Qiu', 'qiang qiu')<br/>('9215658', 'Rama Chellappa', 'rama chellappa')</td><td></td></tr><tr><td>227b1a09b942eaf130d1d84cdcabf98921780a22</td><td>Yang et al. EURASIP Journal on Advances in Signal Processing (2018) 2018:51 +<br/>https://doi.org/10.1186/s13634-018-0572-6 +<br/>EURASIP Journal on Advances +<br/>in Signal Processing +<br/>R ES EAR CH +<br/>Multi-feature shape regression for face +<br/>alignment +<br/>Open Access +</td><td>('3413708', 'Wei-jong Yang', 'wei-jong yang')<br/>('49070426', 'Yi-Chen Chen', 'yi-chen chen')<br/>('1789917', 'Pau-Choo Chung', 'pau-choo chung')<br/>('1749263', 'Jar-Ferr Yang', 'jar-ferr yang')</td><td></td></tr><tr><td>2241eda10b76efd84f3c05bdd836619b4a3df97e</td><td>One-to-many face recognition with bilinear CNNs +<br/>Aruni RoyChowdhury +<br/><b>University of Massachusetts, Amherst</b><br/>Erik Learned-Miller +</td><td>('2144284', 'Tsung-Yu Lin', 'tsung-yu lin')<br/>('35208858', 'Subhransu Maji', 'subhransu maji')</td><td>{arunirc,tsungyulin,smaji,elm}@cs.umass.edu +</td></tr><tr><td>22646cf884cc7093b0db2c1731bd52f43682eaa8</td><td>Human Action Adverb Recognition: ADHA Dataset and A Three-Stream +<br/>Hybrid Model +<br/><b>Shanghai Jiao Tong University, China</b></td><td>('1717692', 'Bo Pang', 'bo pang')<br/>('15376265', 'Kaiwen Zha', 'kaiwen zha')<br/>('1830034', 'Cewu Lu', 'cewu lu')</td><td>pangbo@sjtu.edu.cn,Kevin zha@sjtu.edu.cn,lucewu@cs.sjtu.edu.cn +</td></tr><tr><td>22f94c43dd8b203f073f782d91e701108909690b</td><td>MovieScope: Movie trailer classification using Deep Neural Networks +<br/>Dept of Computer Science +<br/><b>University of Virginia</b></td><td></td><td>{ks6cq, gs9ed}@virginia.edu +</td></tr><tr><td>22dabd4f092e7f3bdaf352edd925ecc59821e168</td><td> Deakin Research Online +<br/>This is the published version: +<br/>An, Senjian, Liu, Wanquan and Venkatesh, Svetha 2008, Exploiting side information in +<br/>locality preserving projection, in CVPR 2008 : Proceedings of the 26th IEEE Conference on +<br/>Computer Vision and Pattern Recognition, IEEE, Washington, D. C., pp. 1-8. +<br/>Available from Deakin Research Online: +<br/>http://hdl.handle.net/10536/DRO/DU:30044576 +<br/> +<br/>Reproduced with the kind permissions of the copyright owner. +<br/>Personal use of this material is permitted. However, permission to reprint/republish this +<br/>material for advertising or promotional purposes or for creating new collective works for +<br/>resale or redistribution to servers or lists, or to reuse any copyrighted component of this work +<br/>in other works must be obtained from the IEEE. +<br/>Copyright : 2008, IEEE +</td><td></td><td></td></tr><tr><td>22f656d0f8426c84a33a267977f511f127bfd7f3</td><td></td><td></td><td></td></tr><tr><td>22143664860c6356d3de3556ddebe3652f9c912a</td><td>Facial Expression Recognition for Human-robot +<br/>Interaction – A Prototype +<br/>1 Department of Informatics, Technische Universitat M¨unchen, Germany +<br/><b>Electrical and Computer Engineering, University of Auckland, New Zealand</b></td><td>('32131501', 'Matthias Wimmer', 'matthias wimmer')<br/>('1761487', 'Bruce A. MacDonald', 'bruce a. macdonald')<br/>('3235721', 'Dinuka Jayamuni', 'dinuka jayamuni')<br/>('2607879', 'Arpit Yadav', 'arpit yadav')</td><td></td></tr><tr><td>2271d554787fdad561fafc6e9f742eea94d35518</td><td>TECHNISCHE UNIVERSIT ¨AT M ¨UNCHEN +<br/>Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +<br/>Multimodale Mensch-Roboter-Interaktion +<br/>f¨ur Ambient Assisted Living +<br/>Tobias F. Rehrl +<br/>Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +<br/>der Technischen Universit¨at M¨unchen zur Erlangung des akademischen Grades eines +<br/>Doktor-Ingenieurs (Dr.-Ing.) +<br/>genehmigten Dissertation. +<br/>Vorsitzende: +<br/>Pr¨ufer der Dissertation: 1. Univ.-Prof. Dr.-Ing. habil. Gerhard Rigoll +<br/>2. Univ.-Prof. Dr.-Ing. Horst-Michael Groß +<br/>Univ.-Prof. Dr.-Ing. Sandra Hirche +<br/>(Technische Universit¨at Ilmenau) +<br/>Die Dissertation wurde am 17. April 2013 bei der Technischen Universit¨at M¨unchen +<br/>eingereicht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik am +<br/>8. Oktober 2013 angenommen. +</td><td></td><td></td></tr><tr><td>22ec256400e53cee35f999244fb9ba6ba11c1d06</td><td></td><td></td><td></td></tr><tr><td>22a7f1aebdb57eecd64be2a1f03aef25f9b0e9a7</td><td></td><td></td><td></td></tr><tr><td>22e189a813529a8f43ad76b318207d9a4b6de71a</td><td>What will Happen Next? +<br/>Forecasting Player Moves in Sports Videos +<br/>UC Berkeley, STATS +<br/>UC Berkeley +<br/>UC Berkeley +</td><td>('2986395', 'Panna Felsen', 'panna felsen')<br/>('33932184', 'Pulkit Agrawal', 'pulkit agrawal')<br/>('1689212', 'Jitendra Malik', 'jitendra malik')</td><td>panna@berkeley.edu +<br/>pulkitag@berkeley.edu +<br/>malik@berkeley.edu +</td></tr><tr><td>25ff865460c2b5481fa4161749d5da8501010aa0</td><td>Seeing What Is Not There: +<br/>Learning Context to Determine Where Objects Are Missing +<br/>Department of Computer Science +<br/><b>University of Maryland</b><br/>Figure 1: When curb ramps (green rectangle) are missing from a segment of sidewalks in an intersection (orange rectangle), +<br/>people with mobility impairments are unable to cross the street. We propose an approach to determine where objects are +<br/>missing by learning a context model so that it can be combined with object detection results. +</td><td>('39516880', 'Jin Sun', 'jin sun')<br/>('34734622', 'David W. Jacobs', 'david w. jacobs')</td><td>{jinsun,djacobs}@cs.umd.edu +</td></tr><tr><td>25d514d26ecbc147becf4117512523412e1f060b</td><td>Annotated Crowd Video Face Database +<br/>IIIT-Delhi, India +</td><td>('2952437', 'Tejas I. Dhamecha', 'tejas i. dhamecha')<br/>('2578160', 'Priyanka Verma', 'priyanka verma')<br/>('3239512', 'Mahek Shah', 'mahek shah')<br/>('39129417', 'Richa Singh', 'richa singh')<br/>('2338122', 'Mayank Vatsa', 'mayank vatsa')</td><td>{tejasd,priyanka13100,mahek13106,rsingh,mayank}@iiitd.ac.in +</td></tr><tr><td>25c19d8c85462b3b0926820ee5a92fc55b81c35a</td><td>Noname manuscript No. +<br/>(will be inserted by the editor) +<br/>Pose-Invariant Facial Expression Recognition +<br/>Using Variable-Intensity Templates +<br/>Received: date / Accepted: date +</td><td>('3325574', 'Shiro Kumano', 'shiro kumano')<br/>('38178548', 'Eisaku Maeda', 'eisaku maeda')</td><td></td></tr><tr><td>258a8c6710a9b0c2dc3818333ec035730062b1a5</td><td>Benelearn 2005 +<br/>Annual Machine Learning Conference of +<br/>Belgium and the Netherlands +<br/>CTIT PROCEEDINGS OF THE FOURTEENTH +<br/>ANNUAL MACHINE LEARNING CONFERENCE +<br/>OF BELGIUM AND THE NETHERLANDS +</td><td>('2541098', 'Martijn van Otterlo', 'martijn van otterlo')<br/>('1688157', 'Mannes Poel', 'mannes poel')<br/>('1745198', 'Anton Nijholt', 'anton nijholt')</td><td></td></tr><tr><td>25695abfe51209798f3b68fb42cfad7a96356f1f</td><td>AN INVESTIGATION INTO COMBINING +<br/>BOTH FACIAL DETECTION AND +<br/>LANDMARK LOCALISATION INTO A +<br/>UNIFIED PROCEDURE USING GPU +<br/>COMPUTING +<br/> MSc by Research +<br/>2016 +</td><td>('32464788', 'J M McDonagh', 'j m mcdonagh')</td><td></td></tr><tr><td>250ebcd1a8da31f0071d07954eea4426bb80644c</td><td>DenseBox: Unifying Landmark Localization with +<br/>End to End Object Detection +<br/><b>Institute of Deep Learning</b><br/>Baidu Research +</td><td>('3168646', 'Lichao Huang', 'lichao huang')<br/>('1698559', 'Yi Yang', 'yi yang')<br/>('1987538', 'Yafeng Deng', 'yafeng deng')<br/>('2278628', 'Yinan Yu', 'yinan yu')</td><td>2{huanglichao01,yangyi05,dengyafeng}@baidu.com +<br/>1alanhuang1990@gmail.com +<br/>3bebekifis@gmail.com +</td></tr><tr><td>25337690fed69033ef1ce6944e5b78c4f06ffb81</td><td>STRATEGIC ENGAGEMENT REGULATION: +<br/>AN INTEGRATION OF SELF-ENHANCEMENT AND ENGAGEMENT +<br/>by +<br/><b>A dissertation submitted to the Faculty of the University of Delaware in partial</b><br/>fulfillment of the requirements for the degree of Doctor of Philosophy in Psychology +<br/>Spring 2014 +<br/>All Rights Reserved +</td><td>('2800616', 'Jordan B. Leitner', 'jordan b. leitner')<br/>('2800616', 'Jordan B. Leitner', 'jordan b. leitner')</td><td></td></tr><tr><td>25c3cdbde7054fbc647d8be0d746373e7b64d150</td><td>ForgetMeNot: Memory-Aware Forensic Facial Sketch Matching +<br/><b>Beijing University of Posts and Telecommunications</b><br/><b>Queen Mary University of London, UK</b></td><td>('2961830', 'Shuxin Ouyang', 'shuxin ouyang')<br/>('1697755', 'Timothy M. Hospedales', 'timothy m. hospedales')<br/>('1705408', 'Yi-Zhe Song', 'yi-zhe song')<br/>('7823169', 'Xueming Li', 'xueming li')</td><td>{s.ouyang, t.hospedales, yizhe.song}@qmul.ac.uk +<br/>lixm@bupt.edu.cn +</td></tr><tr><td>25bf288b2d896f3c9dab7e7c3e9f9302e7d6806b</td><td>Neural Networks with Smooth Adaptive Activation Functions +<br/>for Regression +<br/><b>Stony Brook University, NY, USA</b><br/><b>Stony Brook University, NY, USA</b><br/>3Oak Ridge National Laboratory, USA +<br/>4Department of Applied Mathematics and Statistics, NY, USA +<br/>5Department of Pathology, Stony Brook Hospital, NY, USA +<br/>6Cancer Center, Stony Brook Hospital, NY, USA +<br/>August 24, 2016 +</td><td>('2321406', 'Le Hou', 'le hou')<br/>('1686020', 'Dimitris Samaras', 'dimitris samaras')<br/>('1755448', 'Yi Gao', 'yi gao')<br/>('1735710', 'Joel H. Saltz', 'joel h. saltz')</td><td>{lehhou,samaras}@cs.stonybrook.edu +<br/>{tahsin.kurc,joel.saltz}@stonybrook.edu +<br/>yi.gao@stonybrookmedicine.edu +</td></tr><tr><td>25d3e122fec578a14226dc7c007fb1f05ddf97f7</td><td>The First Facial Expression Recognition and Analysis Challenge +</td><td>('1795528', 'Michel F. Valstar', 'michel f. valstar')<br/>('39532631', 'Bihan Jiang', 'bihan jiang')<br/>('1875347', 'Marc Mehu', 'marc mehu')<br/>('1694605', 'Maja Pantic', 'maja pantic')</td><td></td></tr><tr><td>2597b0dccdf3d89eaffd32e202570b1fbbedd1d6</td><td>Towards predicting the likeability of fashion images +</td><td>('2569065', 'Jinghua Wang', 'jinghua wang')<br/>('2613790', 'Abrar Abdul Nabi', 'abrar abdul nabi')<br/>('22804340', 'Gang Wang', 'gang wang')<br/>('2737180', 'Chengde Wan', 'chengde wan')<br/>('2475944', 'Tian-Tsong Ng', 'tian-tsong ng')</td><td></td></tr><tr><td>2588acc7a730d864f84d4e1a050070ff873b03d5</td><td>Article +<br/>Action Recognition by an Attention-Aware Temporal +<br/>Weighted Convolutional Neural Network +<br/><b>Institute of Arti cial Intelligence and Robotics, Xi an Jiaotong University, Xi an 710049, China</b><br/>Received: 27 April 2018; Accepted: 19 June 2018; Published: 21 June 2018 +</td><td>('40367806', 'Le Wang', 'le wang')<br/>('14800230', 'Jinliang Zang', 'jinliang zang')<br/>('46324995', 'Qilin Zhang', 'qilin zhang')<br/>('1786361', 'Zhenxing Niu', 'zhenxing niu')<br/>('1745420', 'Gang Hua', 'gang hua')<br/>('1715389', 'Nanning Zheng', 'nanning zheng')</td><td>zjl19920904@stu.xjtu.edu.cn (J.Z.); nnzheng@xjtu.edu.cn (N.Z.) +<br/>2 HERE Technologies, Chicago, IL 60606, USA; qilin.zhang@here.com +<br/>3 Alibaba Group, Hangzhou 311121, China; zhenxing.nzx@alibaba-inc.com +<br/>4 Microsoft Research, Redmond, WA 98052, USA; ganghua@microsoft.com +<br/>* Correspondence: lewang@xjtu.edu.cn; Tel.: +86-29-8266-8672 +</td></tr><tr><td>25982e2bef817ebde7be5bb80b22a9864b979fb0</td><td></td><td></td><td></td></tr><tr><td>25c108a56e4cb757b62911639a40e9caf07f1b4f</td><td>Recurrent Scale Approximation for Object Detection in CNN +<br/><b>Multimedia Laboratory at The Chinese University of Hong Kong</b><br/>1SenseTime Group Limited +</td><td>('1715752', 'Yu Liu', 'yu liu')<br/>('1929886', 'Hongyang Li', 'hongyang li')<br/>('1721677', 'Junjie Yan', 'junjie yan')<br/>('22181490', 'Fangyin Wei', 'fangyin wei')<br/>('31843833', 'Xiaogang Wang', 'xiaogang wang')<br/>('1741901', 'Xiaoou Tang', 'xiaoou tang')</td><td>liuyuisanai@gmail.com,{yangli,xgwang}@ee.cuhk.edu.hk, +<br/>{yanjunjie,weifangyin}@sensetime.com, xtang@ie.cuhk.edu.hk +</td></tr><tr><td>2594a77a3f0dd5073f79ba620e2f287804cec630</td><td>TRANSFERRING FACE VERIFICATION NETS TO PAIN AND EXPRESSION REGRESSION +<br/>Dept. of {Computer Science1, Electrical & Computer Engineering2, Radiation Oncology3, Cognitive Science4} +<br/><b>Johns Hopkins University, 3400 N. Charles St, Baltimore, MD 21218, USA</b><br/>5Dept. of EE, UESTC, 2006 Xiyuan Ave, Chengdu, Sichuan 611731, China +<br/><b>Tsinghua University, Beijing 100084, China</b></td><td>('1713335', 'Feng Wang', 'feng wang')<br/>('40031188', 'Xiang Xiang', 'xiang xiang')<br/>('1692867', 'Chang Liu', 'chang liu')<br/>('1709073', 'Trac D. Tran', 'trac d. tran')<br/>('3207112', 'Austin Reiter', 'austin reiter')<br/>('1678633', 'Gregory D. Hager', 'gregory d. hager')<br/>('2095823', 'Harry Quon', 'harry quon')<br/>('1709439', 'Jian Cheng', 'jian cheng')<br/>('1746141', 'Alan L. Yuille', 'alan l. yuille')</td><td></td></tr><tr><td>25e2d3122d4926edaab56a576925ae7a88d68a77</td><td>ORIGINAL RESEARCH +<br/>published: 23 February 2016 +<br/>doi: 10.3389/fpsyg.2016.00166 +<br/>Communicative-Pragmatic +<br/>Treatment in Schizophrenia: A Pilot +<br/>Study +<br/><b>Center for Cognitive Science, University of Turin, Turin, Italy, 2 Neuroscience Institute of Turin</b><br/><b>Turin, Italy, 3 Faculty of Humanities, Research Unit of Logopedics, Child Language Research Center, University of Oulu, Oulu</b><br/>Finland, 4 AslTo2 Department of Mental Health, Turin, Italy, 5 Brain Imaging Group, Turin, Italy +<br/>This paper aims to verify the efficacy of Cognitive Pragmatic Treatment (CPT), a new +<br/>remediation training for the improvement of the communicative-pragmatic abilities, in +<br/>patients with schizophrenia. The CPT program is made up of 20 group sessions, +<br/>focused on a number of communication modalities, i.e., linguistic, extralinguistic and +<br/>paralinguistic, theory of mind (ToM) and other cognitive functions able to play a role +<br/>on the communicative performance, such as awareness and planning. A group of 17 +<br/>patients with schizophrenia took part in the training program. They were evaluated +<br/>before and after training, through the equivalent forms of the Assessment Battery for +<br/>Communication (ABaCo), a tool for testing, both in comprehension and in production, +<br/>a wide range of pragmatic phenomena such as direct and indirect speech acts, +<br/>irony and deceit, and a series of neuropsychological and ToM tests. The results +<br/>showed a significant improvement in patients’ performance on both production and +<br/>comprehension tasks following the program, and in all the communication modalities +<br/>evaluated through the ABaCo, i.e., linguistic, extralinguistic, paralinguistic, and social +<br/>appropriateness. This improvement persisted after 3 months from the end of the training +<br/>program, as shown by the follow-up tests. These preliminary findings provide evidence +<br/>of the efficacy of the CPT program in improving communicative-pragmatic abilities in +<br/>schizophrenic individuals. +<br/>Keywords: rehabilitation, schizophrenia, pragmatic, communication, training +<br/>INTRODUCTION +<br/>People with schizophrenia experience symptoms such as delusions, hallucinations, disorganized +<br/>speech and behavior, that cause difficulty in social relationships (DSM 5; American Psychiatric +<br/>Association [APA], 2013). In the clinical pragmatic domain (Cummings, 2014), the area of study +<br/>of pragmatic impairment in patients with communicative disorders, several studies have reported +<br/>that communicative ability is impaired in patients with schizophrenia (Langdon et al., 2002; Bazin +<br/>et al., 2005; Linscott, 2005; Marini et al., 2008; Colle et al., 2013). For example, Bazin et al. (2005), +<br/>created a structured interview, the Schizophrenia Communication Disorder Scale, which they +<br/>administered to patients with schizophrenia. The authors observed that these patients performed +<br/>less well than those affected by mania or depression in managing a conversation on everyday +<br/>topics, such as family, job, hobbies, and so on. Likewise, non-compliance with conversational +<br/>rules, such as consistency with the agreed purpose of the interaction, giving the partner too little +<br/>Edited by: +<br/>Sayyed Mohsen Fatemi, +<br/><b>Harvard University, USA</b><br/>Reviewed by: +<br/>Silvia Serino, +<br/>IRCCS Istituto Auxologico Italiano, +<br/>Italy +<br/>Michelle Dow Keawphalouk, +<br/><b>Harvard and Massachusetts Institute</b><br/>of Technology, USA +<br/>*Correspondence: +<br/>Specialty section: +<br/>This article was submitted to +<br/>Psychology for Clinical Settings, +<br/>a section of the journal +<br/>Frontiers in Psychology +<br/>Received: 07 October 2015 +<br/>Accepted: 28 January 2016 +<br/>Published: 23 February 2016 +<br/>Citation: +<br/>Bosco FM, Gabbatore I, Gastaldo L +<br/>and Sacco K (2016) +<br/>Communicative-Pragmatic Treatment +<br/>in Schizophrenia: A Pilot Study. +<br/>Front. Psychol. 7:166. +<br/>doi: 10.3389/fpsyg.2016.00166 +<br/>Frontiers in Psychology | www.frontiersin.org +<br/>February 2016 | Volume 7 | Article 166 +</td><td>('2261858', 'Francesca M. Bosco', 'francesca m. bosco')<br/>('3175646', 'Ilaria Gabbatore', 'ilaria gabbatore')<br/>('39551201', 'Luigi Gastaldo', 'luigi gastaldo')<br/>('2159033', 'Katiuscia Sacco', 'katiuscia sacco')<br/>('3175646', 'Ilaria Gabbatore', 'ilaria gabbatore')</td><td>ilaria.gabbatore@oulu.fi; +<br/>ilariagabbatore@gmail.com +</td></tr><tr><td>25e05a1ea19d5baf5e642c2a43cca19c5cbb60f8</td><td>Label Distribution Learning +</td><td>('1735299', 'Xin Geng', 'xin geng')</td><td></td></tr><tr><td>2559b15f8d4a57694a0a33bdc4ac95c479a3c79a</td><td>570 +<br/>Contextual Object Localization With Multiple +<br/>Kernel Nearest Neighbor +<br/>Gert Lanckriet, Member, IEEE +</td><td>('3215419', 'Brian McFee', 'brian mcfee')<br/>('1954793', 'Carolina Galleguillos', 'carolina galleguillos')</td><td></td></tr><tr><td>2574860616d7ffa653eb002bbaca53686bc71cdd</td><td></td><td></td><td></td></tr><tr><td>25f1f195c0efd84c221b62d1256a8625cb4b450c</td><td>1-4244-1017-7/07/$25.00 ©2007 IEEE +<br/>1091 +<br/>ICME 2007 +</td><td></td><td></td></tr><tr><td>25885e9292957feb89dcb4a30e77218ffe7b9868</td><td>JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2016 +<br/>Analyzing the Affect of a Group of People Using +<br/>Multi-modal Framework +</td><td>('18780812', 'Xiaohua Huang', 'xiaohua huang')<br/>('1735697', 'Abhinav Dhall', 'abhinav dhall')<br/>('40357816', 'Xin Liu', 'xin liu')<br/>('1757287', 'Guoying Zhao', 'guoying zhao')<br/>('2473859', 'Jingang Shi', 'jingang shi')</td><td></td></tr><tr><td>259706f1fd85e2e900e757d2656ca289363e74aa</td><td>Improving People Search Using Query Expansions +<br/>How Friends Help To Find People +<br/>LEAR - INRIA Rhˆone Alpes - Grenoble, France +</td><td>('1722052', 'Thomas Mensink', 'thomas mensink')<br/>('34602236', 'Jakob Verbeek', 'jakob verbeek')</td><td>{thomas.mensink,jakob.verbeek}@inria.fr +</td></tr><tr><td>25728e08b0ee482ee6ced79c74d4735bb5478e29</td><td></td><td></td><td></td></tr><tr><td>258a2dad71cb47c71f408fa0611a4864532f5eba</td><td>Discriminative Optimization +<br/>of Local Features for Face Recognition +<br/> +<br/>H O S S E I N A Z I Z P O U R +<br/> +<br/>Master of Science Thesis +<br/>Stockholm, Sweden 2011 +<br/> +</td><td></td><td></td></tr><tr><td>25127c2d9f14d36f03d200a65de8446f6a0e3bd6</td><td>Journal of Theoretical and Applied Information Technology +<br/> 20th May 2016. Vol.87. No.2 +<br/>© 2005 - 2016 JATIT & LLS. All rights reserved. +<br/>ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195 +<br/>EVALUATING THE PERFORMANCE OF DEEP SUPERVISED +<br/>AUTO ENCODER IN SINGLE SAMPLE FACE RECOGNITION +<br/>PROBLEM USING KULLBACK-LEIBLER DIVERGENCE +<br/>SPARSITY REGULARIZER +<br/> Faculty of Computer of Computer Science, Universitas Indonesia, Kampus UI Depok, Indonesia +</td><td>('9324684', 'ARIDA F. SYAFIANDINI', 'arida f. syafiandini')</td><td>E-mail: 1otniel.yosi@ui.ac.id , 2ito.wasito@cs.ui.ac.id, 2arida.ferti@ui.ac.id +</td></tr></table></body></html>
\ No newline at end of file |