From ee3d0d98e19f1d8177d85af1866fd0ee431fe9ea Mon Sep 17 00:00:00 2001 From: Jules Laplace Date: Sun, 25 Nov 2018 22:19:15 +0100 Subject: moving stuff --- scraper/reports/institutions_missing.html | 11693 ++++++++++++++++++++++++++++ 1 file changed, 11693 insertions(+) create mode 100644 scraper/reports/institutions_missing.html (limited to 'scraper/reports/institutions_missing.html') diff --git a/scraper/reports/institutions_missing.html b/scraper/reports/institutions_missing.html new file mode 100644 index 00000000..93a26238 --- /dev/null +++ b/scraper/reports/institutions_missing.html @@ -0,0 +1,11693 @@ +Institutions

Institutions

61084a25ebe736e8f6d7a6e53b2c20d9723c4608
61f04606528ecf4a42b49e8ac2add2e9f92c0defDeep Deformation Network for Object Landmark +
Localization +
NEC Laboratories America, Department of Media Analytics +
614a7c42aae8946c7ad4c36b53290860f62564411 +
Joint Face Detection and Alignment using +
Multi-task Cascaded Convolutional Networks +
0d88ab0250748410a1bc990b67ab2efb370ade5dAuthor(s) : +
ERROR HANDLING IN MULTIMODAL BIOMETRIC SYSTEMS USING +
RELIABILITY MEASURES (ThuPmOR6) +
(EPFL, Switzerland) +
(EPFL, Switzerland) +
(EPFL, Switzerland) +
(EPFL, Switzerland) +
Plamen Prodanov +
0d467adaf936b112f570970c5210bdb3c626a717
0d6b28691e1aa2a17ffaa98b9b38ac3140fb3306Review of Perceptual Resemblance of Local +
Plastic Surgery Facial Images using Near Sets +
1,2 Department of Computer Technology, +
YCCE Nagpur, India +
0db8e6eb861ed9a70305c1839eaef34f2c85bbaf
0dbf4232fcbd52eb4599dc0760b18fcc1e9546e9
0d760e7d762fa449737ad51431f3ff938d6803feLCDet: Low-Complexity Fully-Convolutional Neural Networks for +
Object Detection in Embedded Systems +
UC San Diego ∗ +
Gokce Dane +
Qualcomm Inc. +
UC San Diego +
Qualcomm Inc. +
UC San Diego +
0dd72887465046b0f8fc655793c6eaaac9c03a3dReal-time Head Orientation from a Monocular +
Camera using Deep Neural Network +
KAIST, Republic of Korea +
0d087aaa6e2753099789cd9943495fbbd08437c0
0d8415a56660d3969449e77095be46ef0254a448
0d735e7552af0d1dcd856a8740401916e54b7eee
0d06b3a4132d8a2effed115a89617e0a702c957a
0d2dd4fc016cb6a517d8fb43a7cc3ff62964832e
0d33b6c8b4d1a3cb6d669b4b8c11c2a54c203d1aDetection and Tracking of Faces in Videos: A Review +
© 2016 IJEDR | Volume 4, Issue 2 | ISSN: 2321-9939 +
of Related Work +
1Student, 2Assistant Professor +
1, 2Dept. of Electronics & Comm., S S I E T, Punjab, India +
________________________________________________________________________________________________________ +
956317de62bd3024d4ea5a62effe8d6623a64e53Lighting Analysis and Texture Modification of 3D Human +
Face Scans +
Author +
Zhang, Paul, Zhao, Sanqiang, Gao, Yongsheng +
Published +
2007 +
Conference Title +
Digital Image Computing Techniques and Applications +
DOI +
https://doi.org/10.1109/DICTA.2007.4426825 +
Copyright Statement +
© 2007 IEEE. Personal use of this material is permitted. However, permission to reprint/ +
republish this material for advertising or promotional purposes or for creating new collective +
works for resale or redistribution to servers or lists, or to reuse any copyrighted component of +
this work in other works must be obtained from the IEEE. +
Downloaded from +
http://hdl.handle.net/10072/17889 +
Link to published version +
http://www.ieee.org/ +
Griffith Research Online +
https://research-repository.griffith.edu.au +
956c634343e49319a5e3cba4f2bd2360bdcbc075IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 36, NO. 4, AUGUST 2006 +
873 +
A Novel Incremental Principal Component Analysis +
and Its Application for Face Recognition +
958c599a6f01678513849637bec5dc5dba592394Noname manuscript No. +
(will be inserted by the editor) +
Generalized Zero-Shot Learning for Action +
Recognition with Web-Scale Video Data +
Received: date / Accepted: date +
59fc69b3bc4759eef1347161e1248e886702f8f7Final Report of Final Year Project +
HKU-Face: A Large Scale Dataset for +
Deep Face Recognition +
3035141841 +
COMP4801 Final Year Project +
Project Code: 17007 +
59bfeac0635d3f1f4891106ae0262b81841b06e4Face Verification Using the LARK Face +
Representation +
590628a9584e500f3e7f349ba7e2046c8c273fcf
59eefa01c067a33a0b9bad31c882e2710748ea24IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY +
Fast Landmark Localization +
with 3D Component Reconstruction and CNN for +
Cross-Pose Recognition +
5945464d47549e8dcaec37ad41471aa70001907fNoname manuscript No. +
(will be inserted by the editor) +
Every Moment Counts: Dense Detailed Labeling of Actions in Complex +
Videos +
Received: date / Accepted: date +
59c9d416f7b3d33141cc94567925a447d0662d80Universität des Saarlandes +
Max-Planck-Institut für Informatik +
AG5 +
Matrix factorization over max-times +
algebra for data mining +
Masterarbeit im Fach Informatik +
Master’s Thesis in Computer Science +
von / by +
angefertigt unter der Leitung von / supervised by +
begutachtet von / reviewers +
November 2013 +
UNIVERSITASSARAVIENSIS
59a35b63cf845ebf0ba31c290423e24eb822d245The FaceSketchID System: Matching Facial +
Composites to Mugshots +
tedious, and may not +
59f325e63f21b95d2b4e2700c461f0136aecc1713070 +
978-1-4577-1302-6/11/$26.00 ©2011 IEEE +
FOR FACE RECOGNITION +
1. INTRODUCTION +
5922e26c9eaaee92d1d70eae36275bb226ecdb2eBoosting Classification Based Similarity +
Learning by using Standard Distances +
Departament d’Informàtica, Universitat de València +
Av. de la Universitat s/n. 46100-Burjassot (Spain) +
59031a35b0727925f8c47c3b2194224323489d68Sparse Variation Dictionary Learning for Face Recognition with A Single +
Training Sample Per Person +
ETH Zurich +
Switzerland +
926c67a611824bc5ba67db11db9c05626e79de961913 +
Enhancing Bilinear Subspace Learning +
by Element Rearrangement +
923ede53b0842619831e94c7150e0fc4104e62f7978-1-4799-9988-0/16/$31.00 ©2016 IEEE +
1293 +
ICASSP 2016 +
92b61b09d2eed4937058d0f9494d9efeddc39002Under review in IJCV manuscript No. +
(will be inserted by the editor) +
BoxCars: Improving Vehicle Fine-Grained Recognition using +
3D Bounding Boxes in Traffic Surveillance +
Received: date / Accepted: date +
920a92900fbff22fdaaef4b128ca3ca8e8d54c3eLEARNING PATTERN TRANSFORMATION MANIFOLDS WITH PARAMETRIC ATOM +
SELECTION +
Ecole Polytechnique F´ed´erale de Lausanne (EPFL) +
Signal Processing Laboratory (LTS4) +
Switzerland-1015 Lausanne +
9207671d9e2b668c065e06d9f58f597601039e5eFace Detection Using a 3D Model on +
Face Keypoints +
9282239846d79a29392aa71fc24880651826af72Antonakos et al. EURASIP Journal on Image and Video Processing 2014, 2014:14 +
http://jivp.eurasipjournals.com/content/2014/1/14 +
RESEARCH +
Open Access +
Classification of extreme facial events in sign +
language videos +
92c2dd6b3ac9227fce0a960093ca30678bceb364Provided by the author(s) and NUI Galway in accordance with publisher policies. Please cite the published +
version when available. +
Title +
On color texture normalization for active appearance models +
Author(s) +
Ionita, Mircea C.; Corcoran, Peter M.; Buzuloiu, Vasile +
Publication +
Date +
2009-05-12 +
Publication +
Information +
Ionita, M. C., Corcoran, P., & Buzuloiu, V. (2009). On Color +
Texture Normalization for Active Appearance Models. Image +
Processing, IEEE Transactions on, 18(6), 1372-1378. +
Publisher +
IEEE +
Link to +
publisher's +
version +
http://dx.doi.org/10.1109/TIP.2009.2017163 +
Item record +
http://hdl.handle.net/10379/1350 +
Some rights reserved. For more information, please see the item record link above. +
Downloaded 2018-11-06T00:40:53Z +
92fada7564d572b72fd3be09ea3c39373df3e27c
927ad0dceacce2bb482b96f42f2fe2ad1873f37aInterest-Point based Face Recognition System +
87 +
X +
Interest-Point based Face Recognition System +
Spain +
1. Introduction +
Among all applications of face recognition systems, surveillance is one of the most +
challenging ones. In such an application, the goal is to detect known criminals in crowded +
environments, like airports or train stations. Some attempts have been made, like those of +
Tokio (Engadget, 2006) or Mainz (Deutsche Welle, 2006), with limited success. +
The first task to be carried out in an automatic surveillance system involves the detection of +
all the faces in the images taken by the video cameras. Current face detection algorithms are +
highly reliable and thus, they will not be the focus of our work. Some of the best performing +
examples are the Viola-Jones algorithm (Viola & Jones, 2004) or the Schneiderman-Kanade +
algorithm (Schneiderman & Kanade, 2000). +
The second task to be carried out involves the comparison of all detected faces among the +
database of known criminals. The ideal behaviour of an automatic system performing this +
task would be to get a 100% correct identification rate, but this behaviour is far from the +
capabilities of current face recognition algorithms. Assuming that there will be false +
identifications, supervised surveillance systems seem to be the most realistic option: the +
automatic system issues an alarm whenever it detects a possible match with a criminal, and +
a human decides whether it is a false alarm or not. Figure 1 shows an example. +
However, even in a supervised scenario the requirements for the face recognition algorithm +
are extremely high: the false alarm rate must be low enough as to allow the human operator +
to cope with it; and the percentage of undetected criminals must be kept to a minimum in +
order to ensure security. Fulfilling both requirements at the same time is the main challenge, +
as a reduction in false alarm rate usually implies an increase of the percentage of undetected +
criminals. +
We propose a novel face recognition system based in the use of interest point detectors and +
local descriptors. In order to check the performances of our system, and particularly its +
performances in a surveillance application, we present experimental results in terms of +
Receiver Operating Characteristic curves or ROC curves. From the experimental results, it +
becomes clear that our system outperforms classical appearance based approaches. +
www.intechopen.com +
929bd1d11d4f9cbc638779fbaf958f0efb82e603This is the author’s version of a work that was submitted/accepted for pub- +
lication in the following source: +
Zhang, Ligang & Tjondronegoro, Dian W. (2010) Improving the perfor- +
mance of facial expression recognition using dynamic, subtle and regional +
features. +
In Kok, WaiWong, B. Sumudu, U. Mendis, & Abdesselam , +
Bouzerdoum (Eds.) Neural Information Processing. Models and Applica- +
tions, Lecture Notes in Computer Science, Sydney, N.S.W, pp. 582-589. +
This file was downloaded from: http://eprints.qut.edu.au/43788/ +
c(cid:13) Copyright 2010 Springer-Verlag +
Conference proceedings published, by Springer Verlag, will be available +
via Lecture Notes in Computer Science http://www.springer.de/comp/lncs/ +
Notice: Changes introduced as a result of publishing processes such as +
copy-editing and formatting may not be reflected in this document. For a +
definitive version of this work, please refer to the published source: +
http://dx.doi.org/10.1007/978-3-642-17534-3_72 +
0c36c988acc9ec239953ff1b3931799af388ef70Face Detection Using Improved Faster RCNN +
Huawei Cloud BU, China +
Figure1.Face detection results of FDNet1.0 +
0c5ddfa02982dcad47704888b271997c4de0674b
0cccf576050f493c8b8fec9ee0238277c0cfd69a
0c069a870367b54dd06d0da63b1e3a900a257298Author manuscript, published in "ICANN 2011 - International Conference on Artificial Neural Networks (2011)" +
0c75c7c54eec85e962b1720755381cdca3f57dfb2212 +
Face Landmark Fitting via Optimized Part +
Mixtures and Cascaded Deformable Model +
0ca36ecaf4015ca4095e07f0302d28a5d9424254Improving Bag-of-Visual-Words Towards Effective Facial Expressive +
Image Classification +
1Univ. Grenoble Alpes, CNRS, Grenoble INP∗ , GIPSA-lab, 38000 Grenoble, France +
Keywords: +
BoVW, k-means++, Relative Conjunction Matrix, SIFT, Spatial Pyramids, TF.IDF. +
0cfca73806f443188632266513bac6aaf6923fa8Predictive Uncertainty in Large Scale Classification +
using Dropout - Stochastic Gradient Hamiltonian +
Monte Carlo. +
Vergara, Diego∗1, Hern´andez, Sergio∗2, Valdenegro-Toro, Mat´ıas∗∗3 and Jorquera, Felipe∗4. +
∗Laboratorio de Procesamiento de Informaci´on Geoespacial, Universidad Cat´olica del Maule, Chile. +
∗∗German Research Centre for Artificial Intelligence, Bremen, Germany. +
0c54e9ac43d2d3bab1543c43ee137fc47b77276e
0c5afb209b647456e99ce42a6d9d177764f9a0dd97 +
Recognizing Action Units for +
Facial Expression Analysis +
0c377fcbc3bbd35386b6ed4768beda7b5111eec6258 +
A Unified Probabilistic Framework +
for Spontaneous Facial Action Modeling +
and Understanding +
0cb2dd5f178e3a297a0c33068961018659d0f443
0cf7da0df64557a4774100f6fde898bc4a3c4840Shape Matching and Object Recognition using Low Distortion Correspondences +
Department of Electrical Engineering and Computer Science +
U.C. Berkeley +
0c4659b35ec2518914da924e692deb37e96d62061236 +
Registering a MultiSensor Ensemble of Images +
0c53ef79bb8e5ba4e6a8ebad6d453ecf3672926dSUBMITTED TO JOURNAL +
Weakly Supervised PatchNets: Describing and +
Aggregating Local Patches for Scene Recognition +
0c60eebe10b56dbffe66bb3812793dd514865935
6601a0906e503a6221d2e0f2ca8c3f544a4adab7SRTM-2 2/9/06 3:27 PM Page 321 +
Detection of Ancient Settlement Mounds: +
Archaeological Survey Based on the +
SRTM Terrain Model +
B.H. Menze, J.A. Ur, and A.G. Sherratt +
660b73b0f39d4e644bf13a1745d6ee74424d4a16
66d512342355fb77a4450decc89977efe7e55fa2Under review as a conference paper at ICLR 2018 +
LEARNING NON-LINEAR TRANSFORM WITH DISCRIM- +
INATIVE AND MINIMUM INFORMATION LOSS PRIORS +
Anonymous authors +
Paper under double-blind review +
6643a7feebd0479916d94fb9186e403a4e5f7cbfChapter 8 +
3D Face Recognition +
661ca4bbb49bb496f56311e9d4263dfac8eb96e9Datasheets for Datasets +
66d087f3dd2e19ffe340c26ef17efe0062a59290Dog Breed Identification +
Brian Mittl +
Vijay Singh +
66a2c229ac82e38f1b7c77a786d8cf0d7e369598Proceedings of the 2016 Industrial and Systems Engineering Research Conference +
H. Yang, Z. Kong, and MD Sarder, eds. +
A Probabilistic Adaptive Search System +
for Exploring the Face Space +
Escuela Superior Politecnica del Litoral (ESPOL) +
Guayaquil-Ecuador +
66886997988358847615375ba7d6e9eb0f1bb27f
66837add89caffd9c91430820f49adb5d3f40930
66a9935e958a779a3a2267c85ecb69fbbb75b8dcFAST AND ROBUST FIXED-RANK MATRIX RECOVERY +
Fast and Robust Fixed-Rank Matrix +
Recovery +
Antonio Lopez +
66533107f9abdc7d1cb8f8795025fc7e78eb1122Vi a Sevig f a Ue  h wih E(cid:11)ecive ei Readig +
i a Wheechai baed Rbic A +
W y g Sgy Dae i iy g S g iz ad Ze ga Biey +
y EECS AST 373 1  g Dg Y g G  Taej 305 701 REA +
z VR Cee ETR 161 ajg Dg Y g G  Taej 305 350 REA +
Abac +
Thee exi he c eaive aciviy bewee a h +
a beig ad ehabiiai b beca e he h +
a eae ehabiiai b i he ae evi +
e ad ha he bee(cid:12) f ehabiiai b +
 ch a ai ay  bie f ci. ei +
eadig i e f he eeia f ci f h a +
fiedy ehabiiai b i de  ie he +
cf ad afey f a wh eed he. Fi f +
a he vea  c e f a ew wheechai baed +
bic a ye ARES  ad i h a b +
ieaci echgie ae eeed. Ag he +
echgie we cceae  vi a evig ha +
aw hi bic a  eae a  y via +
vi a feedback. E(cid:11)ecive iei eadig  ch a +
ecgizig he iive ad egaive eaig f he +
e i efed  he bai f chage f he facia +
exei a d i ha i gy eaed  he +
e iei whie hi bic a vide he +
e wih a beveage. F he eÆcie vi a ifa +
i ceig g a aed iage ae ed  +
c he ee caea head ha i caed i he +
ed e(cid:11)ec f he bic a. The vi a evig +
wih e(cid:11)ecive iei eadig i  ccef y aied +
 eve a beveage f he e. +
d ci +
Wheechai baed bic ye ae aiy ed  +
ai he edey ad he diabed wh have hadi +
ca i ey ad  f ci i ib. S ch a +
ye ci f a weed wheechai ad a bic +
a ad ha  y a bie caabiiy h gh +
he wheechai b  a a ai ay f ci via +
he bic a ad h  ake ibe he c +
exiece f a e ad a b i he ae evi +
e. +
 hi cae he e eed  ieac wih +
he bic a i cfabe ad afe way. w +
Fig e 1: The wheechai baed bic a ad i +
h a b ieaci echgie. +
eve i ha bee eed ha ay diÆc ie exi +
i h a bf ieaci i exiig ehabiiai +
b. F exae a a c f he bic +
a ake a high cgiive ad  he e a whie +
hyicay diabed e ay have diÆc ie i  +
eaig jyick dexe y   hig b  f +
deicae vee [4].  addii AUS eva +
ai e eed ha he  diÆc  hig  +
ig ehabiiai b i  ay cad f a +
a adj e ad  ay f ci  kee i +
id a he begiig [4]. Theefe h a fiedy +
h a b ieaci i e f eeia echi e +
i a wheechai baed bic a. +
 hi ae we cide he wheechai baed +
bic ye ARES AST Rehabiiai E +
gieeig Sevice ye  which we ae deveig +
a a evice bic ye f he diabed ad he +
edey ad dic  i h a b ieaci ech +
i e Fig. 1. Ag h a b ieaci ech +
i e vi a evig i dea wih a a aj ic. +
66810438bfb52367e3f6f62c24f5bc127cf92e56Face Recognition of Illumination Tolerance in 2D +
Subspace Based on the Optimum Correlation +
Filter +
Xu Yi +
Department of Information Engineering, Hunan Industry Polytechnic, Changsha, China +
images will be tested to project +
66af2afd4c598c2841dbfd1053bf0c386579234eNoname manuscript No. +
(will be inserted by the editor) +
Context Assisted Face Clustering Framework with +
Human-in-the-Loop +
Received: date / Accepted: date +
66e6f08873325d37e0ec20a4769ce881e04e964eInt J Comput Vis (2014) 108:59–81 +
DOI 10.1007/s11263-013-0695-z +
The SUN Attribute Database: Beyond Categories for Deeper Scene +
Understanding +
Received: 27 February 2013 / Accepted: 28 December 2013 / Published online: 18 January 2014 +
© Springer Science+Business Media New York 2014 +
661da40b838806a7effcb42d63a9624fcd68497653 +
An Illumination Invariant Accurate +
Face Recognition with Down Scaling +
of DCT Coefficients +
Department of Computer Science and Engineering, Amity School of Engineering and Technology, New Delhi, India +
In this paper, a novel approach for illumination normal- +
ization under varying lighting conditions is presented. +
Our approach utilizes the fact that discrete cosine trans- +
form (DCT) low-frequency coefficients correspond to +
illumination variations in a digital image. Under varying +
illuminations, the images captured may have low con- +
trast; initially we apply histogram equalization on these +
for contrast stretching. Then the low-frequency DCT +
coefficients are scaled down to compensate the illumi- +
nation variations. The value of scaling down factor and +
the number of low-frequency DCT coefficients, which +
are to be rescaled, are obtained experimentally. The +
classification is done using k−nearest neighbor classi- +
fication and nearest mean classification on the images +
obtained by inverse DCT on the processed coefficients. +
The correlation coefficient and Euclidean distance ob- +
tained using principal component analysis are used as +
distance metrics in classification. We have tested our +
face recognition method using Yale Face Database B. +
The results show that our method performs without any +
error (100% face recognition performance), even on the +
most extreme illumination variations. There are different +
schemes in the literature for illumination normalization +
under varying lighting conditions, but no one is claimed +
to give 100% recognition rate under all illumination +
variations for this database. The proposed technique is +
computationally efficient and can easily be implemented +
for real time face recognition system. +
Keywords: discrete cosine transform, correlation co- +
efficient, face recognition, illumination normalization, +
nearest neighbor classification +
1. Introduction +
Two-dimensional pattern classification plays a +
crucial role in real-world applications. To build +
high-performance surveillance or information +
security systems, face recognition has been +
known as the key application attracting enor- +
mous researchers highlighting on related topics +
[1,2]. Even though current machine recognition +
systems have reached a certain level of matu- +
rity, their success is limited by the real appli- +
cations constraints, like pose, illumination and +
expression. The FERET evaluation shows that +
the performance of a face recognition system +
decline seriously with the change of pose and +
illumination conditions [31]. +
To solve the variable illumination problem a +
variety of approaches have been proposed [3, 7- +
11, 26-29]. Early work in illumination invariant +
face recognition focused on image representa- +
tions that are mostly insensitive to changes in +
illumination. There were approaches in which +
the image representations and distance mea- +
sures were evaluated on a tightly controlled face +
database that varied the face pose, illumination, +
and expression. The image representations in- +
clude edge maps, 2D Gabor-like filters, first and +
second derivatives of the gray-level image, and +
the logarithmic transformations of the intensity +
image along with these representations [4]. +
The different approaches to solve the prob- +
lem of illumination invariant face recognition +
can be broadly classified into two main cate- +
gories. The first category is named as passive +
approach in which the visual spectrum images +
are analyzed to overcome this problem. The +
approaches belonging to other category named +
active, attempt to overcome this problem by +
employing active imaging techniques to obtain +
face images captured in consistent illumina- +
tion condition, or images of illumination invari- +
ant modalities. There is a hierarchical catego- +
rization of these two approaches. An exten- +
sive review of both approaches is given in [5]. +
3edb0fa2d6b0f1984e8e2c523c558cb026b2a983Automatic Age Estimation Based on +
Facial Aging Patterns +
3ee7a8107a805370b296a53e355d111118e96b7c
3e4acf3f2d112fc6516abcdddbe9e17d839f5d9bDeep Value Networks Learn to +
Evaluate and Iteratively Refine Structured Outputs +
3ea8a6dc79d79319f7ad90d663558c664cf298d4
3e4f84ce00027723bdfdb21156c9003168bc1c801979 +
© EURASIP, 2011 - ISSN 2076-1465 +
19th European Signal Processing Conference (EUSIPCO 2011) +
INTRODUCTION +
3e685704b140180d48142d1727080d2fb9e52163Single Image Action Recognition by Predicting +
Space-Time Saliency +
3e687d5ace90c407186602de1a7727167461194aPhoto Tagging by Collection-Aware People Recognition +
UFF +
UFF +
Asla S´a +
FGV +
IMPA +
501096cca4d0b3d1ef407844642e39cd2ff86b37Illumination Invariant Face Image +
Representation using Quaternions +
Dayron Rizo-Rodr´ıguez, Heydi M´endez-V´azquez, and Edel Garc´ıa-Reyes +
Advanced Technologies Application Center. 7a # 21812 b/ 218 and 222, +
Rpto. Siboney, Playa, P.C. 12200, La Habana, Cuba. +
501eda2d04b1db717b7834800d74dacb7df58f91
5083c6be0f8c85815ead5368882b584e4dfab4d1 Please do not quote. In press, Handbook of affective computing. New York, NY: Oxford +
Automated Face Analysis for Affective Computing +
500b92578e4deff98ce20e6017124e6d2053b451
50ff21e595e0ebe51ae808a2da3b7940549f4035IEEE TRANSACTIONS ON LATEX CLASS FILES, VOL. XX, NO. X, AUGUST 2017 +
Age Group and Gender Estimation in the Wild with +
Deep RoR Architecture +
5042b358705e8d8e8b0655d07f751be6a1565482International Journal of +
Emerging Research in Management &Technology +
ISSN: 2278-9359 (Volume-4, Issue-8) +
Research Article +
August +
2015 +
Review on Emotion Detection in Image +
CSE & PCET, PTU HOD, CSE & PCET, PTU +
Punjab, India Punj ab, India +
50e47857b11bfd3d420f6eafb155199f4b41f6d7International Journal of Computer, Consumer and Control (IJ3C), Vol. 2, No.1 (2013) +
3D Human Face Reconstruction Using a Hybrid of Photometric +
Stereo and Independent Component Analysis +
50eb75dfece76ed9119ec543e04386dfc95dfd13Learning Visual Entities and their Visual Attributes from Text Corpora +
Dept. of Computer Science +
K.U.Leuven, Belgium +
Dept. of Computer Science +
K.U.Leuven, Belgium +
Dept. of Computer Science +
K.U.Leuven, Belgium +
50a0930cb8cc353e15a5cb4d2f41b365675b5ebf
50d15cb17144344bb1879c0a5de7207471b9ff74Divide, Share, and Conquer: Multi-task +
Attribute Learning with Selective Sharing +
5028c0decfc8dd623c50b102424b93a8e9f2e390Published as a conference paper at ICLR 2017 +
REVISITING CLASSIFIER TWO-SAMPLE TESTS +
1Facebook AI Research, 2WILLOW project team, Inria / ENS / CNRS +
505e55d0be8e48b30067fb132f05a91650666c41A Model of Illumination Variation for Robust Face Recognition +
Institut Eur´ecom +
Multimedia Communications Department +
BP 193, 06904 Sophia Antipolis Cedex, France +
680d662c30739521f5c4b76845cb341dce010735Int J Comput Vis (2014) 108:82–96 +
DOI 10.1007/s11263-014-0716-6 +
Part and Attribute Discovery from Relative Annotations +
Received: 25 February 2013 / Accepted: 14 March 2014 / Published online: 26 April 2014 +
© Springer Science+Business Media New York 2014 +
68d2afd8c5c1c3a9bbda3dd209184e368e4376b9Representation Learning by Rotating Your Faces +
68a3f12382003bc714c51c85fb6d0557dcb15467
68d08ed9470d973a54ef7806318d8894d87ba610Drive Video Analysis for the Detection of Traffic Near-Miss Incidents +
68caf5d8ef325d7ea669f3fb76eac58e0170fff0
68d4056765c27fbcac233794857b7f5b8a6a82bfExample-Based Face Shape Recovery Using the +
Zenith Angle of the Surface Normal +
Mario Castel´an1, Ana J. Almaz´an-Delf´ın2, Marco I. Ram´ırez-Sosa-Mor´an3, +
and Luz A. Torres-M´endez1 +
1 CINVESTAV Campus Saltillo, Ramos Arizpe 25900, Coahuila, M´exico +
2 Universidad Veracruzana, Facultad de F´ısica e Inteligencia Artificial, Xalapa 91000, +
3 ITESM, Campus Saltillo, Saltillo 25270, Coahuila, M´exico +
Veracruz, M´exico +
684f5166d8147b59d9e0938d627beff8c9d208ddIEEE TRANS. NNLS, JUNE 2017 +
Discriminative Block-Diagonal Representation +
Learning for Image Recognition +
68cf263a17862e4dd3547f7ecc863b2dc53320d8
68e9c837431f2ba59741b55004df60235e50994dDetecting Faces Using Region-based Fully +
Convolutional Networks +
Tencent AI Lab, China +
687e17db5043661f8921fb86f215e9ca2264d4d2A Robust Elastic and Partial Matching Metric for Face Recognition +
Microsoft Corporate +
One Microsoft Way, Redmond, WA 98052 +
688754568623f62032820546ae3b9ca458ed0870bioRxiv preprint first posted online Sep. 27, 2016; +
doi: +
http://dx.doi.org/10.1101/077784 +
. +
The copyright holder for this preprint (which was not +
peer-reviewed) is the author/funder. It is made available under a +
CC-BY-NC-ND 4.0 International license +
. +
Resting high frequency heart rate variability is not associated with the +
recognition of emotional facial expressions in healthy human adults. +
1 Univ. Grenoble Alpes, LPNC, F-38040, Grenoble, France +
2 CNRS, LPNC UMR 5105, F-38040, Grenoble, France +
3 IPSY, Université Catholique de Louvain, Louvain-la-Neuve, Belgium +
4 Fund for Scientific Research (FRS-FNRS), Brussels, Belgium +
Correspondence concerning this article should be addressed to Brice Beffara, Office E250, Institut +
de Recherches en Sciences Psychologiques, IPSY - Place du Cardinal Mercier, 10 bte L3.05.01 B-1348 +
Author note +
This study explores whether the myelinated vagal connection between the heart and the brain +
is involved in emotion recognition. The Polyvagal theory postulates that the activity of the +
myelinated vagus nerve underlies socio-emotional skills. It has been proposed that the perception +
of emotions could be one of this skills dependent on heart-brain interactions. However, this +
assumption was differently supported by diverging results suggesting that it could be related to +
confounded factors. In the current study, we recorded the resting state vagal activity (reflected by +
High Frequency Heart Rate Variability, HF-HRV) of 77 (68 suitable for analysis) healthy human +
adults and measured their ability to identify dynamic emotional facial expressions. Results show +
that HF-HRV is not related to the recognition of emotional facial expressions in healthy human +
adults. We discuss this result in the frameworks of the polyvagal theory and the neurovisceral +
integration model. +
Keywords: HF-HRV; autonomic flexibility; emotion identification; dynamic EFEs; Polyvagal +
theory; Neurovisceral integration model +
Word count: 9810 +
10 +
11 +
12 +
13 +
14 +
15 +
16 +
17 +
Introduction +
The behavior of an animal is said social when involved in in- +
teractions with other animals (Ward & Webster, 2016). These +
interactions imply an exchange of information, signals, be- +
tween at least two animals. In humans, the face is an efficient +
communication channel, rapidly providing a high quantity of +
information. Facial expressions thus play an important role +
in the transmission of emotional information during social +
interactions. The result of the communication is the combina- +
tion of transmission from the sender and decoding from the +
receiver (Jack & Schyns, 2015). As a consequence, the quality +
of the interaction depends on the ability to both produce and +
identify facial expressions. Emotions are therefore a core +
feature of social bonding (Spoor & Kelly, 2004). Health +
of individuals and groups depend on the quality of social +
bonds in many animals (Boyer, Firat, & Leeuwen, 2015; S. L. +
Brown & Brown, 2015; Neuberg, Kenrick, & Schaller, 2011), +
18 +
19 +
20 +
21 +
22 +
23 +
24 +
25 +
26 +
27 +
28 +
29 +
30 +
31 +
32 +
33 +
34 +
35 +
especially in highly social species such as humans (Singer & +
Klimecki, 2014). +
The recognition of emotional signals produced by others is +
not independent from its production by oneself (Niedenthal, +
2007). The muscles of the face involved in the production of +
a facial expressions are also activated during the perception of +
the same facial expressions (Dimberg, Thunberg, & Elmehed, +
2000). In other terms, the facial mimicry of the perceived +
emotional facial expression (EFE) triggers its sensorimotor +
simulation in the brain, which improves the recognition abili- +
ties (Wood, Rychlowska, Korb, & Niedenthal, 2016). Beyond +
that, the emotion can be seen as the body -including brain- +
dynamic itself (Gallese & Caruana, 2016) which helps to un- +
derstand why behavioral simulation is necessary to understand +
the emotion. +
The interplay between emotion production, emotion percep- +
tion, social communication and body dynamics has been sum- +
marized in the framework of the polyvagal theory (Porges, +
68f9cb5ee129e2b9477faf01181cd7e3099d1824ALDA Algorithms for Online Feature Extraction +
68bf34e383092eb827dd6a61e9b362fcba36a83a
6889d649c6bbd9c0042fadec6c813f8e894ac6ccAnalysis of Robust Soft Learning Vector +
Quantization and an application to Facial +
Expression Recognition +
68c17aa1ecbff0787709be74d1d98d9efd78f410International Journal of Optomechatronics, 6: 92–119, 2012 +
Copyright # Taylor & Francis Group, LLC +
ISSN: 1559-9612 print=1559-9620 online +
DOI: 10.1080/15599612.2012.663463 +
GENDER CLASSIFICATION FROM FACE IMAGES +
USING MUTUAL INFORMATION AND FEATURE +
FUSION +
Department of Electrical Engineering and Advanced Mining Technology +
Center, Universidad de Chile, Santiago, Chile +
In this article we report a new method for gender classification from frontal face images +
using feature selection based on mutual information and fusion of features extracted from +
intensity, shape, texture, and from three different spatial scales. We compare the results of +
three different mutual information measures: minimum redundancy and maximal relevance +
(mRMR), normalized mutual information feature selection (NMIFS), and conditional +
mutual information feature selection (CMIFS). We also show that by fusing features +
extracted from six different methods we significantly improve the gender classification +
results relative to those previously published, yielding 99.13% of the gender classification +
rate on the FERET database. +
Keywords: Feature fusion, feature selection, gender classification, mutual information, real-time gender +
classification +
1. INTRODUCTION +
During the 90’s, one of the main issues addressed in the area of computer +
vision was face detection. Many methods and applications were developed including +
the face detection used in many digital cameras nowadays. Gender classification is +
important in many possible applications including electronic marketing. Displays +
at retail stores could show products and offers according to the person gender as +
the person passes in front of a camera at the store. This is not a simple task since +
faces are not rigid and depend on illumination, pose, gestures, facial expressions, +
occlusions (glasses), and other facial features (makeup, beard). The high variability +
in the appearance of the face directly affects their detection and classification. Auto- +
matic classification of gender from face images has a wide range of possible applica- +
tions, ranging from human-computer interaction to applications in real-time +
electronic marketing in retail stores (Shan 2012; Bekios-Calfa et al. 2011; Chu +
et al. 2010; Perez et al. 2010a). +
Automatic gender classification has a wide range of possible applications for +
improving human-machine interaction and face identification methods (Irick et al. +
ing.uchile.cl +
92 +
6888f3402039a36028d0a7e2c3df6db94f5cb9bbUnder review as a conference paper at ICLR 2018 +
CLASSIFIER-TO-GENERATOR ATTACK: ESTIMATION +
OF TRAINING DATA DISTRIBUTION FROM CLASSIFIER +
Anonymous authors +
Paper under double-blind review +
574751dbb53777101502419127ba8209562c4758
57b8b28f8748d998951b5a863ff1bfd7ca4ae6a5
57101b29680208cfedf041d13198299e2d396314
57893403f543db75d1f4e7355283bdca11f3ab1b
57f8e1f461ab25614f5fe51a83601710142f8e88Region Selection for Robust Face Verification using UMACE Filters +
Department of Electrical, Electronic and Systems Engineering, Faculty of Engineering, +
Universiti Kebangsaan Malaysia, 43600 Bangi, Selangor, Malaysia. +
In this paper, we investigate the verification performances of four subdivided face images with varying expressions. The +
objective of this study is to evaluate which part of the face image is more tolerant to facial expression and still retains its personal +
characteristics due to the variations of the image. The Unconstrained Minimum Average Correlation Energy (UMACE) filter is +
implemented to perform the verification process because of its advantages such as shift–invariance, ability to trade-off between +
discrimination and distortion tolerance, e.g. variations in pose, illumination and facial expression. The database obtained from the +
facial expression database of Advanced Multimedia Processing (AMP) Lab at CMU is used in this study. Four equal +
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 +
half of the face region gives the best performance in terms of the PSR values with zero false accepted rate (FAR) and zero false +
rejection rate (FRR) compared to the other three regions. +
1. Introduction +
Face recognition is a well established field of research, +
and a large number of algorithms have been proposed in the +
literature. Various classifiers have been explored to improve +
the accuracy of face classification. The basic approach is to +
use distance-base methods which measure Euclidean distance +
between any two vectors and then compare it with the preset +
threshold. Neural Networks are often used as classifiers due +
to their powerful generation ability [1]. Support Vector +
Machines (SVM) have been applied with encouraging results +
[2]. +
In biometric applications, one of the important tasks is the +
matching process between an individual biometrics against +
the database that has been prepared during the enrolment +
stage. For biometrics systems such as face authentication that +
use images as personal characteristics, biometrics sensor +
output and image pre-processing play an important role since +
the quality of a biometric input can change significantly due +
to illumination, noise and pose variations. Over the years, +
researchers have studied the role of illumination variation, +
pose variation, facial expression, and occlusions in affecting +
the performance of face verification systems [3]. +
The Minimum Average Correlation Energy (MACE) +
filters have been reported to be an alternative solution to these +
problems because of the advantages such as shift-invariance, +
close-form expressions and distortion-tolerance. MACE +
filters have been successfully applied in the field of automatic +
target recognition as well as in biometric verification [3][4]. +
Face and fingerprint verification using correlation filters have +
been investigated in [5] and [6], respectively. Savvides et.al +
performed face authentication and identification using +
correlation filters based on illumination variation [7]. In the +
process of implementing correlation filters, the number of +
training images used depends on the level of distortions +
applied to the images [5], [6]. +
In this study, we investigate which part of a face image is +
more tolerant to facial expression and retains its personal +
characteristics for the verification process. Four subdivided +
face images, i.e. bottom, top, left and right halves, with +
varying expressions are investigated. By identifying only the +
region of the face that gives the highest verification +
performance, that region can be used instead of the full-face +
to reduce storage requirements. +
2. Unconstrained Minimum Average Correlation +
Energy (UMACE) Filter +
Correlation filter theory and the descriptions of the design +
of the correlation filter can be found in a tutorial survey paper +
[8]. According to [4][6], correlation filter evolves from +
matched filters which are optimal for detecting a known +
reference image in the presence of additive white Gaussian +
noise. However, the detection rate of matched filters +
decreases significantly due to even the small changes of scale, +
rotation and pose of the reference image. +
the pre-specified peak values +
In an effort to solve this problem, the Synthetic +
Discriminant Function (SDF) filter and the Equal Correlation +
Peak SDF (ECP SDF) filter ware introduced which allowed +
several training images to be represented by a single +
correlation filter. SDF filter produces pre-specified values +
called peak constraints. These peak values correspond to the +
authentic class or impostor class when an image is tested. +
However, +
to +
misclassifications when the sidelobes are larger than the +
controlled values at the origin. +
Savvides et.al developed +
the Minimum Average +
Correlation Energy (MACE) filters [5]. This filter reduces the +
large sidelobes and produces a sharp peak when the test +
image is from the same class as the images that have been +
used to design the filter. There are two kinds of variants that +
can be used in order to obtain a sharp peak when the test +
image belongs to the authentic class. The first MACE filter +
variant minimizes the average correlation energy of the +
training images while constraining the correlation output at +
the origin to a specific value for each of the training images. +
The second MACE filter variant is the Unconstrained +
Minimum Average Correlation Energy (UMACE) filter +
which also minimizes the average correlation output while +
maximizing the correlation output at the origin [4]. +
lead +
Proceedings of the International Conference onElectrical Engineering and InformaticsInstitut Teknologi Bandung, Indonesia June 17-19, 2007B-67ISBN 978-979-16338-0-2611
57a1466c5985fe7594a91d46588d969007210581A Taxonomy of Face-models for System Evaluation +
Motivation and Data Types +
Synthetic Data Types +
Unverified – Have no underlying physical or +
statistical basis +
Physics -Based – Based on structure and +
materials combined with the properties +
formally modeled in physics. +
Statistical – Use statistics from real +
data/experiments to estimate/learn model +
parameters. Generally have measurements +
of accuracy +
Guided Synthetic – Individual models based +
on individual people. No attempt to capture +
properties of large groups, a unique model +
per person. For faces, guided models are +
composed of 3D structure models and skin +
textures, capturing many artifacts not +
easily parameterized. Can be combined with +
physics-based rendering to generate samples +
under different conditions. +
Semi–Synethetic – Use measured data such +
as 2D images or 3D facial scans. These are +
not truly synthetic as they are re-rendering’s +
of real measured data. +
Semi and Guided Synthetic data provide +
higher operational relevance while +
maintaining a high degree of control. +
Generating statistically significant size +
datasets for face matching system +
evaluation is both a laborious and +
expensive process. +
There is a gap in datasets that allow for +
evaluation of system issues including: +
 Long distance recognition +
 Blur caused by atmospherics +
 Various weather conditions +
 End to end systems evaluation +
Our contributions: +
 Define a taxonomy of face-models +
for controlled experimentations +
 Show how Synthetic addresses gaps +
in system evaluation +
 Show a process for generating and +
validating synthetic models +
 Use these models in long distance +
face recognition system evaluation +
Experimental Setup +
Results and Conclusions +
Example Models +
Original Pie +
Semi- +
Synthetic +
FaceGen +
Animetrics +
http://www.facegen.com +
http://www.animetrics.com/products/Forensica.php +
Guided- +
Synthetic +
Models +
 Models generated using the well +
known CMU PIE [18] dataset. Each of +
the 68 subjects of PIE were modeled +
using a right profile and frontal +
image from the lights subset. +
 Two modeling programs were used, +
Facegen and Animetrics. Both +
programs create OBJ files and +
textures +
 Models are re-rendered using +
custom display software built with +
OpenGL, GLUT and DevIL libraries +
 Custom Display Box housing a BENQ SP820 high +
powered projector rated at 4000 ANSI Lumens +
 Canon EOS 7D withd a Sigma 800mm F5.6 EX APO +
DG HSM lens a 2x adapter imaging the display +
from 214 meters +
Normalized Example Captures +
Real PIE 1 Animetrics +
FaceGen +
81M inside 214M outside +
Real PIE 2 +
 Pre-cropped images were used for the +
commercial core +
 Ground truth eye points + geometric/lighting +
normalization pre processing before running +
through the implementation of the V1 +
recognition algorithm found in [1]. +
 Geo normalization highlights how the feature +
region of the models looks very similar to +
that of the real person. +
Each test consisted of using 3 approximately frontal gallery images NOT used to +
make the 3D model used as the probe, best score over 3 images determined score. +
Even though the PIE-3D-20100224A–D sets were imaged on the same day, the V1 +
core scored differently on each highlighting the synthetic data’s ability to help +
evaluate data capture methods and effects of varying atmospherics. The ISO setting +
varied which effects the shutter speed, with higher ISO generally yielding less blur. +
Dataset +
Range(m) +
Iso +
V1 +
Comm. +
Original PIE Images +
FaceGen ScreenShots +
Animetrics Screenshots +
PIE-3D-20100210B +
PIE-3D-20100224A +
PIE-3D-20100224B +
PIE-3D-20100224C +
PIE-3D-20100224D +
N/A +
N/A +
N/A +
81m +
214m +
214m +
214m +
214m +
N/A +
N/A +
N/A +
500 +
125 +
125 +
250 +
400 +
100 +
47.76 +
100 +
100 +
58.82 +
45.59 +
81.82 +
79.1 +
100 +
100 +
100 +
100 +
100 +
100 +
 The same (100 percent) recognition rate on screenshots as original images +
validate the Anmetrics guided synthetic models and fails FaceGen Models. +
 100% recognition means dataset is too small/easy; exapanding pose and models +
underway. +
 Expanded the photohead methodology into 3D +
 Developed a robust modeling system allowing for multiple configurations of a +
single real life data set. +
 Gabor+SVM based V1[15] significantly more impacted by atmospheric blur than +
the commercial algorithm +
Key References: +
[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 +
[8 of 21] T. Boult and W. Scheirer. Long range facial image acquisition and quality. In M. Tisarelli, S. Li, and R. Chellappa. +
[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. +
[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. +
5721216f2163d026e90d7cd9942aeb4bebc92334
5753b2b5e442eaa3be066daa4a2ca8d8a0bb1725
574ad7ef015995efb7338829a021776bf9daaa08AdaScan: Adaptive Scan Pooling in Deep Convolutional Neural Networks +
for Human Action Recognition in Videos +
1IIT Kanpur‡ +
2SRI International +
3UCSD +
57d37ad025b5796457eee7392d2038910988655aGEERATVEEETATF +
ERARCCAVETYDETECTR +
by +
DagaEha +
UdeheS eviif +
f.DahaWeiha +
ATheiS biediaiaF (cid:28)efhe +
Re ieefheDegeef +
aefSciece +
a +
TheSchfC eScieceadEgieeig +
ebewUiveiyfe aeae91904 +
Decebe2009 +
3b1260d78885e872cf2223f2c6f3d6f6ea254204
3b1aaac41fc7847dd8a6a66d29d8881f75c91ad5Sparse Representation-based Open Set Recognition +
3bc776eb1f4e2776f98189e17f0d5a78bb755ef4
3b4fd2aec3e721742f11d1ed4fa3f0a86d988a10Glimpse: Continuous, Real-Time Object Recognition on +
Mobile Devices +
MIT CSAIL +
Microsoft Research +
MIT CSAIL +
Microsoft Research +
MIT CSAIL +
3b15a48ffe3c6b3f2518a7c395280a11a5f58ab0On Knowledge Transfer in +
Object Class Recognition +
A dissertation approved by +
TECHNISCHE UNIVERSITÄT DARMSTADT +
Fachbereich Informatik +
for the degree of +
Doktor-Ingenieur (Dr.-Ing.) +
presented by +
Dipl.-Inform. +
born in Mainz, Germany +
Prof. Dr.-Ing. Michael Goesele, examiner +
Prof. Martial Hebert, Ph.D., co-examiner +
Prof. Dr. Bernt Schiele, co-examiner +
Date of Submission: 12th of August, 2010 +
Date of Defense: 23rd of September, 2010 +
Darmstadt, 2010 +
D17 +
3ba8f8b6bfb36465018430ffaef10d2caf3cfa7eLocal Directional Number Pattern for Face +
Analysis: Face and Expression Recognition +
3b80bf5a69a1b0089192d73fa3ace2fbb52a4ad5
3b9d94752f8488106b2c007e11c193f35d941e92CVPR +
#2052 +
000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
045 +
046 +
047 +
048 +
049 +
050 +
051 +
052 +
053 +
CVPR 2013 Submission #2052. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +
CVPR +
#2052 +
Appearance, Visual and Social Ensembles for +
Face Recognition in Personal Photo Collections +
Anonymous CVPR submission +
Paper ID 2052 +
3be7b7eb11714e6191dd301a696c734e8d07435f
3b410ae97e4564bc19d6c37bc44ada2dcd608552Scalability Analysis of Audio-Visual Person +
Identity Verification +
1 Communications Laboratory, +
Universit´e catholique de Louvain, B-1348 Belgium, +
2 IDIAP, CH-1920 Martigny, +
Switzerland +
6f5ce5570dc2960b8b0e4a0a50eab84b7f6af5cbLow Resolution Face Recognition Using a +
Two-Branch Deep Convolutional Neural Network +
Architecture +
6f288a12033fa895fb0e9ec3219f3115904f24deLearning Expressionlets via Universal Manifold +
Model for Dynamic Facial Expression Recognition +
6f2dc51d607f491dbe6338711c073620c85351ac
6f75697a86d23d12a14be5466a41e5a7ffb79fad
6f7d06ced04ead3b9a5da86b37e7c27bfcedbbddPages 51.1-51.12 +
DOI: https://dx.doi.org/10.5244/C.30.51 +
6f7a8b3e8f212d80f0fb18860b2495be4c363eacCreating Capsule Wardrobes from Fashion Images +
UT-Austin +
UT-Austin +
6f6b4e2885ea1d9bea1bb2ed388b099a5a6d9b81Structured Output SVM Prediction of Apparent Age, +
Gender and Smile From Deep Features +
Michal Uˇriˇc´aˇr +
CMP, Dept. of Cybernetics +
FEE, CTU in Prague +
Computer Vision Lab +
D-ITET, ETH Zurich +
Computer Vision Lab +
D-ITET, ETH Zurich +
PSI, ESAT, KU Leuven +
CVL, D-ITET, ETH Zurich +
Jiˇr´ı Matas +
CMP, Dept. of Cybernetics +
FEE, CTU in Prague +
6f35b6e2fa54a3e7aaff8eaf37019244a2d39ed3DOI 10.1007/s00530-005-0177-4 +
R E G U L A R PA P E R +
Learning probabilistic classifiers for human–computer +
interaction applications +
Published online: 10 May 2005 +
c(cid:1) Springer-Verlag 2005 +
intelligent +
interaction, +
6fa3857faba887ed048a9e355b3b8642c6aab1d8Face Recognition in Challenging Environments: +
An Experimental and Reproducible Research +
Survey +
6f7ce89aa3e01045fcd7f1c1635af7a09811a1fe978-1-4673-0046-9/12/$26.00 ©2012 IEEE +
937 +
ICASSP 2012 +
6fe2efbcb860767f6bb271edbb48640adbd806c3SOFT BIOMETRICS: HUMAN IDENTIFICATION USING COMPARATIVE DESCRIPTIONS +
Soft Biometrics; Human Identification using +
Comparative Descriptions +
6fdc0bc13f2517061eaa1364dcf853f36e1ea5aeDAISEE: Dataset for Affective States in +
E-Learning Environments +
1 Microsoft India R&D Pvt. Ltd. +
2 Department of Computer Science, IIT Hyderabad +
6f5151c7446552fd6a611bf6263f14e729805ec75KHHAO /7 %:0 7 +
)>IJH=?J 9EJDE JDA ?JANJ B=?A ANFHAIIE ?=IIE?=JE KIEC JDA +
FH>=>EEJEAI JD=J A=?D A B IALAH= ?O ??KHHEC )7 CHKFI EI +
?=IIIAF=H=>EEJO MAECDJEC +
/=>H M=LAAJI H FHE?EF= ?FAJI ==OIEI 2+) ! 1 JDEI F=FAH MA +
03c56c176ec6377dddb6a96c7b2e95408db65a7aA Novel Geometric Framework on Gram Matrix +
Trajectories for Human Behavior Understanding +
03d9ccce3e1b4d42d234dba1856a9e1b28977640
0322e69172f54b95ae6a90eb3af91d3daa5e36eaFace Classification using Adjusted Histogram in +
Grayscale +
03f7041515d8a6dcb9170763d4f6debd50202c2bClustering Millions of Faces by Identity +
038ce930a02d38fb30d15aac654ec95640fe5cb0Approximate Structured Output Learning for Constrained Local +
Models with Application to Real-time Facial Feature Detection and +
Tracking on Low-power Devices +
03c1fc9c3339813ed81ad0de540132f9f695a0f8Proceedings of Machine Learning Research 81:1–15, 2018 +
Conference on Fairness, Accountability, and Transparency +
Gender Shades: Intersectional Accuracy Disparities in +
Commercial Gender Classification∗ +
MIT Media Lab 75 Amherst St. Cambridge, MA 02139 +
Microsoft Research 641 Avenue of the Americas, New York, NY 10011 +
Editors: Sorelle A. Friedler and Christo Wilson +
0339459a5b5439d38acd9c40a0c5fea178ba52fbD|C|I&I 2009 Prague +
Multimodal recognition of emotions in car +
environments +
03a8f53058127798bc2bc0245d21e78354f6c93bMax-Margin Additive Classifiers for Detection +
Sam Hare +
VGG Reading Group +
October 30, 2009 +
03fc466fdbc8a2efb6e3046fcc80e7cb7e86dc20A Real Time System for Model-based Interpretation of +
the Dynamics of Facial Expressions +
Technische Universit¨at M¨unchen +
Boltzmannstr. 3, 85748 Garching +
1. Motivation +
Recent progress in the field of Computer Vision allows +
intuitive interaction via speech, gesture or facial expressions +
between humans and technical systems.Model-based tech- +
niques facilitate accurately interpreting images with faces +
by exploiting a priori knowledge, such as shape and texture +
information. This renders them an inevitable component +
to realize the paradigm of intuitive human-machine interac- +
tion. +
Our demonstration shows model-based recognition of +
facial expressions in real-time via the state-of-the-art +
Candide-3 face model [1] as visible in Figure 1. This three- +
dimensional and deformable model is highly appropriate +
for real-world face interpretation applications. However, +
its complexity challenges the task of model fitting and we +
tackle this challenge with an algorithm that has been auto- +
matically learned from a large set of images. This solution +
provides both, high accuracy and runtime. Note, that our +
system is not limited to facial expression estimation. Gaze +
direction, gender and age are also estimated. +
2. Face Model Fitting +
Models reduce the large amount of image data to a +
small number of model parameters to describe the im- +
age content, which facilitates and accelerates the subse- +
quent interpretation task. Cootes et al. [3] introduced mod- +
elling shapes with Active Contours. Further enhancements +
emerged the idea of expanding shape models with texture +
information [2]. Recent research considers modelling faces +
in 3D space [1, 10]. +
Fitting the face model is the computational challenge of +
finding the parameters that best describe the face within a +
given image. This task is often addressed by minimizing +
an objective function, such as the pixel error between the +
model’s rendered surface and the underlying image content. +
This section describes the four main components of model- +
based techniques, see [9]. +
The face model contains a parameter vector p that repre- +
sents its configurations. We integrate the complex and de- +
formable 3D wire frame Candide-3 face model [1]. The +
model consists of 116 anatomical landmarks and its param- +
eter vector p = (rx, ry, rz, s, tx, ty, σ, α)T describes the +
affine transformation (rx, ry, rz, s, tx, ty) and the deforma- +
tion (σ, α). The 79 deformation parameters indicate the +
shape of facial components such as the mouth, the eyes, or +
the eye brows, etc., see Figure 2. +
The localization algorithm computes an initial estimate of +
the model parameters that is further refined by the subse- +
quent fitting algorithm. Our system integrates the approach +
of [8], which detects the model’s affine transformation in +
case the image shows a frontal view face. +
The objective function yields a comparable value that +
specifies how accurately a parameterized model matches an +
image. Traditional approaches manually specify the objec- +
tive function in a laborious and erroneous task. In contrast, +
we automatically learn the objective function from a large +
set of training data based on objective information theoretic +
measures [9]. This approach does not require expert knowl- +
edge and it is domain-independently applicable. As a re- +
sult, this approach yields more robust and accurate objective +
functions, which greatly facilitate the task of the associated +
fitting algorithms. Accurately estimated model parameters +
in turn are required to infer correct high-level information, +
such as facial expression or gaze direction. +
Figure 1. Interpreting expressions with the Candide-3 face model. +
03b98b4a2c0b7cc7dae7724b5fe623a43eaf877bAcume: A Novel Visualization Tool for Understanding Facial +
Expression and Gesture Data +
03104f9e0586e43611f648af1132064cadc5cc07
03f14159718cb495ca50786f278f8518c0d8c8c92015 IEEE International Conference on Control System, Computing and Engineering, Nov 27 – Nov 29, 2015 Penang, Malaysia +
2015 IEEE International Conference on Control System, +
Computing and Engineering (ICCSCE2015) +
Technical Session 1A – DAY 1 – 27th Nov 2015 +
Time: 3.00 pm – 4.30 pm +
Venue: Jintan +
Topic: Signal and Image Processing +
3.00 pm – 3.15pm +
3.15 pm – 3.30pm +
3.30 pm – 3.45pm +
3.45 pm – 4.00pm +
4.00 pm – 4.15pm +
4.15 pm – 4.30pm +
4.30 pm – 4.45pm +
1A 01 ID3 +
Can Subspace Based Learning Approach Perform on Makeup Face +
Recognition? +
Khor Ean Yee, Pang Ying Han, Ooi Shih Yin and Wee Kuok Kwee +
1A 02 ID35 +
Performance Evaluation of HOG and Gabor Features for Vision-based +
Vehicle Detection +
1A 03 ID23 +
Experimental Method to Pre-Process Fuzzy Bit Planes before Low-Level +
Feature Extraction in Thermal Images +
Chan Wai Ti and Sim Kok Swee +
1A 04 ID84 +
Fractal-based Texture and HSV Color Features for Fabric Image Retrieval +
Nanik Suciati, Darlis Herumurti and Arya Yudhi Wijaya +
1A 05 ID168 +
Study of Automatic Melody Extraction Methods for Philippine Indigenous +
Music +
Jason Disuanco, Vanessa Tan, Franz de Leon +
1A 06 ID211 +
Acoustical Comparison between Voiced and Voiceless Arabic Phonemes of +
Malay +
Speakers +
Ali Abd Almisreb, Ahmad Farid Abidin, Nooritawati Md Tahir +
*shaded cell is the proposed session chair +
viii +
©Faculty of Electrical Engineering, Universiti Teknologi MARA +
0394040749195937e535af4dda134206aa830258Geodesic Entropic Graphs for Dimension and +
Entropy Estimation in Manifold Learning +
December 16, 2003 +
0334cc0374d9ead3dc69db4816d08c917316c6c4
0394e684bd0a94fc2ff09d2baef8059c2652ffb0Median Robust Extended Local Binary Pattern +
for Texture Classification +
Index Terms— Texture descriptors, rotation invariance, local +
binary pattern (LBP), feature extraction, texture analysis. +
how the texture recognition process works in humans as +
well as in the important role it plays in the wide variety of +
applications of computer vision and image analysis [1], [2]. +
The many applications of texture classification include medical +
image analysis and understanding, object recognition, biomet- +
rics, content-based image retrieval, remote sensing, industrial +
inspection, and document classification. +
As a classical pattern recognition problem, texture classifi- +
cation primarily consists of two critical subproblems: feature +
extraction and classifier designation [1], [2]. It is generally +
agreed that the extraction of powerful texture features plays a +
relatively more important role, since if poor features are used +
even the best classifier will fail to achieve good recognition +
results. Consequently, most research in texture classification +
focuses on the feature extraction part and numerous texture +
feature extraction methods have been developed, with excellent +
surveys given in [1]–[5]. Most existing methods have not, +
however, been capable of performing sufficiently well for +
real-world applications, which have demanding requirements +
including database size, nonideal environmental conditions, +
and running in real-time. +
03e88bf3c5ddd44ebf0e580d4bd63072566613ad
03f4c0fe190e5e451d51310bca61c704b39dcac8J Ambient Intell Human Comput +
DOI 10.1007/s12652-016-0406-z +
O R I G I N A L R E S E A R C H +
CHEAVD: a Chinese natural emotional audio–visual database +
Received: 30 March 2016 / Accepted: 22 August 2016 +
Ó Springer-Verlag Berlin Heidelberg 2016 +
031055c241b92d66b6984643eb9e05fd605f24e2Multi-fold MIL Training for Weakly Supervised Object Localization +
Inria∗ +
0332ae32aeaf8fdd8cae59a608dc8ea14c6e3136Int J Comput Vis +
DOI 10.1007/s11263-017-1009-7 +
Large Scale 3D Morphable Models +
Received: 15 March 2016 / Accepted: 24 March 2017 +
© The Author(s) 2017. This article is an open access publication +
034addac4637121e953511301ef3a3226a9e75fdImplied Feedback: Learning Nuances of User Behavior in Image Search +
Virginia Tech +
03701e66eda54d5ab1dc36a3a6d165389be0ce79179 +
Improved Principal Component Regression for Face +
Recognition Under Illumination Variations +
9b318098f3660b453fbdb7a579778ab5e9118c4c3931 +
Joint Patch and Multi-label Learning for Facial +
Action Unit and Holistic Expression Recognition +
classifiers without +
9b000ccc04a2605f6aab867097ebf7001a52b459
9b474d6e81e3b94e0c7881210e249689139b3e04VG-RAM Weightless Neural Networks for +
Face Recognition +
Departamento de Inform´atica +
Universidade Federal do Esp´ırito Santo +
Av. Fernando Ferrari, 514, 29075-910 - Vit´oria-ES +
Brazil +
1. Introduction +
Computerized human face recognition has many practical applications, such as access control, +
security monitoring, and surveillance systems, and has been one of the most challenging and +
active research areas in computer vision for many decades (Zhao et al.; 2003). Even though +
current machine recognition systems have reached a certain level of maturity, the recognition +
of faces with different facial expressions, occlusions, and changes in illumination and/or pose +
is still a hard problem. +
A general statement of the problem of machine recognition of faces can be formulated as fol- +
lows: given an image of a scene, (i) identify or (ii) verify one or more persons in the scene +
using a database of faces. In identification problems, given a face as input, the system reports +
back the identity of an individual based on a database of known individuals; whereas in veri- +
fication problems, the system confirms or rejects the claimed identity of the input face. In both +
cases, the solution typically involves segmentation of faces from scenes (face detection), fea- +
ture extraction from the face regions, recognition, or verification. In this chapter, we examine +
the recognition of frontal face images required in the context of identification problems. +
Many approaches have been proposed to tackle the problem of face recognition. One can +
roughly divide these into (i) holistic approaches, (ii) feature-based approaches, and (iii) hybrid +
approaches (Zhao et al.; 2003). Holistic approaches use the whole face region as the raw input +
to a recognition system (a classifier). In feature-based approaches, local features, such as the +
eyes, nose, and mouth, are first extracted and their locations and local statistics (geometric +
and/or appearance based) are fed into a classifier. Hybrid approaches use both local features +
and the whole face region to recognize a face. +
Among +
fisher- +
faces (Belhumeur et al.; 1997; Etemad and Chellappa; 1997) have proved to be effective +
(Turk and Pentland; +
eigenfaces +
holistic +
approaches, +
1991) +
and +
9bc01fa9400c231e41e6a72ec509d76ca797207c
9bcfadd22b2c84a717c56a2725971b6d49d3a804How to Detect a Loss of Attention in a Tutoring System +
using Facial Expressions and Gaze Direction +
9bac481dc4171aa2d847feac546c9f7299cc5aa0Matrix Product State for Higher-Order Tensor +
Compression and Classification +
9b7974d9ad19bb4ba1ea147c55e629ad7927c5d7Faical Expression Recognition by Combining +
Texture and Geometrical Features +
9ea73660fccc4da51c7bc6eb6eedabcce7b5ceadTalking Head Detection by Likelihood-Ratio Test† +
MIT Lincoln Laboratory, +
Lexington MA 02420, USA +
9e9052256442f4e254663ea55c87303c85310df9International Journal of Advanced Research in Computer Engineering & Technology (IJARCET) +
Volume 4 Issue 10, October 2015 +
Review On Attribute-assisted Reranking for +
Image Search +
 +
9e0285debd4b0ba7769b389181bd3e0fd7a02af6From face images and attributes to attributes +
Computer Vision Laboratory, ETH Zurich, Switzerland +
9e5c2d85a1caed701b68ddf6f239f3ff941bb707
04bb3fa0824d255b01e9db4946ead9f856cc0b59
040dc119d5ca9ea3d5fc39953a91ec507ed8cc5dNoname manuscript No. +
(will be inserted by the editor) +
Large-scale Bisample Learning on ID vs. Spot Face Recognition +
Received: date / Accepted: date +
04470861408d14cc860f24e73d93b3bb476492d0
0447bdb71490c24dd9c865e187824dee5813a676Manifold Estimation in View-based Feature +
Space for Face Synthesis Across Pose +
Paper 27 +
044ba70e6744e80c6a09fa63ed6822ae241386f2TO APPEAR IN AUTONOMOUS ROBOTS, SPECIAL ISSUE IN LEARNING FOR HUMAN-ROBOT COLLABORATION +
Early Prediction for Physical Human Robot +
Collaboration in the Operating Room +
04dcdb7cb0d3c462bdefdd05508edfcff5a6d315Assisting the training of deep neural networks +
with applications to computer vision +
tesi doctoral està subjecta a +
la +
Aquesta +
CompartirIgual 4.0. Espanya de Creative Commons. +
Esta tesis doctoral está sujeta a la licencia Reconocimiento - NoComercial – CompartirIgual +
4.0. España de Creative Commons. +
This doctoral thesis is licensed under the Creative Commons Attribution-NonCommercial- +
ShareAlike 4.0. Spain License. +
llicència Reconeixement- NoComercial – +
044fdb693a8d96a61a9b2622dd1737ce8e5ff4faDynamic Texture Recognition Using Local Binary +
Patterns with an Application to Facial Expressions +
04250e037dce3a438d8f49a4400566457190f4e2
0431e8a01bae556c0d8b2b431e334f7395dd803aLearning Localized Perceptual Similarity Metrics for Interactive Categorization +
Google Inc. +
google.com +
04b4c779b43b830220bf938223f685d1057368e9Video retrieval based on deep convolutional +
neural network +
Yajiao Dong +
School of Information and Electronics, +
Beijing Institution of Technology, Beijing, China +
Jianguo Li +
School of Information and Electronics, +
Beijing Institution of Technology, Beijing, China +
04616814f1aabe3799f8ab67101fbaf9fd115ae4UNIVERSIT´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
6a3a07deadcaaab42a0689fbe5879b5dfc3ede52Learning to Estimate Pose by Watching Videos +
Department of Computer Science and Engineering +
IIT Kanpur +
6ad107c08ac018bfc6ab31ec92c8a4b234f67d49
6a184f111d26787703f05ce1507eef5705fdda83
6a16b91b2db0a3164f62bfd956530a4206b23feaA Method for Real-Time Eye Blink Detection and Its Application +
Mahidol Wittayanusorn School +
Puttamonton, Nakornpatom 73170, Thailand +
6a806978ca5cd593d0ccd8b3711b6ef2a163d810Facial feature tracking for Emotional Dynamic +
Analysis +
1ISIR, CNRS UMR 7222 +
Univ. Pierre et Marie Curie, Paris +
2LAMIA, EA 4540 +
Univ. of Fr. West Indies & Guyana +
6a8a3c604591e7dd4346611c14dbef0c8ce9ba54ENTERFACE’10, JULY 12TH - AUGUST 6TH, AMSTERDAM, THE NETHERLANDS. +
58 +
An Affect-Responsive Interactive Photo Frame +
6aa43f673cc42ed2fa351cbc188408b724cb8d50
6aefe7460e1540438ffa63f7757c4750c844764dNon-rigid Segmentation using Sparse Low Dimensional Manifolds and +
Deep Belief Networks ∗ +
Instituto de Sistemas e Rob´otica +
Instituto Superior T´ecnico, Portugal +
6a1beb34a2dfcdf36ae3c16811f1aef6e64abff2
322c063e97cd26f75191ae908f09a41c534eba90Noname manuscript No. +
(will be inserted by the editor) +
Improving Image Classification using Semantic Attributes +
Received: date / Accepted: date +
325b048ecd5b4d14dce32f92bff093cd744aa7f8CVPR +
#2670 +
000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
045 +
046 +
047 +
048 +
049 +
050 +
051 +
052 +
053 +
CVPR 2008 Submission #2670. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +
CVPR +
#2670 +
Multi-Image Graph Cut Clothing Segmentation for Recognizing People +
Anonymous CVPR submission +
Paper ID 2670 +
321bd4d5d80abb1bae675a48583f872af3919172Wang et al. EURASIP Journal on Image and Video Processing (2016) 2016:44 +
DOI 10.1186/s13640-016-0152-3 +
EURASIP Journal on Image +
and Video Processing +
R EV I E W +
Entropy-weighted feature-fusion method +
for head-pose estimation +
Open Access +
32b8c9fd4e3f44c371960eb0074b42515f318ee7
32575ffa69d85bbc6aef5b21d73e809b37bf376d-)5741/ *1-641+ 5)2- 37)16; 1 6-45 . *1-641+ 1.4)61 +
7ELAHIEJO B JJ=M= +
)*564)+6 +
IKHA L=HE=JEI E >EAJHE? I=FA GK=EJO 9A >ACE MEJD +
IKHAAJI 9A JDA IDM JD=J JDA >EAJHE? EBH=JE BH +
JA EI JDA A= D(p(cid:107)q) BH = FAHII E JDA FFK=JE 1 +
BH I= ALAI B >KH MEJD = =IOFJJE? >AD=LEH =J =HCAH +
>KH +
 164,7+61 +
*EAJHE? I=FA GK=EJO EI = A=IKHA B JDA KIABKAII B = +
GK=EJO +
F=FAH MA FHFIA = AM =FFH=?D J A=IKHA JDEI GK=JEJO +
JDA EJKEJELA >IAHL=JE JD=J = DECD GK=EJO >EAJHE? E=CA +
>EAJHE? EBH=JE +
EIIKAI E >EAJHE? JA?DCO .H AN=FA A B JDA IJ +
? >EAJHE? GKAIJEI EI JD=J B KEGKAAII AC J MD=J +
ANJAJ =HA CAHFHEJI KEGKA .H JDA FEJ B LEAM B +
=>A EBH=JE EI =L=E=>A BH = CELA JA?DCO IK?D +
  $  " +
1 JDEI F=FAH MA A=>H=JA = =FFH=?D J +
BMI +
AJI +
 >ABHA = >EAJHE? A=IKHAAJ t0 =J MDE?D JEA MA O +
M = FAHI p EI F=HJ B = FFK=JE q MDE?D =O >A JDA +
324b9369a1457213ec7a5a12fe77c0ee9aef1ad4Dynamic Facial Analysis: From Bayesian Filtering to Recurrent Neural Network +
NVIDIA +
32df63d395b5462a8a4a3c3574ae7916b0cd4d1d978-1-4577-0539-7/11/$26.00 ©2011 IEEE +
1489 +
ICASSP 2011 +
35308a3fd49d4f33bdbd35fefee39e39fe6b30b7
352d61eb66b053ae5689bd194840fd5d33f0e9c0Analysis Dictionary Learning based +
Classification: Structure for Robustness +
3538d2b5f7ab393387ce138611ffa325b6400774A DSP-BASED APPROACH FOR THE IMPLEMENTATION OF FACE RECOGNITION +
ALGORITHMS +
A. U. Batur +
B. E. Flinchbaugh +
M. H. Hayes IIl +
Center for Signal and Image Proc. +
Georgia Inst. Of Technology +
Atlanta, GA +
Imaging and Audio Lab. +
Texas Instruments +
Dallas, TX +
Center for Signal and Image Proc. +
Georgia Inst. Of Technology +
Atlanta, CA +
3504907a2e3c81d78e9dfe71c93ac145b1318f9cNoname manuscript No. +
(will be inserted by the editor) +
Unconstrained Still/Video-Based Face Verification with Deep +
Convolutional Neural Networks +
Received: date / Accepted: date +
35b1c1f2851e9ac4381ef41b4d980f398f1aad68Geometry Guided Convolutional Neural Networks for +
Self-Supervised Video Representation Learning +
351c02d4775ae95e04ab1e5dd0c758d2d80c3dddActionSnapping: Motion-based Video +
Synchronization +
Disney Research +
35e4b6c20756cd6388a3c0012b58acee14ffa604Gender Classification in Large Databases +
E. Ram´on-Balmaseda, J. Lorenzo-Navarro, and M. Castrill´on-Santana (cid:63) +
Universidad de Las Palmas de Gran Canaria +
SIANI +
Spain +
357963a46dfc150670061dbc23da6ba7d6da786e
35f1bcff4552632419742bbb6e1927ef5e998eb4
35c973dba6e1225196566200cfafa150dd231fa8
35f084ddee49072fdb6e0e2e6344ce50c02457efA Bilinear Illumination Model +
for Robust Face Recognition +
The Harvard community has made this +
article openly available. Please share how +
this access benefits you. Your story matters +
Citation +
Machiraju. 2005. A bilinear illumination model for robust face +
recognition. Proceedings of the Tenth IEEE International Conference +
on Computer Vision: October 17-21, 2005, Beijing, China. 1177-1184. +
Los Almamitos, C.A.: IEEE Computer Society. +
Published Version +
doi:10.1109/ICCV.2005.5 +
Citable link +
http://nrs.harvard.edu/urn-3:HUL.InstRepos:4238979 +
Terms of Use +

repository, and is made available under the terms and conditions +
applicable to Other Posted Material, as set forth at http:// +
nrs.harvard.edu/urn-3:HUL.InstRepos:dash.current.terms-of- +
use#LAA +
353a89c277cca3e3e4e8c6a199ae3442cdad59b5
352110778d2cc2e7110f0bf773398812fd905eb1TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. X, NO. X, JUNE 2014 +
Matrix Completion for Weakly-supervised +
Multi-label Image Classification +
6964af90cf8ac336a2a55800d9c510eccc7ba8e1Temporal Relational Reasoning in Videos +
MIT CSAIL +
697b0b9630213ca08a1ae1d459fabc13325bdcbb
69d29012d17cdf0a2e59546ccbbe46fa49afcd68Subspace clustering of dimensionality-reduced data +
ETH Zurich, Switzerland +
69de532d93ad8099f4d4902c4cad28db958adfea
69526cdf6abbfc4bcd39616acde544568326d856636 +
[17] B. Moghaddam, T. Jebara, and A. Pentland, “Bayesian face recogni- +
tion,” Pattern Recognit., vol. 33, no. 11, pp. 1771–1782, Nov. 2000. +
[18] A. Nefian, “A hidden Markov model-based approach for face detection +
and recognition,” Ph.D. dissertation, Dept. Elect. Comput. Eng. Elect. +
Eng., Georgia Inst. Technol., Atlanta, 1999. +
[19] P. J. Phillips et al., “Overview of the face recognition grand challenge,” +
presented at the IEEE CVPR, San Diego, CA, Jun. 2005. +
[20] H. T. Tanaka, M. Ikeda, and H. Chiaki, “Curvature-based face surface +
recognition using spherical correlation-principal direction for curved +
object recognition,” in Proc. Int. Conf. Automatic Face and Gesture +
Recognition, 1998, pp. 372–377. +
[21] M. Turk and A. Pentland, “Eigenfaces for recognition,” J. Cognit. Sci., +
pp. 71–86, 1991. +
[22] V. N. Vapnik, Statistical Learning Theory. New York: Wiley, 1998. +
[23] W. Zhao, R. Chellappa, A. Rosenfeld, and P. Phillips, “Face recogni- +
tion: A literature survey,” ACM Comput. Surveys, vol. 35, no. 44, pp. +
399–458, 2003. +
[24] W. Zhao, R. Chellappa, and P. J. Phillips, “Subspace linear discrimi- +
nant analysis for face recognition,” UMD TR4009, 1999. +
Face Verification Using Template Matching +
690d669115ad6fabd53e0562de95e35f1078dfbbProgressive versus Random Projections for Compressive Capture of Images, +
Lightfields and Higher Dimensional Visual Signals +
MIT Media Lab +
75 Amherst St, Cambridge, MA +
MERL +
201 Broadway, Cambridge MA +
MIT Media Lab +
75 Amherst St, Cambridge, MA +
69a9da55bd20ce4b83e1680fbc6be2c976067631
6974449ce544dc208b8cc88b606b03d95c8fd368
3cfbe1f100619a932ba7e2f068cd4c41505c9f58A Realistic Simulation Tool for Testing Face Recognition +
Systems under Real-World Conditions∗ +
M. Correa, J. Ruiz-del-Solar, S. Parra-Tsunekawa, R. Verschae +
Department of Electrical Engineering, Universidad de Chile +
Advanced Mining Technology Center, Universidad de Chile +
3c03d95084ccbe7bf44b6d54151625c68f6e74d0
3cd7b15f5647e650db66fbe2ce1852e00c05b2e4
3ce2ecf3d6ace8d80303daf67345be6ec33b3a93
3c374cb8e730b64dacb9fbf6eb67f5987c7de3c8Measuring Gaze Orientation for Human-Robot +
Interaction +
∗ CNRS; LAAS; 7 avenue du Colonel Roche, 31077 Toulouse Cedex, France +
† Universit´e de Toulouse; UPS; LAAS-CNRS : F-31077 Toulouse, France +
Introduction +
In the context of Human-Robot interaction estimating gaze orientation brings +
useful information about human focus of attention. This is a contextual infor- +
mation : when you point something you usually look at it. Estimating gaze +
orientation requires head pose estimation. There are several techniques to esti- +
mate head pose from images, they are mainly based on training [3, 4] or on local +
face features tracking [6]. The approach described here is based on local face +
features tracking in image space using online learning, it is a mixed approach +
since we track face features using some learning at feature level. It uses SURF +
features [2] to guide detection and tracking. Such key features can be matched +
between images, used for object detection or object tracking [10]. Several ap- +
proaches work on fixed size images like training techniques which mainly work +
on low resolution images because of computation costs whereas approaches based +
on local features tracking work on high resolution images. Tracking face features +
such as eyes, nose and mouth is a common problem in many applications such as +
detection of facial expression or video conferencing [8] but most of those appli- +
cations focus on front face images [9]. We developed an algorithm based on face +
features tracking using a parametric model. First we need face detection, then +
we detect face features in following order: eyes, mouth, nose. In order to achieve +
full profile detection we use sets of SURF to learn what eyes, mouth and nose +
look like once tracking is initialized. Once those sets of SURF are known they +
are used to detect and track face features. SURF have a descriptor which is often +
used to identify a key point and here we add some global geometry information +
by using the relative position between key points. Then we use a particle filter to +
track face features using those SURF based detectors, we compute the head pose +
angles from features position and pass the results through a median filter. This +
paper is organized as follows. Section 2 describes our modeling of visual features, +
section 3 presents our tracking implementation. Section 4 presents results we get +
with our implementation and future works in section 5. +
2 Visual features +
We use some basic properties of facial features to initialize our algorithm : eyes +
are dark and circular, mouth is an horizontal dark line with a specific color,... +
3cb64217ca2127445270000141cfa2959c84d9e7
3cd5da596060819e2b156e8b3a28331ef633036b
3c56acaa819f4e2263638b67cea1ec37a226691dBody Joint guided 3D Deep Convolutional +
Descriptors for Action Recognition +
3c8da376576938160cbed956ece838682fa50e9fChapter 4 +
Aiding Face Recognition with +
Social Context Association Rule +
based Re-Ranking +
Humans are very efficient at recognizing familiar face images even in challenging condi- +
tions. One reason for such capabilities is the ability to understand social context between +
individuals. Sometimes the identity of the person in a photo can be inferred based on the +
identity of other persons in the same photo, when some social context between them is +
known. This chapter presents an algorithm to utilize the co-occurrence of individuals as +
the social context to improve face recognition. Association rule mining is utilized to infer +
multi-level social context among subjects from a large repository of social transactions. +
The results are demonstrated on the G-album and on the SN-collection pertaining to 4675 +
identities prepared by the authors from a social networking website. The results show that +
association rules extracted from social context can be used to augment face recognition and +
improve the identification performance. +
4.1 +
Introduction +
Face recognition capabilities of humans have inspired several researchers to understand +
the science behind it and use it in developing automated algorithms. Recently, it is also +
argued that encoding social context among individuals can be leveraged for improved +
automatic face recognition [175]. As shown in Figure 4.1, often times a person’s identity +
can be inferred based on the identity of other persons in the same photo, when some social +
context between them is known. A subject’s face in consumer photos generally co-occur +
along with their socially relevant people. With the advent of social networking services, +
the social context between individuals is readily available. Face recognition performance +
105 +
56e885b9094391f7d55023a71a09822b38b26447FREQUENCY DECODED LOCAL BINARY PATTERN +
Face Retrieval using Frequency Decoded Local +
Descriptor +
56a653fea5c2a7e45246613049fb16b1d204fc963287 +
Quaternion Collaborative and Sparse Representation +
With Application to Color Face Recognition +
representation-based +
5666ed763698295e41564efda627767ee55cc943Manuscript +
Click here to download Manuscript: template.tex +
Click here to view linked References +
Noname manuscript No. +
(will be inserted by the editor) +
Relatively-Paired Space Analysis: Learning a Latent Common +
Space from Relatively-Paired Observations +
Received: date / Accepted: date +
5615d6045301ecbc5be35e46cab711f676aadf3aDiscriminatively Learned Hierarchical Rank Pooling Networks +
Received: date / Accepted: date +
566038a3c2867894a08125efe41ef0a40824a090978-1-4244-2354-5/09/$25.00 ©2009 IEEE +
1945 +
ICASSP 2009 +
56dca23481de9119aa21f9044efd7db09f618704Riemannian Dictionary Learning and Sparse +
Coding for Positive Definite Matrices +
516a27d5dd06622f872f5ef334313350745eadc3> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < +
1 +
Fine-Grained Facial Expression Analysis Us- +
ing Dimensional Emotion Model +
 +
51c3050fb509ca685de3d9ac2e965f0de1fb21ccFantope Regularization in Metric Learning +
Marc T. Law +
Sorbonne Universit´es, UPMC Univ Paris 06, UMR 7606, LIP6, F-75005, Paris, France +
51c7c5dfda47647aef2797ac3103cf0e108fdfb4CS 395T: Celebrity Look-Alikes ∗ +
519f4eb5fe15a25a46f1a49e2632b12a3b18c94dNon-Lambertian Reflectance Modeling and +
Shape Recovery of Faces using Tensor Splines +
51528cdce7a92835657c0a616c0806594de7513b
5161e38e4ea716dcfb554ccb88901b3d97778f64SSPP-DAN: DEEP DOMAIN ADAPTATION NETWORK FOR +
FACE RECOGNITION WITH SINGLE SAMPLE PER PERSON +
School of Computing, KAIST, Republic of Korea +
51d1a6e15936727e8dd487ac7b7fd39bd2baf5eeJOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
A Fast and Accurate System for Face Detection, +
Identification, and Verification +
5157dde17a69f12c51186ffc20a0a6c6847f1a29Evolutionary Cost-sensitive Extreme Learning +
Machine +
1 +
51dc127f29d1bb076d97f515dca4cc42dda3d25b
3daafe6389d877fe15d8823cdf5ac15fd919676fHuman Action Localization +
with Sparse Spatial Supervision +
3db75962857a602cae65f60f202d311eb4627b41
3d36f941d8ec613bb25e80fb8f4c160c1a2848dfOut-of-sample generalizations for supervised +
manifold learning for classification +
3d5a1be4c1595b4805a35414dfb55716e3bf80d8Hidden Two-Stream Convolutional Networks for +
Action Recognition +
3d62b2f9cef997fc37099305dabff356d39ed477Joint Face Alignment and 3D Face +
Reconstruction with Application to Face +
Recognition +
3dc522a6576c3475e4a166377cbbf4ba389c041f
3dd4d719b2185f7c7f92cc97f3b5a65990fcd5ddEnsemble of Hankel Matrices for +
Face Emotion Recognition +
DICGIM, Universit´a degli Studi di Palermo, +
V.le delle Scienze, Ed. 6, 90128 Palermo, Italy, +
DRAFT +
To appear in ICIAP 2015 +
3dda181be266950ba1280b61eb63ac11777029f9
3d6ee995bc2f3e0f217c053368df659a5d14d5b5
3dd906bc0947e56d2b7bf9530b11351bbdff2358
3d1af6c531ebcb4321607bcef8d9dc6aa9f0dc5a1892 +
Random Multispace Quantization as +
an Analytic Mechanism for BioHashing +
of Biometric and Random Identity Inputs +
3d6943f1573f992d6897489b73ec46df983d776c
3d94f81cf4c3a7307e1a976dc6cb7bf38068a3813846 +
Data-Dependent Label Distribution Learning +
for Age Estimation +
5859774103306113707db02fe2dd3ac9f91f1b9e
5892f8367639e9c1e3cf27fdf6c09bb3247651edEstimating Missing Features to Improve Multimedia Information Retrieval +
5850aab97e1709b45ac26bb7d205e2accc798a87
587f81ae87b42c18c565694c694439c65557d6d5DeepFace: Face Generation using Deep Learning +
580054294ca761500ada71f7d5a78acb0e622f191331 +
A Subspace Model-Based Approach to Face +
Relighting Under Unknown Lighting and Poses +
58081cb20d397ce80f638d38ed80b3384af76869Embedded Real-Time Fall Detection Using Deep +
Learning For Elderly Care +
Samsung Research, Samsung Electronics +
58fa85ed57e661df93ca4cdb27d210afe5d2cdcdCancún Center, Cancún, México, December 4-8, 2016 +
978-1-5090-4847-2/16/$31.00 ©2016 IEEE +
4118 +
58bf72750a8f5100e0c01e55fd1b959b31e7dbcePyramidBox: A Context-assisted Single Shot +
Face Detector. +
Baidu Inc. +
58542eeef9317ffab9b155579256d11efb4610f2International Journal of Science and Research (IJSR) +
ISSN (Online): 2319-7064 +
Index Copernicus Value (2013): 6.14 | Impact Factor (2014): 5.611 +
Face Recognition Revisited on Pose, Alignment, +
Color, Illumination and Expression-PyTen +
Computer Science, BIT Noida, India +
58823377757e7dc92f3b70a973be697651089756Technical Report +
UCAM-CL-TR-861 +
ISSN 1476-2986 +
Number 861 +
Computer Laboratory +
Automatic facial expression analysis +
October 2014 +
15 JJ Thomson Avenue +
Cambridge CB3 0FD +
United Kingdom +
phone +44 1223 763500 +
http://www.cl.cam.ac.uk/ +
58bb77dff5f6ee0fb5ab7f5079a5e788276184ccFacial Expression Recognition with PCA and LBP +
Features Extracting from Active Facial Patches +
 +
58cb1414095f5eb6a8c6843326a6653403a0ee17
677585ccf8619ec2330b7f2d2b589a37146ffad7A flexible model for training action localization +
with varying levels of supervision +
677477e6d2ba5b99633aee3d60e77026fb0b9306
6789bddbabf234f31df992a3356b36a47451efc7Unsupervised Generation of Free-Form and +
Parameterized Avatars +
675b2caee111cb6aa7404b4d6aa371314bf0e647AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions +
Carl Vondrick∗ +
679b72d23a9cfca8a7fe14f1d488363f2139265f
67484723e0c2cbeb936b2e863710385bdc7d5368Anchor Cascade for Efficient Face Detection +
6742c0a26315d7354ab6b1fa62a5fffaea06da14BAS AND SMITH: WHAT DOES 2D GEOMETRIC INFORMATION REALLY TELL US ABOUT 3D FACE SHAPE? +
What does 2D geometric information +
really tell us about 3D face shape? +
67a50752358d5d287c2b55e7a45cc39be47bf7d0
67ba3524e135c1375c74fe53ebb03684754aae56978-1-5090-4117-6/17/$31.00 ©2017 IEEE +
1767 +
ICASSP 2017 +
6769cfbd85329e4815bb1332b118b01119975a95Tied factor analysis for face recognition across +
large pose changes +
0be43cf4299ce2067a0435798ef4ca2fbd255901Title +
A temporal latent topic model for facial expression recognition +
Author(s) +
Shang, L; Chan, KP +
Citation +
The 10th Asian Conference on Computer Vision (ACCV 2010), +
Queenstown, New Zealand, 8-12 November 2010. In Lecture +
Notes in Computer Science, 2010, v. 6495, p. 51-63 +
Issued Date +
2011 +
URL +
http://hdl.handle.net/10722/142604 +
Rights +
Creative Commons: Attribution 3.0 Hong Kong License +
0b2277a0609565c30a8ee3e7e193ce7f79ab48b0944 +
Cost-Sensitive Semi-Supervised Discriminant +
Analysis for Face Recognition +
0ba64f4157d80720883a96a73e8d6a5f5b9f1d9b
0b605b40d4fef23baa5d21ead11f522d7af1df06Label-Embedding for Attribute-Based Classification +
a Computer Vision Group∗, XRCE, France +
b LEAR†, INRIA, France +
0b0eb562d7341231c3f82a65cf51943194add0bb> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < +
Facial Image Analysis Based on Local Binary +
Patterns: A Survey +
+
0b3a146c474166bba71e645452b3a8276ac05998Who’s in the Picture? +
Berkeley, CA 94720 +
Computer Science Division +
U.C. Berkeley +
0b5bd3ce90bf732801642b9f55a781e7de7fdde0
0b0958493e43ca9c131315bcfb9a171d52ecbb8aA Unified Neural Based Model for Structured Output Problems +
Soufiane Belharbi∗1, Cl´ement Chatelain∗1, Romain H´erault∗1, and S´ebastien Adam∗2 +
1LITIS EA 4108, INSA de Rouen, Saint ´Etienne du Rouvray 76800, France +
2LITIS EA 4108, UFR des Sciences, Universit´e de Rouen, France. +
April 13, 2015 +
0b20f75dbb0823766d8c7b04030670ef7147ccdd1 +
Feature selection using nearest attributes +
0b5a82f8c0ee3640503ba24ef73e672d93aeebbfOn Learning 3D Face Morphable Model +
from In-the-wild Images +
0b174d4a67805b8796bfe86cd69a967d357ba9b6 Research Journal of Recent Sciences _________________________________________________ ISSN 2277-2502 +
Vol. 3(4), 56-62, April (2014) +
Res.J.Recent Sci. +
0ba449e312894bca0d16348f3aef41ca01872383
0b572a2b7052b15c8599dbb17d59ff4f02838ff7Automatic Subspace Learning via Principal +
Coefficients Embedding +
0ba99a709cd34654ac296418a4f41a9543928149
0b8c92463f8f5087696681fb62dad003c308ebe2On Matching Sketches with Digital Face Images +
in local +
0bc0f9178999e5c2f23a45325fa50300961e0226Recognizing facial expressions from videos using Deep +
Belief Networks +
CS 229 Project +
0b3f354e6796ef7416bf6dde9e0779b2fcfabed2
93675f86d03256f9a010033d3c4c842a732bf661Universit´edesSciencesetTechnologiesdeLilleEcoleDoctoraleSciencesPourl’ing´enieurUniversit´eLilleNord-de-FranceTHESEPr´esent´ee`al’Universit´edesSciencesetTechnologiesdeLillePourobtenirletitredeDOCTEURDEL’UNIVERSIT´ESp´ecialit´e:MicroetNanotechnologieParTaoXULocalizedgrowthandcharacterizationofsiliconnanowiresSoutenuele25Septembre2009Compositiondujury:Pr´esident:TuamiLASRIRapporteurs:ThierryBARONHenriMARIETTEExaminateurs:EricBAKKERSXavierWALLARTDirecteurdeth`ese:BrunoGRANDIDIER
936c7406de1dfdd22493785fc5d1e5614c6c28822012 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 762–772, +
Montr´eal, Canada, June 3-8, 2012. c(cid:13)2012 Association for Computational Linguistics +
762 +
93721023dd6423ab06ff7a491d01bdfe83db7754ROBUST FACE ALIGNMENT USING CONVOLUTIONAL NEURAL +
NETWORKS +
Orange Labs, 4, Rue du Clos Courtel, 35512 Cesson-S´evign´e, France +
Keywords: +
Face alignment, Face registration, Convolutional Neural Networks. +
93cbb3b3e40321c4990c36f89a63534b506b6dafIEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 35, NO. 3, JUNE 2005 +
477 +
Learning From Examples in the Small Sample Case: +
Face Expression Recognition +
944faf7f14f1bead911aeec30cc80c861442b610Action Tubelet Detector for Spatio-Temporal Action Localization +
9458c518a6e2d40fb1d6ca1066d6a0c73e1d6b735967 +
A Benchmark and Comparative Study of +
Video-Based Face Recognition +
on COX Face Database +
94aa8a3787385b13ee7c4fdd2b2b2a574ffcbd81
94325522c9be8224970f810554611d6a73877c13
9441253b638373a0027a5b4324b4ee5f0dffd670A Novel Scheme for Generating Secure Face +
Templates Using BDA +
P.G. Student, Department of Computer Engineering, +
Associate Professor, Department of Computer +
MCERC, +
Nashik (M.S.), India +
94ac3008bf6be6be6b0f5140a0bea738d4c75579
94a11b601af77f0ad46338afd0fa4ccbab909e82
0e8760fc198a7e7c9f4193478c0e0700950a86cd
0e50fe28229fea45527000b876eb4068abd6ed8cProceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
2936 +
0eff410cd6a93d0e37048e236f62e209bc4383d1Anchorage Convention District +
May 3-8, 2010, Anchorage, Alaska, USA +
978-1-4244-5040-4/10/$26.00 ©2010 IEEE +
4803 +
0ee737085af468f264f57f052ea9b9b1f58d7222SiGAN: Siamese Generative Adversarial Network +
for Identity-Preserving Face Hallucination +
0ee661a1b6bbfadb5a482ec643573de53a9adf5eJOURNAL OF LATEX CLASS FILES, VOL. X, NO. X, MONTH YEAR +
On the Use of Discriminative Cohort Score +
Normalization for Unconstrained Face Recognition +
0e3840ea3227851aaf4633133dd3cbf9bbe89e5b
0e5dad0fe99aed6978c6c6c95dc49c6dca601e6a
0e2ea7af369dbcaeb5e334b02dd9ba5271b10265
0e7c70321462694757511a1776f53d629a1b38f3NIST Special Publication 1136 +
2012 Proceedings of the +
Performance Metrics for Intelligent +
Systems (PerMI ‘12) Workshop +
+
http://dx.doi.org/10.6028/NIST.SP.1136 +
6080f26675e44f692dd722b61905af71c5260af8
60d765f2c0a1a674b68bee845f6c02741a49b44e
60c24e44fce158c217d25c1bae9f880a8bd19fc3Controllable Image-to-Video Translation: +
A Case Study on Facial Expression Generation +
MIT CSAIL +
Wenbing Huang +
Tencent AI Lab +
MIT-Waston Lab +
Tencent AI Lab +
Tencent AI Lab +
60e2b9b2e0db3089237d0208f57b22a3aac932c1Frankenstein: Learning Deep Face Representations +
using Small Data +
60ce4a9602c27ad17a1366165033fe5e0cf68078TECHNICAL NOTE +
DIGITAL & MULTIMEDIA SCIENCES +
J Forensic Sci, 2015 +
doi: 10.1111/1556-4029.12800 +
Available online at: onlinelibrary.wiley.com +
Ph.D. +
Combination of Face Regions in Forensic +
Scenarios* +
6097ea6fd21a5f86a10a52e6e4dd5b78a436d5bf
60efdb2e204b2be6701a8e168983fa666feac1beInt J Comput Vis +
DOI 10.1007/s11263-017-1043-5 +
Transferring Deep Object and Scene Representations for Event +
Recognition in Still Images +
Received: 31 March 2016 / Accepted: 1 September 2017 +
© Springer Science+Business Media, LLC 2017 +
60824ee635777b4ee30fcc2485ef1e103b8e7af9Cascaded Collaborative Regression for Robust Facial +
Landmark Detection Trained using a Mixture of Synthetic and +
Real Images with Dynamic Weighting +
Life Member, IEEE, William Christmas, and Xiao-Jun Wu +
60643bdab1c6261576e6610ea64ea0c0b200a28d
60a20d5023f2bcc241eb9e187b4ddece695c2b9bInvertible Nonlinear Dimensionality Reduction +
via Joint Dictionary Learning +
Department of Electrical and Computer Engineering +
Technische Universit¨at M¨unchen, Germany +
60cdcf75e97e88638ec973f468598ae7f75c59b486 +
Face Annotation Using Transductive +
Kernel Fisher Discriminant +
60b3601d70f5cdcfef9934b24bcb3cc4dde663e7SUBMITTED TO IEEE TRANS. ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +
Binary Gradient Correlation Patterns +
for Robust Face Recognition +
34a41ec648d082270697b9ee264f0baf4ffb5c8d
341002fac5ae6c193b78018a164d3c7295a495e4von Mises-Fisher Mixture Model-based Deep +
learning: Application to Face Verification +
34ec83c8ff214128e7a4a4763059eebac59268a6Action Anticipation By Predicting Future +
Dynamic Images +
Australian Centre for Robotic Vision, ANU, Canberra, Australia +
34b7e826db49a16773e8747bc8dfa48e344e425d
341ed69a6e5d7a89ff897c72c1456f50cfb23c96DAGER: Deep Age, Gender and Emotion +
Recognition Using Convolutional Neural +
Networks +
Computer Vision Lab, Sighthound Inc., Winter Park, FL +
340d1a9852747b03061e5358a8d12055136599b0Audio-Visual Recognition System Insusceptible +
to Illumination Variation over Internet Protocol +
+
5a3da29970d0c3c75ef4cb372b336fc8b10381d7CNN-based Real-time Dense Face Reconstruction +
with Inverse-rendered Photo-realistic Face Images +
5a34a9bb264a2594c02b5f46b038aa1ec3389072Label-Embedding for Image Classification +
5a4c6246758c522f68e75491eb65eafda375b701978-1-4244-4296-6/10/$25.00 ©2010 IEEE +
1118 +
ICASSP 2010 +
5aad5e7390211267f3511ffa75c69febe3b84cc7Driver Gaze Estimation +
Without Using Eye Movement +
MIT AgeLab +
5a029a0b0ae8ae7fc9043f0711b7c0d442bfd372
5a4ec5c79f3699ba037a5f06d8ad309fb4ee682cDownloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Electronic-Imaging on 12/17/2017 Terms of Use: https://www.spiedigitallibrary.org/terms-of-use +
AutomaticageandgenderclassificationusingsupervisedappearancemodelAliMainaBukarHassanUgailDavidConnahAliMainaBukar,HassanUgail,DavidConnah,“Automaticageandgenderclassificationusingsupervisedappearancemodel,”J.Electron.Imaging25(6),061605(2016),doi:10.1117/1.JEI.25.6.061605.
5a7520380d9960ff3b4f5f0fe526a00f63791e99The Indian Spontaneous Expression +
Database for Emotion Recognition +
5fff61302adc65d554d5db3722b8a604e62a8377Additive Margin Softmax for Face Verification +
UESTC +
Georgia Tech +
UESTC +
UESTC +
5fa6e4a23da0b39e4b35ac73a15d55cee8608736IJCV special issue (Best papers of ECCV 2016) manuscript No. +
(will be inserted by the editor) +
RED-Net: +
A Recurrent Encoder-Decoder Network for Video-based Face Alignment +
Submitted: April 19 2017 / Revised: December 12 2017 +
5f871838710a6b408cf647aacb3b198983719c311716 +
Locally Linear Regression for Pose-Invariant +
Face Recognition +
5f64a2a9b6b3d410dd60dc2af4a58a428c5d85f9
5f344a4ef7edfd87c5c4bc531833774c3ed23542c Copyright by Ira Cohen, 2003 +
5fa0e6da81acece7026ac1bc6dcdbd8b204a5f0a
5f27ed82c52339124aa368507d66b71d96862cb7Semi-supervised Learning of Classifiers: Theory, Algorithms +
and Their Application to Human-Computer Interaction +
This work has been partially funded by NSF Grant IIS 00-85980. +
DRAFT +
5fa932be4d30cad13ea3f3e863572372b915bec8
5f5906168235613c81ad2129e2431a0e5ef2b6e4Noname manuscript No. +
(will be inserted by the editor) +
A Unified Framework for Compositional Fitting of +
Active Appearance Models +
Received: date / Accepted: date +
5fc664202208aaf01c9b62da5dfdcd71fdadab29arXiv:1504.05308v1 [cs.CV] 21 Apr 2015 +
5fa1724a79a9f7090c54925f6ac52f1697d6b570Proceedings of the Workshop on Grammar and Lexicon: Interactions and Interfaces, +
pages 41–47, Osaka, Japan, December 11 2016. +
41 +
33aff42530c2fd134553d397bf572c048db12c28From Emotions to Action Units with Hidden and Semi-Hidden-Task Learning +
Universitat Pompeu Fabra +
Centre de Visio per Computador +
Universitat Pompeu Fabra +
Barcelona +
Barcelona +
Barcelona +
33a1a049d15e22befc7ddefdd3ae719ced8394bfFULL PAPER +
International Journal of Recent Trends in Engineering, Vol 2, No. 1, November 2009 +
An Efficient Approach to Facial Feature Detection +
for Expression Recognition +
S.P. Khandait1, P.D. Khandait2 and Dr.R.C.Thool2 +
1Deptt. of Info.Tech., K.D.K.C.E., Nagpur, India +
2Deptt.of Electronics Engg., K.D.K.C.E., Nagpur, India, 2Deptt. of Info.Tech., SGGSIET, Nanded +
3399f8f0dff8fcf001b711174d29c9d4fde89379Face R-CNN +
Tencent AI Lab, China +
333aa36e80f1a7fa29cf069d81d4d2e12679bc67Suggesting Sounds for Images +
from Video Collections +
1Computer Science Department, ETH Z¨urich, Switzerland +
2Disney Research, Switzerland +
33792bb27ef392973e951ca5a5a3be4a22a0d0c6Two-dimensional Whitening Reconstruction for +
Enhancing Robustness of Principal Component +
Analysis +
3328674d71a18ed649e828963a0edb54348ee598IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 34, NO. 6, DECEMBER 2004 +
2405 +
A Face and Palmprint Recognition Approach Based +
on Discriminant DCT Feature Extraction +
339937141ffb547af8e746718fbf2365cc1570c8Facial Emotion Recognition in Real Time +
33aa980544a9d627f305540059828597354b076c
33ae696546eed070717192d393f75a1583cd8e2c
3352426a67eabe3516812cb66a77aeb8b4df4d1bJOURNAL OF LATEX CLASS FILES, VOL. 4, NO. 5, APRIL 2015 +
Joint Multi-view Face Alignment in the Wild +
334d6c71b6bce8dfbd376c4203004bd4464c2099BICONVEX RELAXATION FOR SEMIDEFINITE PROGRAMMING IN +
COMPUTER VISION +
33e20449aa40488c6d4b430a48edf5c4b43afdabTRANSACTIONS ON AFFECTIVE COMPUTING +
The Faces of Engagement: Automatic +
Recognition of Student Engagement from Facial +
Expressions +
333e7ad7f915d8ee3bb43a93ea167d6026aa3c22This 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. +
The final version of record is available at http://dx.doi.org/10.1109/TIFS.2014.2309851 +
DRAFT +
3D Assisted Face Recognition: Dealing With +
Expression Variations +
+
33403e9b4bbd913ae9adafc6751b52debbd45b0e
33ad23377eaead8955ed1c2b087a5e536fecf44eAugmenting CRFs with Boltzmann Machine Shape Priors for Image Labeling +
∗ indicates equal contribution +
05b8673d810fadf888c62b7e6c7185355ffa4121(will be inserted by the editor) +
A Comprehensive Survey to Face Hallucination +
Received: date / Accepted: date +
05e658fed4a1ce877199a4ce1a8f8cf6f449a890
05ad478ca69b935c1bba755ac1a2a90be6679129Attribute Dominance: What Pops Out? +
Georgia Tech +
0562fc7eca23d47096472a1d42f5d4d086e21871
054738ce39920975b8dcc97e01b3b6cc0d0bdf32Towards the Design of an End-to-End Automated +
System for Image and Video-based Recognition +
05e03c48f32bd89c8a15ba82891f40f1cfdc7562Scalable Robust Principal Component +
Analysis using Grassmann Averages +
050fdbd2e1aa8b1a09ed42b2e5cc24d4fe8c7371Contents +
Scale Space and PDE Methods +
Spatio-Temporal Scale Selection in Video Data . . . . . . . . . . . . . . . . . . . . . +
Dynamic Texture Recognition Using Time-Causal Spatio-Temporal +
Scale-Space Filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +
Corner Detection Using the Affine Morphological Scale Space . . . . . . . . . . . +
Luis Alvarez +
Nonlinear Spectral Image Fusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +
Martin Benning, Michael Möller, Raz Z. Nossek, Martin Burger, +
Daniel Cremers, Guy Gilboa, and Carola-Bibiane Schönlieb +
16 +
29 +
41 +
Tubular Structure Segmentation Based on Heat Diffusion. . . . . . . . . . . . . . . +
54 +
Fang Yang and Laurent D. Cohen +
Analytic Existence and Uniqueness Results for PDE-Based Image +
Reconstruction with the Laplacian . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +
Laurent Hoeltgen, Isaac Harris, Michael Breuß, and Andreas Kleefeld +
Combining Contrast Invariant L1 Data Fidelities with Nonlinear +
Spectral Image Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +
Leonie Zeune, Stephan A. van Gils, Leon W.M.M. Terstappen, +
and Christoph Brune +
An Efficient and Stable Two-Pixel Scheme for 2D +
Forward-and-Backward Diffusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +
Martin Welk and Joachim Weickert +
66 +
80 +
94 +
Restoration and Reconstruction +
Blind Space-Variant Single-Image Restoration of Defocus Blur. . . . . . . . . . . +
109 +
Leah Bar, Nir Sochen, and Nahum Kiryati +
Denoising by Inpainting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +
121 +
Robin Dirk Adam, Pascal Peter, and Joachim Weickert +
Stochastic Image Reconstruction from Local Histograms +
of Gradient Orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . +
Agnès Desolneux and Arthur Leclaire +
133 +
056294ff40584cdce81702b948f88cebd731a93e
052880031be0a760a5b606b2ad3d22f237e8af70Datasets on object manipulation and interaction: a survey +
05ea7930ae26165e7e51ff11b91c7aa8d7722002Learning And-Or Model to Represent Context and +
Occlusion for Car Detection and Viewpoint Estimation +
051a84f0e39126c1ebeeb379a405816d5d06604dCogn Comput (2009) 1:257–267 +
DOI 10.1007/s12559-009-9018-7 +
Biometric Recognition Performing in a Bioinspired System +
Joan Fa`bregas Æ Marcos Faundez-Zanuy +
Published online: 20 May 2009 +
Ó Springer Science+Business Media, LLC 2009 +
05f4d907ee2102d4c63a3dc337db7244c570d067
05a7be10fa9af8fb33ae2b5b72d108415519a698Multilayer and Multimodal Fusion of Deep Neural Networks +
for Video Classification +
NVIDIA +
050a149051a5d268fcc5539e8b654c2240070c82MAGISTERSKÉ A DOKTORSKÉSTUDIJNÍ PROGRAMY31. 5. 2018SBORNÍKSTUDENTSKÁ VĚDECKÁ KONFERENCE
0580edbd7865414c62a36da9504d1169dea78d6fBaseline CNN structure analysis for facial expression recognition +
05e96d76ed4a044d8e54ef44dac004f796572f1a
9d839dfc9b6a274e7c193039dfa7166d3c07040bAugmented Faces +
1ETH Z¨urich +
2Kooaba AG +
3K.U. Leuven +
9d60ad72bde7b62be3be0c30c09b7d03f9710c5fA Survey: Face Recognition Techniques +
Assistant Professor, ITM GOI +
M Tech, ITM GOI +
face +
video +
(Eigen +
passport-verification, +
9cfb3a68fb10a59ec2a6de1b24799bf9154a8fd1
9ca7899338129f4ba6744f801e722d53a44e4622Deep Neural Networks Regularization for Structured +
Output Prediction +
Soufiane Belharbi∗ +
INSA Rouen, LITIS +
76000 Rouen, France +
INSA Rouen, LITIS +
76000 Rouen, France +
INSA Rouen, LITIS +
76000 Rouen, France +
INSA Rouen, LITIS +
76000 Rouen, France +
Normandie Univ, UNIROUEN, UNIHAVRE, +
Normandie Univ, UNIROUEN, UNIHAVRE, +
Normandie Univ, UNIROUEN, UNIHAVRE, +
Normandie Univ, UNIROUEN, UNIHAVRE, +
9c1664f69d0d832e05759e8f2f001774fad354d6Action representations in robotics: A +
taxonomy and systematic classification +
Journal Title +
XX(X):1–32 +
c(cid:13)The Author(s) 2016 +
Reprints and permission: +
sagepub.co.uk/journalsPermissions.nav +
DOI: 10.1177/ToBeAssigned +
www.sagepub.com/ +
9c065dfb26ce280610a492c887b7f6beccf27319Learning from Video and Text via Large-Scale Discriminative Clustering +
1 ´Ecole Normale Sup´erieure +
2Inria +
3CIIRC +
02601d184d79742c7cd0c0ed80e846d95def052eGraphical Representation for Heterogeneous +
Face Recognition +
02cc96ad997102b7c55e177ac876db3b91b4e72cMuseumVisitors: a dataset for pedestrian and group detection, gaze estimation +
and behavior understanding +
02fda07735bdf84554c193811ba4267c24fe2e4aIllumination Invariant Face Recognition +
Using Near-Infrared Images +
02dd0af998c3473d85bdd1f77254ebd71e6158c6PPP: Joint Pointwise and Pairwise Image Label Prediction +
1Department of Computer Science, Arizona State Univerity +
2Yahoo Research +
029317f260b3303c20dd58e8404a665c7c5e73391276 +
Character Identification in Feature-Length Films +
Using Global Face-Name Matching +
and Yeh-Min Huang, Member, IEEE +
0273414ba7d56ab9ff894959b9d46e4b2fef7fd0Photographic home styles in Congress: a +
computer vision approach∗ +
December 1, 2016 +
02e133aacde6d0977bca01ffe971c79097097b7f
02567fd428a675ca91a0c6786f47f3e35881bcbdACCEPTED BY IEEE TIP +
Deep Label Distribution Learning +
With Label Ambiguity +
029b53f32079063047097fa59cfc788b2b550c4b
02bd665196bd50c4ecf05d6852a4b9ba027cd9d0
026b5b8062e5a8d86c541cfa976f8eee97b30ab8MDLFace: Memorability Augmented Deep Learning for Video Face Recognition +
IIIT-Delhi, India +
0278acdc8632f463232e961563e177aa8c6d6833Selective Transfer Machine for Personalized +
Facial Expression Analysis +
1 INTRODUCTION +
Index Terms—Facial expression analysis, personalization, domain adaptation, transfer learning, support vector machine (SVM) +
A UTOMATIC facial AU detection confronts a number of +
02c993d361dddba9737d79e7251feca026288c9c
a46283e90bcdc0ee35c680411942c90df130f448
a4a5ad6f1cc489427ac1021da7d7b70fa9a770f2Yudistira and Kurita EURASIP Journal on Image and Video +
Processing (2017) 2017:85 +
DOI 10.1186/s13640-017-0235-9 +
EURASIP Journal on Image +
and Video Processing +
RESEARCH +
Open Access +
Gated spatio and temporal convolutional +
neural network for activity recognition: +
towards gated multimodal deep learning +
a40f8881a36bc01f3ae356b3e57eac84e989eef0End-to-end semantic face segmentation with conditional +
random fields as convolutional, recurrent and adversarial +
networks +
a44590528b18059b00d24ece4670668e86378a79Learning the Hierarchical Parts of Objects by Deep +
Non-Smooth Nonnegative Matrix Factorization +
a4c430b7d849a8f23713dc283794d8c1782198b2Video Concept Embedding +
1. Introduction +
In the area of natural language processing, there has been +
much success in learning distributed representations for +
words as vectors. Doing so has an advantage over using +
simple labels, or a one-hot coding scheme for representing +
individual words. In learning distributed vector representa- +
tions for words, we manage to capture semantic relatedness +
of words in vector distance. For example, the word vector +
for ”car” and ”road” should end up being closer together in +
the vector space representation than ”car” and ”penguin”. +
This has been very useful in NLP areas of machine transla- +
tion and semantic understanding. +
In the computer vision domain, video understanding is a +
very important topic. +
It is made hard due to the large +
amount of high dimensional data in videos. One strategy +
to address this is to summarize a video into concepts (eg. +
running, climbing, cooking). This allows us to represent a +
video in a very natural way to humans, such as a sequence +
of semantic events. However this has the same shortcom- +
ings that one-hot coding of words have. +
The goal of this project is to find a meaningful way to em- +
bed video concepts into a vector space. The hope would +
be to capture semantic relatedness of concepts in a vector +
representation, essentially doing for videos what word2vec +
did for text. Having a vector representation for video con- +
cepts would help in areas such as semantic video retrieval +
and video classification, as it would provide a statistically +
meaningful and robust way of representing videos as lower +
dimensional vectors. An interesting thing would be to ob- +
serve if such a vector representation would result in ana- +
logical reasoning using simple vector arithmetic. +
Figure 1 shows an example of concepts detected at differ- +
ent snapshots in the same video. For example, consider +
the scenario where the concepts Kicking a ball, Soccer and +
Running are detected in the three snapshots respectively +
(from left to right). Since, these snapshots belong in the +
same video, we expect that these concepts are semantically +
similar and that they should lie close in the resulting em- +
bedding space. The aim of this project is to find a vector +
space embedding for the space of concepts such that vector +
representations for semantically similar concepts (in this +
Figure 1. Example snapshots from the same video +
case, Running, Kicking and Soccer) lie in the vicinity of +
each other. +
2. Related Work +
(Mikolov et al., 2013a) introduces the popular skip-gram +
model to learn distributed representations of words from +
very large linguistic datasets. Specifically, it uses each +
word as an input to a log-linear classifier and predict words +
within a certain range before and after the current word in +
the dataset. +
(Mikolov et al., 2013b) extends this model +
to learn representations for phrases, in addition to words, +
and also improve the quality of vectors and training speed. +
These works also show that the skip-gram model exhibits +
a linear structure that enables it to perform reasoning using +
basic vector arithmetic. The skip-gram model from these +
works is the basis of our model in learning representations +
for concepts. +
(Le & Mikolov, 2014) extends the concept of word vectors +
to sentences and paragraphs. Their approach is more in- +
volved than a simple bag of words approach, in that it tries +
to capture the nature of the words in the paragraph. They +
construct the paragraph vector in such a way that it can be +
used to predict the word vectors that are contained inside +
the paragraph. They do this by first learning word vectors, +
such that the probability of a word vector given its context +
is maximized. To learn paragraph vectors, the paragraph +
is essentially treated as a word, and the words it contains +
become the context. This provides a key insight in how +
a set of concept vectors can be used together to provide a +
more meaningful vector representation for videos, which +
can then be used for retrieval. +
(Hu et al.) utilizes structured knowledge in the data to learn +
distributed representations that improve semantic related- +
a4cc626da29ac48f9b4ed6ceb63081f6a4b304a2
a4f37cfdde3af723336205b361aefc9eca688f5cRecent Advances +
in Face Recognition +
a30869c5d4052ed1da8675128651e17f97b87918Fine-Grained Comparisons with Attributes +
a3ebacd8bcbc7ddbd5753935496e22a0f74dcf7bFirst International Workshop on Adaptive Shot Learning +
for Gesture Understanding and Production +
ASL4GUP 2017 +
Held in conjunction with IEEE FG 2017, in May 30, 2017, +
Washington DC, USA +
a3d8b5622c4b9af1f753aade57e4774730787a00Pose-Aware Person Recognition +
Anoop Namboodiri (cid:63) +
(cid:63) CVIT, IIIT Hyderabad, India +
† Facebook AI Research +
a3017bb14a507abcf8446b56243cfddd6cdb542bFace Localization and Recognition in Varied +
Expressions and Illumination +
Hui-Yu Huang, Shih-Hang Hsu +
+
a378fc39128107815a9a68b0b07cffaa1ed32d1fDetermining a Suitable Metric When using Non-negative Matrix Factorization∗ +
Computer Vision Center, Dept. Inform`atica +
Universitat Aut`onoma de Barcelona +
08193 Bellaterra, Barcelona, Spain +
a34d75da87525d1192bda240b7675349ee85c123Naive-Deep Face Recognition: Touching the Limit of LFW Benchmark or Not? +
Face++, Megvii Inc. +
Face++, Megvii Inc. +
Face++, Megvii Inc. +
a3f69a073dcfb6da8038607a9f14eb28b5dab2dbProceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18) +
1184 +
a3f78cc944ac189632f25925ba807a0e0678c4d5Action Recognition in Realistic Sports Videos +
a33f20773b46283ea72412f9b4473a8f8ad751ae
a3a6a6a2eb1d32b4dead9e702824375ee76e3ce7Multiple Local Curvature Gabor Binary +
Patterns for Facial Action Recognition +
Signal Processing Laboratory (LTS5), +
´Ecole Polytechnique F´ed´erale de Lausanne, Switzerland +
a32c5138c6a0b3d3aff69bcab1015d8b043c91fbDownloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Electronic-Imaging on 9/19/2018 +
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use +
Videoredaction:asurveyandcomparisonofenablingtechnologiesShaganSahAmeyaShringiRaymondPtuchaAaronBurryRobertLoceShaganSah,AmeyaShringi,RaymondPtucha,AaronBurry,RobertLoce,“Videoredaction:asurveyandcomparisonofenablingtechnologies,”J.Electron.Imaging26(5),051406(2017),doi:10.1117/1.JEI.26.5.051406.
a3d78bc94d99fdec9f44a7aa40c175d5a106f0b9Recognizing Violence in Movies +
CIS400/401 Project Final Report +
Univ. of Pennsylvania +
Philadelphia, PA +
Univ. of Pennsylvania +
Philadelphia, PA +
Ben Sapp +
Univ. of Pennsylvania +
Philadelphia, PA +
Univ. of Pennsylvania +
Philadelphia, PA +
a3eab933e1b3db1a7377a119573ff38e780ea6a3978-1-4244-4296-6/10/$25.00 ©2010 IEEE +
838 +
ICASSP 2010 +
a35d3ba191137224576f312353e1e0267e6699a1Increasing security in DRM systems +
through biometric authentication. +
ecuring the exchange +
of intellectual property +
and providing protection +
to multimedia contents in +
distribution systems have enabled the +
advent of digital rights management +
(DRM) systems [5], [14], [21], [47], +
[51], [53]. Rights holders should be able to +
license, monitor, and track the usage of rights +
in a dynamic digital trading environment, espe- +
cially in the near future when universal multimedia +
access (UMA) becomes a reality, and any multimedia +
content will be available anytime, anywhere. In such +
DRM systems, encryption algorithms, access control, +
key management strategies, identification and tracing +
of contents, or copy control will play a prominent role +
to supervise and restrict access to multimedia data, +
avoiding unauthorized or fraudulent operations. +
A key component of any DRM system, also known +
as intellectual property management and protection +
(IPMP) systems in the MPEG-21 framework, is user +
authentication to ensure that +
only those with specific rights are +
able to access the digital informa- +
tion. It is here that biometrics can +
play an essential role, reinforcing securi- +
ty at all stages where customer authentica- +
tion is needed. The ubiquity of users and +
devices, where the same user might want to +
access to multimedia contents from different +
environments (home, car, work, jogging, etc.) and +
also from different devices or media (CD, DVD, +
home computer, laptop, PDA, 2G/3G mobile phones, +
game consoles, etc.) strengthens the need for reliable +
and universal authentication of users. +
Classical user authentication systems have been +
based in something that you have (like a key, an identi- +
fication card, etc.) and/or something that you know +
(like a password, or a PIN). With biometrics, a new +
user authentication paradigm is added: something that +
you are (e.g., fingerprints or face) or something that +
you do or produce (e.g., handwritten signature or +
50 +
IEEE SIGNAL PROCESSING MAGAZINE +
1053-5888/04/$20.00©2004IEEE +
MARCH 2004 +
b558be7e182809f5404ea0fcf8a1d1d9498dc01aBottom-up and top-down reasoning with convolutional latent-variable models +
UC Irvine +
UC Irvine +
b5fc4f9ad751c3784eaf740880a1db14843a85baSIViP (2007) 1:225–237 +
DOI 10.1007/s11760-007-0016-5 +
ORIGINAL PAPER +
Significance of image representation for face verification +
Received: 29 August 2006 / Revised: 28 March 2007 / Accepted: 28 March 2007 / Published online: 1 May 2007 +
© Springer-Verlag London Limited 2007 +
b562def2624f59f7d3824e43ecffc990ad780898
b5160e95192340c848370f5092602cad8a4050cdIEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, TO APPEAR +
Video Classification With CNNs: Using The Codec +
As A Spatio-Temporal Activity Sensor +
b52c0faba5e1dc578a3c32a7f5cfb6fb87be06adJournal of Applied Research and +
Technology +
ISSN: 1665-6423 +
Centro de Ciencias Aplicadas y +
Desarrollo Tecnológico +
México +
+
Hussain Shah, Jamal; Sharif, Muhammad; Raza, Mudassar; Murtaza, Marryam; Ur-Rehman, Saeed +
Robust Face Recognition Technique under Varying Illumination +
Journal of Applied Research and Technology, vol. 13, núm. 1, febrero, 2015, pp. 97-105 +
Centro de Ciencias Aplicadas y Desarrollo Tecnológico +
Distrito Federal, México +
Available in: http://www.redalyc.org/articulo.oa?id=47436895009 +
How to cite +
Complete issue +
More information about this article +
Journal's homepage in redalyc.org +
Scientific Information System +
Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal +
Non-profit academic project, developed under the open access initiative +
b52886610eda6265a2c1aaf04ce209c047432b6dMicroexpression Identification and Categorization +
using a Facial Dynamics Map +
b5857b5bd6cb72508a166304f909ddc94afe53e3SSIG and IRISA at Multimodal Person Discovery +
1Department of Computer Science, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil +
2IRISA & Inria Rennes , CNRS, Rennes, France +
b59f441234d2d8f1765a20715e227376c7251cd7
b51e3d59d1bcbc023f39cec233f38510819a2cf9CBMM Memo No. 003 +
March 27, 2014 +
Can a biologically-plausible hierarchy effectively +
replace face detection, alignment, and +
recognition pipelines? +
by +
b54c477885d53a27039c81f028e710ca54c83f111201 +
Semi-Supervised Kernel Mean Shift Clustering +
b2a0e5873c1a8f9a53a199eecae4bdf505816ecbHybrid VAE: Improving Deep Generative Models +
using Partial Observations +
Snap Research +
Microsoft Research +
b2b535118c5c4dfcc96f547274cdc05dde629976JOURNAL OF IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, VOL. XX, NO. X, XXX 2017 +
Automatic Recognition of Facial Displays of +
Unfelt Emotions +
Escalera, Xavier Bar´o, Sylwia Hyniewska, Member, IEEE, J¨uri Allik, +
b235b4ccd01a204b95f7408bed7a10e080623d2eRegularizing Flat Latent Variables with Hierarchical Structures +
b2c25af8a8e191c000f6a55d5f85cf60794c2709Noname manuscript No. +
(will be inserted by the editor) +
A Novel Dimensionality Reduction Technique based on +
Kernel Optimization Through Graph Embedding +
N. Vretos, A. Tefas and I. Pitas +
the date of receipt and acceptance should be inserted later +
d904f945c1506e7b51b19c99c632ef13f340ef4cA scalable 3D HOG model for fast object detection and viewpoint estimation +
KU Leuven, ESAT/PSI - iMinds +
Kasteelpark Arenberg 10 B-3001 Leuven, Belgium +
d94d7ff6f46ad5cab5c20e6ac14c1de333711a0c978-1-5090-4117-6/17/$31.00 ©2017 IEEE +
3031 +
ICASSP 2017 +
d9739d1b4478b0bf379fe755b3ce5abd8c668f89
d9318c7259e394b3060b424eb6feca0f71219179406 +
Face Matching and Retrieval Using Soft Biometrics +
d9a1dd762383213741de4c1c1fd9fccf44e6480d
d9c4b1ca997583047a8721b7dfd9f0ea2efdc42cLearning Inference Models for Computer Vision +
aca232de87c4c61537c730ee59a8f7ebf5ecb14fEBGM VS SUBSPACE PROJECTION FOR FACE RECOGNITION +
19.5 Km Markopoulou Avenue, P.O. Box 68, Peania, Athens, Greece +
Athens Information Technology +
Keywords: +
Human-Machine Interfaces, Computer Vision, Face Recognition. +
ac6a9f80d850b544a2cbfdde7002ad5e25c05ac6779 +
Privacy-Protected Facial Biometric Verification +
Using Fuzzy Forest Learning +
aca273a9350b10b6e2ef84f0e3a327255207d0f5
ac0d3f6ed5c42b7fc6d7c9e1a9bb80392742ad5e
ac820d67b313c38b9add05abef8891426edd5afb
ac26166857e55fd5c64ae7194a169ff4e473eb8bPersonalized Age Progression with Bi-level +
Aging Dictionary Learning +
ac8441e30833a8e2a96a57c5e6fede5df81794afIEEE TRANSACTIONS ON IMAGE PROCESSING +
Hierarchical Representation Learning for Kinship +
Verification +
acb83d68345fe9a6eb9840c6e1ff0e41fa373229Kernel Methods in Computer Vision: +
Object Localization, Clustering, +
and Taxonomy Discovery +
vorgelegt von +
Matthew Brian Blaschko, M.S. +
aus La Jolla +
Von der Fakult¨at IV - Elektrotechnik und Informatik +
der Technischen Universit¨at Berlin +
zur Erlangung des akademischen Grades +
Doktor der Naturwissenschaften +
Dr. rer. nat. +
genehmigte Dissertation +
Promotionsausschuß: +
Vorsitzender: Prof. Dr. O. Hellwich +
Berichter: Prof. Dr. T. Hofmann +
Berichter: Prof. Dr. K.-R. M¨uller +
Berichter: Prof. Dr. B. Sch¨olkopf +
Tag der wissenschaftlichen Aussprache: 23.03.2009 +
Berlin 2009 +
D83 +
adf7ccb81b8515a2d05fd3b4c7ce5adf5377d9beApprentissage de métrique appliqué à la +
détection de changement de page Web et +
aux attributs relatifs +
thieu Cord* +
* Sorbonne Universités, UPMC Univ Paris 06, UMR 7606, LIP6, F-75005, Paris, +
France +
RÉSUMÉ. Nous proposons dans cet article un nouveau schéma d’apprentissage de métrique. +
Basé sur l’exploitation de contraintes qui impliquent des quadruplets d’images, notre approche +
vise à modéliser des relations sémantiques de similarités riches ou complexes. Nous étudions +
comment ce schéma peut être utilisé dans des contextes tels que la détection de régions impor- +
tantes dans des pages Web ou la reconnaissance à partir d’attributs relatifs. +
ada73060c0813d957576be471756fa7190d1e72dVRPBench: A Vehicle Routing Benchmark Tool +
October 19, 2016 +
adaf2b138094981edd615dbfc4b7787693dbc396Statistical Methods For Facial +
Shape-from-shading and Recognition +
Submitted for the degree of Doctor of Philosophy +
Department of Computer Science +
20th February 2007 +
ad6745dd793073f81abd1f3246ba4102046da022
adf62dfa00748381ac21634ae97710bb80fc2922ViFaI: A trained video face indexing scheme +
1. Introduction +
With the increasing prominence of inexpensive +
video recording devices (e.g., digital camcorders and +
video recording smartphones), +
the average user’s +
video collection today is increasing rapidly. With this +
development, there arises a natural desire to rapidly +
access a subset of one’s collection of videos. The solu- +
tion to this problem requires an effective video index- +
ing scheme. In particular, we must be able to easily +
process a video to extract such indexes. +
Today, there also exist large sets of labeled (tagged) +
face images. One important example is an individual’s +
Facebook profile. Such a set of of tagged images of +
one’s self, family, friends, and colleagues represents +
an extremely valuable potential training set. +
In this work, we explore how to leverage the afore- +
mentioned training set to solve the video indexing +
problem. +
2. Problem Statement +
Use a labeled (tagged) training set of face images +
to extract relevant indexes from a collection of videos, +
and use these indexes to answer boolean queries of the +
form: “videos with ‘Person 1’ OP1 ‘Person 2’ OP2 ... +
OP(N-1) ‘Person N’ ”, where ‘Person N’ corresponds +
to a training label (tag) and OPN is a boolean operand +
such as AND, OR, NOT, XOR, and so on. +
3. Proposed Scheme +
In this section, we outline our proposed scheme to +
address the problem we postulate in the previous sec- +
tion. We provide further details about the system im- +
plementation in Section 4. +
At a high level, we subdivide the problem into two +
key phases: the first ”off-line” executed once, and the +
second ”on-line” phase instantiated upon each query. +
For the purposes of this work, we define an index as +
follows:
bba281fe9c309afe4e5cc7d61d7cff1413b29558Social Cognitive and Affective Neuroscience, 2017, 984–992 +
doi: 10.1093/scan/nsx030 +
Advance Access Publication Date: 11 April 2017 +
Original article +
An unpleasant emotional state reduces working +
memory capacity: electrophysiological evidence +
1Laboratorio de Neurofisiologia do Comportamento, Departamento de Fisiologia e Farmacologia, Instituto +
Biome´dico, Universidade Federal Fluminense, Niteroi, Brazil, 2MograbiLab, Departamento de Psicologia, +
Pontifıcia Universidade Catolica do Rio de Janeiro, Rio de Janeiro, Brazil, and 3Laboratorio de Engenharia +
Pulmonar, Programa de Engenharia Biome´dica, COPPE, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil +
bb557f4af797cae9205d5c159f1e2fdfe2d8b096
bb06ef67a49849c169781657be0bb717587990e0Impact of Temporal Subsampling on Accuracy and +
Performance in Practical Video Classification +
F. Scheidegger∗†, L. Cavigelli∗, M. Schaffner∗, A. C. I. Malossi†, C. Bekas†, L. Benini∗‡ +
∗ETH Zürich, 8092 Zürich, Switzerland +
†IBM Research - Zürich, 8803 Rüschlikon, Switzerland +
‡Università di Bologna, Italy +
bb22104d2128e323051fb58a6fe1b3d24a9e9a46IAJ=JE BH ==OIEI 1 AIIA?A ?= EBH=JE =EO B?KIAI  JDA IK>JA +
ABBA?JELAAII B KH =CHEJD +
==OIEI 7IK=O = B=?E= ANFHAIIE ==OIEI IOIJA ?J=EI JDHAA IJ=CAI B=?A =?GKE +
9DAJDAH KIEC *=OAIE= ?=IIEAH " & IKFFHJ LA?JH =?DEA 58  H AKH= +
HACEI E = IECA ?=IIEAH EI = ? IJH=JACO & 0MALAH J = ?= HACEI +
bbe1332b4d83986542f5db359aee1fd9b9ba9967
bb7f2c5d84797742f1d819ea34d1f4b4f8d7c197TO APPEAR IN TPAMI +
From Images to 3D Shape Attributes +
bbf01aa347982592b3e4c9e4f433e05d30e71305
bbf1396eb826b3826c5a800975047beabde2f0de
bbd1eb87c0686fddb838421050007e934b2d74ab
d73d2c9a6cef79052f9236e825058d5d9cdc13212014-ENST-0040 +
EDITE - ED 130 +
Doctorat ParisTech +
T H È S E +
pour obtenir le grade de docteur délivré par +
TELECOM ParisTech +
Spécialité « Signal et Images » +
présentée et soutenue publiquement par +
le 08 juillet 2014 +
Cutting the Visual World into Bigger Slices for Improved Video +
Concept Detection +
Amélioration de la détection des concepts dans les vidéos par de plus grandes tranches du Monde +
Visuel +
Directeur de thèse : Bernard Mérialdo +
Jury +
M. Philippe-Henri Gosselin, Professeur, INRIA +
M. Georges Quénot, Directeur de recherche CNRS, LIG +
M. Georges Linares, Professeur, LIA +
M. François Brémond, Professeur, INRIA +
M. Bernard Mérialdo, Professeur, EURECOM +
Rapporteur +
Rapporteur +
Examinateur +
Examinateur +
Encadrant +
TELECOM ParisTech +
école de l’Institut Télécom - membre de ParisTech +
d78077a7aa8a302d4a6a09fb9737ab489ae169a6
d7312149a6b773d1d97c0c2b847609c07b5255ec
d708ce7103a992634b1b4e87612815f03ba3ab24FCVID: Fudan-Columbia Video Dataset +
Available at: http://bigvid.fudan.edu.cn/FCVID/ +
1 OVERVIEW +
Recognizing visual contents in unconstrained videos +
has become a very important problem for many ap- +
plications, such as Web video search and recommen- +
dation, smart content-aware advertising, robotics, etc. +
Existing datasets for video content recognition are +
either small or do not have reliable manual labels. +
In this work, we construct and release a new Inter- +
net video dataset called Fudan-Columbia Video Dataset +
(FCVID), containing 91,223 Web videos (total duration +
4,232 hours) annotated manually according to 239 +
categories. We believe that the release of FCVID can +
stimulate innovative research on this challenging and +
important problem. +
2 COLLECTION AND ANNOTATION +
The categories in FCVID cover a wide range of topics +
like social events (e.g., “tailgate party”), procedural +
events (e.g., “making cake”), objects (e.g., “panda”), +
scenes (e.g., “beach”), etc. These categories were de- +
fined very carefully. Specifically, we conducted user +
surveys and used the organization structures on +
YouTube and Vimeo as references, and browsed nu- +
merous videos to identify categories that satisfy the +
following three criteria: (1) utility — high relevance +
in supporting practical application needs; (2) cover- +
age — a good coverage of the contents that people +
record; and (3) feasibility — likely to be automatically +
recognized in the next several years, and a high +
frequency of occurrence that is sufficient for training +
a recognition algorithm. +
This definition effort led to a set of over 250 candi- +
date categories. For each category, in addition to the +
official name used in the public release, we manually +
defined another alternative name. Videos were then +
downloaded from YouTube searches using the official +
and the alternative names as search terms. The pur- +
pose of using the alternative names was to expand the +
candidate video sets. For each search, we downloaded +
1,000 videos, and after removing duplicate videos and +
some extremely long ones (longer than 30 minutes), +
there were around 1,000–1,500 candidate videos for +
each category. +
All the videos were annotated manually to ensure +
a high precision of the FCVID labels. In order to min- +
imize subjectivity, nearly 20 annotators were involved +
in the task, and a master annotator was assigned to +
monitor the entire process and double-check all the +
found positive videos. Some of the videos are multi- +
labeled, and thus filtering the 1,000–1,500 videos for +
each category with focus on just the single category +
label is not adequate. As checking the existence of all +
the 250+ classes for each video is extremely difficult, +
we use the following strategy to narrow down the “la- +
bel search space” for each video. We first grouped the +
categories according to subjective predictions of label +
co-occurrences, e.g., “wedding reception” & “wed- +
ding ceremony”, “waterfall” & “river”, “hiking” & +
“mountain”, and even “dog” & “birthday”. We then +
annotated the videos not only based on the target cat- +
egory label, but also according to the identified related +
labels. This helped produce a fairly complete label +
set for FCVID, but largely reduced the annotation +
workload. After removing the rare categories with +
less than 100 videos after annotation, the final FCVID +
dataset contains 91,223 videos and 239 categories, +
where 183 are events and 56 are objects, scenes, etc. +
Figure 1 shows the number of videos per category. +
“Dog” has the largest number of positive videos +
(1,136), while “making egg tarts” is the most infre- +
quent category containing only 108 samples. The total +
duration of FCVID is 4,232 hours with an average +
video duration of 167 seconds. Figure 2 further gives +
the average video duration of each category. +
The categories are organized using a hierarchy con- +
taining 11 high-level groups, as visualized in Figure 3. +
3 COMPARISON WITH RELATED DATASETS +
We compare FCVID with the following datasets. Most +
of them have been widely adopted in the existing +
works on video categorization. +
KTH and Weizmann: The KTH [1] and the Weiz- +
mann [2] datasets are well-known benchmarks for +
human action recognition. The former contains 600 +
videos of 6 human actions performed by 25 people +
in four scenarios, and the latter consists of 81 videos +
associated with 9 actions performed by 9 actors. +
Hollywood Human Action: The Hollywood +
dataset [3] contains 8 action classes collected from +
32 Hollywood movies with a total of 430 videos. +
d7b6bbb94ac20f5e75893f140ef7e207db7cd483Griffith Research Online +
https://research-repository.griffith.edu.au +
Face Recognition across Pose: A +
Review +
Author +
Zhang, Paul, Gao, Yongsheng +
Published +
2009 +
Journal Title +
Pattern Recognition +
DOI +
https://doi.org/10.1016/j.patcog.2009.04.017 +
Copyright Statement +
Copyright 2009 Elsevier. This is the author-manuscript version of this paper. Reproduced in accordance +
with the copyright policy of the publisher. Please refer to the journal's website for access to the +
definitive, published version. +
Downloaded from +
http://hdl.handle.net/10072/30193 +
d78373de773c2271a10b89466fe1858c3cab677f
d03265ea9200a993af857b473c6bf12a095ca178Multiple deep convolutional neural +
networks averaging for face +
alignment +
Zhouping Yin +
Downloaded From: http://electronicimaging.spiedigitallibrary.org/ on 05/28/2015 Terms of Use: http://spiedl.org/terms
d0eb3fd1b1750242f3bb39ce9ac27fc8cc7c5af0
d03baf17dff5177d07d94f05f5791779adf3cd5f
d0144d76b8b926d22411d388e7a26506519372ebImproving Regression Performance with Distributional Losses +
d02e27e724f9b9592901ac1f45830341d37140feDA-GAN: Instance-level Image Translation by Deep Attention Generative +
Adversarial Networks +
The State Universtiy of New York at Buffalo +
The State Universtiy of New York at Buffalo +
Microsoft Research +
Microsoft Research +
d0a21f94de312a0ff31657fd103d6b29db823caaFacial Expression Analysis +
d03e4e938bcbc25aa0feb83d8a0830f9cd3eb3eaFace Recognition with Patterns of Oriented +
Edge Magnitudes +
1 Vesalis Sarl, Clermont Ferrand, France +
2 Gipsa-lab, Grenoble INP, France +
d00787e215bd74d32d80a6c115c4789214da5edbFaster and Lighter Online +
Sparse Dictionary Learning +
Project report +
be8c517406528edc47c4ec0222e2a603950c2762Harrigan / The new handbook of methods in nonverbal behaviour research 02-harrigan-chap02 Page Proof page 7 +
17.6.2005 +
5:45pm +
B A S I C R E S E A RC H +
M E T H O D S A N D +
P RO C E D U R E S +
be48b5dcd10ab834cd68d5b2a24187180e2b408fFOR PERSONAL USE ONLY +
Constrained Low-rank Learning Using Least +
Squares Based Regularization +
be437b53a376085b01ebd0f4c7c6c9e40a4b1a75ISSN (Online) 2321 – 2004 +
ISSN (Print) 2321 – 5526 +
INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN ELECTRICAL, ELECTRONICS, INSTRUMENTATION AND CONTROL ENGINEERING +
Vol. 4, Issue 5, May 2016 +
IJIREEICE +
Face Recognition and Retrieval Using Cross +
Age Reference Coding +
BE, DSCE, Bangalore1 +
Assistant Professor, DSCE, Bangalore2 +
bebea83479a8e1988a7da32584e37bfc463d32d4Discovery of Latent 3D Keypoints via +
End-to-end Geometric Reasoning +
Google AI +
bef503cdfe38e7940141f70524ee8df4afd4f954
beab10d1bdb0c95b2f880a81a747f6dd17caa9c2DeepDeblur: Fast one-step blurry face images restoration +
Tsinghua Unversity +
b331ca23aed90394c05f06701f90afd550131fe3Zhou et al. EURASIP Journal on Image and Video Processing (2018) 2018:49 +
https://doi.org/10.1186/s13640-018-0287-5 +
EURASIP Journal on Image +
and Video Processing +
R ES EAR CH +
Double regularized matrix factorization for +
image classification and clustering +
Open Access +
b3cb91a08be4117d6efe57251061b62417867de9T. Swearingen and A. Ross. "A label propagation approach for predicting missing biographic labels in +
A Label Propagation Approach for +
Predicting Missing Biographic Labels +
in Face-Based Biometric Records +
b3c60b642a1c64699ed069e3740a0edeabf1922cMax-Margin Object Detection +
b3f7c772acc8bc42291e09f7a2b081024a172564 www.ijmer.com Vol. 3, Issue. 5, Sep - Oct. 2013 pp-3225-3230 ISSN: 2249-6645 +
International Journal of Modern Engineering Research (IJMER) +
A novel approach for performance parameter estimation of face +
recognition based on clustering, shape and corner detection +

+
b32631f456397462b3530757f3a73a2ccc362342Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
3069 +
b3afa234996f44852317af382b98f5f557cab25a
df90850f1c153bfab691b985bfe536a5544e438bFACE TRACKING ALGORITHM ROBUST TO POSE, +
ILLUMINATION AND FACE EXPRESSION CHANGES: A 3D +
PARAMETRIC MODEL APPROACH +

via Bramante 65 - 26013, Crema (CR), Italy +
Luigi Arnone, Fabrizio Beverina +
STMicroelectronics - Advanced System Technology Group +
via Olivetti 5 - 20041, Agrate Brianza, Italy +
Keywords: +
Face tracking, expression changes, FACS, illumination changes. +
df8da144a695269e159fb0120bf5355a558f4b02International Journal of Computer Applications (0975 – 8887) +
International Conference on Recent Trends in engineering & Technology - 2013(ICRTET'2013) +
Face Recognition using PCA and Eigen Face +
Approach +
ME EXTC [VLSI & Embedded System] +
Sinhgad Academy of Engineering +
EXTC Department +
Pune, India +
df577a89830be69c1bfb196e925df3055cafc0edShift: A Zero FLOP, Zero Parameter Alternative to Spatial Convolutions +
UC Berkeley +
dfabe7ef245ca68185f4fcc96a08602ee1afb3f7
df51dfe55912d30fc2f792561e9e0c2b43179089Face Hallucination using Linear Models of Coupled +
Sparse Support +
grid and fuse them to suppress the aliasing caused by under- +
sampling [5], [6]. On the other hand, learning based meth- +
ods use coupled dictionaries to learn the mapping relations +
between low- and high- resolution image pairs to synthesize +
high-resolution images from low-resolution images [4], [7]. +
The research community has lately focused on the latter +
category of super-resolution methods, since they can provide +
higher quality images and larger magnification factors. +
df80fed59ffdf751a20af317f265848fe6bfb9c91666 +
Learning Deep Sharable and Structural +
Detectors for Face Alignment +
dfa80e52b0489bc2585339ad3351626dee1a8395Human Action Forecasting by Learning Task Grammars +
dfecaedeaf618041a5498cd3f0942c15302e75c3Noname manuscript No. +
(will be inserted by the editor) +
A Recursive Framework for Expression Recognition: From +
Web Images to Deep Models to Game Dataset +
Received: date / Accepted: date +
df5fe0c195eea34ddc8d80efedb25f1b9034d07dRobust Modified Active Shape Model for Automatic Facial Landmark +
Annotation of Frontal Faces +
df674dc0fc813c2a6d539e892bfc74f9a761fbc8IOSR Journal of Computer Engineering (IOSR-JCE) +
e-ISSN: 2278-0661, p- ISSN: 2278-8727Volume 10, Issue 6 (May. - Jun. 2013), PP 21-29 +
www.iosrjournals.org +
An Image Mining System for Gender Classification & Age +
Prediction Based on Facial Features +
1.Ms.Dhanashri Shirkey , 2Prof.Dr.S.R.Gupta, +
M.E(Scholar),Department Computer Science & Engineering, PRMIT & R, Badnera +
Asstt.Prof. Department Computer Science & Engineering, PRMIT & R, Badnera +
da4170c862d8ae39861aa193667bfdbdf0ecb363Multi-task CNN Model for Attribute Prediction +
da15344a4c10b91d6ee2e9356a48cb3a0eac6a97
da5bfddcfe703ca60c930e79d6df302920ab9465
dac2103843adc40191e48ee7f35b6d86a02ef019854 +
Unsupervised Celebrity Face Naming in Web Videos +
dae420b776957e6b8cf5fbbacd7bc0ec226b3e2eRECOGNIZING EMOTIONS IN SPONTANEOUS FACIAL EXPRESSIONS +
Institut f¨ur Nachrichtentechnik +
Universit¨at Karlsruhe (TH), Germany +
daba8f0717f3f47c272f018d0a466a205eba6395
daefac0610fdeff415c2a3f49b47968d84692e87New Orleans, Louisiana, June 1 - 6, 2018. c(cid:13)2018 Association for Computational Linguistics +
Proceedings of NAACL-HLT 2018, pages 1481–1491 +
1481 +
b49affdff167f5d170da18de3efa6fd6a50262a2Author manuscript, published in "Workshop on Faces in 'Real-Life' Images: Detection, Alignment, and Recognition, Marseille : France +
(2008)" +
b41374f4f31906cf1a73c7adda6c50a78b4eb498This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. +
Iterative Gaussianization: From ICA to +
Random Rotations +
b4d7ca26deb83cec1922a6964c1193e8dd7270e7
b4ee64022cc3ccd14c7f9d4935c59b16456067d3Unsupervised Cross-Domain Image Generation +
b40290a694075868e0daef77303f2c4ca1c43269第 40 卷 第 4 期 +
2014 年 4 月 +
自 动 化 学 报 +
ACTA AUTOMATICA SINICA +
Vol. 40, No. 4 +
April, 2014 +
融合局部与全局信息的头发形状模型 +
王 楠 1 艾海舟 1 +
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 +
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 +
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 +
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 +
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. +
关键词 头发形状建模, 部件模型, 部件配置算法, 支持向量机 +
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 +
DOI 10.3724/SP.J.1004.2014.00615 +
Combining Local and Global Information for Hair Shape Modeling +
AI Hai-Zhou1 +
a2359c0f81a7eb032cff1fe45e3b80007facaa2aTowards Structured Analysis of Broadcast Badminton Videos +
C.V.Jawahar +
CVIT, KCIS, IIIT Hyderabad +
a2d9c9ed29bbc2619d5e03320e48b45c15155195
a2b54f4d73bdb80854aa78f0c5aca3d8b56b571d
a27735e4cbb108db4a52ef9033e3a19f4dc0e5faIntention from Motion +
a50b4d404576695be7cd4194a064f0602806f3c4In Proceedings of BMVC, Edimburgh, UK, September 2006 +
Efficiently estimating facial expression and +
illumination in appearance-based tracking +
†ESCET, U. Rey Juan Carlos +
C/ Tulip´an, s/n +
28933 M´ostoles, Spain +
‡Facultad Inform´atica, UPM +
Campus de Montegancedo s/n +
28660 Boadilla del Monte, Spain +
http://www.dia.fi.upm.es/~pcr +
a56c1331750bf3ac33ee07004e083310a1e63ddcVol. xx, pp. x +
c(cid:13) xxxx Society for Industrial and Applied Mathematics +
x–x +
Efficient Point-to-Subspace Query in (cid:96)1 with Application to Robust Object +
Instance Recognition +
a54e0f2983e0b5af6eaafd4d3467b655a3de52f4Face Recognition Using Convolution Filters and +
Neural Networks +
Head, Dept. of E&E,PEC +
Sec-12, Chandigarh – 160012 +
Department of CSE & IT, PEC +
Sec-12, Chandigarh – 160012 +
C.P. Singh +
Physics Department, CFSL, +
Sec-36, Chandigarh - 160036 +
a +
of +
to: (a) +
potential method +
a55efc4a6f273c5895b5e4c5009eabf8e5ed0d6a818 +
Continuous Head Movement Estimator for +
Driver Assistance: Issues, Algorithms, +
and On-Road Evaluations +
Mohan Manubhai Trivedi, Fellow, IEEE +
a5c04f2ad6a1f7c50b6aa5b1b71c36af76af06be
a503eb91c0bce3a83bf6f524545888524b29b166
a5a44a32a91474f00a3cda671a802e87c899fbb4Moments in Time Dataset: one million +
videos for event understanding +
bd9eb65d9f0df3379ef96e5491533326e9dde315
bd07d1f68486052b7e4429dccecdb8deab1924db
bd8e2d27987be9e13af2aef378754f89ab20ce10
bd2d7c7f0145028e85c102fe52655c2b6c26aeb5Attribute-based People Search: Lessons Learnt from a +
Practical Surveillance System +
Rogerio Feris +
IBM Watson +
http://rogerioferis.com +
Russel Bobbitt +
IBM Watson +
Lisa Brown +
IBM Watson +
IBM Watson +
bdbba95e5abc543981fb557f21e3e6551a563b45International Journal of Computational Intelligence and Applications +
Vol. 17, No. 2 (2018) 1850008 (15 pages) +
#.c The Author(s) +
DOI: 10.1142/S1469026818500086 +
Speeding up the Hyperparameter Optimization of Deep +
Convolutional Neural Networks +
Knowledge Technology, Department of Informatics +
Universit€at Hamburg +
Vogt-K€olln-Str. 30, Hamburg 22527, Germany +
Received 15 August 2017 +
Accepted 23 March 2018 +
Published 18 June 2018 +
Most learning algorithms require the practitioner to manually set the values of many hyper- +
parameters before the learning process can begin. However, with modern algorithms, the +
evaluation of a given hyperparameter setting can take a considerable amount of time and the +
search space is often very high-dimensional. We suggest using a lower-dimensional represen- +
tation of the original data to quickly identify promising areas in the hyperparameter space. This +
information can then be used to initialize the optimization algorithm for the original, higher- +
dimensional data. We compare this approach with the standard procedure of optimizing the +
hyperparameters only on the original input. +
We perform experiments with various state-of-the-art hyperparameter optimization algo- +
rithms such as random search, the tree of parzen estimators (TPEs), sequential model-based +
algorithm con¯guration (SMAC), and a genetic algorithm (GA). Our experiments indicate that +
it is possible to speed up the optimization process by using lower-dimensional data repre- +
sentations at the beginning, while increasing the dimensionality of the input later in the opti- +
mization process. This is independent of the underlying optimization procedure, making the +
approach promising for many existing hyperparameter optimization algorithms. +
Keywords: Hyperparameter optimization; hyperparameter importance; convolutional neural +
networks; genetic algorithm; Bayesian optimization. +
1. Introduction +
The performance of many contemporary machine learning algorithms depends cru- +
cially on the speci¯c initialization of hyperparameters such as the general architec- +
ture, the learning rate, regularization parameters, and many others.1,2 Indeed, +
This is an Open Access article published by World Scienti¯c Publishing Company. It is distributed under +
the terms of the Creative Commons Attribution 4.0 (CC-BY) License. Further distribution of this work is +
permitted, provided the original work is properly cited. +
1850008-1 +
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.
d1dfdc107fa5f2c4820570e369cda10ab1661b87Super SloMo: High Quality Estimation of Multiple Intermediate Frames +
for Video Interpolation +
Erik Learned-Miller1 +
1UMass Amherst +
2NVIDIA 3UC Merced +
d1a43737ca8be02d65684cf64ab2331f66947207IJB–S: IARPA Janus Surveillance Video Benchmark (cid:3) +
Kevin O’Connor z +
d1082eff91e8009bf2ce933ac87649c686205195(will be inserted by the editor) +
Pruning of Error Correcting Output Codes by +
Optimization of Accuracy-Diversity Trade off +
S¨ureyya ¨Oz¨o˘g¨ur Aky¨uz · Terry +
Windeatt · Raymond Smith +
Received: date / Accepted: date +
d69df51cff3d6b9b0625acdcbea27cd2bbf4b9c0
d6102a7ddb19a185019fd2112d2f29d9258f6decProceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
3721 +
d6bfa9026a563ca109d088bdb0252ccf33b76bc6Unsupervised Temporal Segmentation of Facial Behaviour +
Department of Computer Science and Engineering, IIT Kanpur +
d6fb606e538763282e3942a5fb45c696ba38aee6
bc9003ad368cb79d8a8ac2ad025718da5ea36bc4Technische Universit¨at M¨unchen +
Bildverstehen und Intelligente Autonome Systeme +
Facial Expression Recognition With A +
Three-Dimensional Face Model +
Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Informatik der Technischen Uni- +
versit¨at M¨unchen zur Erlangung des akademischen Grades eines +
Doktors der Naturwissenschaften +
genehmigten Dissertation. +
Vorsitzender: +
Univ.-Prof. Dr. Johann Schlichter +
Pr¨ufer der Dissertation: 1. Univ.-Prof. Dr. Bernd Radig (i.R.) +
2. Univ.-Prof. Gudrun J. Klinker, Ph.D. +
Die Dissertation wurde am 04.07.2011 bei der Technischen Universit¨at M¨unchen +
eingereicht und durch die Fakult¨at f¨ur Informatik am 02.12.2011 angenommen. +
bcc346f4a287d96d124e1163e4447bfc47073cd8
bcc172a1051be261afacdd5313619881cbe0f676978-1-5090-4117-6/17/$31.00 ©2017 IEEE +
2197 +
ICASSP 2017 +
bcfeac1e5c31d83f1ed92a0783501244dde5a471
bc2852fa0a002e683aad3fb0db5523d1190d0ca5
bcb99d5150d792001a7d33031a3bd1b77bea706b
bc811a66855aae130ca78cd0016fd820db1603ecTowards three-dimensional face recognition in the real +
To cite this version: +
HAL Id: tel-00998798 +
https://tel.archives-ouvertes.fr/tel-00998798 +
Submitted on 2 Jun 2014 +
archive for the deposit and dissemination of sci- +
entific research documents, whether they are pub- +
teaching and research institutions in France or +
destin´ee au d´epˆot et `a la diffusion de documents +
recherche fran¸cais ou ´etrangers, des laboratoires +
bc9af4c2c22a82d2c84ef7c7fcc69073c19b30abMoCoGAN: Decomposing Motion and Content for Video Generation +
Snap Research +
NVIDIA +
bcac3a870501c5510df80c2a5631f371f2f6f74aCVPR +
#1387 +
000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
045 +
046 +
047 +
048 +
049 +
050 +
051 +
052 +
053 +
CVPR 2013 Submission #1387. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +
CVPR +
#1387 +
Structured Face Hallucination +
Anonymous CVPR submission +
Paper ID 1387 +
aed321909bb87c81121c841b21d31509d6c78f69
ae936628e78db4edb8e66853f59433b8cc83594f
ae2cf545565c157813798910401e1da5dc8a6199Mahkonen et al. EURASIP Journal on Image and Video +
Processing (2018) 2018:61 +
https://doi.org/10.1186/s13640-018-0303-9 +
EURASIP Journal on Image +
and Video Processing +
RESEARCH +
Open Access +
Cascade of Boolean detector +
combinations +
aebb9649bc38e878baef082b518fa68f5cda23a5 +
aeff403079022683b233decda556a6aee3225065DeepFace: Face Generation using Deep Learning +
ae753fd46a744725424690d22d0d00fb05e53350000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
Describing Clothing by Semantic Attributes +
Anonymous ECCV submission +
Paper ID 727 +
ae4e2c81c8a8354c93c4b21442c26773352935dd
ae85c822c6aec8b0f67762c625a73a5d08f5060dThis 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. +
The final version of record is available at http://dx.doi.org/10.1109/TPAMI.2014.2353624 +
IEEE TRANSACTION ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. M, NO. N, MONTH YEAR +
Retrieving Similar Styles to Parse Clothing +
d861c658db2fd03558f44c265c328b53e492383aAutomated Face Extraction and Normalization of 3D Mesh Data +
d83d2fb5403c823287f5889b44c1971f049a1c93Motiv Emot +
DOI 10.1007/s11031-013-9353-6 +
O R I G I N A L P A P E R +
Introducing the sick face +
Ó Springer Science+Business Media New York 2013 +
d8b568392970b68794a55c090c4dd2d7f90909d2PDA Face Recognition System +
Using Advanced Correlation +
Filters +
Chee Kiat Ng +
2005 +
Advisor: Prof. Khosla/Reviere +
d83ae5926b05894fcda0bc89bdc621e4f21272daversion of the following thesis: +
Frugal Forests: Learning a Dynamic and Cost Sensitive +
Feature Extraction Policy for Anytime Activity Classification +
d89cfed36ce8ffdb2097c2ba2dac3e2b2501100dRobust Face Recognition via Multimodal Deep +
Face Representation +
ab8f9a6bd8f582501c6b41c0e7179546e21c5e91Nonparametric Face Verification Using a Novel +
Face Representation +
ab58a7db32683aea9281c188c756ddf969b4cdbdEfficient Solvers for Sparse Subspace Clustering +
ab989225a55a2ddcd3b60a99672e78e4373c0df1Sample, Computation vs Storage Tradeoffs for +
Classification Using Tensor Subspace Models +
ab6776f500ed1ab23b7789599f3a6153cdac84f7International Journal of Scientific & Engineering Research, Volume 6, Issue 4, April-2015 1212 +
ISSN 2229-5518 +
A Survey on Various Facial Expression +
Techniques +
ab2b09b65fdc91a711e424524e666fc75aae7a51Multi-modal Biomarkers to Discriminate Cognitive State* +
1MIT Lincoln Laboratory, Lexington, Massachusetts, USA +
2USARIEM, 3NSRDEC +
1. Introduction +
Multimodal biomarkers based on behavorial, neurophysiolgical, and cognitive measurements have +
recently obtained increasing popularity in the detection of cognitive stress- and neurological-based +
disorders. Such conditions are significantly and adversely affecting human performance and quality +
of life for a large fraction of the world’s population. Example modalities used in detection of these +
conditions include voice, facial expression, physiology, eye tracking, gait, and EEG analysis. +
Toward the goal of finding simple, noninvasive means to detect, predict and monitor cognitive +
stress and neurological conditions, MIT Lincoln Laboratory is developing biomarkers that satisfy +
three criteria. First, we seek biomarkers that reflect core components of cognitive status such as +
working memory capacity, processing speed, attention, and arousal. Second, and as importantly, we +
seek biomarkers that reflect timing and coordination relations both within components of each +
modality and across different modalities. This is based on the hypothesis that neural coordination +
across different parts of the brain is essential in cognition (Figure 1). An example of timing and +
coordination within a modality is the set of finely timed and synchronized physiological +
components of speech production, while an example of coordination across modalities is the timing +
and synchrony that occurs across speech and facial expression while speaking. Third, we seek +
multimodal biomarkers that contribute in a complementary fashion under various channel and +
background conditions. In this chapter, as an illustration of this biomarker approach we focus on +
cognitive stress and the particular case of detecting different cognitive load levels. We also briefly +
show how similar feature-extraction principles can be applied to a neurological condition through +
the example of major depression disorder (MDD). MDD is one of several neurological disorders +
where multi-modal biomarkers based on principles of timing and coordination are important for +
detection [11]-[22]. In our cognitive load experiments, we use two easily obtained noninvasive +
modalities, voice and face, and show how these two modalities can be fused to produce results on +
par with more invasive, “gold-standard” EEG measurements. Vocal and facial biomarkers will also +
be used in our MDD case study. In both application areas we focus on timing and coordination +
relations within the components of each modality. +
* Distribution A: public release.This work is sponsored by the Assistant Secretary of Defense for Research & Engineering under Air Force contract +
#FA8721-05-C-0002. Opinions,interpretations, conclusions, and recommendations are those of the authors and are not necessarily endorsed by the United States +
Government. +
ab87dfccb1818bdf0b41d732da1f9335b43b74aeSUBMITTED TO IEEE TRANSACTIONS ON SIGNAL PROCESSING +
Structured Dictionary Learning for Classification +
ab1dfcd96654af0bf6e805ffa2de0f55a73c025d
abeda55a7be0bbe25a25139fb9a3d823215d7536UNIVERSITATPOLITÈCNICADECATALUNYAProgramadeDoctorat:AUTOMÀTICA,ROBÒTICAIVISIÓTesiDoctoralUnderstandingHuman-CentricImages:FromGeometrytoFashionEdgarSimoSerraDirectors:FrancescMorenoNoguerCarmeTorrasMay2015
ab1900b5d7cf3317d17193e9327d57b97e24d2fc
ab8fb278db4405f7db08fa59404d9dd22d38bc83UNIVERSITÉ DE GENÈVE +
Département d'Informatique +
FACULTÉ DES SCIENCES +
Implicit and Automated Emotional +
Tagging of Videos +
THÈSE +
présenté à la Faculté des sciences de l'Université de Genève +
pour obtenir le grade de Docteur ès sciences, mention informatique +
par +
de +
Téhéran (IRAN) +
Thèse No 4368 +
GENÈVE +
Repro-Mail - Université de Genève +
2011 +
e5737ffc4e74374b0c799b65afdbf0304ff344cb
e5823a9d3e5e33e119576a34cb8aed497af20eeaDocFace+: ID Document to Selfie* Matching +
e5dfd17dbfc9647ccc7323a5d62f65721b318ba9
e56c4c41bfa5ec2d86c7c9dd631a9a69cdc05e69Human Activity Recognition Based on Wearable +
Sensor Data: A Standardization of the +
State-of-the-Art +
Smart Surveillance Interest Group, Computer Science Department +
Universidade Federal de Minas Gerais, Brazil +
e27c92255d7ccd1860b5fb71c5b1277c1648ed1e
e200c3f2849d56e08056484f3b6183aa43c0f13a
f437b3884a9e5fab66740ca2a6f1f3a5724385eaHuman Identification Technical Challenges +
DARPA +
3701 N. Fairfax Dr +
Arlington, VA 22203 +
f442a2f2749f921849e22f37e0480ac04a3c3fec
f4f6fc473effb063b7a29aa221c65f64a791d7f4Downloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Electronic-Imaging on 4/20/2018 Terms of Use: https://www.spiedigitallibrary.org/terms-of-use +
FacialexpressionrecognitioninthewildbasedonmultimodaltexturefeaturesBoSunLiandongLiGuoyanZhouJunHeBoSun,LiandongLi,GuoyanZhou,JunHe,“Facialexpressionrecognitioninthewildbasedonmultimodaltexturefeatures,”J.Electron.Imaging25(6),061407(2016),doi:10.1117/1.JEI.25.6.061407.
f4c01fc79c7ead67899f6fe7b79dd1ad249f71b0
f4373f5631329f77d85182ec2df6730cbd4686a9Soft Computing manuscript No. +
(will be inserted by the editor) +
Recognizing Gender from Human Facial Regions using +
Genetic Algorithm +
Received: date / Accepted: date +
f47404424270f6a20ba1ba8c2211adfba032f405International Journal of Emerging Technology and Advanced Engineering +
Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 5, May 2012) +
Identification of Face Age range Group using Neural +
Network +
f3fcaae2ea3e998395a1443c87544f203890ae15
f3d9e347eadcf0d21cb0e92710bc906b22f2b3e7NosePose: a competitive, landmark-free +
methodology for head pose estimation in the wild +
IMAGO Research Group - Universidade Federal do Paran´a +
f355e54ca94a2d8bbc598e06e414a876eb62ef99
f3ea181507db292b762aa798da30bc307be95344Covariance Pooling for Facial Expression Recognition +
†Computer Vision Lab, ETH Zurich, Switzerland +
‡VISICS, KU Leuven, Belgium +
f3cf10c84c4665a0b28734f5233d423a65ef1f23Title +
Temporal Exemplar-based Bayesian Networks for facial +
expression recognition +
Author(s) +
Shang, L; Chan, KP +
Citation +
Proceedings - 7Th International Conference On Machine +
Learning And Applications, Icmla 2008, 2008, p. 16-22 +
Issued Date +
2008 +
URL +
http://hdl.handle.net/10722/61208 +
Rights +
This work is licensed under a Creative Commons Attribution- +
NonCommercial-NoDerivatives 4.0 International License.; +
International Conference on Machine Learning and Applications +
Proceedings. Copyright © IEEE.; ©2008 IEEE. Personal use of +
this material is permitted. However, permission to +
reprint/republish this material for advertising or promotional +
purposes or for creating new collective works for resale or +
redistribution to servers or lists, or to reuse any copyrighted +
component of this work in other works must be obtained from +
the IEEE. +
f3b7938de5f178e25a3cf477107c76286c0ad691JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, MARCH 2017 +
Object Detection with Deep Learning: A Review +
ebedc841a2c1b3a9ab7357de833101648281ff0e
eb526174fa071345ff7b1fad1fad240cd943a6d7Deeply Vulnerable – A Study of the Robustness of Face Recognition to +
Presentation Attacks +
eb566490cd1aa9338831de8161c6659984e923fdFrom Lifestyle Vlogs to Everyday Interactions +
EECS Department, UC Berkeley +
eb9312458f84a366e98bd0a2265747aaed40b1a61-4244-1437-7/07/$20.00 ©2007 IEEE +
IV - 473 +
ICIP 2007 +
eb716dd3dbd0f04e6d89f1703b9975cad62ffb09Copyright +
by +
2012 +
ebabd1f7bc0274fec88a3dabaf115d3e226f198fDriver drowsiness detection system based on feature +
representation learning using various deep networks +
School of Electrical Engineering, KAIST, +
Guseong-dong, Yuseong-gu, Dajeon, Rep. of Korea +
ebb9d53668205c5797045ba130df18842e3eadef
eb48a58b873295d719827e746d51b110f5716d6cFace Alignment Using K-cluster Regression Forests +
With Weighted Splitting +
c7e4c7be0d37013de07b6d829a3bf73e1b95ad4eThe International Journal of Multimedia & Its Applications (IJMA) Vol.5, No.5, October 2013 +
DYNEMO: A VIDEO DATABASE OF NATURAL FACIAL +
EXPRESSIONS OF EMOTIONS +
1LIP, Univ. Grenoble Alpes, BP 47 - 38040 Grenoble Cedex 9, France +
2LIG, Univ. Grenoble Alpes, BP 53 - 38041 Grenoble Cedex 9, France +
c7c5f0fe1fcaf3787c7f78f7dc62f3497dcfdf3cTHE IMPACT OF PRODUCT PHOTO ON ONLINE CONSUMER +
PURCHASE INTENTION: AN IMAGE-PROCESSING ENABLED +
EMPIRICAL STUDY +
c758b9c82b603904ba8806e6193c5fefa57e9613Heterogeneous Face Recognition with CNNs +
INRIA Grenoble, Laboratoire Jean Kuntzmann +
c7c8d150ece08b12e3abdb6224000c07a6ce7d47DeMeshNet: Blind Face Inpainting for Deep MeshFace Verification +
National Laboratory of Pattern Recognition, CASIA +
Center for Research on Intelligent Perception and Computing, CASIA +
c038beaa228aeec174e5bd52460f0de75e9cccbeTemporal Segment Networks for Action +
Recognition in Videos +
c043f8924717a3023a869777d4c9bee33e607fb5Emotion Separation Is Completed Early and It Depends +
on Visual Field Presentation +
Lab for Human Brain Dynamics, RIKEN Brain Science Institute, Wakoshi, Saitama, Japan, 2 Lab for Human Brain Dynamics, AAI Scientific Cultural Services Ltd., Nicosia
Cyprus +
c05a7c72e679745deab9c9d7d481f7b5b9b36bddNPS-CS-11-005 +
+
+
NAVAL +
POSTGRADUATE +
SCHOOL +
MONTEREY, CALIFORNIA +
by +
BIOMETRIC CHALLENGES FOR FUTURE DEPLOYMENTS: +
A STUDY OF THE IMPACT OF GEOGRAPHY, CLIMATE, CULTURE, +
AND SOCIAL CONDITIONS ON THE EFFECTIVE +
COLLECTION OF BIOMETRICS +
April 2011 +
Approved for public release; distribution is unlimited +
c02847a04a99a5a6e784ab580907278ee3c12653Fine Grained Video Classification for +
Endangered Bird Species Protection +
Non-Thesis MS Final Report +
1. Introduction +
1.1 Background +
This project is about detecting eagles in videos. Eagles are endangered species at the brim of +
extinction since 1980s. With the bans of harmful pesticides, the number of eagles keep increasing. +
However, recent studies on golden eagles’ activities in the vicinity of wind turbines have shown +
significant number of turbine blade collisions with eagles as the major cause of eagles’ mortality. [1] +
This project is a part of a larger research project to build an eagle detection and deterrent system +
on wind turbine toward reducing eagles’ mortality. [2] The critical component of this study is a +
computer vision system for eagle detection in videos. The key requirement are that the system should +
work in real time and detect eagles at a far distance from the camera (i.e. in low resolution). +
There are three different bird species in my dataset - falcon, eagle and seagull. The reason for +
involving only these three species is based on the real world situation. Wind turbines are always +
installed near coast and mountain hill where falcons and seagulls will be the majority. So my model +
will classify the minority eagles out of other bird species during the immigration season and protecting +
them by using the deterrent system. +
1.2 Brief Approach +
Our approach represents a unified deep-learning architecture for eagle detection. Given videos, +
our goal is to detect eagle species at far distance from the camera, using both appearance and bird +
motion cues, so as to meet the recall-precision rates set by the user. Detecting eagle is a challenging +
task because of the following reasons. Frist, an eagle flies fast and high in the sky which means that +
we need a lens with wide angle such that captures their movement. However, a camera with wide +
angle produces a low resolution and low quality video and the detailed appearance of bird is +
compromised. Second, current neural network typically take as input low resolution images. This is +
because a higher resolution image will require larger filters and deeper networks which is turn hard to +
train [3]. So it is not clear whether the low resolution will cause challenge for fine-grained +
classification task. Last but not the least, there is not a large training database like PASCAL, MNIST +
c0c8d720658374cc1ffd6116554a615e846c74b5JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
Modeling Multimodal Clues in a Hybrid Deep +
Learning Framework for Video Classification +
c0d5c3aab87d6e8dd3241db1d931470c15b9e39d
eee8a37a12506ff5df72c402ccc3d59216321346Uredniki: +
dr. Tomaž Erjavec +
Odsek za tehnologije znanja +
Institut »Jožef Stefan«, Ljubljana +
dr. Jerneja Žganec Gros +
Alpineon d.o.o, Ljubljana +
Založnik: Institut »Jožef Stefan«, Ljubljana +
Tisk: Birografika BORI d.o.o. +
Priprava zbornika: Mitja Lasič +
Oblikovanje naslovnice: dr. Damjan Demšar +
Tiskano iz predloga avtorjev +
Naklada: 50 +
Ljubljana, oktober 2008 +
Konferenco IS 2008 sofinancirata +
Ministrstvo za visoko šolstvo, znanost in tehnologijo +
Institut »Jožef Stefan« +
ISSN 1581-9973 +
CIP - Kataložni zapis o publikaciji +
Narodna in univerzitetna knjižnica, Ljubljana +
004.934(082) +
81'25:004.6(082) +
004.8(063) +
oktober 2008, Ljubljana, Slovenia : zbornik 11. mednarodne +
Proceedings of the Sixth Language Technologies Conference, October +
16th-17th, 2008 : proceedings of the 11th International +
Multiconference Information Society - IS 2008, volume C / uredila, +
edited by Tomaž Erjavec, Jerneja Žganec Gros. - Ljubljana : +
1581-9973) +
ISBN 978-961-264-006-4 +
družba 4. Information society 5. Erjavec, Tomaž, 1960- 6. +
Ljubljana) +
241520896 +
ee18e29a2b998eddb7f6663bb07891bfc72622481119 +
Local Linear Discriminant Analysis Framework +
Using Sample Neighbors +
ee461d060da58d6053d2f4988b54eff8655ecede
eefb8768f60c17d76fe156b55b8a00555eb40f4dSubspace Scores for Feature Selection in Computer Vision +
eed1dd2a5959647896e73d129272cb7c3a2e145c
ee92d36d72075048a7c8b2af5cc1720c7bace6ddFACE RECOGNITION USING MIXTURES OF PRINCIPAL COMPONENTS +
Video and Display Processing +
Philips Research USA +
Briarcliff Manor, NY 10510 +
eedfb384a5e42511013b33104f4cd3149432bd9eMultimodal Probabilistic Person +
Tracking and Identification +
in Smart Spaces +
zur Erlangung des akademischen Grades eines +
Doktors der Ingenieurwissenschaften +
der Fakultät für Informatik +
der Universität Fridericiana zu Karlsruhe (TH) +
genehmigte +
Dissertation +
von +
aus Karlsruhe +
Tag der mündlichen Prüfung: 20.11.2009 +
Erster Gutachter: +
Zweiter Gutachter: +
Prof. Dr. A. Waibel +
Prof. Dr. R. Stiefelhagen +
c9424d64b12a4abe0af201e7b641409e182bababArticle +
Which, When, and How: Hierarchical Clustering with +
Human–Machine Cooperation +
Academic Editor: Tom Burr +
Received: 3 November 2016; Accepted: 14 December 2016; Published: 21 December 2016 +
c903af0d69edacf8d1bff3bfd85b9470f6c4c243
fc1e37fb16006b62848def92a51434fc74a2431aDRAFT +
A Comprehensive Analysis of Deep Regression +
fc516a492cf09aaf1d319c8ff112c77cfb55a0e5
fcd3d69b418d56ae6800a421c8b89ef363418665Effects of Aging over Facial Feature Analysis and Face +
Recognition +
Bogaziçi Un. Electronics Eng. Dept. March 2010 +
fcd77f3ca6b40aad6edbd1dab9681d201f85f365c(cid:13)Copyright 2014 +
fcf8bb1bf2b7e3f71fb337ca3fcf3d9cf18daa46MANUSCRIPT SUBMITTED TO IEEE TRANS. PATTERN ANAL. MACH. INTELL., JULY 2010 +
Feature Selection via Sparse Approximation for +
Face Recognition +
fcbf808bdf140442cddf0710defb2766c2d25c30IJCV manuscript No. +
(will be inserted by the editor) +
Unsupervised Semantic Action Discovery from Video +
Collections +
Received: date / Accepted: date +
fd4ac1da699885f71970588f84316589b7d8317bJOURNAL OF LATEX CLASS FILES, VOL. 6, NO. 1, JANUARY 2007 +
Supervised Descent Method +
for Solving Nonlinear Least Squares +
Problems in Computer Vision +
fdf533eeb1306ba418b09210387833bdf27bb756951 +
fdda5852f2cffc871fd40b0cb1aa14cea54cd7e3Im2Flow: Motion Hallucination from Static Images for Action Recognition +
UT Austin +
UT Austin +
UT Austin +
fdfaf46910012c7cdf72bba12e802a318b5bef5aComputerized Face Recognition in Renaissance +
Portrait Art +
fd15e397629e0241642329fc8ee0b8cd6c6ac807Semi-Supervised Clustering with Neural Networks +
IIIT-Delhi, India +
fdca08416bdadda91ae977db7d503e8610dd744f +
ICT-2009.7.1 +
KSERA Project +
2010-248085 +
Deliverable D3.1 +
Deliverable D3.1 +
Human Robot Interaction +
Human Robot Interaction +
18 October 2010 +
Public Document +
The KSERA project (http://www.ksera +
KSERA project (http://www.ksera-project.eu) has received funding from the European Commission +
project.eu) has received funding from the European Commission +
under the 7th Framework Programme (FP7) for Research and Technological Development under grant +
under the 7th Framework Programme (FP7) for Research and Technological Development under grant +
under the 7th Framework Programme (FP7) for Research and Technological Development under grant +
agreement n°2010-248085. +
fdaf65b314faee97220162980e76dbc8f32db9d6Accepted Manuscript +
Face recognition using both visible light image and near-infrared image and a deep +
network +
PII: +
DOI: +
Reference: +
S2468-2322(17)30014-8 +
10.1016/j.trit.2017.03.001 +
TRIT 41 +
To appear in: +
CAAI Transactions on Intelligence Technology +
Received Date: 30 January 2017 +
Accepted Date: 28 March 2017 +
Please cite this article as: K. Guo, S. Wu, Y. Xu, Face recognition using both visible light image and +
near-infrared image and a deep network, CAAI Transactions on Intelligence Technology (2017), doi: +
10.1016/j.trit.2017.03.001. +
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to +
our customers we are providing this early version of the manuscript. The manuscript will undergo +
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please +
note that during the production process errors may be discovered which could affect the content, and all +
legal disclaimers that apply to the journal pertain. +
f2e9494d0dca9fb6b274107032781d435a508de6
f2c568fe945e5743635c13fe5535af157b1903d1
f26097a1a479fb6f32b27a93f8f32609cfe30fdc
f231046d5f5d87e2ca5fae88f41e8d74964e8f4fWe are IntechOpen, +
the first native scientific +
publisher of Open Access books +
3,350 +
108,000 +
1.7 M +
Open access books available +
International authors and editors +
Downloads +
Our authors are among the +
151 +
Countries delivered to +
TOP 1% +
12.2% +
most cited scientists +
Contributors from top 500 universities +
Selection of our books indexed in the Book Citation Index +
in Web of Science™ Core Collection (BKCI) +
Interested in publishing with us? +
Numbers displayed above are based on latest data collected. +
For more information visit www.intechopen.com +
f214bcc6ecc3309e2efefdc21062441328ff6081
f5770dd225501ff3764f9023f19a76fad28127d4Real Time Online Facial Expression Transfer +
with Single Video Camera +
f519723238701849f1160d5a9cedebd31017da89Impact of multi-focused images on recognition of soft biometric traits +
aEURECOM, Campus SophiaTech, 450 Route des Chappes, CS 50193 - 06904 Biot Sophia +
+
Antipolis cedex, FRANCE +
f558af209dd4c48e4b2f551b01065a6435c3ef33International Journal of Emerging Technology in Computer Science & Electronics (IJETCSE) +
ISSN: 0976-1353 Volume 23 Issue 1 –JUNE 2016. +
AN ENHANCED ATTRIBUTE +
RERANKING DESIGN FOR WEB IMAGE +
SEARCH +
#Student,Cse, CIET, Lam,Guntur, India +
* Assistant Professort,Cse, CIET, Lam,Guntur , India +
e393a038d520a073b9835df7a3ff104ad610c552Automatic temporal segment +
detection via bilateral long short- +
term memory recurrent neural +
networks +
detection via bilateral long short-term memory recurrent neural networks,” J. +
Electron. Imaging 26(2), 020501 (2017), doi: 10.1117/1.JEI.26.2.020501. +
Downloaded From: http://electronicimaging.spiedigitallibrary.org/ on 03/03/2017 Terms of Use: http://spiedigitallibrary.org/ss/termsofuse.aspx
e3657ab4129a7570230ff25ae7fbaccb4ba9950c
e315959d6e806c8fbfc91f072c322fb26ce0862bAn Efficient Face Recognition System Based on Sub-Window +
International Journal of Soft Computing and Engineering (IJSCE) +
ISSN: 2231-2307, Volume-1, Issue-6, January 2012 +
Extraction Algorithm +
e3c011d08d04c934197b2a4804c90be55e21d572How to Train Triplet Networks with 100K Identities? +
Orion Star +
Beijing, China +
Orion Star +
Beijing, China +
Orion Star +
Beijing, China +
e39a0834122e08ba28e7b411db896d0fdbbad9ba1368 +
Maximum Likelihood Estimation of Depth Maps +
Using Photometric Stereo +
e3917d6935586b90baae18d938295e5b089b5c62152 +
Face Localization and Authentication +
Using Color and Depth Images +
cfa572cd6ba8dfc2ee8ac3cc7be19b3abff1a8a2
cfffae38fe34e29d47e6deccfd259788176dc213TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. X, NO. X, DECEMBER 2012 +
Matrix Completion for Weakly-supervised +
Multi-label Image Classification +
cfd4004054399f3a5f536df71f9b9987f060f434IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. ??, NO. ??, ?? 20?? +
Person Recognition in Personal Photo Collections +
cfb8bc66502fb5f941ecdb22aec1fdbfdb73adce
cf875336d5a196ce0981e2e2ae9602580f3f62437 What 1 +
Rosalind W. Picard +
It Mean for a Computer to "Have" Emotions? +
There is a lot of talk about giving machines emotions, some of +
it fluff. Recently at a large technical meeting, a researcher stood up +
and talked of how a Bamey stuffed animal [the purple dinosaur for +
kids) "has emotions." He did not define what he meant by this, but +
after repeating it several times, it became apparent that children +
attributed emotions to Barney, and that Barney had deliberately +
expressive behaviors that would encourage the kids to think. Bar- +
ney had emotions. But kids have attributed emotions to dolls and +
stuffed animals for as long a s we know; and most of my technical +
colleagues would agree that such toys have never had and still do +
not have emotions. What is different now that prompts a researcher +
to make such a claim? Is the computational plush an example of a +
computer that really does have emotions? +
If not Barney, then what would be an example of a computa- +
tional system that has emotions? I am not a philosopher, and this +
paper will not be a discussion of the meaning of this question in +
any philosophical sense. However, as an engineer I am interested +
in what capabilities I would require a machine to have before I +
would say that it "has emotions," if that is even possible. +
Theorists still grappl~ with the problem of defining emotion, +
after many decades of discussion, and no clean definition looks +
likely to emerge. Even without a precise definition, one can still +
begin to say concrete things about certain components of emotion, +
at least based on what is known about human and animal emo- +
tions. Of course, much is still u d a o w n about human emotions, so +
we are nowhere near being able to model them, much less dupli- +
cate all their functions in machines.'~lso, all scientific findings are +
subject to revision-history has certainly taught us humility, that +
what scientists believed to be true at one point has often been +
changed at a later date. +
I wish to begin by mentioning four motivations for giving +
machines certain emotional abilities (and there are more). One goal +
is to build robots and synthetic characters that can emulate living +
humans and animals-for example, to build a humanoid robot. A +
I +
cf54a133c89f730adc5ea12c3ac646971120781c
cfbb2d32586b58f5681e459afd236380acd86e28Improving Alignment of Faces for Recognition +
Christopher J. Pal +
D´epartement de g´enie informatique et g´enie logiciel +
´Ecole Polytechnique de Montr´eal, +
D´epartement de g´enie informatique et g´enie logiciel +
´Ecole Polytechnique de Montr´eal, +
Qu´ebec, Canada +
Qu´ebec, Canada +
cfa92e17809e8d20ebc73b4e531a1b106d02b38cAdvances in Data Analysis and Classification manuscript No. +
(will be inserted by the editor) +
Parametric Classification with Soft Labels using the +
Evidential EM Algorithm +
Linear Discriminant Analysis vs. Logistic Regression +
Received: date / Accepted: date +
cfdc632adcb799dba14af6a8339ca761725abf0aProbabilistic Formulations of Regression with Mixed +
Guidance +
cfc30ce53bfc204b8764ebb764a029a8d0ad01f4Regularizing Deep Neural Networks by Noise: +
Its Interpretation and Optimization +
Dept. of Computer Science and Engineering, POSTECH, Korea +
cf86616b5a35d5ee777585196736dfafbb9853b5This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. +
Learning Multiscale Active Facial Patches for +
Expression Analysis +
cad52d74c1a21043f851ae14c924ac689e197d1fFrom Ego to Nos-vision: +
Detecting Social Relationships in First-Person Views +
Universit`a degli Studi di Modena e Reggio Emilia +
Via Vignolese 905, 41125 Modena - Italy +
cac8bb0e393474b9fb3b810c61efdbc2e2c25c29
cad24ba99c7b6834faf6f5be820dd65f1a755b29Understanding hand-object +
manipulation by modeling the +
contextual relationship between actions, +
grasp types and object attributes +
Journal Title +
XX(X):1–14 +
c(cid:13)The Author(s) 2016 +
Reprints and permission: +
sagepub.co.uk/journalsPermissions.nav +
DOI: 10.1177/ToBeAssigned +
www.sagepub.com/ +
cadba72aa3e95d6dcf0acac828401ddda7ed8924THÈSE PRÉSENTÉE À LA FACULTÉ DES SCIENCES +
POUR L’OBTENTION DU GRADE DE DOCTEUR ÈS SCIENCES +
Algorithms and VLSI Architectures +
for Low-Power Mobile Face Verification +
par +
Acceptée sur proposition du jury: +
Prof. F. Pellandini, directeur de thèse +
PD Dr. M. Ansorge, co-directeur de thèse +
Prof. P.-A. Farine, rapporteur +
Dr. C. Piguet, rapporteur +
Soutenue le 2 juin 2005 +
INSTITUT DE MICROTECHNIQUE +
UNIVERSITÉ DE NEUCHÂTEL +
2006 +
ca606186715e84d270fc9052af8500fe23befbdaUsing Subclass Discriminant Analysis, Fuzzy Integral and Symlet Decomposition for +
Face Recognition +
Department of Electrical Engineering, +
Iran Univ. of Science and Technology, +
Narmak, Tehran, Iran +
Department of Electrical Engineering, +
Iran Univ. of Science and Technology, +
Department of Electrical Engineering, +
Iran Univ. of Science and Technology, +
Narmak, Tehran, Iran +
Narmak, Tehran, Iran +
e465f596d73f3d2523dbf8334d29eb93a35f6da0
e4aeaf1af68a40907fda752559e45dc7afc2de67
e4c3d5d43cb62ac5b57d74d55925bdf76205e306
e4a1b46b5c639d433d21b34b788df8d81b518729JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
Side Information for Face Completion: a Robust +
PCA Approach +
e4c81c56966a763e021938be392718686ba9135e
e4e95b8bca585a15f13ef1ab4f48a884cd6ecfccFace Recognition with Independent Component Based +
Super-resolution +
aFaculty of Engineering and Natural Sciences, Sabanci Univ., Istanbul, Turkiye, 34956 +
bSchool of Elec. and Comp. Eng. , Georgia Inst. of Tech., Atlanta, GA, USA, 30332-0250 +
e43ea078749d1f9b8254e0c3df4c51ba2f4eebd5Facial Expression Recognition Based on Constrained +
Local Models and Support Vector Machines +
e476cbcb7c1de73a7bcaeab5d0d59b8b3c4c1cbf
e475e857b2f5574eb626e7e01be47b416deff268Facial Emotion Recognition Using Nonparametric +
Weighted Feature Extraction and Fuzzy Classifier +
e4391993f5270bdbc621b8d01702f626fba36fc2Author manuscript, published in "18th Scandinavian Conference on Image Analysis (2013)" +
DOI : 10.1007/978-3-642-38886-6_31 +
e4d8ba577cabcb67b4e9e1260573aea708574886UM SISTEMA DE RECOMENDAC¸ ˜AO INTELIGENTE BASEADO EM V´IDIO +
AULAS PARA EDUCAC¸ ˜AO A DIST ˆANCIA +
Gaspare Giuliano Elias Bruno +
Tese de Doutorado apresentada ao Programa +
de P´os-gradua¸c˜ao em Engenharia de Sistemas e +
Computa¸c˜ao, COPPE, da Universidade Federal +
do Rio de Janeiro, como parte dos requisitos +
necess´arios `a obten¸c˜ao do t´ıtulo de Doutor em +
Engenharia de Sistemas e Computa¸c˜ao. +
Orientadores: Edmundo Albuquerque de +
Souza e Silva +
Rosa Maria Meri Le˜ao +
Rio de Janeiro +
Janeiro de 2016 +
e475deadd1e284428b5e6efd8fe0e6a5b83b9dcdAccepted in Pattern Recognition Letters +
Pattern Recognition Letters +
journal homepage: www.elsevier.com +
Are you eligible? Predicting adulthood from face images via class specific mean +
autoencoder +
IIIT-Delhi, New Delhi, 110020, India +
Article history: +
Received 15 March 2017 +
e4d0e87d0bd6ead4ccd39fc5b6c62287560bac5bImplicit Video Multi-Emotion Tagging by Exploiting Multi-Expression +
Relations +
fe9c460d5ca625402aa4d6dd308d15a40e1010faNeural Architecture for Temporal Emotion +
Classification +
Universit¨at Ulm, Neuroinformatik, Germany +
fe7c0bafbd9a28087e0169259816fca46db1a837
fe48f0e43dbdeeaf4a03b3837e27f6705783e576
fea83550a21f4b41057b031ac338170bacda8805Learning a Metric Embedding +
for Face Recognition +
using the Multibatch Method +
Orcam Ltd., Jerusalem, Israel +
feeb0fd0e254f38b38fe5c1022e84aa43d63f7ccEURECOM +
Multimedia Communications Department +
and +
Mobile Communications Department +
2229, route des Crˆetes +
B.P. 193 +
06904 Sophia-Antipolis +
FRANCE +
Research Report RR-11-255 +
Search Pruning with Soft Biometric Systems: +
Efficiency-Reliability Tradeoff +
June 1st, 2011 +
Last update June 1st, 2011 +
1EURECOM’s research is partially supported by its industrial members: BMW Group, Cisco, +
Monaco Telecom, Orange, SAP, SFR, Sharp, STEricsson, Swisscom, Symantec, Thales. +
fe108803ee97badfa2a4abb80f27fa86afd9aad9
fe0c51fd41cb2d5afa1bc1900bbbadb38a0de139Rahman et al. EURASIP Journal on Image and Video Processing (2015) 2015:35 +
DOI 10.1186/s13640-015-0090-5 +
RESEARCH +
Open Access +
Bayesian face recognition using 2D +
Gaussian-Hermite moments +
c8db8764f9d8f5d44e739bbcb663fbfc0a40fb3dModeling for part-based visual object +
detection based on local features +
Von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +
der Rheinisch-Westf¨alischen Technischen Hochschule Aachen +
zur Erlangung des akademischen Grades eines Doktors +
der Ingenieurwissenschaften genehmigte Dissertation +
vorgelegt von +
Diplom-Ingenieur +
aus Neuss +
Berichter: +
Univ.-Prof. Dr.-Ing. Jens-Rainer Ohm +
Univ.-Prof. Dr.-Ing. Til Aach +
Tag der m¨undlichen Pr¨ufung: 28. September 2011 +
Diese Dissertation ist auf den Internetseiten der +
Hochschulbibliothek online verf¨ugbar. +
c86e6ed734d3aa967deae00df003557b6e937d3dGenerative Adversarial Networks with +
Decoder-Encoder Output Noise +
conditional distribution of their neighbors. In [32], Portilla and +
Simoncelli proposed a parametric texture model based on joint +
statistics, which uses a decomposition method that is called +
steerable pyramid decomposition to decompose the texture +
of images. An example-based super-resolution algorithm [11] +
was proposed in 2002, which uses a Markov network to model +
the spatial relationship between the pixels of an image. A +
scene completion algorithm [16] was proposed in 2007, which +
applied a semantic scene match technique. These traditional +
algorithms can be applied to particular image generation tasks, +
such as texture synthesis and super-resolution. Their common +
characteristic is that they predict the images pixel by pixel +
rather than generate an image as a whole, and the basic idea +
of them is to make an interpolation according to the existing +
part of the images. Here, the problem is, given a set of images, +
can we generate totally new images with the same distribution +
of the given ones? +
c8a4b4fe5ff2ace9ab9171a9a24064b5a91207a3LOCATING FACIAL LANDMARKS WITH BINARY MAP CROSS-CORRELATIONS +
J´er´emie Nicolle +
K´evin Bailly +
Univ. Pierre & Marie Curie, ISIR - CNRS UMR 7222, F-75005, Paris - France +
c866a2afc871910e3282fd9498dce4ab20f6a332Noname manuscript No. +
(will be inserted by the editor) +
Surveillance Face Recognition Challenge +
Received: date / Accepted: date +
c82c147c4f13e79ad49ef7456473d86881428b89
c84233f854bbed17c22ba0df6048cbb1dd4d3248Exploring Locally Rigid Discriminative +
Patches for Learning Relative Attributes +
http://researchweb.iiit.ac.in/~yashaswi.verma/ +
http://www.iiit.ac.in/~jawahar/ +
CVIT +
IIIT-Hyderabad, India +
http://cvit.iiit.ac.in +
c8adbe00b5661ab9b3726d01c6842c0d72c8d997Deep Architectures for Face Attributes +
Computer Vision and Machine Learning Group, Flickr, Yahoo, +
fb4545782d9df65d484009558e1824538030bbb1
fb5280b80edcf088f9dd1da769463d48e7b08390
fba464cb8e3eff455fe80e8fb6d3547768efba2f +
International Journal of Engineering and Applied Sciences (IJEAS) +
ISSN: 2394-3661, Volume-3, Issue-2, February 2016 +
Survey Paper on Emotion Recognition +
 +
fbb2f81fc00ee0f257d4aa79bbef8cad5000ac59Reading Hidden Emotions: Spontaneous +
Micro-expression Spotting and Recognition +
fb9ad920809669c1b1455cc26dbd900d8e719e613D Gaze Estimation from Remote RGB-D Sensors +
THÈSE NO 6680 (2015) +
PRÉSENTÉE LE 9 OCTOBRE 2015 +
À LA FACULTÉ DES SCIENCES ET TECHNIQUES DE L'INGÉNIEUR +
LABORATOIRE DE L'IDIAP +
PROGRAMME DOCTORAL EN GÉNIE ÉLECTRIQUE +
ÉCOLE POLYTECHNIQUE FÉDÉRALE DE LAUSANNE +
POUR L'OBTENTION DU GRADE DE DOCTEUR ÈS SCIENCES +
PAR +
acceptée sur proposition du jury: +
Prof. K. Aminian, président du jury +
Dr J.-M. Odobez, directeur de thèse +
Prof. L.-Ph. Morency, rapporteur +
Prof. D. Witzner Hansen, rapporteur +
Dr R. Boulic, rapporteur +
Suisse +
2015 +
edef98d2b021464576d8d28690d29f5431fd5828Pixel-Level Alignment of Facial Images +
for High Accuracy Recognition +
Using Ensemble of Patches +
ed04e161c953d345bcf5b910991d7566f7c486f7Combining facial expression analysis and synthesis on a +
Mirror my emotions! +
robot +
c178a86f4c120eca3850a4915134fff44cbccb48
c1d2d12ade031d57f8d6a0333cbe8a772d752e01Journal of Math-for-Industry, Vol.2(2010B-5), pp.147–156 +
Convex optimization techniques for the efficient recovery of a sparsely +
corrupted low-rank matrix +
D 案 +
Received on August 10, 2010 / Revised on August 31, 2010 +
E 案 +
c10a15e52c85654db9c9343ae1dd892a2ac4a279Int J Comput Vis (2012) 100:134–153 +
DOI 10.1007/s11263-011-0494-3 +
Learning the Relative Importance of Objects from Tagged Images +
for Retrieval and Cross-Modal Search +
Received: 16 December 2010 / Accepted: 23 August 2011 / Published online: 18 October 2011 +
© Springer Science+Business Media, LLC 2011 +
c1fc70e0952f6a7587b84bf3366d2e57fc572fd7
c1dfabe36a4db26bf378417985a6aacb0f769735Journal of Computer Vision and Image Processing, NWPJ-201109-50 +
1 +
Describing Visual Scene through EigenMaps +
 +
c1482491f553726a8349337351692627a04d5dbe
c1ff88493721af1940df0d00bcfeefaa14f1711fCVPR +
#1369 +
000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
045 +
046 +
047 +
048 +
049 +
050 +
051 +
052 +
053 +
CVPR 2010 Submission #1369. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +
CVPR +
#1369 +
Subspace Regression: Predicting a Subspace from one Sample +
Anonymous CVPR submission +
Paper ID 1369 +
c11eb653746afa8148dc9153780a4584ea529d28Global and Local Consistent Wavelet-domain Age +
Synthesis +
c1ebbdb47cb6a0ed49c4d1cf39d7565060e6a7eeRobust Facial Landmark Localization Based on +
c17a332e59f03b77921942d487b4b102b1ee73b6Learning an appearance-based gaze estimator +
from one million synthesised images +
Tadas Baltruˇsaitis2 +
c1e76c6b643b287f621135ee0c27a9c481a99054
c6f3399edb73cfba1248aec964630c8d54a9c534A Comparison of CNN-based Face and Head Detectors for +
Real-Time Video Surveillance Applications +
1 ´Ecole de technologie sup´erieure, Universit´e du Qu´ebec, Montreal, Canada +
2 Genetec Inc., Montreal, Canada +
c62c07de196e95eaaf614fb150a4fa4ce49588b4Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18) +
1078 +
ec1e03ec72186224b93b2611ff873656ed4d2f74JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
3D Reconstruction of “In-the-Wild” Faces in +
Images and Videos +
ec22eaa00f41a7f8e45ed833812d1ac44ee1174e
ec54000c6c0e660dd99051bdbd7aed2988e27ab8TWO IN ONE: JOINT POSE ESTIMATION AND FACE RECOGNITION WITH P2CA1 +
*Dept. Teoria del Senyal i Comunicacions - Universitat Politècnica de Catalunya, Barcelona, Spain +
+Dipartimento di Elettronica e Informazione - Politecnico di Milano, Meiland, Italy +
ec0104286c96707f57df26b4f0a4f49b774c486b758 +
An Ensemble CNN2ELM for Age Estimation +
4e32fbb58154e878dd2fd4b06398f85636fd0cf4A Hierarchical Matcher using Local Classifier Chains +
L. Zhang and I.A. Kakadiaris +
Computational Biomedicine Lab, 4849 Calhoun Rd, Rm 373, Houston, TX 77204 +
4e27fec1703408d524d6b7ed805cdb6cba6ca132SSD-Sface: Single shot multibox detector for small faces +
C. Thuis +
4e6c9be0b646d60390fe3f72ce5aeb0136222a10Long-term Temporal Convolutions +
for Action Recognition +
4e444db884b5272f3a41e4b68dc0d453d4ec1f4c
4ef0a6817a7736c5641dc52cbc62737e2e063420International Journal of Advanced Computer Research (ISSN (Print): 2249-7277 ISSN (Online): 2277-7970) +
Volume-4 Number-4 Issue-17 December-2014 +
Study of Face Recognition Techniques +
Received: 10-November-2014; Revised: 18-December-2014; Accepted: 23-December-2014 +
©2014 ACCENTS +
4e7ebf3c4c0c4ecc48348a769dd6ae1ebac3bf1b
4e0e49c280acbff8ae394b2443fcff1afb9bdce6Automatic learning of gait signatures for people identification +
F.M. Castro +
Univ. of Malaga +
fcastrouma.es +
M.J. Mar´ın-Jim´enez +
Univ. of Cordoba +
mjmarinuco.es +
N. Guil +
Univ. of Malaga +
nguiluma.es +
N. P´erez de la Blanca +
Univ. of Granada +
nicolasugr.es +
4e4e8fc9bbee816e5c751d13f0d9218380d74b8f
20a88cc454a03d62c3368aa1f5bdffa73523827b
20a432a065a06f088d96965f43d0055675f0a6c1In: Proc. of the 25th Int. Conference on Artificial Neural Networks (ICANN) +
Part II, LNCS 9887, pp. 80-87, Barcelona, Spain, September 2016 +
The final publication is available at Springer via +
http://dx.doi.org//10.1007/978-3-319-44781-0_10 +
The Effects of Regularization on Learning Facial +
Expressions with Convolutional Neural Networks +

Vogt-Koelln-Strasse 30, 22527 Hamburg, Germany +
http://www.informatik.uni-hamburg.de/WTM +
20e504782951e0c2979d9aec88c76334f7505393Robust LSTM-Autoencoders for Face De-Occlusion +
in the Wild +
20ade100a320cc761c23971d2734388bfe79f7c5IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +
Subspace Clustering via Good Neighbors +
20767ca3b932cbc7b8112db21980d7b9b3ea43a3
20c2a5166206e7ffbb11a23387b9c5edf42b5230
2098983dd521e78746b3b3fa35a22eb2fa630299
206e24f7d4b3943b35b069ae2d028143fcbd0704Learning Structure and Strength of CNN Filters for Small Sample Size Training +
IIIT-Delhi, India +
2059d2fecfa61ddc648be61c0cbc9bc1ad8a9f5bTRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING, VOL. 23, NO. 4, APRIL 2015 +
Co-Localization of Audio Sources in Images Using +
Binaural Features and Locally-Linear Regression +
∗ INRIA Grenoble Rhˆone-Alpes, Montbonnot Saint-Martin, France +
† Univ. Grenoble Alpes, GIPSA-Lab, France +
‡ Dept. Electrical Eng., Technion-Israel Inst. of Technology, Haifa, Israel +
206fbe6ab6a83175a0ef6b44837743f8d5f9b7e8
20111924fbf616a13d37823cd8712a9c6b458cd6International Journal of Computer Applications (0975 – 8887) +
Volume 130 – No.11, November2015 +
Linear Regression Line based Partial Face Recognition +
Naveena M. +
Department of Studies in +
Computer Science, +
Manasagagothri, +
Mysore. +
Department of Studies in +
Computer Science, +
Manasagagothri, +
Mysore. +
P. Nagabhushan +
Department of Studies in +
Computer Science, +
Manasagagothri, +
Mysore. +
images. In +
20532b1f80b509f2332b6cfc0126c0f80f438f10A deep matrix factorization method for learning +
attribute representations +
Bj¨orn W. Schuller, Senior member, IEEE +
205af28b4fcd6b569d0241bb6b255edb325965a4Intel Serv Robotics (2008) 1:143–157 +
DOI 10.1007/s11370-007-0014-z +
SPECIAL ISSUE +
Facial expression recognition and tracking for intelligent human-robot +
interaction +
Received: 27 June 2007 / Accepted: 6 December 2007 / Published online: 23 January 2008 +
© Springer-Verlag 2008 +
20a0b23741824a17c577376fdd0cf40101af5880Learning to track for spatio-temporal action localization +
Zaid Harchaouia,b +
b NYU +
a Inria∗ +
18c72175ddbb7d5956d180b65a96005c100f6014IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 23, NO. 6, +
JUNE 2001 +
643 +
From Few to Many: Illumination Cone +
Models for Face Recognition under +
Variable Lighting and Pose +
18636347b8741d321980e8f91a44ee054b051574978-1-4244-5654-3/09/$26.00 ©2009 IEEE +
37 +
ICIP 2009 +
18206e1b988389eaab86ef8c852662accf3c3663
181045164df86c72923906aed93d7f2f987bce6cRHEINISCH-WESTFÄLISCHE TECHNISCHE +
HOCHSCHULE AACHEN +
KNOWLEDGE-BASED SYSTEMS GROUP +
Detection and Recognition of Human +
Faces using Random Forests for a +
Mobile Robot +
MASTER OF SCIENCE THESIS +
MATRICULATION NUMBER: 26 86 51 +
SUPERVISOR: +
SECOND SUPERVISOR: +
PROF. ENRICO BLANZIERI, PH. D. +
ADVISERS: +
18d5b0d421332c9321920b07e0e8ac4a240e5f1fCollaborative Representation Classification +
Ensemble for Face Recognition +
18d51a366ce2b2068e061721f43cb798177b4bb7Cognition and Emotion +
ISSN: 0269-9931 (Print) 1464-0600 (Online) Journal homepage: http://www.tandfonline.com/loi/pcem20 +
Looking into your eyes: observed pupil size +
influences approach-avoidance responses +
eyes: observed pupil size influences approach-avoidance responses, Cognition and Emotion, DOI: +
10.1080/02699931.2018.1472554 +
To link to this article: https://doi.org/10.1080/02699931.2018.1472554 +
View supplementary material +
Published online: 11 May 2018. +
Submit your article to this journal +
View related articles +
View Crossmark data +
Full Terms & Conditions of access and use can be found at +
http://www.tandfonline.com/action/journalInformation?journalCode=pcem20 +
1885acea0d24e7b953485f78ec57b2f04e946eafCombining Local and Global Features for 3D Face Tracking +
Megvii (face++) Research +
184750382fe9b722e78d22a543e852a6290b3f70
18a849b1f336e3c3b7c0ee311c9ccde582d7214fInt J Comput Vis +
DOI 10.1007/s11263-012-0564-1 +
Efficiently Scaling up Crowdsourced Video Annotation +
A Set of Best Practices for High Quality, Economical Video Labeling +
Received: 31 October 2011 / Accepted: 20 August 2012 +
© Springer Science+Business Media, LLC 2012 +
1886b6d9c303135c5fbdc33e5f401e7fc4da6da4Knowledge Guided Disambiguation for Large-Scale +
Scene Classification with Multi-Resolution CNNs +
1888bf50fd140767352158c0ad5748b501563833PA R T 1 +
THE BASICS +
185360fe1d024a3313042805ee201a75eac50131299 +
Person De-Identification in Videos +
18dfc2434a95f149a6cbb583cca69a98c9de9887
27d709f7b67204e1e5e05fe2cfac629afa21699d
275b5091c50509cc8861e792e084ce07aa906549Institut für Informatik +
der Technischen +
Universität München +
Dissertation +
Leveraging the User’s Face as a Known Object +
in Handheld Augmented Reality +
Sebastian Bernhard Knorr +
270733d986a1eb72efda847b4b55bc6ba9686df4We are IntechOpen, +
the first native scientific +
publisher of Open Access books +
3,350 +
108,000 +
1.7 M +
Open access books available +
International authors and editors +
Downloads +
Our authors are among the +
151 +
Countries delivered to +
TOP 1% +
12.2% +
most cited scientists +
Contributors from top 500 universities +
Selection of our books indexed in the Book Citation Index +
in Web of Science™ Core Collection (BKCI) +
Interested in publishing with us? +
Numbers displayed above are based on latest data collected. +
For more information visit www.intechopen.com +
27da432cf2b9129dce256e5bf7f2f18953eef5a5
2770b095613d4395045942dc60e6c560e882f887GridFace: Face Rectification via Learning Local +
Homography Transformations +
Face++, Megvii Inc. +
27cccf992f54966feb2ab4831fab628334c742d8International Journal of Computer Applications (0975 – 8887) +
Volume 64– No.18, February 2013 +
Facial Expression Recognition by Statistical, Spatial +
Features and using Decision Tree +
Assistant Professor +
CSIT Department +
GGV BIlaspur, Chhattisgarh +
India +
Assistant Professor +
Electronics (ECE) Department +
JECRC Jaipur, Rajasthan India +
IshanBhardwaj +
Student of Ph.D. +
Electrical Department +
NIT Raipur, Chhattisgarh India +
27f8b01e628f20ebfcb58d14ea40573d351bbaadDEPARTMENT OF INFORMATION ENGINEERING AND COMPUTER SCIENCE +
ICT International Doctoral School +
Events based Multimedia Indexing +
and Retrieval +
SUBMITTED TO THE DEPARTMENT OF +
INFORMATION ENGINEERING AND COMPUTER SCIENCE (DISI) +
IN THE PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE +
OF +
DOCTOR OF PHILOSOPHY +
Advisor: +
Examiners: Prof. Marco Carli, Universit`a degli Studi di Roma Tre, Italy +
Prof. Nicola Conci, Universit`a degli Studi di Trento, Italy +
Prof. Pietro Zanuttigh, Universit`a degli Studi di Padova, Italy +
Prof. Giulia Boato, Universit`a degli Studi di Trento, Italy +
December 2017 +
274f87ad659cd90382ef38f7c6fafc4fc7f0d74d
27ee8482c376ef282d5eb2e673ab042f5ded99d7Scale Normalization for the Distance Maps AAM. +
Avenue de la boulaie, BP 81127, +
35 511 Cesson-S´evign´e, France +
Sup´elec, IETR-SCEE Team +
4b89cf7197922ee9418ae93896586c990e0d2867LATEX Author Guidelines for CVPR Proceedings +
First Author +
Institution1 +
Institution1 address +
4b04247c7f22410681b6aab053d9655cf7f3f888Robust Face Recognition by Constrained Part-based +
Alignment +
4b60e45b6803e2e155f25a2270a28be9f8bec130Attribute Based Object Identification +
4b48e912a17c79ac95d6a60afed8238c9ab9e553JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
Minimum Margin Loss for Deep Face Recognition +
4b5eeea5dd8bd69331bd4bd4c66098b125888deaHuman Activity Recognition Using Conditional +
Random Fields and Privileged Information +
submitted to +
the designated by the General Assembly Composition of the +
Department of Computer Science & Engineering Inquiry +
Committee +
by +
in partial fulfillment of the Requirements for the Degree of +
DOCTOR OF PHILOSOPHY +
February 2016 +
4bbbee93519a4254736167b31be69ee1e537f942
4b6be933057d939ddfa665501568ec4704fabb39
4be03fd3a76b07125cd39777a6875ee59d9889bdCONTENT-BASED ANALYSIS FOR ACCESSING AUDIOVISUAL ARCHIVES: +
ALTERNATIVES FOR CONCEPT-BASED INDEXING AND SEARCH +
ESAT/PSI - IBBT +
KU Leuven, Belgium +
113e5678ed8c0af2b100245057976baf82fcb907Facing Imbalanced Data +
Recommendations for the Use of Performance Metrics +
11f17191bf74c80ad0b16b9f404df6d03f7c8814Recognition of Visually Perceived Compositional +
Human Actions by Multiple Spatio-Temporal Scales +
Recurrent Neural Networks +
11367581c308f4ba6a32aac1b4a7cdb32cd63137
1198572784788a6d2c44c149886d4e42858d49e4Learning Discriminative Features using Encoder/Decoder type Deep +
Neural Nets +
11fe6d45aa2b33c2ec10d9786a71c15ec4d3dca8970 +
JUNE 2008 +
Tied Factor Analysis for Face Recognition +
across Large Pose Differences +
112780a7fe259dc7aff2170d5beda50b2bfa7bda
111a9645ad0108ad472b2f3b243ed3d942e7ff16Facial Expression Classification Using +
Combined Neural Networks +
DEE/PUC-Rio, Marquês de São Vicente 225, Rio de Janeiro – RJ - Brazil +
111d0b588f3abbbea85d50a28c0506f74161e091International Journal of Computer Applications (0975 – 8887) +
Volume 134 – No.10, January 2016 +
Facial Expression Recognition from Visual Information +
using Curvelet Transform +
Surabhi Group of Institution Bhopal +
systems. Further applications +
7de6e81d775e9cd7becbfd1bd685f4e2a5eebb22Labeled Faces in the Wild: A Survey +
7d73adcee255469aadc5e926066f71c93f51a1a5978-1-4799-9988-0/16/$31.00 ©2016 IEEE +
1283 +
ICASSP 2016 +
7dffe7498c67e9451db2d04bb8408f376ae86992LEAR-INRIA submission for the THUMOS workshop +
LEAR, INRIA, France +
7d3f6dd220bec883a44596ddec9b1f0ed4f6aca22106 +
Linear Regression for Face Recognition +
29ce6b54a87432dc8371f3761a9568eb3c5593b0Kent Academic Repository +
Full text document (pdf) +
Citation for published version +
Yassin, DK H. PHM and Hoque, Sanaul and Deravi, Farzin (2013) Age Sensitivity of Face Recognition +
pp. 12-15. +
DOI +
https://doi.org/10.1109/EST.2013.8 +
Link to record in KAR +
http://kar.kent.ac.uk/43222/ +
Document Version +
Author's Accepted Manuscript +
Copyright & reuse +
Content in the Kent Academic Repository is made available for research purposes. Unless otherwise stated all +
content is protected by copyright and in the absence of an open licence (eg Creative Commons), permissions +
for further reuse of content should be sought from the publisher, author or other copyright holder. +
Versions of research +
The version in the Kent Academic Repository may differ from the final published version. +
Users are advised to check http://kar.kent.ac.uk for the status of the paper. Users should always cite the +
published version of record. +
Enquiries +
For any further enquiries regarding the licence status of this document, please contact: +
If you believe this document infringes copyright then please contact the KAR admin team with the take-down +
information provided at http://kar.kent.ac.uk/contact.html +
292eba47ef77495d2613373642b8372d03f7062bDeep Secure Encoding: An Application to Face Recognition +
29e96ec163cb12cd5bd33bdf3d32181c136abaf9Report No. UIUCDCS-R-2006-2748 +
UILU-ENG-2006-1788 +
Regularized Locality Preserving Projections with Two-Dimensional +
Discretized Laplacian Smoothing +
by +
July 2006 +
29c1f733a80c1e07acfdd228b7bcfb136c1dff98
29f27448e8dd843e1c4d2a78e01caeaea3f46a2d
294d1fa4e1315e1cf7cc50be2370d24cc6363a412008 SPIE Digital Library -- Subscriber Archive Copy +
29d414bfde0dfb1478b2bdf67617597dd2d57fc6Multidim Syst Sign Process (2010) 21:213–229 +
DOI 10.1007/s11045-009-0099-y +
Perfect histogram matching PCA for face recognition +
Received: 10 August 2009 / Revised: 21 November 2009 / Accepted: 29 December 2009 / +
Published online: 14 January 2010 +
© Springer Science+Business Media, LLC 2010 +
290136947fd44879d914085ee51d8a4f433765faOn a Taxonomy of Facial Features +
2957715e96a18dbb5ed5c36b92050ec375214aa6Improving Face Attribute Detection with Race and Gender Diversity +
InclusiveFaceNet: +
291265db88023e92bb8c8e6390438e5da148e8f5MS-Celeb-1M: A Dataset and Benchmark for +
Large-Scale Face Recognition +
Microsoft Research +
2921719b57544cfe5d0a1614d5ae81710ba804faFace Recognition Enhancement Based on Image +
File Formats and Wavelet De-noising +
 +
29a013b2faace976f2c532533bd6ab4178ccd348This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. +
Hierarchical Manifold Learning With Applications +
to Supervised Classification for High-Resolution +
Remotely Sensed Images +
29756b6b16d7b06ea211f21cdaeacad94533e8b4Thresholding Approach based on GPU for Facial +
Expression Recognition +
1 Benemérita Universidad Autónoma de Puebla, Faculty of Computer Science, Puebla, México +
2Instituto Tecnológico de Puebla, Puebla, México +
293193d24d5c4d2975e836034bbb2329b71c4fe7Building a Corpus of Facial Expressions +
for Learning-Centered Emotions +
Instituto Tecnológico de Culiacán, Culiacán, Sinaloa, +
Mexico +
2988f24908e912259d7a34c84b0edaf7ea50e2b3A Model of Brightness Variations Due to +
Illumination Changes and Non-rigid Motion +
Using Spherical Harmonics +
Jos´e M. Buenaposada +
Dep. Ciencias de la Computaci´on, +
U. Rey Juan Carlos, Spain +
http://www.dia.fi.upm.es/~pcr +
Inst. for Systems and Robotics +
Inst. Superior T´ecnico, Portugal +
http://www.isr.ist.utl.pt/~adb +
Enrique Mu˜noz +
Facultad de Inform´atica, +
U. Complutense de Madrid, Spain +
Dep. de Inteligencia Artificial, +
U. Polit´ecnica de Madrid, Spain +
http://www.dia.fi.upm.es/~pcr +
http://www.dia.fi.upm.es/~pcr +
29156e4fe317b61cdcc87b0226e6f09e416909e0
293ade202109c7f23637589a637bdaed06dc37c9
7c7ab59a82b766929defd7146fd039b89d67e984Improving Multiview Face Detection with +
Multi-Task Deep Convolutional Neural Networks +
Microsoft Research +
One Microsoft Way, Redmond WA 98052 +
7c45b5824645ba6d96beec17ca8ecfb22dfcdd7fNews image annotation on a large parallel text-image corpus +
Universit´e de Rennes 1/IRISA, CNRS/IRISA, INRIA Rennes-Bretagne Atlantique +
Campus de Beaulieu +
35042 Rennes Cedex, France +
7c0a6824b556696ad7bdc6623d742687655852db18th Telecommunications forum TELFOR 2010 +
Serbia, Belgrade, November 23-25, 2010. +
MPCA+DATER: A Novel Approach for Face +
Recognition Based on Tensor Objects +
Ali. A. Shams Baboli, Member, IEEE, G. Rezai-rad, Member, IEEE, Aref. Shams Baboli +
7c95449a5712aac7e8c9a66d131f83a038bb7caaThis is an author produced version of Facial first impressions from another angle: How +
social judgements are influenced by changeable and invariant facial properties. +
White Rose Research Online URL for this paper: +
http://eprints.whiterose.ac.uk/102935/ +
Article: +
Rhodes (2017) Facial first impressions from another angle: How social judgements are +
influenced by changeable and invariant facial properties. British journal of psychology. pp. +
397-415. ISSN 0007-1269 +
https://doi.org/10.1111/bjop.12206 +
promoting access to +
White Rose research papers +
http://eprints.whiterose.ac.uk/ +
7c3e09e0bd992d3f4670ffacb4ec3a911141c51fNoname manuscript No. +
(will be inserted by the editor) +
Transferring Object-Scene Convolutional Neural Networks for +
Event Recognition in Still Images +
Received: date / Accepted: date +
7c7b0550ec41e97fcfc635feffe2e53624471c591051-4651/14 $31.00 © 2014 IEEE +
DOI 10.1109/ICPR.2014.124 +
660 +
7ce03597b703a3b6754d1adac5fbc98536994e8f
7c9a65f18f7feb473e993077d087d4806578214eSpringerLink - Zeitschriftenbeitrag +
http://www.springerlink.com/content/93hr862660nl1164/?p=abe5352... +
Deutsch +
Deutsch +
Go +
Vorherige Beitrag Nächste Beitrag +
Beitrag markieren +
In den Warenkorb legen +
Zu gespeicherten Artikeln +
hinzufügen +
Permissions & Reprints +
Diesen Artikel empfehlen +
Ergebnisse +
finden +
Erweiterte Suche +
Go +
im gesamten Inhalt +
in dieser Zeitschrift +
in diesem Heft +
Diesen Beitrag exportieren +
Diesen Beitrag exportieren als RIS +
| Text +
Text +
PDF +
PDF ist das gebräuchliche Format +
für Online Publikationen. Die Größe +
dieses Dokumentes beträgt 564 +
Kilobyte. Je nach Art Ihrer +
Internetverbindung kann der +
Download einige Zeit in Anspruch +
nehmen. +
öffnen: Gesamtdokument +
Publikationsart Subject Collections +
Zurück zu: Journal Issue +
Athens Authentication Point +
Zeitschriftenbeitrag +
Willkommen! +
Um unsere personalisierten +
Angebote nutzen zu können, +
müssen Sie angemeldet sein. +
Login +
Jetzt registrieren +
Zugangsdaten vergessen? +
Hilfe. +
Mein Menü +
Markierte Beiträge +
Alerts +
Meine Bestellungen +
Private emotions versus social interaction: a data-driven approach towards +
analysing emotion in speech +
Zeitschrift +
Verlag +
ISSN +
Heft +
Kategorie +
DOI +
Seiten +
Subject Collection +
SpringerLink Date +
User Modeling and User-Adapted Interaction +
Springer Netherlands +
0924-1868 (Print) 1573-1391 (Online) +
Volume 18, Numbers 1-2 / Februar 2008 +
Original Paper +
10.1007/s11257-007-9039-4 +
175-206 +
Informatik +
Freitag, 12. Oktober 2007 +
Gespeicherte Beiträge +
Alle +
Favoriten +
(1) Lehrstuhl für Mustererkennung, FAU Erlangen – Nürnberg, Martensstr. 3, 91058 Erlangen, +
Germany +
Received: 3 July 2006 Accepted: 14 January 2007 Published online: 12 October 2007 +
7c1e1c767f7911a390d49bed4f73952df8445936NON-RIGID OBJECT DETECTION WITH LOCAL INTERLEAVED SEQUENTIAL ALIGNMENT (LISA) +
Non-Rigid Object Detection with Local +
Interleaved Sequential Alignment (LISA) +
and Tom´aˇs Svoboda, Member, IEEE +
7c349932a3d083466da58ab1674129600b12b81c
1648cf24c042122af2f429641ba9599a2187d605Boosting Cross-Age Face Verification via Generative Age Normalization +
(cid:2) Orange Labs, 4 rue Clos Courtel, 35512 Cesson-S´evign´e, France +
† Eurecom, 450 route des Chappes, 06410 Biot, France +
162403e189d1b8463952fa4f18a291241275c354Action Recognition with Spatio-Temporal +
Visual Attention on Skeleton Image Sequences +
With a strong ability of modeling sequential data, Recur- +
rent Neural Networks (RNN) with Long Short-Term Memory +
(LSTM) neurons outperform the previous hand-crafted feature +
based methods [9], [10]. Each skeleton frame is converted into +
a feature vector and the whole sequence is fed into the RNN. +
Despite the strong ability in modeling temporal sequences, +
RNN structures lack the ability to efficiently learn the spatial +
relations between the joints. To better use spatial information, +
a hierarchical structure is proposed in [11], [12] that feeds +
the joints into the network as several pre-defined body part +
groups. However, +
limit +
the effectiveness of representing spatial relations. A spatio- +
temporal 2D LSTM (ST-LSTM) network [13] is proposed +
to learn the spatial and temporal relations simultaneously. +
Furthermore, a two-stream RNN structure [14] is proposed to +
learn the spatio-temporal relations with two RNN branches. +
the pre-defined body regions still +
160259f98a6ec4ec3e3557de5e6ac5fa7f2e7f2bDiscriminant Multi-Label Manifold Embedding for Facial Action Unit +
Detection +
Signal Procesing Laboratory (LTS5), ´Ecole Polytechnique F´ed´erale de Lausanne, Switzerland +
16671b2dc89367ce4ed2a9c241246a0cec9ec10e2006 +
Detecting the Number of Clusters +
in n-Way Probabilistic Clustering +
16de1324459fe8fdcdca80bba04c3c30bb789bdf
16892074764386b74b6040fe8d6946b67a246a0b
16395b40e19cbc6d5b82543039ffff2a06363845Action Recognition in Video Using Sparse Coding and Relative Features +
Anal´ı Alfaro +
P. Universidad Catolica de Chile +
P. Universidad Catolica de Chile +
P. Universidad Catolica de Chile +
Santiago, Chile +
Santiago, Chile +
Santiago, Chile +
16286fb0f14f6a7a1acc10fcd28b3ac43f12f3ebJ Nonverbal Behav +
DOI 10.1007/s10919-008-0059-5 +
O R I G I N A L P A P E R +
All Smiles are Not Created Equal: Morphology +
and Timing of Smiles Perceived as Amused, Polite, +
and Embarrassed/Nervous +
Ó Springer Science+Business Media, LLC 2008 +
166186e551b75c9b5adcc9218f0727b73f5de899Volume 4, Issue 2, February 2016 +
International Journal of Advance Research in +
Computer Science and Management Studies +
Research Article / Survey Paper / Case Study +
Available online at: www.ijarcsms.com +
ISSN: 2321-7782 (Online) +
Automatic Age and Gender Recognition in Human Face Image +
Dataset using Convolutional Neural Network System +
Subhani Shaik1 +
Assoc. Prof & Head of the Department +
Department of CSE, +
Associate Professor +
Department of CSE, +
St.Mary’s Group of Institutions Guntur +
St.Mary’s Group of Institutions Guntur +
Chebrolu(V&M),Guntur(Dt), +
Andhra Pradesh - India +
Chebrolu(V&M),Guntur(Dt), +
Andhra Pradesh - India +
16d9b983796ffcd151bdb8e75fc7eb2e31230809EUROGRAPHICS 2018 / D. Gutierrez and A. Sheffer +
(Guest Editors) +
Volume 37 (2018), Number 2 +
GazeDirector: Fully Articulated Eye Gaze Redirection in Video +
ID: paper1004 +
1679943d22d60639b4670eba86665371295f52c3
169076ffe5e7a2310e98087ef7da25aceb12b62d
161eb88031f382e6a1d630cd9a1b9c4bc6b476521 +
Automatic Facial Expression Recognition +
Using Features of Salient Facial Patches +
4209783b0cab1f22341f0600eed4512155b1dee6Accurate and Efficient Similarity Search for Large Scale Face Recognition +
BUPT +
BUPT +
BUPT +
42e3dac0df30d754c7c7dab9e1bb94990034a90dPANDA: Pose Aligned Networks for Deep Attribute Modeling +
2EECS, UC Berkeley +
1Facebook AI Research +
429c3588ce54468090cc2cf56c9b328b549a86dc
42cc9ea3da1277b1f19dff3d8007c6cbc0bb9830Coordinated Local Metric Learning +
Inria∗ +
42350e28d11e33641775bef4c7b41a2c3437e4fd212 +
Multilinear Discriminant Analysis +
for Face Recognition +
42e155ea109eae773dadf74d713485be83fca105
4270460b8bc5299bd6eaf821d5685c6442ea179aInt J Comput Vis (2009) 84: 163–183 +
DOI 10.1007/s11263-008-0147-3 +
Partial Similarity of Objects, or How to Compare a Centaur +
to a Horse +
Received: 30 September 2007 / Accepted: 3 June 2008 / Published online: 26 July 2008 +
© Springer Science+Business Media, LLC 2008 +
429d4848d03d2243cc6a1b03695406a6de1a7abdFace Recognition based on Logarithmic Fusion +
International Journal of Soft Computing and Engineering (IJSCE) +
ISSN: 2231-2307, Volume-2, Issue-3, July 2012 +
of SVD and KT +
Ramachandra A C, Raja K B, Venugopal K R, L M Patnaik +
to +
 +
424259e9e917c037208125ccc1a02f8276afb667
42ecfc3221c2e1377e6ff849afb705ecd056b6ffPose Invariant Face Recognition under Arbitrary +
Unknown Lighting using Spherical Harmonics +
Department of Computer Science, +
SUNY at Stony Brook, NY, 11790 +
421955c6d2f7a5ffafaf154a329a525e21bbd6d3570 +
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 22, NO. 6, +
JUNE 2000 +
Evolutionary Pursuit and Its +
Application to Face Recognition +
42e0127a3fd6a96048e0bc7aab6d0ae88ba00fb0
42df75080e14d32332b39ee5d91e83da8a914e344280 +
Illumination Compensation Using Oriented +
Local Histogram Equalization and +
Its Application to Face Recognition +
89945b7cd614310ebae05b8deed0533a9998d212Divide-and-Conquer Method for L1 Norm Matrix +
Factorization in the Presence of Outliers and +
Missing Data +
89de30a75d3258816c2d4d5a733d2bef894b66b9
8913a5b7ed91c5f6dec95349fbc6919deee4fc75BigBIRD: A Large-Scale 3D Database of Object Instances +
89d3a57f663976a9ac5e9cdad01267c1fc1a7e06Neural Class-Specific Regression for face +
verification +
891b10c4b3b92ca30c9b93170ec9abd71f6099c4Facial landmark detection using structured output deep +
neural networks +
Soufiane Belharbi ∗1, Cl´ement Chatelain∗1, Romain H´erault∗1, and S´ebastien +
1LITIS EA 4108, INSA de Rouen, Saint ´Etienne du Rouvray 76800, France +
2LITIS EA 4108, UFR des Sciences, Universit´e de Rouen, France. +
September 24, 2015 +
45c340c8e79077a5340387cfff8ed7615efa20fd
45e7ddd5248977ba8ec61be111db912a4387d62fCHEN ET AL.: ADVERSARIAL POSENET +
Adversarial Learning of Structure-Aware Fully +
Convolutional Networks for Landmark +
Localization +
45f3bf505f1ce9cc600c867b1fb2aa5edd5feed8
4560491820e0ee49736aea9b81d57c3939a69e12Investigating the Impact of Data Volume and +
Domain Similarity on Transfer Learning +
Applications +
State Farm Insurance, Bloomington IL 61710, USA, +
4571626d4d71c0d11928eb99a3c8b10955a74afeGeometry Guided Adversarial Facial Expression Synthesis +
1National Laboratory of Pattern Recognition, CASIA +
2Center for Research on Intelligent Perception and Computing, CASIA +
3Center for Excellence in Brain Science and Intelligence Technology, CAS +
4534d78f8beb8aad409f7bfcd857ec7f19247715Under review as a conference paper at ICLR 2017 +
TRANSFORMATION-BASED MODELS OF VIDEO +
SEQUENCES +
Facebook AI Research +
459e840ec58ef5ffcee60f49a94424eb503e8982One-shot Face Recognition by Promoting Underrepresented Classes +
Microsoft +
One Microsoft Way, Redmond, Washington, United States +
45fbeed124a8956477dbfc862c758a2ee2681278
451c42da244edcb1088e3c09d0f14c064ed9077e1964 +
© EURASIP, 2011 - ISSN 2076-1465 +
19th European Signal Processing Conference (EUSIPCO 2011) +
INTRODUCTION +
4511e09ee26044cb46073a8c2f6e1e0fbabe33e8
45a6333fc701d14aab19f9e2efd59fe7b0e89fecHAND POSTURE DATASET CREATION FOR GESTURE +
RECOGNITION +
Luis Anton-Canalis +
Instituto de Sistemas Inteligentes y Aplicaciones Numericas en Ingenieria +
Campus Universitario de Tafira, 35017 Gran Canaria, Spain +
Elena Sanchez-Nielsen +
Departamento de E.I.O. y Computacion +
38271 Universidad de La Laguna, Spain +
Keywords: +
Image understanding, Gesture recognition, Hand dataset. +
1ffe20eb32dbc4fa85ac7844178937bba97f4bf0Face Clustering: Representation and Pairwise +
Constraints +
1f8304f4b51033d2671147b33bb4e51b9a1e16feNoname manuscript No. +
(will be inserted by the editor) +
Beyond Trees: +
MAP Inference in MRFs via Outer-Planar Decomposition +
Received: date / Accepted: date +
1f9ae272bb4151817866511bd970bffb22981a49An Iterative Regression Approach for Face Pose Estima- +
tion from RGB Images +
This paper presents a iterative optimization method, explicit shape regression, for face pose +
detection and localization. The regression function is learnt to find out the entire facial shape +
and minimize the alignment errors. A cascaded learning framework is employed to enhance +
shape constraint during detection. A combination of a two-level boosted regression, shape +
performance. In this paper, we have explain the advantage of ESR for deformable object like +
face pose estimation and reveal its generic applications of the method. In the experiment, +
we compare the results with different work and demonstrate the accuracy and robustness in +
different scenarios. +
Introduction +
Pose estimation is an important problem in computer vision, and has enabled many practical ap- +
plication from face expression 1 to activity tracking 2. Researchers design a new algorithm called +
explicit shape regression (ESR) to find out face alignment from a picture 3. Figure 1 shows how +
the system uses ESR to learn a shape of a human face image. A simple way to identify a face is to +
find out facial landmarks like eyes, nose, mouth and chin. The researchers define a face shape S +
and S is composed of Nf p facial landmarks. Therefore, they get S = [x1, y1, ..., xNf p, yNf p]T . The +
objective of the researchers is to estimate a shape S of a face image. The way to know the accuracy +
1fc249ec69b3e23856b42a4e591c59ac60d77118Evaluation of a 3D-aided Pose Invariant 2D Face Recognition System +
Computational Biomedicine Lab +
4800 Calhoun Rd. Houston, TX, USA +
1fbde67e87890e5d45864e66edb86136fbdbe20eThe Action Similarity Labeling Challenge +
1f41a96589c5b5cee4a55fc7c2ce33e1854b09d6Demographic Estimation from Face Images: +
Human vs. Machine Performance +
1fd2ed45fb3ba77f10c83f0eef3b66955645dfe0
1f2d12531a1421bafafe71b3ad53cb080917b1a7
1fefb2f8dd1efcdb57d5c2966d81f9ab22c1c58dvExplorer: A Search Method to Find Relevant YouTube Videos for Health +
Researchers +
IBM Research, Cambridge, MA, USA +
1f94734847c15fa1da68d4222973950d6b683c9eEmbedding Label Structures for Fine-Grained Feature Representation +
UNC Charlotte +
Charlotte, NC 28223 +
NEC Lab America +
Cupertino, CA 95014 +
NEC Lab America +
Cupertino, CA 95014 +
UNC Charlotte +
Charlotte, NC 28223 +
1f745215cda3a9f00a65166bd744e4ec35644b02Facial Cosmetics Database and Impact Analysis on +
Automatic Face Recognition +
# Computer Science Department, TU Muenchen +
Boltzmannstr. 3, 85748 Garching b. Muenchen, Germany +
∗ Multimedia Communications Department, EURECOM +
450 Route des Chappes, 06410 Biot, France +
1fff309330f85146134e49e0022ac61ac60506a9Data-Driven Sparse Sensor Placement for Reconstruction +
7323b594d3a8508f809e276aa2d224c4e7ec5a80JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
An Experimental Evaluation of Covariates +
Effects on Unconstrained Face Verification +
732e8d8f5717f8802426e1b9debc18a8361c1782Unimodal Probability Distributions for Deep Ordinal Classification +
73ed64803d6f2c49f01cffef8e6be8fc9b5273b8Noname manuscript No. +
(will be inserted by the editor) +
Cooking in the kitchen: Recognizing and Segmenting Human +
Activities in Videos +
Received: date / Accepted: date +
7306d42ca158d40436cc5167e651d7ebfa6b89c1Noname manuscript No. +
(will be inserted by the editor) +
Transductive Zero-Shot Action Recognition by +
Word-Vector Embedding +
Received: date / Accepted: date +
734cdda4a4de2a635404e4c6b61f1b2edb3f501dTie and Guan EURASIP Journal on Image and Video Processing 2013, 2013:8 +
http://jivp.eurasipjournals.com/content/2013/1/8 +
R ES EAR CH +
Open Access +
Automatic landmark point detection and tracking +
for human facial expressions +
732686d799d760ccca8ad47b49a8308b1ab381fbRunning head: TEACHERS’ DIFFERING BEHAVIORS +
1 +
Graduate School of Psychology +
RESEARCH MASTER’S PSYCHOLOGY THESıS REPORT +
+
Teachers’ differing classroom behaviors: +
The role of emotional sensitivity and cultural tolerance +
Research Master’s, Social Psychology +
Ethics Committee Reference Code: 2016-SP-7084 +
73fbdd57270b9f91f2e24989178e264f2d2eb7ae978-1-4673-0046-9/12/$26.00 ©2012 IEEE +
1945 +
ICASSP 2012 +
73c9cbbf3f9cea1bc7dce98fce429bf0616a1a8c
871f5f1114949e3ddb1bca0982086cc806ce84a8Discriminative Learning of Apparel Features +
1 Computer Vision Laboratory, D-ITET, ETH Z¨urich, Switzerland +
2 ESAT - PSI / IBBT, K.U. Leuven, Belgium +
878169be6e2c87df2d8a1266e9e37de63b524ae7CBMM Memo No. 089 +
May 10, 2018 +
Image interpretation above and below the object level +
878301453e3d5cb1a1f7828002ea00f59cbeab06Faceness-Net: Face Detection through +
Deep Facial Part Responses +
87e592ee1a7e2d34e6b115da08700a1ae02e9355Deep Pictorial Gaze Estimation +
AIT Lab, Department of Computer Science, ETH Zurich +
87bb183d8be0c2b4cfceb9ee158fee4bbf3e19fdCraniofacial Image Analysis +
8006219efb6ab76754616b0e8b7778dcfb46603dCONTRIBUTIONSTOLARGE-SCALELEARNINGFORIMAGECLASSIFICATIONZeynepAkataPhDThesisl’´EcoleDoctoraleMath´ematiques,SciencesetTechnologiesdel’Information,InformatiquedeGrenoble
80193dd633513c2d756c3f568ffa0ebc1bb5213e
804b4c1b553d9d7bae70d55bf8767c603c1a09e3978-1-4799-9988-0/16/$31.00 ©2016 IEEE +
1831 +
ICASSP 2016 +
800cbbe16be0f7cb921842d54967c9a94eaa2a65MULTIMODAL RECOGNITION OF +
EMOTIONS +
803c92a3f0815dbf97e30c4ee9450fd005586e1aMax-Mahalanobis Linear Discriminant Analysis Networks +
80345fbb6bb6bcc5ab1a7adcc7979a0262b8a923Research Article +
Soft Biometrics for a Socially Assistive Robotic +
Platform +
Open Access +
80a6bb337b8fdc17bffb8038f3b1467d01204375Proceedings of the International Conference on Computer and Information Science and Technology +
Ottawa, Ontario, Canada, May 11 – 12, 2015 +
Paper No. 126 +
Subspace LDA Methods for Solving the Small Sample Size +
Problem in Face Recognition +

101 KwanFu Rd., Sec. 2, Hsinchu, Taiwan +
80097a879fceff2a9a955bf7613b0d3bfa68dc23Active Self-Paced Learning for Cost-Effective and +
Progressive Face Identification +
74408cfd748ad5553cba8ab64e5f83da14875ae8Facial Expressions Tracking and Recognition: Database Protocols for Systems Validation +
and Evaluation +
747d5fe667519acea1bee3df5cf94d9d6f874f20
74dbe6e0486e417a108923295c80551b6d759dbeInternational Journal of Computer Applications (0975 – 8887) +
Volume 45– No.11, May 2012 +
An HMM based Model for Prediction of Emotional +
Composition of a Facial Expression using both +
Significant and Insignificant Action Units and +
Associated Gender Differences +
Department of Management and Information +
Department of Management and Information +
Systems Science +
1603-1 Kamitomioka, Nagaoka +
Niigata, Japan +
Systems Science +
1603-1 Kamitomioka, Nagaoka +
Niigata, Japan +
747c25bff37b96def96dc039cc13f8a7f42dbbc7EmoNets: Multimodal deep learning approaches for emotion +
recognition in video +
74b0095944c6e29837c208307a67116ebe1231c8
74156a11c2997517061df5629be78428e1f09cbdCancún Center, Cancún, México, December 4-8, 2016 +
978-1-5090-4846-5/16/$31.00 ©2016 IEEE +
2784 +
745b42050a68a294e9300228e09b5748d2d20b81
749d605dd12a4af58de1fae6f5ef5e65eb06540eMulti-Task Video Captioning with Video and Entailment Generation +
UNC Chapel Hill +
74c19438c78a136677a7cb9004c53684a4ae56ffRESOUND: Towards Action Recognition +
without Representation Bias +
UC San Diego +
7480d8739eb7ab97c12c14e75658e5444b852e9fNEGREL ET AL.: REVISITED MLBOOST FOR FACE RETRIEVAL +
MLBoost Revisited: A Faster Metric +
Learning Algorithm for Identity-Based Face +
Retrieval +
Frederic Jurie +
Normandie Univ, UNICAEN, +
ENSICAEN, CNRS +
France +
74ba4ab407b90592ffdf884a20e10006d2223015Partial Face Detection in the Mobile Domain +
7405ed035d1a4b9787b78e5566340a98fe4b63a0Self-Expressive Decompositions for +
Matrix Approximation and Clustering +
744db9bd550bf5e109d44c2edabffec28c867b91FX e-Makeup for Muscle Based Interaction +
1 Department of Informatics, PUC-Rio, Rio de Janeiro, Brazil +
2 Department of Mechanical Engineering, PUC-Rio, Rio de Janeiro, Brazil +
3 Department of Administration, PUC-Rio, Rio de Janeiro, Brazil +
744d23991a2c48d146781405e299e9b3cc14b731This 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 +
Transactions on Image Processing +
Aging Face Recognition: A Hierarchical Learning +
Model Based on Local Patterns Selection +
1a45ddaf43bcd49d261abb4a27977a952b5fff12LDOP: Local Directional Order Pattern for Robust +
Face Retrieval +
 +
1aa766bbd49bac8484e2545c20788d0f86e73ec2 +
Baseline Face Detection, Head Pose Estimation, and Coarse +
Direction Detection for Facial Data in the SHRP2 Naturalistic +
Driving Study +
J. Paone, D. Bolme, R. Ferrell, Member, IEEE, D. Aykac, and +
T. Karnowski, Member, IEEE +
Oak Ridge National Laboratory, Oak Ridge, TN +
1a849b694f2d68c3536ed849ed78c82e979d64d5This is a repository copy of Symmetric Shape Morphing for 3D Face and Head Modelling. +
White Rose Research Online URL for this paper: +
http://eprints.whiterose.ac.uk/131760/ +
Version: Accepted Version +
Proceedings Paper: +
Dai, Hang, Pears, Nicholas Edwin orcid.org/0000-0001-9513-5634, Smith, William Alfred +
Peter orcid.org/0000-0002-6047-0413 et al. (1 more author) (2018) Symmetric Shape +
Morphing for 3D Face and Head Modelling. In: The 13th IEEE Conference on Automatic +
Face and Gesture Recognition. IEEE . +
Reuse +
Items deposited in White Rose Research Online are protected by copyright, with all rights reserved unless +
indicated otherwise. They may be downloaded and/or printed for private study, or other acts as permitted by +
national copyright laws. The publisher or other rights holders may allow further reproduction and re-use of +
the full text version. This is indicated by the licence information on the White Rose Research Online record +
for the item. +
Takedown +
If you consider content in White Rose Research Online to be in breach of UK law, please notify us by +
https://eprints.whiterose.ac.uk/ +
1a3eee980a2252bb092666cf15dd1301fa84860ePCA GAUSSIANIZATION FOR IMAGE PROCESSING +
Image Processing Laboratory (IPL), Universitat de Val`encia +
Catedr´atico A. Escardino - 46980 Paterna, Val`encia, Spain +
1a031378cf1d2b9088a200d9715d87db8a1bf041Workshop track - ICLR 2018 +
DEEP DICTIONARY LEARNING: SYNERGIZING RE- +
CONSTRUCTION AND CLASSIFICATION +
1a9337d70a87d0e30966ecd1d7a9b0bbc7be161f
1a9a192b700c080c7887e5862c1ec578012f9ed1IEEE TRANSACTIONS ON SYSTEM, MAN AND CYBERNETICS, PART B +
Discriminant Subspace Analysis for Face +
Recognition with Small Number of Training +
Samples +
1a8ccc23ed73db64748e31c61c69fe23c48a2bb1Extensive Facial Landmark Localization +
with Coarse-to-fine Convolutional Network Cascade +
Megvii Inc. +
1ad97cce5fa8e9c2e001f53f6f3202bddcefba22Grassmann Averages for Scalable Robust PCA +
DIKU and MPIs T¨ubingen∗ +
Denmark and Germany +
DTU Compute∗ +
Lyngby, Denmark +
1a1118cd4339553ad0544a0a131512aee50cf7de
1a7a2221fed183b6431e29a014539e45d95f0804Person Identification Using Text and Image Data +
David S. Bolme, J. Ross Beveridge and Adele E. Howe +
Computer Science Department +
Colorado State Univeristy +
Fort Collins, Colorado 80523 +
28e0ed749ebe7eb778cb13853c1456cb6817a166
28b9d92baea72ec665c54d9d32743cf7bc0912a7
28d7029cfb73bcb4ad1997f3779c183972a406b4Discriminative Nonlinear Analysis Operator +
Learning: When Cosparse Model Meets Image +
Classification +
280d59fa99ead5929ebcde85407bba34b1fcfb59978-1-4799-9988-0/16/$31.00 ©2016 IEEE +
2662 +
ICASSP 2016 +
28cd46a078e8fad370b1aba34762a874374513a5CVPAPER.CHALLENGE IN 2016, JULY 2017 +
cvpaper.challenge in 2016: Futuristic Computer +
Vision through 1,600 Papers Survey +
282a3ee79a08486f0619caf0ada210f5c3572367
288dbc40c027af002298b38954d648fddd4e2fd3
28312c3a47c1be3a67365700744d3d6665b86f22
28b5b5f20ad584e560cd9fb4d81b0a22279b2e7bA New Fuzzy Stacked Generalization Technique +
and Analysis of its Performance +
28bc378a6b76142df8762cd3f80f737ca2b79208Understanding Objects in Detail with Fine-grained Attributes +
Ross Girshick5 +
David Weiss7 +
287900f41dd880802aa57f602e4094a8a9e5ae56
28d4e027c7e90b51b7d8908fce68128d1964668a
2866cbeb25551257683cf28f33d829932be651feIn Proceedings of the 2018 IEEE International Conference on Image Processing (ICIP) +
The final publication is available at: http://dx.doi.org/10.1109/ICIP.2018.8451026 +
A TWO-STEP LEARNING METHOD FOR DETECTING LANDMARKS +
ON FACES FROM DIFFERENT DOMAINS +
Erickson R. Nascimento +
Universidade Federal de Minas Gerais (UFMG), Brazil +
28aa89b2c827e5dd65969a5930a0520fdd4a3dc7
28b061b5c7f88f48ca5839bc8f1c1bdb1e6adc68Predicting User Annoyance Using Visual Attributes +
Virginia Tech +
Goibibo +
Virginia Tech +
Virginia Tech +
17a85799c59c13f07d4b4d7cf9d7c7986475d01cADVERTIMENT. La consulta d’aquesta tesi queda condicionada a l’acceptació de les següents +
condicions d'ús: La difusió d’aquesta tesi per mitjà del servei TDX (www.tesisenxarxa.net) ha +
estat autoritzada pels titulars dels drets de propietat intel·lectual únicament per a usos privats +
emmarcats en activitats d’investigació i docència. No s’autoritza la seva reproducció amb finalitats +
de lucre ni la seva difusió i posada a disposició des d’un lloc aliè al servei TDX. No s’autoritza la +
presentació del seu contingut en una finestra o marc aliè a TDX (framing). Aquesta reserva de +
drets afecta tant al resum de presentació de la tesi com als seus continguts. En la utilització o cita +
de parts de la tesi és obligat indicar el nom de la persona autora. +
ADVERTENCIA. La consulta de esta tesis queda condicionada a la aceptación de las siguientes +
condiciones de uso: La difusión de esta tesis por medio del servicio TDR (www.tesisenred.net) ha +
sido autorizada por los titulares de los derechos de propiedad intelectual únicamente para usos +
privados enmarcados en actividades de investigación y docencia. No se autoriza su reproducción +
con finalidades de lucro ni su difusión y puesta a disposición desde un sitio ajeno al servicio TDR. +
No se autoriza la presentación de su contenido en una ventana o marco ajeno a TDR (framing). +
Esta reserva de derechos afecta tanto al resumen de presentación de la tesis como a sus +
contenidos. En la utilización o cita de partes de la tesis es obligado indicar el nombre de la +
persona autora. +
WARNING. On having consulted this thesis you’re accepting the following use conditions: +
Spreading this thesis by the TDX (www.tesisenxarxa.net) service has been authorized by the +
titular of the intellectual property rights only for private uses placed in investigation and teaching +
activities. Reproduction with lucrative aims is not authorized neither its spreading and availability +
from a site foreign to the TDX service. Introducing its content in a window or frame foreign to the +
TDX service is not authorized (framing). This rights affect to the presentation summary of the +
thesis as well as to its contents. In the using or citation of parts of the thesis it’s obliged to indicate +
the name of the author +
176f26a6a8e04567ea71677b99e9818f8a8819d0MEG: Multi-Expert Gender classification from +
face images in a demographics-balanced dataset +
17035089959a14fe644ab1d3b160586c67327db2
17a995680482183f3463d2e01dd4c113ebb31608IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. X, NO. Y, MONTH Z +
Structured Label Inference for +
Visual Understanding +
17aa78bd4331ef490f24bdd4d4cd21d22a18c09c
17c0d99171efc957b88c31a465c59485ab033234
1742ffea0e1051b37f22773613f10f69d2e4ed2c
1791f790b99471fc48b7e9ec361dc505955ea8b1
174930cac7174257515a189cd3ecfdd80ee7dd54Multi-view Face Detection Using Deep Convolutional +
Neural Networks +
Yahoo +
Mohammad Saberian +
inc.com +
Yahoo +
Yahoo +
17fad2cc826d2223e882c9fda0715fcd5475acf3
1750db78b7394b8fb6f6f949d68f7c24d28d934fDetecting Facial Retouching Using Supervised +
Deep Learning +
Bowyer, Fellow, IEEE +
173657da03e3249f4e47457d360ab83b3cefbe63HKU-Face: A Large Scale Dataset for +
Deep Face Recognition +
Final Report +
3035140108 +
COMP4801 Final Year Project +
Project Code: 17007 +
7ba0bf9323c2d79300f1a433ff8b4fe0a00ad889
7bfe085c10761f5b0cc7f907bdafe1ff577223e0
7b9b3794f79f87ca8a048d86954e0a72a5f97758DOI 10.1515/jisys-2013-0016      Journal of Intelligent Systems 2013; 22(4): 365–415 +
Passing an Enhanced Turing Test – +
Interacting with Lifelike Computer +
Representations of Specific Individuals  +
7b0f1fc93fb24630eb598330e13f7b839fb46cceLearning to Find Eye Region Landmarks for Remote Gaze +
Estimation in Unconstrained Settings +
ETH Zurich +
MPI for Informatics +
MPI for Informatics +
ETH Zurich +
7bdcd85efd1e3ce14b7934ff642b76f017419751289 +
Learning Discriminant Face Descriptor +
7b3b7769c3ccbdf7c7e2c73db13a4d32bf93d21fOn the Design and Evaluation of Robust Head Pose for +
Visual User Interfaces: Algorithms, Databases, and +
Comparisons +
Laboratory of Intelligent and +
Safe Automobiles +
UCSD - La Jolla, CA, USA +
Laboratory of Intelligent and +
Safe Automobiles +
UCSD - La Jolla, CA, USA +
Laboratory of Intelligent and +
Safe Automobiles +
UCSD - La Jolla, CA, USA +
Laboratory of Intelligent and +
Safe Automobiles +
UCSD - La Jolla, CA, USA +
Mohan Trivedi +
Laboratory of Intelligent and +
Safe Automobiles +
UCSD - La Jolla, CA, USA +
8f772d9ce324b2ef5857d6e0b2a420bc93961196MAHPOD et al.: CFDRNN +
Facial Landmark Point Localization using +
Coarse-to-Fine Deep Recurrent Neural Network +
8fb611aca3bd8a3a0527ac0f38561a5a9a5b8483
8fda2f6b85c7e34d3e23927e501a4b4f7fc15b2aFeature Selection with Annealing for Big Data +
Learning +
8f9c37f351a91ed416baa8b6cdb4022b231b9085Generative Adversarial Style Transfer Networks for Face Aging +
Sveinn Palsson +
D-ITET, ETH Zurich +
Eirikur Agustsson +
D-ITET, ETH Zurich +
8f8c0243816f16a21dea1c20b5c81bc223088594
8f89aed13cb3555b56fccd715753f9ea72f27f05Attended End-to-end Architecture for Age +
Estimation from Facial Expression Videos +
8f9f599c05a844206b1bd4947d0524234940803d
8fd9c22b00bd8c0bcdbd182e17694046f245335f   +
Recognizing Facial Expressions in Videos +
8a866bc0d925dfd8bb10769b8b87d7d0ff01774dWikiArt Emotions: An Annotated Dataset of Emotions Evoked by Art +
National Research Council Canada +
8a40b6c75dd6392ee0d3af73cdfc46f59337efa9
8a91ad8c46ca8f4310a442d99b98c80fb8f7625f2592 +
2D Segmentation Using a Robust Active +
Shape Model With the EM Algorithm +
8aed6ec62cfccb4dba0c19ee000e6334ec585d70Localizing and Visualizing Relative Attributes +
8a336e9a4c42384d4c505c53fb8628a040f2468eWang and Luo EURASIP Journal on Bioinformatics +
and Systems Biology (2016) 2016:13 +
DOI 10.1186/s13637-016-0048-7 +
R ES EAR CH +
Detecting Visually Observable Disease +
Symptoms from Faces +
Open Access +
7e600faee0ba11467d3f7aed57258b0db0448a72
7e8016bef2c180238f00eecc6a50eac473f3f138TECHNISCHE UNIVERSIT ¨AT M ¨UNCHEN +
Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +
Immersive Interactive Data Mining and Machine +
Learning Algorithms for Big Data Visualization +
Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +
der Technischen Universit¨at M¨unchen zur Erlangung des akademischen Grades eines +
Doktor-Ingenieurs (Dr.-Ing.) +
genehmigten Dissertation. +
Vorsitzender: +
Univ.-Prof. Dr. sc.techn. Andreas Herkersdorf +
Pr¨ufer der Dissertation: +
1. Univ.-Prof. Dr.-Ing. habil. Gerhard Rigoll +
2. Univ.-Prof. Dr.-Ing. habil. Dirk Wollherr +
3. Prof. Dr. Mihai Datcu +
Die Dissertation wurde am 13.08.2015 bei der Technischen Universit¨at M¨unchen eingerei- +
cht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik am 16.02.2016 +
angenommen. +
7e3367b9b97f291835cfd0385f45c75ff84f4dc5Improved Local Binary Pattern Based Action Unit Detection Using +
Morphological and Bilateral Filters +
1Signal Processing Laboratory (LTS5) +
´Ecole Polytechnique F´ed´erale de Lausanne, +
Switzerland +
2nViso SA +
Lausanne, Switzerland +
7ed6ff077422f156932fde320e6b3bd66f8ffbcbState of 3D Face Biometrics for Homeland Security Applications +
Chaudhari4 +
7e507370124a2ac66fb7a228d75be032ddd083ccThis 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 +
Transactions on Affective Computing +
Dynamic Pose-Robust Facial Expression +
Recognition by Multi-View Pairwise Conditional +
Random Forests +
1 Sorbonne Universit´es, UPMC Univ Paris 06 +
CNRS, UMR 7222, F-75005, Paris, France +
1056347fc5e8cd86c875a2747b5f84fd570ba232
10e7dd3bbbfbc25661213155e0de1a9f043461a2Cross Euclidean-to-Riemannian Metric Learning +
with Application to Face Recognition from Video +
10ab1b48b2a55ec9e2920a5397febd84906a7769
10ce3a4724557d47df8f768670bfdd5cd5738f95Fihe igh Fied f Face Recgii +
Ac e ad  iai +
Rah G ai ahew ad Si Bake +
The Rbic i e Caegie e Uiveiy +
5000 Fbe Ave e ib gh A 15213 +
Abac.  ay face ecgii ak he e ad i iai +
cdii f he be ad gaey iage ae di(cid:11)ee.  he cae +
 ie gaey  be iage ay be avaiabe each ca ed f +
a di(cid:11)ee e ad de a di(cid:11)ee i iai. We e a face +
ecgii agih which ca e ay  be f gaey iage e +
 bjec ca ed a abiay e ad de abiay i iai +
ad ay  be f be iage agai ca ed a abiay e ad +
de abiay i iai. The agih eae by eiaig he +
Fihe igh (cid:12)ed f he  bjec head f he i  gaey  be +
iage. achig bewee he be ad gaey i he efed ig +
he Fihe igh (cid:12)ed. +
d ci +
 ay face ecgii ceai he e f he be ad gaey iage ae +
di(cid:11)ee. The gaey cai he iage ed d ig aiig f he agih. +
The agih ae eed wih he iage i he be e. F exae he +
gaey iage igh be a fa \ g h" ad he be iage igh be a 3/4 +
view ca ed f a caea i he ce f he . The  be f gaey +
ad be iage ca a vay. F exae he gaey ay ci f a ai f +
iage f each  bjec a fa  g h ad f (cid:12)e view ike he iage +
yicay ca ed by ice deae. The be ay be a iia ai f +
iage a ige 3/4 view  eve a ceci f view f ad e. +
Face ecgii ac e i.e. face ecgii whee he gaey ad be +
iage d  have he ae e ha eceived vey ie aei. Agih +
have bee ed which ca ecgize face [1]  e geea bjec [2] +
a a vaiey f e. weve  f hee agih e ie gaey iage +
a evey e. Agih have bee ed which d geeaize ac e +
f exae [3] b  hi agih c e 3D head de ig a gaey +
caiig a age  be f iage e  bjec ca ed ig ced i +
iai vaiai.  ca be ed wih abiay gaey ad be e. +
Afe e vaiai he ex  igi(cid:12)ca fac a(cid:11)ecig he aea +
ace f face i i iai. A  be f agih have bee deveed f +
face ecgii ac i iai b  hey yicay y dea wih fa +
face [4 5]. y a few aache have bee ed  hade bh e ad +
i iai vaiai a he ae ie. F exae [3] c e a 3D head +
102e374347698fe5404e1d83f441630b1abf62d9Facial Image Analysis for Fully-Automatic +
Prediction of Difficult Endotracheal Intubation +
100641ed8a5472536dde53c1f50fa2dd2d4e9be9Visual Attributes for Enhanced Human-Machine Communication* +
10195a163ab6348eef37213a46f60a3d87f289c5
10e704c82616fb5d9c48e0e68ee86d4f83789d96
101569eeef2cecc576578bd6500f1c2dcc0274e2Multiaccuracy: Black-Box Post-Processing for Fairness in +
Classification +
James Zou +
106732a010b1baf13c61d0994552aee8336f8c85Expanded Parts Model for Semantic Description +
of Humans in Still Images +
10e70a34d56258d10f468f8252a7762950830d2b
102b27922e9bd56667303f986404f0e1243b68abWang et al. Appl Inform (2017) 4:13 +
DOI 10.1186/s40535-017-0042-5 +
RESEARCH +
Multiscale recurrent regression networks +
for face alignment +
Open Access +
*Correspondence: +
3 State Key Lab of Intelligent +
Technologies and Systems, +
Beijing 100084, People’s +
Republic of China +
Full list of author information +
is available at the end of the +
article +
10fcbf30723033a5046db791fec2d3d286e34daaOn-Line Cursive Handwriting Recognition: A Survey of Methods +
and Performances +
*Faculty of Computer Science & Information Systems, Universiti Teknologi Malaysia (UTM) , 81310 +
Skudai, Johor, Malaysia. +
108b2581e07c6b7ca235717c749d45a1fa15bb24Using Stereo Matching with General Epipolar +
Geometry for 2D Face Recognition +
across Pose +
10d334a98c1e2a9e96c6c3713aadd42a557abb8bScene Text Recognition using Part-based Tree-structured Character Detection +
State Key Laboratory of Management and Control for Complex Systems, CASIA, Beijing, China +
192723085945c1d44bdd47e516c716169c06b7c0This 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 +
Vision and Attention Theory Based Sampling +
for Continuous Facial Emotion Recognition +
Ninad S. Thakoor, Member, IEEE +
10 +
11 +
12 +
13 +
14 +
15 +
16 +
17 +
18 +
19 +
20 +
21 +
22 +
23 +
24 +
25 +
26 +
27 +
28 +
29 +
30 +
31 +
32 +
33 +
34 +
35 +
36 +
37 +
19fb5e5207b4a964e5ab50d421e2549ce472baa8International Conference on Computer Systems and Technologies - CompSysTech’14 +
Online Emotional Facial Expression Dictionary +
Léon Rothkrantz +
1962e4c9f60864b96c49d85eb897141486e9f6d1Neural Comput & Applic (2011) 20:565–573 +
DOI 10.1007/s00521-011-0577-7 +
O R I G I N A L A R T I C L E +
Locality preserving embedding for face and handwriting digital +
recognition +
Received: 3 December 2008 / Accepted: 11 March 2011 / Published online: 1 April 2011 +
Ó Springer-Verlag London Limited 2011 +
supervised manifold +
the local sub-manifolds. +
19af008599fb17bbd9b12288c44f310881df951cDiscriminative Local Sparse Representations for +
Robust Face Recognition +
19296e129c70b332a8c0a67af8990f2f4d4f44d1Metric Learning Approaches for Face Identification +
Is that you? +
M. Guillaumin, J. Verbeek and C. Schmid +
LEAR team, INRIA Rhˆone-Alpes, France +
Supplementary Material +
19666b9eefcbf764df7c1f5b6938031bcf777191Group Component Analysis for Multi-block Data: +
Common and Individual Feature Extraction +
190b3caa2e1a229aa68fd6b1a360afba6f50fde4
19c0c7835dba1a319b59359adaa738f0410263e8228 +
Natural Image Statistics and +
Low-Complexity Feature Selection +
19808134b780b342e21f54b60095b181dfc7a600
19d583bf8c5533d1261ccdc068fdc3ef53b9ffb9FaceNet: A Unified Embedding for Face Recognition and Clustering +
Google Inc. +
Google Inc. +
Google Inc. +
197c64c36e8a9d624a05ee98b740d87f94b4040cRegularized Greedy Column Subset Selection +
aDepartment of Computer Systems, Universidad Polit´ecnica de Madrid +
bDepartment of Applied Mathematics, Universidad Polit´ecnica de Madrid +
19d4855f064f0d53cb851e9342025bd8503922e2Learning SURF Cascade for Fast and Accurate Object Detection +
Intel Labs China +
19eb486dcfa1963c6404a9f146c378fc7ae3a1df
4c6daffd092d02574efbf746d086e6dc0d3b1e91
4c6e1840451e1f86af3ef1cb551259cb259493baHAND POSTURE DATASET CREATION FOR GESTURE +
RECOGNITION +
Instituto de Sistemas Inteligentes y Aplicaciones Numericas en Ingenieria +
Campus Universitario de Tafira, 35017 Gran Canaria, Spain +
Departamento de E.I.O. y Computacion +
38271 Universidad de La Laguna, Spain +
Keywords: +
Image understanding, Gesture recognition, Hand dataset. +
4c29e1f31660ba33e46d7e4ffdebb9b8c6bd5adc
4c815f367213cc0fb8c61773cd04a5ca8be2c959978-1-4244-4296-6/10/$25.00 ©2010 IEEE +
2470 +
ICASSP 2010 +
4c4236b62302957052f1bbfbd34dbf71ac1650ecSEMI-SUPERVISED FACE RECOGNITION WITH LDA SELF-TRAINING +
Multimedia Communications Department, EURECOM +
2229 Route des Crêtes , BP 193, F-06560 Sophia-Antipolis Cedex, France +
2661f38aaa0ceb424c70a6258f7695c28b97238aIEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 42, NO. 4, AUGUST 2012 +
1027 +
Multilayer Architectures for Facial +
Action Unit Recognition +
2609079d682998da2bc4315b55a29bafe4df414eON RANK AGGREGATION FOR FACE RECOGNITION FROM VIDEOS +
IIIT-Delhi, India +
26a72e9dd444d2861298d9df9df9f7d147186bcdDOI 10.1007/s00138-016-0768-4 +
ORIGINAL PAPER +
Collecting and annotating the large continuous action dataset +
Received: 18 June 2015 / Revised: 18 April 2016 / Accepted: 22 April 2016 / Published online: 21 May 2016 +
© The Author(s) 2016. This article is published with open access at Springerlink.com +
265af79627a3d7ccf64e9fe51c10e5268fee2aae1817 +
A Mixture of Transformed Hidden Markov +
Models for Elastic Motion Estimation +
267c6e8af71bab68547d17966adfaab3b4711e6b
26a89701f4d41806ce8dbc8ca00d901b68442d45
26ad6ceb07a1dc265d405e47a36570cb69b2ace6RESEARCH AND EXPLOR ATORY +
DEVELOPMENT DEPARTMENT +
REDD-2015-384 +
Neural Correlates of Cross-Cultural +
How to Improve the Training and Selection for +
Military Personnel Involved in Cross-Cultural +
Operating Under Grant #N00014-12-1-0629/113056 +
Adaptation +
September, 2015 +
Interactions +
Prepared for: +
Office of Naval Research +
26e570049aaedcfa420fc8c7b761bc70a195657cJ Sign Process Syst +
DOI 10.1007/s11265-017-1276-0 +
Hybrid Facial Regions Extraction for Micro-expression +
Recognition System +
Received: 2 February 2016 / Revised: 20 October 2016 / Accepted: 10 August 2017 +
© Springer Science+Business Media, LLC 2017 +
21ef129c063bad970b309a24a6a18cbcdfb3aff5POUR 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
218b2c5c9d011eb4432be4728b54e39f366354c1Enhancing Training Collections for Image +
Annotation: An Instance-Weighted Mixture +
Modeling Approach +
21e828071249d25e2edaca0596e27dcd63237346
2162654cb02bcd10794ae7e7d610c011ce0fb51b4697 +
978-1-4799-5751-4/14/$31.00 ©2014 IEEE +
1http://www.skype.com/ +
2http://www.google.com/hangouts/ +
tification, sparse coding +
21f3c5b173503185c1e02a3eb4e76e13d7e9c5bcm 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 — 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 +
Rotation Invariant Real-time +
Face Detection and +
Recognition System +
AI Memo 2001-010 +
CBCL Memo 197 +
May 31, 2001 +
© 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 +
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 +
21bd9374c211749104232db33f0f71eab4df35d5Integrating Facial Makeup Detection Into +
Multimodal Biometric User Verification System +
CuteSafe Technology Inc. +
Gebze, Kocaeli, Turkey +
Eurecom Digital Security Department +
06410 Biot, France +
213a579af9e4f57f071b884aa872651372b661fdInt J Comput Vis +
DOI 10.1007/s11263-013-0672-6 +
Automatic and Efficient Human Pose Estimation for Sign +
Language Videos +
Received: 4 February 2013 / Accepted: 29 October 2013 +
© Springer Science+Business Media New York 2013 +
21626caa46cbf2ae9e43dbc0c8e789b3dbb420f1978-1-4673-2533-2/12/$26.00 ©2012 IEEE +
1437 +
ICIP 2012 +
4d49c6cff198cccb21f4fa35fd75cbe99cfcbf27Topological Principal Component Analysis for +
face encoding and recognition +
Juan J. Villanueva +
Computer Vision Center and Departament d’Inform(cid:18)atica, Edi(cid:12)ci O, Universitat +
Aut(cid:18)onoma de Barcelona  , Cerdanyola, Spain +
4da735d2ed0deeb0cae4a9d4394449275e316df2Gothenburg, Sweden, June 19-22, 2016 +
978-1-5090-1820-8/16/$31.00 ©2016 IEEE +
1410 +
4d530a4629671939d9ded1f294b0183b56a513efInternational Journal of Machine Learning and Computing, Vol. 2, No. 4, August 2012 +
Facial Expression Classification Method Based on Pseudo +
Zernike Moment and Radial Basis Function Network +
+
4d2975445007405f8cdcd74b7fd1dd547066f9b8Image and Video Processing +
for Affective Applications +
4df889b10a13021928007ef32dc3f38548e5ee56
4d423acc78273b75134e2afd1777ba6d3a398973
4db9e5f19366fe5d6a98ca43c1d113dac823a14dCombining Crowdsourcing and Face Recognition to Identify Civil War Soldiers +
Are 1,000 Features Worth A Picture? +
Department of Computer Science and Center for Human-Computer Interaction +
Virginia Tech, Arlington, VA, USA +
4dd6d511a8bbc4d9965d22d79ae6714ba48c8e41
4d7e1eb5d1afecb4e238ba05d4f7f487dff96c11978-1-5090-4117-6/17/$31.00 ©2017 IEEE +
2352 +
ICASSP 2017 +
4d90bab42806d082e3d8729067122a35bbc15e8d
4d6ad0c7b3cf74adb0507dc886993e603c863e8cHuman Activity Recognition Based on Wearable +
Sensor Data: A Standardization of the +
State-of-the-Art +
Smart Surveillance Interest Group, Computer Science Department +
Universidade Federal de Minas Gerais, Brazil +
4d0ef449de476631a8d107c8ec225628a67c87f9© 2010 IEEE. Personal use of this material is permitted. Permission from IEEE +
must be obtained for all other uses, in any current or future media, including +
reprinting/republishing this material for advertising or promotional purposes, +
creating new collective works, for resale or redistribution to servers or lists, or +
reuse of any copyrighted component of this work in other works. +
Pre-print of article that appeared at BTAS 2010. +
The published article can be accessed from: +
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5634517 +
4d47261b2f52c361c09f7ab96fcb3f5c22cafb9fDeep multi-frame face super-resolution +
Evgeniya Ustinova, Victor Lempitsky +
October 17, 2017 +
75879ab7a77318bbe506cb9df309d99205862f6cAnalysis Of Emotion Recognition From Facial +
Expressions Using Spatial And Transform Domain +
Methods +
7574f999d2325803f88c4915ba8f304cccc232d1Transfer Learning For Cross-Dataset Recognition: A Survey +
This paper summarises and analyses the cross-dataset recognition transfer learning techniques with the +
emphasis on what kinds of methods can be used when the available source and target data are presented +
in different forms for boosting the target task. This paper for the first time summarises several transferring +
criteria in details from the concept level, which are the key bases to guide what kind of knowledge to transfer +
between datasets. In addition, a taxonomy of cross-dataset scenarios and problems is proposed according the +
properties of data that define how different datasets are diverged, thereby review the recent advances on +
each specific problem under different scenarios. Moreover, some real world applications and corresponding +
commonly used benchmarks of cross-dataset recognition are reviewed. Lastly, several future directions are +
identified. +
Additional Key Words and Phrases: Cross-dataset, transfer learning, domain adaptation +
1. INTRODUCTION +
It has been explored how human would transfer learning in one context to another +
similar context [Woodworth and Thorndike 1901; Perkins et al. 1992] in the field of +
Psychology and Education. For example, learning to drive a car helps a person later +
to learn more quickly to drive a truck, and learning mathematics prepares students to +
study physics. The machine learning algorithms are mostly inspired by human brains. +
However, most of them require a huge amount of training examples to learn a new +
model from scratch and fail to apply knowledge learned from previous domains or +
tasks. This may be due to that a basic assumption of statistical learning theory is +
that the training and test data are drawn from the same distribution and belong to +
the same task. Intuitively, learning from scratch is not realistic and practical, because +
it violates how human learn things. In addition, manually labelling a large amount +
of data for new domain or task is labour extensive, especially for the modern “data- +
hungry” and “data-driven” learning techniques (i.e. deep learning). However, the big +
data era provides a huge amount available data collected for other domains and tasks. +
Hence, how to use the previously available data smartly for the current task with +
scarce data will be beneficial for real world applications. +
To reuse the previous knowledge for current tasks, the differences between old data +
and new data need to be taken into account. Take the object recognition as an ex- +
ample. As claimed by Torralba and Efros [2011], despite the great efforts of object +
datasets creators, the datasets appear to have strong build-in bias caused by various +
factors, such as selection bias, capture bias, category or label bias, and negative set +
bias. This suggests that no matter how big the dataset is, it is impossible to cover +
the complexity of the real visual world. Hence, the dataset bias needs to be consid- +
ered before reusing data from previous datasets. Pan and Yang [2010] summarise that +
the differences between different datasets can be caused by domain divergence (i.e. +
distribution shift or feature space difference) or task divergence (i.e. conditional dis- +
tribution shift or label space difference), or both. For example, in visual recognition, +
the distributions between the previous and current data can be discrepant due to the +
different environments, lighting, background, sensor types, resolutions, view angles, +
and post-processing. Those external factors may cause the distribution divergence or +
even feature space divergence between different domains. On the other hand, the task +
divergence between current and previous data is also ubiquitous. For example, it is +
highly possible that an animal species that we want to recognize have not been seen +
ACM Journal Name, Vol. V, No. N, Article A, Publication date: January YYYY. +
75e9a141b85d902224f849ea61ab135ae98e7bfb
75503aff70a61ff4810e85838a214be484a674baImproved Facial Expression Recognition via Uni-Hyperplane Classification +
S.W. Chew∗, S. Lucey†, P. Lucey‡, S. Sridharan∗, and J.F. Cohn‡ +
75cd81d2513b7e41ac971be08bbb25c63c37029a
75e5ba7621935b57b2be7bf4a10cad66a9c445b9
75859ac30f5444f0d9acfeff618444ae280d661dMultibiometric Cryptosystems based on Feature +
Level Fusion +
758d7e1be64cc668c59ef33ba8882c8597406e53IEEE TRANSACTIONS ON AFFECTIVE COMPUTING +
AffectNet: A Database for Facial Expression, +
Valence, and Arousal Computing in the Wild +
754f7f3e9a44506b814bf9dc06e44fecde599878Quantized Densely Connected U-Nets for +
Efficient Landmark Localization +
75249ebb85b74e8932496272f38af274fbcfd696Face Identification in Large Galleries +
Smart Surveillance Interest Group, Department of Computer Science +
Universidade Federal de Minas Gerais, Belo Horizonte, Brazil +
81a142c751bf0b23315fb6717bc467aa4fdfbc92978-1-5090-4117-6/17/$31.00 ©2017 IEEE +
1767 +
ICASSP 2017 +
8147ee02ec5ff3a585dddcd000974896cb2edc53Angular Embedding: +
A Robust Quadratic Criterion +
Stella X. Yu, Member, +
IEEE +
8199803f476c12c7f6c0124d55d156b5d91314b6The iNaturalist Species Classification and Detection Dataset +
1Caltech +
2Google +
3Cornell Tech +
4iNaturalist +
81831ed8e5b304e9d28d2d8524d952b12b4cbf55
81b2a541d6c42679e946a5281b4b9dc603bc171cUniversit¨at Ulm | 89069 Ulm | Deutschland +
Fakult¨at f¨ur Ingenieurwissenschaften und Informatik +
Institut f¨ur Neuroinformatik +
Direktor: Prof. Dr. G¨unther Palm +
Semi-Supervised Learning with Committees: +
Exploiting Unlabeled Data Using Ensemble +
Learning Algorithms +
Dissertation zur Erlangung des Doktorgrades +
Doktor der Naturwissenschaften (Dr. rer. nat.) +
der Fakult¨at f¨ur Ingenieurwissenschaften und Informatik +
der Universit¨at Ulm +
vorgelegt von +
aus Kairo, ¨Agypten +
Ulm, Deutschland +
2010 +
8160b3b5f07deaa104769a2abb7017e9c031f1c1683 +
Exploiting Discriminant Information in Nonnegative +
Matrix Factorization With Application +
to Frontal Face Verification +
816eff5e92a6326a8ab50c4c50450a6d02047b5efLRR: Fast Low-Rank Representation Using +
Frobenius Norm +
Low Rank Representation (LRR) intends to find the representation +
with lowest-rank of a given data set, which can be formulated as a +
rank minimization problem. Since the rank operator is non-convex and +
discontinuous, most of the recent works use the nuclear norm as a convex +
relaxation. This letter theoretically shows that under some conditions, +
Frobenius-norm-based optimization problem has an unique solution that +
is also a solution of the original LRR optimization problem. In other +
words, it is feasible to apply Frobenius-norm as a surrogate of the +
nonconvex matrix rank function. This replacement will largely reduce the +
time-costs for obtaining the lowest-rank solution. Experimental results +
show that our method (i.e., fast Low Rank Representation, fLRR), +
performs well in terms of accuracy and computation speed in image +
clustering and motion segmentation compared with nuclear-norm-based +
LRR algorithm. +
Introduction: Given a data set X ∈ Rm×n(m < n) composed of column +
vectors, let A be a data set composed of vectors with the same dimension +
as those in X. Both X and A can be considered as matrices. A linear +
representation of X with respect to A is a matrix Z that satisfies the +
equation X = AZ. The data set A is called a dictionary. In general, this +
linear matrix equation will have infinite solutions, and any solution can be +
considered to be a representation of X associated with the dictionary A. To +
obtain an unique Z and explore the latent structure of the given data set, +
various assumptions could be enforced over Z. +
Liu et al. recently proposed Low Rank Representation (LRR) [1] by +
assuming that data are approximately sampled from an union of low-rank +
subspaces. Mathematically, LRR aims at solving +
min rank(Z) +
s.t. X = AZ, +
(1) +
where rank(Z) could be defined as the number of nonzero eigenvalues of +
the matrix Z. Clearly, (1) is non-convex and discontinuous, whose convex +
relaxation is as follows, +
min kZk∗ +
s.t. X = AZ, +
(2) +
where kZk∗ is the nuclear norm, which is a convex and continuous +
optimization problem. +
Considering the possible corruptions, the objective function of LRR is +
min kZk∗ + λkEkp +
s.t. X = AZ + E, +
(3) +
where k · kp could be ℓ1-norm for describing sparse corruption or ℓ2,1- +
norm for characterizing sample-specified corruption. +
The above nuclear-norm-based optimization problems are generally +
solved using Augmented Lagrange Multiplier algorithm (ALM) [2] which +
requires repeatedly performing Single Value Decomposition (SVD) over +
Z. Hence, this optimization program is inefficient. +
Beyond the nuclear-norm, do other norms exist that can be used as +
a surrogates for rank-minimization problem in LRR? Can we develop +
a fast algorithm to calculate LRR? This letter addresses these problems +
by theoretically showing the equivalence between the solutions of a +
Frobenius-norm-based problem and the original LRR problem. And we +
further develop fast Low Rank Representation (fLRR) based on the +
theoretical results. +
Theoretical Analysis: In the following analyses, Theorem 1 and +
Theorem 3 prove that Frobenius-norm-based problem is a surrogate of +
the rank-minimization problem of LRR in the case of clean data and +
corrupted ones, respectively. Theorem 2 shows that our Frobenius-norm- +
based method could produce a block-diagonal Z under some conditions. +
This property is helpful to subspace clustering. +
Let A ∈ Rm×n be a matrix with rank r. The full SVD and skinny +
SVD of A are A = U ΣV T and A = UrΣrV T +
r , where U and V are two +
orthogonal matrices with the size of m × m and n × n, respectively. In +
addition, Σ is an m × n rectangular diagonal matrix, its diagonal elements +
are nonnegative real numbers. Σr is a r × r diagonal matrix with singular +
values located on the diagonal in decreasing order, Ur and Vr consist of the +
first r columns of U and V , respectively. Clearly, Ur and Vr are column +
orthogonal matrices, i.e., U T +
r Vr = Ir, where Ir denotes the +
r Ur = Ir, V T +
identity matrix with the size of r × r. The pseudoinverse of A is defined +
by A† = VrΣ−1 +
r U T +
r . +
Given a matrix M ∈ Rm×n, the Frobenius norm of M is defined by +
kM kF =ptrace (M T M ) =qPmin{m,n} +
value of M . Clearly, kM kF = 0 if and only if M = 0. +
i=1 +
σ2 +
i , where σi is a singular +
Lemma 1: Suppose P is a column orthogonal matrix, i.e., P T P = I. Then, +
kP M kF = kM kF . +
Lemma 2: For the matrices M and N with same number of columns, it +
holds that +
= kM k2 +
F + kN k2 +
F . +
(4) +
N (cid:21)(cid:13)(cid:13)(cid:13)(cid:13) +
(cid:13)(cid:13)(cid:13)(cid:13) +
(cid:20) M +
The proofs of the above two lemmas are trivial. +
Theorem 1: +
minimization problem +
Suppose +
that X ∈ span{A}, +
the Frobenius norm +
min kZkF +
s.t. X = AZ, +
(5) +
has an unique solution Z ∗ = A†X which is also the lowest-rank solution +
of LRR in terms of (1). +
Proof: Let the full and skinny SVDs of A be A = U ΣV T and A = +
r U T +
UrΣrV T +
r . +
r , respectively. Then, the pseudoinverse of A is A† = VrΣ−1 +
Defining Vc by V T =(cid:20) V T +
V T +
(cid:21) and V T +
c Vr = 0. Moreover, it can be easily +
checked that Z ∗ satisfies X = AZ ∗ owing to X ∈ span{A}. +
To prove that Z ∗ is the unique solution of the optimization problem +
(5), two steps are required. First, we will prove that, for any solution Z of +
X = AZ, it must hold that kZkF ≥ kZ ∗kF . Using Lemma 1, we have +
kZkF = (cid:13)(cid:13)(cid:13)(cid:13) +
= (cid:13)(cid:13)(cid:13)(cid:13) +
V T +
(cid:20) V T +
(cid:20) V T +
(cid:21) [Z ∗ + (Z − Z ∗)](cid:13)(cid:13)(cid:13)(cid:13)F +
c (Z − Z ∗) (cid:21)(cid:13)(cid:13)(cid:13)(cid:13)F +
r (Z − Z ∗) +
r Z ∗ + V T +
c Z ∗ + V T +
V T +
As A (Z − Z ∗) = 0, +
r (Z − Z ∗) = 0. Denote B = Σ−1 +
V T +
V T +
c Vr = 0, we have V T +
i.e., UrΣrV T +
r U T +
c VrB = 0. Then, +
r (Z − Z ∗) = 0, +
r X, +
follows that +
then Z ∗ = VrB. Because +
it +
c Z ∗ = V T +
(cid:20) +
kZkF =(cid:13)(cid:13)(cid:13)(cid:13) +
V T +
c (Z − Z ∗) (cid:21)(cid:13)(cid:13)(cid:13)(cid:13)F +
By Lemma 2, +
kZk2 +
F = kBk2 +
F + kV T +
c (Z − Z ∗)k2 +
F , +
then, kZkF ≥ kBkF . +
By Lemma 1, +
kBkF = kVrBkF = kZ ∗kF , +
(6) +
(7) +
(8) +
thus, kZkF ≥ kZ ∗kF for any solution Z of X = AZ. +
In the second step, we will prove that if there exists another solution Z +
of (5), Z = Z ∗ must hold. Clearly, Z is a solution of (5) which implies that +
X = AZ and kZkF = kZ ∗kF . From (7) and (8), +
kZk2 +
F + kV T +
F = kZ ∗k2 +
Since kZkF = kZ ∗kF , +
c (Z − Z ∗) k2 +
F . +
c (Z − Z ∗) kF = 0, +
r (Z − Z ∗) = 0, this gives +
and so V T +
V T (Z − Z ∗) = 0. Because V is an orthogonal matrix, it must hold +
that Z = Z ∗. The above proves that Z ∗ is the unique solution of the +
optimization problem (5). +
c (Z − Z ∗) = 0. Together with V T +
it must hold that kV T +
(9) +
Next, we prove that Z ∗ is also a solution of the LRR optimization +
problem (1). Clearly, for any solution Z of X = AZ, +
it holds that +
rank(Z) ≥ rank(AZ) = rank(X). On the other hand, rank(Z ∗) = +
rank(A†X) ≤ rank(X). Thus, rank(Z ∗) = rank(X). This shows that +
Z ∗ is the lowest-rank solution of the LRR optimization problem (1). The +
proof is complete. +
(cid:4) +
In the following, Theorem 2 will show that the optimal Z of (5) will +
be block-diagonal if the data are sampled from a set of independent +
subspaces {S1, S2, · · · , Sk}, where the dimensionality of Si is ri and +
i = {1, 2, · · · , k}. Note that, {S1, S2, · · · , Sk} are independent if and +
only if SiTPj6=i Sj = {0}. Suppose that X = [X1, X2, · · · , Xk] and +
A = [A1, A2, · · · , Ak], where Ai and Xi contain mi and ni data points +
ELECTRONICS LETTERS 12th December 2011 Vol. 00 No. 00 +
8149c30a86e1a7db4b11965fe209fe0b75446a8cSemi-Supervised Multiple Instance Learning based +
Domain Adaptation for Object Detection +
Siemens Corporate Research +
Siemens Corporate Research +
Siemens Corporate Research +
Amit Kale +
Bangalore +
Bangalore +
{chhaya.methani, +
Bangalore +
rahul.thota, +
86b69b3718b9350c9d2008880ce88cd035828432Improving Face Image Extraction by Using Deep Learning Technique +
National Library of Medicine, NIH, Bethesda, MD +
86904aee566716d9bef508aa9f0255dc18be3960Learning Anonymized Representations with +
Adversarial Neural Networks +
867e709a298024a3c9777145e037e239385c0129 INTERNATIONAL JOURNAL +
OF PROFESSIONAL ENGINEERING STUDIES Volume VIII /Issue 2 / FEB 2017 +
ANALYTICAL REPRESENTATION OF UNDERSAMPLED FACE +
RECOGNITION APPROACH BASED ON DICTIONARY LEARNING +
AND SPARSE REPRESENTATION +
(M.Tech)1, Assistant Professor2, Assistant Professor3, HOD of CSE Department4 +
86c053c162c08bc3fe093cc10398b9e64367a100Cascade of Forests for Face Alignment +
86b985b285c0982046650e8d9cf09565a939e4f9
861802ac19653a7831b314cd751fd8e89494ab12Time-of-Flight and Depth Imaging. Sensors, Algorithms +
and Applications: Dagstuhl Seminar 2012 and GCPR +
Workshop on Imaging New Modalities (Lecture ... Vision, +
Pattern Recognition, and Graphics) +
Publisher: Springer; 2013 edition +
(November 8, 2013) +
Language: English +
Pages: 320 +
ISBN: 978-3642449635 +
Size: 20.46 MB +
Format: PDF / ePub / Kindle +
Cameras for 3D depth imaging, using +
either time-of-flight (ToF) or +
structured light sensors, have received +
a lot of attention recently and have +
been improved considerably over the +
last few years. The present +
techniques... +
861b12f405c464b3ffa2af7408bff0698c6c9bf0International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 +
Volume: 3 Issue: 5 +
3337 - 3342 +
_______________________________________________________________________________________________ +
An Effective Technique for Removal of Facial Dupilcation by SBFA +
Computer Department, +
GHRCEM, +
Pune, India +
Computer Department, +
GHRCEM, +
Pune, India +
86e1bdbfd13b9ed137e4c4b8b459a3980eb257f6The Kinetics Human Action Video Dataset +
Jo˜ao Carreira +
Paul Natsev +
86b105c3619a433b6f9632adcf9b253ff98aee871­4244­0367­7/06/$20.00 ©2006 IEEE +
1013 +
ICME 2006 +
86b51bd0c80eecd6acce9fc538f284b2ded5bcdd
8699268ee81a7472a0807c1d3b1db0d0ab05f40d
869583b700ecf33a9987447aee9444abfe23f343
72a00953f3f60a792de019a948174bf680cd6c9fStat Comput (2007) 17:57–70 +
DOI 10.1007/s11222-006-9004-9 +
Understanding the role of facial asymmetry in human face +
identification +
Received: May 2005 / Accepted: September 2006 / Published online: 30 January 2007 +
C(cid:1) Springer Science + Business Media, LLC 2007 +
726b8aba2095eef076922351e9d3a724bb71cb51
721b109970bf5f1862767a1bec3f9a79e815f79a
72ecaff8b57023f9fbf8b5b2588f3c7019010ca7Facial Keypoints Detection +
72591a75469321074b072daff80477d8911c3af3Group Component Analysis for Multi-block Data: +
Common and Individual Feature Extraction +
729dbe38538fbf2664bc79847601f00593474b05
729a9d35bc291cc7117b924219bef89a864ce62cRecognizing Material Properties from Images +
721d9c387ed382988fce6fa864446fed5fb23173
72c0c8deb9ea6f59fde4f5043bff67366b86bd66Age progression in Human Faces : A Survey +
445461a34adc4bcdccac2e3c374f5921c93750f8Emotional Expression Classification using Time-Series Kernels∗ +
4414a328466db1e8ab9651bf4e0f9f1fe1a163e41164 +
© EURASIP, 2010 ISSN 2076-1465 +
18th European Signal Processing Conference (EUSIPCO-2010) +
INTRODUCTION +
442f09ddb5bb7ba4e824c0795e37cad754967208
446a99fdedd5bb32d4970842b3ce0fc4f5e5fa03A Pose-Adaptive Constrained Local Model For +
Accurate Head Pose Tracking +
Eikeo +
11 rue Leon Jouhaux, +
F-75010, Paris, France +
Sorbonne Universit´es +
UPMC Univ Paris 06 +
CNRS UMR 7222, ISIR +
F-75005, Paris, France +
Eikeo +
11 rue Leon Jouhaux, +
F-75010, Paris, France +
44b1399e8569a29eed0d22d88767b1891dbcf987This 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. +
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +
Learning Multi-modal Latent Attributes +
446dc1413e1cfaee0030dc74a3cee49a47386355Recent Advances in Zero-shot Recognition +
44a3ec27f92c344a15deb8e5dc3a5b3797505c06A Taxonomy of Part and Attribute Discovery +
Techniques +
44aeda8493ad0d44ca1304756cc0126a2720f07bFace Alive Icons +
449b1b91029e84dab14b80852e35387a9275870e
44078d0daed8b13114cffb15b368acc467f96351
44dd150b9020b2253107b4a4af3644f0a51718a3An Analysis of the Sensitivity of Active Shape +
Models to Initialization when Applied to Automatic +
Facial Landmarking +
447d8893a4bdc29fa1214e53499ffe67b28a6db5
44f65e3304bdde4be04823fd7ca770c1c05c2cefSIViP +
DOI 10.1007/s11760-009-0125-4 +
ORIGINAL PAPER +
On the use of phase of the Fourier transform for face recognition +
under variations in illumination +
Received: 17 November 2008 / Revised: 20 February 2009 / Accepted: 7 July 2009 +
© Springer-Verlag London Limited 2009 +
44eb4d128b60485377e74ffb5facc0bf4ddeb022
448ed201f6fceaa6533d88b0b29da3f36235e131
447a5e1caf847952d2bb526ab2fb75898466d1bcUnder review as a conference paper at ICLR 2018 +
LEARNING NON-LINEAR TRANSFORM WITH DISCRIM- +
INATIVE AND MINIMUM INFORMATION LOSS PRIORS +
Anonymous authors +
Paper under double-blind review +
2a7bca56e2539c8cf1ae4e9da521879b7951872dExploiting Unrelated Tasks in Multi-Task Learning +
Anonymous Author 1 +
Unknown Institution 1 +
Anonymous Author 2 +
Unknown Institution 2 +
Anonymous Author 3 +
Unknown Institution 3 +
2aaa6969c03f435b3ea8431574a91a0843bd320b
2ad7cef781f98fd66101fa4a78e012369d064830
2ad29b2921aba7738c51d9025b342a0ec770c6ea
2a6bba2e81d5fb3c0fd0e6b757cf50ba7bf8e924
2aec012bb6dcaacd9d7a1e45bc5204fac7b63b3cRobust Registration and Geometry Estimation from Unstructured +
Facial Scans +
2ae139b247057c02cda352f6661f46f7feb38e45Combining Modality Specific Deep Neural Networks for +
Emotion Recognition in Video +
1École Polytechique de Montréal, Université de Montréal, Montréal, Canada +
2Laboratoire d’Informatique des Systèmes Adaptatifs, Université de Montréal, Montréal, Canada +
2a5903bdb3fdfb4d51f70b77f16852df3b8e5f83121 +
The Effect of Computer-Generated Descriptions +
on Photo-Sharing Experiences of People With +
Visual Impairments +
Like sighted people, visually impaired people want to share photographs on social networking services, but +
find it difficult to identify and select photos from their albums. We aimed to address this problem by +
incorporating state-of-the-art computer-generated descriptions into Facebook’s photo-sharing feature. We +
interviewed 12 visually impaired participants to understand their photo-sharing experiences and designed a +
photo description feature for the Facebook mobile application. We evaluated this feature with six +
participants in a seven-day diary study. We found that participants used the descriptions to recall and +
organize their photos, but they hesitated to upload photos without a sighted person’s input. In addition to +
basic information about photo content, participants wanted to know more details about salient objects and +
people, and whether the photos reflected their personal aesthetic. We discuss these findings from the lens of +
self-disclosure and self-presentation theories and propose new computer vision research directions that will +
better support visual content sharing by visually impaired people. +
CCS Concepts: • Information interfaces and presentations → Multimedia and information systems; • +
Social and professional topics → People with disabilities +
KEYWORDS +
Visual impairments; computer-generated descriptions; SNSs; photo sharing; self-disclosure; self-presentation +
ACM Reference format: +
The Effect of Computer-Generated Descriptions On Photo-Sharing Experiences of People With Visual +
Impairments. Proc. ACM Hum.-Comput. Interact. 1, CSCW. 121 (November 2017), 22 pages. +
DOI: 10.1145/3134756 +
1 INTRODUCTION +
Sharing memories and experiences via photos is a common way to engage with others on social networking +
services (SNSs) [39,46,51]. For instance, Facebook users uploaded more than 350 million photos a day [24] +
and Twitter, which initially supported only text in tweets, now has more than 28.4% of tweets containing +
images [39]. Visually impaired people (both blind and low vision) have a strong presence on SNS and are +
interested in sharing photos [50]. They take photos for the same reasons that sighted people do: sharing +
daily moments with their sighted friends and family [30,32]. A prior study showed that visually impaired +
people shared a relatively large number of photos on Facebook—only slightly less than their sighted +
counterparts [50]. +
+
PACM on Human-Computer Interaction, Vol. 1, No. 2, Article 121. Publication date: November 2017 +
2a02355c1155f2d2e0cf7a8e197e0d0075437b19
2aea27352406a2066ddae5fad6f3f13afdc90be9
2ad0ee93d029e790ebb50574f403a09854b65b7eAcquiring Linear Subspaces for Face +
Recognition under Variable Lighting +
David Kriegman, Senior Member, IEEE +
2ff9618ea521df3c916abc88e7c85220d9f0ff06Facial Tic Detection Using Computer Vision +
Christopher D. Leveille +
March 20, 2014 +
2fda461869f84a9298a0e93ef280f79b9fb76f94OpenFace: an open source facial behavior analysis toolkit +
Tadas Baltruˇsaitis +
2fdce3228d384456ea9faff108b9c6d0cf39e7c7
2f7e9b45255c9029d2ae97bbb004d6072e70fa79Noname manuscript No. +
(will be inserted by the editor) +
cvpaper.challenge in 2015 +
A review of CVPR2015 and DeepSurvey +
Nakamura +
Received: date / Accepted: date +
2f489bd9bfb61a7d7165a2f05c03377a00072477JIA, YANG: STRUCTURED SEMI-SUPERVISED FOREST +
Structured Semi-supervised Forest for +
Facial Landmarks Localization with Face +
Mask Reasoning +
1 Department of Computer Science +
The Univ. of Hong Kong, HK +
2 School of EECS +
Queen Mary Univ. of London, UK +
Angran Lin1 +
2f16459e2e24dc91b3b4cac7c6294387d4a0eacf
2f59f28a1ca3130d413e8e8b59fb30d50ac020e2Children Gender Recognition Under Unconstrained +
Conditions Based on Contextual Information +
Joint Research Centre, European Commission, Ispra, Italy +
2f88d3189723669f957d83ad542ac5c2341c37a5Downloaded From: https://www.spiedigitallibrary.org/journals/Journal-of-Electronic-Imaging on 9/13/2018 +
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use +
Attribute-correlatedlocalregionsfordeeprelativeattributeslearningFenZhangXiangweiKongZeJiaFenZhang,XiangweiKong,ZeJia,“Attribute-correlatedlocalregionsfordeeprelativeattributeslearning,”J.Electron.Imaging27(4),043021(2018),doi:10.1117/1.JEI.27.4.043021.
2fda164863a06a92d3a910b96eef927269aeb730Names and Faces in the News +
Computer Science Division +
U.C. Berkeley +
Berkeley, CA 94720 +
2fa057a20a2b4a4f344988fee0a49fce85b0dc33
2f8ef26bfecaaa102a55b752860dbb92f1a11dc6A Graph Based Approach to Speaker Retrieval in Talk +
Show Videos with Transcript-Based Supervision +
2f184c6e2c31d23ef083c881de36b9b9b6997ce9Polichotomies on Imbalanced Domains +
by One-per-Class Compensated Reconstruction Rule +
Integrated Research Centre, Universit´a Campus Bio-Medico of Rome, Rome, Italy +
2f9c173ccd8c1e6b88d7fb95d6679838bc9ca51d
2f8183b549ec51b67f7dad717f0db6bf342c9d02
2fa1fc116731b2b5bb97f06d2ac494cb2b2fe475A novel approach to personal photo album representation +
and management +
Universit`a di Palermo - Dipartimento di Ingegneria Informatica +
Viale delle Scienze, 90128, Palermo, Italy +
2f882ceaaf110046e63123b495212d7d4e99f33dHigh Frequency Component Compensation based Super-resolution +
Algorithm for Face Video Enhancement +
CVRR Lab, UC San Diego, La Jolla, CA 92093, USA +
2f95340b01cfa48b867f336185e89acfedfa4d92Face Expression Recognition with a 2-Channel +
Convolutional Neural Network +

Vogt-K¨olln-Straße 30, 22527 Hamburg, Germany +
http://www.informatik.uni-hamburg.de/WTM/ +
2fea258320c50f36408032c05c54ba455d575809
2faa09413162b0a7629db93fbb27eda5aeac54caNISTIR 7674 +
Quantifying How Lighting and Focus +
Affect Face Recognition Performance +
Phillips, P. J. +
Beveridge, J. R. +
Draper, B. +
Bolme, D. +
Givens, G. H. +
Lui, Y. M. +
1 +
433bb1eaa3751519c2e5f17f47f8532322abbe6d
4300fa1221beb9dc81a496cd2f645c990a7ede53
439ac8edfa1e7cbc65474cab544a5b8c4c65d5dbSIViP (2011) 5:401–413 +
DOI 10.1007/s11760-011-0244-6 +
ORIGINAL PAPER +
Face authentication with undercontrolled pose and illumination +
Received: 15 September 2010 / Revised: 14 December 2010 / Accepted: 17 February 2011 / Published online: 7 August 2011 +
© Springer-Verlag London Limited 2011 +
43f6953804964037ff91a4f45d5b5d2f8edfe4d5Multi-Feature Fusion in Advanced Robotics Applications +
Institut für Informatik +
Technische Universität München +
D-85748 Garching, Germany +
439ec47725ae4a3660e509d32828599a495559bfFacial Expressions Tracking and Recognition: Database Protocols for Systems Validation +
and Evaluation +
43a03cbe8b704f31046a5aba05153eb3d6de4142Towards Robust Face Recognition from Video +
Image Science and Machine Vision Group +
Oak Ridge National Laboratory +
Oak Ridge, TN 37831-6010 +
43836d69f00275ba2f3d135f0ca9cf88d1209a87Ozaki et al. IPSJ Transactions on Computer Vision and +
Applications (2017) 9:20 +
DOI 10.1186/s41074-017-0030-7 +
IPSJ Transactions on Computer +
Vision and Applications +
RESEARCH PAPER +
Open Access +
Effective hyperparameter optimization +
using Nelder-Mead method in deep learning +
43aa40eaa59244c233f83d81f86e12eba8d74b59
4362368dae29cc66a47114d5ffeaf0534bf0159cUACEE International Journal of Artificial Intelligence and Neural Networks ISSN:- 2250-3749 (online) +
Performance Analysis of FDA Based Face +
Recognition Using Correlation, ANN and SVM +
Department of Computer Engineering +
Department of Computer Engineering +
Department of Computer Engineering +
Anand, INDIA +
Anand, INDIA +
Anand, INDIA +
43e268c118ac25f1f0e984b57bc54f0119ded520
43476cbf2a109f8381b398e7a1ddd794b29a9a16A Practical Transfer Learning Algorithm for Face Verification +
David Wipf +
4353d0dcaf450743e9eddd2aeedee4d01a1be78bLearning Discriminative LBP-Histogram Bins +
for Facial Expression Recognition +
Philips Research, High Tech Campus 36, Eindhoven 5656 AE, The Netherlands +
437a720c6f6fc1959ba95e48e487eb3767b4e508
436d80cc1b52365ed7b2477c0b385b6fbbb51d3b
43b8b5eeb4869372ef896ca2d1e6010552cdc4d4Large-scale Supervised Hierarchical Feature Learning for Face Recognition +
Intel Labs China +
43ae4867d058453e9abce760ff0f9427789bab3a951 +
Graph Embedded Nonparametric Mutual +
Information For Supervised +
Dimensionality Reduction +
430c4d7ad76e51d83bbd7ec9d3f856043f054915
438b88fe40a6f9b5dcf08e64e27b2719940995e0Building a Classi(cid:2)cation Cascade for Visual Identi(cid:2)cation from One Example +
Computer Science, U.C. Berkeley +
Computer Science, UMass Amherst +
Computer Science, U.C. Berkeley +
43fb9efa79178cb6f481387b7c6e9b0ca3761da8Mixture of Parts Revisited: Expressive Part Interactions for Pose Estimation +
Anoop R Katti +
IIT Madras +
Chennai, India +
IIT Madras +
Chennai, India +
43d7d0d0d0e2d6cf5355e60c4fe5b715f0a1101aPobrane z czasopisma Annales AI- Informatica http://ai.annales.umcs.pl +
Data: 04/05/2018 16:53:32 +
U M CS +
889bc64c7da8e2a85ae6af320ae10e05c4cd6ce7174 +
Using Support Vector Machines to Enhance the +
Performance of Bayesian Face Recognition +
8812aef6bdac056b00525f0642702ecf8d57790bA Unified Features Approach to Human Face Image +
Analysis and Interpretation +
Department of Informatics, +
Technische Universit¨at M¨unchen +
85748 Garching, Germany +
881066ec43bcf7476479a4146568414e419da804From Traditional to Modern : Domain Adaptation for +
Action Classification in Short Social Video Clips +
Center for Visual Information Technology, IIIT Hyderabad, India +
8813368c6c14552539137aba2b6f8c55f561b75fTrunk-Branch Ensemble Convolutional Neural +
Networks for Video-based Face Recognition +
883006c0f76cf348a5f8339bfcb649a3e46e2690Weakly Supervised Pain Localization using Multiple Instance Learning +
88f2952535df5859c8f60026f08b71976f8e19ecA neural network framework for face +
recognition by elastic bunch graph matching +
8818b12aa0ff3bf0b20f9caa250395cbea0e8769Fashion Conversation Data on Instagram +
∗Graduate School of Culture Technology, KAIST, South Korea +
†Department of Communication Studies, UCLA, USA +
8878871ec2763f912102eeaff4b5a2febfc22fbe3781 +
Human Action Recognition in Unconstrained +
Videos by Explicit Motion Modeling +
8855d6161d7e5b35f6c59e15b94db9fa5bbf2912COGNITION IN PREGNANCY AND THE POSTPARTUM PERIOD +
88bee9733e96958444dc9e6bef191baba4fa6efaExtending Face Identification to +
Open-Set Face Recognition +
Department of Computer Science +
Universidade Federal de Minas Gerais +
Belo Horizonte, Brazil +
88fd4d1d0f4014f2b2e343c83d8c7e46d198cc79978-1-4799-9988-0/16/$31.00 ©2016 IEEE +
2697 +
ICASSP 2016 +
9fa1be81d31fba07a1bde0275b9d35c528f4d0b8Identifying Persons by Pictorial and +
Contextual Cues +
Nicholas Leonard Pi¨el +
Thesis submitted for the degree of Master of Science +
Supervisor: +
April 2009 +
9f094341bea610a10346f072bf865cb550a1f1c1Recognition and Volume Estimation of Food Intake using a Mobile Device +
Sarnoff Corporation +
201 Washington Rd, +
Princeton, NJ, 08540 +
6b333b2c6311e36c2bde920ab5813f8cfcf2b67b
6b9aa288ce7740ec5ce9826c66d059ddcfd8dba9
6b089627a4ea24bff193611e68390d1a4c3b3644CROSS-POLLINATION OF NORMALISATION +
TECHNIQUES FROM SPEAKER TO FACE +
AUTHENTICATION USING GAUSSIAN +
MIXTURE MODELS +
Idiap-RR-03-2012 +
JANUARY 2012 +
Centre du Parc, Rue Marconi 19, P.O. Box 592, CH - 1920 Martigny +
6be0ab66c31023762e26d309a4a9d0096f72a7f0Enhance Visual Recognition under Adverse +
Conditions via Deep Networks +
6b18628cc8829c3bf851ea3ee3bcff8543391819Face recognition based on subset selection via metric learning on manifold. +
1058. [doi:10.1631/FITEE.1500085] +
Face recognition based on subset +
selection via metric learning on manifold +
Key words: Face recognition, Sparse representation, Manifold structure, +
Metric learning, Subset selection +
ORCID: http://orcid.org/0000-0001-7441-4749 +
Front Inform Technol & Electron Eng
6b1b43d58faed7b457b1d4e8c16f5f7e7d819239
6b35b15ceba2f26cf949f23347ec95bbbf7bed64
6b6493551017819a3d1f12bbf922a8a8c8cc2a03Pose Normalization for Local Appearance-Based +
Face Recognition +
Computer Science Department, Universit¨at Karlsruhe (TH) +
Am Fasanengarten 5, Karlsruhe 76131, Germany +
http://isl.ira.uka.de/cvhci +
6bb630dfa797168e6627d972560c3d438f71ea99
0728f788107122d76dfafa4fb0c45c20dcf523caThe Best of Both Worlds: Combining Data-independent and Data-driven +
Approaches for Action Recognition +
071099a4c3eed464388c8d1bff7b0538c7322422FACIAL EXPRESSION RECOGNITION IN THE WILD USING RICH DEEP FEATURES +
Microsoft Advanced Technology labs, Microsoft Technology and Research, Cairo, Egypt +
+
071af21377cc76d5c05100a745fb13cb2e40500f
0754e769eb613fd3968b6e267a301728f52358beTowards a Watson That Sees: Language-Guided Action Recognition for +
Robots +
0717b47ab84b848de37dbefd81cf8bf512b544acInternational Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 +
International Conference on Humming Bird ( 01st March 2014) +
RESEARCH ARTICLE +
OPEN ACCESS +
Robust Face Recognition and Tagging in Visual Surveillance +
System +
0750a816858b601c0dbf4cfb68066ae7e788f05dCosFace: Large Margin Cosine Loss for Deep Face Recognition +
Tencent AI Lab +
0716e1ad868f5f446b1c367721418ffadfcf0519Interactively Guiding Semi-Supervised +
Clustering via Attribute-Based Explanations +
Virginia Tech, Blacksburg, VA, USA +
073eaa49ccde15b62425cda1d9feab0fea03a842
0726a45eb129eed88915aa5a86df2af16a09bcc1Introspective Perception: Learning to Predict Failures in Vision Systems +
38d56ddcea01ce99902dd75ad162213cbe4eaab7Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
2648 +
389334e9a0d84bc54bcd5b94b4ce4c5d9d6a2f26FACIAL PARAMETER EXTRACTION SYSTEM BASED ON ACTIVE CONTOURS +
Universitat Politècnica de Catalunya, Barcelona, Spain +
380dd0ddd5d69adc52defc095570d1c22952f5cc
38679355d4cfea3a791005f211aa16e76b2eaa8dTitle +
Evolutionary cross-domain discriminative Hessian Eigenmaps +
Author(s) +
Si, S; Tao, D; Chan, KP +
Citation +
1086 +
Issued Date +
2010 +
URL +
http://hdl.handle.net/10722/127357 +
Rights +
This work is licensed under a Creative Commons Attribution- +
NonCommercial-NoDerivatives 4.0 International License.; ©2010 +
IEEE. Personal use of this material is permitted. However, +
permission to reprint/republish this material for advertising or +
promotional purposes or for creating new collective works for +
resale or redistribution to servers or lists, or to reuse any +
copyrighted component of this work in other works must be +
obtained from the IEEE. +
38682c7b19831e5d4f58e9bce9716f9c2c29c4e7International Journal of Computer Trends and Technology (IJCTT) – Volume 18 Number 5 – Dec 2014 +
Movie Character Identification Using Graph Matching +
Algorithm +
M.Tech Scholar, Dept of CSE, QISCET, ONGOLE, Dist: Prakasam, AP, India. +
Associate Professor, Department of CSE, QISCET, ONGOLE, Dist: Prakasam, AP, India +
3803b91e784922a2dacd6a18f61b3100629df932Temporal Multimodal Fusion +
for Video Emotion Classification in the Wild +
Orange Labs +
Cesson-Sévigné, France +
Orange Labs +
Cesson-Sévigné, France +
Normandie Univ., UNICAEN, +
ENSICAEN, CNRS +
Caen, France +
38eea307445a39ee7902c1ecf8cea7e3dcb7c0e7Noname manuscript No. +
(will be inserted by the editor) +
Multi-distance Support Matrix Machine +
Received: date / Accepted: date +
385750bcf95036c808d63db0e0b14768463ff4c6
384f972c81c52fe36849600728865ea50a0c46701 +
Multi-Fold Gabor, PCA and ICA Filter +
Convolution Descriptor for Face Recognition +
+
380d5138cadccc9b5b91c707ba0a9220b0f39271Deep Imbalanced Learning for Face Recognition +
and Attribute Prediction +
38861d0d3a0292c1f54153b303b0d791cbba1d50
38192a0f9261d9727b119e294a65f2e25f72d7e6
00fb2836068042c19b5197d0999e8e93b920eb9c
0077cd8f97cafd2b389783858a6e4ab7887b0b6bMAI et al.: ON THE RECONSTRUCTION OF DEEP FACE TEMPLATES +
On the Reconstruction of Deep Face Templates +
00214fe1319113e6649435cae386019235474789Bachelorarbeit im Fach Informatik +
Face Recognition using +
Distortion Models +
Mathematik, Informatik und Naturwissenschaften der +
RHEINISCH-WESTFÄLISCHEN TECHNISCHEN HOCHSCHULE AACHEN +
Der Fakultät für +
Lehrstuhl für Informatik VI +
Prof. Dr.-Ing. H. Ney +
vorgelegt von: +
Matrikelnummer 252400 +
Gutachter: +
Prof. Dr.-Ing. H. Ney +
Prof. Dr. B. Leibe +
Betreuer: +
September 2009 +
0004f72a00096fa410b179ad12aa3a0d10fc853c
00f0ed04defec19b4843b5b16557d8d0ccc5bb42
0037bff7be6d463785d4e5b2671da664cd7ef746Author manuscript, published in "European Conference on Computer Vision (ECCV '10) 6311 (2010) 634--647" +
DOI : 10.1007/978-3-642-15549-9_46 +
00d9d88bb1bdca35663946a76d807fff3dc1c15fSubjects and Their Objects: Localizing Interactees for a +
Person-Centric View of Importance +
00a967cb2d18e1394226ad37930524a31351f6cfFully-adaptive Feature Sharing in Multi-Task Networks with Applications in +
Person Attribute Classification +
UC San Diego +
IBM Research +
IBM Research +
Binghamton Univeristy, SUNY +
UC San Diego +
Rogerio Feris +
IBM Research +
00a3cfe3ce35a7ffb8214f6db15366f4e79761e3Kinect for real-time emotion recognition via facial expressions. Frontiers of +
Information Technology & Electronic Engineering, 16(4):272-282. +
[doi:10.1631/FITEE.1400209] +
Using Kinect for real-time emotion +
recognition via facial expressions +
Key words: Kinect, Emotion recognition, Facial expression, Real-time +
classification, Fusion algorithm, Support vector machine (SVM) +
ORCID: http://orcid.org/0000-0002-5021-9057 +
Front Inform Technol & Electron Eng
004a1bb1a2c93b4f379468cca6b6cfc6d8746cc4Balanced k-Means and Min-Cut Clustering +
00d94b35ffd6cabfb70b9a1d220b6823ae9154eeDiscriminative Bayesian Dictionary Learning +
for Classification +
006f283a50d325840433f4cf6d15876d475bba77756 +
Preserving Structure in Model-Free Tracking +
0059b3dfc7056f26de1eabaafd1ad542e34c2c2e
6e198f6cc4199e1c4173944e3df6f39a302cf787MORPH-II: Inconsistencies and Cleaning Whitepaper +
NSF-REU Site at UNC Wilmington, Summer 2017 +
6eaf446dec00536858548fe7cc66025b70ce20eb
6e91be2ad74cf7c5969314b2327b513532b1be09Dimensionality Reduction with Subspace Structure +
Preservation +
Department of Computer Science +
SUNY Buffalo +
Buffalo, NY 14260 +
6eba25166fe461dc388805cc2452d49f5d1cdaddPages 122.1-122.12 +
DOI: https://dx.doi.org/10.5244/C.30.122 +
6e8a81d452a91f5231443ac83e4c0a0db4579974Illumination robust face representation based on intrinsic geometrical +
information +
Soyel, H; Ozmen, B; McOwan, PW +
This is a pre-copyedited, author-produced PDF of an article accepted for publication in IET +
Conference on Image Processing (IPR 2012). The version of record is available +
http://ieeexplore.ieee.org/document/6290632/?arnumber=6290632&tag=1 +
For additional information about this publication click this link. +
http://qmro.qmul.ac.uk/xmlui/handle/123456789/16147 +
Information about this research object was correct at the time of download; we occasionally +
make corrections to records, please therefore check the published record when citing. For +
6ecd4025b7b5f4894c990614a9a65e3a1ac347b2International Journal on Recent and Innovation Trends in Computing and Communication +
+
ISSN: 2321-8169 +
Volume: 2 Issue: 5 +
1275– 1281 +
_______________________________________________________________________________________________ +
Automatic Naming of Character using Video Streaming for Face +
Recognition with Graph Matching +
Nivedita.R.Pandey +
Ranjan.P.Dahake +
PG Student at MET’s IOE Bhujbal Knowledge City, +
PG Student at MET’s IOE Bhujbal Knowledge City, +
Nasik, Maharashtra, India, +
Nasik, Maharashtra, India, +
6eaeac9ae2a1697fa0aa8e394edc64f32762f578
6ee2ea416382d659a0dddc7a88fc093accc2f8ee
6e3a181bf388dd503c83dc324561701b19d37df1Finding a low-rank basis in a matrix subspace +
Andr´e Uschmajew +
6e8c3b7d25e6530a631ea01fbbb93ac1e8b69d2fDeep Episodic Memory: Encoding, Recalling, and Predicting +
Episodic Experiences for Robot Action Execution +
6e911227e893d0eecb363015754824bf4366bdb7Wasserstein Divergence for GANs +
1 Computer Vision Lab, ETH Zurich, Switzerland +
2 VISICS, KU Leuven, Belgium +
6ee8a94ccba10062172e5b31ee097c846821a822Submitted 3/13; Revised 10/13; Published 12/13 +
How to Solve Classification and Regression Problems on +
High-Dimensional Data with a Supervised +
Extension of Slow Feature Analysis +
Institut f¨ur Neuroinformatik +
Ruhr-Universit¨at Bochum +
Bochum D-44801, Germany +
Editor: David Dunson +
6e379f2d34e14efd85ae51875a4fa7d7ae63a662A NEW MULTI-MODAL BIOMETRIC SYSTEM +
BASED ON FINGERPRINT AND FINGER +
VEIN RECOGNITION +
Master's Thesis +
Department of Software Engineering +
JULY-2014 +
I +
6e1802874ead801a7e1072aa870681aa2f555f351­4244­0728­1/07/$20.00 ©2007 IEEE +
I ­ 629 +
ICASSP 2007 +
6ed22b934e382c6f72402747d51aa50994cfd97bCustomized Expression Recognition for Performance-Driven +
Cutout Character Animation +
†NEC Laboratories America +
‡Snapchat +
6e93fd7400585f5df57b5343699cb7cda20cfcc2http://journalofvision.org/9/2/22/ +
Comparing a novel model based on the transferable +
belief model with humans during the recognition of +
partially occluded facial expressions +
Département de Psychologie, Université de Montréal, +
Canada +
Département de Psychologie, Université de Montréal, +
Canada +
Département de Psychologie, Université de Montréal, +
Canada +
Humans recognize basic facial expressions effortlessly. Yet, despite a considerable amount of research, this task remains +
elusive for computer vision systems. Here, we compared the behavior of one of the best computer models of facial +
expression recognition (Z. Hammal, L. Couvreur, A. Caplier, & M. Rombaut, 2007) with the behavior of human observers +
during the M. Smith, G. Cottrell, F. Gosselin, and P. G. Schyns (2005) facial expression recognition task performed on +
stimuli randomly sampled using Gaussian apertures. The modelVwhich we had to significantly modify in order to give the +
ability to deal with partially occluded stimuliVclassifies the six basic facial expressions (Happiness, Fear, Sadness, +
Surprise, Anger, and Disgust) plus Neutral from static images based on the permanent facial feature deformations and the +
Transferable Belief Model (TBM). Three simulations demonstrated the suitability of the TBM-based model to deal with +
partially occluded facial parts and revealed the differences between the facial information used by humans and by the +
model. This opens promising perspectives for the future development of the model. +
Keywords: facial features behavior, facial expressions classification, Transferable Belief Model, Bubbles +
Citation: Hammal, Z., Arguin, M., & Gosselin, F. (2009). Comparing a novel model based on the transferable belief +
http://journalofvision.org/9/2/22/, doi:10.1167/9.2.22. +
Introduction +
Facial expressions communicate information from +
which we can quickly infer the state of mind of our peers +
and adjust our behavior accordingly (Darwin, 1872). To +
illustrate, take a person like patient SM with complete +
bilateral damage to the amygdala nuclei that prevents her +
from recognizing facial expressions of fear. SM would be +
incapable of interpreting the fearful expression on the face +
of a bystander, who has encountered a furious Grizzly +
bear, as a sign of potential +
threat (Adolphs, Tranel, +
Damasio, & Damasio, 1994). +
Facial expressions are typically arranged into six +
universally recognized basic categories Happiness, Sur- +
prise, Disgust, Anger, Sadness, and Fear that are similarly +
expressed across different backgrounds and cultures +
(Cohn, 2006; Ekman, 1999; Izard, 1971, 1994). Facial +
expressions result +
from the precisely choreographed +
deformation of facial features, which are often described +
using the 46 Action Units (AUs; Ekman & Friesen, +
1978). +
Facial expression recognition and computer +
vision +
The study of human facial expressions has an impact in +
several areas of life such as art, social interaction, cognitive +
science, medicine, security, affective computing, and +
human-computer interaction (HCI). An automatic facial +
expressions classification system may contribute signifi- +
cantly to the development of all these disciplines. However, +
the development of such a system constitutes a significant +
challenge because of the many constraints that are imposed +
by its application in a real-world context (Pantic & Bartlett, +
2007; Pantic & Patras, 2006). In particular, such systems +
need to provide great accuracy and robustness without +
demanding too many interventions from the user. +
There have been major advances in computer vision +
over the past 15 years for the recognition of the six basic +
facial expressions (for reviews, see Fasel & Luettin, 2003; +
Pantic & Rothkrantz, 2000b). The main approaches can be +
divided in two classes: Model-based and fiducial points +
approaches. The model-based approach requires the +
design of a deterministic physical model that can represent +
doi: 10.1167/9.2.22 +
Received January 28, 2008; published February 26, 2009 +
ISSN 1534-7362 * ARVO +
6e12ba518816cbc2d987200c461dc907fd19f533
9ab463d117219ed51f602ff0ddbd3414217e3166Weighted Transmedia +
Relevance Feedback for +
Image Retrieval and +
Auto-annotation +
TECHNICAL +
REPORT +
N° 0415 +
December 2011 +
Project-Teams LEAR - INRIA +
and TVPA - XRCE +
9ac82909d76b4c902e5dde5838130de6ce838c16Recognizing Facial Expressions Automatically +
from Video +
1 Introduction +
Facial expressions, resulting from movements of the facial muscles, are the face +
changes in response to a person’s internal emotional states, intentions, or social +
communications. There is a considerable history associated with the study on fa- +
cial expressions. Darwin (1872) was the first to describe in details the specific fa- +
cial expressions associated with emotions in animals and humans, who argued that +
all mammals show emotions reliably in their faces. Since that, facial expression +
analysis has been a area of great research interest for behavioral scientists (Ekman, +
Friesen, and Hager, 2002). Psychological studies (Mehrabian, 1968; Ambady and +
Rosenthal, 1992) suggest that facial expressions, as the main mode for non-verbal +
communication, play a vital role in human face-to-face communication. For illus- +
tration, we show some examples of facial expressions in Fig. 1. +
Computer recognition of facial expressions has many important applications in +
intelligent human-computer interaction, computer animation, surveillance and se- +
curity, medical diagnosis, law enforcement, and awareness systems (Shan, 2007). +
Therefore, it has been an active research topic in multiple disciplines such as psy- +
chology, cognitive science, human-computer interaction, and pattern recognition. +
Meanwhile, as a promising unobtrusive solution, automatic facial expression analy- +
sis from video or images has received much attention in last two decades (Pantic and +
Rothkrantz, 2000a; Fasel and Luettin, 2003; Tian, Kanade, and Cohn, 2005; Pantic +
and Bartlett, 2007). +
This chapter introduces recent advances in computer recognition of facial expres- +
sions. Firstly, we describe the problem space, which includes multiple dimensions: +
level of description, static versus dynamic expression, facial feature extraction and +
9ac15845defcd0d6b611ecd609c740d41f0c341dCopyright +
by +
2011 +
9af1cf562377b307580ca214ecd2c556e20df000Feb. 28 +
International Journal of Advanced Studies in Computer Science and Engineering +
IJASCSE, Volume 4, Issue 2, 2015 +
Video-Based Facial Expression Recognition +
Using Local Directional Binary Pattern +
Electrical Engineering Dept., AmirKabir Univarsity of Technology +
Tehran, Iran +
9a23a0402ae68cc6ea2fe0092b6ec2d40f667adbHigh-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs +
1NVIDIA Corporation +
2UC Berkeley +
Figure 1: We propose a generative adversarial framework for synthesizing 2048 × 1024 images from semantic label maps +
(lower left corner in (a)). Compared to previous work [5], our results express more natural textures and details. (b) We can +
change labels in the original label map to create new scenes, like replacing trees with buildings. (c) Our framework also +
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. +
Please visit our website for more side-by-side comparisons as well as interactive editing demos. +
9a7858eda9b40b16002c6003b6db19828f94a6c6MOONEY FACE CLASSIFICATION AND PREDICTION BY LEARNING ACROSS TONE +
(cid:63) UC Berkeley / †ICSI +
9a276c72acdb83660557489114a494b86a39f6ffEmotion Classification through Lower Facial Expressions using Adaptive +
Support Vector Machines +
Department of Information Technology, Faculty of Industrial Technology and Management, +
9a42c519f0aaa68debbe9df00b090ca446d25bc4Face Recognition via Centralized Coordinate +
Learning +
9aad8e52aff12bd822f0011e6ef85dfc22fe8466Temporal-Spatial Mapping for Action Recognition +
36b40c75a3e53c633c4afb5a9309d10e12c292c7
3646b42511a6a0df5470408bc9a7a69bb3c5d742International Journal of Computer Applications (0975 – 8887) +
Applications of Computers and Electronics for the Welfare of Rural Masses (ACEWRM) 2015 +
Detection of Facial Parts based on ABLATA +
Technical Campus, Bhilai +
Vikas Singh +
Technical Campus, Bhilai +
Abha Choubey +
Technical Campus, Bhilai +
365f67fe670bf55dc9ccdcd6888115264b2a2c56
36fe39ed69a5c7ff9650fd5f4fe950b5880760b0Tracking von Gesichtsmimik +
mit Hilfe von Gitterstrukturen +
zur Klassifikation von schmerzrelevanten Action +
Units +
1Fraunhofer-Institut f¨ur Integrierte Schaltungen IIS, Erlangen, +
2Otto-Friedrich-Universit¨at Bamberg, 3Universit¨atsklinkum Erlangen +
Kurzfassung. In der Schmerzforschung werden schmerzrelevante Mi- +
mikbewegungen von Probanden mittels des Facial Action Coding System +
klassifiziert. Die manuelle Klassifikation hierbei ist aufw¨andig und eine +
automatische (Vor-)klassifikation k¨onnte den diagnostischen Wert dieser +
Analysen erh¨ohen sowie den klinischen Workflow unterst¨utzen. Der hier +
vorgestellte regelbasierte Ansatz erm¨oglicht eine automatische Klassifika- +
tion ohne große Trainingsmengen vorklassifizierter Daten. Das Verfahren +
erkennt und verfolgt Mimikbewegungen, unterst¨utzt durch ein Gitter, +
und ordnet diese Bewegungen bestimmten Gesichtsarealen zu. Mit die- +
sem Wissen kann aus den Bewegungen auf die zugeh¨origen Action Units +
geschlossen werden. +
1 Einleitung +
Menschliche Empfindungen wie Emotionen oder Schmerz l¨osen spezifische Mu- +
ster von Kontraktionen der Gesichtsmuskulatur aus, die Grundlage dessen sind, +
was wir Mimik nennen. Aus der Beobachtung der Mimik kann wiederum auf +
menschliche Empfindungen r¨uckgeschlossen werden. Im Rahmen der Schmerz- +
forschung werden Videoaufnahmen von Probanden hinsichtlich des mimischen +
Schmerzausdrucks analysiert. Zur Beschreibung des mimischen Ausdrucks und +
dessen Ver¨anderungen wird das Facial Action Coding System (FACS) [1] verwen- +
det, das anatomisch begr¨undet, kleinste sichtbare Muskelbewegungen im Gesicht +
beschreibt und als einzelne Action Units (AUs) kategorisiert. Eine Vielzahl von +
Untersuchungen hat gezeigt, dass spezifische Muster von Action Units auftre- +
ten, wenn Probanden Schmerzen angeben [2]. Die manuelle Klassifikation und +
Markierung der Action Units von Probanden in Videosequenzen bedarf einer +
langwierigen Beobachtung durch ausgebildete FACS-Coder. Eine automatische +
(Vor-)klassifikation kann hierbei den klinischen Workflow unterst¨utzen und dieses +
Verfahren zum brauchbaren diagnostischen Instrument machen. Bisher realisier- +
te Ans¨atze zum Erkennen von Gesichtsausdr¨ucken basieren auf der Klassifikation +
36ce0b68a01b4c96af6ad8c26e55e5a30446f360Multimed Tools Appl +
DOI 10.1007/s11042-014-2322-6 +
Facial expression recognition based on a mlp neural +
network using constructive training algorithm +
Received: 5 February 2014 / Revised: 22 August 2014 / Accepted: 13 October 2014 +
© Springer Science+Business Media New York 2014 +
3674f3597bbca3ce05e4423611d871d09882043bISSN 1796-2048 +
Volume 7, Number 4, August 2012 +
Contents +
Special Issue: Multimedia Contents Security in Social Networks Applications +
Guest Editors: Zhiyong Zhang and Muthucumaru Maheswaran +
Guest Editorial +
Zhiyong Zhang and Muthucumaru Maheswaran +
SPECIAL ISSUE PAPERS +
DRTEMBB: Dynamic Recommendation Trust Evaluation Model Based on Bidding +
Gang Wang and Xiao-lin Gui +
Block-Based Parallel Intra Prediction Scheme for HEVC +
Jie Jiang, Baolong, Wei Mo, and Kefeng Fan +
Optimized LSB Matching Steganography Based on Fisher Information +
Yi-feng Sun, Dan-mei Niu, Guang-ming Tang, and Zhan-zhan Gao +
A Novel Robust Zero-Watermarking Scheme Based on Discrete Wavelet Transform +
Yu Yang, Min Lei, Huaqun Liu, Yajian Zhou, and Qun Luo +
Stego Key Estimation in LSB Steganography +
Jing Liu and Guangming Tang +
REGULAR PAPERS +
Facial Expression Spacial Charts for Describing Dynamic Diversity of Facial Expressions +
277 +
279 +
289 +
295 +
303 +
309 +
314 +
362a70b6e7d55a777feb7b9fc8bc4d40a57cde8c978-1-4799-9988-0/16/$31.00 ©2016 IEEE +
2792 +
ICASSP 2016 +
366d20f8fd25b4fe4f7dc95068abc6c6cabe1194
3630324c2af04fd90f8668f9ee9709604fe980fdThis article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TCSVT.2016.2607345, IEEE +
Transactions on Circuits and Systems for Video Technology +
Image Classification with Tailored Fine-Grained +
Dictionaries +
362ba8317aba71c78dafca023be60fb71320381d
36cf96fe11a2c1ea4d999a7f86ffef6eea7b5958RGB-D Face Recognition with Texture and +
Attribute Features +
Member, IEEE +
36018404263b9bb44d1fddaddd9ee9af9d46e560OCCLUDED FACE RECOGNITION BY USING GABOR +
FEATURES +
1 Department of Electrical And Electronics Engineering, METU, Ankara, Turkey +
2 7h%ł7$.(cid:3)%ł/7(1(cid:15)(cid:3)$QNDUD(cid:15)(cid:3)7XUNH\ +
5c4ce36063dd3496a5926afd301e562899ff53ea
5c2a7518fb26a37139cebff76753d83e4da25159
5c2e264d6ac253693469bd190f323622c457ca05978-1-4799-2341-0/13/$31.00 ©2013 IEEE +
4367 +
ICIP 2013 +
5c473cfda1d7c384724fbb139dfe8cb39f79f626
5c5e1f367e8768a9fb0f1b2f9dbfa060a22e75c02132 +
Reference Face Graph for Face Recognition +
5c35ac04260e281141b3aaa7bbb147032c887f0cFace Detection and Tracking Control with Omni Car +
CS 231A Final Report +
June 31, 2016 +
5c717afc5a9a8ccb1767d87b79851de8d3016294978-1-4673-0046-9/12/$26.00 ©2012 IEEE +
1845 +
ICASSP 2012 +
0952ac6ce94c98049d518d29c18d136b1f04b0c0
09137e3c267a3414314d1e7e4b0e3a4cae801f45Noname manuscript No. +
(will be inserted by the editor) +
Two Birds with One Stone: Transforming and Generating +
Facial Images with Iterative GAN +
Received: date / Accepted: date +
09926ed62511c340f4540b5bc53cf2480e8063f8Action Tubelet Detector for Spatio-Temporal Action Localization +
09718bf335b926907ded5cb4c94784fd20e5ccd8875 +
Recognizing Partially Occluded, Expression Variant +
Faces From Single Training Image per Person +
With SOM and Soft k-NN Ensemble +
0903bb001c263e3c9a40f430116d1e629eaa616fCVPR +
#987 +
000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
045 +
046 +
047 +
048 +
049 +
050 +
051 +
052 +
053 +
CVPR 2009 Submission #987. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +
An Empirical Study of Context in Object Detection +
Anonymous CVPR submission +
Paper ID 987 +
09df62fd17d3d833ea6b5a52a232fc052d4da3f5ISSN: 1405-5546 +
Instituto Politécnico Nacional +
México +
+
Rivas Araiza, Edgar A.; Mendiola Santibañez, Jorge D.; Herrera Ruiz, Gilberto; González Gutiérrez, +
Carlos A.; Trejo Perea, Mario; Ríos Moreno, G. J. +
Mejora de Contraste y Compensación en Cambios de la Iluminación +
Instituto Politécnico Nacional +
Distrito Federal, México +
Disponible en: http://www.redalyc.org/articulo.oa?id=61509703 +
Cómo citar el artículo +
Número completo +
Más información del artículo +
Página de la revista en redalyc.org +
Sistema de Información Científica +
Red de Revistas Científicas de América Latina, el Caribe, España y Portugal +
Proyecto académico sin fines de lucro, desarrollado bajo la iniciativa de acceso abierto +
097104fc731a15fad07479f4f2c4be2e071054a2
09f853ce12f7361c4b50c494df7ce3b9fad1d221myjournal manuscript No. +
(will be inserted by the editor) +
Random forests for real time 3D face analysis +
Received: date / Accepted: date +
09111da0aedb231c8484601444296c50ca0b5388
09750c9bbb074bbc4eb66586b20822d1812cdb20978-1-4673-0046-9/12/$26.00 ©2012 IEEE +
1385 +
ICASSP 2012 +
097f674aa9e91135151c480734dda54af5bc4240Proc. VIIth Digital Image Computing: Techniques and Applications, Sun C., Talbot H., Ourselin S. and Adriaansen T. (Eds.), 10-12 Dec. 2003, Sydney +
Face Recognition Based on Multiple Region Features +
CSIRO Telecommunications & Industrial Physics +
Australia +
Tel: 612 9372 4104, Fax: 612 9372 4411, Email: +
5d485501f9c2030ab33f97972aa7585d3a0d59a7
5de5848dc3fc35e40420ffec70a407e4770e3a8dWebVision Database: Visual Learning and Understanding from Web Data +
1 Computer Vision Laboratory, ETH Zurich +
2 Google Switzerland +
5da139fc43216c86d779938d1c219b950dd82a4c1-4244-1437-7/07/$20.00 ©2007 IEEE +
II - 205 +
ICIP 2007 +
5dc056fe911a3e34a932513abe637076250d96da
5d233e6f23b1c306cf62af49ce66faac2078f967RESEARCH ARTICLE +
Optimal Geometrical Set for Automated +
Marker Placement to Virtualized Real-Time +
Facial Emotions +
School of Mechatronic Engineering, Universiti Malaysia Perlis, 02600, Ulu Pauh, Arau, Perlis, West Malaysia +
5d7f8eb73b6a84eb1d27d1138965eb7aef7ba5cfRobust Registration of Dynamic Facial Sequences +
5dcf78de4d3d867d0fd4a3105f0defae2234b9cb
5db4fe0ce9e9227042144758cf6c4c2de2042435INTERNATIONAL JOURNAL OF ELECTRICAL AND ELECTRONIC SYSTEMS RESEARCH, VOL.3, JUNE 2010 +
Recognition of Facial Expression Using Haar +
Wavelet Transform +
for +
paper +
features +
investigates +
+
5d5cd6fa5c41eb9d3d2bab3359b3e5eb60ae194eFace Recognition Algorithms +
June 16, 2010 +
Ion Marqu´es +
Supervisor: +
Manuel Gra˜na +
5d09d5257139b563bd3149cfd5e6f9eae3c34776Optics Communications 338 (2015) 77–89 +
Contents lists available at ScienceDirect +
Optics Communications +
journal homepage: www.elsevier.com/locate/optcom +
Pattern recognition with composite correlation filters designed with +
multi-objective combinatorial optimization +
a Instituto Politécnico Nacional – CITEDI, Ave. del Parque 1310, Mesade Otay, Tijuana B.C. 22510, México +
b Department of Computer Science, CICESE, Carretera Ensenada-Tijuana 3918, Ensenada B.C. 22860, México +
c Instituto Tecnológico de Tijuana, Blvd. Industrial y Ave. ITR TijuanaS/N, Mesa de Otay, Tijuana B.C. 22500, México +
d National Ignition Facility, Lawrence Livermore National Laboratory, Livermore, CA 94551, USA +
a r t i c l e i n f o +
a b s t r a c t +
Article history: +
Received 12 July 2014 +
Accepted 16 November 2014 +
Available online 23 October 2014 +
Keywords: +
Object recognition +
Composite correlation filters +
Multi-objective evolutionary algorithm +
Combinatorial optimization +
Composite correlation filters are used for solving a wide variety of pattern recognition problems. These +
filters are given by a combination of several training templates chosen by a designer in an ad hoc manner. +
In this work, we present a new approach for the design of composite filters based on multi-objective +
combinatorial optimization. Given a vast search space of training templates, an iterative algorithm is used +
to synthesize a filter with an optimized performance in terms of several competing criteria. Moreover, by +
employing a suggested binary-search procedure a filter bank with a minimum number of filters can be +
constructed, for a prespecified trade-off of performance metrics. Computer simulation results obtained +
with the proposed method in recognizing geometrically distorted versions of a target in cluttered and +
noisy scenes are discussed and compared in terms of recognition performance and complexity with +
existing state-of-the-art filters. +
& Elsevier B.V. All rights reserved. +
1. +
Introduction +
Nowadays, object recognition receives much research interest +
due to its high impact in real-life activities, such as robotics, bio- +
metrics, and target tracking [1,2]. Object recognition consists in +
solving two essential tasks: detection of a target within an ob- +
served scene and determination of the exact position of the de- +
tected object. Different approaches can be utilized to address these +
tasks, that is feature-based methods [3–6] and template matching +
algorithms [7,8]. In feature-based methods the observed scene is +
processed to extract relevant features of potential targets within +
the scene. Next, the extracted features are processed and analyzed +
to make decisions. Feature-based methods yield good results in +
many applications. However, they depend on several subjective +
decisions which often require optimization [9,10]. On the other +
hand, correlation filtering is a template matching processing. In +
this approach, the coordinates of the maximum of the filter output +
are taken as estimates of the target coordinates in the observed +
scene. Correlation filters possess a good mathematical basis and +
they can be implemented by exploiting massive parallelism either +
in hybrid opto-digital correlators [11,12] or in high-performance +
n Corresponding author. Tel.: þ52 664 623 1344x82856. +
http://dx.doi.org/10.1016/j.optcom.2014.10.038 +
0030-4018/& Elsevier B.V. All rights reserved. +
hardware such as graphics processing units (GPUs) [13] or field +
programmable gate arrays (FPGAs) [14] at high rate. Additionally, +
these filters are capable to reliably recognize a target in highly +
cluttered and noisy environments [8,15,16]. Moreover, they are +
able to estimate very accurately the position of the target within +
the scene [17]. Correlation filters are usually designed by a opti- +
mization of various criteria [18,19]. The filters can be broadly +
classified in to two main categories: analytical and composite fil- +
ters. Analytical filters optimize a performance criterion using +
mathematical models of signals and noise [20,21]. Composite fil- +
ters are constructed by combination of several training templates, +
each of them representing an expected target view in the observed +
scene [22,21]. In practice, composite filters are effective for real- +
life degradations of targets such as rotations and scaling. Compo- +
site filters are synthesized by means of a supervised training +
process. Thus, the performance of the filters highly depends on a +
proper selection of image templates used for training [20,23]. +
Normally, the training templates are chosen by a designer in an ad +
hoc manner. Such a subjective procedure is not optimal. In addi- +
tion, Kumar and Pochavsky [24] showed that the signal to noise +
ratio of a composite filter gradually reduces when the number of +
training templates increases. In order to synthesize composite +
filters with improved performance in terms of several competing +
metrics, a search and optimization strategy is required to auto- +
matically choose the set of training templates. +
5d01283474b73a46d80745ad0cc0c4da14aae194
5d197c8cd34473eb6cde6b65ced1be82a3a1ed14AFaceImageDatabaseforEvaluatingOut-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
31aa20911cc7a2b556e7d273f0bdd5a2f0671e0a
31b05f65405534a696a847dd19c621b7b8588263
318e7e6daa0a799c83a9fdf7dd6bc0b3e89ab24aSparsity in Dynamics of Spontaneous +
Subtle Emotions: Analysis & Application +
31c0968fb5f587918f1c49bf7fa51453b3e89cf7Deep Transfer Learning for Person Re-identification +
31e57fa83ac60c03d884774d2b515813493977b9
316e67550fbf0ba54f103b5924e6537712f06beeMultimodal semi-supervised learning +
for image classification +
LEAR team, INRIA Grenoble, France +
31ef5419e026ef57ff20de537d82fe3cfa9ee741Facial Expression Analysis Based on +
High Dimensional Binary Features +
´Ecole Polytechique de Montr´eal, Universit´e de Montr´eal, Montr´eal, Canada +
31b58ced31f22eab10bd3ee2d9174e7c14c27c01
31ace8c9d0e4550a233b904a0e2aabefcc90b0e3Learning Deep Face Representation +
Megvii Inc. +
Megvii Inc. +
Megvii Inc. +
Megvii Inc. +
Megvii Inc. +
312afff739d1e0fcd3410adf78be1c66b3480396
31bb49ba7df94b88add9e3c2db72a4a98927bb05
91811203c2511e919b047ebc86edad87d985a4faExpression Subspace Projection for Face +
Recognition from Single Sample per Person +
91e57667b6fad7a996b24367119f4b22b6892ecaProbabilistic Corner Detection for Facial Feature +
Extraction +
Article +
Accepted version +
E. Ardizzone, M. La Cascia, M. Morana +
In Lecture Notes in Computer Science Volume 5716, 2009 +
It is advisable to refer to the publisher's version if you intend to cite +
from the work. +
Publisher: Springer +
http://link.springer.com/content/pdf/10.1007%2F978-3- +
642-04146-4_50.pdf +
91883dabc11245e393786d85941fb99a6248c1fb
917bea27af1846b649e2bced624e8df1d9b79d6fUltra Power-Efficient CNN Domain Specific Accelerator with 9.3TOPS/Watt for +
Mobile and Embedded Applications +
Gyrfalcon Technology Inc. +
1900 McCarthy Blvd. Milpitas, CA 95035 +
91b1a59b9e0e7f4db0828bf36654b84ba53b0557This 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 +
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < +
 +
Simultaneous Hallucination and Recognition of +
Low-Resolution Faces Based on Singular Value +
Decomposition +
(SVD) +
for performing both +
911bef7465665d8b194b6b0370b2b2389dfda1a1RANJAN, ROMERO, BLACK: LEARNING HUMAN OPTICAL FLOW +
Learning Human Optical Flow +
1 MPI for Intelligent Systems +
Tübingen, Germany +
2 Amazon Inc. +
91ead35d1d2ff2ea7cf35d15b14996471404f68dCombining and Steganography of 3D Face Textures +
919d0e681c4ef687bf0b89fe7c0615221e9a1d30
912a6a97af390d009773452814a401e258b77640
91d513af1f667f64c9afc55ea1f45b0be7ba08d4Automatic Face Image Quality Prediction +
918b72a47b7f378bde0ba29c908babf6dab6f833
91e58c39608c6eb97b314b0c581ddaf7daac075ePixel-wise Ear Detection with Convolutional +
Encoder-Decoder Networks +
91d2fe6fdf180e8427c65ffb3d895bf9f0ec4fa0
9131c990fad219726eb38384976868b968ee9d9cDeep Facial Expression Recognition: A Survey +
915d4a0fb523249ecbc88eb62cb150a60cf60fa0Comparison of Feature Extraction Techniques in Automatic +
Face Recognition Systems for Security Applications +
S . Cruz-Llanas, J. Ortega-Garcia, E. Martinez-Torrico, J. Gonzalez-Rodriguez +
Dpto. Ingenieria Audiovisual y Comunicaciones, EUIT Telecomunicacion, Univ. PolitCcnica de Madrid, Spain +
http://www.atvs.diac.upm.es +
65b737e5cc4a565011a895c460ed8fd07b333600Transfer Learning For Cross-Dataset Recognition: A Survey +
This paper summarises and analyses the cross-dataset recognition transfer learning techniques with the +
emphasis on what kinds of methods can be used when the available source and target data are presented +
in different forms for boosting the target task. This paper for the first time summarises several transferring +
criteria in details from the concept level, which are the key bases to guide what kind of knowledge to transfer +
between datasets. In addition, a taxonomy of cross-dataset scenarios and problems is proposed according the +
properties of data that define how different datasets are diverged, thereby review the recent advances on +
each specific problem under different scenarios. Moreover, some real world applications and corresponding +
commonly used benchmarks of cross-dataset recognition are reviewed. Lastly, several future directions are +
identified. +
Additional Key Words and Phrases: Cross-dataset, transfer learning, domain adaptation +
1. INTRODUCTION +
It has been explored how human would transfer learning in one context to another +
similar context [Woodworth and Thorndike 1901; Perkins et al. 1992] in the field of +
Psychology and Education. For example, learning to drive a car helps a person later +
to learn more quickly to drive a truck, and learning mathematics prepares students to +
study physics. The machine learning algorithms are mostly inspired by human brains. +
However, most of them require a huge amount of training examples to learn a new +
model from scratch and fail to apply knowledge learned from previous domains or +
tasks. This may be due to that a basic assumption of statistical learning theory is +
that the training and test data are drawn from the same distribution and belong to +
the same task. Intuitively, learning from scratch is not realistic and practical, because +
it violates how human learn things. In addition, manually labelling a large amount +
of data for new domain or task is labour extensive, especially for the modern “data- +
hungry” and “data-driven” learning techniques (i.e. deep learning). However, the big +
data era provides a huge amount available data collected for other domains and tasks. +
Hence, how to use the previously available data smartly for the current task with +
scarce data will be beneficial for real world applications. +
To reuse the previous knowledge for current tasks, the differences between old data +
and new data need to be taken into account. Take the object recognition as an ex- +
ample. As claimed by Torralba and Efros [2011], despite the great efforts of object +
datasets creators, the datasets appear to have strong build-in bias caused by various +
factors, such as selection bias, capture bias, category or label bias, and negative set +
bias. This suggests that no matter how big the dataset is, it is impossible to cover +
the complexity of the real visual world. Hence, the dataset bias needs to be consid- +
ered before reusing data from previous datasets. Pan and Yang [2010] summarise that +
the differences between different datasets can be caused by domain divergence (i.e. +
distribution shift or feature space difference) or task divergence (i.e. conditional dis- +
tribution shift or label space difference), or both. For example, in visual recognition, +
the distributions between the previous and current data can be discrepant due to the +
different environments, lighting, background, sensor types, resolutions, view angles, +
and post-processing. Those external factors may cause the distribution divergence or +
even feature space divergence between different domains. On the other hand, the task +
divergence between current and previous data is also ubiquitous. For example, it is +
highly possible that an animal species that we want to recognize have not been seen +
ACM Journal Name, Vol. V, No. N, Article A, Publication date: January YYYY. +
6582f4ec2815d2106957215ca2fa298396dde274JUNE 2007 +
1005 +
Discriminative Learning and Recognition +
of Image Set Classes Using +
Canonical Correlations +
655d9ba828eeff47c600240e0327c3102b9aba7cIEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 35, NO. 3, JUNE 2005 +
489 +
Kernel Pooled Local Subspaces for Classification +
656a59954de3c9fcf82ffcef926af6ade2f3fdb5Convolutional Network Representation +
for Visual Recognition +
Doctoral Thesis +
Stockholm, Sweden, 2017 +
656f05741c402ba43bb1b9a58bcc5f7ce2403d9a
65817963194702f059bae07eadbf6486f18f4a0ahttp://dx.doi.org/10.1007/s11263-015-0814-0 +
WhittleSearch: Interactive Image Search with Relative Attribute +
Feedback +
Received: date / Accepted: date +
6581c5b17db7006f4cc3575d04bfc6546854a785Contextual Person Identification +
in Multimedia Data +
zur Erlangung des akademischen Grades eines +
Doktors der Ingenieurwissenschaften +
der Fakultät für Informatik +
des Karlsruher Instituts für Technologie (KIT) +
genehmigte +
Dissertation +
von +
aus Erlangen +
Tag der mündlichen Prüfung: +
18. November 2014 +
Hauptreferent: +
Korreferent: +
Prof. Dr. Rainer Stiefelhagen +
Karlsruher Institut für Technologie +
Prof. Dr. Gerhard Rigoll +
Technische Universität München +
KIT – Universität des Landes Baden-Württemberg und nationales Forschungszentrum in der Helmholtz-Gemeinschaft +
www.kit.edu +
65babb10e727382b31ca5479b452ee725917c739Label Distribution Learning +
62dccab9ab715f33761a5315746ed02e48eed2a0A Short Note about Kinetics-600 +
Jo˜ao Carreira +
62d1a31b8acd2141d3a994f2d2ec7a3baf0e6dc4Ding et al. EURASIP Journal on Image and Video Processing (2017) 2017:43 +
DOI 10.1186/s13640-017-0188-z +
EURASIP Journal on Image +
and Video Processing +
R ES EAR CH +
Noise-resistant network: a deep-learning +
method for face recognition under noise +
Open Access +
62694828c716af44c300f9ec0c3236e98770d7cfPadrón-Rivera, G., Rebolledo-Mendez, G., Parra, P. P., & Huerta-Pacheco, N. S. (2016). Identification of Action Units Related to +
Identification of Action Units Related to Affective States in a Tutoring System +
1Facultad de Estadística e Informática, Universidad Veracruzana, Mexico // 2Universidad Juárez Autónoma de +
for Mathematics +
Huerta-Pacheco1 +
*Corresponding author +
620339aef06aed07a78f9ed1a057a25433faa58b
62b3598b401c807288a113796f424612cc5833ca
628a3f027b7646f398c68a680add48c7969ab1d9Plan for Final Year Project: +
HKU-Face: A Large Scale Dataset for Deep Face +
Recognition +
3035140108 +
3035141841 +
Introduction +
Face recognition has been one of the most successful techniques in the field of artificial intelligence +
because of its surpassing human-level performance in academic experiments and broad application in +
the industrial world. Gaussian-face[1] and Facenet[2] hold state-of-the-art record using statistical +
method and deep-learning method respectively. What’s more, face recognition has been applied +
in various areas like authority checking and recording, fostering a large number of start-ups like +
Face++. +
Our final year project will deal with the face recognition task by building a large-scaled and carefully- +
filtered dataset. Our project plan specifies our roadmap and current research process. This plan first +
illustrates the significance and potential enhancement in constructing large-scale face dataset for +
both academics and companies. Then objectives to accomplish and related literature review will be +
expressed in detail. Next, methodologies used, scope of our project and challenges faced by us are +
described. The detailed timeline for this project follows as well as a small summary. +
2 Motivation +
Nowadays most of the face recognition tasks are supervised learning tasks which use dataset annotated +
by human beings. This contains mainly two drawbacks: (1) limited size of dataset due to limited +
human effort; (2) accuracy problem resulted from human perceptual bias. +
Parkhi et al.[3] discuss the first problem, showing that giant companies hold private face databases +
with larger size of data (See the comparison in Table 1). Other research institution could only get +
access to public but smaller databases like LFW[4, 5], which acts like a barricade to even higher +
performance. +
Dataset +
IJB-A [6] +
LFW [4, 5] +
YFD [7] +
CelebFaces [8] +
CASIA-WebFace [9] +
MS-Celeb-1M [10] +
Facebook +
Google +
Availability +
public +
public +
public +
public +
public +
public +
private +
private +
identities +
500 +
5K +
1595 +
10K +
10K +
100K +
4K +
8M +
images +
5712 +
13K +
3425 videos +
202K +
500K +
about 10M +
4400K +
100-200M +
Table 1: Face recognition datasets +
6257a622ed6bd1b8759ae837b50580657e676192
626859fe8cafd25da13b19d44d8d9eb6f0918647Activity Recognition based on a +
Magnitude-Orientation Stream Network +
Smart Surveillance Interest Group, Department of Computer Science +
Universidade Federal de Minas Gerais, Belo Horizonte, Brazil +
620e1dbf88069408b008347cd563e16aeeebeb83
62007c30f148334fb4d8975f80afe76e5aef8c7fEye In-Painting with Exemplar Generative Adversarial Networks +
Facebook Inc. +
1 Hacker Way, Menlo Park (CA), USA +
62a30f1b149843860938de6dd6d1874954de24b7418 +
Fast Algorithm for Updating the Discriminant Vectors +
of Dual-Space LDA +
62e0380a86e92709fe2c64e6a71ed94d152c6643Facial Emotion Recognition With Expression Energy +
Albert Cruz +
Center for Research in +
Intelligent Systems +
216 Winston Chung Hall +
Center for Research in +
Intelligent Systems +
216 Winston Chung Hall +
Center for Research in +
Intelligent Systems +
216 Winston Chung Hall +
Riverside, CA, 92521-0425, +
Riverside, CA, 92521-0425, +
Riverside, CA, 92521-0425, +
USA +
USA +
USA +
961a5d5750f18e91e28a767b3cb234a77aac8305Face Detection without Bells and Whistles +
1 ESAT-PSI/VISICS, iMinds, KU Leuven, Belgium +
2 MPI Informatics, Saarbrücken, Germany +
3 D-ITET/CVL, ETH Zürich, Switzerland +
9626bcb3fc7c7df2c5a423ae8d0a046b2f69180cUPTEC STS 17033 +
Examensarbete 30 hp +
November 2017 +
A deep learning approach for +
action classification in American +
football video sequences +
9696b172d66e402a2e9d0a8d2b3f204ad8b98cc4J Inf Process Syst, Vol.9, No.1, March 2013 +
pISSN 1976-913X +
eISSN 2092-805X +
Region-Based Facial Expression Recognition in +
Still Images +
964a3196d44f0fefa7de3403849d22bbafa73886
9606b1c88b891d433927b1f841dce44b8d3af066Principal Component Analysis with Tensor Train +
Subspace +
96b1000031c53cd4c1c154013bb722ffd87fa7daContextVP: Fully Context-Aware Video +
Prediction +
1 NVIDIA, Santa Clara, CA, USA +
2 ETH Zurich, Zurich, Switzerland +
3 The Swiss AI Lab IDSIA, Manno, Switzerland +
4 NNAISENSE, Lugano, Switzerland +
968f472477a8afbadb5d92ff1b9c7fdc89f0c009Firefly-based Facial Expression Recognition +
9636c7d3643fc598dacb83d71f199f1d2cc34415
3a2fc58222870d8bed62442c00341e8c0a39ec87Probabilistic Local Variation +
Segmentation +
Technion - Computer Science Department - M.Sc. Thesis MSC-2014-02 - 2014
3abc833f4d689f37cc8a28f47fb42e32deaa4b17Noname manuscript No. +
(will be inserted by the editor) +
Large Scale Retrieval and Generation of Image Descriptions +
Received: date / Accepted: date +
3a60678ad2b862fa7c27b11f04c93c010cc6c430JANUARY-MARCH 2012 +
A Multimodal Database for +
Affect Recognition and Implicit Tagging +
3a0a839012575ba455f2b84c2d043a35133285f9444 +
Proceedings of the 2011 Conference on Empirical Methods in Natural Language Processing, pages 444–454, +
Edinburgh, Scotland, UK, July 27–31, 2011. c(cid:13)2011 Association for Computational Linguistics +
3a9681e2e07be7b40b59c32a49a6ff4c40c962a2Biometrics & Biostatistics International Journal +
Comparing treatment means: overlapping standard +
errors, overlapping confidence intervals, and tests of +
hypothesis +
3a846704ef4792dd329a5c7a2cb8b330ab6b8b4ein any current or +
future media, +
for all other uses, +
© 2010 IEEE. Personal use of this material is permitted. Permission from IEEE must be +
obtained +
including +
reprinting/republishing this material for advertising or promotional purposes, creating +
new collective works, for resale or redistribution to servers or lists, or reuse of any +
copyrighted component of this work in other works. +
Pre-print of article that appeared at the IEEE Computer Society Workshop on Biometrics +
2010. +
The published article can be accessed from: +
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5544597 +
3a95eea0543cf05670e9ae28092a114e3dc3ab5cConstructing the L2-Graph for Robust Subspace +
Learning and Subspace Clustering +
3a4f522fa9d2c37aeaed232b39fcbe1b64495134ISSN (Online) 2321 – 2004 +
ISSN (Print) 2321 – 5526 +
INTERNATIONAL JOURNAL OF INNOVATIVE RESEARCH IN ELECTRICAL, ELECTRONICS, INSTRUMENTATION AND CONTROL ENGINEERING +
Vol. 4, Issue 5, May 2016 +
IJIREEICE +
Face Recognition and Retrieval Using Cross +
Age Reference Coding +
Sricharan H S1, Srinidhi K S1, Rajath D N1, Tejas J N1, Chandrakala B M2 +
BE, DSCE, Bangalore1 +
Assistant Professor, DSCE, Bangalore2 +
54969bcd728b0f2d3285866c86ef0b4797c2a74dIEEE TRANSACTION SUBMISSION +
Learning for Video Compression +
5456166e3bfe78a353df988897ec0bd66cee937fImproved Boosting Performance by Exclusion +
of Ambiguous Positive Examples +
Computer Vision and Active Perception, KTH, Stockholm 10800, Sweden +
Keywords: +
Boosting, Image Classification, Algorithm Evaluation, Dataset Pruning, VOC2007. +
54aacc196ffe49b3450059fccdf7cd3bb6f6f3c3A Joint Learning Framework for Attribute Models and Object Descriptions +
Dhruv Mahajan +
Yahoo! Labs, Bangalore, India +
541bccf19086755f8b5f57fd15177dc49e77d675
549c719c4429812dff4d02753d2db11dd490b2aeYouTube-BoundingBoxes: A Large High-Precision +
Human-Annotated Data Set for Object Detection in Video +
Google Brain +
Google Brain +
Google Research +
Google Brain +
Google Brain +
98b2f21db344b8b9f7747feaf86f92558595990c
988d1295ec32ce41d06e7cf928f14a3ee079a11eSemantic Deep Learning +
September 29, 2015 +
981449cdd5b820268c0876477419cba50d5d1316Learning Deep Features for One-Class +
Classification +
98127346920bdce9773aba6a2ffc8590b9558a4aNoname manuscript No. +
(will be inserted by the editor) +
Efficient Human Action Recognition using +
Histograms of Motion Gradients and +
VLAD with Descriptor Shape Information +
Received: date / Accepted: date +
982fed5c11e76dfef766ad9ff081bfa25e62415a
98519f3f615e7900578bc064a8fb4e5f429f3689Dictionary-based Domain Adaptation Methods +
for the Re-identification of Faces +
9825aa96f204c335ec23c2b872855ce0c98f9046International Journal of Ethics in Engineering & Management Education +
Website: www.ijeee.in (ISSN: 2348-4748, Volume 1, Issue 5, May2014) +
FACE AND FACIAL EXPRESSION +
RECOGNITION IN 3-D USING MASKED +
PROJECTION UNDER OCCLUSION +
Jyoti patil * +
M.Tech (CSE) +
GNDEC Bidar-585401 +
BIDAR, INDIA +
M.Tech (CSE) +
GNDEC Bidar- 585401 +
BIDAR, INDIA +
M.Tech (CSE) +
VKIT, Bangalore- 560040 +
BANGALORE, INDIA +
5334ac0a6438483890d5eef64f6db93f44aacdf4
53dd25350d3b3aaf19beb2104f1e389e3442df61
530243b61fa5aea19b454b7dbcac9f463ed0460e
539ca9db570b5e43be0576bb250e1ba7a727d640
53c8cbc4a3a3752a74f79b74370ed8aeed97db85
5366573e96a1dadfcd4fd592f83017e378a0e185Böhlen, Chandola and Salunkhe +
Server, server in the cloud. +
Who is the fairest in the crowd? +
533bfb82c54f261e6a2b7ed7d31a2fd679c56d18Technical Report MSU-CSE-14-1 +
Unconstrained Face Recognition: Identifying a +
Person of Interest from a Media Collection +
530ce1097d0681a0f9d3ce877c5ba31617b1d709
3fbd68d1268922ee50c92b28bd23ca6669ff87e5598 +
IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 10, NO. 4, APRIL 2001 +
A Shape- and Texture-Based Enhanced Fisher +
Classifier for Face Recognition +
3f22a4383c55ceaafe7d3cfed1b9ef910559d639JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
Robust Kronecker Component Analysis +
3fdcc1e2ebcf236e8bb4a6ce7baf2db817f30001A top-down approach for a synthetic +
autobiographical memory system +
1Sheffield Centre for Robotics (SCentRo), Univ. of Sheffield, Sheffield, S10 2TN, UK +
2Dept. of Computer Science, Univ. of Sheffield, Sheffield, S1 4DP, UK +
3 CVAP Lab, KTH, Stockholm, Sweden +
3f848d6424f3d666a1b6dd405a48a35a797dd147GHODRATI et al.: IS 2D INFORMATION ENOUGH FOR VIEWPOINT ESTIMATION? +
Is 2D Information Enough For Viewpoint +
Estimation? +
KU Leuven, ESAT - PSI, iMinds +
Leuven, Belgium +
3fa738ab3c79eacdbfafa4c9950ef74f115a3d84DaMN – Discriminative and Mutually Nearest: +
Exploiting Pairwise Category Proximity +
for Video Action Recognition +
1 Center for Research in Computer Vision at UCF, Orlando, USA +
2 Google Research, Mountain View, USA +
http://crcv.ucf.edu/projects/DaMN/ +
3fb98e76ffd8ba79e1c22eda4d640da0c037e98aConvolutional Neural Networks for Crop Yield Prediction using Satellite Images +
H. Russello +
3f5cf3771446da44d48f1d5ca2121c52975bb3d3
3f14b504c2b37a0e8119fbda0eff52efb2eb24615727 +
Joint Facial Action Unit Detection and Feature +
Fusion: A Multi-Conditional Learning Approach +
3f9a7d690db82cf5c3940fbb06b827ced59ec01eVIP: Finding Important People in Images +
Virginia Tech +
Google Inc. +
Virginia Tech +
Project: https://computing.ece.vt.edu/~mclint/vip/ +
Demo: http://cloudcv.org/vip/ +
3fd90098551bf88c7509521adf1c0ba9b5dfeb57Page 1 of 21 +
*****For Peer Review Only***** +
10 +
11 +
12 +
13 +
14 +
15 +
16 +
17 +
18 +
19 +
20 +
21 +
22 +
23 +
24 +
25 +
26 +
27 +
28 +
29 +
30 +
31 +
32 +
33 +
34 +
35 +
36 +
37 +
38 +
39 +
40 +
41 +
42 +
43 +
44 +
45 +
46 +
47 +
48 +
49 +
50 +
51 +
52 +
53 +
54 +
55 +
56 +
57 +
58 +
59 +
60 +
Attribute-Based Classification for Zero-Shot +
Visual Object Categorization +
3f63f9aaec8ba1fa801d131e3680900680f14139Facial Expression Recognition using Local Binary +
Patterns and Kullback Leibler Divergence +
AnushaVupputuri, SukadevMeher +
+
divergence. +
role +
3f0e0739677eb53a9d16feafc2d9a881b9677b63Efficient Two-Stream Motion and Appearance 3D CNNs for +
Video Classification +
ESAT-KU Leuven +
Ali Pazandeh +
Sharif UTech +
ESAT-KU Leuven, ETH Zurich +
30870ef75aa57e41f54310283c0057451c8c822bOvercoming Catastrophic Forgetting with Hard Attention to the Task +
303065c44cf847849d04da16b8b1d9a120cef73a
3046baea53360a8c5653f09f0a31581da384202eDeformable Face Alignment via Local +
Measurements and Global Constraints +
3028690d00bd95f20842d4aec84dc96de1db6e59Leveraging Union of Subspace Structure to Improve Constrained Clustering +
30c96cc041bafa4f480b7b1eb5c45999701fe0661090 +
Discrete Cosine Transform Locality-Sensitive +
Hashes for Face Retrieval +
306957285fea4ce11a14641c3497d01b46095989FACE RECOGNITION UNDER VARYING LIGHTING BASED ON +
DERIVATES OF LOG IMAGE +
2ICT-ISVISION Joint R&D Laboratory for Face Recognition, CAS, Beijing 100080, China +
1Graduate School, CAS, Beijing, 100039, China +
302c9c105d49c1348b8f1d8cc47bead70e2acf08This 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 +
Transactions on Circuits and Systems for Video Technology +
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY +
Unconstrained Face Recognition Using A Set-to-Set +
Distance Measure +
304a306d2a55ea41c2355bd9310e332fa76b3cb0
5e7e055ef9ba6e8566a400a8b1c6d8f827099553
5e28673a930131b1ee50d11f69573c17db8fff3eAuthor manuscript, published in "Workshop on Faces in 'Real-Life' Images: Detection, Alignment, and Recognition, Marseille : France +
(2008)" +
5e6ba16cddd1797853d8898de52c1f1f44a73279Face Identification with Second-Order Pooling +
5e821cb036010bef259046a96fe26e681f20266e
5bfc32d9457f43d2488583167af4f3175fdcdc03International Journal of Science and Research (IJSR), India Online ISSN: 2319-7064 +
Local Gray Code Pattern (LGCP): A Robust +
Feature Descriptor for Facial Expression +
Recognition +
5ba7882700718e996d576b58528f1838e5559225This 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 +
Transactions on Affective Computing +
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, VOL. X, NO. X, OCTOBER 2016 +
Predicting Personalized Image Emotion +
Perceptions in Social Networks +
5bb684dfe64171b77df06ba68997fd1e8daffbe1
5bae9822d703c585a61575dced83fa2f4dea1c6dMOTChallenge 2015: +
Towards a Benchmark for Multi-Target Tracking +
5babbad3daac5c26503088782fd5b62067b94fa5Are You Sure You Want To Do That? +
Classification with Verification +
5b9d9f5a59c48bc8dd409a1bd5abf1d642463d65Evolving Systems. manuscript No. +
(will be inserted by the editor) +
An evolving spatio-temporal approach for gender and age +
group classification with Spiking Neural Networks +
Received: date / Accepted: date +
5bf70c1afdf4c16fd88687b4cf15580fd2f26102Accepted in Pattern Recognition Letters +
Pattern Recognition Letters +
journal homepage: www.elsevier.com +
Residual Codean Autoencoder for Facial Attribute Analysis +
IIIT-Delhi, New Delhi, India +
Article history: +
Received 29 March 2017 +
5b2cfee6e81ef36507ebf3c305e84e9e0473575a
5be3cc1650c918da1c38690812f74573e66b1d32Relative Parts: Distinctive Parts for Learning Relative Attributes +
Center for Visual Information Technology, IIIT Hyderabad, India - 500032 +
5b0ebb8430a04d9259b321fc3c1cc1090b8e600e
3765c26362ad1095dfe6744c6d52494ea106a42c
3727ac3d50e31a394b200029b2c350073c1b69e3
37f2e03c7cbec9ffc35eac51578e7e8fdfee3d4eWACV +
#394 +
000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
045 +
046 +
047 +
048 +
049 +
050 +
051 +
052 +
053 +
WACV 2015 Submission #394. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +
Co-operative Pedestrians Group Tracking in Crowded Scenes using an MST +
Approach +
Anonymous WACV submission +
Paper ID 394 +
377a1be5113f38297716c4bb951ebef7a93f949aDear Faculty, IGERT Fellows, IGERT Associates and Students, +
You are cordially invited to attend a Seminar presented by Albert Cruz. Please +
plan to attend. +
Albert Cruz +
IGERT Fellow +
Electrical Engineering +
+
Date: Friday, October 11, 2013 +
Location: Bourns A265 +
Time: 11:00am +
Facial emotion recognition with anisotropic +
inhibited gabor energy histograms +
377c6563f97e76a4dc836a0bd23d7673492b1aae
370e0d9b89518a6b317a9f54f18d5398895a7046IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY, VOL. X, NO. X, XXXXXXX 20XX +
Cross-pollination of normalisation techniques +
from speaker to face authentication +
using Gaussian mixture models +
and S´ebastien Marcel, Member, IEEE +
37eb666b7eb225ffdafc6f318639bea7f0ba9a24MSU Technical Report (2014): MSU-CSE-14-5 +
Age, Gender and Race Estimation from +
Unconstrained Face Images +
375435fb0da220a65ac9e82275a880e1b9f0a557This 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 +
From Pixels to Response Maps: Discriminative Image +
Filtering for Face Alignment in the Wild +
37b6d6577541ed991435eaf899a2f82fdd72c790Vision-based Human Gender Recognition: A Survey +
Universiti Tunku Abdul Rahman, Kuala Lumpur, Malaysia. +
370b5757a5379b15e30d619e4d3fb9e8e13f3256Labeled Faces in the Wild: A Database for Studying +
Face Recognition in Unconstrained Environments +
08d2f655361335bdd6c1c901642981e650dff5ecThis is the published version:   +
 Arandjelovic, Ognjen and Cipolla, R. 2006, Automatic cast listing in feature‐length films with +
Anisotropic Manifold Space, in CVPR 2006 : Proceedings of the Computer Vision and Pattern +
Recognition Conference 2006, IEEE, Piscataway, New Jersey, pp. 1513‐1520. +
+
http://hdl.handle.net/10536/DRO/DU:30058435 +
Reproduced with the kind permission of the copyright owner. +
Copyright : 2006, IEEE +
Available from Deakin Research Online:  +
08ae100805d7406bf56226e9c3c218d3f9774d19Gavrilescu and Vizireanu EURASIP Journal on Image and Video Processing (2017) 2017:59 +
DOI 10.1186/s13640-017-0211-4 +
EURASIP Journal on Image +
and Video Processing +
R ES EAR CH +
Predicting the Sixteen Personality Factors +
(16PF) of an individual by analyzing facial +
features +
Open Access +
08c18b2f57c8e6a3bfe462e599a6e1ce03005876A Least-Squares Framework +
for Component Analysis +
081a431107eb38812b74a8cd036ca5e97235b499
0831a511435fd7d21e0cceddb4a532c35700a622
080c204edff49bf85b335d3d416c5e734a861151CLAD: A Complex and Long Activities +
Dataset with Rich Crowdsourced +
Annotations +
Journal Title +
XX(X):1–6 +
c(cid:13)The Author(s) 2016 +
Reprints and permission: +
sagepub.co.uk/journalsPermissions.nav +
DOI: 10.1177/ToBeAssigned +
www.sagepub.com/ +
08f4832507259ded9700de81f5fd462caf0d5be8International Journal of Computer Applications (0975 – 8887) +
Volume 118 – No.14, May 2015 +
Geometric Approach for Human Emotion +
Recognition using Facial Expression +
S. S. Bavkar +
Assistant Professor +
J. S. Rangole +
Assistant Professor +
V. U. Deshmukh +
Assistant Professor +
08d40ee6e1c0060d3b706b6b627e03d4b123377aHuman Action Localization +
with Sparse Spatial Supervision +
08c1f8f0e69c0e2692a2d51040ef6364fb263a40
088aabe3da627432fdccf5077969e3f6402f0a80Under review as a conference paper at ICLR 2018 +
CLASSIFIER-TO-GENERATOR ATTACK: ESTIMATION +
OF TRAINING DATA DISTRIBUTION FROM CLASSIFIER +
Anonymous authors +
Paper under double-blind review +
08903bf161a1e8dec29250a752ce9e2a508a711cJoint Dimensionality Reduction and Metric Learning: A Geometric Take +
08e24f9df3d55364290d626b23f3d42b4772efb6ENHANCING FACIAL EXPRESSION CLASSIFICATION BY INFORMATION +
FUSION +
I. Buciu1, Z. Hammal 2, A. Caplier2, N. Nikolaidis 1, and I. Pitas 1 +

GR-54124, Thessaloniki, Box 451, Greece +
2 Laboratoire des Images et des Signaux / Institut National Polytechnique de Grenoble +
web: http://www.aiia.csd.auth.gr +
38031 Grenoble, France +
web: http://www.lis.inpg.fr +
0830c9b9f207007d5e07f5269ffba003235e4eff
081fb4e97d6bb357506d1b125153111b673cc128
0857281a3b6a5faba1405e2c11f4e17191d3824dChude-Olisah et al. EURASIP Journal on Advances in Signal Processing 2014, 2014:102 +
http://asp.eurasipjournals.com/content/2014/1/102 +
R ES EAR CH +
Face recognition via edge-based Gabor feature +
representation for plastic surgery-altered images +
Open Access +
08f6745bc6c1b0fb68953ea61054bdcdde6d2fc7Understanding Kin Relationships in a Photo +
082ad50ac59fc694ba4369d0f9b87430553b11db
6dd052df6b0e89d394192f7f2af4a3e3b8f89875International Journal of Engineering and Advanced Technology (IJEAT) +
ISSN: 2249 – 8958, Volume-2, Issue-4, April 2013 +
A literature survey on Facial Expression +
Recognition using Global Features +
 +
6dd5dbb6735846b214be72983e323726ef77c7a9Josai Mathematical Monographs +
vol. 7 (2014), pp. 25-40 +
A Survey on Newer Prospective +
Biometric Authentication Modalities +
6d10beb027fd7213dd4bccf2427e223662e20b7d
6dddf1440617bf7acda40d4d75c7fb4bf9517dbbJOURNAL OF LATEX CLASS FILES, VOL. XX, NO. X, MM YY +
Beyond Counting: Comparisons of Density Maps for Crowd +
Analysis Tasks - Counting, Detection, and Tracking +
6de18708218988b0558f6c2f27050bb4659155e4
6d91da37627c05150cb40cac323ca12a91965759
6d8c9a1759e7204eacb4eeb06567ad0ef4229f93Face Alignment Robust to Pose, Expressions and +
Occlusions +
6d66c98009018ac1512047e6bdfb525c35683b16IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 25, NO. 9, SEPTEMBER 2003 +
1063 +
Face Recognition Based on +
Fitting a 3D Morphable Model +
016cbf0878db5c40566c1fbc237686fbad666a33
01bef320b83ac4405b3fc5b1cff788c124109fb9de Lausanne +
RLC D1 740, CH-1015 +
Lausanne +
de Lausanne +
RLC D1 740, CH-1015 +
Lausanne +
de Lausanne +
RLC D1 740, CH-1015 +
Lausanne +
Translating Head Motion into Attention - Towards +
Processing of Student’s Body-Language +
CHILI Laboratory +
Łukasz Kidzi´nski +
CHILI Laboratory +
CHILI Laboratory +
École polytechnique fédérale +
École polytechnique fédérale +
École polytechnique fédérale +
01c8d7a3460422412fba04e7ee14c4f6cdff9ad7(IJACSA) International Journal of Advanced Computer Science and Applications, +
Vol. 4, No. 7, 2013 +
Rule Based System for Recognizing Emotions Using +
Multimodal Approach +
Information System +
SBM, SVKM’s NMIMS +
Mumbai, India +
+
01e12be4097fa8c94cabeef0ad61498c8e7762f2
0163d847307fae508d8f40ad193ee542c1e051b4JOURNAL OF LATEX CLASS FILES, VOL. 6, NO. 1, JANUARY 2007 +
Classemes and Other Classifier-based +
Features for Efficient Object Categorization +
- Supplementary material - +
1 LOW-LEVEL FEATURES +
We extract the SIFT [1] features for our descriptor +
according to the following pipeline. We first convert +
each image to gray-scale, then we normalize the con- +
trast by forcing the 0.01% of lightest and darkest pixels +
to be mapped to white and black respectively, and +
linearly rescaling the values in between. All images +
exceeding 786,432 pixels of resolution are downsized +
to this maximum value while keeping the aspect ratio. +
The 128-dimensional SIFT descriptors are computed +
from the interest points returned by a DoG detec- +
tor [2]. We finally compute a Bag-Of-Word histogram +
of these descriptors, using a K-means vocabulary of +
500 words. +
2 CLASSEMES +
The LSCOM categories were developed specifically +
for multimedia annotation and retrieval, and have +
been used in the TRECVID video retrieval series. +
We took the LSCOM CYC ontology dated 2006-06-30, +
which contains 2832 unique categories. We removed +
01c4cf9c7c08f0ad3f386d88725da564f3c54679Interpretability Beyond Feature Attribution: +
Quantitative Testing with Concept Activation Vectors (TCAV) +
017ce398e1eb9f2eed82d0b22fb1c21d3bcf9637FACE RECOGNITION WITH HARMONIC DE-LIGHTING +
2ICT-ISVISION Joint R&D Laboratory for Face Recognition, CAS, Beijing, China, 100080 +
1Graduate School, CAS, Beijing, China, 100080 +
Emails: {lyqing, sgshan, wgao}jdl.ac.cn +
014e3d0fa5248e6f4634dc237e2398160294edceInt J Comput Vis manuscript No. +
(will be inserted by the editor) +
What does 2D geometric information really tell us about +
3D face shape? +
Received: date / Accepted: date +
01beab8f8293a30cf48f52caea6ca0fb721c8489
0178929595f505ef7655272cc2c339d7ed0b9507
01b4b32c5ef945426b0396d32d2a12c69c282e29
0113b302a49de15a1d41ca4750191979ad756d2f1­4244­0367­7/06/$20.00 ©2006 IEEE +
537 +
ICME 2006 +
064b797aa1da2000640e437cacb97256444dee82Coarse-to-fine Face Alignment with Multi-Scale Local Patch Regression +
Megvii Inc. +
Megvii Inc. +
Megvii Inc. +
06f146dfcde10915d6284981b6b84b85da75acd4Scalable Face Image Retrieval using +
Attribute-Enhanced Sparse Codewords +
0697bd81844d54064d992d3229162fe8afcd82cbUser-driven mobile robot storyboarding: Learning image interest and +
saliency from pairwise image comparisons +
06e7e99c1fdb1da60bc3ec0e2a5563d05b63fe32WhittleSearch: Image Search with Relative Attribute Feedback +
(Supplementary Material) +
1 Comparative Qualitative Search Results +
We present three qualitative search results for human-generated feedback, in addition to those +
shown in the paper. Each example shows one search iteration, where the 20 reference images are +
randomly selected (rather than ones that match a keyword search, as the image examples in the +
main paper illustrate). For each result, the first figure shows our method and the second figure +
shows the binary feedback result for the corresponding target image. Note that for our method, +
“more/less X” (where X is an attribute) means that the target image is more/less X than the +
reference image which is shown. +
Figures 1 and 2 show results for human-generated relative attribute and binary feedback, re- +
spectively, when both methods are used to target the same “mental image” of a shoe shown in the +
top left bubble. The top right grid of 20 images are the reference images displayed to the user, and +
those outlined and annotated with constraints are the ones chosen by the user to give feedback. +
The bottom row of images in either figure shows the top-ranked images after integrating the user’s +
feedback into the scoring function, revealing the two methods’ respective performance. We see that +
while both methods retrieve high-heeled shoes, only our method retrieves images that are as “open” +
as the target image. This is because using the proposed approach, the user was able to comment +
explicitly on the desired openness property. +
066d71fcd997033dce4ca58df924397dfe0b5fd1(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) +
(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) +
(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) +
(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) +
(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)/ +
(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) +
(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) +
(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) +
(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) +
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) +
((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) +
((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) +
(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) +
(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) +
(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) +
(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:8)(cid:12)(cid:18) 1=(cid:21)(cid:1) ’(cid:16)(cid:4)(cid:11)(cid:4)(cid:1) +
(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) +
(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) +
(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) +
’(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
A(cid:8)(cid:5)(cid:4)(cid:1)<0=(cid:14)(cid:1)(cid:3)(cid:26)’(cid:1)(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) +
(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) +
’(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) +
.(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) +
!(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) +
<(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/, +
(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) +
(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) +
(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) +
/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) +
((cid:15)(cid:11)(cid:4)(cid:11)$(cid:1)(cid:20)(cid:6)-(cid:16)(cid:24)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:5)(cid:4)(cid:25)(cid:24)(cid:1)((cid:20)(cid:15)(cid:25)(cid:6)(cid:5)(cid:4)(cid:14)(cid:1)(cid:20)(cid:6)-(cid:16)(cid:24)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1)(cid:5)(cid:4)(cid:25)(cid:24)(cid:1);)(cid:8)(cid:20)(cid:24)(cid:4)(cid:20)(cid:1)((cid:20)(cid:15)(cid:25)(cid:6)(cid:5)(cid:4)(cid:1)(cid:8)(cid:12)(cid:18)(cid:1) +
(cid:20)(cid:6)-(cid:16)(cid:24)(cid:1) (cid:8)(cid:12)(cid:18)(cid:1) (cid:5)(cid:4)(cid:25)(cid:24)(cid:1) (cid:16)(cid:8)(cid:5)(cid:25)(cid:1) ((cid:20)(cid:15)(cid:25)(cid:6)(cid:5)(cid:4)(cid:21)(cid:1) (cid:26)(cid:12)(cid:1) (cid:24)(cid:16)(cid:4)(cid:11)(cid:4)(cid:1) (cid:27)(cid:8)(cid:24)(cid:4)-(cid:15)(cid:20)(cid:6)(cid:4)(cid:11)(cid:1) (cid:6)(cid:17)(cid:8)-(cid:4)(cid:11)(cid:1) .(cid:4)(cid:20)(cid:4)(cid:1) +
(cid:20)(cid:4)(cid:27)(cid:15)(cid:20)(cid:18)(cid:4)(cid:18)(cid:1)(cid:25)(cid:15)(cid:20)(cid:1)1,0(cid:1)(cid:24)(cid:15)(cid:1)B0,(cid:1)(cid:11))(cid:10) (cid:4)(cid:27)(cid:24)(cid:11)(cid:21) +
(cid:30) (cid:30)(cid:6)(cid:7)#$(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) +
(cid:2) +
’(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) +
#)(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) +
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) +
!(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) +
(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) +
(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,, +
((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) +
(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) +
(cid:8)(((cid:4)(cid:8)(cid:20)(cid:8)(cid:12)(cid:27)(cid:4)(cid:1)<(cid:2)> =(cid:21) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
((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) +
((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) +
(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) +
% (cid:10)(cid:9)(cid:13)(cid:8)(cid:2)(cid:5)&(cid:11)(cid:5)(cid:19)(cid:4)(cid:6) ’((cid:6) +
(cid:10)(cid:3)(cid:11)(cid:3)(cid:12)(cid:3)(cid:13)(cid:9) +
(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: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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
(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) +
.(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) +
(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) +
(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) +
(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) +
’(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) +
(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) +
#(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) +
(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) +
.(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) +
(cid:24)(cid:16)(cid:4)(cid:1) +
(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) +
<(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) +
(cid:6)(cid:17)((cid:15)(cid:20)(cid:24)(cid:8)(cid:12)(cid:24)(cid:1) +
(cid:25)(cid:15)(cid:20)(cid:1) +
(cid:6)(cid:11)(cid:1) +
06526c52a999fdb0a9fd76e84f9795a69480cecf
06fe63b34fcc8ff68b72b5835c4245d3f9b8a016Mach Learn +
DOI 10.1007/s10994-013-5336-9 +
Learning semantic representations of objects +
and their parts +
Received: 24 May 2012 / Accepted: 26 February 2013 +
© The Author(s) 2013 +
06aab105d55c88bd2baa058dc51fa54580746424Image Set based Collaborative Representation for +
Face Recognition +
06262d14323f9e499b7c6e2a3dec76ad9877ba04Real-Time Pose Estimation Piggybacked on Object Detection +
Brno, Czech Republic +
062c41dad67bb68fefd9ff0c5c4d296e796004dcTemporal Generative Adversarial Nets with Singular Value Clipping +
Preferred Networks inc., Japan +
06400a24526dd9d131dfc1459fce5e5189b7baecEvent Recognition in Photo Collections with a Stopwatch HMM +
1Computer Vision Lab +
ETH Z¨urich, Switzerland +
2ESAT, PSI-VISICS +
K.U. Leuven, Belgium +
0653dcdff992ad980cd5ea5bc557efb6e2a53ba1
063a3be18cc27ba825bdfb821772f9f59038c207This is a repository copy of The development of spontaneous facial responses to others’ +
emotions in infancy. An EMG study. +
White Rose Research Online URL for this paper: +
http://eprints.whiterose.ac.uk/125231/ +
Version: Published Version +
Article: +
Kaiser, Jakob, Crespo-Llado, Maria Magdalena, Turati, Chiara et al. (1 more author) +
(2017) The development of spontaneous facial responses to others’ emotions in infancy. +
An EMG study. Scientific Reports. ISSN 2045-2322 +
https://doi.org/10.1038/s41598-017-17556-y +
Reuse +
This article is distributed under the terms of the Creative Commons Attribution (CC BY) licence. This licence +
allows you to distribute, remix, tweak, and build upon the work, even commercially, as long as you credit the +
authors for the original work. More information and the full terms of the licence here: +
https://creativecommons.org/licenses/ +
Takedown +
If you consider content in White Rose Research Online to be in breach of UK law, please notify us by +
https://eprints.whiterose.ac.uk/ +
06a9ed612c8da85cb0ebb17fbe87f5a137541603Deep Learning of Player Trajectory Representations for Team +
Activity Analysis +
06ad99f19cf9cb4a40741a789e4acbf4433c19aeSenTion: A framework for Sensing Facial +
Expressions +
6c304f3b9c3a711a0cca5c62ce221fb098dccff0Attentive Semantic Video Generation using Captions +
IIT Hyderabad +
IIT Hyderabad +
6c2b392b32b2fd0fe364b20c496fcf869eac0a98DOI 10.1007/s00138-012-0423-7 +
ORIGINAL PAPER +
Fully automatic face recognition framework based +
on local and global features +
Received: 30 May 2011 / Revised: 21 February 2012 / Accepted: 29 February 2012 / Published online: 22 March 2012 +
© Springer-Verlag 2012 +
6cddc7e24c0581c50adef92d01bb3c73d8b80b41Face Verification Using the LARK +
Representation +
6c8c7065d1041146a3604cbe15c6207f486021baAttention Modeling for Face Recognition via Deep Learning +
Department of Computing, Hung Hom, Kowloon +
Hong Kong, 999077 CHINA +
Department of Computing, Hung Hom, Kowloon +
Hong Kong, 99907 CHINA +
Department of Computing, Hung Hom, Kowloon +
Hong Kong, 99907 CHINA +
Department of Computing, Hung Hom, Kowloon +
Hong Kong, 99907 CHINA +
390f3d7cdf1ce127ecca65afa2e24c563e9db93bLearning Deep Representation for Face +
Alignment with Auxiliary Attributes +
3918b425bb9259ddff9eca33e5d47bde46bd40aaCopyright +
by +
David Lieh-Chiang Chen +
2012 +
39ce143238ea1066edf0389d284208431b53b802
39ce2232452c0cd459e32a19c1abe2a2648d0c3f
3998c5aa6be58cce8cb65a64cb168864093a9a3e
397aeaea61ecdaa005b09198942381a7a11cd129
39b22bcbd452d5fea02a9ee63a56c16400af2b83
399a2c23bd2592ebe20aa35a8ea37d07c14199da
39c8b34c1b678235b60b648d0b11d241a34c8e32Learning to Deblur Images with Exemplars +
3986161c20c08fb4b9b791b57198b012519ea58bInternational Journal of Soft Computing and Engineering (IJSCE) +
ISSN: 2231-2307, Volume-4 Issue-4, September 2014 +
An Efficient Method for Face Recognition based on +
Fusion of Global and Local Feature Extraction +
392425be1c9d9c2ee6da45de9df7bef0d278e85f
392c3cabe516c0108b478152902a9eee94f4c81eComputer Science and Artificial Intelligence Laboratory +
Technical Report +
MIT-CSAIL-TR-2007-024 +
April 23, 2007 +
Tiny images +
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 +
3947b64dcac5bcc1d3c8e9dcb50558efbb8770f1
3965d61c4f3b72044f43609c808f8760af8781a2
395bf182983e0917f33b9701e385290b64e22f9a
3933e323653ff27e68c3458d245b47e3e37f52fdEvaluation of a 3D-aided Pose Invariant 2D Face Recognition System +
Computational Biomedicine Lab +
4800 Calhoun Rd. Houston, TX, USA +
39b452453bea9ce398613d8dd627984fd3a0d53c
3958db5769c927cfc2a9e4d1ee33ecfba86fe054Describable Visual Attributes for +
Face Verification and Image Search +
39b5f6d6f8d8127b2b97ea1a4987732c0db6f9df
994f7c469219ccce59c89badf93c0661aae342641 +
Model Based Face Recognition Across Facial +
Expressions +
+
screens, embedded into mobiles and installed into everyday +
living and working environments they become valuable tools +
for human system interaction. A particular important aspect of +
this interaction is detection and recognition of faces and +
interpretation of facial expressions. These capabilities are +
deeply rooted in the human visual system and a crucial +
building block for social interaction. Consequently, these +
capabilities are an important step towards the acceptance of +
many technical systems. +
trees as a classifier +
lies not only +
9949ac42f39aeb7534b3478a21a31bc37fe2ffe3Parametric Stereo for Multi-Pose Face Recognition and +
3D-Face Modeling +
PSI ESAT-KUL +
Leuven, Belgium +
9958942a0b7832e0774708a832d8b7d1a5d287aeThe Sparse Matrix Transform for Covariance +
Estimation and Analysis of High Dimensional +
Signals +
9931c6b050e723f5b2a189dd38c81322ac0511de
9993f1a7cfb5b0078f339b9a6bfa341da76a3168JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
A Simple, Fast and Highly-Accurate Algorithm to +
Recover 3D Shape from 2D Landmarks on a Single +
Image +
99c20eb5433ed27e70881d026d1dbe378a12b342ISCA Archive +
http://www.isca-speech.org/archive +
First Workshop on Speech, Language +
and Audio in Multimedia +
Marseille, France +
August 22-23, 2013 +
Proceedings of the First Workshop on Speech, Language and Audio in Multimedia (SLAM), Marseille, France, August 22-23, 2013. +
78 +
9990e0b05f34b586ffccdc89de2f8b0e5d427067International Journal of Modeling and Optimization, Vol. 3, No. 2, April 2013 +
Auto-Optimized Multimodal Expression Recognition +
Framework Using 3D Kinect Data for ASD Therapeutic +
Aid +
 +
regarding +
emotion +
and +
to +
recognize +
99d7678039ad96ee29ab520ff114bb8021222a91Political image analysis with deep neural +
networks +
November 28, 2017 +
529e2ce6fb362bfce02d6d9a9e5de635bde81191This 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. +
> TIP-05732-2009< +
1 +
Normalization of Face Illumination Based +
on Large- and Small- Scale Features +
52887969107956d59e1218abb84a1f834a3145781283 +
Travel Recommendation by Mining People +
Attributes and Travel Group Types From +
Community-Contributed Photos +
521482c2089c62a59996425603d8264832998403
521b625eebea73b5deb171a350e3709a4910eebf
527dda77a3864d88b35e017d542cb612f275a4ec
52f23e1a386c87b0dab8bfdf9694c781cd0a3984
529baf1a79cca813f8c9966ceaa9b3e42748c058Triangle Wise Mapping Technique to Transform one Face Image into Another Face Image +
+
{tag} {/tag} +
+
International Journal of Computer Applications +
+
© 2014 by IJCA Journal +
Volume 87 - Number 6 +
+
Year of Publication: 2014 +
+
+
+
Authors: +
+
Bhogeswar Borah +
+
+
+
+
+
+
+
+
+
+
10.5120/15209-3714 +
{bibtex}pxc3893714.bib{/bibtex} +
5239001571bc64de3e61be0be8985860f08d7e7eSUBMITTED TO IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, JUNE 2016 +
Deep Appearance Models: A Deep Boltzmann +
Machine Approach for Face Modeling +
550858b7f5efaca2ebed8f3969cb89017bdb739f
554b9478fd285f2317214396e0ccd81309963efdSpatio-Temporal Action Localization For Human Action +
Recognition in Large Dataset +
1L2TI, Institut Galil´ee, Universit´e Paris 13, France; +
2SERCOM, Ecole Polytechnique de Tunisie +
55c68c1237166679d2cb65f266f496d1ecd4bec6Learning to Score Figure Skating Sport Videos +
5502dfe47ac26e60e0fb25fc0f810cae6f5173c0Affordance Prediction via Learned Object Attributes +
55a158f4e7c38fe281d06ae45eb456e05516af50The 22nd International Conference on Computer Graphics and Vision +
108 +
GraphiCon’2012 +
5506a1a1e1255353fde05d9188cb2adc20553af5
55c81f15c89dc8f6eedab124ba4ccab18cf38327
551fa37e8d6d03b89d195a5c00c74cc52ff1c67aGeThR-Net: A Generalized Temporally Hybrid +
Recurrent Neural Network for Multimodal +
Information Fusion +
1 Xerox Research Centre India; 2 Amazon Development Center India +
55c40cbcf49a0225e72d911d762c27bb1c2d14aaIndian Face Age Database: A Database for Face Recognition with Age Variation +
{tag} {/tag} +
International Journal of Computer Applications +
+
Foundation of Computer Science (FCS), NY, USA +
+
+
Volume 126 +
- +
Number 5 +
+
+
Year of Publication: 2015 +
+
+
+
+
Authors: +
+
+
+
+
+
+
+
+
+
+
+
10.5120/ijca2015906055 +
{bibtex}2015906055.bib{/bibtex} +
973e3d9bc0879210c9fad145a902afca07370b86(IJACSA) International Journal of Advanced Computer Science and Applications, +
Vol. 7, No. 7, 2016 +
From Emotion Recognition to Website +
Customizations +
O.B. Efremides +
School of Web Media +
Bahrain Polytechnic +
Isa Town, Kingdom of Bahrain +
97b8249914e6b4f8757d22da51e8347995a4063728 +
Large-Scale Vehicle Detection, Indexing, +
and Search in Urban Surveillance Videos +
97032b13f1371c8a813802ade7558e816d25c73fTotal Recall Final Report +
Supervisor: Professor Duncan Gillies +
January 11, 2006 +
97cf04eaf1fc0ac4de0f5ad4a510d57ce12544f5manuscript No. +
(will be inserted by the editor) +
Deep Affect Prediction in-the-wild: Aff-Wild Database and Challenge, +
Deep Architectures, and Beyond +
Zafeiriou4 +
97d1d561362a8b6beb0fdbee28f3862fb48f13801955 +
Age Synthesis and Estimation via Faces: +
A Survey +
97540905e4a9fdf425989a794f024776f28a3fa9
9755554b13103df634f9b1ef50a147dd02eab02fHow Transferable are CNN-based Features for +
Age and Gender Classification? +
1 +
635158d2da146e9de559d2742a2fa234e06b52db
63cf5fc2ee05eb9c6613043f585dba48c5561192Prototype Selection for +
Classification in Standard +
and Generalized +
Dissimilarity Spaces +
63d8d69e90e79806a062cb8654ad78327c8957bb
631483c15641c3652377f66c8380ff684f3e365cSync-DRAW: Automatic Video Generation using Deep Recurrent +
A(cid:130)entive Architectures +
Gaurav Mi(cid:138)al∗ +
IIT Hyderabad +
Vineeth N Balasubramanian +
IIT Hyderabad +
63eefc775bcd8ccad343433fc7a1dd8e1e5ee796
632fa986bed53862d83918c2b71ab953fd70d6ccGÜNEL ET AL.: WHAT FACE AND BODY SHAPES CAN TELL ABOUT HEIGHT +
What Face and Body Shapes Can Tell +
About Height +
CVLab +
EPFL, +
Lausanne, Switzerland +
63340c00896d76f4b728dbef85674d7ea8d5ab261732 +
Discriminant Subspace Analysis: +
A Fukunaga-Koontz Approach +
63d865c66faaba68018defee0daf201db8ca79edDeep Regression for Face Alignment +
1Dept. of Electronics and Information Engineering, Huazhong Univ. of Science and Technology, China +
2Microsoft Research, Beijing, China +
634541661d976c4b82d590ef6d1f3457d2857b19AAllmmaa MMaatteerr SSttuuddiioorruumm –– UUnniivveerrssiittàà ddii BBoollooggnnaa +
in cotutela con Università di Sassari +
DOTTORATO DI RICERCA IN +
INGEGNERIA ELETTRONICA, INFORMATICA E DELLE +
TELECOMUNICAZIONI +
Ciclo XXVI +
Settore Concorsuale di afferenza: 09/H1 +
Settore Scientifico disciplinare: ING-INF/05 +
ADVANCED TECHNIQUES FOR FACE RECOGNITION +
UNDER CHALLENGING ENVIRONMENTS +
TITOLO TESI +
Presentata da: +
Coordinatore Dottorato +
ALESSANDRO VANELLI-CORALLI +
+
Relatore +
DAVIDE MALTONI +
Relatore +
MASSIMO TISTARELLI +
Esame finale anno 2014 +
6332a99e1680db72ae1145d65fa0cccb37256828MASTER IN COMPUTER VISION AND ARTIFICIAL INTELLIGENCE +
REPORT OF THE RESEARCH PROJECT +
OPTION: COMPUTER VISION +
Pose and Face Recovery via +
Spatio-temporal GrabCut Human +
Segmentation +
Date: 13/07/2010 +
63c022198cf9f084fe4a94aa6b240687f21d8b41425 +
0f65c91d0ed218eaa7137a0f6ad2f2d731cf8dabMulti-Directional Multi-Level Dual-Cross +
Patterns for Robust Face Recognition +
0f112e49240f67a2bd5aaf46f74a924129f03912947 +
Age-Invariant Face Recognition +
0f4cfcaca8d61b1f895aa8c508d34ad89456948eLOCAL APPEARANCE BASED FACE RECOGNITION USING +
DISCRETE COSINE TRANSFORM (WedPmPO4) +
Author(s) : +
0fad544edfc2cd2a127436a2126bab7ad31ec333Decorrelating Semantic Visual Attributes by Resisting the Urge to Share +
UT Austin +
USC +
UT Austin +
0f32df6ae76402b98b0823339bd115d33d3ec0a0Emotion recognition from embedded bodily +
expressions and speech during dyadic interactions +
0fd1715da386d454b3d6571cf6d06477479f54fcJ Intell Robot Syst (2016) 82:101–133 +
DOI 10.1007/s10846-015-0259-2 +
A Survey of Autonomous Human Affect Detection Methods +
for Social Robots Engaged in Natural HRI +
Received: 10 December 2014 / Accepted: 11 August 2015 / Published online: 23 August 2015 +
© Springer Science+Business Media Dordrecht 2015 +
0f9bf5d8f9087fcba419379600b86ae9e9940013
0f92e9121e9c0addc35eedbbd25d0a1faf3ab529MORPH-II: A Proposed Subsetting Scheme +
NSF-REU Site at UNC Wilmington, Summer 2017 +
0fd1bffb171699a968c700f206665b2f8837d953Weakly Supervised Object Localization with +
Multi-fold Multiple Instance Learning +
0a511058edae582e8327e8b9d469588c25152dc6
0a4f3a423a37588fde9a2db71f114b293fc09c50
0a3863a0915256082aee613ba6dab6ede962cdcdEarly and Reliable Event Detection Using Proximity Space Representation +
LTCI, CNRS, T´el´ecom ParisTech, Universit´e Paris-Saclay, 75013, Paris, France +
J´erˆome Gauthier +
LADIS, CEA, LIST, 91191, Gif-sur-Yvette, France +
Normandie Universit´e, UR, LITIS EA 4108, Avenue de l’universit´e, 76801, Saint-Etienne-du-Rouvray, France +
0ad90118b4c91637ee165f53d557da7141c3fde0
0af48a45e723f99b712a8ce97d7826002fe4d5a52982 +
Toward Wide-Angle Microvision Sensors +
Todd Zickler, Member, IEEE +
0aa8a0203e5f406feb1815f9b3dd49907f5fd05bMixture subclass discriminant analysis +
0a1138276c52c734b67b30de0bf3f76b0351f097This 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. +
The final version of record is available at +
http://dx.doi.org/10.1109/TIP.2016.2539502 +
Discriminant Incoherent Component Analysis +
0a6a25ee84fc0bf7284f41eaa6fefaa58b5b329a
0ae9cc6a06cfd03d95eee4eca9ed77b818b59cb7Noname manuscript No. +
(will be inserted by the editor) +
Multi-task, multi-label and multi-domain learning with +
residual convolutional networks for emotion recognition +
Received: date / Accepted: date +
0acf23485ded5cb9cd249d1e4972119239227ddbDual coordinate solvers for large-scale structural SVMs +
UC Irvine +
This manuscript describes a method for training linear SVMs (including binary SVMs, SVM regression, +
and structural SVMs) from large, out-of-core training datasets. Current strategies for large-scale learning fall +
into one of two camps; batch algorithms which solve the learning problem given a finite datasets, and online +
algorithms which can process out-of-core datasets. The former typically requires datasets small enough to fit +
in memory. The latter is often phrased as a stochastic optimization problem [4, 15]; such algorithms enjoy +
strong theoretical properties but often require manual tuned annealing schedules, and may converge slowly +
for problems with large output spaces (e.g., structural SVMs). We discuss an algorithm for an “intermediate” +
regime in which the data is too large to fit in memory, but the active constraints (support vectors) are small +
enough to remain in memory. +
In this case, one can design rather efficient learning algorithms that are +
as stable as batch algorithms, but capable of processing out-of-core datasets. We have developed such a +
MATLAB-based solver and used it to train a series of recognition systems [19, 7, 21, 12] for articulated pose +
estimation, facial analysis, 3D object recognition, and action classification, all with publicly-available code. +
This writeup describes the solver in detail. +
Approach: Our approach is closely based on data-subsampling algorithms for collecting hard exam- +
ples [9, 10, 6], combined with the dual coordinate quadratic programming (QP) solver described in liblinear +
[8]. The latter appears to be current fastest method for learning linear SVMs. We make two extensions (1) +
We show how to generalize the solver to other types of SVM problems such as (latent) structural SVMs (2) +
We show how to modify it to behave as a partially-online algorithm, which only requires access to small +
amounts of data at a time. +
Overview: Sec. 1 describes a general formulation of an SVM problem that encompasses many standard +
tasks such as multi-class classification and (latent) structural prediction. Sec. 2 derives its dual QP, and Sec. 3 +
describes a dual coordinate descent optimization algorithm. Sec. 4 describes modifications for optimizing +
in an online fashion, allowing one to learn near-optimal models with a single pass over large, out-of-core +
datasets. Sec. 5 briefly touches on some theoretical issues that are necessary to ensure convergence. Finally, +
Sec. 6 and Sec. 7 describe modifications to our basic formulation to accommodate non-negativity constraints +
and flexible regularization schemes during learning. +
1 Generalized SVMs +
We first describe a general formulation of a SVM which encompasses various common problems such as +
binary classification, regression, and structured prediction. Assume we are given training data where the ith +
example is described by a set of Ni vectors {xij} and a set of Ni scalars {lij}, where j varies from 1 to Ni. +
We wish to solve the following optimization problem: +
(0, lij − wT xij) +
max +
j∈Ni +
(1) +
(cid:88) +
argmin +
L(w) = +
||w||2 + +
0ad4a814b30e096ad0e027e458981f812c835aa0
6448d23f317babb8d5a327f92e199aaa45f0efdc
6412d8bbcc01f595a2982d6141e4b93e7e982d0fDeep Convolutional Neural Network using Triplets of Faces, Deep Ensemble, and +
Score-level Fusion for Face Recognition +
1Department of Creative IT Engineering, POSTECH, Korea +
2Department of Computer Science and Engineering, POSTECH, Korea +
649eb674fc963ce25e4e8ce53ac7ee20500fb0e3
642c66df8d0085d97dc5179f735eed82abf110d0
641f34deb3bdd123c6b6e7b917519c3e56010cb7
645de797f936cb19c1b8dba3b862543645510544Deep Temporal Linear Encoding Networks +
1ESAT-PSI, KU Leuven, 2CVL, ETH Z¨urich +
6462ef39ca88f538405616239471a8ea17d76259
90ac0f32c0c29aa4545ed3d5070af17f195d015f
90cb074a19c5e7d92a1c0d328a1ade1295f4f311MIT. Media Laboratory Affective Computing Technical Report #571 +
Appears in IEEE International Workshop on Analysis and Modeling of Faces and Gestures , Oct 2003 +
Fully Automatic Upper Facial Action Recognition +
MIT Media Laboratory +
Cambridge, MA 02139 +
90b11e095c807a23f517d94523a4da6ae6b12c76
9028fbbd1727215010a5e09bc5758492211dec19Solving the Uncalibrated Photometric Stereo +
Problem using Total Variation +
1 IRIT, UMR CNRS 5505, Toulouse, France +
2 Dept. of Computer Science, Univ. of Copenhagen, Denmark +
bf1e0279a13903e1d43f8562aaf41444afca4fdc International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056 +
Volume: 04 Issue: 10 | Oct -2017 www.irjet.net p-ISSN: 2395-0072 +
Different Viewpoints of Recognizing Fleeting Facial Expressions with +
DWT +
information +
to get desired +
information +
Introduction +
---------------------------------------------------------------------***--------------------------------------------------------------------- +
bf5940d57f97ed20c50278a81e901ae4656f0f2cQuery-free Clothing Retrieval via Implicit +
Relevance Feedback +
bfb98423941e51e3cd067cb085ebfa3087f3bfbeSparseness helps: Sparsity Augmented +
Collaborative Representation for Classification +
d3b73e06d19da6b457924269bb208878160059daProceedings of the 5th International Conference on Computing and Informatics, ICOCI 2015 +
11-13 August, 2015 Istanbul, Turkey. Universiti Utara Malaysia (http://www.uum.edu.my ) +
Paper No. +
065 +
IMPLEMENTATION OF AN AUTOMATED SMART HOME +
CONTROL FOR DETECTING HUMAN EMOTIONS VIA FACIAL +
DETECTION +
Osman4 +
d3d71a110f26872c69cf25df70043f7615edcf922736 +
Learning Compact Feature Descriptor and Adaptive +
Matching Framework for Face Recognition +
improvements +
d309e414f0d6e56e7ba45736d28ee58ae2bad478Efficient Two-Stream Motion and Appearance 3D CNNs for +
Video Classification +
Ali Diba +
ESAT-KU Leuven +
Ali Pazandeh +
Sharif UTech +
Luc Van Gool +
ESAT-KU Leuven, ETH Zurich +
d3f5a1848b0028d8ab51d0b0673732cad2e3c8c9
d33fcdaf2c0bd0100ec94b2c437dccdacec66476Neurons with Paraboloid Decision Boundaries for +
Improved Neural Network Classification +
Performance +
d444368421f456baf8c3cb089244e017f8d32c41CNN for IMU Assisted Odometry Estimation using Velodyne LiDAR +
d4c7d1a7a03adb2338704d2be7467495f2eb6c7b
d4ebf0a4f48275ecd8dbc2840b2a31cc07bd676d
d44a93027208816b9e871101693b05adab576d89
d4b88be6ce77164f5eea1ed2b16b985c0670463aTECHNICAL REPORT JAN.15.2016 +
A Survey of Different 3D Face Reconstruction +
Methods +
Department of Computer Science and Engineering +
d44ca9e7690b88e813021e67b855d871cdb5022fQUT Digital Repository: +
http://eprints.qut.edu.au/ +
Zhang, Ligang and Tjondronegoro, Dian W. (2009) Selecting, optimizing and +
fusing ‘salient’ Gabor features for facial expression recognition. In: Neural +
Information Processing (Lecture Notes in Computer Science), 1-5 December +
2009, Hotel Windsor Suites Bangkok, Bangkok. +
+
© Copyright 2009 Springer-Verlag GmbH Berlin Heidelberg +
+
bafb8812817db7445fe0e1362410a372578ec1fc805 +
Image-Quality-Based Adaptive Face Recognition +
ba816806adad2030e1939450226c8647105e101cMindLAB at the THUMOS Challenge +
Fabi´an P´aez +
Fabio A. Gonz´alez +
MindLAB Research Group +
MindLAB Research Group +
MindLAB Research Group +
Bogot´a, Colombia +
Bogot´a, Colombia +
Bogot´a, Colombia +
badcd992266c6813063c153c41b87babc0ba36a3Recent Advances in Object Detection in the Age +
of Deep Convolutional Neural Networks +
,1,2), Fr´ed´eric Jurie(1) +
(∗) equal contribution +
(1)Normandie Univ, UNICAEN, ENSICAEN, CNRS +
(2)Safran Electronics and Defense +
September 11, 2018 +
ba788365d70fa6c907b71a01d846532ba3110e31
ba8a99d35aee2c4e5e8a40abfdd37813bfdd0906ELEKTROTEHNI ˇSKI VESTNIK 78(1-2): 12–17, 2011 +
EXISTING SEPARATE ENGLISH EDITION +
Uporaba emotivno pogojenega raˇcunalniˇstva v +
priporoˇcilnih sistemih +
Marko Tkalˇciˇc, Andrej Koˇsir, Jurij Tasiˇc +
1Univerza v Ljubljani, Fakulteta za elektrotehniko, Trˇzaˇska 25, 1000 Ljubljana, Slovenija +
2Univerza v Ljubljani, Fakulteta za raˇcunalniˇstvo in informatiko, Trˇzaˇska 25, 1000 Ljubljana, Slovenija +
Povzetek. V ˇclanku predstavljamo rezultate treh raziskav, vezanih na izboljˇsanje delovanja multimedijskih +
priporoˇcilnih sistemov s pomoˇcjo metod emotivno pogojenega raˇcunalniˇstva (ang. affective computing). +
Vsebinski priporoˇcilni sistem smo izboljˇsali s pomoˇcjo metapodatkov, ki opisujejo emotivne odzive uporabnikov. +
Pri skupinskem priporoˇcilnem sistemu smo dosegli znaˇcilno izboljˇsanje v obmoˇcju hladnega zagona z uvedbo +
nove mere podobnosti, ki temelji na osebnostnem modelu velikih pet (ang. five factor model). Razvili smo tudi +
sistem za neinvazivno oznaˇcevanje vsebin z emotivnimi parametri, ki pa ˇse ni zrel za uporabo v priporoˇcilnih +
sistemih. +
Kljuˇcne besede: priporoˇcilni sistemi, emotivno pogojeno raˇcunalniˇstvo, strojno uˇcenje, uporabniˇski profil, +
emocije +
Uporaba emotivnega raˇcunalniˇstva v priporoˇcilnih +
sistemih +
In this paper we present the results of three investigations of +
our broad research on the usage of affect and personality in +
recommender systems. We improved the accuracy of content- +
based recommender system with the inclusion of affective +
parameters of user and item modeling. We improved the +
accuracy of a content filtering recommender system under the +
cold start conditions with the introduction of a personality +
based user similarity measure. Furthermore we developed a +
system for implicit tagging of content with affective metadata. +
1 UVOD +
Uporabniki (porabniki) multimedijskih (MM) vsebin so +
v ˇcedalje teˇzjem poloˇzaju, saj v veliki koliˇcini vse- +
bin teˇzko najdejo zanje primerne. Pomagajo si s pri- +
poroˇcilnimi sistemi, ki na podlagi osebnih preferenc +
uporabnikov izberejo manjˇso koliˇcino relevantnih MM +
vsebin, med katerimi uporabnik laˇze izbira. Noben danes +
znan priporoˇcilni sistem ne zadoˇsˇca v celoti potrebam +
uporabnikov, saj je izbor priporoˇcenih vsebin obiˇcajno +
nezadovoljive kakovosti [10]. Cilj tega ˇclanka je pred- +
staviti metode emotivno pogojenega raˇcunalniˇstva (ang. +
affective computing - glej [12]) za izboljˇsanje kakovosti +
priporoˇcilnih sistemov in utrditi za slovenski prostor +
novo terminologijo. +
1.1 Opis problema +
Za izboljˇsanje kakovosti priporoˇcilnih sistemov sta +
na voljo dve poti: (i) optimizacija algoritmov ali (ii) +
uporaba boljˇsih znaˇcilk, ki bolje razloˇzijo neznano +
Prejet 13. oktober, 2010 +
Odobren 1. februar, 2011 +
varianco [8]. V tem ˇclanku predstavljamo izboljˇsanje +
priporoˇcilnih sistemov z uporabo novih znaˇcilk, ki te- +
meljijo na emotivnih odzivih uporabnikov in na njiho- +
vih osebnostnih lastnostih. Te znaˇcilke razloˇzijo velik +
del uporabnikovih preferenc, ki se izraˇzajo v obliki +
ocen posameznih vsebin (npr. Likertova lestvica, binarne +
ocene itd.). Ocene vsebin se pri priporoˇcilnih sistemih +
zajemajo eksplicitno (ocena) ali implicitno, pri ˇcemer o +
oceni sklepamo na podlagi opazovanj (npr. ˇcas gledanja +
kot indikator vˇseˇcnosti [7]. +
Izboljˇsanja uˇcinkovitosti priporoˇcilnih sistemov smo +
se lotili na treh podroˇcjih: (i) uporaba emotivnega +
modeliranja uporabnikov v vsebinskem priporoˇcilnem +
sistemu, (ii) neinvazivna (implicitna) detekcija emocij za +
emotivno modeliranje in (iii) uporaba osebnostne mere +
podobnosti v skupinskem priporoˇcilnem sistemu. Slika 1 +
prikazuje arhitekturo emotivnega priporoˇcilnega sistema +
in mesta, kjer smo vnesli opisane izboljˇsave. +
Preostanek ˇclanka je strukturiran tako: v razdelku +
2 je predstavljen zajem podatkov. V razdelku 3 je +
predstavljen vsebinski priporoˇcilni sistem z emotivnimi +
metapodatki. V razdelku 4 je predstavljen skupinski +
priporoˇcilni sistem, ki uporablja mero podobnosti na +
podlagi osebnosti, v razdelku 5 pa algoritem za razpo- +
znavo emocij. Vsak od teh razdelov je sestavljen iz opisa +
eksperimenta in predstavitve rezultatov. V razdelku 6 so +
predstavljeni sklepi. +
1.2 Sorodno delo +
Najbolj groba delitev priporoˇcilnih sistemov je na vse- +
binske, skupinske ter hibridne sisteme [1]. Z izjemo vse- +
binskih priporoˇcilnih sistemov, ki sta ga razvila Arapakis +
[2] in Tkalˇciˇc [14], sorodnega dela na podroˇcju emotivno +
pogojenih priporoˇcilnih sistemov takorekoˇc ni. Panti´c in +
ba29ba8ec180690fca702ad5d516c3e43a7f0bb8
bab88235a30e179a6804f506004468aa8c28ce4f
badd371a49d2c4126df95120902a34f4bee01b00GONDA, WEI, PARAG, PFISTER: PARALLEL SEPARABLE 3D CONVOLUTION +
Parallel Separable 3D Convolution for Video +
and Volumetric Data Understanding +
Harvard John A. Paulson School of +
Engineering and Applied Sciences +
Camabridge MA, USA +
Toufiq Parag +
Hanspeter Pfister +
a0f94e9400938cbd05c4b60b06d9ed58c34583031118 +
Value-Directed Human Behavior Analysis +
from Video Using Partially Observable +
Markov Decision Processes +
a022eff5470c3446aca683eae9c18319fd2406d52017-ENST-0071 +
EDITE - ED 130 +
Doctorat ParisTech +
T H È S E +
pour obtenir le grade de docteur délivré par +
TÉLÉCOM ParisTech +
Spécialité « SIGNAL et IMAGES » +
présentée et soutenue publiquement par +
le 15 décembre 2017 +
Apprentissage Profond pour la Description Sémantique des Traits +
Visuels Humains +
Directeur de thèse : Jean-Luc DUGELAY +
Co-encadrement de la thèse : Moez BACCOUCHE +
Jury +
Mme Bernadette DORIZZI, PRU, Télécom SudParis +
Mme Jenny BENOIS-PINEAU, PRU, Université de Bordeaux +
M. Christian WOLF, MC/HDR, INSA de Lyon +
M. Patrick PEREZ, Chercheur/HDR, Technicolor Rennes +
M. Moez BACCOUCHE, Chercheur/Docteur, Orange Labs Rennes +
M. Jean-Luc DUGELAY, PRU, Eurecom Sophia Antipolis +
M. Sid-Ahmed BERRANI, Directeur de l’Innovation/HDR, Algérie Télécom +
Présidente +
Rapporteur +
Rapporteur +
Examinateur +
Encadrant +
Directeur de Thèse +
Invité +
TÉLÉCOM ParisTech +
école de l’Institut Télécom - membre de ParisTech +
N°: 2009 ENAM XXXX T H È S E
a0c37f07710184597befaa7e6cf2f0893ff440e9
a0fb5b079dd1ee5ac6ac575fe29f4418fdb0e670
a0fd85b3400c7b3e11122f44dc5870ae2de9009aLearning Deep Representation for Face +
Alignment with Auxiliary Attributes +
a0dfb8aae58bd757b801e2dcb717a094013bc178Reconocimiento de expresiones faciales con base +
en la din´amica de puntos de referencia faciales +
Instituto Nacional de Astrof´ısica ´Optica y Electr´onica, +
Divisi´on de Ciencias Computacionales, Tonantzintla, Puebla, +
M´exico +
Resumen. Las expresiones faciales permiten a las personas comunicar +
emociones, y es pr´acticamente lo primero que observamos al interactuar +
con alguien. En el ´area de computaci´on, el reconocimiento de expresiones +
faciales es importante debido a que su an´alisis tiene aplicaci´on directa en +
´areas como psicolog´ıa, medicina, educaci´on, entre otras. En este articulo +
se presenta el proceso de dise˜no de un sistema para el reconocimiento de +
expresiones faciales utilizando la din´amica de puntos de referencia ubi- +
cados en el rostro, su implementaci´on, experimentos realizados y algunos +
de los resultados obtenidos hasta el momento. +
Palabras clave: Expresiones faciales, clasificaci´on, m´aquinas de soporte +
vectorial,modelos activos de apariencia. +
Facial Expressions Recognition Based on Facial +
Landmarks Dynamics +
a03cfd5c0059825c87d51f5dbf12f8a76fe9ff60Simultaneous Learning and Alignment: +
Multi-Instance and Multi-Pose Learning? +
1 Comp. Science & Eng. +
Univ. of CA, San Diego +
2 Electrical Engineering +
California Inst. of Tech. +
3 Lab of Neuro Imaging +
Univ. of CA, Los Angeles +
a000149e83b09d17e18ed9184155be140ae1266eChapter 9 +
Action Recognition in Realistic +
Sports Videos +
a784a0d1cea26f18626682ab108ce2c9221d1e53Anchored Regression Networks applied to Age Estimation and Super Resolution +
D-ITET, ETH Zurich +
Switzerland +
D-ITET, ETH Zurich +
Merantix GmbH +
D-ITET, ETH Zurich +
ESAT, KU Leuven +
a74251efa970b92925b89eeef50a5e37d9281ad0
a7664247a37a89c74d0e1a1606a99119cffc41d4Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
3287 +
a7a6eb53bee5e2224f2ecd56a14e3a5a717e55b911th International Symposium of Robotics Research (ISRR2003), pp.192-201, 2003 +
Face Recognition Using Multi-viewpoint Patterns for +
Robot Vision +
Corporate Research and Development Center, TOSHIBA Corporation +
1, KomukaiToshiba-cho, Saiwai-ku, Kawasaki 212-8582 Japan +
a75ee7f4c4130ef36d21582d5758f953dba03a01DD2427 Final Project Report +
DD2427 Final Project Report +
Human face attributes prediction with Deep +
Learning +
a775da3e6e6ea64bffab7f9baf665528644c7ed3International Journal of Computer Applications (0975 – 8887) +
Volume 142 – No.9, May 2016 +
Human Face Pose Estimation based on Feature +
Extraction Points +
Research scholar, +
Department of ECE +
SBSSTC, Moga Road, +
Ferozepur, Punjab, India +
b8dba0504d6b4b557d51a6cf4de5507141db60cfComparing Performances of Big Data Stream +
Processing Platforms with RAM3S +
b8378ab83bc165bc0e3692f2ce593dcc713df34a
b8f3f6d8f188f65ca8ea2725b248397c7d1e662dSelfie Detection by Synergy-Constriant Based +
Convolutional Neural Network +
Electrical and Electronics Engineering, NITK-Surathkal, India. +
b81cae2927598253da37954fb36a2549c5405cdbExperiments on Visual Information Extraction with the Faces of Wikipedia +
D´epartement de g´enie informatique et g´enie logiciel, Polytechnique Montr´eal +
2500, Chemin de Polytechnique, Universit´e de Montr´eal, Montr`eal, Qu´ebec, Canada +
b8a829b30381106b806066d40dd372045d49178d1872 +
A Probabilistic Framework for Joint Pedestrian Head +
and Body Orientation Estimation +
b1d89015f9b16515735d4140c84b0bacbbef19acToo Far to See? Not Really! +
— Pedestrian Detection with Scale-aware +
Localization Policy +
b14b672e09b5b2d984295dfafb05604492bfaec5LearningImageClassificationandRetrievalModelsThomasMensink
b171f9e4245b52ff96790cf4f8d23e822c260780
b1a3b19700b8738b4510eecf78a35ff38406df22This 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 +
Transactions on Affective Computing +
JOURNAL OF LATEX CLASS FILES, VOL. 13, NO. 9, SEPTEMBER 2014 +
Automatic Analysis of Facial Actions: A Survey +
and Maja Pantic, Fellow, IEEE +
b1301c722886b6028d11e4c2084ee96466218be4
b1c5581f631dba78927aae4f86a839f43646220c
b1444b3bf15eec84f6d9a2ade7989bb980ea7bd1LOCAL DIRECTIONAL RELATION PATTERN +
Local Directional Relation Pattern for +
Unconstrained and Robust Face Retrieval +
b19e83eda4a602abc5a8ef57467c5f47f493848dJOURNAL OF LATEX CLASS FILES +
Heat Kernel Based Local Binary Pattern for +
Face Representation +
dd8084b2878ca95d8f14bae73e1072922f0cc5daModel Distillation with Knowledge Transfer from +
Face Classification to Alignment and Verification +
Beijing Orion Star Technology Co., Ltd. Beijing, China +
dd0760bda44d4e222c0a54d41681f97b3270122b
ddea3c352f5041fb34433b635399711a90fde0e8Facial Expression Classification using Visual Cues and Language +
Department of Computer Science and Engineering, IIT Kanpur +
ddbd24a73ba3d74028596f393bb07a6b87a469c0Multi-region two-stream R-CNN +
for action detection +
Inria(cid:63) +
ddf099f0e0631da4a6396a17829160301796151cIEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY +
Learning Face Image Quality from +
Human Assessments +
dd0a334b767e0065c730873a95312a89ef7d1c03Eigenexpressions: Emotion Recognition using Multiple +
Eigenspaces +
Luis Marco-Gim´enez1, Miguel Arevalillo-Herr´aez1, and Cristina Cuhna-P´erez2 +

Burjassot. Valencia 46100, Spain, +
2 Universidad Cat´olica San Vicente M´artir de Valencia (UCV), +
Burjassot. Valencia. Spain +
dd2f6a1ba3650075245a422319d86002e1e87808
dd8d53e67668067fd290eb500d7dfab5b6f730dd69 +
A Parameter-Free Framework for General +
Supervised Subspace Learning +
ddbb6e0913ac127004be73e2d4097513a8f02d37264 +
IEEE TRANSACTIONS ON MULTIMEDIA, VOL. 1, NO. 3, SEPTEMBER 1999 +
Face Detection Using Quantized Skin Color +
Regions Merging and Wavelet Packet Analysis +
dd600e7d6e4443ebe87ab864d62e2f4316431293
dcb44fc19c1949b1eda9abe998935d567498467dProceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
1916 +
dc77287bb1fcf64358767dc5b5a8a79ed9abaa53Fashion Conversation Data on Instagram +
∗Graduate School of Culture Technology, KAIST, South Korea +
†Department of Communication Studies, UCLA, USA +
dc2e805d0038f9d1b3d1bc79192f1d90f6091ecb
dc974c31201b6da32f48ef81ae5a9042512705feAm I done? Predicting Action Progress in Video +
1 Media Integration and Communication Center, Univ. of Florence, Italy +
2 Department of Mathematics “Tullio Levi-Civita”, Univ. of Padova, Italy +
b6c047ab10dd86b1443b088029ffe05d79bbe257
b6c53891dff24caa1f2e690552a1a5921554f994
b613b30a7cbe76700855479a8d25164fa7b6b9f11 +
Identifying User-Specific Facial Affects from +
Spontaneous Expressions with Minimal Annotation +
b6f682648418422e992e3ef78a6965773550d36bFebruary 8, 2017 +
b656abc4d1e9c8dc699906b70d6fcd609fae8182
a9eb6e436cfcbded5a9f4b82f6b914c7f390adbd(IJARAI) International Journal of Advanced Research in Artificial Intelligence, +
Vol. 5, No.6, 2016 +
A Model for Facial Emotion Inference Based on +
Planar Dynamic Emotional Surfaces +
Ruivo, J. P. P. +
Escola Polit´ecnica +
Negreiros, T. +
Escola Polit´ecnica +
Barretto, M. R. P. +
Escola Polit´ecnica +
Tinen, B. +
Escola Polit´ecnica +
Universidade de S˜ao Paulo +
Universidade de S˜ao Paulo +
Universidade de S˜ao Paulo +
Universidade de S˜ao Paulo +
S˜ao Paulo, Brazil +
S˜ao Paulo, Brazil +
S˜ao Paulo, Brazil +
S˜ao Paulo, Brazil +
a92adfdd8996ab2bd7cdc910ea1d3db03c66d34f
a98316980b126f90514f33214dde51813693fe0dCollaborations on YouTube: From Unsupervised Detection to the +
Impact on Video and Channel Popularity +
Multimedia Communications Lab (KOM), Technische Universität Darmstadt, Germany +
a93781e6db8c03668f277676d901905ef44ae49fRecent Datasets on Object Manipulation: A Survey +
a9adb6dcccab2d45828e11a6f152530ba8066de6Aydınlanma Alt-uzaylarına dayalı Gürbüz Yüz Tanıma +
Illumination Subspaces based Robust Face Recognition +
Interactive Systems Labs, Universität Karlsruhe (TH) +
76131 Karlsruhe, Almanya +
web: http://isl.ira.uka.de/face_recognition +
Özetçe +
yönlerine +
aydınlanma +
kaynaklanan +
sonra, yüz uzayı +
Bu çalışmada aydınlanma alt-uzaylarına dayalı bir yüz tanıma +
sistemi sunulmuştur. Bu sistemde, +
ilk olarak, baskın +
aydınlanma yönleri, bir topaklandırma algoritması kullanılarak +
öğrenilmiştir. Topaklandırma algoritması sonucu önden, sağ +
ve sol yanlardan olmak üzere üç baskın aydınlanma yönü +
gözlemlenmiştir. Baskın +
karar +
-yüzün görünümündeki +
kılındıktan +
aydınlanmadan +
kişi +
kimliklerinden kaynaklanan değişimlerden ayırmak için- bu üç +
aydınlanma uzayına bölünmüştür. Daha sonra, ek aydınlanma +
yönü bilgisinden faydalanmak için aydınlanma alt-uzaylarına +
dayalı yüz +
tanıma algoritması kullanılmıştır. Önerilen +
yaklaşım, CMU PIE veritabanında, “illumination” ve +
“lighting” kümelerinde yer alan yüz +
imgeleri üzerinde +
sınanmıştır. Elde edilen deneysel sonuçlar, aydınlanma +
yönünden yararlanmanın ve aydınlanma alt-uzaylarına dayalı +
yüz tanıma algoritmasının yüz tanıma başarımını önemli +
ölçüde arttırdığını göstermiştir. +
değişimleri, +
farklı +
a95dc0c4a9d882a903ce8c70e80399f38d2dcc89 TR-IIS-14-003 +
Review and Implementation of +
High-Dimensional Local Binary +
Patterns and Its Application to +
Face Recognition +
July. 24, 2014 || Technical Report No. TR-IIS-14-003 +
http://www.iis.sinica.edu.tw/page/library/TechReport/tr2014/tr14.html +
a9286519e12675302b1d7d2fe0ca3cc4dc7d17f6Learning to Succeed while Teaching to Fail: +
Privacy in Closed Machine Learning Systems +
a92b5234b8b73e06709dd48ec5f0ec357c1aabed
d50c6d22449cc9170ab868b42f8c72f8d31f9b6cProceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
1668 +
d522c162bd03e935b1417f2e564d1357e98826d2He et al. EURASIP Journal on Advances in Signal Processing 2013, 2013:19 +
http://asp.eurasipjournals.com/content/2013/1/19 +
RESEARCH +
Open Access +
Weakly supervised object extraction with +
iterative contour prior for remote sensing +
images +
d59f18fcb07648381aa5232842eabba1db52383eInternational Conference on Systemics, Cybernetics and Informatics, February 12–15, 2004 +
ROBUST FACIAL EXPRESSION RECOGNITION USING SPATIALLY +
LOCALIZED GEOMETRIC MODEL +
Department of Electrical Engineering +
Dept. of Computer Sc. and Engg. +
IIT Kanpur +
Kanpur 208016, India +
Kanpur 208016, India +
IIT Kanpur +
Dept. of Computer Sc. and Engg. +
IIT Kanpur +
Kanpur 208016, India +
While approaches based on 3D deformable facial model have +
achieved expression recognition rates of as high as 98% [2], they +
are computationally inefficient and require considerable apriori +
training based on 3D information, which is often unavailable. +
Recognition from 2D images remains a difficult yet important +
problem for areas such as +
image database querying and +
classification. The accuracy rates achieved for 2D images are +
around 90% [3,4,5,11]. In a recent review of expression +
recognition, Fasel [1] considers the problem along several +
dimensions: whether features such as lips or eyebrows are first +
identified in the face (local [4] vs holistic [11]), or whether the +
image model used is 2D or 3D. Methods proposed for expression +
recognition from 2D images include the Gabor-Wavelet [5] or +
Holistic Optical flow [11] approach. +
This paper describes a more robust system for facial expression +
recognition from image sequences using 2D appearance-based +
local approach for the extraction of intransient facial features, i.e. +
features such as eyebrows, lips, or mouth, which are always +
present in the image, but may be deformed [1] (in contrast, +
transient features are wrinkles or bulges that disappear at other +
times). The main advantages of such an approach is low +
computational requirements, ability to work with both colored and +
grayscale images and robustness in handling partial occlusions +
[3]. +
Edge projection analysis which is used here for feature extraction +
(eyebrows and lips) is well known [6]. Unlike [6] which describes +
a template based matching as an essential starting point, we use +
contours analysis. Our system computes a feature vector based on +
geometrical model of the face and then classifies it into four +
expression classes using a feed-forward basis function net. The +
system detects open and closed state of the mouth as well. The +
algorithm presented here works on both color and grayscale image +
sequences. An important aspect of our work is the use of color +
information for robust and more accurate segmentation of lip +
region in case of color images. The novel lip-enhancement +
transform is based on Gaussian modeling of skin and lip color. +
To place the work in a larger context of face analysis and +
recognition, the overall task requires that the part of the image +
involving the face be detected and segmented. We assume that a +
near-frontal view of the face is available. Tests on a grayscale +
and two color face image databases ([8] and [9,10]) demonstrate a +
superior recognition rate for four facial expressions (smile, +
surprise, disgust and sad against neutral). +
image sequences +
d588dd4f305cdea37add2e9bb3d769df98efe880 +
Audio-Visual Authentication System over the +
Internet Protocol +
abandoned. +
in +
illumination based +
is developed with the objective to +
d5444f9475253bbcfef85c351ea9dab56793b9eaIEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS +
BoxCars: Improving Fine-Grained Recognition +
of Vehicles using 3D Bounding Boxes +
in Traffic Surveillance +
in contrast +
d5ab6aa15dad26a6ace5ab83ce62b7467a18a88eWorld Journal of Computer Application and Technology 2(7): 133-138, 2014 +
DOI: 10.13189/wjcat.2014.020701 +
http://www.hrpub.org +
Optimized Structure for Facial Action Unit Relationship +
Using Bayesian Network +
Intelligent Biometric Group, School of Electrical and Electronic Engineering, Engineering Campus, Universiti Sains Malaysia, Pulau +
Pinang, Malaysia +
Copyright © 2014 Horizon Research Publishing All rights reserved. +
d56fe69cbfd08525f20679ffc50707b738b88031Training of multiple classifier systems utilizing +
partially labelled sequences +

89069 Ulm - Germany +
d50751da2997e7ebc89244c88a4d0d18405e8507
d511e903a882658c9f6f930d6dd183007f508eda
d59404354f84ad98fa809fd1295608bf3d658bdcInternational Journal of Computer Vision manuscript No. +
(will be inserted by the editor) +
Face Synthesis from Visual Attributes via Sketch using +
Conditional VAEs and GANs +
Received: date / Accepted: date +
d5e1173dcb2a51b483f86694889b015d55094634
d2eb1079552fb736e3ba5e494543e67620832c52ANNUNZIATA, SAGONAS, CALÌ: DENSELY FUSED SPATIAL TRANSFORMER NETWORKS1 +
DeSTNet: Densely Fused Spatial +
Transformer Networks1 +
Onfido Research +
3 Finsbury Avenue +
London, UK +
d24dafe10ec43ac8fb98715b0e0bd8e479985260J Nonverbal Behav (2018) 42:81–99 +
https://doi.org/10.1007/s10919-017-0266-z +
O R I G I N A L P A P E R +
Effects of Social Anxiety on Emotional Mimicry +
and Contagion: Feeling Negative, but Smiling Politely +
• Gerben A. van Kleef2 +
• Agneta H. Fischer2 +
Published online: 25 September 2017 +
Ó The Author(s) 2017. This article is an open access publication +
d278e020be85a1ccd90aa366b70c43884dd3f798Learning From Less Data: Diversified Subset Selection and +
Active Learning in Image Classification Tasks +
IIT Bombay +
Mumbai, Maharashtra, India +
AITOE Labs +
Mumbai, Maharashtra, India +
AITOE Labs +
Mumbai, Maharashtra, India +
Rishabh Iyer +
AITOE Labs +
Seattle, Washington, USA +
AITOE Labs +
Seattle, Washington, USA +
Narsimha Raju +
IIT Bombay +
Mumbai, Maharashtra, India +
IIT Bombay +
Mumbai, Maharashtra, India +
IIT Bombay +
Mumbai, Maharashtra, India +
May 30, 2018 +
aafb271684a52a0b23debb3a5793eb618940c5dd
aa52910c8f95e91e9fc96a1aefd406ffa66d797dFACE RECOGNITION SYSTEM BASED +
ON 2DFLD AND PCA +
E&TC Department +
Sinhgad Academy of Engineering +
Pune, India +
Mr. Hulle Rohit Rajiv +
ME E&TC [Digital System] +
Sinhgad Academy of Engineering +
Pune, India +
aadfcaf601630bdc2af11c00eb34220da59b7559Multi-view Hybrid Embedding: +
A Divide-and-Conquer Approach +
aaa4c625f5f9b65c7f3df5c7bfe8a6595d0195a5Biometrics in Ambient Intelligence +
aa331fe378056b6d6031bb8fe6676e035ed60d6d
aae0e417bbfba701a1183d3d92cc7ad550ee59c3844 +
A Statistical Method for 2-D Facial Landmarking +
aa577652ce4dad3ca3dde44f881972ae6e1acce7Deep Attribute Networks +
Department of EE, KAIST +
Daejeon, South Korea +
Department of EE, KAIST +
Daejeon, South Korea +
Department of EE, KAIST +
Daejeon, South Korea +
Department of EE, KAIST +
Daejeon, South Korea +
aa94f214bb3e14842e4056fdef834a51aecef39cReconhecimento de padrões faciais: Um estudo +
Universidade Federal +
Rural do Semi-Árido +
Departamento de Ciências Naturais +
Mossoró, RN - 59625-900 +
Resumo—O reconhecimento facial tem sido utilizado em di- +
versas áreas para identificação e autenticação de usuários. Um +
dos principais mercados está relacionado a segurança, porém há +
uma grande variedade de aplicações relacionadas ao uso pessoal, +
conveniência, aumento de produtividade, etc. O rosto humano +
possui um conjunto de padrões complexos e mutáveis. Para +
reconhecer esses padrões, são necessárias técnicas avançadas de +
reconhecimento de padrões capazes, não apenas de reconhecer, +
mas de se adaptar às mudanças constantes das faces das pessoas. +
Este documento apresenta um método de reconhecimento facial +
proposto a partir da análise comparativa de trabalhos encontra- +
dos na literatura. +
biométrica é o uso da biometria para reconhecimento, identi- +
ficação ou verificação, de um ou mais traços biométricos de +
um indivíduo com o objetivo de autenticar sua identidade. Os +
traços biométricos são os atributos analisados pelas técnicas +
de reconhecimento biométrico. +
A tarefa de reconhecimento facial é composta por três +
processos distintos: Registro, verificação e identificação bio- +
métrica. Os processos se diferenciam pela forma de determinar +
a identidade de um indivíduo. Na Figura 1 são descritos os +
processos de registro, verificação e identificação biométrica. +
I. INTRODUÇÃO +
Biometria é a ciência que estabelece a identidade de um +
indivíduo baseada em seus atributos físicos, químicos ou +
comportamentais [1]. Possui inúmeras aplicações em diver- +
sas áreas, se destacando mais na área de segurança, como +
por exemplo sistemas de gerenciamento de identidade, cuja +
funcionalidade é autenticar a identidade de um indivíduo no +
contexto de uma aplicação. +
O reconhecimento facial é uma técnica biométrica que +
consiste em identificar padrões em características faciais como +
formato da boca, do rosto, distância dos olhos, entre outros. +
Um humano é capaz de reconhecer uma pessoa familiar +
mesmo com muitos obstáculos com distância, sombras ou +
apenas a visão parcial do rosto. Uma máquina, no entanto, +
precisa realizar inúmeros processos para detectar e reconhecer +
um conjunto de padrões específicos para rotular uma face +
como conhecida ou desconhecida. Para isso, exitem métodos +
capazes de detectar, extrair e classificar as características +
faciais, fornecendo um reconhecimento automático de pessoas. +
II. RECONHECIMENTO FACIAL +
A tecnologia biométrica oferece vantagens em relação a +
outros métodos tradicionais de identificação como senhas, +
documentos e tokens. Entre elas estão o fato de que os +
traços biométricos não podem ser perdidos ou esquecidos, são +
difíceis de serem copiados, compartilhados ou distribuídos. Os +
métodos requerem que a pessoa autenticada esteja presente +
na hora e lugar da autenticação, evitando que pessoas má +
intencionadas tenham acesso sem autorização. +
A autenticação é o ato de estabelecer ou confirmar alguém, +
ou alguma coisa, como autêntico, isto é, que as alegações +
feitas por ou sobre a coisa é verdadeira [2]. Autenticação +
(a) +
(b) +
(c) +
Figura 1: Registro biométrico (a), identificação biométrica (b) +
e verificação biométrica (c) +
A Figura 1a descreve o processo de registro de dados +
af8fe1b602452cf7fc9ecea0fd4508ed4149834e
af6e351d58dba0962d6eb1baf4c9a776eb73533fHow to Train Your Deep Neural Network with +
Dictionary Learning +
*IIIT Delhi +
Okhla Phase 3 +
Delhi, 110020, India +
+IIIT Delhi +
Okhla Phase 3 +
#IIIT Delhi +
Okhla Phase 3 +
Delhi, 110020, India +
Delhi, 110020, India +
af6cae71f24ea8f457e581bfe1240d5fa63faaf7
af54dd5da722e104740f9b6f261df9d4688a9712
afc7092987f0d05f5685e9332d83c4b27612f964Person-Independent Facial Expression Detection using Constrained +
Local Models +
b730908bc1f80b711c031f3ea459e4de09a3d3242024 +
Active Orientation Models for Face +
Alignment In-the-Wild +
b7cf7bb574b2369f4d7ebc3866b461634147041aNeural Comput & Applic (2012) 21:1575–1583 +
DOI 10.1007/s00521-011-0728-x +
O R I G I N A L A R T I C L E +
From NLDA to LDA/GSVD: a modified NLDA algorithm +
Received: 2 August 2010 / Accepted: 3 August 2011 / Published online: 19 August 2011 +
Ó Springer-Verlag London Limited 2011 +
b7eead8586ffe069edd190956bd338d82c69f880A VIDEO DATABASE FOR FACIAL +
BEHAVIOR UNDERSTANDING +
D. Freire-Obreg´on and M. Castrill´on-Santana. +
SIANI, Universidad de Las Palmas de Gran Canaria, Spain +
b75cee96293c11fe77ab733fc1147950abbe16f9
b7f05d0771da64192f73bdb2535925b0e238d233 MVA2005 IAPR Conference on Machine VIsion Applications, May 16-18, 2005 Tsukuba Science City, Japan +
4-3 +
Robust Active Shape Model using AdaBoosted Histogram Classifiers +
W ataru Ito +
Imaging Software Technology Center +
Imaging Software Technology Center +
FUJI PHOTO FILM CO., LTD. +
FUJI PHOTO FILM CO., LTD. +
b755505bdd5af078e06427d34b6ac2530ba69b12To appear in the International Joint Conf. Biometrics, Washington D.C., October, 2011 +
NFRAD: Near-Infrared Face Recognition at a Distance +
aDept. of Brain and Cognitive Eng. Korea Univ., Seoul, Korea +
bDept. of Comp. Sci. & Eng. Michigan State Univ., E. Lansing, MI, USA 48824 +
b73fdae232270404f96754329a1a18768974d3f6
b76af8fcf9a3ebc421b075b689defb6dc4282670Face Mask Extraction in Video Sequence +
b747fcad32484dfbe29530a15776d0df5688a7db
b7f7a4df251ff26aca83d66d6b479f1dc6cd1085Bouges et al. EURASIP Journal on Image and Video Processing 2013, 2013:55 +
http://jivp.eurasipjournals.com/content/2013/1/55 +
RESEARCH +
Open Access +
Handling missing weak classifiers in boosted +
cascade: application to multiview and +
occluded face detection +
db227f72bb13a5acca549fab0dc76bce1fb3b948International Refereed Journal of Engineering and Science (IRJES) +
ISSN (Online) 2319-183X, (Print) 2319-1821 +
Volume 4, Issue 6 (June 2015), PP.169-169-174 +
Characteristic Based Image Search using Re-Ranking method +
1Chitti Babu, 2Yasmeen Jaweed, 3G.Vijay Kumar +
dbaf89ca98dda2c99157c46abd136ace5bdc33b3Nonlinear Cross-View Sample Enrichment for +
Action Recognition +
Institut Mines-T´el´ecom; T´el´ecom ParisTech; CNRS LTCI +
dbab6ac1a9516c360cdbfd5f3239a351a64adde7
dbe255d3d2a5d960daaaba71cb0da292e0af36a7Evolutionary Cost-sensitive Extreme Learning +
Machine +
1 +
dbb0a527612c828d43bcb9a9c41f1bf7110b1dc8Chapter 7 +
Machine Learning Techniques +
for Face Analysis +
dbb7f37fb9b41d1aa862aaf2d2e721a470fd2c57Face Image Analysis With +
Convolutional Neural Networks +
Dissertation +
Zur Erlangung des Doktorgrades +
der Fakult¨at f¨ur Angewandte Wissenschaften +
an der Albert-Ludwigs-Universit¨at Freiburg im Breisgau +
von +
Stefan Duffner +
2007 +
a83fc450c124b7e640adc762e95e3bb6b423b310Deep Face Feature for Face Alignment +
a85e9e11db5665c89b057a124547377d3e1c27efDynamics of Driver’s Gaze: Explorations in +
Behavior Modeling & Maneuver Prediction +
a8117a4733cce9148c35fb6888962f665ae65b1eIEEE TRANSACTIONS ON XXXX, VOL. XX, NO. XX, XX 201X +
A Good Practice Towards Top Performance of Face +
Recognition: Transferred Deep Feature Fusion +
a8035ca71af8cc68b3e0ac9190a89fed50c92332000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
IIIT-CFW: A Benchmark Database of +
Cartoon Faces in the Wild +
1 IIIT Chittoor, Sri City, India +
2 CVIT, KCIS, IIIT Hyderabad, India +
a88640045d13fc0207ac816b0bb532e42bcccf36ARXIV VERSION +
Simultaneously Learning Neighborship and +
Projection Matrix for Supervised +
Dimensionality Reduction +
a8a30a8c50d9c4bb8e6d2dd84bc5b8b7f2c84dd8This is a repository copy of Modelling of Orthogonal Craniofacial Profiles. +
White Rose Research Online URL for this paper: +
http://eprints.whiterose.ac.uk/131767/ +
Version: Published Version +
Article: +
Dai, Hang, Pears, Nicholas Edwin orcid.org/0000-0001-9513-5634 and Duncan, Christian +
(2017) Modelling of Orthogonal Craniofacial Profiles. Journal of Imaging. ISSN 2313-433X +
https://doi.org/10.3390/jimaging3040055 +
Reuse +
This article is distributed under the terms of the Creative Commons Attribution (CC BY) licence. This licence +
allows you to distribute, remix, tweak, and build upon the work, even commercially, as long as you credit the +
authors for the original work. More information and the full terms of the licence here: +
https://creativecommons.org/licenses/ +
Takedown +
If you consider content in White Rose Research Online to be in breach of UK law, please notify us by +
https://eprints.whiterose.ac.uk/ +
a8e75978a5335fd3deb04572bb6ca43dbfad4738Sparse Graphical Representation based Discriminant +
Analysis for Heterogeneous Face Recognition +
ded968b97bd59465d5ccda4f1e441f24bac7ede5Noname manuscript No. +
(will be inserted by the editor) +
Large scale 3D Morphable Models +
Zafeiriou +
Received: date / Accepted: date +
de0eb358b890d92e8f67592c6e23f0e3b2ba3f66ACCEPTED BY IEEE TRANS. PATTERN ANAL. AND MACH. INTELL. +
Inference-Based Similarity Search in +
Randomized Montgomery Domains for +
Privacy-Preserving Biometric Identification +
dee406a7aaa0f4c9d64b7550e633d81bc66ff451Content-Adaptive Sketch Portrait Generation by +
Decompositional Representation Learning +
dedabf9afe2ae4a1ace1279150e5f1d495e565da3294 +
Robust Face Recognition With Structurally +
Incoherent Low-Rank Matrix Decomposition +
de398bd8b7b57a3362c0c677ba8bf9f1d8ade583Hierarchical Bayesian Theme Models for +
Multi-pose Facial Expression Recognition +
ded41c9b027c8a7f4800e61b7cfb793edaeb2817
defa8774d3c6ad46d4db4959d8510b44751361d8FEBEI - Face Expression Based Emoticon Identification +
CS - B657 Computer Vision +
Robert J Henderson - rojahend +
b0c512fcfb7bd6c500429cbda963e28850f2e948
b09b693708f412823053508578df289b8403100aWANG et al.: TWO-STREAM SR-CNNS FOR ACTION RECOGNITION IN VIDEOS +
Two-Stream SR-CNNs for Action +
Recognition in Videos +
1 Advanced Interactive Technologies Lab +
ETH Zurich +
Zurich, Switzerland +
2 Computer Vision Lab +
ETH Zurich +
Zurich, Switzerland +
b07582d1a59a9c6f029d0d8328414c7bef64dca0Employing Fusion of Learned and Handcrafted +
Features for Unconstrained Ear Recognition +
Maur´ıcio Pamplona Segundo∗† +
October 24, 2017 +
b03d6e268cde7380e090ddaea889c75f64560891
b0c1615ebcad516b5a26d45be58068673e2ff217How Image Degradations Affect Deep CNN-based Face +
Recognition? +
S¸amil Karahan1 Merve Kılınc¸ Yıldırım1 Kadir Kırtac¸1 Ferhat S¸ ¨ukr¨u Rende1 +
G¨ultekin B¨ut¨un1Hazım Kemal Ekenel2 +
b0de0892d2092c8c70aa22500fed31aa7eb4dd3f(will be inserted by the editor) +
A robust and efficient video representation for action recognition +
Received: date / Accepted: date +
a66d89357ada66d98d242c124e1e8d96ac9b37a0Failure Detection for Facial Landmark Detectors +
Computer Vision Lab, D-ITET, ETH Zurich, Switzerland +
a608c5f8fd42af6e9bd332ab516c8c2af7063c612408 +
Age Estimation via Grouping and Decision Fusion +
a6eb6ad9142130406fb4ffd4d60e8348c2442c29Video Description: A Survey of Methods, +
Datasets and Evaluation Metrics +
a6583c8daa7927eedb3e892a60fc88bdfe89a486
a6590c49e44aa4975b2b0152ee21ac8af3097d80https://doi.org/10.1007/s11263-018-1074-6 +
3D Interpreter Networks for Viewer-Centered Wireframe Modeling +
Received: date / Accepted: date +
a694180a683f7f4361042c61648aa97d222602dbFace Recognition using Scattering Wavelet under Illicit Drug Abuse Variations +
IIIT-Delhi India +
a6db73f10084ce6a4186363ea9d7475a9a658a11
a6634ff2f9c480e94ed8c01d64c9eb70e0d98487
b9d0774b0321a5cfc75471b62c8c5ef6c15527f5Fishy Faces: Crafting Adversarial Images to Poison Face Authentication +
imec-DistriNet, KU Leuven +
imec-DistriNet, KU Leuven +
imec-DistriNet, KU Leuven +
imec-DistriNet, KU Leuven +
imec-DistriNet, KU Leuven +
b908edadad58c604a1e4b431f69ac8ded350589aDeep Face Feature for Face Alignment +
b9f2a755940353549e55690437eb7e13ea226bbfUnsupervised Feature Learning from Videos for Discovering and Recognizing Actions +
b9cedd1960d5c025be55ade0a0aa81b75a6efa61INEXACT KRYLOV SUBSPACE ALGORITHMS FOR LARGE +
MATRIX EXPONENTIAL EIGENPROBLEM FROM +
DIMENSIONALITY REDUCTION +
b971266b29fcecf1d5efe1c4dcdc2355cb188ab0MAI et al.: ON THE RECONSTRUCTION OF FACE IMAGES FROM DEEP FACE TEMPLATES +
On the Reconstruction of Face Images from +
Deep Face Templates +
a158c1e2993ac90a90326881dd5cb0996c20d4f3OPEN ACCESS +
ISSN 2073-8994 +
Article +
1 DMA, Università degli Studi di Palermo, via Archirafi 34, 90123 Palermo, Italy +
2 CITC, Università degli Studi di Palermo, via Archirafi 34, 90123 Palermo, Itlay +
3 Istituto Nazionale di Ricerche Demopolis, via Col. Romey 7, 91100 Trapani, Italy +
† Deceased on 15 March 2009. +
Received: 4 March 2010; in revised form: 23 March 2010 / Accepted: 29 March 2010 / +
Published: 1 April 2010 +
a15d9d2ed035f21e13b688a78412cb7b5a04c469Object Detection Using +
Strongly-Supervised Deformable Part Models +
1Computer Vision and Active Perception Laboratory (CVAP), KTH, Sweden +
2INRIA, WILLOW, Laboratoire d’Informatique de l’Ecole Normale Superieure +
a1b1442198f29072e907ed8cb02a064493737158456 +
Crowdsourcing Facial Responses +
to Online Videos +
a15c728d008801f5ffc7898568097bbeac8270a4Concise Preservation by Combining Managed Forgetting +
and Contextualized Remembering +
Grant Agreement No. 600826 +
Deliverable D4.4 +
Work-package +
Deliverable +
Deliverable Leader +
Quality Assessor +
Dissemination level +
Delivery date in Annex I +
Actual delivery date +
Revisions +
Status +
Keywords +
Information Consolidation and Con- +
WP4: +
centration +
D4.4: +
Information analysis, consolidation +
and concentration techniques, and evalua- +
tion - Final release. +
Vasileios Mezaris (CERTH) +
Walter Allasia (EURIX) +
PU +
31-01-2016 (M36) +
31-01-2016 +
Final +
multidocument summarization, semantic en- +
richment, +
feature extraction, concept de- +
tection, event detection, image/video qual- +
ity, image/video aesthetic quality, face de- +
tection/clustering, +
im- +
age/video summarization, image/video near +
duplicate detection, data deduplication, con- +
densation, consolidation +
image clustering, +
a1132e2638a8abd08bdf7fc4884804dd6654fa636 +
Real-Time Video Face Recognition +
for Embedded Devices +
Tessera, Galway, +
Ireland +
1. Introduction +
This chapter will address the challenges of real-time video face recognition systems +
implemented in embedded devices. Topics to be covered include: the importance and +
challenges of video face recognition in real life scenarios, describing a general architecture of +
a generic video face recognition system and a working solution suitable for recognizing +
faces in real-time using low complexity devices. Each component of the system will be +
described together with the system’s performance on a database of video samples that +
resembles real life conditions. +
2. Video face recognition +
Face recognition remains a very active topic in computer vision and receives attention from +
a large community of researchers in that discipline. Many reasons feed this interest; the +
main being the wide range of commercial, law enforcement and security applications that +
require authentication. The progress made in recent years on the methods and algorithms +
for data processing as well as the availability of new technologies makes it easier to study +
these algorithms and turn them into commercially viable product. Biometric based security +
systems are becoming more popular due to their non-invasive nature and their increasing +
reliability. Surveillance applications based on face recognition are gaining increasing +
attention after the United States’ 9/11 events and with the ongoing security threats. The +
Face Recognition Vendor Test (FRVT) (Phillips et al., 2003) includes video face recognition +
testing starting with the 2002 series of tests. +
Recently, face recognition technology was deployed in consumer applications such as +
organizing a collection of images using the faces present in the images (Picassa; Corcoran & +
Costache, 2005), prioritizing family members for best capturing conditions when taking +
pictures, or directly annotating the images as they are captured (Costache et al., 2006). +
Video face recognition, compared with more traditional still face recognition, has the main +
advantage of using multiple instances of the same individual in sequential frames for +
recognition to occur. In still recognition case, the system has only one input image to make +
the decision if the person is or is not in the database. If the image is not suitable for +
recognition (due to face orientation, expression, quality or facial occlusions) the recognition +
result will most likely be incorrect. In the video image there are multiple frames which can +
www.intechopen.com +
a14ae81609d09fed217aa12a4df9466553db4859REVISED VERSION, JUNE 2011 +
Face Identification Using Large Feature Sets +
a1e97c4043d5cc9896dc60ae7ca135782d89e5fcIEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +
Re-identification of Humans in Crowds using +
Personal, Social and Environmental Constraints +
efd308393b573e5410455960fe551160e1525f49Tracking Persons-of-Interest via +
Unsupervised Representation Adaptation +
ef4ecb76413a05c96eac4c743d2c2a3886f2ae07Modeling the Importance of Faces in Natural Images +
Jin B.a, Yildirim G.a, Lau C.a, Shaji A.a, Ortiz Segovia M.b and S¨usstrunk S.a +
aEPFL, Lausanne, Switzerland; +
bOc´e, Paris, France +
ef032afa4bdb18b328ffcc60e2dc5229cc1939bcFang and Yuan EURASIP Journal on Image and Video +
Processing (2018) 2018:44 +
https://doi.org/10.1186/s13640-018-0282-x +
EURASIP Journal on Image +
and Video Processing +
RESEARCH +
Open Access +
Attribute-enhanced metric learning for +
face retrieval +
ef5531711a69ed687637c48930261769465457f0Studio2Shop: from studio photo shoots to fashion articles +
Zalando Research, Muehlenstr. 25, 10243 Berlin, Germany +
Keywords: +
computer vision, deep learning, fashion, item recognition, street-to-shop +
efa08283656714911acff2d5022f26904e451113Active Object Localization in Visual Situations +
ef999ab2f7b37f46445a3457bf6c0f5fd7b5689dCalhoun: The NPS Institutional Archive +
DSpace Repository +
Theses and Dissertations +
1. Thesis and Dissertation Collection, all items +
2017-12 +
Improving face verification in photo albums by +
combining facial recognition and metadata +
with cross-matching +
Monterey, California: Naval Postgraduate School +
http://hdl.handle.net/10945/56868 +
Downloaded from NPS Archive: Calhoun +
c3beae515f38daf4bd8053a7d72f6d2ed3b05d88
c3dc4f414f5233df96a9661609557e341b71670dTao et al. EURASIP Journal on Advances in Signal Processing 2011, 2011:4 +
http://asp.eurasipjournals.com/content/2011/1/4 +
RESEARCH +
Utterance independent bimodal emotion +
recognition in spontaneous communication +
Open Access +
c398684270543e97e3194674d9cce20acaef3db3Chapter 2 +
Comparative Face Soft Biometrics for +
Human Identification +
c3285a1d6ec6972156fea9e6dc9a8d88cd001617
c3418f866a86dfd947c2b548cbdeac8ca5783c15
c32383330df27625592134edd72d69bb6b5cff5c422 +
IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 42, NO. 2, APRIL 2012 +
Intrinsic Illumination Subspace for Lighting +
Insensitive Face Recognition +
c3a3f7758bccbead7c9713cb8517889ea6d04687
c30e4e4994b76605dcb2071954eaaea471307d80
c37a971f7a57f7345fdc479fa329d9b425ee02beA Novice Guide towards Human Motion Analysis and Understanding +
c3638b026c7f80a2199b5ae89c8fcbedfc0bd8af
c3fb2399eb4bcec22723715556e31c44d086e054499 +
2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +
978-1-4799-2893-4/14/$31.00 ©2014 IEEE +
1. INTRODUCTION +
c37de914c6e9b743d90e2566723d0062bedc9e6a©2016 Society for Imaging Science and Technology +
DOI: 10.2352/ISSN.2470-1173.2016.11.IMAWM-455 +
Joint and Discriminative Dictionary Learning +
Expression Recognition +
for Facial +
c4f1fcd0a5cdaad8b920ee8188a8557b6086c1a4Int J Comput Vis (2014) 108:3–29 +
DOI 10.1007/s11263-014-0698-4 +
The Ignorant Led by the Blind: A Hybrid Human–Machine Vision +
System for Fine-Grained Categorization +
Received: 7 March 2013 / Accepted: 8 January 2014 / Published online: 20 February 2014 +
© Springer Science+Business Media New York 2014 +
c4dcf41506c23aa45c33a0a5e51b5b9f8990e8ad Understanding Activity: Learning the Language of Action +
Univ. of Rochester and Maryland +
1.1 Overview +
Understanding observed activity is an important +
problem, both from the standpoint of practical applications, +
and as a central issue in attempting to describe the +
phenomenon of intelligence. On the practical side, there are a +
large number of applications that would benefit from +
improved machine ability to analyze activity. The most +
prominent are various surveillance scenarios. The current +
emphasis on homeland security has brought this issue to the +
forefront, and resulted in considerable work on mostly low- +
level detection schemes. There are also applications in +
medical diagnosis and household assistants that, in the long +
run, may be even more important. In addition, there are +
numerous scientific projects, ranging from monitoring of +
weather conditions to observation of animal behavior that +
would be facilitated by automatic understanding of activity. +
From a scientific standpoint, understanding activity +
understanding is central to understanding intelligence. +
Analyzing what is happening in the environment, and acting +
on the results of that analysis is, to a large extent, what +
natural intelligent systems do, whether they are human or +
animal. Artificial intelligences, if we want them to work with +
people in the natural world, will need commensurate abilities. +
The importance of the problem has not gone unrecognized. +
There is a substantial body of work on various components of +
the problem, most especially on change detection, motion +
analysis, and tracking. More recently, in the context of +
surveillance applications, there have been some preliminary +
efforts to come up with a general ontology of human activity. +
These efforts have largely been top-down in the classic AI +
tradition, and, as with earlier analogous effort in areas such +
as object recognition and scene understanding, have seen +
limited practical application because of the difficulty in +
robustly extracting the putative primitives on which the top- +
down formalism is based. We propose a novel alternative +
approach, where understanding activity is centered on +
c49aed65fcf9ded15c44f9cbb4b161f851c6fa88Multiscale Facial Expression Recognition using Convolutional Neural Networks +
IDIAP, Martigny, Switzerland +
eac6aee477446a67d491ef7c95abb21867cf71fcJOURNAL +
A survey of sparse representation: algorithms and +
applications +
ea482bf1e2b5b44c520fc77eab288caf8b3f367aProceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
2592 +
eafda8a94e410f1ad53b3e193ec124e80d57d095Jeffrey F. Cohn +
13 +
Observer-Based Measurement of Facial Expression +
With the Facial Action Coding System +
Facial expression has been a focus of emotion research for over +
a hundred years (Darwin, 1872/1998). It is central to several +
leading theories of emotion (Ekman, 1992; Izard, 1977; +
Tomkins, 1962) and has been the focus of at times heated +
debate about issues in emotion science (Ekman, 1973, 1993; +
Fridlund, 1992; Russell, 1994). Facial expression figures +
prominently in research on almost every aspect of emotion, +
including psychophysiology (Levenson, Ekman, & Friesen, +
1990), neural bases (Calder et al., 1996; Davidson, Ekman, +
Saron, Senulis, & Friesen, 1990), development (Malatesta, +
Culver, Tesman, & Shephard, 1989; Matias & Cohn, 1993), +
perception (Ambadar, Schooler, & Cohn, 2005), social pro- +
cesses (Hatfield, Cacioppo, & Rapson, 1992; Hess & Kirouac, +
2000), and emotion disorder (Kaiser, 2002; Sloan, Straussa, +
Quirka, & Sajatovic, 1997), to name a few. +
Because of its importance to the study of emotion, a num- +
ber of observer-based systems of facial expression measure- +
ment have been developed (Ekman & Friesen, 1978, 1982; +
Ekman, Friesen, & Tomkins, 1971; Izard, 1979, 1983; Izard +
& Dougherty, 1981; Kring & Sloan, 1991; Tronick, Als, & +
Brazelton, 1980). Of these various systems for describing +
facial expression, the Facial Action Coding System (FACS; +
Ekman & Friesen, 1978; Ekman, Friesen, & Hager, 2002) is +
the most comprehensive, psychometrically rigorous, and +
widely used (Cohn & Ekman, 2005; Ekman & Rosenberg, +
2005). Using FACS and viewing video-recorded facial behav- +
ior at frame rate and slow motion, coders can manually code +
nearly all possible facial expressions, which are decomposed +
into action units (AUs). Action units, with some qualifica- +
tions, are the smallest visually discriminable facial move- +
ments. By comparison, other systems are less thorough +
(Malatesta et al., 1989), fail to differentiate between some +
anatomically distinct movements (Oster, Hegley, & Nagel, +
1992), consider movements that are not anatomically dis- +
tinct as separable (Oster et al., 1992), and often assume a one- +
to-one mapping between facial expression and emotion (for +
a review of these systems, see Cohn & Ekman, in press). +
Unlike systems that use emotion labels to describe ex- +
pression, FACS explicitly distinguishes between facial actions +
and inferences about what they mean. FACS itself is descrip- +
tive and includes no emotion-specified descriptors. Hypoth- +
eses and inferences about the emotional meaning of facial +
actions are extrinsic to FACS. If one wishes to make emo- +
tion-based inferences from FACS codes, a variety of related +
resources exist. These include the FACS Investigators’ Guide +
(Ekman et al., 2002), the FACS interpretive database (Ekman, +
Rosenberg, & Hager, 1998), and a large body of empirical +
research.(Ekman & Rosenberg, 2005). These resources sug- +
gest combination rules for defining emotion-specified expres- +
sions from FACS action units, but this inferential step remains +
extrinsic to FACS. Because of its descriptive power, FACS +
is regarded by many as the standard measure for facial be- +
havior and is used widely in diverse fields. Beyond emo- +
tion science, these include facial neuromuscular disorders +
(Van Swearingen & Cohn, 2005), neuroscience (Bruce & +
Young, 1998; Rinn, 1984, 1991), computer vision (Bartlett, +
203 +
UNPROOFED PAGES
ea85378a6549bb9eb9bcc13e31aa6a61b655a9afDiplomarbeit +
Template Protection for PCA-LDA-based 3D +
Face Recognition System +
von +
Technische Universität Darmstadt +
Fachbereich Informatik +
Fachgebiet Graphisch-Interaktive Systeme +
Fraunhoferstraße 5 +
64283 Darmstadt +
ea2ee5c53747878f30f6d9c576fd09d388ab0e2bViola-Jones based Detectors: How much affects +
the Training Set? +
SIANI +
Edif. Central del Parque Cient´ıfico Tecnol´ogico +
Universidad de Las Palmas de Gran Canaria +
35017 - Spain +
ea96bc017fb56593a59149e10d5f14011a3744a0
e10a257f1daf279e55f17f273a1b557141953ce2
e171fba00d88710e78e181c3e807c2fdffc6798a
e1ab3b9dee2da20078464f4ad8deb523b5b1792ePre-Training CNNs Using Convolutional +
Autoencoders +
TU Berlin +
TU Berlin +
Sabbir Ahmmed +
TU Berlin +
TU Berlin +
e16efd2ae73a325b7571a456618bfa682b51aef8
e19ebad4739d59f999d192bac7d596b20b887f78Learning Gating ConvNet for Two-Stream based Methods in Action +
Recognition +
e13360cda1ebd6fa5c3f3386c0862f292e4dbee4
e1d726d812554f2b2b92cac3a4d2bec678969368J Electr Eng Technol.2015; 10(?): 30-40 +
http://dx.doi.org/10.5370/JEET.2015.10.2.030 +
ISSN(Print) +
1975-0102 +
ISSN(Online) 2093-7423 +
Human Action Recognition Bases on Local Action Attributes +
and Mohan S Kankanhalli** +
e1e6e6792e92f7110e26e27e80e0c30ec36ac9c2TSINGHUA SCIENCE AND TECHNOLOGY +
ISSNll1007-0214 +
0?/?? pp???–??? +
DOI: 10.26599/TST.2018.9010000 +
Volume 1, Number 1, Septembelr 2018 +
Ranking with Adaptive Neighbors +
cd9666858f6c211e13aa80589d75373fd06f6246A Novel Time Series Kernel for +
Sequences Generated by LTI Systems +
V.le delle Scienze Ed.6, DIID, Universit´a degli studi di Palermo, Italy +
cd4c047f4d4df7937aff8fc76f4bae7718004f40
cd596a2682d74bdfa7b7160dd070b598975e89d9Mood Detection: Implementing a facial +
expression recognition system +
1. Introduction +
Facial expressions play a significant role in human dialogue. As a result, there has been +
considerable work done on the recognition of emotional expressions and the application of this +
research will be beneficial in improving human-machine dialogue. One can imagine the +
improvements to computer interfaces, automated clinical (psychological) research or even +
interactions between humans and autonomous robots. +
Unfortunately, a lot of the literature does not focus on trying to achieve high recognition rates +
across multiple databases. In this project we develop our own mood detection system that +
addresses this challenge. The system involves pre-processing image data by normalizing and +
applying a simple mask, extracting certain (facial) features using PCA and Gabor filters and then +
using SVMs for classification and recognition of expressions. Eigenfaces for each class are used +
to determine class-specific masks which are then applied to the image data and used to train +
multiple, one against the rest, SVMs. We find that simply using normalized pixel intensities +
works well with such an approach. +
Figure 1 – Overview of our system design +
2. Image pre-processing +
We performed pre-processing on the images used to train and test our algorithms as follows: +
1. The location of the eyes is first selected manually +
2. Images are scaled and cropped to a fixed size (170 x 130) keeping the eyes in all images +
aligned +
3. The image is histogram equalized using the mean histogram of all the training images to +
make it invariant to lighting, skin color etc. +
4. A fixed oval mask is applied to the image to extract face region. This serves to eliminate +
the background, hair, ears and other extraneous features in the image which provide no +
information about facial expression. +
This approach works reasonably well in capturing expression-relevant facial information across +
all databases. Examples of pre-processed images from the various datasets are shown in Figure- +
2a below. +
cda4fb9df653b5721ad4fe8b4a88468a410e55ecGabor wavelet transform and its application +
cd3005753012409361aba17f3f766e33e3a7320dMultilinear Biased Discriminant Analysis: A Novel Method for Facial +
Action Unit Representation +
cd7a7be3804fd217e9f10682e0c0bfd9583a08dbWomen also Snowboard: +
Overcoming Bias in Captioning Models +
ccfcbf0eda6df876f0170bdb4d7b4ab4e7676f18JOURNAL OF LATEX CLASS FILES, VOL. 6, NO. 1, JUNE 2011 +
A Dynamic Appearance Descriptor Approach to +
Facial Actions Temporal Modelling +
ccbfc004e29b3aceea091056b0ec536e8ea7c47e
cc3c273bb213240515147e8be68c50f7ea22777cGaining Insight Into Films +
Via Topic Modeling & Visualization +
KEYWORDS Collaboration, computer vision, cultural +
analytics, economy of abundance, interactive data +
visualization +
We moved beyond misuse when the software actually +
became useful for film analysis with the addition of audio +
analysis, subtitle analysis, facial recognition, and topic +
modeling. Using multiple types of visualizations and +
a back-and-fourth workflow between people and AI +
we arrived at an approach for cultural analytics that +
can be used to review and develop film criticism. Finally, +
we present ways to apply these techniques to Database +
Cinema and other aspects of film and video creation. +
PROJECT DATE 2014 +
URL http://misharabinovich.com/soyummy.html +
cc8e378fd05152a81c2810f682a78c5057c8a735International Journal of Computer Sciences and Engineering Open Access +
Research Paper Volume-5, Issue-12 E-ISSN: 2347-2693 +
Expression Invariant Face Recognition System based on Topographic +
Independent Component Analysis and Inner Product Classifier +
+
Department of Electrical Engineering, IIT Delhi, New Delhi, India +
Available online at: www.ijcseonline.org +
Received: 07/Nov/2017, Revised: 22/Nov/2017, Accepted: 14/Dec/2017, Published: 31/Dec/2017 +
cc31db984282bb70946f6881bab741aa841d3a7cALBANIE, VEDALDI: LEARNING GRIMACES BY WATCHING TV +
Learning Grimaces by Watching TV +
http://www.robots.ox.ac.uk/~albanie +
http://www.robots.ox.ac.uk/~vedaldi +
Engineering Science Department +
Univeristy of Oxford +
Oxford, UK +
cc8bf03b3f5800ac23e1a833447c421440d92197
cc96eab1e55e771e417b758119ce5d7ef1722b43An Empirical Study of Recent +
Face Alignment Methods +
e64b683e32525643a9ddb6b6af8b0472ef5b6a37Face Recognition and Retrieval in Video +
e6b45d5a86092bbfdcd6c3c54cda3d6c3ac6b227Pairwise Relational Networks for Face +
Recognition +
1 Department of Creative IT Engineering, POSTECH, Korea +
2 Department of Computer Science and Engineering, POSTECH, Korea +
e6865b000cf4d4e84c3fe895b7ddfc65a9c4aaecChapter 15. The critical role of the +
cold-start problem and incentive systems +
in emotional Web 2.0 services +
e6dc1200a31defda100b2e5ddb27fb7ecbbd4acd1921 +
Flexible Manifold Embedding: A Framework +
for Semi-Supervised and Unsupervised +
Dimension Reduction +
0 = +
, the linear regression function ( +
e6e5a6090016810fb902b51d5baa2469ae28b8a1Title +
Energy-Efficient Deep In-memory Architecture for NAND +
Flash Memories +
Archived version +
Accepted manuscript: the content is same as the published +
paper but without the final typesetting by the publisher +
Published version +
DOI +
Published paper +
URL +
Authors (contact) +
10.1109/ISCAS.2018.8351458 +
e6540d70e5ffeed9f447602ea3455c7f0b38113e
e6ee36444038de5885473693fb206f49c1369138
f913bb65b62b0a6391ffa8f59b1d5527b7eba948
f96bdd1e2a940030fb0a89abbe6c69b8d7f6f0c1
f0cee87e9ecedeb927664b8da44b8649050e1c86
f0f4f16d5b5f9efe304369120651fa688a03d495Temporal Generative Adversarial Nets +
Preferred Networks inc., Japan +
f06b015bb19bd3c39ac5b1e4320566f8d83a0c84
f0a3f12469fa55ad0d40c21212d18c02be0d1264Sparsity Sharing Embedding for Face +
Verification +
Department of Electrical Engineering, KAIST, Daejeon, Korea +
f7dea4454c2de0b96ab5cf95008ce7144292e52a
f7b422df567ce9813926461251517761e3e6cda0FACE AGING WITH CONDITIONAL GENERATIVE ADVERSARIAL NETWORKS +
(cid:63) Orange Labs, 4 rue Clos Courtel, 35512 Cesson-S´evign´e, France +
† Eurecom, 450 route des Chappes, 06410 Biot, France +
f79c97e7c3f9a98cf6f4a5d2431f149ffacae48fProvided by the author(s) and NUI Galway in accordance with publisher policies. Please cite the published +
version when available. +
Title +
On color texture normalization for active appearance models +
Author(s) +
Ionita, Mircea C.; Corcoran, Peter M.; Buzuloiu, Vasile +
Publication +
Date +
2009-05-12 +
Publication +
Information +
Ionita, M. C., Corcoran, P., & Buzuloiu, V. (2009). On Color +
Texture Normalization for Active Appearance Models. Image +
Processing, IEEE Transactions on, 18(6), 1372-1378. +
Publisher +
IEEE +
Link to +
publisher's +
version +
http://dx.doi.org/10.1109/TIP.2009.2017163 +
Item record +
http://hdl.handle.net/10379/1350 +
Some rights reserved. For more information, please see the item record link above. +
Downloaded 2017-06-17T22:38:27Z +
f7452a12f9bd927398e036ea6ede02da79097e6e
f7dcadc5288653ec6764600c7c1e2b49c305dfaaCopyright +
by +
Adriana Ivanova Kovashka +
2014 +
f7de943aa75406fe5568fdbb08133ce0f9a765d4Project 1.5: Human Identification at a Distance - Hornak, Adjeroh, Cukic, Gautum, & Ross +
Project 1.5 +
Biometric Identification and Surveillance1 +
Year 5 Deliverable  +
Technical Report:  +
and +
Research Challenges in Biometrics +
Indexed biography of relevant biometric research literature +
Donald Adjeroh, Bojan Cukic, Arun Ross  +
April, 2014   +
                                                             +
1 "This research was supported by the United States Department of Homeland Security through the National Center for Border Security +
and Immigration (BORDERS) under grant number 2008-ST-061-BS0002. However, any opinions, findings, and conclusions or +
recommendations in this document are those of the authors and do not necessarily reflect views of the United States Department of +
Homeland Security." +
f75852386e563ca580a48b18420e446be45fcf8dILLUMINATION INVARIANT FACE RECOGNITION +
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
ENEE 631: Digital Image and Video Processing +
Instructor: Dr. K. J. Ray Liu +
Term Project - Spring 2006 +
1. +
INTRODUCTION +
+
+
The performance of the Face Recognition algorithms is severely affected by two +
important factors: the change in Pose and Illumination conditions of the subjects. The +
changes in Illumination conditions of the subjects can be so drastic that, the variation in +
lighting will be of the similar order as that of the variation due to the change in subjects +
[1] and this can result in misclassification. +
+
For example, in the acquisition of the face of a person from a real time video, the +
ambient conditions will cause different lighting variations on the tracked face. Some +
examples of images with different illumination conditions are shown in Fig. 1. In this +
project, we study some algorithms that are capable of performing Illumination Invariant +
Face Recognition. The performances of these algorithms were compared on the CMU- +
Illumination dataset [13], by using the entire face as the input to the algorithms. Then, a +
model of dividing the face into four regions is proposed and the performance of the +
algorithms on these new features is analyzed. +
+
+
f78863f4e7c4c57744715abe524ae4256be884a9
f77c9bf5beec7c975584e8087aae8d679664a1ebLocal Deep Neural Networks for Age and Gender Classification +
March 27, 2017 +
e8410c4cd1689829c15bd1f34995eb3bd4321069
e8fdacbd708feb60fd6e7843b048bf3c4387c6dbDeep Learning +
Hinnerup Net A/S +
www.hinnerup.net +
July 4, 2014 +
Introduction +
Deep learning is a topic in the field of artificial intelligence (AI) and is a relatively +
new research area although based on the popular artificial neural networks (supposedly +
mirroring brain function). With the development of the perceptron in the 1950s and +
1960s by Frank RosenBlatt, research began on artificial neural networks. To further +
mimic the architectural depth of the brain, researchers wanted to train a deep multi- +
layer neural network – this, however, did not happen until Geoffrey Hinton in 2006 +
introduced Deep Belief Networks [1]. +
Recently, the topic of deep learning has gained public interest. Large web companies such +
as Google and Facebook have a focused research on AI and an ever increasing amount +
of compute power, which has led to researchers finally being able to produce results +
that are of interest to the general public. In July 2012 Google trained a deep learning +
network on YouTube videos with the remarkable result that the network learned to +
recognize humans as well as cats [6], and in January this year Google successfully used +
deep learning on Street View images to automatically recognize house numbers with +
an accuracy comparable to that of a human operator [5]. In March this year Facebook +
announced their DeepFace algorithm that is able to match faces in photos with Facebook +
users almost as accurately as a human can do [9]. +
Deep learning and other AI are here to stay and will become more and more present in +
our daily lives, so we had better make ourselves acquainted with the technology. Let’s +
dive into the deep water and try not to drown! +
Data Representations +
Before presenting data to an AI algorithm, we would normally prepare the data to make +
it feasible to work with. For instance, if the data consists of images, we would take each +
e8b2a98f87b7b2593b4a046464c1ec63bfd13b51CMS-RCNN: Contextual Multi-Scale +
Region-based CNN for Unconstrained Face +
Detection +
e8c6c3fc9b52dffb15fe115702c6f159d955d30813 +
Linear Subspace Learning for +
Facial Expression Analysis +
Philips Research +
The Netherlands +
1. Introduction +
Facial expression, resulting from movements of the facial muscles, is one of the most +
powerful, natural, and immediate means for human beings to communicate their emotions +
and intentions. Some examples of facial expressions are shown in Fig. 1. Darwin (1872) was +
the first to describe in detail the specific facial expressions associated with emotions in +
animals and humans; he argued that all mammals show emotions reliably in their faces. +
Psychological studies (Mehrabian, 1968; Ambady & Rosenthal, 1992) indicate that facial +
expressions, with other non-verbal cues, play a major and fundamental role in face-to-face +
communication. +
Fig. 1. Facial expressions of George W. Bush. +
Machine analysis of facial expressions, enabling computers to analyze and interpret facial +
expressions as humans do, has many important applications including intelligent human- +
computer interaction, computer animation, surveillance and security, medical diagnosis, +
law enforcement, and awareness system (Shan, 2007). Driven by its potential applications +
and theoretical interests of cognitive and psychological scientists, automatic facial +
expression analysis has attracted much attention in last two decades (Pantic & Rothkrantz, +
2000a; Fasel & Luettin, 2003; Tian et al, 2005; Pantic & Bartlett, 2007). It has been studied in +
multiple disciplines such as psychology, cognitive science, computer vision, pattern +
Source: Machine Learning, Book edited by: Abdelhamid Mellouk and Abdennacer Chebira, +
ISBN 978-3-902613-56-1, pp. 450, February 2009, I-Tech, Vienna, Austria +
www.intechopen.com +
fab83bf8d7cab8fe069796b33d2a6bd70c8cefc6Draft: Evaluation Guidelines for Gender +
Classification and Age Estimation +
July 1, 2011 +
Introduction +
In previous research on gender classification and age estimation did not use a +
standardised evaluation procedure. This makes comparison the different ap- +
proaches difficult. +
Thus we propose here a benchmarking and evaluation protocol for gender +
classification as well as age estimation to set a common ground for future re- +
search in these two areas. +
The evaluations are designed such that there is one scenario under controlled +
labratory conditions and one under uncontrolled real life conditions. +
The datasets were selected with the criteria of being publicly available for +
research purposes. +
File lists for the folds corresponding to the individual benchmarking proto- +
cols will be provided over our website at http://face.cs.kit.edu/befit. We +
will provide two kinds of folds for each of the tasks and conditions: one set of +
folds using the whole dataset and one set of folds using a reduced dataset, which +
is approximately balanced in terms of age, gender and ethnicity. +
2 Gender Classification +
In this task the goal is to determine the gender of the persons depicted in the +
individual images. +
2.1 Data +
In previous works one of the most commonly used databases is the Feret database [1, +
2]. We decided here not to take this database, because of its low number of im- +
ages. +
fa08a4da5f2fa39632d90ce3a2e1688d147ece61Supplementary material for +
“Unsupervised Creation of Parameterized Avatars” +
1 Summary of Notations +
Tab. 1 itemizes the symbols used in the submission. Fig. 2,3,4 of the main text illustrate many of these +
symbols. +
2 DANN results +
Fig. 1 shows side by side samples of the original image and the emoji generated by the method of [1]. +
As can be seen, these results do not preserve the identity very well, despite considerable effort invested in +
finding suitable architectures. +
3 Multiple Images Per Person +
Following [4], we evaluate the visual quality that is obtained per person and not just per image, by testing +
TOS on the Facescrub dataset [3]. For each person p, we considered the set of their images Xp, and selected +
the emoji that was most similar to their source image, i.e., the one for which: +
||f (x) − f (e(c(G(x))))||. +
argmin +
x∈Xp +
(1) +
Fig. 2 depicts the results obtained by this selection method on sample images form the Facescrub dataset +
(it is an extension of Fig. 7 of the main text). The figure also shows, for comparison, the DTN [4] result for +
the same image. +
4 Detailed Architecture of the Various Networks +
In this section we describe the architectures of the networks used in for the emoji and avatar experiments. +
4.1 TOS +
Network g maps DeepFace’s 256-dimensional representation [5] into 64 × 64 RGB emoji images. Follow- +
ing [4], this is done through a network with 9 blocks, each consisting of a convolution, batch-normalization +
and ReLU, except the last layer which employs Tanh activation. The odd blocks 1,3,5,7,9 perform upscaling +
convolutions with 512-256-128-64-3 filters respectively of spatial size 4 × 4. The even ones perform 1 × 1 +
convolutions [2]. The odd blocks use a stride of 2 and padding of 1, excluding the first one which does not +
use stride or padding. +
Network e maps emoji parameterization into the matching 64× 64 RGB emoji. The parameterization is +
given as binary vectors in R813 for emojis; Avatar parameterization is in R354. While there are dependencies +
among the various dimensions (an emoji cannot have two hairstyles at once), the binary representation is +
chosen for its simplicity and generality. e is trained in a fully supervised way, using pairs of matching +
parameterization vectors and images in a supervised manner. +
The architecture of e employs five upscaling convolutions with 512-256-128-64-3 filters respectively, +
each of spatial size 4×4. All layers except the last one are batch normalized followed by a ReLU activation. +
The last layer is followed by Tanh activation, generating an RGB image with values in range [−1, 1]. All +
the layers use a stride of 2 and padding of 1, excluding the first one which does not use stride or padding. +
faead8f2eb54c7bc33bc7d0569adc7a4c2ec4c3b
faf5583063682e70dedc4466ac0f74eeb63169e7
fad895771260048f58d12158a4d4d6d0623f4158Audio-Visual Emotion +
Recognition For Natural +
Human-Robot Interaction +
Dissertation zur Erlangung des akademischen Grades +
Doktor der Ingenieurwissenschaften (Dr.-Ing.) +
vorgelegt von +
an der Technischen Fakultät der Universität Bielefeld +
15. März 2010 +
ff8315c1a0587563510195356c9153729b533c5b432 +
Zapping Index:Using Smile to Measure +
Advertisement Zapping Likelihood +
ff44d8938c52cfdca48c80f8e1618bbcbf91cb2aTowards Video Captioning with Naming: a +
Novel Dataset and a Multi-Modal Approach +
Dipartimento di Ingegneria “Enzo Ferrari” +
Universit`a degli Studi di Modena e Reggio Emilia +
fffefc1fb840da63e17428fd5de6e79feb726894Fine-Grained Age Estimation in the wild with +
Attention LSTM Networks +
ff398e7b6584d9a692e70c2170b4eecaddd78357
ffd81d784549ee51a9b0b7b8aaf20d5581031b74Performance Analysis of Retina and DoG +
Filtering Applied to Face Images for Training +
Correlation Filters +
Everardo Santiago Ram(cid:19)(cid:16)rez1, Jos(cid:19)e (cid:19)Angel Gonz(cid:19)alez Fraga1, Omar (cid:19)Alvarez +
1 Facultad de Ciencias, Universidad Aut(cid:19)onoma de Baja California, +
Carretera Transpeninsular Tijuana-Ensenada, N(cid:19)um. 3917, Colonia Playitas, +
Ensenada, Baja California, C.P. 22860 +
{everardo.santiagoramirez,angel_fraga, +
2 Facultad de Ingenier(cid:19)(cid:16)a, Arquitectura y Dise~no, Universidad Aut(cid:19)onoma de Baja +
California, Carretera Transpeninsular Tijuana-Ensenada, N(cid:19)um. 3917, Colonia +
Playitas, Ensenada, Baja California, C.P. 22860 +
ff60d4601adabe04214c67e12253ea3359f4e082
ff8ef43168b9c8dd467208a0b1b02e223b731254BreakingNews: Article Annotation by +
Image and Text Processing +
ffcbedb92e76fbab083bb2c57d846a2a96b5ae30
c50d73557be96907f88b59cfbd1ab1b2fd696d41JournalofElectronicImaging13(3),474–485(July2004). +
Semiconductor sidewall shape estimation +
Oak Ridge National Laboratory +
Oak Ridge, Tennessee 37831-6010 +
c54f9f33382f9f656ec0e97d3004df614ec56434
c574c72b5ef1759b7fd41cf19a9dcd67e5473739Zlatintsi et al. EURASIP Journal on Image and Video Processing (2017) 2017:54 +
DOI 10.1186/s13640-017-0194-1 +
EURASIP Journal on Image +
and Video Processing +
RESEARCH +
Open Access +
COGNIMUSE: a multimodal video +
database annotated with saliency, events, +
semantics and emotion with application to +
summarization +
c5a561c662fc2b195ff80d2655cc5a13a44ffd2dUsing Language to Learn Structured Appearance +
Models for Image Annotation +
c5fe40875358a286594b77fa23285fcfb7bda68e
c5be0feacec2860982fbbb4404cf98c654142489Semi-Qualitative Probabilistic Networks in Computer +
Vision Problems +
Troy, NY 12180, USA. +
Troy, NY 12180, USA. +
Troy, NY 12180, USA. +
Troy, NY 12180, USA. +
Received: *** +
Revised: *** +
c5844de3fdf5e0069d08e235514863c8ef900eb7Lam S K et al. / (IJCSE) International Journal on Computer Science and Engineering +
Vol. 02, No. 08, 2010, 2659-2665 +
A Study on Similarity Computations in Template +
Matching Technique for Identity Verification +
Lam, S. K., Yeong, C. Y., Yew, C. T., Chai, W. S., Suandi, S. A. +
Intelligent Biometric Group, School of Electrical and Electronic Engineering +
Engineering Campus, Universiti Sains Malaysia +
14300 Nibong Tebal, Pulau Pinang, MALAYSIA +
c220f457ad0b28886f8b3ef41f012dd0236cd91aJOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
Crystal Loss and Quality Pooling for +
Unconstrained Face Verification and Recognition +
c254b4c0f6d5a5a45680eb3742907ec93c3a222bA Fusion-based Gender Recognition Method +
Using Facial Images +
c28461e266fe0f03c0f9a9525a266aa3050229f0Automatic Detection of Facial Feature Points via +
HOGs and Geometric Prior Models +
1 Computer Vision Center , Universitat Aut`onoma de Barcelona +
2 Universitat Oberta de Catalunya +
3 Dept. de Matem`atica Aplicada i An`alisi +
Universitat de Barcelona +
c29e33fbd078d9a8ab7adbc74b03d4f830714cd0
f68ed499e9d41f9c3d16d843db75dc12833d988d
f6ca29516cce3fa346673a2aec550d8e671929a6International Journal of Engineering and Advanced Technology (IJEAT) +
ISSN: 2249 – 8958, Volume-2, Issue-4, April 2013 +
Algorithm for Face Matching Using Normalized +
Cross-Correlation +
 +
f6c70635241968a6d5fd5e03cde6907022091d64
f6ce34d6e4e445cc2c8a9b8ba624e971dd4144caCross-label Suppression: A Discriminative and Fast +
Dictionary Learning with Group Regularization +
April 24, 2017 +
f6abecc1f48f6ec6eede4143af33cc936f14d0d0
f6fa97fbfa07691bc9ff28caf93d0998a767a5c1k2-means for fast and accurate large scale clustering +
Computer Vision Lab +
D-ITET +
ETH Zurich +
Computer Vision Lab +
D-ITET +
ETH Zurich +
ESAT, KU Leuven +
D-ITET, ETH Zurich +
e9ed17fd8bf1f3d343198e206a4a7e0561ad7e66International Journal of Enhanced Research in Science Technology & Engineering, ISSN: 2319-7463 +
Vol. 3 Issue 1, January-2014, pp: (362-365), Impact Factor: 1.252, Available online at: www.erpublications.com +
Cognitive Learning for Social Robot through +
Facial Expression from Video Input +
1Department of Automation & Robotics, 2Department of Computer Science & Engg. +
e988be047b28ba3b2f1e4cdba3e8c94026139fcfMulti-Task Convolutional Neural Network for +
Pose-Invariant Face Recognition +
e9d43231a403b4409633594fa6ccc518f035a135Deformable Part Models with CNN Features +
Kokkinos1,2 +
1 Ecole Centrale Paris,2 INRIA, 3TTI-Chicago (cid:63) +
e9fcd15bcb0f65565138dda292e0c71ef25ea8bbRepositorio Institucional de la Universidad Autónoma de Madrid +
https://repositorio.uam.es +
Esta es la versión de autor de la comunicación de congreso publicada en: +
This is an author produced version of a paper published in: +
Highlights on Practical Applications of Agents and Multi-Agent Systems: +
International Workshops of PAAMS. Communications in Computer and +
Information Science, Volumen 365. Springer, 2013. 223-230 +
DOI: http://dx.doi.org/10.1007/978-3-642-38061-7_22 +
Copyright: © 2013 Springer-Verlag +
El acceso a la versión del editor puede requerir la suscripción del recurso +
Access to the published version may require subscription +
e9363f4368b04aeaa6d6617db0a574844fc59338BENCHIP: Benchmarking Intelligence +
Processors +
1ICT CAS,2Cambricon,3Alibaba Infrastructure Service, Alibaba Group +
4IFLYTEK,5JD,6RDA Microelectronics,7AMD +
f16a605abb5857c39a10709bd9f9d14cdaa7918fFast greyscale road sign model matching +
and recognition +
Centre de Visió per Computador +
Edifici O – Campus UAB, 08193 Bellaterra, Barcelona, Catalonia, Spain +
f1748303cc02424704b3a35595610890229567f9
f19ab817dd1ef64ee94e94689b0daae0f686e849TECHNISCHE UNIVERSIT¨AT M ¨UNCHEN +
Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +
Blickrichtungsunabh¨angige Erkennung von +
Personen in Bild- und Tiefendaten +
Andre St¨ormer +
Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +
der Technischen Universit¨at M¨unchen zur Erlangung des akademischen Grades eines +
Doktor-Ingenieurs (Dr.-Ing.) +
genehmigten Dissertation. +
Vorsitzender: +
Univ.-Prof. Dr.-Ing. Thomas Eibert +
Pr¨ufer der Dissertation: +
1. Univ.-Prof. Dr.-Ing. habil. Gerhard Rigoll +
2. Univ.-Prof. Dr.-Ing. Horst-Michael Groß, +
Technische Universit¨at Ilmenau +
Die Dissertation wurde am 16.06.2009 bei der Technischen Universit¨at M¨unchen einge- +
reicht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik am 30.10.2009 +
angenommen. +
e76798bddd0f12ae03de26b7c7743c008d505215
e726acda15d41b992b5a41feabd43617fab6dc23
e7b6887cd06d0c1aa4902335f7893d7640aef823Modelling of Facial Aging and Kinship: A Survey +
cb004e9706f12d1de83b88c209ac948b137caae0Face Aging Effect Simulation using Hidden Factor +
Analysis Joint Sparse Representation +
cb9092fe74ea6a5b2bb56e9226f1c88f96094388
cb08f679f2cb29c7aa972d66fe9e9996c8dfae00JOURNAL OF LATEX CLASS FILES, VOL. 13, NO. 9, SEPTEMBER 2014 +
Action Understanding +
with Multiple Classes of Actors +
cb84229e005645e8623a866d3d7956c197f85e11IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. X, NO. X, MONTH 201X +
Disambiguating Visual Verbs +
cbe859d151466315a050a6925d54a8d3dbad591fGAZE SHIFTS AS DYNAMICAL RANDOM SAMPLING +
Dipartimento di Scienze dell’Informazione +
Universit´a di Milano +
Via Comelico 39/41 +
20135 Milano, Italy +
f8c94afd478821681a1565d463fc305337b02779 +
www.semargroup.org, +
www.ijsetr.com +
+
ISSN 2319-8885 +
Vol.03,Issue.25 +
September-2014, +
Pages:5079-5085 +
Design and Implementation of Robust Face Recognition System for +
Uncontrolled Pose and Illumination Changes +
2 +
f8ec92f6d009b588ddfbb47a518dd5e73855547dJ Inf Process Syst, Vol.10, No.3, pp.443~458, September 2014 +
+
ISSN 1976-913X (Print) +
ISSN 2092-805X (Electronic) +
Extreme Learning Machine Ensemble Using +
Bagging for Facial Expression Recognition +
f869601ae682e6116daebefb77d92e7c5dd2cb15
f8ed5f2c71e1a647a82677df24e70cc46d2f12a8International Journal of Scientific & Engineering Research, Volume 2, Issue 12, December-2011 1 +
ISSN 2229-5518 +
Artificial Neural Network Design and Parameter +
Optimization for Facial Expressions Recognition +
cef841f27535c0865278ee9a4bc8ee113b4fb9f3
ce85d953086294d989c09ae5c41af795d098d5b2This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. +
Bilinear Analysis for Kernel Selection and +
Nonlinear Feature Extraction +
ce691a37060944c136d2795e10ed7ba751cd8394
ce3f3088d0c0bf236638014a299a28e492069753
ce9a61bcba6decba72f91497085807bface02dafEigen-Harmonics Faces: Face Recognition under Generic Lighting +
1Graduate School, CAS, Beijing, China, 100080 +
2ICT-ISVISION Joint R&D Laboratory for Face Recognition, CAS, Beijing, China, 100080 +
Emails: {lyqing, sgshan, wgao}jdl.ac.cn +
cef6cffd7ad15e7fa5632269ef154d32eaf057afEmotion Detection Through Facial Feature +
Recognition +
through consistent +
cebfafea92ed51b74a8d27c730efdacd65572c40JANUARY 2006 +
31 +
Matching 2.5D Face Scans to 3D Models +
ce54e891e956d5b502a834ad131616786897dc91International Journal of Science and Research (IJSR) +
ISSN (Online): 2319-7064 +
Index Copernicus Value (2013): 6.14 | Impact Factor (2014): 5.611 +
Face Recognition Using LTP Algorithm +
1ECE & KUK +
2Assistant Professor (ECE) +
Volume 4 Issue 12, December 2015 +
Licensed Under Creative Commons Attribution CC BY +
www.ijsr.net +
 Variation in luminance: Third main challenge that +
appears in face recognition process is the luminance. Due +
to variation in the luminance the representation get varied +
from the original image. The person with same poses +
expression and seen from same viewpoint can be appear +
very different due to variation in lightening. +
e0dedb6fc4d370f4399bf7d67e234dc44deb4333Supplementary Material: Multi-Task Video Captioning with Video and +
Entailment Generation +
UNC Chapel Hill +
1 Experimental Setup +
1.1 Datasets +
1.1.1 Video Captioning Datasets +
YouTube2Text or MSVD The Microsoft Re- +
search Video Description Corpus (MSVD) or +
YouTube2Text (Chen and Dolan, 2011) is used +
for our primary video captioning experiments. It +
has 1970 YouTube videos in the wild with many +
diverse captions in multiple languages for each +
video. Caption annotations to these videos are +
collected using Amazon Mechanical Turk (AMT). +
All our experiments use only English captions. On +
average, each video has 40 captions, and the over- +
all dataset has about 80, 000 unique video-caption +
pairs. The average clip duration is roughly 10 sec- +
onds. We used the standard split as stated in Venu- +
gopalan et al. (2015), i.e., 1200 videos for training, +
100 videos for validation, and 670 for testing. +
MSR-VTT MSR-VTT is a recent collection of +
10, 000 video clips of 41.2 hours duration (i.e., +
average duration of 15 seconds), which are an- +
notated by AMT workers. It has 200, 000 video +
clip-sentence pairs covering diverse content from +
a commercial video search engine. On average, +
each clip is annotated with 20 natural language +
captions. We used the standard split as provided +
in (Xu et al., 2016), i.e., 6, 513 video clips for +
training, 497 for validation, and 2, 990 for testing. +
M-VAD M-VAD is a movie description dataset +
with 49, 000 video clips collected from 92 movies, +
with the average clip duration being 6 seconds. +
Alignment of descriptions to video clips is done +
through an automatic procedure using Descrip- +
tive Video Service (DVS) provided for the movies. +
Each video clip description has only 1 or 2 sen- +
tences, making most evaluation metrics (except +
paraphrase-based METEOR) infeasible. Again, +
we used the standard train/val/test split as pro- +
vided in Torabi et al. (2015). +
1.1.2 Video Prediction Dataset +
For our unsupervised video representation learn- +
ing task, we use the UCF-101 action videos +
dataset (Soomro et al., 2012), which contains +
13, 320 video clips of 101 action categories and +
with an average clip length of 7.21 seconds each. +
This dataset suits our video captioning task well +
because both contain short video clips of a sin- +
gle action or few actions, and hence using future +
frame prediction on UCF-101 helps learn more ro- +
bust and context-aware video representations for +
our short clip video captioning task. We use the +
standard split of 9, 500 videos for training (we +
don’t need any validation set in our setup because +
we directly tune on the validation set of the video +
captioning task). +
the +
three +
video +
captioning +
1.2 Pre-trained Visual Frame Features +
For +
datasets +
(Youtube2Text, MSR-VTT, M-VAD) and the +
unsupervised video prediction dataset (UCF-101), +
we fix our sampling rate to 3f ps to bring uni- +
formity in the temporal representation of actions +
across all videos. These sampled frames are then +
converted into features using several state-of-the- +
art pre-trained models on ImageNet (Deng et al., +
2009) – VGGNet +
(Simonyan and Zisserman, +
2015), GoogLeNet (Szegedy et al., 2015; Ioffe +
and Szegedy, 2015), and Inception-v4 (Szegedy +
et al., 2016). For VGGNet, we use its f c7 layer +
features with dimension 4096. For GoogLeNet +
and Inception-v4, we use the layer before the fully +
connected layer with dimensions 1024 and 1536, +
respectively. We follow standard preprocessing +
and convert all the natural language descriptions +
to lower case and tokenize the sentences and +
remove punctuations. +
e096b11b3988441c0995c13742ad188a80f2b461Noname manuscript No. +
(will be inserted by the editor) +
DeepProposals: Hunting Objects and Actions by Cascading +
Deep Convolutional Layers +
Van Gool +
Received: date / Accepted: date +
e0c081a007435e0c64e208e9918ca727e2c1c44e
e00d4e4ba25fff3583b180db078ef962bf7d6824Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 20 March 2017 doi:10.20944/preprints201703.0152.v1 +
Article +
Face Verification with Multi-Task and Multi-Scale +
Features Fusion +
e0939b4518a5ad649ba04194f74f3413c793f28eTechnical Report +
UCAM-CL-TR-636 +
ISSN 1476-2986 +
Number 636 +
Computer Laboratory +
Mind-reading machines: +
automated inference +
of complex mental states +
July 2005 +
15 JJ Thomson Avenue +
Cambridge CB3 0FD +
United Kingdom +
phone +44 1223 763500 +
http://www.cl.cam.ac.uk/ +
e0765de5cabe7e287582532456d7f4815acd74c1
e013c650c7c6b480a1b692bedb663947cd9d260f860 +
Robust Image Analysis With Sparse Representation +
on Quantized Visual Features +
46a4551a6d53a3cd10474ef3945f546f45ef76ee2014 IEEE Intelligent Vehicles Symposium (IV) +
June 8-11, 2014. Dearborn, Michigan, USA +
978-1-4799-3637-3/14/$31.00 ©2014 IEEE +
344 +
4686bdcee01520ed6a769943f112b2471e436208Utsumi et al. IPSJ Transactions on Computer Vision and +
Applications (2017) 9:11 +
DOI 10.1186/s41074-017-0024-5 +
IPSJ Transactions on Computer +
Vision and Applications +
EXPRESS PAPER +
Open Access +
Fast search based on generalized +
similarity measure +
4688787d064e59023a304f7c9af950d192ddd33eInvestigating the Discriminative Power of Keystroke +
Sound +
and Dimitris Metaxas, Member, IEEE +
46e86cdb674440f61b6658ef3e84fea95ea51fb4
464de30d3310123644ab81a1f0adc51598586fd2
466a5add15bb5f91e0cfd29a55f5fb159a7980e5Video Repeat Recognition and Mining by Visual +
Features +
46538b0d841654a0934e4c75ccd659f6c5309b72Signal & Image Processing : An International Journal (SIPIJ) Vol.5, No.1, February 2014 +
A NOVEL APPROACH TO GENERATE FACE +
BIOMETRIC TEMPLATE USING BINARY +
DISCRIMINATING ANALYSIS +
1P.G. Student, Department of Computer Engineering, MCERC, Nashik (M.S.), India. +
2Associate Professor, Department of Computer Engineering, +
MCERC, Nashik (M.S.), India +
46196735a201185db3a6d8f6e473baf05ba7b68f
4682fee7dc045aea7177d7f3bfe344aabf153bd5Tabula Rasa: Model Transfer for +
Object Category Detection +
Department of Engineering Science +
Oxford +
(Presented by Elad Liebman) +
2cbb4a2f8fd2ddac86f8804fd7ffacd830a66b58
2c8743089d9c7df04883405a31b5fbe494f175b4Washington State Convention Center +
Seattle, Washington, May 26-30, 2015 +
978-1-4799-6922-7/15/$31.00 ©2015 IEEE +
3039 +
2c61a9e26557dd0fe824909adeadf22a6a0d86b0
2c93c8da5dfe5c50119949881f90ac5a0a4f39feAdvanced local motion patterns for macro and micro facial +
expression recognition +
B. Allaerta,∗, IM. Bilascoa, C. Djerabaa +
aUniv. Lille, CNRS, Centrale Lille, UMR 9189 - CRIStAL - +
Centre de Recherche en Informatique Signal et Automatique de Lille, F-59000 Lille, France +
2c2786ea6386f2d611fc9dbf209362699b104f83
2c848cc514293414d916c0e5931baf1e8583eabcAn automatic facial expression recognition system +
evaluated by different classifiers +
∗Programa de P´os-Graduac¸˜ao em Mecatrˆonica +
Universidade Federal da Bahia, +
†Department of Electrical Engineering - EESC/USP +
2cdd9e445e7259117b995516025fcfc02fa7eebbTitle +
Temporal Exemplar-based Bayesian Networks for facial +
expression recognition +
Author(s) +
Shang, L; Chan, KP +
Citation +
Proceedings - 7Th International Conference On Machine +
Learning And Applications, Icmla 2008, 2008, p. 16-22 +
Issued Date +
2008 +
URL +
http://hdl.handle.net/10722/61208 +
Rights +
This work is licensed under a Creative Commons Attribution- +
NonCommercial-NoDerivatives 4.0 International License.; +
International Conference on Machine Learning and Applications +
Proceedings. Copyright © IEEE.; ©2008 IEEE. Personal use of +
this material is permitted. However, permission to +
reprint/republish this material for advertising or promotional +
purposes or for creating new collective works for resale or +
redistribution to servers or lists, or to reuse any copyrighted +
component of this work in other works must be obtained from +
the IEEE. +
2c5d1e0719f3ad7f66e1763685ae536806f0c23bAENet: Learning Deep Audio Features for Video +
Analysis +
2c8f24f859bbbc4193d4d83645ef467bcf25adc2845 +
Classification in the Presence of +
Label Noise: a Survey +
2cdde47c27a8ecd391cbb6b2dea64b73282c7491ORDER-AWARE CONVOLUTIONAL POOLING FOR VIDEO BASED ACTION RECOGNITION +
Order-aware Convolutional Pooling for Video Based +
Action Recognition +
2c7c3a74da960cc76c00965bd3e343958464da45
2cf5f2091f9c2d9ab97086756c47cd11522a6ef3MPIIGaze: Real-World Dataset and Deep +
Appearance-Based Gaze Estimation +
79581c364cefe53bff6bdd224acd4f4bbc43d6d4
790aa543151312aef3f7102d64ea699a1d15cb29Confidence-Weighted Local Expression Predictions for +
Occlusion Handling in Expression Recognition and Action +
Unit detection +
1 Sorbonne Universités, UPMC Univ Paris 06, CNRS, ISIR UMR 7222 +
4 place Jussieu 75005 Paris +
79f6a8f777a11fd626185ab549079236629431acCopyright +
by +
2013 +
795ea140df2c3d29753f40ccc4952ef24f46576c
79dc84a3bf76f1cb983902e2591d913cee5bdb0e
79b669abf65c2ca323098cf3f19fa7bdd837ff31 Deakin Research Online +
This is the published version: +
Rana, Santu, Liu, Wanquan, Lazarescu, Mihai and Venkatesh, Svetha 2008, Efficient tensor +
based face recognition, in ICPR 2008 : Proceedings of the 19th International Conference on +
Pattern Recognition, IEEE, Washington, D. C., pp. 1-4. +
Available from Deakin Research Online: +
http://hdl.handle.net/10536/DRO/DU:30044585 +
+
Reproduced with the kind permissions of the copyright owner. +
Personal use of this material is permitted. However, permission to reprint/republish this +
material for advertising or promotional purposes or for creating new collective works for +
resale or redistribution to servers or lists, or to reuse any copyrighted component of this work +
in other works must be obtained from the IEEE. +
Copyright : 2008, IEEE +
79c3a7131c6c176b02b97d368cd0cd0bc713ff7e
79dd787b2877cf9ce08762d702589543bda373beFace Detection Using SURF Cascade +
Intel Labs China +
793e7f1ba18848908da30cbad14323b0389fd2a8
2dd6c988b279d89ab5fb5155baba65ce4ce53c1e
2d294c58b2afb529b26c49d3c92293431f5f98d04413 +
Maximum Margin Projection Subspace Learning +
for Visual Data Analysis +
2d1f86e2c7ba81392c8914edbc079ac64d29b666
2d05e768c64628c034db858b7154c6cbd580b2d5Available Online at www.ijcsmc.com +
International Journal of Computer Science and Mobile Computing +
A Monthly Journal of Computer Science and Information Technology +
IJCSMC, Vol. 4, Issue. 8, August 2015, pg.431 – 446 +
RESEARCH ARTICLE +
ISSN 2320–088X +
FACIAL EXPRESSION RECOGNITION: +
Machine Learning using C# +
2d072cd43de8d17ce3198fae4469c498f97c6277Random Cascaded-Regression Copse for Robust +
Facial Landmark Detection +
and Xiao-Jun Wu +
2d71e0464a55ef2f424017ce91a6bcc6fd83f6c3International Journal of Computer Applications (0975 – 8887) +
National Conference on Advancements in Computer & Information Technology (NCACIT-2016) +
A Survey on: Image Process using Two- Stage Crawler +
Assistant Professor +
SPPU, Pune +
Department of Computer Engg +
Department of Computer Engg +
Department of Computer Engg +
BE Student +
SPPU, Pune +
BE Student +
SPPU, Pune +
BE Student +
Department of Computer Engg +
SPPU, Pune +
additional +
analysis +
for +
information +
2d8d089d368f2982748fde93a959cf5944873673Proceedings of NAACL-HLT 2018, pages 788–794 +
New Orleans, Louisiana, June 1 - 6, 2018. c(cid:13)2018 Association for Computational Linguistics +
788 +
2df4d05119fe3fbf1f8112b3ad901c33728b498aFacial landmark detection using structured output deep +
neural networks +
Soufiane Belharbi ∗1, Cl´ement Chatelain∗1, Romain H´erault∗1, and S´ebastien +
Adam∗2 +
1LITIS EA 4108, INSA de Rouen, Saint ´Etienne du Rouvray 76800, France +
2LITIS EA 4108, UFR des Sciences, Universit´e de Rouen, France. +
September 24, 2015 +
4188bd3ef976ea0dec24a2512b44d7673fd4ad261050 +
Nonlinear Non-Negative Component +
Analysis Algorithms +
41000c3a3344676513ef4bfcd392d14c7a9a7599A NOVEL APPROACH FOR GENERATING FACE +
TEMPLATE USING BDA +
1P.G. Student, Department of Computer Engineering, MCERC, Nashik (M.S.), India. +
2Associate Professor, Department of Computer Engineering, MCERC, Nashik (M.S.), +
India +
414715421e01e8c8b5743c5330e6d2553a08c16dPoTion: Pose MoTion Representation for Action Recognition +
1Inria∗ +
2NAVER LABS Europe +
41ab4939db641fa4d327071ae9bb0df4a612dc89Interpreting Face Images by Fitting a Fast +
Illumination-Based 3D Active Appearance +
Model +
Instituto Nacional de Astrof´ısica, ´Optica y Electr´onica, +
Luis Enrique Erro #1, 72840 Sta Ma. Tonantzintla. Pue., M´exico +
Coordinaci´on de Ciencias Computacionales +
41a6196f88beced105d8bc48dd54d5494cc156fb2015 International Conference on +
Communications, Signal +
Processing, and their Applications +
(ICCSPA 2015) +
Sharjah, United Arab Emirates +
17-19 February 2015 +
IEEE Catalog Number: +
ISBN: +
CFP1574T-POD +
978-1-4799-6533-5 +
41de109bca9343691f1d5720df864cdbeeecd9d0Article +
Facial Emotion Recognition: A Survey and +
Real-World User Experiences in Mixed Reality +
Received: 10 December 2017; Accepted: 26 January 2018; Published: 1 Febuary 2018 +
41d9a240b711ff76c5448d4bf4df840cc5dad5fcJOURNAL DRAFT, VOL. X, NO. X, APR 2013 +
Image Similarity Using Sparse Representation +
and Compression Distance +
419a6fca4c8d73a1e43003edc3f6b610174c41d2A Component Based Approach Improves Classification of Discrete +
Facial Expressions Over a Holistic Approach +
4180978dbcd09162d166f7449136cb0b320adf1fReal-time head pose classification in uncontrolled environments +
with Spatio-Temporal Active Appearance Models +
∗ Matematica Aplicada i Analisi ,Universitat de Barcelona, Barcelona, Spain +
+ Matematica Aplicada i Analisi, Universitat de Barcelona, Barcelona, Spain +
+ Matematica Aplicada i Analisi, Universitat de Barcelona, Barcelona, Spain +
41b997f6cec7a6a773cd09f174cb6d2f036b36cd
413a184b584dc2b669fbe731ace1e48b22945443Human Pose Co-Estimation and Applications +
83ca4cca9b28ae58f461b5a192e08dffdc1c76f3DETECTING EMOTIONAL STRESS FROM FACIAL EXPRESSIONS FOR DRIVING SAFETY +
Signal Processing Laboratory (LTS5), +
´Ecole Polytechnique F´ed´erale de Lausanne, Switzerland +
831fbef657cc5e1bbf298ce6aad6b62f00a5b5d9
832e1d128059dd5ed5fa5a0b0f021a025903f9d5Pairwise Conditional Random Forests for Facial Expression Recognition +
S´everine Dubuisson1 +
1 Sorbonne Universit´es, UPMC Univ Paris 06, CNRS, ISIR UMR 7222, 4 place Jussieu 75005 Paris +
83e093a07efcf795db5e3aa3576531d61557dd0dFacial Landmark Localization using Robust +
Relationship Priors and Approximative Gibbs +
Sampling +
Institut f¨ur Informationsverarbeitung (tnt) +
Leibniz Universit¨at Hannover, Germany +
83b4899d2899dd6a8d956eda3c4b89f27f1cd3081-4244-1437-7/07/$20.00 ©2007 IEEE +
I - 377 +
ICIP 2007 +
830e5b1043227fe189b3f93619ef4c58868758a7
8395cf3535a6628c3bdc9b8d0171568d551f5ff0Entropy Non-increasing Games for the +
Improvement of Dataflow Programming +
Norbert B´atfai, Ren´at´o Besenczi, Gerg˝o Bogacsovics, +
February 16, 2017 +
83ac942d71ba908c8d76fc68de6173151f012b38
834f5ab0cb374b13a6e19198d550e7a32901a4b2Face Translation between Images and Videos using Identity-aware CycleGAN +
†Computer Vision Lab, ETH Zurich, Switzerland +
‡VISICS, KU Leuven, Belgium +
834b15762f97b4da11a2d851840123dbeee51d33Landmark-free smile intensity estimation +
IMAGO Research Group - Universidade Federal do Paran´a +
Fig. 1. Overview of our method for smile intensity estimation +
833f6ab858f26b848f0d747de502127406f06417978-1-4244-5654-3/09/$26.00 ©2009 IEEE +
61 +
ICIP 2009 +
8309e8f27f3fb6f2ac1b4343a4ad7db09fb8f0ffGeneric versus Salient Region-based Partitioning +
for Local Appearance Face Recognition +
Computer Science Depatment, Universit¨at Karlsruhe (TH) +
Am Fasanengarten 5, Karlsruhe 76131, Germany +
http://isl.ira.uka.de/cvhci +
1b55c4e804d1298cbbb9c507497177014a923d22Incremental Class Representation +
Learning for Face Recognition +
Degree’s Thesis +
Audiovisual Systems Engineering +
Author: +
Universitat Politècnica de Catalunya (UPC) +
2016 - 2017 +
1bd50926079e68a6e32dc4412e9d5abe331daefb
1b150248d856f95da8316da868532a4286b9d58eAnalyzing 3D Objects in Cluttered Images +
UC Irvine +
UC Irvine +
1be498d4bbc30c3bfd0029114c784bc2114d67c0Age and Gender Estimation of Unfiltered Faces +
1b300a7858ab7870d36622a51b0549b1936572d4This 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 +
Transactions on Image Processing +
Dynamic Facial Expression Recognition with Atlas +
Construction and Sparse Representation +
1b1173a3fb33f9dfaf8d8cc36eb0bf35e364913dDICTA +
#147 +
000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
045 +
046 +
047 +
048 +
049 +
050 +
051 +
052 +
053 +
DICTA 2010 Submission #147. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +
Registration Invariant Representations for Expression Detection +
Anonymous DICTA submission +
Paper ID 147 +
1b0a071450c419138432c033f722027ec88846eaWindsor Oceanico Hotel, Rio de Janeiro, Brazil, November 1-4, 2016 +
978-1-5090-1889-5/16/$31.00 ©2016 IEEE +
649 +
1b3b01513f99d13973e631c87ffa43904cd8a821HMM RECOGNITION OF EXPRESSIONS IN UNRESTRAINED VIDEO INTERVALS +
Universitat Politècnica de Catalunya, Barcelona, Spain +
1bc214c39536c940b12c3a2a6b78cafcbfddb59a
1be18a701d5af2d8088db3e6aaa5b9b1d54b6fd3ENHANCEMENT OF FAST FACE DETECTION ALGORITHM BASED ON A CASCADE OF +
DECISION TREES +
Commission II, WG II/5 +
KEY WORDS: Face Detection, Cascade Algorithm, Decision Trees. +
1b79628af96eb3ad64dbb859dae64f31a09027d5
1b4f6f73c70353869026e5eec1dd903f9e26d43fRobust Subjective Visual Property Prediction +
from Crowdsourced Pairwise Labels +
1bc23c771688109bed9fd295ce82d7e702726327
1b589016fbabe607a1fb7ce0c265442be9caf3a9
1b27ca161d2e1d4dd7d22b1247acee5c53db5104
7711a7404f1f1ac3a0107203936e6332f50ac30cAction Classification and Highlighting in Videos +
Disney Research Pittsburgh +
Disney Research Pittsburgh +
778c9f88839eb26129427e1b8633caa4bd4d275ePose Pooling Kernels for Sub-category Recognition +
ICSI & UC Berkeley +
ICSI & UC Berkeley +
Trever Darrell +
ICSI & UC Berkeley +
7789a5d87884f8bafec8a82085292e87d4e2866fA Unified Tensor-based Active Appearance Face +
Model +
Member, IEEE +
776835eb176ed4655d6e6c308ab203126194c41e
778bff335ae1b77fd7ec67404f71a1446624331bHough Forest-based Facial Expression Recognition from +
Video Sequences +
BIWI, ETH Zurich http://www.vision.ee.ethz.ch +
VISICS, K.U. Leuven http://www.esat.kuleuven.be/psi/visics +
7726a6ab26a1654d34ec04c0b7b3dd80c5f84e0dCONTENT-AWARE COMPRESSION USING SALIENCY-DRIVEN IMAGE RETARGETING +
*Disney Research Zurich +
†ETH Zurich +
7754b708d6258fb8279aa5667ce805e9f925dfd0Facial Action Unit Recognition by Exploiting +
Their Dynamic and Semantic Relationships +
77db171a523fc3d08c91cea94c9562f3edce56e1Poursaberi et al. EURASIP Journal on Image and Video Processing 2012, 2012:17 +
http://jivp.eurasipjournals.com/content/2012/1/17 +
R ES EAR CH +
Open Access +
Gauss–Laguerre wavelet textural feature fusion +
with geometrical information for facial expression +
identification +
77037a22c9b8169930d74d2ce6f50f1a999c1221Robust Face Recognition With Kernelized +
Locality-Sensitive Group Sparsity Representation +
77d31d2ec25df44781d999d6ff980183093fb3deThe Multiverse Loss for Robust Transfer Learning +
Supplementary +
1. Omitted proofs +
for which the joint loss: +
m(cid:88) +
r=1 +
L(F r, br, D, y) +
(2) +
J(F 1, b1...F m, bm, D, y) = +
is bounded by: +
mL∗(D, y) ≤ J(F 1, b1...F m, bm, D, y) +
m−1(cid:88) +
≤ mL∗(D, y) + +
Alλd−j+1 +
(3) +
l=1 +
where [A1 . . . Am−1] are bounded parameters. +
We provide proofs that were omitted from the paper for +
lack of space. We follow the same theorem numbering as in +
the paper. +
Lemma 1. The minimizers F ∗, b∗ of L are not unique, and +
it holds that for any vector v ∈ Rc and scalar s, the solu- +
tions F ∗ + v1(cid:62) +
Proof. denoting V = v1(cid:62) +
c , b∗ + s1c are also minimizers of L. +
c , s = s1c, +
i v+byi +s +
i v+bj +s +
i fyi +byi +
i v+sed(cid:62) +
i fj +bj +
i=1 +
log( +
L(F ∗ + V, b∗ + s, D, y) = +
i fyi +d(cid:62) +
ed(cid:62) +
i fj +d(cid:62) +
j=1 ed(cid:62) +
i v+sed(cid:62) +
ed(cid:62) +
j=1 ed(cid:62) +
i v+sed(cid:62) +
ed(cid:62) +
(cid:80)c +
(cid:80)c +
i v+s(cid:80)c +
− n(cid:88) +
= − n(cid:88) +
= − n(cid:88) +
(cid:80)c +
= − n(cid:88) +
ed(cid:62) +
i fyi +byi +
j=1 ed(cid:62) +
i fj +bj +
ed(cid:62) +
log( +
log( +
log( +
i=1 +
i=1 +
i=1 +
i fj +bj +
i fyi +byi +
j=1 ed(cid:62) +
) = L(F ∗, b∗, D, y) +
The following simple lemma was not part of the paper. +
However, it is the reasoning behind the statement at the end +
of the proof of Thm. 1. “Since ∀i, j pi(j) > 0 and since +
rank(D) is full,(cid:80)n +
Lemma 2. Let K =(cid:80)n +
such that ∀i qi > 0, the matrix ˆK =(cid:80)n +
i be a full rank d×d matrix, +
i.e., it is PD and not just PSD, then for all vector q ∈ Rn +
is also +
i pi(j)pi(j(cid:48)) is PD.” +
i=1 did(cid:62) +
i=1 did(cid:62) +
i=1 qidid(cid:62) +
full rank. +
Proof. For +
(miniqi)v(cid:62)Kv > 0. +
every vector v +
(cid:2)f 1 +
(cid:3) , b1, F 2 = (cid:2)f 2 +
Theorem 3. There exist a set of weights F 1 = +
j ⊥ f s +
C ] , bm which are orthogonal ∀jrs f r +
2 , ..., f 1 +
2 , ..., f m +
1 , f 1 +
1 , f m +
2 , ..., f 2 +
1 , f 2 +
[f m +
(cid:3) , b2...F m = +
Proof. We again prove the theorem by constructing such a +
solution. Denoting by vd−m+2...vd the eigenvectors of K +
corresponding to λd−m+2 . . . λd. Given F 1 = F ∗, b1 = b∗, +
we can construct each pair F r, br as follows: +
(1) +
∀j, r +
fj +
r = f1 +
1 + +
m−1(cid:88) +
l=1 +
αjlrvd−l+1 +
br = b1 +
(4) +
The tensor of parameters αjlr is constructed to insure the +
orthogonality condition. Formally, αjlr has to satisfy: +
Rd, +
v(cid:62) ˆKv +
∀j, r (cid:54)= s +
(f 1 +
j + +
m−1(cid:88) +
l=1 +
αjlrvd−l+1)(cid:62)f s +
j = 0 +
(5) +
2 m(m− 1) equations, it +
Noticing that 5 constitutes a set of 1 +
can be satisfied by the tensor αjlr which contains m(m − +
c ] = F r − +
1)c parameters. Defining Ψr = [ψr +
1, ψr +
2, . . . , ψr +
486840f4f524e97f692a7f6b42cd19019ee71533DeepVisage: Making face recognition simple yet with powerful generalization +
skills +
1Laboratoire LIRIS, ´Ecole centrale de Lyon, 69134 Ecully, France. +
2Safran Identity & Security, 92130 Issy-les-Moulineaux, France. +
48186494fc7c0cc664edec16ce582b3fcb5249c0P-CNN: Pose-based CNN Features for Action Recognition +
Guilhem Ch´eron∗ † +
INRIA +
48499deeaa1e31ac22c901d115b8b9867f89f952Interim Report of Final Year Project +
HKU-Face: A Large Scale Dataset for +
Deep Face Recognition +
3035140108 +
Haoyu Li +
3035141841 +
COMP4801 Final Year Project +
Project Code: 17007 +
486a82f50835ea888fbc5c6babf3cf8e8b9807bcMSU TECHNICAL REPORT MSU-CSE-15-11, JULY 24, 2015 +
Face Search at Scale: 80 Million Gallery +
4866a5d6d7a40a26f038fc743e16345c064e9842
487df616e981557c8e1201829a1d0ec1ecb7d275Acoustic Echo Cancellation Using a Vector-Space-Based +
Adaptive Filtering Algorithm +
48f211a9764f2bf6d6dda4a467008eda5680837a
4858d014bb5119a199448fcd36746c413e60f295
48cfc5789c246c6ad88ff841701204fc9d6577edJ Inf Process Syst, Vol.12, No.3, pp.392~409, September 2016 +
+
+
ISSN 1976-913X (Print) +
ISSN 2092-805X (Electronic) +
Age Invariant Face Recognition Based on DCT +
Feature Extraction and Kernel Fisher Analysis +
70f189798c8b9f2b31c8b5566a5cf3107050b349The Challenge of Face Recognition from Digital Point-and-Shoot Cameras +
David Bolme‡ +
70109c670471db2e0ede3842cbb58ba6be804561Noname manuscript No. +
(will be inserted by the editor) +
Zero-Shot Visual Recognition via Bidirectional Latent Embedding +
Received: date / Accepted: date +
703890b7a50d6535900a5883e8d2a6813ead3a03
706236308e1c8d8b8ba7749869c6b9c25fa9f957Crowdsourced Data Collection of Facial Responses +
MIT Media Lab +
Cambridge +
02139, USA +
Rosalind Picard +
MIT Media Lab +
Cambridge +
02139, USA +
MIT Media Lab +
Cambridge +
02139, USA +
70569810e46f476515fce80a602a210f8d9a2b95Apparent Age Estimation from Face Images Combining General and +
Children-Specialized Deep Learning Models +
1Orange Labs – France Telecom, 4 rue Clos Courtel, 35512 Cesson-S´evign´e, France +
2Eurecom, 450 route des Chappes, 06410 Biot, France +
70e79d7b64f5540d309465620b0dab19d9520df1International Journal of Scientific & Engineering Research, Volume 8, Issue 3, March-2017 +
ISSN 2229-5518 +
Facial Expression Recognition System +
Using Extreme Learning Machine +
7003d903d5e88351d649b90d378f3fc5f211282bInternational Journal of Computer Applications (0975 – 8887) +
Volume 68– No.23, April 2013 +
Facial Expression Recognition using Gabor Wavelet +
ENTC SVERI’S COE (Poly), +
Pandharpur, +
Solapur, India +
ENTC SVERI’S COE, +
Pandharpur, +
Solapur, India +
ENTC SVERI’S COE (Poly), +
Pandharpur, +
Solapur, India +
70bf1769d2d5737fc82de72c24adbb7882d2effdFace detection in intelligent ambiences with colored illumination +
Department of Intelligent Systems +
TU Delft +
Delft, The Netherlands +
1e799047e294267087ec1e2c385fac67074ee5c8IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 21, NO. 12, DECEMBER 1999 +
1357 +
Short Papers___________________________________________________________________________________________________ +
Automatic Classification of +
Single Facial Images +
1ef4815f41fa3a9217a8a8af12cc385f6ed137e1Rendering of Eyes for Eye-Shape Registration and Gaze Estimation +
1e7ae86a78a9b4860aa720fb0fd0bdc199b092c3Article +
A Brief Review of Facial Emotion Recognition Based +
on Visual Information +
Byoung Chul Ko ID +
Tel.: +82-10-3559-4564 +
Received: 6 December 2017; Accepted: 25 January 2018; Published: 30 January 2018 +
1e8eee51fd3bf7a9570d6ee6aa9a09454254689dThis 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 +
Transactions on Pattern Analysis and Machine Intelligence +
Face Search at Scale +
1ea8085fe1c79d12adffb02bd157b54d799568e4
1ebdfceebad642299e573a8995bc5ed1fad173e3
1eec03527703114d15e98ef9e55bee5d6eeba736UNIVERSITÄT KARLSRUHE (TH) +
FAKULTÄT FÜR INFORMATIK +
INTERACTIVE SYSTEMS LABS +
DIPLOMA THESIS +
Automatic identification +
of persons in TV series +
SUBMITTED BY +
MAY 2008 +
ADVISORS +
1e8394cc9fe7c2392aa36fb4878faf7e78bbf2deTO APPEAR IN IEEE THMS +
Zero-Shot Object Recognition System +
based on Topic Model +
1ef4aac0ebc34e76123f848c256840d89ff728d0
1ecb56e7c06a380b3ce582af3a629f6ef0104457List of Contents Vol.8 +
Contents of +
Journal of Advanced Computational +
Intelligence and Intelligent Informatics +
Volume 8 +
Vol.8 No.1, January 2004 +
Editorial: +
o Special Issue on Selected Papers from Humanoid, +
Papers: +
o Dynamic Color Object Recognition Using Fuzzy +
Nano-technology, Information Technology, +
Communication and Control, Environment, and +
Management (HNICEM’03). +
. 1 +
Elmer P. Dadios +
Papers: +
o A New Way of Discovery of Belief, Desire and +
Intention in the BDI Agent-Based Software +
Modeling . +
. 2 +
o Integration of Distributed Robotic Systems +
. 7 +
Fakhri Karray, Rogelio Soto, Federico Guedea, +
and Insop Song +
o A Searching and Tracking Framework for +
Multi-Robot Observation of Multiple Moving +
Targets . +
. 14 +
Zheng Liu, Marcelo H. Ang Jr., and Winston +
Khoon Guan Seah +
Development Paper: +
o Possibilistic Uncertainty Propagation and +
Compromise Programming in the Life Cycle +
Analysis of Alternative Motor Vehicle Fuels +
Raymond R. Tan, Alvin B. Culaba, and +
Michael R. I. Purvis +
. 23 +
Logic . +
Napoleon H. Reyes, and Elmer P. Dadios +
. 29 +
o A Optical Coordinate Measuring Machine for +
Nanoscale Dimensional Metrology . +
. 39 +
Eric Kirkland, Thomas R. Kurfess, and Steven +
Y. Liang +
o Humanoid Robot HanSaRam: Recent Progress +
and Developments . +
. 45 +
Jong-Hwan Kim, Dong-Han Kim, Yong-Jae +
Kim, Kui-Hong Park, Jae-Ho Park, +
Choon-Kyoung Moon, Jee-Hwan Ryu, Kiam +
Tian Seow, and Kyoung-Chul Koh +
o Generalized Associative Memory Models: Their +
Memory Capacities and Potential Application +
. 56 +
Teddy N. Yap, Jr., and Arnulfo P. Azcarraga +
o Hybrid Fuzzy Logic Strategy for Soccer Robot +
Game. +
. 65 +
Elmer A. Maravillas , Napoleon H. Reyes, and +
Elmer P. Dadios +
o Image Compression and Reconstruction Based on +
Fuzzy Relation and Soft Computing +
Technology . +
. 72 +
Kaoru Hirota, Hajime Nobuhara, Kazuhiko +
Kawamoto, and Shin’ichi Yoshida +
Vol.8 No.2, March 2004 +
Editorial: +
o Special Issue on Pattern Recognition . +
. 83 +
Papers: +
o Operation of Spatiotemporal Patterns Stored in +
Osamu Hasegawa +
Review: +
o Support Vector Machine and Generalization . 84 +
Takio Kurita +
o Bayesian Network: Probabilistic Reasoning, +
Statistical Learning, and Applications . +
. 93 +
Yoichi Motomura +
Living Neuronal Networks Cultured on a +
Microelectrode Array . +
Suguru N. Kudoh, and Takahisa Taguchi +
o Rapid Discriminative Learning . +
. 100 +
. 108 +
Jun Rokui +
o Robust Fuzzy Clustering Based on Similarity +
between Data . +
Kohei Inoue, and Kiichi Urahama +
Vol.8 No.6, 2004 +
Journal of Advanced Computational Intelligence +
and Intelligent Informatics +
. 115 +
I-1 +
1e64b2d2f0a8a608d0d9d913c4baee6973995952DOMINANT AND +
COMPLEMENTARY MULTI- +
EMOTIONAL FACIAL +
EXPRESSION RECOGNITION +
USING C-SUPPORT VECTOR +
CLASSIFICATION +
1e21b925b65303ef0299af65e018ec1e1b9b8d60Under review as a conference paper at ICLR 2017 +
UNSUPERVISED CROSS-DOMAIN IMAGE GENERATION +
Facebook AI Research +
Tel-Aviv, Israel +
1ee27c66fabde8ffe90bd2f4ccee5835f8dedbb9Entropy Regularization +
The problem of semi-supervised induction consists in learning a decision rule from +
labeled and unlabeled data. This task can be undertaken by discriminative methods, +
provided that learning criteria are adapted consequently. In this chapter, we moti- +
vate the use of entropy regularization as a means to bene(cid:12)t from unlabeled data in +
the framework of maximum a posteriori estimation. The learning criterion is derived +
from clearly stated assumptions and can be applied to any smoothly parametrized +
model of posterior probabilities. The regularization scheme favors low density sep- +
aration, without any modeling of the density of input features. The contribution +
of unlabeled data to the learning criterion induces local optima, but this problem +
can be alleviated by deterministic annealing. For well-behaved models of posterior +
probabilities, deterministic annealing EM provides a decomposition of the learning +
problem in a series of concave subproblems. Other approaches to the semi-supervised +
problem are shown to be close relatives or limiting cases of entropy regularization. +
A series of experiments illustrates the good behavior of the algorithm in terms of +
performance and robustness with respect to the violation of the postulated low den- +
sity separation assumption. The minimum entropy solution bene(cid:12)ts from unlabeled +
data and is able to challenge mixture models and manifold learning in a number of +
situations. +
9.1 Introduction +
semi-supervised +
induction +
This chapter addresses semi-supervised induction, which refers to the learning of +
a decision rule, on the entire input domain X, from labeled and unlabeled data. +
The objective is identical to the one of supervised classi(cid:12)cation: generalize from +
examples. The problem di(cid:11)ers in the respect that the supervisor’s responses are +
missing for some training examples. This characteristic is shared with transduction, +
which has however a di(cid:11)erent goal, that is, of predicting labels on a set of prede(cid:12)ned +
1ee3b4ba04e54bfbacba94d54bf8d05fd202931dIndonesian Journal of Electrical Engineering and Computer Science +
Vol. 12, No. 2, November 2018, pp. 476~481 +
ISSN: 2502-4752, DOI: 10.11591/ijeecs.v12.i2.pp476-481 +
 476 +
Celebrity Face Recognition using Deep Learning +
1,2,3Faculty of Computer and Mathematical Sciences, UniversitiTeknologi MARA (UiTM), +
4Faculty of Computer and Mathematical Sciences, UniversitiTeknologi MARA (UiTM), +
Shah Alam, Selangor, Malaysia +
Campus Jasin, Melaka, Malaysia +
Article Info +
Article history: +
Received May 29, 2018 +
Revised Jul 30, 2018 +
Accepted Aug 3, 2018 +
Keywords: +
AlexNet +
Convolutional neural network +
Deep learning +
Face recognition +
GoogLeNet +
1e41a3fdaac9f306c0ef0a978ae050d884d77d2a411 +
Robust Object Recognition with +
Cortex-Like Mechanisms +
Tomaso Poggio, Member, IEEE +
1e1e66783f51a206509b0a427e68b3f6e40a27c8SEMI-SUPERVISED ESTIMATION OF PERCEIVED AGE +
FROM FACE IMAGES +
VALWAY Technology Center, NEC Soft, Ltd., Tokyo, Japan +
Keywords: +
1efaa128378f988965841eb3f49d1319a102dc36JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 +
Hierarchical binary CNNs for landmark +
localization with limited resources +
8451bf3dd6bcd946be14b1a75af8bbb65a42d4b2Consensual and Privacy-Preserving Sharing of +
Multi-Subject and Interdependent Data +
EPFL, UNIL–HEC Lausanne +
K´evin Huguenin +
UNIL–HEC Lausanne +
EPFL +
EPFL +
84fe5b4ac805af63206012d29523a1e033bc827e
84e4b7469f9c4b6c9e73733fa28788730fd30379Duong et al. EURASIP Journal on Advances in Signal Processing (2018) 2018:10 +
DOI 10.1186/s13634-017-0521-9 +
EURASIP Journal on Advances +
in Signal Processing +
R ES EAR CH +
Projective complex matrix factorization for +
facial expression recognition +
Open Access +
84dcf04802743d9907b5b3ae28b19cbbacd97981
84fa126cb19d569d2f0147bf6f9e26b54c9ad4f1Improved Boosting Performance by Explicit +
Handling of Ambiguous Positive Examples +
841a5de1d71a0b51957d9be9d9bebed33fb5d9fa5017 +
PCANet: A Simple Deep Learning Baseline for +
Image Classification? +
849f891973ad2b6c6f70d7d43d9ac5805f1a1a5bDetecting Faces Using Region-based Fully +
Convolutional Networks +
Tencent AI Lab, China +
4adca62f888226d3a16654ca499bf2a7d3d11b71Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics, pages 572–582, +
Sofia, Bulgaria, August 4-9 2013. c(cid:13)2013 Association for Computational Linguistics +
572 +
4a2d54ea1da851151d43b38652b7ea30cdb6dfb2Direct Recognition of Motion Blurred Faces +
4a3758f283b7c484d3f164528d73bc8667eb1591Attribute Enhanced Face Aging with Wavelet-based Generative Adversarial +
Networks +
Center for Research on Intelligent Perception and Computing, CASIA +
National Laboratory of Pattern Recognition, CASIA +
4abd49538d04ea5c7e6d31701b57ea17bc349412Recognizing Fine-Grained and Composite Activities +
using Hand-Centric Features and Script Data +
4a0f98d7dbc31497106d4f652968c708f7da6692Real-time Eye Gaze Direction Classification Using +
Convolutional Neural Network +
4acd683b5f91589002e6f50885df51f48bc985f4BRIDGING COMPUTER VISION AND SOCIAL SCIENCE : A MULTI-CAMERA VISION +
SYSTEM FOR SOCIAL INTERACTION TRAINING ANALYSIS +
Peter Tu +
GE Global Research, Niskayuna NY USA +
4aeb87c11fb3a8ad603311c4650040fd3c088832Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence (IJCAI-17) +
1816 +
4a3d96b2a53114da4be3880f652a6eef3f3cc0352666 +
A Dictionary Learning-Based +
3D Morphable Shape Model +
4a6fcf714f663618657effc341ae5961784504c7Scaling up Class-Specific Kernel Discriminant +
Analysis for large-scale Face Verification +
24115d209e0733e319e39badc5411bbfd82c5133Long-term Recurrent Convolutional Networks for +
Visual Recognition and Description +
24c442ac3f6802296d71b1a1914b5d44e48b4f29Pose and expression-coherent face recovery in the wild +
Technicolor, Cesson-S´evign´e, France +
Franc¸ois Le Clerc +
Patrick P´erez +
24aac045f1e1a4c13a58eab4c7618dccd4c0e671
240d5390af19bb43761f112b0209771f19bfb696
24e099e77ae7bae3df2bebdc0ee4e00acca71250Robust face alignment under occlusion via regional predictive power +
estimation. +
© 2015 IEEE +
For additional information about this publication click this link. +
http://qmro.qmul.ac.uk/xmlui/handle/123456789/22467 +
Information about this research object was correct at the time of download; we occasionally +
make corrections to records, please therefore check the published record when citing. For +
2450c618cca4cbd9b8cdbdb05bb57d67e63069b1A Connexionist Approach for Robust and Precise Facial Feature Detection in +
Complex Scenes +
Stefan Duffner and Christophe Garcia +
France Telecom Research & Development +
4, rue du Clos Courtel +
35512 Cesson-S´evign´e, France +
244b57cc4a00076efd5f913cc2833138087e1258Warped Convolutions: Efficient Invariance to Spatial Transformations +
24869258fef8f47623b5ef43bd978a525f0af60eUNIVERSITÉDEGRENOBLENoattribuéparlabibliothèqueTHÈSEpourobtenirlegradedeDOCTEURDEL’UNIVERSITÉDEGRENOBLESpécialité:MathématiquesetInformatiquepréparéeauLaboratoireJeanKuntzmanndanslecadredel’ÉcoleDoctoraleMathématiques,SciencesetTechnologiesdel’Information,InformatiqueprésentéeetsoutenuepubliquementparMatthieuGuillauminle27septembre2010ExploitingMultimodalDataforImageUnderstandingDonnéesmultimodalespourl’analysed’imageDirecteursdethèse:CordeliaSchmidetJakobVerbeekJURYM.ÉricGaussierUniversitéJosephFourierPrésidentM.AntonioTorralbaMassachusettsInstituteofTechnologyRapporteurMmeTinneTuytelaarsKatholiekeUniversiteitLeuvenRapporteurM.MarkEveringhamUniversityofLeedsExaminateurMmeCordeliaSchmidINRIAGrenobleExaminatriceM.JakobVerbeekINRIAGrenobleExaminateur
24d376e4d580fb28fd66bc5e7681f1a8db3b6b78
24ff832171cb774087a614152c21f54589bf7523Beat-Event Detection in Action Movie Franchises +
Jerome Revaud +
Zaid Harchaoui +
24bf94f8090daf9bda56d54e42009067839b20df
230527d37421c28b7387c54e203deda64564e1b7Person Re-identification: System Design and +
Evaluation Overview +
23fdbef123bcda0f07d940c72f3b15704fd49a98
23ebbbba11c6ca785b0589543bf5675883283a57
23172f9a397f13ae1ecb5793efd81b6aba9b4537Proceedings of the 2015 Workshop on Vision and Language (VL’15), pages 10–17, +
Lisbon, Portugal, 18 September 2015. c(cid:13)2015 Association for Computational Linguistics. +
10 +
236a4f38f79a4dcc2183e99b568f472cf45d27f41632 +
Randomized Clustering Forests +
for Image Classification +
Frederic Jurie, Member, IEEE Computer Society +
230c4a30f439700355b268e5f57d15851bcbf41fEM Algorithms for Weighted-Data Clustering +
with Application to Audio-Visual Scene Analysis +
237fa91c8e8098a0d44f32ce259ff0487aec02cfIEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 36, NO. 4, AUGUST 2006 +
863 +
Bidirectional PCA With Assembled Matrix +
Distance Metric for Image Recognition +
23ba9e462151a4bf9dfc3be5d8b12dbcfb7fe4c3CS 229 Project, Fall 2014 +
Determining Mood from Facial Expressions +
Introduction +
I +
Facial expressions play an extremely important role in human communication. As +
society continues to make greater use of human-machine interactions, it is important for +
machines to be able to interpret facial expressions in order to improve their +
authenticity. If machines can be trained to determine mood to a better extent than +
humans can, especially for more subtle moods, then this could be useful in fields such as +
counseling. This could also be useful for gauging reactions of large audiences in various +
contexts, such as political talks. +
The results of this project could also be applied to recognizing other features of facial +
expressions, such as determining when people are purposefully suppressing emotions or +
lying. The ability to recognize different facial expressions could also improve technology +
that recognizes to whom specific faces belong. This could in turn be used to search a +
large number of pictures for a specific photo, which is becoming increasingly difficult, as +
storing photos digitally has been extremely common in the past decade. The possibilities +
are endless. +
II Data and Features +
2.1 Data +
Our data consists of 1166 frontal images of +
people’s faces from three databases, with each +
image labeled with one of eight emotions: +
anger, contempt, disgust, fear, happiness, +
neutral, sadness, and surprise. The TFEID [1], +
CK+ [2], and JAFFE [3] databases primarily +
consist of Taiwanese, Caucasian, and Japanese +
subjects, respectively. The TFEID and JAFFE +
images are both cropped with the faces +
centered. Each image has a subject posing with +
one of the emotions. The JAFFE database does +
not have any images for contempt. +
2.2 Features +
On each face, there are many different facial landmarks. While some of these landmarks +
(pupil position, nose tip, and face contour) are not as indicative of emotion, others +
(eyebrow, mouth, and eye shape) are. To extract landmark data from images, we used +
Happiness +
Figure 1 +
Anger +
238fc68b2e0ef9f5ec043d081451902573992a032656 +
Enhanced Local Gradient Order Features and +
Discriminant Analysis for Face Recognition +
role in robust face recognition [5]. Many algorithms have +
been proposed to deal with the effectiveness of feature design +
and extraction [6], [7]; however, the performance of many +
existing methods is still highly sensitive to variations of +
imaging conditions, such as outdoor illumination, exaggerated +
expression, and continuous occlusion. These complex varia- +
tions are significantly affecting the recognition accuracy in +
recent years [8]–[10]. +
Appearance-based subspace learning is one of the sim- +
plest approach for feature extraction, and many methods +
are usually based on linear correlation of pixel intensities. +
For example, Eigenface [11] uses eigen system of pixel +
intensities to estimate the lower rank linear subspace of +
a set of training face images by minimizing the (cid:2)2 dis- +
tance metric. The solution enjoys optimality properties when +
noise is independent +
identically distributed Gaussian only. +
Fisherface [12] will suffer more due to the estimation of +
inverse within-class covariance matrix [13], +
thus the per- +
formance will degenerate rapidly in the cases of occlusion +
and small sample size. Laplacianfaces [14] refer to another +
appearance-based approach which learns a locality preserv- +
ing subspace and seeks to capture the intrinsic geometry +
and local structure of the data. Other methods such as those +
in [5] and [15] also provide valuable approaches to supervised +
or unsupervised dimension reduction tasks. +
A fundamental problem of appearance-based methods for +
face recognition, however, is that they are sensitive to imag- +
ing conditions [10]. As for data corrupted by illumination +
changes, occlusions, and inaccurate alignment, the estimated +
subspace will be biased, thus much of the efforts concentrate +
on removing/shrinking the noise components. In contrast, local +
feature descriptors [15]–[19] have certain advantages as they +
are more stable to local changes. In the view of image pro- +
cessing and vision, the basic imaging system can be simply +
formulated as +
(x, y) = A(x, y) × L(x, y) +
(1) +
23d55061f7baf2ffa1c847d356d8f76d78ebc8c1Solmaz et al. IPSJ Transactions on Computer Vision and +
Applications (2017) 9:22 +
DOI 10.1186/s41074-017-0033-4 +
IPSJ Transactions on Computer +
Vision and Applications +
RESEARCH PAPER +
Open Access +
Generic and attribute-specific deep +
representations for maritime vessels +
23a8d02389805854cf41c9e5fa56c66ee4160ce3Multimed Tools Appl +
DOI 10.1007/s11042-013-1568-8 +
Influence of low resolution of images on reliability +
of face detection and recognition +
© The Author(s) 2013. This article is published with open access at SpringerLink.com +
4fd29e5f4b7186e349ba34ea30738af7860cf21f
4f051022de100241e5a4ba8a7514db9167eabf6eFace Parsing via a Fully-Convolutional Continuous +
CRF Neural Network +
4f6adc53798d9da26369bea5a0d91ed5e1314df2IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. , NO. , 2016 +
Online Nonnegative Matrix Factorization with +
General Divergences +
4fbef7ce1809d102215453c34bf22b5f9f9aab26
4fa0d73b8ba114578744c2ebaf610d2ca9694f45
4f591e243a8f38ee3152300bbf42899ac5aae0a5SUBMITTED TO TPAMI +
Understanding Higher-Order Shape +
via 3D Shape Attributes +
4f9958946ad9fc71c2299847e9ff16741401c591Facial Expression Recognition with Recurrent Neural Networks +
Robotics and Embedded Systems Lab, Department of Computer Science +
Image Understanding and Knowledge-Based Systems, Department of Computer Science +
Technische Universit¨at M¨unchen, Germany +
4f0bf2508ae801aee082b37f684085adf0d06d23
4f4f920eb43399d8d05b42808e45b56bdd36a929International Journal of Computer Applications (0975 – 8887) +
Volume 123 – No.4, August 2015 +
A Novel Method for 3D Image Segmentation with Fusion +
of Two Images using Color K-means Algorithm +
Neelam Kushwah +
Dept. of CSE +
ITM Universe +
Gwalior +
Priusha Narwariya +
Dept. of CSE +
ITM Universe +
Gwalior +
two +
8d71872d5877c575a52f71ad445c7e5124a4b174
8de06a584955f04f399c10f09f2eed77722f6b1cAuthor manuscript, published in "International Conference on Computer Vision Theory and Applications (VISAPP 2013) (2013)" +
8d4f0517eae232913bf27f516101a75da3249d15ARXIV SUBMISSION, MARCH 2018 +
Event-based Dynamic Face Detection and +
Tracking Based on Activity +
8de2dbe2b03be8a99628ffa000ac78f8b66a1028´Ecole Nationale Sup´erieure dInformatique et de Math´ematiques Appliqu´ees de Grenoble +
INP Grenoble – ENSIMAG +
UFR Informatique et Math´ematiques Appliqu´ees de Grenoble +
Rapport de stage de Master 2 et de projet de fin d’´etudes +
Effectu´e au sein de l’´equipe LEAR, I.N.R.I.A., Grenoble +
Action Recognition in Videos +
3e ann´ee ENSIMAG – Option I.I.I. +
M2R Informatique – sp´ecialit´e I.A. +
04 f´evrier 2008 – 04 juillet 2008 +
LEAR, +
I.N.R.I.A., Grenoble +
655 avenue de l’Europe +
38 334 Montbonnot +
France +
Responsable de stage +
Mme. Cordelia Schmid +
Tuteur ´ecole +
Jury +
8d42a24d570ad8f1e869a665da855628fcb1378fCVPR +
#987 +
000 +
001 +
002 +
003 +
004 +
005 +
006 +
007 +
008 +
009 +
010 +
011 +
012 +
013 +
014 +
015 +
016 +
017 +
018 +
019 +
020 +
021 +
022 +
023 +
024 +
025 +
026 +
027 +
028 +
029 +
030 +
031 +
032 +
033 +
034 +
035 +
036 +
037 +
038 +
039 +
040 +
041 +
042 +
043 +
044 +
045 +
046 +
047 +
048 +
049 +
050 +
051 +
052 +
053 +
CVPR 2009 Submission #987. CONFIDENTIAL REVIEW COPY. DO NOT DISTRIBUTE. +
An Empirical Study of Context in Object Detection +
Anonymous CVPR submission +
Paper ID 987 +
8d8461ed57b81e05cc46be8e83260cd68a2ebb4dAge identification of Facial Images using Neural +
Network +
CSE Department,CSVTU +
RIT, Raipur, Chhattisgarh , INDIA +
8dbe79830713925affc48d0afa04ed567c54724b
8d1adf0ac74e901a94f05eca2f684528129a630aFacial Expression Recognition Using Facial +
Movement Features +
8d712cef3a5a8a7b1619fb841a191bebc2a17f15
8dffbb6d75877d7d9b4dcde7665888b5675deee1Emotion Recognition with Deep-Belief +
Networks +
Introduction +
For our CS229 project, we studied the problem of +
reliable computerized emotion recognition in images of +
human +
faces. First, we performed a preliminary +
exploration using SVM classifiers, and then developed an +
approach based on Deep Belief Nets. Deep Belief Nets, or +
DBNs, are probabilistic generative models composed of +
multiple layers of stochastic latent variables, where each +
“building block” layer is a Restricted Boltzmann Machine +
(RBM). DBNs have a greedy layer-wise unsupervised +
learning algorithm as well as a discriminative fine-tuning +
procedure for optimizing performance on classification +
tasks. [1]. +
We trained our classifier on three databases: the +
Cohn-Kanade Extended Database (CK+) [2], the Japanese +
Female Facial Expression Database (JAFFE) [3], and the +
Yale Face Database (YALE) [4]. We tested several +
different database configurations, image pre-processing +
settings, and DBN parameters, and obtained test errors as +
low as 20% on a limited subset of the emotion labels. +
Finally, we created a real-time system which takes +
images of a single subject using a computer webcam and +
classifies the emotion shown by the subject. +
Part 1: Exploration of SVM-based approaches +
To set a baseline for comparison, we applied an +
SVM classifier to the emotion images in the CK+ +
database, using the LIBLINEAR library and its MATLAB +
interface [5]. This database contains 593 image sequences +
across 123 human subjects, beginning with a “neutral +
“expression and showing the progression to one of seven +
“peak” emotions. When given both a neutral and an +
expressive face to compare, the SVM obtained accuracy +
as high as 90%. This +
the +
implementation of the SVM classifier. For additional +
details on this stage of the project, please see our +
Milestone document. +
Part 1.1 Choice of labels (emotion numbers vs. FACS +
features) +
The CK+ database offers two sets of emotion +
features: “emotion numbers” and FACS features. Emotion +
numbers are integer values representing the main emotion +
shown in the “peak emotion” image. The emotions are +
coded as follows: 1=anger, 2=contempt, 3=disgust, +
4=fear, 5=happiness, 6=sadness, and 7=surprise. +
The other labeling option is called FACS, or the +
Facial Action Coding System. FACS decomposes every +
summarizes +
section +
facial emotion into a set of Action Units (AUs), which +
describe the specific muscle groups involved in forming +
the emotion. We chose not to use FACS because accurate +
labeling currently requires trained human experts [8], and +
we are interesting in creating an automated system. +
+
Part 1.2 Features +
Part 1.2.1 Norm of differences between neutral face +
and full emotion +
Each of the CK+ images has been hand-labeled with +
68 standard Active Appearance Models (AAM) face +
landmarks that describe the X and Y position of these +
landmarks on the image (Figure 1). +
Figure 1. AAM Facial Landmarks +
We initially trained the SVM on the norm of the +
vector differences in landmark positions between the +
neutral and peak expressions. With this approach, the +
training error was approximately 35% for hold out cross +
validation (see Figure 2). +
with +
Figure 3. Accuracy of +
SVM with separate X, Y +
displacement features. +
Figure 2. Accuracy of +
SVM +
norm- +
displacement features. +
Part 1.2.2 Separate X and Y differences between +
neutral face and full emotion +
Because the initial approach did not differentiate +
between displacements of +
in different +
directions, we also provided the differences in the X and +
Y components of each landmark separately. This doubled +
the size of our feature vector, and resulting in a significant +
(about 20%) improvement in accuracy (Figure 3). +
Part 1.2.3 Feature Selection +
landmarks +
Finally, we visualized which features were the most +
important for classifying each emotion; the results can be +
seen in Figure 4. The figure shows the X and Y +
153f5ad54dd101f7f9c2ae17e96c69fe84aa9de4Overview of algorithms for face detection and +
tracking +
Nenad Markuˇs +
15136c2f94fd29fc1cb6bedc8c1831b7002930a6Deep Learning Architectures for Face +
Recognition in Video Surveillance +
153e5cddb79ac31154737b3e025b4fb639b3c9e7PREPRINT SUBMITTED TO IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS +
Active Dictionary Learning in Sparse +
Representation Based Classification +
15e0b9ba3389a7394c6a1d267b6e06f8758ab82bXu et al. IPSJ Transactions on Computer Vision and +
Applications (2017) 9:24 +
DOI 10.1186/s41074-017-0035-2 +
IPSJ Transactions on Computer +
Vision and Applications +
TECHNICAL NOTE +
Open Access +
The OU-ISIR Gait Database comprising the +
Large Population Dataset with Age and +
performance evaluation of age estimation +
15aa6c457678e25f6bc0e818e5fc39e42dd8e533
15f3d47b48a7bcbe877f596cb2cfa76e798c6452Automatic face analysis tools for interactive digital games +
Anonymised for blind review +
Anonymous +
Anonymous +
Anonymous +
15728d6fd5c9fc20b40364b733228caf63558c31
1513949773e3a47e11ab87d9a429864716aba42d
153c8715f491272b06dc93add038fae62846f498
122ee00cc25c0137cab2c510494cee98bd504e9fThe Application of +
Active Appearance Models to +
Comprehensive Face Analysis +
Technical Report +
TU M¨unchen +
April 5, 2007 +
1287bfe73e381cc8042ac0cc27868ae086e1ce3b
12cb3bf6abf63d190f849880b1703ccc183692feGuess Who?: A game to crowdsource the labeling of affective facial +
expressions is comparable to expert ratings. +
Graduation research project, june 2012 +
Supervised by: Dr. Joost Broekens +
12cd96a419b1bd14cc40942b94d9c4dffe5094d229 +
Proceedings of the 5th Workshop on Vision and Language, pages 29–38, +
Berlin, Germany, August 12 2016. c(cid:13)2016 Association for Computational Linguistics +
12055b8f82d5411f9ad196b60698d76fbd07ac1e1475 +
Multiview Facial Landmark Localization in RGB-D +
Images via Hierarchical Regression +
With Binary Patterns +
120785f9b4952734818245cc305148676563a99bDiagnostic automatique de l’état dépressif +
S. Cholet +
H. Paugam-Moisy +
Laboratoire de Mathématiques Informatique et Applications (LAMIA - EA 4540) +
Université des Antilles, Campus de Fouillole - Guadeloupe +
Résumé +
Les troubles psychosociaux sont un problème de santé pu- +
blique majeur, pouvant avoir des conséquences graves sur +
le court ou le long terme, tant sur le plan professionnel que +
personnel ou familial. Le diagnostic de ces troubles doit +
être établi par un professionnel. Toutefois, l’IA (l’Intelli- +
gence Artificielle) peut apporter une contribution en four- +
nissant au praticien une aide au diagnostic, et au patient +
un suivi permanent rapide et peu coûteux. Nous proposons +
une approche vers une méthode de diagnostic automatique +
de l’état dépressif à partir d’observations du visage en +
temps réel, au moyen d’une simple webcam. A partir de +
vidéos du challenge AVEC’2014, nous avons entraîné un +
classifieur neuronal à extraire des prototypes de visages +
selon différentes valeurs du score de dépression de Beck +
(BDI-II). +
12c713166c46ac87f452e0ae383d04fb44fe4eb2
12150d8b51a2158e574e006d4fbdd3f3d01edc93Deep End2End Voxel2Voxel Prediction +
Presented by: Ahmed Osman +
Ahmed Osman +
8c13f2900264b5cf65591e65f11e3f4a35408b48A GENERIC FACE REPRESENTATION APPROACH FOR +
LOCAL APPEARANCE BASED FACE VERIFICATION +
Interactive Systems Labs, Universität Karlsruhe (TH) +
76131 Karlsruhe, Germany +
web: http://isl.ira.uka.de/face_recognition/ +
8cb3f421b55c78e56c8a1c1d96f23335ebd4a5bf
8c955f3827a27e92b6858497284a9559d2d0623aBuletinul Ştiinţific al Universităţii "Politehnica" din Timişoara +
Seria ELECTRONICĂ şi TELECOMUNICAŢII +
TRANSACTIONS on ELECTRONICS and COMMUNICATIONS +
Tom 53(67), Fascicola 1-2, 2008 +
Facial Expression Recognition under Noisy Environment +
Using Gabor Filters +
8ce9b7b52d05701d5ef4a573095db66ce60a7e1cStructured Sparse Subspace Clustering: A Joint +
Affinity Learning and Subspace Clustering +
Framework +
8c6c0783d90e4591a407a239bf6684960b72f34eSESSION +
KNOWLEDGE ENGINEERING AND +
MANAGEMENT + KNOWLEDGE ACQUISITION +
Chair(s) +
TBA +
Int'l Conf. Information and Knowledge Engineering | IKE'13 |1
8509abbde2f4b42dc26a45cafddcccb2d370712fImproving precision and recall of face recognition in SIPP with combination of +
modified mean search and LSH +
Xihua.Li +
855bfc17e90ec1b240efba9100fb760c068a8efa
858ddff549ae0a3094c747fb1f26aa72821374ecSurvey on RGB, 3D, Thermal, and Multimodal +
Approaches for Facial Expression Recognition: +
History, Trends, and Affect-related Applications +
858901405086056361f8f1839c2f3d65fc86a748ON TENSOR TUCKER DECOMPOSITION: THE CASE FOR AN +
ADJUSTABLE CORE SIZE +
85188c77f3b2de3a45f7d4f709b6ea79e36bd0d9Author manuscript, published in "Workshop on Faces in 'Real-Life' Images: Detection, Alignment, and Recognition, Marseille : +
France (2008)" +
8518b501425f2975ea6dcbf1e693d41e73d0b0afRelative Hidden Markov Models for Evaluating Motion Skills +
Computer Science and Engineering +
Arizona State Univerisity, Tempe, AZ 85281 +
854dbb4a0048007a49df84e3f56124d387588d99JOURNAL OF LATEX CLASS FILES, VOL. 13, NO. 9, SEPTEMBER 2014 +
Spatial-Temporal Recurrent Neural Network for +
Emotion Recognition +
1dbbec4ad8429788e16e9f3a79a80549a0d7ac7b
1d7ecdcb63b20efb68bcc6fd99b1c24aa6508de91860 +
The Hidden Sides of Names—Face Modeling +
with First Name Attributes +
1d846934503e2bd7b8ea63b2eafe00e29507f06a
1d0dd20b9220d5c2e697888e23a8d9163c7c814bNEGREL ET AL.: BOOSTED METRIC LEARNING FOR FACE RETRIEVAL +
Boosted Metric Learning for Efficient +
Identity-Based Face Retrieval +
Frederic Jurie +
GREYC, CNRS UMR 6072, ENSICAEN +
Université de Caen Basse-Normandie +
France +
1d776bfe627f1a051099997114ba04678c45f0f5Deployment of Customized Deep Learning based +
Video Analytics On Surveillance Cameras +
AitoeLabs (www.aitoelabs.com) +
1d3e01d5e2721dcfafe5a3b39c54ee1c980350bb
1de8f38c35f14a27831130060810cf9471a62b45Int J Comput Vis +
DOI 10.1007/s11263-017-0989-7 +
A Branch-and-Bound Framework for Unsupervised Common +
Event Discovery +
Received: 3 June 2016 / Accepted: 12 January 2017 +
© Springer Science+Business Media New York 2017 +
1da83903c8d476c64c14d6851c85060411830129Iterated Support Vector Machines for Distance +
Metric Learning +
1d6068631a379adbcff5860ca2311b790df3a70f
1d58d83ee4f57351b6f3624ac7e727c944c0eb8dEnhanced Local Texture +
Feature Sets for Face +
Recognition under Difficult +
Lighting Conditions +
INRIA & Laboratoire Jean +
Kuntzmann, +
655 avenue de l'Europe, Montbonnot 38330, France +
71b376dbfa43a62d19ae614c87dd0b5f1312c966The Temporal Connection Between Smiles and Blinks +
714d487571ca0d676bad75c8fa622d6f50df953beBear: An Expressive Bear-Like Robot +
710011644006c18291ad512456b7580095d628a2Learning Residual Images for Face Attribute Manipulation +
Fujitsu Research & Development Center, Beijing, China. +
76fd801981fd69ff1b18319c450cb80c4bc78959Proceedings of the 11th International Conference on Computational Semantics, pages 76–81, +
London, UK, April 15-17 2015. c(cid:13)2015 Association for Computational Linguistics +
76 +
76dc11b2f141314343d1601635f721fdeef86fdbWeighted Decoding ECOC for Facial +
Action Unit Classification +
760a712f570f7a618d9385c0cee7e4d0d6a78ed2
76b9fe32d763e9abd75b427df413706c4170b95c
76d9f5623d3a478677d3f519c6e061813e58e833FAST ALGORITHMS FOR THE GENERALIZED FOLEY-SAMMON +
DISCRIMINANT ANALYSIS +
765b2cb322646c52e20417c3b44b81f89860ff71PoseShop: Human Image Database +
Construction and Personalized +
Content Synthesis +
7644d90efef157e61fe4d773d8a3b0bad5feccec
760ba44792a383acd9ca8bef45765d11c55b48d4~ +
I . +
INTRODUCTION AND BACKGROUND +
The purpose of this article is to introduce the +
reader to the basic principles of classification with +
class-specific features. It is written both for readers +
interested in only the basic concepts as well as those +
interested in getting started in applying the method. +
For in-depth coverage, the reader is referred to a more +
detailed article [l]. +
Class-Specific Classifier: +
Avoiding the Curse of +
Dimensionality +
PAUL M. BAGGENSTOSS, Member. lEEE +
US. Naval Undersea Warfare Center +
This article describes a new probabilistic method called the +
“class-specific method” (CSM). CSM has the potential to avoid +
the “curse of dimensionality” which plagues most clmiiiers +
which attempt to determine the decision boundaries in a +
highdimensional featue space. In contrast, in CSM, it is possible +
to build classifiers without a ” n o n feature space. Separate +
Law-dimensional features seta may be de6ned for each class, while +
the decision funetions are projected back to the common raw data +
space. CSM eflectively extends the classical classification theory +
to handle multiple feature spaw.. It is completely general, and +
requires no s i m p l i n g assumption such as Gaussianity or that +
data lies in linear subspaces. +
Manuscript received September 26, 2W2; revised February 12, +
2003. +
This work was supported by the Office of Naval Research. +
Author’s address: US. Naval Undersea Warfare Center, Newport +
Classification is the process of assigning data +
to one of a set of pre-determined class labels [2]. +
Classification is a fundamental problem that has +
to be solved if machines are to approximate the +
human functions of recognizing sounds, images, or +
other sensory inputs. This is why classification is so +
important for automation in today’s commercial and +
military arenas. +
Many of us have first-hand knowledge of +
successful automated recognition systems from +
cameras that recognize faces in airports to computers +
that can scan and read printed and handwritten text, +
or systems that can recognize human speech. These +
systems are becoming more and more reliable and +
accurate. Given reasonably clean input data, the +
performance is often quite good if not perfect. But +
many of these systems fail in applications where +
clean, uncorrupted data is not available or if the +
problem is complicated by variability of conditions +
or by proliferation of inputs from unknown sources. +
In military environments, the targets to he recognized +
are often uncooperative and hidden in clutter and +
interference. In short, military uses of such systems +
still fall far short of what a well-trained alert human +
operator can achieve. +
We are often perplexed by the wide gap of +
as a car door slamming. From +
performance between humans and automated systems. +
Allow a human listener to hear two or three examples +
of a sound-such +
these few examples, the human can recognize +
the sound again and not confuse it with similar +
interfering sounds. But try the same experiment with +
general-purpose classifiers using neural networks +
and the story is quite different. Depending on the +
problem, the automated system may require hundreds, +
thousands, even millions of examples for training +
before it becomes both robust and reliable. +
Why? The answer lies in what is known as the +
“curse of dimensionality.” General-purpose classifiers +
need to extract a large number of measurements, +
or features, from the data to account for all the +
different possibilities of data types. The large +
collection of features form a high-dimensional space +
that the classifier has to sub-divide into decision +
boundaries. It is well-known that the complexity of +
a high-dimensional space increases exponentially +
with the number of measurements [31-and +
so does +
the difficulty of finding the hest decision boundaries +
from a fixed amount of training data. Unless a lot +
EEE A&E SYSTEMS MAGAZINE VOL. 19, NO. 1 JANUARY 2004 PART 2: TUTORIALS-BAGGENSTOSS +
37 +
766728bac030b169fcbc2fbafe24c6e22a58ef3cA survey of deep facial landmark detection +
Yongzhe Yan1,2 +
Thierry Chateau1 +
1 Université Clermont Auvergne, France +
2 Wisimage, France +
3 Université de Lyon, CNRS, INSA Lyon, LIRIS, UMR5205, Lyon, France +
Résumé +
La détection de landmarks joue un rôle crucial dans de +
nombreuses applications d’analyse du visage comme la +
reconnaissance de l’identité, des expressions, l’animation +
d’avatar, la reconstruction 3D du visage, ainsi que pour +
les applications de réalité augmentée comme la pose de +
masque ou de maquillage virtuel. L’avènement de l’ap- +
prentissage profond a permis des progrès très importants +
dans ce domaine, y compris sur les corpus non contraints +
(in-the-wild). Nous présentons ici un état de l’art cen- +
tré sur la détection 2D dans une image fixe, et les mé- +
thodes spécifiques pour la vidéo. Nous présentons ensuite +
les corpus existants pour ces trois tâches, ainsi que les mé- +
triques d’évaluations associées. Nous exposons finalement +
quelques résultats, ainsi que quelques pistes de recherche. +
Mots Clef +
Détection de landmark facial, Alignement de visage, Deep +
learning +
7697295ee6fc817296bed816ac5cae97644c2d5bDetecting and Recognizing Human-Object Interactions +
Facebook AI Research (FAIR) +
1c80bc91c74d4984e6422e7b0856cf3cf28df1fbNoname manuscript No. +
(will be inserted by the editor) +
Hierarchical Adaptive Structural SVM for Domain Adaptation +
Received: date / Accepted: date +
1ce4587e27e2cf8ba5947d3be7a37b4d1317fbeeDeep fusion of visual signatures +
for client-server facial analysis +
Normandie Univ, UNICAEN, +
ENSICAEN, CNRS, GREYC +
Computer Sc. & Engg. +
IIT Kanpur, India +
Frederic Jurie +
Normandie Univ, UNICAEN, +
ENSICAEN, CNRS, GREYC +
Facial analysis is a key technology for enabling human- +
machine interaction. +
In this context, we present a client- +
server framework, where a client transmits the signature of +
a face to be analyzed to the server, and, in return, the server +
sends back various information describing the face e.g. is the +
person male or female, is she/he bald, does he have a mus- +
tache, etc. We assume that a client can compute one (or a +
combination) of visual features; from very simple and effi- +
cient features, like Local Binary Patterns, to more complex +
and computationally heavy, like Fisher Vectors and CNN +
based, depending on the computing resources available. The +
challenge addressed in this paper is to design a common uni- +
versal representation such that a single merged signature is +
transmitted to the server, whatever be the type and num- +
ber of features computed by the client, ensuring nonetheless +
an optimal performance. Our solution is based on learn- +
ing of a common optimal subspace for aligning the different +
face features and merging them into a universal signature. +
We have validated the proposed method on the challenging +
CelebA dataset, on which our method outperforms existing +
state-of-art methods when rich representation is available at +
test time, while giving competitive performance when only +
simple signatures (like LBP) are available at test time due +
to resource constraints on the client. +
1. +
INTRODUCTION +
We propose a novel method in a heterogeneous server- +
client framework for the challenging and important task of +
analyzing images of faces. Facial analysis is a key ingredient +
for assistive computer vision and human-machine interaction +
methods, and systems and incorporating high-performing +
methods in daily life devices is a challenging task. The ob- +
jective of the present paper is to develop state-of-the-art +
technologies for recognizing facial expressions and facial at- +
tributes on mobile and low cost devices. Depending on their +
computing resources, the clients (i.e. the devices on which +
the face image is taken) are capable of computing different +
types of face signatures, from the simplest ones (e.g. LPB) +
to the most complex ones (e.g. very deep CNN features), and +
should be able to eventually combine them into a single rich +
signature. Moreover, it is convenient if the face analyzer, +
which might require significant computing resources, is im- +
plemented on a server receiving face signatures and comput- +
ing facial expressions and attributes from these signatures. +
Keeping the computation of the signatures on the client is +
safer in terms of privacy, as the original images are not trans- +
mitted, and keeping the analysis part on the server is also +
beneficial for easy model upgrades in the future. To limit +
the transmission costs, the signatures have to be made as +
compact as possible. +
In summary, the technology needed +
for this scenario has to be able to merge the different avail- +
able features – the number of features available at test time +
is not known in advance but is dependent on the computing +
resources available on the client – producing a unique rich +
and compact signature of the face, which can be transmitted +
and analyzed by a server. Ideally, we would like the univer- +
sal signature to have the following properties: when all the +
features are available, we would like the performance of the +
signature to be better than the one of a system specifically +
optimized for any single type of feature. +
In addition, we +
would like to have reasonable performance when only one +
type of feature is available at test time. +
For developing such a system, we propose a hybrid deep +
neural network and give a method to carefully fine-tune the +
network parameters while learning with all or a subset of +
features available. Thus, the proposed network can process a +
number of wide ranges of feature types such as hand-crafted +
LBP and FV, or even CNN features which are learned end- +
to-end. +
While CNNs have been quite successful in computer vi- +
sion [1], representing images with CNN features is relatively +
time consuming, much more than some simple hand-crafted +
features such as LBP. Thus, the use of CNN in real-time ap- +
plications is still not feasible. In addition, the use of robust +
hand-crafted features such as FV in hybrid architectures can +
give performance comparable to Deep CNN features [2]. The +
main advantage of learning hybrid architectures is to avoid +
having large numbers of convolutional and pooling layers. +
Again from [2], we can also observe that hybrid architec- +
tures improve the performance of hand-crafted features e.g. +
FVs. Therefore, hybrid architectures are useful for the cases +
where only hand-crafted features, and not the original im- +
ages, are available during training and testing time. This +
scenario is useful when it is not possible to share training +
images due to copyright or privacy issues. +
Hybrid networks are particularly adapted to our client- +
1c3073b57000f9b6dbf1c5681c52d17c55d60fd7THÈSEprésentéepourl’obtentiondutitredeDOCTEURDEL’ÉCOLENATIONALEDESPONTSETCHAUSSÉESSpécialité:InformatiqueparCharlotteGHYSAnalyse,Reconstruction3D,&AnimationduVisageAnalysis,3DReconstruction,&AnimationofFacesSoutenancele19mai2010devantlejurycomposéde:Rapporteurs:MajaPANTICDimitrisSAMARASExaminateurs:MichelBARLAUDRenaudKERIVENDirectiondethèse:NikosPARAGIOSBénédicteBASCLE
1c93b48abdd3ef1021599095a1a5ab5e0e020dd5JOURNAL OF LATEX CLASS FILES, VOL. *, NO. *, JANUARY 2009 +
A Compositional and Dynamic Model for Face Aging +
1c6be6874e150898d9db984dd546e9e85c85724e
1c65f3b3c70e1ea89114f955624d7adab620a013
1c6e22516ceb5c97c3caf07a9bd5df357988ceda
82bef8481207de9970c4dc8b1d0e17dced706352
825f56ff489cdd3bcc41e76426d0070754eab1a8Making Convolutional Networks Recurrent for Visual Sequence Learning +
NVIDIA +
82d2af2ffa106160a183371946e466021876870dA Novel Space-Time Representation on the Positive Semidefinite Cone +
for Facial Expression Recognition +
1IMT Lille Douai, Univ. Lille, CNRS, UMR 9189 – CRIStAL – +
Centre de Recherche en Informatique Signal et Automatique de Lille, F-59000 Lille, France +
2Univ. Lille, CNRS, UMR 8524, Laboratoire Paul Painlev´e, F-59000 Lille, France. +
8210fd10ef1de44265632589f8fc28bc439a57e6Single Sample Face Recognition via Learning Deep +
Supervised Auto-Encoders +
Shenghua Gao, Yuting Zhang, Kui Jia, Jiwen Lu, Yingying Zhang +
82a4a35b2bae3e5c51f4d24ea5908c52973bd5beReal-time emotion recognition for gaming using +
deep convolutional network features +
S´ebastien Ouellet +
82f4e8f053d20be64d9318529af9fadd2e3547efTechnical Report: +
Multibiometric Cryptosystems +
82d781b7b6b7c8c992e0cb13f7ec3989c8eafb3d141 +
REFERENCES +
1. +
2. +
3. +
4. +
5. +
6. +
7. +
8. +
9. +
Adler A., Youmaran R. and Loyka S., “Towards a Measure of +
Biometric Information”, Canadian Conference on Electrical and +
Computer Engineering, pp. 210-213, 2006. +
Military Academy, West Point, New York, pp. 452-458, 2005. +
Security and Trust, St. Andrews, New Brunswick, Canada, pp. 1-8, +
2005. +
Structural Model for Biometric Sketch Recognition”, Proceedings of +
DAGM, Magdeburg, Germany, Vol. 2781, pp. 187-195, 2003. +
of Security”, The First UAE International Conference on Biological +
and Medical Physics, pp. 1-4, 2005. +
Avraam Kasapis., “MLPs and Pose, Expression Classification”, +
Proceedings of UNiS Report, pp. 1-87, 2003. +
Detection for Storage Area Networks (SANs)”, Proceedings of 22nd +
IEEE / 13th NASA Goddard Conference on Mass Storage Systems and +
Technologies, pp. 118-127, 2005. +
Black M.J. and Yacoob Y., “Recognizing Facial Expressions in Image +
Sequences using Local Parameterized Models of Image Motion”, Int. +
Journal Computer Vision, Vol. 25, No. 1, pp. 23-48, 1997. +
10. +
Recognition using a State-Based Model of Spatially-Localized Facial +
82417d8ec8ac6406f2d55774a35af2a1b3f4b66eSome faces are more equal than others: +
Hierarchical organization for accurate and +
efficient large-scale identity-based face retrieval +
GREYC, CNRS UMR 6072, Universit´e de Caen Basse-Normandie, France1 +
Technicolor, Rennes, France2 +
826c66bd182b54fea3617192a242de1e4f16d020978-1-5090-4117-6/17/$31.00 ©2017 IEEE +
1602 +
ICASSP 2017 +
4972aadcce369a8c0029e6dc2f288dfd0241e144Multi-target Unsupervised Domain Adaptation +
without Exactly Shared Categories +
49dd4b359f8014e85ed7c106e7848049f852a304
49e85869fa2cbb31e2fd761951d0cdfa741d95f3253 +
Adaptive Manifold Learning +
49659fb64b1d47fdd569e41a8a6da6aa76612903
49a7949fabcdf01bbae1c2eb38946ee99f491857A CONCATENATING FRAMEWORK OF SHORTCUT +
CONVOLUTIONAL NEURAL NETWORKS +
49e1aa3ecda55465641b2c2acc6583b32f3f1fc6International Journal of Emerging Technology and Advanced Engineering +
Website: www.ijetae.com (ISSN 2250-2459, Volume 2, Issue 5, May 2012) +
Support Vector Machine for age classification +
1Assistant Professor, CSE, RSR RCET, Kohka Bhilai +
2,3 Sr. Assistant Professor, CSE, SSCET, Junwani Bhilai +
49df381ea2a1e7f4059346311f1f9f45dd9971642018 +
On the Use of Client-Specific Information for Face +
Presentation Attack Detection Based on Anomaly +
Detection +
40205181ed1406a6f101c5e38c5b4b9b583d06bcUsing Context to Recognize People in Consumer Images +
40dab43abef32deaf875c2652133ea1e2c089223Noname manuscript No. +
(will be inserted by the editor) +
Facial Communicative Signals +
Valence Recognition in Task-Oriented Human-Robot Interaction +
Received: date / Accepted: date +
405b43f4a52f70336ac1db36d5fa654600e9e643What can we learn about CNNs from a large scale controlled object dataset? +
UWM +
AUT +
USC +
40b86ce698be51e36884edcc8937998979cd02ecYüz ve İsim İlişkisi kullanarak Haberlerdeki Kişilerin Bulunması +
Finding Faces in News Photos Using Both Face and Name Information +
Derya Ozkan, Pınar Duygulu +
Bilgisayar Mühendisliği Bölümü, Bilkent Üniversitesi, 06800, Ankara +
Özetçe +
Bu çalışmada, haber fotoğraflarından oluşan geniş veri +
kümelerinde kişilerin sorgulanmasını sağlayan bir yöntem +
sunulmuştur. Yöntem isim ve yüzlerin ilişkilendirilmesine +
dayanmaktadır. Haber başlığında kişinin ismi geçiyor ise +
fotoğrafta da o kişinin yüzünün bulunacağı varsayımıyla, ilk +
olarak sorgulanan isim ile ilişkilendirilmiş, fotoğraflardaki +
tüm yüzler seçilir. Bu yüzler arasında sorgu kişisine ait farklı +
koşul, poz ve zamanlarda çekilmiş pek çok resmin yanında, +
haberde ismi geçen başka kişilere ait yüzler ya da kullanılan +
yüz bulma yönteminin hatasından kaynaklanan yüz olmayan +
resimler de bulunabilir. Yine de, çoğu zaman, sorgu kişisine +
ait resimler daha çok olup, bu resimler birbirine diğerlerine +
olduğundan daha çok benzeyeceklerdir. Bu nedenle, yüzler +
arasındaki benzerlikler çizgesel olarak betimlendiğinde , +
birbirine en çok benzeyen yüzler bu çizgede en yoğun bileşen +
olacaktır. Bu çalışmada, sorgu ismiyle ilişkilendirilmiş, +
yüzler arasında birbirine en çok benzeyen alt kümeyi bulan, +
çizgeye dayalı bir yöntem sunulmaktadır. +
402f6db00251a15d1d92507887b17e1c50feebca3D Facial Action Units Recognition for Emotional +
Expression +
1Department of Information Technology and Communication, Politeknik Kuching, Sarawak, Malaysia +
2Faculty of Computer Science and Information Technology, Universiti Malaysia Sarawak, Kota Samarahan, Sarawak, Malaysia +
The muscular activities caused the activation of certain AUs for every facial expression at the certain duration of time +
throughout the facial expression. This paper presents the methods to recognise facial Action Unit (AU) using facial distance +
of the facial features which activates the muscles. The seven facial action units involved are AU1, AU4, AU6, AU12, AU15, +
AU17 and AU25 that characterises happy and sad expression. The recognition is performed on each AU according to rules +
defined based on the distance of each facial points. The facial distances chosen are extracted from twelve facial features. +
Then the facial distances are trained using Support Vector Machine (SVM) and Neural Network (NN). Classification result +
using SVM is presented with several different SVM kernels while result using NN is presented for each training, validation +
and testing phase. +
Keywords: Facial action units recognition, 3D AU recognition, facial expression +
+
40fb4e8932fb6a8fef0dddfdda57a3e142c3e823A Mixed Generative-Discriminative Framework for Pedestrian Classification +
Dariu M. Gavrila2,3 +
1 Image & Pattern Analysis Group, Dept. of Math. and Comp. Sc., Univ. of Heidelberg, Germany +
2 Environment Perception, Group Research, Daimler AG, Ulm, Germany +
3 Intelligent Systems Lab, Faculty of Science, Univ. of Amsterdam, The Netherlands +
40cd062438c280c76110e7a3a0b2cf5ef675052c
40a1935753cf91f29ffe25f6c9dde2dc49bf2a3a80 +
40a34d4eea5e32dfbcef420ffe2ce7c1ee0f23cdBridging Heterogeneous Domains With Parallel Transport For Vision and +
Multimedia Applications +
Dept. of Video and Multimedia Technologies Research +
AT&T Labs-Research +
San Francisco, CA 94108 +
40389b941a6901c190fb74e95dc170166fd7639dAutomatic Facial Expression Recognition +
Emotient +
http://emotient.com +
February 12, 2014 +
Imago animi vultus est, indices oculi. (Cicero) +
Introduction +
The face is innervated by two different brain systems that compete for control of its muscles: +
a cortical brain system related to voluntary and controllable behavior, and a sub-cortical +
system responsible for involuntary expressions. The interplay between these two systems +
generates a wealth of information that humans constantly use to read the emotions, inten- +
tions, and interests [25] of others. +
Given the critical role that facial expressions play in our daily life, technologies that can +
interpret and respond to facial expressions automatically are likely to find a wide range of +
applications. For example, in pharmacology, the effect of new anti-depression drugs could +
be assessed more accurately based on daily records of the patients’ facial expressions than +
asking the patients to fill out a questionnaire, as it is currently done [7]. Facial expression +
recognition may enable a new generation of teaching systems to adapt to the expression +
of their students in the way good teachers do [61]. Expression recognition could be used +
to assess the fatigue of drivers and air-pilots [58, 59]. Daily-life robots with automatic +
expression recognition will be able to assess the states and intentions of humans and respond +
accordingly [41]. Smart phones with expression analysis may help people to prepare for +
important meetings and job interviews. +
Thanks to the introduction of machine learning methods, recent years have seen great +
progress in the field of automatic facial expression recognition. Commercial real-time ex- +
pression recognition systems are starting to be used in consumer applications, e.g., smile +
detectors embedded in digital cameras [62]. Nonetheless, considerable progress has yet to be +
made: Methods for face detection and tracking (the first step of automated face analysis) +
work well for frontal views of adult Caucasian and Asian faces [50], but their performance +
40273657e6919455373455bd9a5355bb46a7d614Anonymizing k-Facial Attributes via Adversarial Perturbations +
1 IIIT Delhi, New Delhi, India +
2 Ministry of Electronics and Information Technology, New Delhi, India +
40b10e330a5511a6a45f42c8b86da222504c717fImplementing the Viola-Jones +
Face Detection Algorithm +
Kongens Lyngby 2008 +
IMM-M.Sc.-2008-93 +
40ca925befa1f7e039f0cd40d57dbef6007b4416Sampling Matters in Deep Embedding Learning +
UT Austin +
A9/Amazon +
Amazon +
Philipp Kr¨ahenb¨uhl +
UT Austin +
4042bbb4e74e0934f4afbedbe92dd3e37336b2f4
40f127fa4459a69a9a21884ee93d286e99b54c5fOptimizing Apparent Display Resolution +
Enhancement for Arbitrary Videos +
401e6b9ada571603b67377b336786801f5b54eeeActive Image Clustering: Seeking Constraints from +
Humans to Complement Algorithms +
November 22, 2011 +
2e20ed644e7d6e04dd7ab70084f1bf28f93f75e9
2e8e6b835e5a8f55f3b0bdd7a1ff765a0b7e1b87International Journal of Computer Vision manuscript No. +
(will be inserted by the editor) +
Pointly-Supervised Action Localization +
Received: date / Accepted: date +
2eb37a3f362cffdcf5882a94a20a1212dfed25d94 +
Local Feature Based Face Recognition +
R.I.T., Rajaramnagar and S.G.G.S. COE &T, Nanded +
India +
1. Introduction +
A reliable automatic face recognition (AFR) system is a need of time because in today's +
networked world, maintaining the security of private information or physical property is +
becoming increasingly important and difficult as well. Most of the time criminals have been +
taking the advantage of fundamental flaws in the conventional access control systems i.e. +
the systems operating on credit card, ATM etc. do not grant access by "who we are", but by +
"what we have”. The biometric based access control systems have a potential to overcome +
most of the deficiencies of conventional access control systems and has been gaining the +
importance in recent years. These systems can be designed with biometric traits such as +
fingerprint, face, iris, signature, hand geometry etc. But comparison of different biometric +
traits shows that face is very attractive biometric because of its non-intrusiveness and social +
acceptability. It provides automated methods of verifying or recognizing the identity of a +
living person based on its facial characteristics. +
In last decade, major advances occurred in face recognition, with many systems capable of +
achieving recognition rates greater than 90%. However real-world scenarios remain a +
challenge, because face acquisition process can undergo to a wide range of variations. Hence +
the AFR can be thought as a very complex object recognition problem, where the object to be +
recognized is the face. This problem becomes even more difficult because the search is done +
among objects belonging to the same class and very few images of each class are available to +
train the system. Moreover different problems arise when images are acquired under +
uncontrolled conditions such as illumination variations, pose changes, occlusion, person +
appearance at different ages, expression changes and face deformations. The numbers of +
approaches has been proposed by various researchers to deal with these problems but still +
reported results cannot suffice the need of the reliable AFR system in presence of all facial +
image variations. A recent survey paper (Abate et al., 2007) states that the sensibility of the +
AFR systems to illumination and pose variations are the main problems researchers have +
been facing up till. +
2. Face recognition methods +
The existing face recognition methods can be divided into two categories: holistic matching +
methods and local matching methods.The holistic matching methods use complete face +
region as a input to face recognition system and constructs a lower dimensional subspace +
using principal component analysis (PCA) (Turk & Pentland, 1991), linear discriminant +
www.intechopen.com +
2e5cfa97f3ecc10ae8f54c1862433285281e6a7c
2e0e056ed5927a4dc6e5c633715beb762628aeb0
2e68190ebda2db8fb690e378fa213319ca915cf8Generating Videos with Scene Dynamics +
MIT +
UMBC +
MIT +
2e0d56794379c436b2d1be63e71a215dd67eb2caImproving precision and recall of face recognition in SIPP with combination of +
modified mean search and LSH +
Xihua.Li +
2ee8900bbde5d3c81b7ed4725710ed46cc7e91cd
2ef51b57c4a3743ac33e47e0dc6a40b0afcdd522Leveraging Billions of Faces to Overcome +
Performance Barriers in Unconstrained Face +
Recognition +
face.com +
2e19371a2d797ab9929b99c80d80f01a1fbf9479
2ebc35d196cd975e1ccbc8e98694f20d7f52faf3This 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. +
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE +
Towards Wide-angle Micro Vision Sensors +
2e3d081c8f0e10f138314c4d2c11064a981c1327
2e86402b354516d0a8392f75430156d629ca6281
2e0f5e72ad893b049f971bc99b67ebf254e194f7Apparel Classification with Style +
1ETH Z¨urich, Switzerland 2Microsoft, Austria 3Kooaba AG, Switzerland +
4KU Leuven, Belgium +
2ec7d6a04c8c72cc194d7eab7456f73dfa501c8cInternational Journal of Scientific Research and Management Studies (IJSRMS) +
ISSN: 2349-3771 +
+
Volume 3 Issue 4, pg: 164-169 +
A REVIEW ON TEXTURE BASED EMOTION RECOGNITION +
FROM FACIAL EXPRESSION +
1U.G. Scholars, 2Assistant Professor, +
Dept. of E & C Engg., MIT Moradabad, Ram Ganga Vihar, Phase II, Moradabad, India. +
2e1b1969ded4d63b69a5ec854350c0f74dc4de36
2b0ff4b82bac85c4f980c40b3dc4fde05d3cc23fAn Effective Approach for Facial Expression Recognition with Local Binary +
Pattern and Support Vector Machine +
2b3ceb40dced78a824cf67054959e250aeaa573b
2b1327a51412646fcf96aa16329f6f74b42aba89Under review as a conference paper at ICLR 2016 +
IMPROVING PERFORMANCE OF RECURRENT NEURAL +
NETWORK WITH RELU NONLINEARITY +
Qualcomm Research +
San Diego, CA 92121, USA +
2b5cb5466eecb131f06a8100dcaf0c7a0e30d391A Comparative Study of Active Appearance Model +
Annotation Schemes for the Face +
Face Aging Group +
UNCW, USA +
Face Aging Group +
UNCW, USA +
Face Aging Group +
UNCW, USA +
2b632f090c09435d089ff76220fd31fd314838aeEarly Adaptation of Deep Priors in Age Prediction from Face Images +
Computer Vision Lab +
D-ITET, ETH Zurich +
Computer Vision Lab +
D-ITET, ETH Zurich +
CVL, D-ITET, ETH Zurich +
Merantix GmbH +
2b8dfbd7cae8f412c6c943ab48c795514d53c4a7529 +
2014 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) +
978-1-4799-2893-4/14/$31.00 ©2014 IEEE +
RECOGNITION +
1. INTRODUCTION +
(d1,d2)∈[0;d]2 +
d1+d2≤d +
2baec98c19804bf19b480a9a0aa814078e28bb3d
470dbd3238b857f349ebf0efab0d2d6e9779073aUnsupervised Simultaneous Orthogonal Basis Clustering Feature Selection +
School of Electrical Engineering, KAIST, South Korea +
In this paper, we propose a novel unsupervised feature selection method: Si- +
multaneous Orthogonal basis Clustering Feature Selection (SOCFS). To per- +
form feature selection on unlabeled data effectively, a regularized regression- +
based formulation with a new type of target matrix is designed. The target +
matrix captures latent cluster centers of the projected data points by per- +
forming the orthogonal basis clustering, and then guides the projection ma- +
trix to select discriminative features. Unlike the recent unsupervised feature +
selection methods, SOCFS does not explicitly use the pre-computed local +
structure information for data points represented as additional terms of their +
objective functions, but directly computes latent cluster information by the +
target matrix conducting orthogonal basis clustering in a single unified term +
of the proposed objective function. +
Since the target matrix is put in a single unified term for regression of +
the proposed objective function, feature selection and clustering are simul- +
taneously performed. In this way, the projection matrix for feature selection +
is more properly computed by the estimated latent cluster centers of the +
projected data points. To the best of our knowledge, this is the first valid +
formulation to consider feature selection and clustering together in a sin- +
gle unified term of the objective function. The proposed objective function +
has fewer parameters to tune and does not require complicated optimization +
tools so just a simple optimization algorithm is sufficient. Substantial ex- +
periments are performed on several publicly available real world datasets, +
which shows that SOCFS outperforms various unsupervised feature selec- +
tion methods and that latent cluster information by the target matrix is ef- +
fective for regularized regression-based feature selection. +
Problem Formulation: Given training data, let X = [x1, . . . ,xn] ∈ Rd×n +
denote the data matrix with n instances where dimension is d and T = +
[t1, . . . ,tn] ∈ Rm×n denote the corresponding target matrix where dimension +
is m. We start from the regularized regression-based formulation to select +
maximum r features is minW (cid:107)WT X− T(cid:107)2 +
s.t. (cid:107)W(cid:107)2,0 ≤ r. To exploit +
such formulation on unlabeled data more effectively, it is crucial for the tar- +
get matrix T to have discriminative destinations for projected clusters. To +
this end, a new type of target matrix T is proposed to conduct clustering di- +
rectly on the projected data points WT X. We allow extra degrees of freedom +
to T by decomposing it into two other matrices B ∈ Rm×c and E ∈ Rn×c as +
T = BET with additional constraints as +
(1) +
F + λ(cid:107)W(cid:107)2,1 +
(cid:107)WT X− BET(cid:107)2 +
s.t. BT B = I, ET E = I, E ≥ 0, +
min +
W,B,E +
where λ > 0 is a weighting parameter for the relaxed regularizer (cid:107)W(cid:107)2,1 +
that induces row sparsity of the projection matrix W. The meanings of the +
constraints BT B = I,ET E = I,E ≥ 0 are as follows: 1) the orthogonal con- +
straint of B lets each column of B be independent; 2) the orthogonal and +
the nonnegative constraint of E make each row of E has only one non-zero +
element [2]. From 1) and 2), we can clearly interpret B as the basis matrix, +
which has orthogonality and E as the encoding matrix, where the non-zero +
element of each column of ET selects one column in B. +
While optimizing problem (1), T = BET acts like clustering of projected +
data points WT X with orthogonal basis B and encoder E, so T can estimate +
latent cluster centers of the WT X. Then, W successively projects X close +
to corresponding latent cluster centers, which are estimated by T. Note that +
the orthogonal constraint of B makes each projected cluster in WT X be sep- +
arated (independent of each other), and it helps W to be a better projection +
matrix for selecting more discriminative features. If the clustering is directly +
performed on X not on WT X, the orthogonal constraint of B extremely re- +
stricts the degree of freedom of B. However, since features are selected by +
W and the clustering is carried out on WT X in our formulation, so the or- +
thogonal constraint of B is highly reasonable. A schematic illustration of +
the proposed method is shown in Figure 1. +
47541d04ec24662c0be438531527323d983e958eAffective Information Processing +
474b461cd12c6d1a2fbd67184362631681defa9e2014 IEEE International +
Conference on Systems, Man +
and Cybernetics +
(SMC 2014) +
San Diego, California, USA +
5-8 October 2014 +
Pages 1-789 +
IEEE Catalog Number: +
ISBN: +
CFP14SMC-POD +
978-1-4799-3841-4 +
1/5 +
47d4838087a7ac2b995f3c5eba02ecdd2c28ba14JOURNAL OF IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, VOL. XX, NO. X, XXX 2017 +
Automatic Recognition of Facial Displays of +
Unfelt Emotions +
Escalera, Xavier Bar´o, Sylwia Hyniewska, Member, IEEE, J¨uri Allik, +
47a2727bd60e43f3253247b6d6f63faf2b67c54bSemi-supervised Vocabulary-informed Learning +
Disney Research +
47e3029a3d4cf0a9b0e96252c3dc1f646e750b14International Conference on Computer Systems and Technologies - CompSysTech’07 +
Facial Expression Recognition in still pictures and videos using Active +
Appearance Models. A comparison approach. +
Drago(cid:1) Datcu +
Léon Rothkrantz +
475e16577be1bfc0dd1f74f67bb651abd6d63524DAiSEE: Towards User Engagement Recognition in the Wild +
Microsoft +
Vineeth N Balasubramanian +
Indian Institution of Technology Hyderabad +
471befc1b5167fcfbf5280aa7f908eff0489c72b570 +
Class-Specific Kernel-Discriminant +
Analysis for Face Verification +
class problems ( +
47f8b3b3f249830b6e17888df4810f3d189daac1
47e8db3d9adb79a87c8c02b88f432f911eb45dc5MAGMA: Multi-level accelerated gradient mirror descent algorithm for +
large-scale convex composite minimization +
July 15, 2016 +
47aeb3b82f54b5ae8142b4bdda7b614433e69b9a
477811ff147f99b21e3c28309abff1304106dbbe
47e14fdc6685f0b3800f709c32e005068dfc8d47
782188821963304fb78791e01665590f0cd869e8
78a4cabf0afc94da123e299df5b32550cd638939
78f08cc9f845dc112f892a67e279a8366663e26dTECHNISCHE UNIVERSIT ¨AT M ¨UNCHEN +
Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +
Semi-Autonomous Data Enrichment and +
Optimisation for Intelligent Speech Analysis +
Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +
der Technischen Universit¨at M¨unchen zur Erlangung des akademischen Grades eines +
Doktor-Ingenieurs (Dr.-Ing.) +
genehmigten Dissertation. +
Vorsitzender: +
Univ.-Prof. Dr.-Ing. habil. Dr. h.c. Alexander W. Koch +
Pr¨ufer der Dissertation: +
1. +
Univ.-Prof. Dr.-Ing. habil. Bj¨orn W. Schuller, +
Universit¨at Passau +
2. Univ.-Prof. Gordon Cheng, Ph.D. +
Die Dissertation wurde am 30.09.2014 bei der Technischen Universit¨at M¨unchen einge- +
reicht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik am 07.04.2015 +
angenommen. +
783f3fccde99931bb900dce91357a6268afecc52Hindawi Publishing Corporation +
EURASIP Journal on Image and Video Processing +
Volume 2009, Article ID 945717, 14 pages +
doi:10.1155/2009/945717 +
Research Article +
Adapted Active Appearance Models +
1 SUP ´ELEC/IETR, Avenue de la Boulaie, 35511 Cesson-S´evign´e, France +
2 Orange Labs—TECH/IRIS, 4 rue du clos courtel, 35 512 Cesson S´evign´e, France +
Received 5 January 2009; Revised 2 September 2009; Accepted 20 October 2009 +
Recommended by Kenneth M. Lam +
Active Appearance Models (AAMs) are able to align efficiently known faces under duress, when face pose and illumination are +
controlled. We propose Adapted Active Appearance Models to align unknown faces in unknown poses and illuminations. Our +
proposal is based on the one hand on a specific transformation of the active model texture in an oriented map, which changes the +
AAM normalization process; on the other hand on the research made in a set of different precomputed models related to the most +
adapted AAM for an unknown face. Tests on public and private databases show the interest of our approach. It becomes possible +
to align unknown faces in real-time situations, in which light and pose are not controlled. +
Copyright © 2009 Renaud S´eguier et al. This is an open access article distributed under the Creative Commons Attribution +
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly +
cited. +
1. Introduction +
All applications related to face analysis and synthesis (Man- +
Machine Interaction, compression in video communication, +
augmented reality) need to detect and then to align the user’s +
face. This latest process consists in the precise localization of +
the eyes, nose, and mouth gravity center. Face detection can +
now be realized in real time and in a rather efficient manner +
[1, 2]; the technical bottleneck lies now in the face alignment +
when it is done in real conditions, which is precisely the +
object of this paper. +
Since such Active Appearance Models (AAMs) as those +
described in [3] exist, it is therefore possible to align faces +
in real time. The AAMs exploit a set of face examples in +
order to extract a statistical model. To align an unknown +
face in new image, the models parameters must be tuned, in +
order to match the analyzed face features in the best possible +
way. There is no difficulty to align a face featuring the same +
characteristics (same morphology, illumination, and pose) +
as those constituting the example data set. Unfortunately, +
AAMs are less outstanding when illumination, pose, and +
face type changes. We suggest in this paper a robust Active +
Appearance Model allowing a real-time implementation. In +
the next section, we will survey the different techniques, +
which aim to increase the AAM robustness. We will see +
that none of them address at the same time the three types +
of robustness, we are interested in pose, illumination, and +
identity. It must be pointed out that we do not consider the +
robustness against occlusion as [4] does, for example, when +
a person moves his hand around the face. +
After a quick introduction of the Active Appearance +
Models and their limitations (Section 3), we will present our +
two main contributions in Section 4.1 in order to improve +
AAM robustness in illumination, pose, and identity. Exper- +
iments will be conducted and discussed in Section 5 before +
drawing a conclusion, suggesting new research directions in +
the last section. +
2. State of the Art +
We propose to classify the methods which lead to an increase +
of the AAM robustness as follows. The specific types of +
dedicated robustness are in italic. +
(i) Preprocess +
(1) Invariant features (illumination) +
(2) Canonical representation (illumination) +
(ii) Parameter space extension +
(1) Light modeling (illumination) +
(2) 3D modeling (pose) +
7897c8a9361b427f7b07249d21eb9315db189496
78f438ed17f08bfe71dfb205ac447ce0561250c6
78a11b7d2d7e1b19d92d2afd51bd3624eca86c3c
781c2553c4ed2a3147bbf78ad57ef9d0aeb6c7edInt J Comput Vis +
DOI 10.1007/s11263-017-1023-9 +
Tubelets: Unsupervised Action Proposals from Spatiotemporal +
Super-Voxels +
Cees G. M. Snoek1 +
Received: 25 June 2016 / Accepted: 18 May 2017 +
© The Author(s) 2017. This article is an open access publication +
78df7d3fdd5c32f037fb5cc2a7c104ac1743d74eTEMPORAL PYRAMID POOLING CNN FOR ACTION RECOGNITION +
Temporal Pyramid Pooling Based Convolutional +
Neural Network for Action Recognition +
78fdf2b98cf6380623b0e20b0005a452e736181e
788a7b59ea72e23ef4f86dc9abb4450efefeca41
8b7191a2b8ab3ba97423b979da6ffc39cb53f46bSearch Pruning in Video Surveillance Systems: Efficiency-Reliability Tradeoff +
EURECOM +
Sophia Antipolis, France +
8b8728edc536020bc4871dc66b26a191f6658f7c
8b744786137cf6be766778344d9f13abf4ec0683978-1-4799-9988-0/16/$31.00 ©2016 IEEE +
2697 +
ICASSP 2016 +
8bf647fed40bdc9e35560021636dfb892a46720eLearning to Hash-tag Videos with Tag2Vec +
CVIT, KCIS, IIIT Hyderabad, India +
P J Narayanan +
http://cvit.iiit.ac.in/research/projects/tag2vec +
Figure 1. Learning a direct mapping from videos to hash-tags : sample frames from short video clips with user-given hash-tags +
(left); a sample frame from a query video and hash-tags suggested by our system for this query (right). +
8bb21b1f8d6952d77cae95b4e0b8964c9e0201b0Methoden +
at 11/2013 +
(cid:2)(cid:2)(cid:2) +
Multimodale Interaktion +
auf einer sozialen Roboterplattform +
Multimodal Interaction on a Social Robotic Platform +
Zusammenfassung Dieser Beitrag beschreibt die multimo- +
dalen Interaktionsmöglichkeiten mit der Forschungsroboter- +
plattform ELIAS. Zunächst wird ein Überblick über die Ro- +
boterplattform sowie die entwickelten Verarbeitungskompo- +
nenten gegeben, die Einteilung dieser Komponenten erfolgt +
nach dem Konzept von wahrnehmenden und agierenden Mo- +
dalitäten. Anschließend wird das Zusammenspiel der Kom- +
ponenten in einem multimodalen Spieleszenario näher be- +
trachtet. (cid:2)(cid:2)(cid:2) Summary +
This paper presents the mul- +
timodal +
interaction capabilities of the robotic research plat- +
form ELIAS. An overview of the robotic platform as well +
as the developed processing components is presented, the +
classification of the components follows the concept of sen- +
sing and acting modalities. Finally, +
the interplay between +
those components within a multimodal gaming scenario is +
described. +
Schlagwörter Mensch-Roboter-Interaktion, Multimodalität, Gesten, Blick (cid:2)(cid:2)(cid:2) Keywords Human-robot interaction, +
multimodal, gestures, gaze +
1 Einleitung +
Eine intuitive und natürliche Bedienbarkeit der zuneh- +
mend komplexeren Technik wird für den Menschen +
immer wichtiger, da im heutigen Alltag eine Vielzahl an +
technischen Geräten mit wachsendem Funktionsumfang +
anzutreffen ist. Unterschiedliche Aktivitäten in der For- +
schungsgemeinschaft haben sich schon seit längerer Zeit +
mit verbalen sowie nonverbalen Kommunikationsformen +
(bspw. Emotions- und Gestenerkennung) in der Mensch- +
Maschine-Interaktion beschäftigt. Gerade in der jüngeren +
Zeit trugen auf diesem Forschungsfeld unterschiedliche +
Innovationen (bspw. Touchscreen, Gestensteuerung im +
Fernseher) dazu bei, dass intuitive und natürliche Bedien- +
konzepte mehr und mehr im Alltag Verwendung finden. +
Auch Möglichkeiten zur Sprach- und Gestensteuerung +
von Konsolen und Mobiltelefonen finden heute vermehr- +
ten Einsatz in der Gerätebedienung. Diese natürlicheren +
und multimodalen Benutzerschnittstellen sind dem Nut- +
zer schnell zugänglich und erlauben eine intuitivere +
Interaktion mit komplexen technischen Geräten. +
Auch für Robotersysteme bietet sich eine multimodale +
Interaktion an, um die Benutzung und den Zugang zu +
den Funktionalitäten zu vereinfachen. Der Mensch soll +
in seiner Kommunikation idealerweise vollkommene Ent- +
scheidungsfreiheit bei der Wahl der Modalitäten haben, +
um sein gewünschtes Ziel zu erreichen. Dafür werden +
in diesem Beitrag die wahrnehmenden und agieren- +
den Modalitäten einer, rein auf Kommunikationsaspekte +
reduzierten, Forschungsroboterplattform beispielhaft in +
einer Spieleanwendung untersucht. +
1.1 Struktur des Beitrags +
In diesem Beitrag wird zunächst ein kurzer Über- +
blick über die multimodale Interaktion im Allgemeinen +
gegeben, hierbei erfolgt eine Betrachtung nach wahr- +
nehmenden und agierenden Modalitäten. Im nächsten +
Abschnitt werden Arbeiten vorgestellt, die sich auch mit +
multimodalen Robotersystemen beschäftigen. Im darauf +
folgenden Abschnitt wird die Roboterplattform ELIAS +
mit den wahrnehmenden, verarbeitenden und agierenden +
at – Automatisierungstechnik 61 (2013) 11 / DOI 10.1515/auto.2013.1062 © Oldenbourg Wissenschaftsverlag +
- 10.1515/auto.2013.1062 +
Downloaded from De Gruyter Online at 09/27/2016 10:08:34PM +
via Technische Universität München +
737 +
8b1db0894a23c4d6535b5adf28692f795559be90Biometric and Surveillance Technology for Human and Activity Identification X, edited by Ioannis Kakadiaris, +
Walter J. Scheirer, Laurence G. Hassebrook, Proc. of SPIE Vol. 8712, 87120Q · © 2013 SPIE +
CCC code: 0277-786X/13/$18 · doi: 10.1117/12.2018974 +
Proc. of SPIE Vol. 8712 87120Q-1 +
134db6ca13f808a848321d3998e4fe4cdc52fbc2IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS—PART B: CYBERNETICS, VOL. 36, NO. 2, APRIL 2006 +
433 +
Dynamics of Facial Expression: Recognition of +
Facial Actions and Their Temporal Segments +
From Face Profile Image Sequences +
133dd0f23e52c4e7bf254e8849ac6f8b17fcd22dThis 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 +
Active Clustering with Model-Based +
Uncertainty Reduction +
1369e9f174760ea592a94177dbcab9ed29be1649Geometrical Facial Modeling for Emotion Recognition +
133900a0e7450979c9491951a5f1c2a403a180f0JOURNAL OF LATEX CLASS FILES +
Social Grouping for Multi-target Tracking and +
Head Pose Estimation in Video +
13141284f1a7e1fe255f5c2b22c09e32f0a4d465Object Tracking by +
Oversampling Local Features +
133da0d8c7719a219537f4a11c915bf74c320da7International Journal of Computer Applications (0975 – 8887) +
Volume 123 – No.4, August 2015 +
A Novel Method for 3D Image Segmentation with Fusion +
of Two Images using Color K-means Algorithm +
Dept. of CSE +
ITM Universe +
Gwalior +
Dept. of CSE +
ITM Universe +
Gwalior +
two +
133f01aec1534604d184d56de866a4bd531dac87Effective Unconstrained Face Recognition by +
Combining Multiple Descriptors and Learned +
Background Statistics +
13841d54c55bd74964d877b4b517fa94650d9b65Generalised Ambient Reflection Models for Lambertian and +
Phong Surfaces +
Author +
Zhang, Paul, Gao, Yongsheng +
Published +
2009 +
Conference Title +
Proceedings of the 2009 IEEE International Conference on Image Processing (ICIP 2009) +
DOI +
https://doi.org/10.1109/ICIP.2009.5413812 +
Copyright Statement +
© 2009 IEEE. Personal use of this material is permitted. However, permission to reprint/ +
republish this material for advertising or promotional purposes or for creating new collective +
works for resale or redistribution to servers or lists, or to reuse any copyrighted component of +
this work in other works must be obtained from the IEEE. +
Downloaded from +
http://hdl.handle.net/10072/30001 +
Griffith Research Online +
https://research-repository.griffith.edu.au +
131e395c94999c55c53afead65d81be61cd349a4
1384a83e557b96883a6bffdb8433517ec52d0bea
13fd0a4d06f30a665fc0f6938cea6572f3b496f7
13afc4f8d08f766479577db2083f9632544c7ea6Multiple Kernel Learning for +
Emotion Recognition in the Wild +
Machine Perception Laboratory +
UCSD +
EmotiW Challenge, ICMI, 2013 +
1 +
13d9da779138af990d761ef84556e3e5c1e0eb94Int J Comput Vis (2008) 77: 3–24 +
DOI 10.1007/s11263-007-0093-5 +
Learning to Locate Informative Features for Visual Identification +
Received: 18 August 2005 / Accepted: 11 September 2007 / Published online: 9 November 2007 +
© Springer Science+Business Media, LLC 2007 +
7f533bd8f32525e2934a66a5b57d9143d7a89ee1Audio-Visual Identity Grounding for Enabling Cross Media Search +
Paper ID 22 +
7f44f8a5fd48b2d70cc2f344b4d1e7095f4f1fe5Int J Comput Vis (2016) 119:60–75 +
DOI 10.1007/s11263-015-0839-4 +
Sparse Output Coding for Scalable Visual Recognition +
Received: 15 May 2013 / Accepted: 16 June 2015 / Published online: 26 June 2015 +
© Springer Science+Business Media New York 2015 +
7f4bc8883c3b9872408cc391bcd294017848d0cf +
+
Computer +
Sciences +
Department +
The Multimodal Focused Attribute Model: A Nonparametric +
Bayesian Approach to Simultaneous Object Classification and +
Attribute Discovery +
Technical Report #1697 +
January 2012 +
+
7f6061c83dc36633911e4d726a497cdc1f31e58aYouTube-8M: A Large-Scale Video Classification +
Benchmark +
Paul Natsev +
Google Research +
7f36dd9ead29649ed389306790faf3b390dc0aa2MOVEMENT DIFFERENCES BETWEEN DELIBERATE +
AND SPONTANEOUS FACIAL EXPRESSIONS: +
ZYGOMATICUS MAJOR ACTION IN SMILING +
7f6cd03e3b7b63fca7170e317b3bb072ec9889e0A Face Recognition Signature Combining Patch-based +
Features with Soft Facial Attributes +
L. Zhang, P. Dou, I.A. Kakadiaris +
Computational Biomedicine Lab, 4849 Calhoun Rd, Rm 373, Houston, TX 77204 +
7f3a73babe733520112c0199ff8d26ddfc7038a0
7f205b9fca7e66ac80758c4d6caabe148deb8581Page 1 of 47 +
Computing Surveys +
A Survey on Mobile Social Signal Processing +
Understanding human behaviour in an automatic but non-intrusive manner is an important area for various applications. This requires the +
collaboration of information technology with human sciences to transfer existing knowledge of human behaviour into self-acting tools. These +
tools will reduce human error that is introduced by current obtrusive methods such as questionnaires. To achieve unobtrusiveness, we focus on +
exploiting the pervasive and ubiquitous character of mobile devices. +
In this article, a survey of existing techniques for extracting social behaviour through mobile devices is provided. Initially we expose the +
terminology used in the area and introduce a concrete architecture for social signal processing applications on mobile phones, constituted by +
sensing, social interaction detection, behavioural cues extraction, social signal inference and social behaviour understanding. Furthermore, we +
present state-of-the-art techniques applied to each stage of the process. Finally, potential applications are shown while arguing about the main +
challenges of the area. +
Categories and Subject Descriptors: General and reference [Document Types]: Surveys and Overviews; Human-centered computing [Collab- +
orative and social computing, Ubiquitous and mobile computing] +
General Terms: Design, Theory, Human Factors, Performance +
Additional Key Words and Phrases: Social Signal Processing, mobile phones, social behaviour +
ACM Reference Format: +
Processing. ACM V, N, Article A (January YYYY), 35 pages. +
DOI:http://dx.doi.org/10.1145/0000000.0000000 +
1. INTRODUCTION +
Human behaviour understanding has received a great deal of interest since the beginning of the previous century. +
People initially conducted research on the way animals behave when they are surrounded by creatures of the same +
species. Acquiring basic underlying knowledge of animal relations led to extending this information to humans +
in order to understand social behaviour, social relations etc. Initial experiments were conducted by empirically +
observing people and retrieving feedback from them. These methods gave rise to well-established psychological +
approaches for understanding human behaviour, such as surveys, questionnaires, camera recordings and human +
observers. Nevertheless, these methods introduce several limitations including various sources of error. Complet- +
ing surveys and questionnaires induces partiality, unconcern etc. [Groves 2004], human error [Reason 1990], and +
additional restrictions in scalability of the experiments. Accumulating these research problems leads to a common +
challenge, the lack of automation in an unobtrusive manner. +
An area that has focussed on detecting social behaviour automatically and has received a great amount of at- +
tention is Social Signal Processing (SSP). The main target of the field is to model, analyse and synthesise human +
behaviour with limited user intervention. To achieve these targets, researchers presented three key terms which +
7a9ef21a7f59a47ce53b1dff2dd49a8289bb5098
7af38f6dcfbe1cd89f2307776bcaa09c54c30a8beaig i C e Vii ad Beyd: +
Devee +
h . Weg +
Deae f C e Sciece +
ichiga Sae Uiveiy +
Ea aig  48824 +
Abac +
Thi chae id ce wha i caed he deveea aach  c e vii i +
aic a ad ai(cid:12)cia ieigece i geea.  dic e he c e baic aadig f de +
veig a ye ad i f daea iiai. The deveea aach i ivaed +
by h a cgiive devee f ifacy  ad hd. A deveea eaig ag +
ih i deeied befe he \bih" f he ye. Afe he \bih" i eabe he ye +
 ea ew ak wih  a eed f egaig. The aj ga f he deveea +
aach i  eaize a ai f geea  e eaig ha eabe achie  ef +
deveea eaig ve a g eid. S ch eaig i cd ced i a de iia  he +
way aia ad h a ea. The achie   ea diecy f ci   ey i +
  ea whie ieacig wih he evie ic dig h a eache.  hi eaig +
de deveig ieige ga f vai  ak i eaized h gh ea ie ieac +
7a81967598c2c0b3b3771c1af943efb1defd4482Do We Need More Training Data? +
7ad77b6e727795a12fdacd1f328f4f904471233fSupervised Local Descriptor Learning +
for Human Action Recognition +
7a97de9460d679efa5a5b4c6f0b0a5ef68b56b3b
7aa4c16a8e1481629f16167dea313fe9256abb42978-1-5090-4117-6/17/$31.00 ©2017 IEEE +
2981 +
ICASSP 2017 +
7a85b3ab0efb6b6fcb034ce13145156ee9d10598
7ab930146f4b5946ec59459f8473c700bcc89233
7ad7897740e701eae455457ea74ac10f8b307bedRandom Subspace Two-dimensional LDA for Face Recognition* +
7a7b1352d97913ba7b5d9318d4c3d0d53d6fb697Attend and Rectify: a Gated Attention +
Mechanism for Fine-Grained Recovery +
†Computer Vision Center and Universitat Aut`onoma de Barcelona (UAB), +
Campus UAB, 08193 Bellaterra, Catalonia Spain +
‡Visual Tagging Services, Parc de Recerca, Campus UAB +
1451e7b11e66c86104f9391b80d9fb422fb11c01IET Signal Processing +
Research Article +
Image privacy protection with secure JPEG +
transmorphing +
ISSN 1751-9675 +
Received on 30th December 2016 +
Revised 13th July 2017 +
Accepted on 11th August 2017 +
doi: 10.1049/iet-spr.2016.0756 +
www.ietdl.org +
1Multimedia Signal Processing Group, Electrical Engineering Department, EPFL, Station 11, Lausanne, Switzerland +
14761b89152aa1fc280a33ea4d77b723df4e3864
14fa27234fa2112014eda23da16af606db7f3637
1459d4d16088379c3748322ab0835f50300d9a38Cross-Domain Visual Matching via Generalized +
Similarity Measure and Feature Learning +
14e949f5754f9e5160e8bfa3f1364dd92c2bb8d6
1450296fb936d666f2f11454cc8f0108e2306741Learning to Discover Cross-Domain Relations +
with Generative Adversarial Networks +
14fdce01c958043140e3af0a7f274517b235adf3
141eab5f7e164e4ef40dd7bc19df9c31bd200c5e
14e759cb019aaf812d6ac049fde54f40c4ed1468Subspace Methods +
Synonyms +
{ Multiple similarity method +
Related Concepts +
{ Principal component analysis (PCA) +
{ Subspace analysis +
{ Dimensionality reduction +
De(cid:12)nition +
Subspace analysis in computer vision is a generic name to describe a general +
framework for comparison and classification of subspaces. A typical approach in +
subspace analysis is the subspace method (SM) that classify an input pattern +
vector into several classes based on the minimum distance or angle between the +
input pattern vector and each class subspace, where a class subspace corresponds +
to the distribution of pattern vectors of the class in high dimensional vector +
space. +
Background +
Comparison and classification of subspaces has been one of the central prob- +
lems in computer vision, where an image set of an object to be classified is +
compactly represented by a subspace in high dimensional vector space. +
The subspace method is one of the most effective classification method in +
subspace analysis, which was developed by two Japanese researchers, Watanabe +
and Iijima around 1970, independently [1, 2]. Watanabe and Iijima named their +
methods the CLAFIC [3] and the multiple similarity method [4], respectively. +
The concept of the subspace method is derived from the observation that pat- +
terns belonging to a class forms a compact cluster in high dimensional vector +
space, where, for example, a w×h pixels image pattern is usually represented as a +
vector in w×h-dimensional vector space. The compact cluster can be represented +
by a subspace, which is generated by using Karhunen-Lo`eve (KL) expansion, also +
known as the principal component analysis (PCA). Note that a subspace is gen- +
erated for each class, unlike the Eigenface Method [5] in which only one subspace +
(called eigenspace) is generated. +
The SM has been known as one of the most useful methods in pattern recog- +
nition field, since its algorithm is very simple and it can handle classification +
of multiple classes. However, its classification performance was not sufficient for +
many applications in practice, because class subspaces are generated indepen- +
dently of each other [1]. There is no reason to assume a priori that each class +
148eb413bede35487198ce7851997bf8721ea2d6People Search in Surveillance Videos +
Four Eyes Lab, UCSB +
IBM Research +
IBM Research +
IBM Research +
Four Eyes Lab, UCSB +
INTRODUCTION +
1. +
In traditional surveillance scenarios, users are required to +
watch video footage corresponding to extended periods of +
time in order to find events of interest. However, this pro- +
cess is resource-consuming, and suffers from high costs of +
employing security personnel. The field of intelligent vi- +
sual surveillance [2] seeks to address these issues by applying +
computer vision techniques to automatically detect specific +
events in long video streams. The events can then be pre- +
sented to the user or be indexed into a database to allow +
queries such as “show me the red cars that entered a given +
parking lot from 7pm to 9pm on Monday” or “show me the +
faces of people who left the city’s train station last week.” +
In this work, we are interested in analyzing people, by ex- +
tracting information that can be used to search for them in +
surveillance videos. Current research on this topic focuses +
on approaches based on face recognition, where the goal is +
to establish the identity of a person given an image of a +
face. However, face recognition is still a very challenging +
problem, especially in low resolution images with variations +
in pose and lighting, which is often the case in surveillance +
data. State-of-the-art face recognition systems [1] require +
a fair amount of resolution in order to produce reliable re- +
sults, but in many cases this level of detail is not available +
in surveillance applications. +
We approach the problem in an alternative way, by avoiding +
face recognition and proposing a framework for finding peo- +
ple based on parsing the human body and exploiting part +
attributes. Those include visual attributes such as facial hair +
type (beards, mustaches, absence of facial hair), type of eye- +
wear (sunglasses, eyeglasses, absence of glasses), hair type +
(baldness, hair, wearing a hat), and clothing color. While +
face recognition is still a difficult problem, accurate and ef- +
ficient face detectors1 based on learning approaches [6] are +
available. Those have been demonstrated to work well on +
challenging low-resolution images, with variations in pose +
and lighting. In our method, we employ this technology to +
design detectors for facial attributes from large sets of train- +
ing data. +
1The face detection problem consists of localizing faces in +
images, while face recognition aims to establish the identity +
of a person given an image of a face. Face detection is a +
challenging problem, but it is arguably not as complex as +
face recognition. +
Our technique falls into the category of short term recogni- +
tion methods, taking advantage of features present in brief +
intervals in time, such as clothing color, hairstyle, and makeup, +
which are generally considered an annoyance in face recogni- +
tion methods. There are several applications that naturally +
fit within a short term recognition framework. An example +
is in criminal investigation, when the police are interested in +
locating a suspect. In those cases, eyewitnesses typically fill +
out a suspect description form, where they indicate personal +
traits of the suspect as seen at the moment when the crime +
was committed. Those include facial hair type, hair color, +
clothing type, etc. Based on that description, the police +
manually scan the entire video archive looking for a person +
with similar characteristics. This process is tedious and time +
consuming, and could be drastically accelerated by the use +
of our technique. Another application is on finding missing +
people. Parents looking for their children in an amusement +
park could provide a description including clothing and eye- +
wear type, and videos from multiple cameras in the park +
would then be automatically searched. +
1473a233465ea664031d985e10e21de927314c94
140c95e53c619eac594d70f6369f518adfea12efPushing the Frontiers of Unconstrained Face Detection and Recognition: IARPA Janus Benchmark A +
The development of accurate and scalable unconstrained face recogni- +
tion algorithms is a long term goal of the biometrics and computer vision +
communities. The term “unconstrained” implies a system can perform suc- +
cessful identifications regardless of face image capture presentation (illumi- +
nation, sensor, compression) or subject conditions (facial pose, expression, +
occlusion). While automatic, as well as human, face identification in certain +
scenarios may forever be elusive, such as when a face is heavily occluded or +
captured at very low resolutions, there still remains a large gap between au- +
tomated systems and human performance on familiar faces. In order to close +
this gap, large annotated sets of imagery are needed that are representative +
of the end goals of unconstrained face recognition. This will help continue +
to push the frontiers of unconstrained face detection and recognition, which +
are the primary goals of the IARPA Janus program. +
The current state of the art in unconstrained face recognition is high +
accuracy (roughly 99% true accept rate at a false accept rate of 1.0%) on +
faces that can be detected with a commodity face detectors, but unknown +
accuracy on other faces. Despite the fact that face detection and recognition +
research generally has advanced somewhat independently, the frontal face +
detector filtering approach used for key in the wild face recognition datasets +
means that progress in face recognition is currently hampered by progress +
in face detection. Hence, a major need exists for a face recognition dataset +
that captures as wide of a range of variations as possible to offer challenges +
to both face detection as well as face recognition. +
In this paper we introduce the IARPA Janus Benchmark A (IJB-A), +
which is publicly available for download. The IJB-A contains images and +
videos from 500 subjects captured from “in the wild” environment. All la- +
belled subjects have been manually localized with bounding boxes for face +
detection, as well as fiducial landmarks for the center of the two eyes (if +
visible) and base of the nose. Manual bounding box annotations for all non- +
labelled subjects (i.e., other persons captured in the imagery) have been cap- +
tured as well. All imagery is Creative Commons licensed, which is a license +
that allows open re-distribution provided proper attribution is made to the +
data creator. The subjects have been intentionally sampled to contain wider +
geographic distribution than previous datasets. Recognition and detection +
protocols are provided which are motivated by operational deployments of +
face recognition systems. An example of images and video from IJB-A can +
be found in Figure 3. +
The IJB-A dataset has the following claimed contributions: (i) The most +
unconstrained database released to date; (ii) The first joint face detection and +
face recognition benchmark dataset collected in the wild; (iii) Meta-data +
providing subject gender and skin color, and occlusion (eyes, mouth/nose, +
and forehead), facial hear, and coarse pose information for each imagery +
instance; (iv) Widest geographic distribution of any public face dataset; (v) +
The first in the wild dataset to contain a mixture of images and videos; (vi) +
Clear authority for re-distribution; (vii) Protocols for identification (search) +
and verification (compare); (viii) Baseline accuracies from off the shelf de- +
tectors and recognition algorithms; and (ix) Protocols for both template and +
model-based face recognition. +
Every subject in the dataset contains at least five images and one video. +
IJB-A consists of a total of 5,712 images and 2,085 videos, with an average +
of 11.4 images and 4.2 videos per subject. +
142dcfc3c62b1f30a13f1f49c608be3e62033042Adaptive Region Pooling for Object Detection +
UC Merced +
Qualcomm Research, San Diego +
UC Merced +
14e428f2ff3dc5cf96e5742eedb156c1ea12ece1Facial Expression Recognition Using Neural Network Trained with Zernike +
Moments +
Dept. Génie-Electrique +
Université M.C.M Souk-Ahras +
Souk-Ahras, Algeria +
14a5feadd4209d21fa308e7a942967ea7c13b7b6978-1-4673-0046-9/12/$26.00 ©2012 IEEE +
1025 +
ICASSP 2012 +
14fee990a372bcc4cb6dc024ab7fc4ecf09dba2bModeling Spatio-Temporal Human Track Structure for Action +
Localization +
14ee4948be56caeb30aa3b94968ce663e7496ce4Jang, Y; Gunes, H; Patras, I +
© Copyright 2018 IEEE +
For additional information about this publication click this link. +
http://qmro.qmul.ac.uk/xmlui/handle/123456789/36405 +
Information about this research object was correct at the time of download; we occasionally +
make corrections to records, please therefore check the published record when citing. For +
8ee62f7d59aa949b4a943453824e03f4ce19e500Robust Head-Pose Estimation Based on +
Partially-Latent Mixture of Linear Regression +
∗INRIA Grenoble Rhˆone-Alpes, Montbonnot Saint-Martin, France +
†INRIA Rennes Bretagne Atlantique, Rennes, France +
8e33183a0ed7141aa4fa9d87ef3be334727c76c0– COS429 Written Report, Fall 2017 – +
Robustness of Face Recognition to Image Manipulations +
1. Motivation +
We can often recognize pictures of people we know even if the image has low resolution or obscures +
part of the face, if the camera angle resulted in a distorted image of the subject’s face, or if the +
subject has aged or put on makeup since we last saw them. Although this is a simple recognition task +
for a human, when we think about how we accomplish this task, it seems non-trivial for computer +
algorithms to recognize faces despite visual changes. +
Computer facial recognition is relied upon for many application where accuracy is important. +
Facial recognition systems have applications ranging from airport security and suspect identification +
to personal device authentication and face tagging [7]. In these real-world applications, the system +
must continue to recognize images of a person who looks slightly different due to the passage of +
time, a change in environment, or a difference in clothing. +
Therefore, we are interested in investigating face recognition algorithms and their robustness to +
image changes resulting from realistically plausible manipulations. Furthermore, we are curious +
about whether the impact of image manipulations on computer algorithms’ face recognition ability +
mirrors related insights from neuroscience about humans’ face recognition abilities. +
2. Goal +
In this project, we implement both face recognition algorithms and image manipulations. We then +
analyze the impact of each image manipulation on the recognition accuracy each algorithm, and +
how these influences depend on the accuracy of each algorithm on non-manipulated images. +
3. Background and Related Work +
Researchers have developed a wide variety of face recognition algorithms, such as traditional +
statistical methods such as PCA, more opaque methods such as deep neural networks, and proprietary +
systems used by governments and corporations [1][13][14]. +
Similarly, others have developed image manipulations using principles from linear algebra, such +
as mimicking distortions from lens distortions, as well as using neural networks, such as a system +
for transforming images according to specified characteristics [12][16]. +
Furthermore, researchers in psychology have studied face recognition in humans. A study of +
“super-recognizers” (people with extraordinarily high powers of face recognition) and “developmen- +
tal prosopagnosics” (people with severely impaired face recognition abilities) found that inverting +
images of faces impaired recognition ability more for people with stronger face recognition abilities +
[11]. This could indicate that image manipulations tend to equalize face recognition abilities, and +
we investigate whether this is the case with the manipulations and face recognition algorithms we +
test. +
8e3d0b401dec8818cd0245c540c6bc032f169a1dMcGan: Mean and Covariance Feature Matching GAN +
8e94ed0d7606408a0833e69c3185d6dcbe22bbbe© 2012 IEEE. Personal use of this material is permitted. Permission from IEEE +
must be obtained for all other uses, in any current or future media, including +
reprinting/republishing this material for advertising or promotional purposes, +
creating new collective works, for resale or redistribution to servers or lists, or +
reuse of any copyrighted component of this work in other works. +
Pre-print of article that will appear at WACV 2012. +
8e461978359b056d1b4770508e7a567dbed49776LOMo: Latent Ordinal Model for Facial Analysis in Videos +
Marian Bartlett1,∗,‡ +
1UCSD, USA +
2MPI for Informatics, Germany +
3IIT Kanpur, India +
8ea30ade85880b94b74b56a9bac013585cb4c34bFROM TURBO HIDDEN MARKOV MODELS TO TURBO STATE-SPACE MODELS +
Institut Eur´ecom +
Multimedia Communications Department +
BP 193, 06904 Sophia Antipolis Cedex, France +
8ed32c8fad924736ebc6d99c5c319312ba1fa80b
8e8e3f2e66494b9b6782fb9e3f52aeb8e1b0d125in any current or +
future media, +
for all other uses, +
 2012 IEEE. Personal use of this material is permitted. Permission from IEEE must be +
obtained +
including +
reprinting/republishing this material for advertising or promotional purposes, creating +
new collective works, for resale or redistribution to servers or lists, or reuse of any +
copyrighted component of this work in other works. +
Pre-print of article that will appear at BTAS 2012.!! +
8e378ef01171b33c59c17ff5798f30293fe30686Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +
der Technischen Universit¨at M¨unchen +
A System for Automatic Face Analysis +
Based on +
Statistical Shape and Texture Models +
Ronald M¨uller +
Vollst¨andiger Abdruck der von der Fakult¨at +
f¨ur Elektrotechnik und Informationstechnik +
der Technischen Universit¨at M¨unchen +
zur Erlangung des akademischen Grades eines +
Doktor-Ingenieurs +
genehmigten Dissertation +
Vorsitzender: Prof. Dr. rer. nat. Bernhard Wolf +
Pr¨ufer der Dissertation: +
1. Prof. Dr.-Ing. habil. Gerhard Rigoll +
2. Prof. Dr.-Ing. habil. Alexander W. Koch +
Die Dissertation wurde am 28.02.2008 bei der Technischen Universit¨at M¨unchen +
eingereicht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik +
am 18.09.2008 angenommen. +
8ed051be31309a71b75e584bc812b71a0344a019Class-based feature matching across unrestricted +
transformations +
8e36100cb144685c26e46ad034c524b830b8b2f2Modeling Facial Geometry using Compositional VAEs +
1 ´Ecole Polytechnique F´ed´erale de Lausanne +
2Facebook Reality Labs, Pittsburgh +
8e0becfc5fe3ecdd2ac93fabe34634827b21ef2bInternational Journal of Computer Vision manuscript No. +
(will be inserted by the editor) +
Learning from Longitudinal Face Demonstration - +
Where Tractable Deep Modeling Meets Inverse Reinforcement Learning +
Savvides · Tien D. Bui +
Received: date / Accepted: date +
225fb9181545f8750061c7693661b62d715dc542
22043cbd2b70cb8195d8d0500460ddc00ddb1a62Separability-Oriented Subclass Discriminant +
Analysis +
22137ce9c01a8fdebf92ef35407a5a5d18730dde
22dada4a7ba85625824489375184ba1c3f7f0c8f
223ec77652c268b98c298327d42aacea8f3ce23fTR-CS-11-02 +
Acted Facial Expressions In The Wild +
Database +
September 2011 +
ANU Computer Science Technical Report Series +
227b18fab568472bf14f9665cedfb95ed33e5fceCompositional Dictionaries for Domain Adaptive +
Face Recognition +
227b1a09b942eaf130d1d84cdcabf98921780a22Yang et al. EURASIP Journal on Advances in Signal Processing (2018) 2018:51 +
https://doi.org/10.1186/s13634-018-0572-6 +
EURASIP Journal on Advances +
in Signal Processing +
R ES EAR CH +
Multi-feature shape regression for face +
alignment +
Open Access +
22dabd4f092e7f3bdaf352edd925ecc59821e168 Deakin Research Online +
This is the published version: +
An, Senjian, Liu, Wanquan and Venkatesh, Svetha 2008, Exploiting side information in +
locality preserving projection, in CVPR 2008 : Proceedings of the 26th IEEE Conference on +
Computer Vision and Pattern Recognition, IEEE, Washington, D. C., pp. 1-8. +
Available from Deakin Research Online: +
http://hdl.handle.net/10536/DRO/DU:30044576 +
+
Reproduced with the kind permissions of the copyright owner. +
Personal use of this material is permitted. However, permission to reprint/republish this +
material for advertising or promotional purposes or for creating new collective works for +
resale or redistribution to servers or lists, or to reuse any copyrighted component of this work +
in other works must be obtained from the IEEE. +
Copyright : 2008, IEEE +
22f656d0f8426c84a33a267977f511f127bfd7f3
2271d554787fdad561fafc6e9f742eea94d35518TECHNISCHE UNIVERSIT ¨AT M ¨UNCHEN +
Lehrstuhl f¨ur Mensch-Maschine-Kommunikation +
Multimodale Mensch-Roboter-Interaktion +
f¨ur Ambient Assisted Living +
Tobias F. Rehrl +
Vollst¨andiger Abdruck der von der Fakult¨at f¨ur Elektrotechnik und Informationstechnik +
der Technischen Universit¨at M¨unchen zur Erlangung des akademischen Grades eines +
Doktor-Ingenieurs (Dr.-Ing.) +
genehmigten Dissertation. +
Vorsitzende: +
Pr¨ufer der Dissertation: 1. Univ.-Prof. Dr.-Ing. habil. Gerhard Rigoll +
2. Univ.-Prof. Dr.-Ing. Horst-Michael Groß +
Univ.-Prof. Dr.-Ing. Sandra Hirche +
(Technische Universit¨at Ilmenau) +
Die Dissertation wurde am 17. April 2013 bei der Technischen Universit¨at M¨unchen +
eingereicht und durch die Fakult¨at f¨ur Elektrotechnik und Informationstechnik am +
8. Oktober 2013 angenommen. +
22ec256400e53cee35f999244fb9ba6ba11c1d06
22a7f1aebdb57eecd64be2a1f03aef25f9b0e9a7
22e189a813529a8f43ad76b318207d9a4b6de71aWhat will Happen Next? +
Forecasting Player Moves in Sports Videos +
UC Berkeley, STATS +
UC Berkeley +
UC Berkeley +
25d514d26ecbc147becf4117512523412e1f060bAnnotated Crowd Video Face Database +
IIIT-Delhi, India +
25c19d8c85462b3b0926820ee5a92fc55b81c35aNoname manuscript No. +
(will be inserted by the editor) +
Pose-Invariant Facial Expression Recognition +
Using Variable-Intensity Templates +
Received: date / Accepted: date +
258a8c6710a9b0c2dc3818333ec035730062b1a5Benelearn 2005 +
Annual Machine Learning Conference of +
Belgium and the Netherlands +
CTIT PROCEEDINGS OF THE FOURTEENTH +
ANNUAL MACHINE LEARNING CONFERENCE +
OF BELGIUM AND THE NETHERLANDS +
25695abfe51209798f3b68fb42cfad7a96356f1fAN INVESTIGATION INTO COMBINING +
BOTH FACIAL DETECTION AND +
LANDMARK LOCALISATION INTO A +
UNIFIED PROCEDURE USING GPU +
COMPUTING +
MSc by Research +
2016 +
25d3e122fec578a14226dc7c007fb1f05ddf97f7The First Facial Expression Recognition and Analysis Challenge +
2597b0dccdf3d89eaffd32e202570b1fbbedd1d6Towards predicting the likeability of fashion images +
25982e2bef817ebde7be5bb80b22a9864b979fb0
25e05a1ea19d5baf5e642c2a43cca19c5cbb60f8Label Distribution Learning +
2559b15f8d4a57694a0a33bdc4ac95c479a3c79a570 +
Contextual Object Localization With Multiple +
Kernel Nearest Neighbor +
Gert Lanckriet, Member, IEEE +
2574860616d7ffa653eb002bbaca53686bc71cdd
25f1f195c0efd84c221b62d1256a8625cb4b450c1-4244-1017-7/07/$25.00 ©2007 IEEE +
1091 +
ICME 2007 +
25885e9292957feb89dcb4a30e77218ffe7b9868JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2016 +
Analyzing the Affect of a Group of People Using +
Multi-modal Framework +
259706f1fd85e2e900e757d2656ca289363e74aaImproving People Search Using Query Expansions +
How Friends Help To Find People +
LEAR - INRIA Rhˆone Alpes - Grenoble, France +
25728e08b0ee482ee6ced79c74d4735bb5478e29
258a2dad71cb47c71f408fa0611a4864532f5ebaDiscriminative Optimization +
of Local Features for Face Recognition +
+
H O S S E I N A Z I Z P O U R +
+
Master of Science Thesis +
Stockholm, Sweden 2011 +
+
25127c2d9f14d36f03d200a65de8446f6a0e3bd6Journal of Theoretical and Applied Information Technology +
20th May 2016. Vol.87. No.2 +
© 2005 - 2016 JATIT & LLS. All rights reserved. +
ISSN: 1992-8645 www.jatit.org E-ISSN: 1817-3195 +
EVALUATING THE PERFORMANCE OF DEEP SUPERVISED +
AUTO ENCODER IN SINGLE SAMPLE FACE RECOGNITION +
PROBLEM USING KULLBACK-LEIBLER DIVERGENCE +
SPARSITY REGULARIZER +
Faculty of Computer of Computer Science, Universitas Indonesia, Kampus UI Depok, Indonesia +
\ No newline at end of file -- cgit v1.2.3-70-g09d2