diff options
| -rw-r--r-- | notes/utils/image_utils.md | 7 | ||||
| -rw-r--r-- | site/content/pages/datasets/lfw/index.md | 22 | ||||
| -rw-r--r-- | site/content/pages/datasets/youtube_celebrities/index.md | 40 | ||||
| -rw-r--r-- | site/public/about/style/index.html | 85 | ||||
| -rw-r--r-- | site/public/datasets/lfw/what/index.html | 142 | ||||
| -rw-r--r-- | site/public/research/from_1_to_100_pixels/index.html | 101 |
6 files changed, 65 insertions, 332 deletions
diff --git a/notes/utils/image_utils.md b/notes/utils/image_utils.md index 3ccc2e78..d8a466f9 100644 --- a/notes/utils/image_utils.md +++ b/notes/utils/image_utils.md @@ -112,6 +112,10 @@ Get the size of a PSD `convert website-design.psd -print "Size: %wx%h\n" /dev/null` +Convert all images to RGB + +`mogrify -resize 256x256! -colorspace sRGB -type truecolor /data_store_hdd/datasets/people/lfw/progan/faces_256_full/*.jpg` + ## Animate GIFs With ImageMagick @@ -185,10 +189,11 @@ convert -define jpeg:size=200x200 original.jpeg -thumbnail 100x100^ -gravity ce Resize all directories to square crops for Caffe ``` find . -name "*.jpg" -print0 | xargs -0 mogrify -resize 256x256^ -gravity Center -crop 256x256+0+0 +repage - ``` +Move all images in subdirectories to another directory: +`find . -name "*.jpg" -print0 | xargs -0 mv -t new_location/` Replace spaces in filenames diff --git a/site/content/pages/datasets/lfw/index.md b/site/content/pages/datasets/lfw/index.md index f52b1be6..e85c7556 100644 --- a/site/content/pages/datasets/lfw/index.md +++ b/site/content/pages/datasets/lfw/index.md @@ -16,7 +16,7 @@ authors: Adam Harvey + Images: 13,233 + Identities: 5,749 + Origin: Yahoo News Images -+ Funding: TBD ++ Funding: (Possibly, partially CIA*)  @@ -67,8 +67,19 @@ Browse or download the geocoded citation data collected for the LFW dataset. - The faces in the LFW dataset were detected using the Viola-Jones haarcascade face detector [^lfw_website] [^lfw-survey] - The LFW dataset is used by several of the largest tech companies in the world including "Google, Facebook, Microsoft Research Asia, Baidu, Tencent, SenseTime, Face++ and Chinese University of Hong Kong." [^lfw_pingan] - All images in the LFW dataset were copied from Yahoo News between 2002 - 2004 +<<<<<<< HEAD - In 2014, two of the four original authors of the LFW dataset received funding from IARPA and ODNI for their follow up paper [Labeled Faces in the Wild: Updates and New Reporting Procedures](https://www.semanticscholar.org/paper/Labeled-Faces-in-the-Wild-%3A-Updates-and-New-Huang-Learned-Miller/2d3482dcff69c7417c7b933f22de606a0e8e42d4) via IARPA contract number 2014-14071600010 - The dataset includes 2 images of [George Tenet](http://vis-www.cs.umass.edu/lfw/person/George_Tenet.html), the former Director of Central Intelligence (DCI) for the Central Intelligence Agency whose facial biometrics were eventually used to help train facial recognition software in China and Russia +======= +- In 2014, 2/4 of the original authors of the LFW dataset received funding from IARPA and ODNI for their follow up paper "Labeled Faces in the Wild: Updates and New Reporting Procedures" via IARPA contract number 2014-14071600010 +- The LFW dataset was used Center for Intelligent Information Retrieval, the Central Intelligence Agency, the National Security Agency and National + +TODO (need citations for the following) + +- SenseTime, who has relied on LFW for benchmarking their facial recognition performance, is one the leading provider of surveillance to the Chinese Government [need citation for this fact. is it the most? or is that Tencent?] +- Two out of 4 of the original authors received funding from the Office of Director of National Intelligence and IARPA for their 2016 LFW survey follow up report + +>>>>>>> 13d7a450affe8ea4f368a97ea2014faa17702a4c   @@ -137,9 +148,14 @@ Ignore text below these lines Research -> This research is based upon work supported in part by the Office of the Director of National Intelligence (ODNI), Intelligence Advanced Research Projects Activity (IARPA), via contract number 2014-14071600010. +- "In our experiments, we used 10000 images and associated captions from the Faces in the wilddata set [3]." +- "Ths work was supported in part by the Center for Intelligent Information Retrieval, the Central Intelligence Agency, the National Security Agency and National Science Foundation under CAREER award IIS-0546666 and grant IIS-0326249." +- From: "People-LDA: Anchoring Topics to People using Face Recognition" +- This paper was presented at IEEE 11th ICCV conference Oct 14-21 but the main LFW paper "Labeled Faces in the Wild: A Database forStudying Face Recognition in Unconstrained Environments" was not published until 2008 +- 10f17534dba06af1ddab96c4188a9c98a020a459 -"Labeled Faces in the Wild: Updates and New Reporting Procedures" +- This research is based upon work supported in part by the Office of the Director of National Intelligence (ODNI), Intelligence Advanced Research Projects Activity (IARPA), via contract number 2014-14071600010. +- From "Labeled Faces in the Wild: Updates and New Reporting Procedures" [^lfw_www]: <http://vis-www.cs.umass.edu/lfw/results.html> [^lfw_baidu]: Jingtuo Liu, Yafeng Deng, Tao Bai, Zhengping Wei, Chang Huang. Targeting Ultimate Accuracy: Face Recognition via Deep Embedding. <https://arxiv.org/abs/1506.07310> diff --git a/site/content/pages/datasets/youtube_celebrities/index.md b/site/content/pages/datasets/youtube_celebrities/index.md new file mode 100644 index 00000000..f5a7128d --- /dev/null +++ b/site/content/pages/datasets/youtube_celebrities/index.md @@ -0,0 +1,40 @@ +------------ + +status: published +title: YouTube Celebrities +desc: YouTube Celebrities +slug: lfw +published: 2019-2-23 +updated: 2019-2-23 +authors: Adam Harvey + +------------ + +# YouTube Celebrities + ++ Years: TBD ++ Images: TBD ++ Identities: TBD ++ Origin: YouTube.com ++ Funded by: CIA, US Army + + +TODO + + +-------- + +RESEARCH below these lines + +-------- + +> Selected dataset sequences: (a) MBGC, (b) CMU MoBo, (c) First +Honda/UCSD, and (d) YouTube Celebrities. +> This research is supported by the Central Intelligence Agency, the Biometrics +Task Force and the Technical Support Working Group through US Army contract +W91CRB-08-C-0093. The opinions, (cid:12)ndings, and conclusions or recommendations +expressed in this publication are those of the authors and do not necessarily re(cid:13)ect +the views of our sponsors. + +- in "Face Recognition From Video Draft 17" +- International Journal of Pattern Recognition and Artifcial Intelligence WorldScientific Publishing Company
\ No newline at end of file diff --git a/site/public/about/style/index.html b/site/public/about/style/index.html deleted file mode 100644 index da0d718f..00000000 --- a/site/public/about/style/index.html +++ /dev/null @@ -1,85 +0,0 @@ -<!doctype html> -<html> -<head> - <title>MegaPixels</title> - <meta charset="utf-8" /> - <meta name="author" content="Adam Harvey" /> - <meta name="description" content="MegaPixels Style" /> - <meta name="referrer" content="no-referrer" /> - <meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes" /> - <link rel='stylesheet' href='/assets/css/fonts.css' /> - <link rel='stylesheet' href='/assets/css/tabulator.css' /> - <link rel='stylesheet' href='/assets/css/css.css' /> - <link rel='stylesheet' href='/assets/css/applets.css' /> -</head> -<body> - <header> - <a class='slogan' href="/"> - <div class='logo'></div> - <div class='site_name'>MegaPixels</div> - <span class='sub'>The Darkside of Datasets</span> - </a> - <div class='links'> - <a href="/datasets/">Datasets</a> - <a href="/research/">Research</a> - <a href="/about/">About</a> - </div> - </header> - <div class="content"> - - <section><h1>Style Examples</h1> -</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/site/about/assets/test.jpg' alt='Alt text here'><div class='caption'>Alt text here</div></div></section><section><div class='meta'><div><div class='gray'>Date</div><div>17-Jan-2019</div></div><div><div class='gray'>Numbers</div><div>17</div></div><div><div class='gray'>Identities</div><div>12,139</div></div><div><div class='gray'>But also</div><div>This is a test of the stylesheet</div></div></div></section><section><h1>Header 1</h1> -<h2>Header 2</h2> -<h3>Header 3</h3> -<h4>Header 4</h4> -<h5>Header 5</h5> -<h6>Header 6</h6> -<p><strong>Bold text</strong>, <em>italic text</em>, <strong><em>bold italic text</em></strong></p> -<p>At vero eos et et iusto qui blanditiis <a href="#">praesentium voluptatum</a> deleniti atque corrupti[^1], quos dolores et quas molestias excepturi sint, obcaecati cupiditate non-provident, similique sunt in culpa, qui officia deserunt mollitia animi, id est laborum et dolorum fuga. Et harum quidem rerum facilis est et expedita distinctio[^2]. Nam libero tempore, cum soluta nobis est eligendi optio, cumque nihil impedit, quo minus id, quod maxime placeat, facere possimus, omnis voluptas assumenda est, omnis dolor repellendus[^3].</p> -<ul> -<li>Sed ut perspiciatis, unde omnis iste natus error sit voluptatem accusantium doloremque laudantium</li> -<li>Totam rem aperiam eaque ipsa, quae ab illo inventore veritatis et quasi architecto beatae vitae dicta sunt, explicabo</li> -<li>Nemo enim ipsam voluptatem, quia voluptas sit, aspernatur aut</li> -<li>Odit aut fugit, sed quia consequuntur magni dolores eos</li> -<li>Qui ratione voluptatem sequi nesciunt, neque porro quisquam </li> -</ul> -<h2>single image test</h2> -</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/site/about/assets/man.jpg' alt='This person is alone'><div class='caption'>This person is alone</div></div></section><section><h2>double image test</h2> -</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/site/about/assets/man.jpg' alt='This person is on the left'><div class='caption'>This person is on the left</div></div> -<div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/site/about/assets/man.jpg' alt='This person is on the right'><div class='caption'>This person is on the right</div></div></section><section><h2>triple image test</h2> -</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/site/about/assets/man.jpg' alt='Person 1'><div class='caption'>Person 1</div></div> -<div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/site/about/assets/man.jpg' alt='Person 2'><div class='caption'>Person 2</div></div> -<div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/site/about/assets/man.jpg' alt='Person 3. Let me tell you about Person 3. This person has a very long description with text which wraps like crazy'><div class='caption'>Person 3. Let me tell you about Person 3. This person has a very long description with text which wraps like crazy</div></div></section><section><blockquote><p>est, qui dolorem ipsum, quia dolor sit amet consectetur adipisci[ng] velit, sed quia non-numquam [do] eius modi tempora inci[di]dunt, ut labore et dolore magnam aliquam quaerat voluptatem.</p> -</blockquote> -</section><section class='wide'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/site/about/assets/wide-test.jpg' alt='This image is extremely wide and the text beneath it will wrap but thats fine because it can also contain <a href="https://example.com/">hyperlinks</a>! Yes, you read that right—hyperlinks! Lorem ipsum dolor sit amet ad volotesque sic hoc ad nauseam'><div class='caption'>This image is extremely wide and the text beneath it will wrap but that's fine because it can also contain <a href="https://example.com/">hyperlinks</a>! Yes, you read that right—hyperlinks! Lorem ipsum dolor sit amet ad volotesque sic hoc ad nauseam</div></div></section><section><p>Inline <code>code</code> has <code>back-ticks around</code> it.</p> -</section><section><div class='applet' data-payload='{"command": "javascript"}'></div></section><section><pre><code class="lang-python">s = "Python syntax highlighting" -print(s) -</code></pre> -</section><section><div class='applet' data-payload='{"command": "No language indicated, so no syntax highlighting. "}'></div></section><section><p>Horizontal rule</p> -<hr> -<p>Citations below here</p> -<div class="footnotes"> -<hr> -<ol></ol> -</div> -</section> - - </div> - <footer> - <div> - <a href="/">MegaPixels.cc</a> - <a href="/about/disclaimer/">Disclaimer</a> - <a href="/about/terms/">Terms of Use</a> - <a href="/about/privacy/">Privacy</a> - <a href="/about/">About</a> - <a href="/about/team/">Team</a> - </div> - <div> - MegaPixels ©2017-19 Adam R. Harvey / - <a href="https://ahprojects.com">ahprojects.com</a> - </div> - </footer> -</body> - -<script src="/assets/js/dist/index.js"></script> -</html>
\ No newline at end of file diff --git a/site/public/datasets/lfw/what/index.html b/site/public/datasets/lfw/what/index.html deleted file mode 100644 index ceafb35a..00000000 --- a/site/public/datasets/lfw/what/index.html +++ /dev/null @@ -1,142 +0,0 @@ -<!doctype html> -<html> -<head> - <title>MegaPixels</title> - <meta charset="utf-8" /> - <meta name="author" content="Adam Harvey" /> - <meta name="description" content="LFW: Labeled Faces in The Wild" /> - <meta name="referrer" content="no-referrer" /> - <meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes" /> - <link rel='stylesheet' href='/assets/css/fonts.css' /> - <link rel='stylesheet' href='/assets/css/css.css' /> -</head> -<body> - <header> - <a class='slogan' href="/"> - <div class='logo'></div> - <div class='site_name'>MegaPixels</div> - <span class='sub'>The Darkside of Datasets</span> - </a> - <div class='links'> - <a href="/search/">Face Search</a> - <a href="/datasets/">Datasets</a> - <a href="/research/">Research</a> - <a href="/about/">About</a> - </div> - </header> - <div class="content"> - - <section><h1>Labeled Faces in The Wild</h1> -<ul> -<li>Created 2007 (auto)</li> -<li>Images 13,233 (auto)</li> -<li>People 5,749 (auto)</li> -<li>Created From Yahoo News images (auto)</li> -<li>Analyzed and searchable (auto)</li> -</ul> -<p><em>Labeled Faces in The Wild</em> is amongst the most widely used facial recognition training datasets in the world and is the first facial recognition dataset [^lfw_names_faces] of its kind to be created entirely from Internet photos. It includes 13,233 images of 5,749 people that appeared on Yahoo News between 2002 - 2004.</p> -</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/lfw/assets/lfw_grid_preview.jpg' alt='Eight out of 5,749 people in the Labeled Faces in the Wild dataset. The face recognition training dataset is created entirely from photos downloaded from the Internet.'><div class='caption'>Eight out of 5,749 people in the Labeled Faces in the Wild dataset. The face recognition training dataset is created entirely from photos downloaded from the Internet.</div></div></section><section><h2>INTRO</h2> -<p>It began in 2002. Researchers at University of Massachusetts Amherst were developing algorithms for facial recognition and they needed more data. Between 2002-2004 they scraped Yahoo News for images of public figures. Two years later they cleaned up the dataset and repackaged it as Labeled Faces in the Wild (LFW).</p> -<p>Since then the LFW dataset has become one of the most widely used datasets used for evaluating face recognition algorithms. The associated research paper “Labeled Faces in the Wild: A Database for Studying Face Recognition in Unconstrained Environments” has been cited 996 times reaching 45 different countries throughout the world.</p> -<p>The faces come from news stories and are mostly celebrities from the entertainment industry, politicians, and villains. It’s a sampling of current affairs and breaking news that has come to pass. The images, detached from their original context now server a new purpose: to train, evaluate, and improve facial recognition.</p> -<p>As the most widely used facial recognition dataset, it can be said that each individual in LFW has, in a small way, contributed to the current state of the art in facial recognition surveillance. John Cusack, Julianne Moore, Barry Bonds, Osama bin Laden, and even Moby are amongst these biometric pillars, exemplar faces provided the visual dimensions of a new computer vision future.</p> -</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/lfw/assets/lfw_a_to_c.jpg' alt='From Aaron Eckhart to Zydrunas Ilgauskas. A small sampling of the LFW dataset'><div class='caption'>From Aaron Eckhart to Zydrunas Ilgauskas. A small sampling of the LFW dataset</div></div></section><section><p>In addition to commercial use as an evaluation tool, all of the faces in LFW dataset are prepackaged into a popular machine learning code framework called scikit-learn.</p> -<h2>Usage</h2> -<pre><code class="lang-python">#!/usr/bin/python -from matplotlib import plt -from sklearn.datasets import fetch_lfw_people -lfw_people = fetch_lfw_people() -lfw_person = lfw_people[0] -plt.imshow(lfw_person) -</code></pre> -<h2>Commercial Use</h2> -<p>The LFW dataset is used by numerous companies for benchmarking algorithms and in some cases training. According to the benchmarking results page [^lfw_results] provided by the authors, over 2 dozen companies have contributed their benchmark results</p> -<pre><code>load file: lfw_commercial_use.csv -name_display,company_url,example_url,country,description -</code></pre> -<table> -<thead><tr> -<th style="text-align:left">Company</th> -<th style="text-align:left">Country</th> -<th style="text-align:left">Industries</th> -</tr> -</thead> -<tbody> -<tr> -<td style="text-align:left"><a href="http://www.aratek.co">Aratek</a></td> -<td style="text-align:left">China</td> -<td style="text-align:left">Biometric sensors for telecom, civil identification, finance, education, POS, and transportation</td> -</tr> -<tr> -<td style="text-align:left"><a href="http://www.aratek.co">Aratek</a></td> -<td style="text-align:left">China</td> -<td style="text-align:left">Biometric sensors for telecom, civil identification, finance, education, POS, and transportation</td> -</tr> -<tr> -<td style="text-align:left"><a href="http://www.aratek.co">Aratek</a></td> -<td style="text-align:left">China</td> -<td style="text-align:left">Biometric sensors for telecom, civil identification, finance, education, POS, and transportation</td> -</tr> -</tbody> -</table> -<p>Add 2-4 screenshots of companies mentioning LFW here</p> -</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/lfw/assets/lfw_screenshot_01.png' alt='ReadSense'><div class='caption'>ReadSense</div></div></section><section><p>In benchmarking, companies use a dataset to evaluate their algorithms which are typically trained on other data. After training, researchers will use LFW as a benchmark to compare results with other algorithms.</p> -<p>For example, Baidu (est. net worth $13B) uses LFW to report results for their "Targeting Ultimate Accuracy: Face Recognition via Deep Embedding". According to the three Baidu researchers who produced the paper:</p> -<blockquote><p>LFW has been the most popular evaluation benchmark for face recognition, and played a very important role in facilitating the face recognition society to improve algorithm. <sup class="footnote-ref" id="fnref-baidu_lfw"><a href="#fn-baidu_lfw">1</a></sup>.</p> -</blockquote> -<h2>Citations</h2> -<table> -<thead><tr> -<th style="text-align:left">Title</th> -<th style="text-align:left">Organization</th> -<th style="text-align:left">Country</th> -<th style="text-align:left">Type</th> -</tr> -</thead> -<tbody> -<tr> -<td style="text-align:left">3D-aided face recognition from videos</td> -<td style="text-align:left">University of Lyon</td> -<td style="text-align:left">France</td> -<td style="text-align:left">edu</td> -</tr> -<tr> -<td style="text-align:left">A Community Detection Approach to Cleaning Extremely Large Face Database</td> -<td style="text-align:left">National University of Defense Technology, China</td> -<td style="text-align:left">China</td> -<td style="text-align:left">edu</td> -</tr> -</tbody> -</table> -<h2>Conclusion</h2> -<p>The LFW face recognition training and evaluation dataset is a historically important face dataset as it was the first popular dataset to be created entirely from Internet images, paving the way for a global trend towards downloading anyone’s face from the Internet and adding it to a dataset. As will be evident with other datasets, LFW’s approach has now become the norm.</p> -<p>For all the 5,000 people in this datasets, their face is forever a part of facial recognition history. It would be impossible to remove anyone from the dataset because it is so ubiquitous. For their rest of the lives and forever after, these 5,000 people will continue to be used for training facial recognition surveillance.</p> -<h2>Notes</h2> -<p>According to BiometricUpdate.com<sup class="footnote-ref" id="fnref-biometric_update_lfw"><a href="#fn-biometric_update_lfw">2</a></sup>, LFW is "the most widely used evaluation set in the field of facial recognition, LFW attracts a few dozen teams from around the globe including Google, Facebook, Microsoft Research Asia, Baidu, Tencent, SenseTime, Face++ and Chinese University of Hong Kong."</p> -<div class="footnotes"> -<hr> -<ol><li id="fn-baidu_lfw"><p>"Chinese tourist town uses face recognition as an entry pass". New Scientist. November 17, 2016. <a href="https://www.newscientist.com/article/2113176-chinese-tourist-town-uses-face-recognition-as-an-entry-pass/">https://www.newscientist.com/article/2113176-chinese-tourist-town-uses-face-recognition-as-an-entry-pass/</a><a href="#fnref-baidu_lfw" class="footnote">↩</a></p></li> -<li id="fn-biometric_update_lfw"><p>"PING AN Tech facial recognition receives high score in latest LFW test results". <a href="https://www.biometricupdate.com/201702/ping-an-tech-facial-recognition-receives-high-score-in-latest-lfw-test-results">https://www.biometricupdate.com/201702/ping-an-tech-facial-recognition-receives-high-score-in-latest-lfw-test-results</a><a href="#fnref-biometric_update_lfw" class="footnote">↩</a></p></li> -</ol> -</div> -</section> - - </div> - <footer> - <div> - <a href="/">MegaPixels.cc</a> - <a href="/about/disclaimer/">Disclaimer</a> - <a href="/about/terms/">Terms of Use</a> - <a href="/about/privacy/">Privacy</a> - <a href="/about/">About</a> - <a href="/about/team/">Team</a> - </div> - <div> - MegaPixels ©2017-19 Adam R. Harvey / - <a href="https://ahprojects.com">ahprojects.com</a> - </div> - </footer> -</body> - -<script src="/assets/js/app/site.js"></script> -</html>
\ No newline at end of file diff --git a/site/public/research/from_1_to_100_pixels/index.html b/site/public/research/from_1_to_100_pixels/index.html deleted file mode 100644 index 751e885b..00000000 --- a/site/public/research/from_1_to_100_pixels/index.html +++ /dev/null @@ -1,101 +0,0 @@ -<!doctype html> -<html> -<head> - <title>MegaPixels</title> - <meta charset="utf-8" /> - <meta name="author" content="Adam Harvey<br> Berit Gilma<br> Matthew Stender" /> - <meta name="description" content="High resolution insights from low resolution imagery" /> - <meta name="referrer" content="no-referrer" /> - <meta name="viewport" content="width=device-width, initial-scale=1.0, user-scalable=yes" /> - <link rel='stylesheet' href='/assets/css/fonts.css' /> - <link rel='stylesheet' href='/assets/css/css.css' /> -</head> -<body> - <header> - <a class='slogan' href="/"> - <div class='logo'></div> - <div class='site_name'>MegaPixels</div> - <span class='sub'>The Darkside of Datasets</span> - </a> - <div class='links'> - <a href="/search">Face Search</a> - <a href="/datasets">Datasets</a> - <a href="/research/from_1_to_100_pixels/">Research</a> - <a href="/about">About</a> - </div> - </header> - <div class="content"> - - <section> - <h1>From 1 to 100 Pixels</h1> - <div class='meta'> - <div> - <div class='gray'>Posted</div> - <div>2018-12-04</div> - </div> - <div> - <div class='gray'>By</div> - <div>Adam Harvey<br> Berit Gilma<br> Matthew Stender</div> - </div> - - </div> - </section> - - <section><h2>High resolution insights from low resolution data</h2> -<p>This post will be about the meaning of "face". How do people define it? How to biometrics researchers define it? How has it changed during the last decade.</p> -<p>What can you know from a very small amount of information?</p> -<ul> -<li>1 pixel grayscale</li> -<li>2x2 pixels grayscale, font example</li> -<li>4x4 pixels</li> -<li>8x8 yotta yotta</li> -<li>5x7 face recognition</li> -<li>12x16 activity recognition</li> -<li>6/5 (up to 124/106) pixels in height/width, and the average is 24/20 for QMUL SurvFace</li> -<li>20x16 tiny faces paper</li> -<li>20x20 MNIST handwritten images <a href="http://yann.lecun.com/exdb/mnist/">http://yann.lecun.com/exdb/mnist/</a></li> -<li>24x24 haarcascade detector idealized images</li> -<li>32x32 CIFAR image dataset</li> -<li>40x40 can do emotion detection, face recognition at scale, 3d modeling of the face. include datasets with faces at this resolution including pedestrian.</li> -<li>need more material from 60-100</li> -<li>60x60 show how texture emerges and pupils, eye color, higher resolution of features and compare to lower resolution faces</li> -<li>100x100 0.5% of one Instagram photo</li> -</ul> -<p>Find specific cases of facial resolution being used in legal cases, forensic investigations, or military footage</p> -<p>Research</p> -<ul> -<li>NIST report on sres states several resolutions</li> -<li>"Results show that the tested face recognition systems yielded similar performance for query sets with eye-to-eye distance from 60 pixels to 30 pixels" <sup class="footnote-ref" id="fnref-nist_sres"><a href="#fn-nist_sres">1</a></sup></li> -</ul> -<div class="footnotes"> -<hr> -<ol><li id="fn-nist_sres"><p>NIST 906932. Performance Assessment of Face Recognition Using Super-Resolution. Shuowen Hu, Robert Maschal, S. Susan Young, Tsai Hong Hong, Jonathon P. Phillips<a href="#fnref-nist_sres" class="footnote">↩</a></p></li> -</ol> -</div> -</section> - - <section> - <h3>MORE RESEARCH</h3> - <div class='blogposts'> - - </div> - </section> - - </div> - <footer> - <div> - <a href="/">MegaPixels.cc</a> - <a href="/about/disclaimer/">Disclaimer</a> - <a href="/about/terms/">Terms of Use</a> - <a href="/about/privacy/">Privacy</a> - <a href="/about/">About</a> - <a href="/about/team/">Team</a> - </div> - <div> - MegaPixels ©2017-19 Adam R. Harvey / - <a href="https://ahprojects.com">ahprojects.com</a> - </div> - </footer> -</body> -<script src="/assets/js/app/site.js"></script> -</html>
\ No newline at end of file |