Merge branch 'master' of asdf.us:megapixels_dev

1 files changed, 87 insertions, 9 deletions

diff --git a/site/public/datasets/helen/index.html b/site/public/datasets/helen/index.html
index 44ef462e..08791d29 100644
--- a/site/public/datasets/helen/index.html
+++ b/site/public/datasets/helen/index.html
@@ -4,7 +4,7 @@
   <title>MegaPixels: HELEN</title>
   <meta charset="utf-8" />
   <meta name="author" content="Adam Harvey" />
-  <meta name="description" content="HELEN Face Dataset" />
+  <meta name="description" content="HELEN is a dataset of face images from Flickr used for training facial component localization algorithms" />
   <meta property="og:title" content="MegaPixels: HELEN"/>
   <meta property="og:type" content="website"/>
   <meta property="og:summary" content="MegaPixels is an art and research project about face recognition datasets created \"in the wild\"/>
@@ -55,8 +55,7 @@
   </header>
   <div class="content content-dataset">
     
-  <section class='intro_section' style='background-image: url(https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/background.jpg)'><div class='inner'><div class='hero_desc'><span class='bgpad'>HELEN Face Dataset</span></div><div class='hero_subdesc'><span class='bgpad'>HELEN (under development)
-</span></div></div></section><section><h2>HELEN</h2>
+  <section class='intro_section' style='background-image: url(https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/background.jpg)'></section><section><div class='image'><div class='intro-caption caption'>Example images from the HELEN dataset</div></div></section><section><h1>HELEN Dataset</h1>
 </section><section><div class='right-sidebar'><div class='meta'>
     <div class='gray'>Published</div>
     <div>2012</div>
@@ -69,8 +68,74 @@
   </div><div class='meta'>
     <div class='gray'>Website</div>
     <div><a href='http://www.ifp.illinois.edu/~vuongle2/helen/' target='_blank' rel='nofollow noopener'>illinois.edu</a></div>
-  </div></div><p>[ page under development ]</p>
-</section><section>
+  </div></div><p>Helen is a dataset of annotated face images used for facial component localization. It includes 2,330 images from Flickr found by searching for "portrait" combined with terms such as "family", "wedding", "boy", "outdoor", and "studio".<a class="footnote_shim" name="[^orig_paper]_1"> </a><a href="#[^orig_paper]" class="footnote" title="Footnote 1">1</a></p>
+<p>The dataset was published in 2012 with the primary motivation listed as facilitating "high quality editing of portraits". However, the paper's introduction also mentions that facial feature localization "is an essential component for face recognition, tracking and expression analysis."<a class="footnote_shim" name="[^orig_paper]_2"> </a><a href="#[^orig_paper]" class="footnote" title="Footnote 1">1</a></p>
+<p>Irregardless of the authors' primary motivations, the HELEN dataset has become one of the most widely used datasets for training facial landmark algorithms, which are essential parts of most facial recogntion processing systems. Facial landmarking are used to isolate facial features such as the eyes, nose, jawline, and mouth in order to align faces to match a templated pose.</p>
+</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/montage_lms_21_14_14_14_26.png' alt=' An example annotation from the HELEN dataset showing 194 points that were originally annotated by Mechanical Turk workers. Graphic &copy; 2019 MegaPixels.cc based on data from HELEN dataset by  Le, Vuong et al.'><div class='caption'> An example annotation from the HELEN dataset showing 194 points that were originally annotated by Mechanical Turk workers. Graphic &copy; 2019 MegaPixels.cc based on data from HELEN dataset by  Le, Vuong et al.</div></div></section><section><p>This analysis shows that since its initial publication in 2012, the HELEN dataset has been used in over 200 research projects related to facial recognition with the vast majority of research taking place in China.</p>
+<p>Commercial use includes IBM, NVIDIA, NEC, Microsoft Research Asia, Google, Megvii, Microsoft, Intel, Daimler, Tencent, Baidu, Adobe, Facebook</p>
+<p>Military and Defense Usage includes NUDT</p>
+<p><a href="http://eccv2012.unifi.it/">http://eccv2012.unifi.it/</a></p>
+<p>TODO</p>
+<ul>
+<li>add proof of use in dlib and openface</li>
+<li>add proof of use in commercial use of dlib? ibm dif</li>
+<li>make landmark over blurred images</li>
+<li>add 6x6 gride for landmarks</li>
+<li>highlight key findings</li>
+<li>highlight key commercial usage</li>
+<li>look for most interesting research papers to provide example of how it's used for face recognition</li>
+<li>estimated time: 6 hours</li>
+<li>add data to github repo?</li>
+</ul>
+<table>
+<thead><tr>
+<th>Organization</th>
+<th>Paper</th>
+<th>Link</th>
+<th>Year</th>
+<th>Used Duke MTMC</th>
+</tr>
+</thead>
+<tbody>
+<tr>
+<td>SenseTime, Amazon</td>
+<td><a href="https://arxiv.org/pdf/1805.10483.pdf">Look at Boundary: A Boundary-Aware Face Alignment Algorithm</a></td>
+</tr>
+<tr>
+<td>2018</td>
+<td>year</td>
+<td>&#x2714;</td>
+</tr>
+<tr>
+<td>SenseTime</td>
+<td><a href="https://arxiv.org/pdf/1807.11079.pdf">ReenactGAN: Learning to Reenact Faces via Boundary Transfer</a></td>
+<td>2018</td>
+<td>year</td>
+<td>&#x2714;</td>
+</tr>
+</tbody>
+</table>
+<p>The dataset was used for training the OpenFace software "we used the HELEN and LFPW training subsets for training and the rest for testing" <a href="https://github.com/TadasBaltrusaitis/OpenFace/wiki/Datasets">https://github.com/TadasBaltrusaitis/OpenFace/wiki/Datasets</a></p>
+<p>The popular dlib facial landmark detector was trained using HELEN</p>
+<p>In addition to the 200+ verified citations, the HELEN dataset was used for</p>
+<ul>
+<li><a href="https://github.com/memoiry/face-alignment">https://github.com/memoiry/face-alignment</a></li>
+<li><a href="http://www.dsp.toronto.edu/projects/face_analysis/">http://www.dsp.toronto.edu/projects/face_analysis/</a></li>
+</ul>
+<p>It's been converted into new datasets including</p>
+<ul>
+<li><a href="https://github.com/JPlin/Relabeled-HELEN-Dataset">https://github.com/JPlin/Relabeled-HELEN-Dataset</a></li>
+<li><a href="https://www.kaggle.com/kmader/helen-eye-dataset">https://www.kaggle.com/kmader/helen-eye-dataset</a></li>
+</ul>
+<p>The original site</p>
+<ul>
+<li><a href="http://www.ifp.illinois.edu/~vuongle2/helen/">http://www.ifp.illinois.edu/~vuongle2/helen/</a></li>
+</ul>
+<h3>Example Images</h3>
+</section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/feature_outdoor_02.jpg' alt=' An image from the HELEN dataset "wedding" category used for training face recognition  2839127417_1.jpg for outdoor studio'><div class='caption'> An image from the HELEN dataset "wedding" category used for training face recognition  2839127417_1.jpg for outdoor studio</div></div>
+<div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/feature_graduation.jpg' alt=' An image from the HELEN dataset "wedding" category used for training face recognition 2325274893_1 '><div class='caption'> An image from the HELEN dataset "wedding" category used for training face recognition 2325274893_1 </div></div></section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/feature_wedding.jpg' alt=' An image from the HELEN dataset "wedding" category used for training face recognition 2325274893_1 '><div class='caption'> An image from the HELEN dataset "wedding" category used for training face recognition 2325274893_1 </div></div>
+<div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/feature_wedding_02.jpg' alt=' An image from the HELEN dataset "wedding" category used for training face recognition 2325274893_1 '><div class='caption'> An image from the HELEN dataset "wedding" category used for training face recognition 2325274893_1 </div></div></section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/feature_family.jpg' alt=' Original Flickr image used in HELEN facial analysis and recognition dataset for the keyword "family". 296814969'><div class='caption'> Original Flickr image used in HELEN facial analysis and recognition dataset for the keyword "family". 296814969</div></div>
+<div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/feature_family_05.jpg' alt=' Original Flickr image used in HELEN facial analysis and recognition dataset for the keyword "family". 296814969'><div class='caption'> Original Flickr image used in HELEN facial analysis and recognition dataset for the keyword "family". 296814969</div></div></section><section>
   <h3>Who used Helen Dataset?</h3>
 
   <p>
@@ -91,10 +156,10 @@
 
 <section>
 	
-	<h3>Information Supply chain</h3>
+	<h3>Information Supply Chain</h3>
 
 	<p>
-		To help understand how Helen Dataset has been used around the world by commercial, military, and academic organizations; existing publicly available research citing Helen Dataset was collected, verified, and geocoded to show the biometric trade routes of people appearing in the images. Click on the markers to reveal research projects at that location.
+		To help understand how Helen Dataset has been used around the world by commercial, military, and academic organizations; existing publicly available research citing Helen Dataset was collected, verified, and geocoded to show how AI training data has proliferated around the world. Click on the markers to reveal research projects at that location.
 	</p>
  
  </section>
@@ -109,7 +174,7 @@
 	<li class="com">Commercial</li>
 	<li class="gov">Military / Government</li>
 	</ul>
-	<div class="source">Citation data is collected using <a href="https://semanticscholar.org" target="_blank">SemanticScholar.org</a> then dataset usage verified and geolocated.</div >
+	<div class="source">Citation data is collected using SemanticScholar.org then dataset usage verified and geolocated. Citations are used to provide overview of how and where images were used.</div>
 </div>
 
 
@@ -130,7 +195,10 @@
 
   <h2>Supplementary Information</h2>
   
+</section><section><h3>Age and Gender Distribution</h3>
 </section><section>
+	<p>Age and gender estimation distribution were calculated by anlayzing all faces in the dataset images. This may include additional faces appearing next to an annotated face, or this may skip false faces that were erroneously included as part of the original dataset. These numbers are provided as an estimation and not a factual representation of the exact gender and age of all faces.</p>
+</section><section><div class='columns columns-2'><section class='applet_container'><div class='applet' data-payload='{"command": "single_pie_chart /datasets/helen/assets/age.csv", "fields": ["Caption: HELEN dataset age distribution", "Top: 10", "OtherLabel: Other"]}'></div></section><section class='applet_container'><div class='applet' data-payload='{"command": "single_pie_chart /datasets/helen/assets/gender.csv", "fields": ["Caption: HELEN dataset gender distribution", "Top: 10", "OtherLabel: Other"]}'></div></section></div></section><section class='images'><div class='image'><img src='https://nyc3.digitaloceanspaces.com/megapixels/v1/datasets/helen/assets/montage_lms_21_15_15_7_26_0.png' alt=' Visualization of the HELEN dataset 194-point facial landmark annotations. Credit: graphic &copy; MegaPixels.cc 2019, data from HELEN dataset by Zhou, Brand, Lin 2013. If you use this image please credit both the graphic and data source.'><div class='caption'> Visualization of the HELEN dataset 194-point facial landmark annotations. Credit: graphic &copy; MegaPixels.cc 2019, data from HELEN dataset by Zhou, Brand, Lin 2013. If you use this image please credit both the graphic and data source.</div></div></section><section>
 
   <h4>Cite Our Work</h4>
   <p>
@@ -147,7 +215,17 @@
 }</pre>
 
 	</p>
-</section>
+</section><section><h4>Cite the Original Author's Work</h4>
+<p>If you find the HELEN dataset useful or reference it in your work, please cite the author's original work as:</p>
+<pre>
+@inproceedings{Le2012InteractiveFF,
+ title={Interactive Facial Feature Localization},
+ author={Vuong Le and Jonathan Brandt and Zhe L. Lin and Lubomir D. Bourdev and Thomas S. Huang},
+ booktitle={ECCV},
+ year={2012}
+}
+</pre></section><section><h3>References</h3><section><ul class="footnotes"><li>1 <a name="[^orig_paper]" class="footnote_shim"></a><span class="backlinks"><a href="#[^orig_paper]_1">a</a><a href="#[^orig_paper]_2">b</a></span>Le, Vuong et al. “Interactive Facial Feature Localization.” ECCV (2012).
+</li></ul></section></section>
 
   </div>
   <footer>