1
2
3
4
5
6
7
8
9
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
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
|
<!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>
|
