b3do/index.html at master · UCBAIR/b3do · GitHub

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
<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN"
   "http://www.w3.org/TR/html4/strict.dtd">

<html lang="en">
  <head>
    <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
    <title>Berkeley 3-D Object Dataset</title>

    <!-- Framework CSS -->
    <link rel="stylesheet" href="blueprint/screen.css" type="text/css" media="screen, projection">
    <link rel="stylesheet" href="blueprint/print.css" type="text/css" media="print">
    <!--[if lt IE 8]><link rel="stylesheet" href="blueprint/ie.css" type="text/css" media="screen, projection"><![endif]-->

    <!-- Import fancy-type plugin for the sample page. -->
    <link rel="stylesheet" href="blueprint/plugins/fancy-type/screen.css" type="text/css" media="screen, projection">
    <script type="text/javascript">

      var _gaq = _gaq || [];
      _gaq.push(['_setAccount', 'UA-23543539-1']);
      _gaq.push(['_trackPageview']);

      (function() {
        var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true;
        ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js';
        var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s);
      })();

    </script>
  </head>
  <body>
    <div class="container">
      <div class="span-24">
        &nbsp;
        <h1>B3DO: Berkeley 3-D Object Dataset</h1>
      </div>
      <hr>
      <h2 class="alt">A quality depth sensor, the Microsoft Kinect, is now in millions of homes. Yet robust household object detection is still not a reality. To get there, we are collecting a massive, crowd-sourced, and challenging 3-D object dataset.</h2>
      <hr>
      <div class="span-10 colborder">
         <p>
            <img width="400px" src="images/example.png" alt-="Example of object detection" />
         </p>
      </div>
      <div class="span-12 last">
         <h3>Download the dataset</h3>
         <p>
            The current, September 13, 2011 version of the dataset is available as a zip archive <a href="https://www.dropbox.com/s/yzrjtc87tfcr2c1/VOCB3DO.zip?dl=1">here</a>.
            Results in the <a href="http://sergeykarayev.com/files/iccv2011.pdf">paper</a> refer to this version.
         </p>
      </div>
      <hr>
      <h3>Details</h3>
      <p>
         For the first few decades of the field's existence, computer vision has been focused on algorithmic, logical approaches to perception. But it was only with the arrival of the era of large datasets, collected and annotated using the Internet, that truly significant progress has been made in object recognition.
      </p>
      <p>
         With the recent proliferation of the Microsoft Kinect, we can for the very first time collect an "Internet-sized" dataset of matched color and depth images. We are making this dataset public to hasten progress in the field of robust, cheap, household object recognition. While our paper is under review in an academic conference, we cannot share much more information--please stay tuned. In the meantime, please participate in the data collection!
      </p>
      <p><strong>
         Annotations of the 3D center points of all objects in the dataset were recently made available by <a href="http://www.cs.bgu.ac.il/~icvl/lab_projects/symmetric-object-detection-in-3d/">Ehud Barnea</a>.
      </strong></p>
      <hr>
      <img width="980px" src="images/numpos_just.png" alt="Current statistics of object distribution in our dataset." />
      <hr>
      <img width="980px" src="images/smoothing_example1.png" alt="Example of a color, registered, and smoothed depth images in our dataset." />
      <hr>
      <h2 class="alt">Thank you for your interest.</h2>
      <hr>
      <p class="quiet right">
        Made by <a href="http://sergeykarayev.com/">Sergey Karayev</a> &amp; <a href="http://www.eecs.berkeley.edu/~allie/">Allison Janoch</a>.
      </p>
    </div>
  </body>
</html>