-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy path360SOD.html
More file actions
150 lines (148 loc) · 6.4 KB
/
Copy path360SOD.html
File metadata and controls
150 lines (148 loc) · 6.4 KB
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
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<title>360SOD</title>
<link rel="stylesheet" type="text/css" href="assets/scripts/bulma.min.css">
<link rel="stylesheet" type="text/css" href="assets/scripts/theme.css">
<link rel="stylesheet" type="text/css" href="https://cdn.bootcdn.net/ajax/libs/font-awesome/4.7.0/css/font-awesome.min.css">
</head>
<body>
<section class="hero is-light" style="">
<div class="hero-body" style="padding-top: 50px;">
<div class="container" style="text-align: center;margin-bottom:5px;">
<h1 class="title">
Distortion-adaptive Salient Object Detection in
</h1>
<h1 class="title">
360° Omnidirectional Images
</h1>
<div class="author">Jia Li<sup>1</sup></div>
<div class="author">Jinming Su<sup>1</sup></div>
<div class="author">Changqun Xia<sup>3</sup></div>
<div class="author">Yonghong Tian<sup>2</sup></div>
<div class="group">
<a href="http://cvteam.net/">CVTEAM</a>
</div>
<div class="aff">
<p><sup>1</sup>State Key Laboratory of Virtual Reality Technology and Systems, SCSE, Beihang University, Beijing, China</p>
<p><sup>2</sup>School of Electronics Engineering and Computer Science,Peking University, Beijing, China.</p>
<p><sup>3</sup>Peng Cheng Laboratory, China</p>
</div>
<div class="con">
<p style="font-size: 24px; margin-top:5px; margin-bottom: 15px;">
JSTSP 2020
</p>
</div>
<div class="columns">
<div class="column"></div>
<div class="column"></div>
<div class="column">
<a href="https://arxiv.org/abs/1909.04913" target="_blank">
<p class="link">Paper</p>
</a>
</div>
<div class="column">
<a href="https://github.com/iCVTEAM/360SOD/" target="_blank">
<p class="link">Code</p>
</a>
</div>
<div class="column"></div>
<div class="column"></div>
</div>
</div>
</div>
</section>
<div style="text-align: center;">
<div class="container" style="max-width:850px">
<div style="text-align: center;">
<img src="assets/360SOD/head.png" class="centerImage">
</div>
</div>
<div class="head_cap">
<p style="color:gray;">
The framework of our baseline model.
</p>
</div>
</div>
<section class="hero">
<div class="hero-body">
<div class="container" style="max-width: 800px" >
<h1 style="">Abstract</h1>
<p style="text-align: justify; font-size: 17px;">
Image-based salient object detection (SOD) has been
extensively explored in the past decades. However, SOD on 360°
omnidirectional images is less studied owing to the lack of
datasets with pixel-level annotations. Toward this end, this paper
proposes a 360° image-based SOD dataset that contains 500
high-resolution equirectangular images. We collect the representative
equirectangular images from five mainstream 360° video datasets
and manually annotate all objects and regions over these images
with precise masks with a free-viewpoint way. To the best of our
knowledge, it is the first public available dataset for salient object
detection on 360° scenes. By observing this dataset, we find that
distortion from projection, large-scale complex scene and small
salient objects are the most prominent characteristics. Inspired
by these foundings, this paper proposes a baseline model for
SOD on equirectangular images. In the proposed approach, we
construct a distortion-adaptive module to deal with the distortion
caused by the equirectangular projection. In addition, a
multiscale contextual integration block is introduced to perceive and
distinguish the rich scenes and objects in omnidirectional scenes.
The whole network is organized in a progressively manner with
deep supervision. Experimental results show the proposed baseline
approach outperforms the top-performanced state-of-the-art methods on 360°
SOD dataset. Moreover, benchmarking results
of the proposed baseline approach and other methods on 360°
SOD dataset show the proposed dataset is very challenging, which
also validate the usefulness of the proposed dataset and approach
to boost the development of SOD on 360° omnidirectional scenes.
</p>
</div>
</div>
</section>
<section class="hero is-light" style="background-color:#FFFFFF;">
<div class="hero-body">
<div class="container" style="max-width:800px;margin-bottom:20px;">
<h1>
Representative examples
</h1>
</div>
<div class="container" style="max-width:800px">
<div style="text-align: center;">
<img src="assets/360SOD/comp.png" class="centerImage">
</div>
</div>
</div>
</section>
<section class="hero" style="padding-top:0px;">
<div class="hero-body">
<div class="container" style="max-width:800px;">
<div class="card">
<header class="card-header">
<p class="card-header-title">
BibTex Citation
</p>
<a class="card-header-icon button-clipboard" style="border:0px; background: inherit;" data-clipboard-target="#bibtex-info" >
<i class="fa fa-copy" height="20px"></i>
</a>
</header>
<div class="card-content">
<pre style="background-color:inherit;padding: 0px;" id="bibtex-info">@article{li2019distortion,
title={Distortion-Adaptive Salient Object Detection in 360$\^{}$\backslash$circ $ Omnidirectional Images},
author={Li, Jia and Su, Jinming and Xia, Changqun and Tian, Yonghong},
journal={IEEE Journal of Selected Topics in Signal Processing},
volume={14},
number={1},
pages={38--48},
year={2019},
publisher={IEEE}
}</pre>
</div>
</section>
<script type="text/javascript" src="assets/scripts/clipboard.min.js"></script>
<script>
new ClipboardJS('.button-clipboard');
</script>
</body>
</html>