-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
221 lines (208 loc) · 8.81 KB
/
index.html
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
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1">
<meta name="description" content="width=device-width, initial-scale=1">
<meta name="description"
content="SEM leverages geometry priors (epipolar constraint) to filtering out irrelevant areas in attention and matching, and fuses relative positional information with anchor points to improve the robustness in textureless and repetitive texture areas. Accepted in CVPR 2023 IMW."/>
<title>SEM: Structured Epipolar Matcher for Local Feature Matching</title>
<!-- Bootstrap -->
<link href="css/bootstrap-4.4.1.css" rel="stylesheet">
<link href="https://fonts.googleapis.com/css?family=Open+Sans" rel="stylesheet" type="text/css">
<script type="text/javascript" id="MathJax-script" async
src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js">
</script>
<style>
body {
background: no-repeat fixed top left;
font-family:'Open Sans', sans-serif;
}
</style>
<script>
MathJax = {
tex: {inlineMath: [['$', '$'], ['\\(', '\\)']]}
};
</script>
</head>
<body>
<!-- cover -->
<section>
<div class="jumbotron text-center mt-0">
<div class="container">
<div class="row">
<div class="col">
<h2>SEM: Structured Epipolar Matcher for Local Feature Matching</h2>
<h4 style="color:#6e6e6e;"> CVPR 2023 Image Matching Workshop</h4>
<hr>
<h6>
<a href="https://scholar.google.com/citations?user=HA5zLp4AAAAJ" target="_blank">Jiahao Chang</a><sup>*</sup>,
<a href="https://monsoon-cs.moe" target="_blank">Jiahuan Yu</a><sup>*</sup>,
<a href="http://staff.ustc.edu.cn/~tzzhang/" target="_blank">Tianzhu Zhang</a>
</h6>
<p>University of Science and Technology of China
<br>
<sup>*</sup> denotes equal contribution
</p>
<div class="row justify-content-center">
<div class="column">
<p class="mb-5"><a class="btn btn-large btn-light"
href="https://arxiv.org/abs/2303.16646" role="button"
target="_blank">
<i class="fa fa-file"></i> Paper </a></p>
</div>
<div class="column">
<p class="mb-5"><a class="btn btn-large btn-light"
href="files/SEM_slides.pdf" role="button"
target="_blank">
<i class="fa fa-file"></i> Slides </a></p>
</div>
<div class="column">
<p class="mb-5"><a class="btn btn-large btn-light" id="code_soon"
href="https://github.com/SEM2023/SEM" role="button"
target="_blank">
<i class="fa fa-github-alt"></i> Code (Coming Soon) </a></p>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section>
<div class="container">
<div class="row">
<div class="col-12 text-center">
<h5 style="color:green">Thanks to <a href="https://zju3dv.github.io/loftr/" target="_blank">LoFTR</a>
for their great work and well-organized code.</h5>
<h5 style="color:red">Also see our another work <a href="https://astr2023.github.io" target="_blank">ASTR</a> in local feature matching, accepted by CVPR 2023</h5>
</div>
</div>
</div>
</section>
<br>
<!-- abstract -->
<section>
<div class="container">
<div class="row">
<div class="col-12 text-center">
<h3>Abstract</h3>
<hr style="margin-top:0px">
<h6 style="color:#8899a5" class="text-center">
TL;DR: SEM leverages geometry priors (epipolar constraint) to filtering out irrelevant areas in attention and matching,
and fuses relative positional information with anchor points to improve the robustness in textureless and repetitive texture areas.
</h6>
<br>
<div><b style="color:#fd5638; font-size:large" id="demo-warning"></b>
<br>
<p class="text-justify">
Local feature matching is challenging due to textureless
and repetitive patterns. Existing methods focus on using
appearance features and global interaction and matching,
while the importance of geometry priors in local feature
matching has not been fully exploited. Different from these
methods, in this paper, we delve into the importance of
geometry prior and propose Structured Epipolar Matcher
(SEM) for local feature matching, which can leverage the
geometric information in an iterative matching way. The
proposed model enjoys several merits. First, our proposed
Structured Feature Extractor can model the relative po-
sitional relationship between pixels and high-confidence
anchor points. Second, our proposed Epipolar Attention
and Matching can filter out irrelevant areas by utilizing
the epipolar constraint. Extensive experimental results on
five standard benchmarks demonstrate the superior performance
of our SEM compared to state-of-the-art methods.
</p>
</div>
</div>
</div>
</div>
</section>
<br>
<!-- overview video -->
<section>
<div class="container">
<div class="row">
<div class="col-12 text-center">
<h3>Overview video (6:24)</h3>
<hr style="margin-top:0px">
<div class="embed-responsive embed-responsive-16by9">
<iframe width="560" height="315" src="https://www.youtube.com/embed/fm1LH8wv-sU" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen></iframe>
</div>
</div>
</div>
</div>
</section>
<br>
<!-- pipeline -->
<section>
<div class="container">
<div class="row">
<div class="col-12 text-center">
<h3>Pipeline Overview</h3>
<hr style="margin-top:0px">
WIP
<p class="text-justify">
</p>
<br>
</div>
</div>
</div>
</section>
<!-- performance -->
<section>
<div class="container">
<div class="row">
<div class="col-12 text-center">
<h3>Performance</h3>
<hr style="margin-top:0px">
</div>
</div>
<div class="row justify-content-center">
WIP
</div>
</div>
</section>
<br>
<!-- visualizations -->
<section>
<div class="container">
<div class="row">
<div class="col-12 text-center">
<h3>Visualizations</h3>
<hr style="margin-top:0px">
</div>
</div>
<div class="row justify-content-center">
WIP
</div>
</div>
</section>
<br>
<!-- citing -->
<div class="container">
<div class="row ">
<div class="col-12">
<h3>Citation</h3>
<hr style="margin-top:0px">
<pre style="background-color: #e9eeef;padding: 1.25em 1.5em">
<code>@inproceedings{chang2023structured,
title={Structured Epipolar Matcher for Local Feature Matching},
author={Chang, Jiahao and Yu, Jiahuan and Zhang, Tianzhu},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
pages={6176--6185},
year={2023}
}</code>
</pre>
</div>
</div>
</div>
<footer class="text-center" style="margin-bottom:10px; font-size: medium;">
<hr>
Thanks to <a href="https://lioryariv.github.io/" target="_blank">Lior Yariv</a> for the <a
href="https://lioryariv.github.io/idr/" target="_blank">website template</a>.
</footer>
</body>
</html>