-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
216 lines (197 loc) · 13.6 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
<!DOCTYPE HTML>
<html lang="en"><head><meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>CVIR IIT Kharagpur</title>
<meta name="author" content="CVIR IIT Kharagpur">
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="stylesheet" type="text/css" href="stylesheet.css">
<link rel="icon" type="image/png" href="images/cvir.png">
</head>
<body>
<table style="width:100%;max-width:800px;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr style="padding:0px">
<td style="padding:0px">
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr style="padding:0px">
<td style="padding:2.5%;width:63%;vertical-align:middle">
<p style="text-align:center">
<name>CVIR IIT Kharagpur</name>
</p>
<p>This is the laboratory webpage of Computer Vision and Intelligence Research Group at Department of Computer Science and Engineering, IIT Kharagpur, India, led by Prof. Abir Das.
</p>
<!-- <p>
<font color="red">The site is under construction and will be updated soon.</font>
</p> -->
<p style="text-align:center">
</p>
</td>
<td style="padding:2.5%;width:30%;max-width:40%">
<a href="images/cvir.png"><img style="width:100%;max-width:100%" alt="profile photo" src="images/cvir.png" class="hoverZoomLink"></a>
</td>
</tr>
</tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr> <td style="padding:20px;width:100%;vertical-align:middle">
<heading>News</heading>
<ul>
<li> [Feb'24] Convolutional Prompting meets Language Models for Continual Learning accepted at <a href="https://cvpr.thecvf.com/Conferences/2024">CVPR 2024</a>.<br>
<li> [Aug'23] Exemplar-Free Continual Transformer with Convolutions accepted at <a href="https://iccv2023.thecvf.com">ICCV 2023</a>.<br>
<li> [Jan'23] Select, Label, and Mix (SLM) received the <font color="red">Best Paper Honorable Mention Award at <a href="https://wacv2023.thecvf.com/">WACV 2023</a>!</font>.<br>
<li> [Oct'22] Paper on Partial Domain Adaptation accepted at <a href="https://wacv2023.thecvf.com/">WACV 2023</a>.<br>
<li> [Sept'21] Paper on Domain Adaptation in Action Recognition accepted at <a href="https://neurips.cc/">NeurIPS 2021</a>.<br>
<li> [Feb'21] Paper on Semi-Supervised Action Recognition accepted at <a href="https://cvpr2021.thecvf.com/">CVPR 2021</a>.<br>
</ul>
</td> </tr>
</tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr>
<td style="padding:20px;width:100%;vertical-align:middle">
<heading>Research</heading>
<p>
</p>
</td>
</tr>
</tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr>
<td style="padding:20px;width:100%;vertical-align:middle">
<heading>Publications</heading>
</td>
</tr>
</tbody></table>
<!-- Papers list -->
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr onmouseout="nlt_stop()" onmouseover="nlt_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/CVPR2024_convprompt.png" width="160" height="100">
</td>
<td width="75%" valign="middle">
<!-- <a href="https://arxiv.org/pdf/2308.11357.pdf"> -->
<papertitle>Convolutional Prompting meets Language Models for Continual Learning</papertitle>
</a>
<br>
Anurag Roy, Riddhiman Moulick, Vinay K. Verma, Saptarshi Ghosh, Abir Das<br>
<em>Computer Vision and Pattern Recognition (<strong>CVPR</strong>)</em>, 2024.<br>
<a href="https://cvir.github.io/projects/convprompt.html">project page</a> /
<a href="https://arxiv.org/pdf/2403.20317.pdf">pdf</a> /
<!-- <a href="https://video.vast.uccs.edu/WACV23/1177_wacv.mp4">video presentation</a> / -->
<!-- <a href="https://docs.google.com/presentation/d/1JlORqqL4LlOOqaAYqriaJ5g5UAIumURyrGF0BAj0gQY/edit?usp=sharing">slides</a> / -->
<a href="https://github.com/CVIR/ConvPrompt">code</a>
<p> We develop a parameter and compute efficient approach called ConvPrompt that leverages convolutional prompt creation and Large Language Models for enhanced knowledge sharing and concept transfer in Continual Learning.</p>
</td>
</tr> </tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr onmouseout="nlt_stop()" onmouseover="nlt_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/ICCV2023_contracon.png" width="160" height="100">
</td>
<td width="75%" valign="middle">
<a href="https://arxiv.org/pdf/2308.11357.pdf">
<papertitle>Exemplar-Free Continual Transformer with Convolutions</papertitle>
</a>
<br>
Anurag Roy, Vinay K. Verma, Sravan Voonna, Kripabandhu Ghosh, Saptarshi Ghosh, Abir Das<br>
<em>International Conference on Computer Vision (<strong>ICCV</strong>)</em>, 2023.<br>
<a href="https://cvir.github.io/projects/contracon.html">project page</a> /
<a href="https://arxiv.org/pdf/2308.11357.pdf">pdf</a> /
<!-- <a href="https://video.vast.uccs.edu/WACV23/1177_wacv.mp4">video presentation</a> / -->
<!-- <a href="https://docs.google.com/presentation/d/1JlORqqL4LlOOqaAYqriaJ5g5UAIumURyrGF0BAj0gQY/edit?usp=sharing">slides</a> / -->
<a href="https://github.com/CVIR/contracon">code</a>
<p> We develop a novel approach called ConTraCon that leverages convolutions on transformer weights for continual learning without requiring task identifiers or storing examples from previous tasks.</p>
</td>
</tr> </tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr onmouseout="nlt_stop()" onmouseover="nlt_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/arXiv_2020_SLM.png" width="160" height="100">
</td>
<td width="75%" valign="middle">
<a href="https://openaccess.thecvf.com/content/WACV2023/papers/Sahoo_Select_Label_and_Mix_Learning_Discriminative_Invariant_Feature_Representations_for_WACV_2023_paper.pdf">
<papertitle>Select, Label, and Mix: Learning Discriminative Invariant Feature Representations for Partial Domain Adaptation</papertitle>
</a>
<br>
Aadarsh Sahoo, Rameswar Panda, Rogerio Feris, Kate Saenko, Abir Das<br>
<em>NeurIPS DistShift Workshop (<strong>NeurIPS-W</strong>)</em>, 2021.<br>
<em>Winter Conference on Applications of Computer Vision (<strong>WACV</strong>)</em>, 2023.<br>
<em><font color="red"><strong>(Best Paper Honorable Mention)</strong></font>.</em><br>
<a href="https://cvir.github.io/projects/slm.html">project page</a> /
<a href="https://video.vast.uccs.edu/WACV23/1177-wacv-post.pdf">poster</a> /
<a href="https://video.vast.uccs.edu/WACV23/1177_wacv.mp4">video presentation</a> /
<a href="https://docs.google.com/presentation/d/1JlORqqL4LlOOqaAYqriaJ5g5UAIumURyrGF0BAj0gQY/edit?usp=sharing">slides</a> /
<a href="https://github.com/CVIR/SLM">code</a>
<p> We develop a novel 'Select, Label, and Mix' (SLM) framework that aims to learn discriminative invariant feature representations for partial domain adaptation.</p>
</td>
</tr> </tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr onmouseout="nlt_stop()" onmouseover="nlt_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/NeurIPS_2021_CoMix.png" width="160" height="100">
</td>
<td width="75%" valign="middle">
<a href="https://proceedings.neurips.cc/paper/2021/file/c47e93742387750baba2e238558fa12d-Paper.pdf">
<papertitle>Contrast and Mix: Temporal Contrastive Video Domain Adaptation with Background Mixing</papertitle>
</a>
<br>
Aadarsh Sahoo, Rutav Shah, Rameswar Panda, Kate Saenko, Abir Das<br>
<em>35th Conference on Neural Information Processing Systems (NeurIPS)</em>, 2021.<br>
<a href="https://cvir.github.io/projects/comix.html">project page</a> /
<a href="https://cvir.github.io/projects/images/comix_neurips2021_poster.png">poster</a> /
<a href="https://slideslive.com/38967894/contrast-and-mix-temporal-contrastive-video-domain-adaptation-with-background-mixing">video presentation</a> /
<a href="https://docs.google.com/presentation/d/1UBvqIRdl7DkmZN7_3lumpMg5kVyjcOl6B507pw4Ul98/edit#slide=id.gf354b2a2b3_0_351">slides</a> /
<a href="https://github.com/CVIR/CoMix">code</a>
<p> We introduce a novel temporal contrastive learning approach for unsupervised video domain adaptation, which is achieved by jointly leveraging video speed, background mixing, and target pseudo-labels.</p>
</td>
</tr> </tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr onmouseout="nlt_stop()" onmouseover="nlt_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/rexl.jpg" width="160" height="100">
</td>
<td width="75%" valign="middle">
<a href="https://cvir.github.io/">
<papertitle>Reinforcement Explanation Learning</papertitle>
</a>
<br>
Siddhant Agarwal, Owais Iqbal, Sree Aditya Buridi, Mada Manjusha, Abir Das<br>
<em>35th Conference on Neural Information Processing Systems (NeurIPS)</em>, 2021.<br>
<a href="https://cvir.github.io/projects/rexl.html">project page</a> /
<a href="https://cvir.github.io/">arXiv</a> /
<!-- <a href="https://youtu.be/EpH175PY1A0">video</a> / -->
<a href="https://cvir.github.io/">code</a>
<p> We reformulate the process of generating saliency maps using perturbation based methods for black box models as a Markov Decsion Process and use RL to optimally search for the best saliency map, thereby reducing the inference time without hurting the performance.</p>
</td>
</tr> </tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr onmouseout="nlt_stop()" onmouseover="nlt_start()">
<td style="padding:20px;width:25%;vertical-align:middle">
<img src="images/TCL_Framework.png" width="160" height="100">
</td>
<td width="75%" valign="middle">
<a href="https://arxiv.org/pdf/2102.02751.pdf">
<papertitle>Semi-Supervised Action Recognition with Temporal Contrastive Learning</papertitle>
</a>
<br>
Ankit Singh, Omprakash Chakraborty, Ashutosh Varshney, Rameswar Panda, Rogerio Feris, Kate Saenko, Abir Das<br>
<em>Computer Vision and Pattern Recognition (CVPR)</em>, 2021.<br>
<a href="https://cvir.github.io/TCL/">project page</a> /
<a href="https://cse.iitkgp.ac.in/~adas/papers/TCL_poster.pdf">poster</a> /
<a href="https://youtu.be/_qIYu3EU2kY">video presentation</a> /
<!-- <a href="https://docs.google.com/presentation/d/1UBvqIRdl7DkmZN7_3lumpMg5kVyjcOl6B507pw4Ul98/edit#slide=id.gf354b2a2b3_0_351">slides</a> / -->
<a href="https://github.com/CVIR/TCL">code</a>
<p> We address semi-supervised video action recognition by learning a two-pathway temporal contrastive model where the similarity between representations of the same video at two different speeds are maximized while the similarity between different videos played at different speeds are minimized.</p>
</td>
</tr> </tbody></table>
<table style="width:100%;border:0px;border-spacing:0px;border-collapse:separate;margin-right:auto;margin-left:auto;"><tbody>
<tr>
<td style="padding:0px">
<br>
<p style="text-align:right;font-size:small;">
Webpage template courtesy: <a href="https://jonbarron.info/">Jon Barron</a>
</p>
</td>
</tr>
</tbody></table>
</td>
</tr>
</table>
</body>
</html>