-
Notifications
You must be signed in to change notification settings - Fork 5
/
index.html
766 lines (694 loc) · 38.8 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
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
<title>Xiang Li</title>
<meta content="Xiang Li, https://implus.github.io" name="keywords">
<style media="screen" type="text/css">
html,
body,
div,
span,
applet,
object,
iframe,
h1,
h2,
h3,
h4,
h5,
h6,
p,
blockquote,
pre,
a,
abbr,
acronym,
address,
big,
cite,
code,
del,
dfn,
em,
font,
img,
ins,
kbd,
q,
s,
samp,
small,
strike,
strong,
sub,
tt,
var,
dl,
dt,
dd,
ol,
ul,
li,
fieldset,
form,
label,
legend,
table,
caption,
tbody,
tfoot,
thead,
tr,
th,
td {
border: 0pt none;
font-family: inherit;
font-size: 100%;
font-style: inherit;
font-weight: inherit;
margin: 0pt;
outline-color: invert;
outline-style: none;
outline-width: 0pt;
padding: 0pt;
vertical-align: baseline;
}
a {
color: #1772d0;
text-decoration: none;
}
a:focus,
a:hover {
color: #f09228;
text-decoration: none;
}
a.paper {
font-weight: bold;
font-size: 12pt;
}
b.paper {
font-weight: bold;
font-size: 12pt;
}
* {
margin: 0pt;
padding: 0pt;
}
body {
position: relative;
margin: 3em auto 2em auto;
width: 800px;
font-family: Verdana, Helvetica, sans-serif;
font-size: 14px;
background: #eee;
}
h2 {
font-family: Verdana, Helvetica, sans-serif;
font-size: 15pt;
font-weight: 700;
}
h3 {
font-family: Verdana, Helvetica, sans-serif;
font-size: 16px;
font-weight: 700;
}
strong {
font-family: Verdana, Helvetica, sans-serif;
font-size: 13px;
font-weight: bold;
}
ul {
list-style: circle;
}
img {
border: none;
}
li {
padding-bottom: 0.5em;
margin-left: 1.4em;
}
alert {
font-family: Verdana, Helvetica, sans-serif;
font-size: 13px;
font-weight: bold;
color: #FF0000;
}
em,
i {
font-style: italic;
}
div.section {
clear: both;
margin-bottom: 1.5em;
background: #eee;
}
div.spanner {
clear: both;
}
div.paper {
clear: both;
margin-top: 0.5em;
margin-bottom: 1em;
border: 1px solid #ddd;
background: #fff;
padding: 1em 1em 1em 1em;
}
div.paper div {
padding-left: 230px;
}
img.paper {
margin-bottom: 0.5em;
float: left;
width: 200px;
}
span.blurb {
font-style: italic;
display: block;
margin-top: 0.75em;
margin-bottom: 0.5em;
}
pre,
code {
font-family: 'Lucida Console', 'Andale Mono', 'Courier', monospaced;
margin: 1em 0;
padding: 0;
}
div.paper pre {
font-size: 0.9em;
}
</style>
<link href="http://fonts.googleapis.com/css?family=Lato:400,700,400italic,700italic" rel="stylesheet"
type="text/css"/>
<!-- Global site tag (gtag.js) - Google Analytics -->
<script async src="https://www.googletagmanager.com/gtag/js?id=UA-164510176-1"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'UA-164510176-1');
</script>
</head>
<body>
<div style="margin-bottom: 1em; border: 1px solid #ddd; background-color: #fff; padding: 1em; height: 130px;">
<div style="margin: 0px auto; width: 100%;">
<img title="implus" style="float: left; padding-left: .01em; height: 130px;"
src="./resources/images/me.png">
<div style="padding-left: 12em; vertical-align: top; height: 120px;">
<span style="line-height: 150%; font-size: 20pt;">Xiang Li (李翔)</span><br>
<span> <a href="https://cc.nankai.edu.cn/2021/0323/c13620a490349/page.htm">Associate Professor, College of Computer Science, Nankai University</a></span><br>
<span><strong>Address</strong>: No. 38, Tongyan Road, Haihe Education Park, Tianjin, China</span><br>
<span><strong>Email</strong>: xiang.li.implus [at] {nankai.edu.cn} </span> <br>
<span><strong>Research Group Page</strong>: <a href="https://github.com/IMPlus-PCALab">IMPlus@PCALab</a> </span> <br>
</div>
</div>
</div>
<!--<div style="clear: both; background-color: #fff; margin-top: 1.5em; padding: .2em; padding-left: .3em;">-->
<div style="clear: both;">
<div class="section">
<h2>About Me [<a href="https://github.com/implus">GitHub</a>]
[<a href="https://scholar.google.com/citations?user=oamjJdYAAAAJ&hl=zh-CN">Google Scholar</a>]
[<a href="https://github.com/IMPlus-PCALab">Research Group</a>]
<!--[<a href="./resources/cv/wwh_cv.pdf">CV</a>])-->
</h2>
<div class="paper">
I'm an <a href="https://cc.nankai.edu.cn/2021/0323/c13620a490349/page.htm">Associate Professor</a> in College of Computer Science, Nankai University, in the Team of <a href="https://mmcheng.net/cmm/">Ming-Ming Cheng</a>.
I got my PhD degree from the Department of Computer Science and Technology, Nanjing University of Science and Technology (NJUST) in 2020.
My advisor is <a href="http://202.119.85.163/open/TutorInfo.aspx?dsbh=tLbjVM9T1OzsoNduSpyHQg==&yxsh=4iVdgPyuKTE=&zydm=L-3Jh59wXco=">Prof. Jian Yang</a> from NJUST, who is a Changjiang Scholar. My vice-advisor is <a href="http://www.xlhu.cn/">Prof. Xiaolin Hu</a> from Tsinghua University.
I started my postdoctoral career in NJUST as a candidate for the <a href="https://zhuanlan.zhihu.com/p/147471409">2020 Postdoctoral Innovative Talent Program</a>, supervised by <a href="https://imag-njust.net/jinhui-tang/">Prof. Jinhui Tang</a>.
In 2016, I spent 8 months as a research intern in Microsoft Research Asia, supervised by <a href="https://scholar.google.com/citations?user=Bl4SRU0AAAAJ&hl=zh-CN">Prof. Tao Qin</a> and <a href="https://scholar.google.com/citations?user=Nh832fgAAAAJ&hl=zh-CN">Prof. Tie-Yan Liu</a>.
I was a visiting scholar at <a href="https://www.momenta.cn/">Momenta</a>, mainly focusing on monocular perception algorithm.
<br><br>
My recent works are mainly on:
<ul>
<li>neural architecture design, CNN/Transformer</li>
<li>object detection/recognition</li>
<li>unsupervised learning</li>
<li>knowledge distillation</li>
</ul>
<p style='color:red'><strong>
We are looking for self-motivated PhD candidates! Please feel free to contact me through the email (attach your CV).
During the PhD career, you can have:
<ul>
<li>joint supervision with well-known research institute (e.g., Megvii, SenseTime Research, Huawei Noah's Ark Lab, BAAI)</li>
<li>hand in hand guidance to publish earlier papers</li>
<li>relatively flexible and free research space</li>
</ul>
We would not push hard, but you should always be self-driven for your own target, i.e., making solid and impactful contributions to the CV/AI community.
</strong></p>
</div>
</div>
</div>
<div style="clear: both;">
<div class="section">
<h2 id="confpapers">Honor</h2>
(See more details (codes, solutions, summaries) in <a href="https://github.com/IMPlus-PCALab/AICompetition">[AICompetition of Group Page]</a>)
<div class="paper">
<ul>
<li>
1st place of of Change Detection in High-resolution and Multi-temporal Optical Images, 2nd place of Forgery Detection in Multi-scenario Remote Sensing Images of Typical Objects in 2024 TC I Contest on <a href="https://mp.weixin.qq.com/s/tnx1cWeyt3siIjBD0civsQ">Intelligent Interpretation for Multi-modal Remote Sensing Application</a>, total <strong>13,000 RMB bonus</strong>
</li>
<li>
Second place of IACC International Algorithm Case Competition, namely the <a href="https://www.cvmart.net/race/10345/base">remote sensing detection</a>, <strong>100,000 RMB bonus (2nd from 116 teams)</strong>
</li>
<li>
Champion of 2022 Jittor AI competition, namely the <a href="https://www.educoder.net/competitions/index/Jittor-3">landscape picture generation</a>, <strong>50,000 RMB bonus (1st from 154 teams)</strong>
</li>
<li>
Second place of 2020 Zhengtu Cup's first AI competition, namely the industrial defect detection algorithm, <strong>150,000 RMB bonus (2nd from 900 teams)</strong>
</li>
<li>
Champion of 2016 Didi Tech Di-Tech's first big data competition, namely the travel demand prediction algorithm, <strong>100,000 US dollars bonus (1st from 7664 team)</strong>
</li>
<li>
Champion of 2015 Alibaba Tianchi's first big data competition, namely Ali mobile recommendation algorithm, <strong>300,000 RMB bonus (1st from 7186 team)</strong>
</li>
<li>
2015 Dean Medal of School of Computer Science, Nanjing University of Science and Technology, 2016 Presidential Medal of Nanjing University of Science and Technology, 2016 National Scholarship
</li>
<li>
ACM-ICPC Asia Regional Contest, Silver Medal (1st)
</li>
</ul>
<div class="spanner"></div>
</div>
</div>
</div>
<div style="clear: both;">
<div class="section">
<h2 id="experience">News</h2>
<div class="paper">
<ul>
<li>
2024-09-26: 3 papers accepted in NeurIPS 2024, including <a href="https://github.com/zcablii/SARDet_100K">SARDet100K</a>.
</li>
<li>
2024-09-16: I was selected for the 2024 World’s Top 2% Scientists list released by Stanford University, USA. See details <a href="https://mp.weixin.qq.com/s/eJIK5x-nEzy07Ltz5Ypi3g">here</a>.
</li>
<li>
2024-07-02: 2 papers accepted in ECCV 2024.
</li>
<li>
2024-06-22: JianLun (减论) IP Project is officially started, focusing on efficient AI understanding and education.
</li>
<li>
2024-05-29: 1 paper <a href="https://github.com/Zzh-tju/ZoneEval">Zone Evaluation</a> accepted in TPAMI.
</li>
<li>
2024-03-07: <a href="http://www.wuwenjunkejijiang.cn/a/2237.html">Final round</a> of The Wu Wenjun AI Outstanding Youth Award.
</li>
<li>
2024-02-27: 2 papers (<a href="https://github.com/zhengli97/PromptKD">PromptKD</a>, <a href="https://github.com/jbwang1997/CrossKD">CrossKD</a>) accepted in CVPR 2024.
</li>
<li>
2023-09-23: 1 paper (<a href="https://openreview.net/pdf?id=l6R4Go3noz">FGVP</a>) accepted in NeurIPS 2023.
</li>
<li>
2023-07-14: 3 papers (including <a href="https://arxiv.org/abs/2303.09030.pdf">LSKNet</a>, ADNet) accepted in ICCV 2023.
</li>
<li>
2023-04-25: 1 paper <a href="">DUAL for Panoramic Depth Completion</a> accepted in ICML 2023.
</li>
<li>
2023-01: Nomination Award for the CCF Excellent Doctoral Dissertation Incentive Plan. See <a href="https://www.ccf.org.cn/Awards/Awards/2022-12-08/781242.shtml">first round</a>, <a href="https://www.ccf.org.cn/Awards/Awards/2023-01-04/783561.shtml">final result</a>.
</li>
<li>
2022-11-19: 4 papers (including <a href="https://arxiv.org/pdf/2211.16231.pdf">Curriculum Temperature</a>, <a href="https://arxiv.org/pdf/2211.10994.pdf">DesNet</a>) accepted in AAAI 2023.
</li>
<li>
2022-09-15: 2 papers (<a href="https://arxiv.org/pdf/2203.06844.pdf">RecursiveMix</a>, <a href="https://arxiv.org/pdf/2207.05536.pdf">DTG-SSOD</a>) accepted in NeurIPS 2022.
</li>
<li>
2022-07-05: 3 papers (<a href="https://arxiv.org/pdf/2107.13802.pdf">RigNet</a>, <a href="https://arxiv.org/pdf/2203.09855.pdf">M3PT</a>, <a href="https://arxiv.org/pdf/2203.16317.pdf">PseCo</a>) accepted in ECCV 2022.
</li>
<li>
2022-05-20: 1 paper (<a href="https://arxiv.org/pdf/2205.10063.pdf">UM-MAE</a>) is publicly available in arXiv.
</li>
<li>
2022-03-02: 1 paper (<a href="https://arxiv.org/pdf/2203.03253.pdf">dynamicMLP</a>) accepted (oral) in CVPR 2022.
</li>
<li>
2021-12-01: 1 paper (<a href="https://arxiv.org/pdf/2112.04840.pdf">KD for object detection</a>) accepted in AAAI 2022.
</li>
<li>
2021-05-05: 1 paper (<a href="https://ieeexplore.ieee.org/document/9423611">PAN++</a>) is accepted by TPAMI 2021.
</li>
<li>
2021-03-01: 1 paper (<a href="https://arxiv.org/pdf/2011.12885.pdf">GFocalv2</a>) accepted in CVPR 2021.
</li>
<li>
2020-09-25: 1 paper (<a href="https://arxiv.org/pdf/2006.04388.pdf">GFocal</a>) accepted in NeurIPS 2020.
</li>
<li> 2019-12-01: 1 paper (<a href="https://ojs.aaai.org/index.php/AAAI/article/download/5904/5760">Understanding the disharmony v2</a>) accepted in AAAI 2020. </li>
<li> 2019-03-15: 3 papers (<a href="http://openaccess.thecvf.com/content_CVPR_2019/papers/Li_Selective_Kernel_Networks_CVPR_2019_paper.pdf">SKNet</a>,
<a href="https://openaccess.thecvf.com/content_CVPR_2019/papers/Li_Understanding_the_Disharmony_Between_Dropout_and_Batch_Normalization_by_Variance_CVPR_2019_paper.pdf">Understanding the disharmony v1</a>,
<a href="http://openaccess.thecvf.com/content_CVPR_2019/papers/Wang_Shape_Robust_Text_Detection_With_Progressive_Scale_Expansion_Network_CVPR_2019_paper.pdf">PSENet</a>) accepted in CVPR 2019. </li>
<li> 2017-03-01: 1 paper (<a href="https://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_Stacked_Conditional_Generative_CVPR_2018_paper.pdf">ST-CGAN</a>) accepted in CVPR 2018. </li>
<li> 2018-06-16: 1 paper (<a href="https://www.ijcai.org/Proceedings/2018/0391.pdf">MixNet</a>) accepted in IJCAI 2018. </li>
<li> 2016-09-30: 1 paper (<a href="https://arxiv.org/pdf/1610.09893.pdf">LightRNN</a>) accepted in NeurIPS 2016. </li>
</ul>
<div class="spanner"></div>
</div>
</div>
</div>
<div style="clear: both;">
<div class="section">
<h2 id="confpapers">Selected Publications</h2>
(* indicates equal contribution, # corresponding author)
<div class="paper"><img class="paper" src="./resources/paper_icon/ICCV_2023_LSKNet.png"
title="Large Selective Kernel Network for Remote Sensing Object Detection">
<div><strong>Large Selective Kernel Network for Remote Sensing Object Detection</strong><br>
Yuxuan Li, Qibin Hou, Zhaohui Zheng, Ming-Ming Cheng, Jian Yang#, Xiang Li#<br>
in ICCV, 2023<br>
<a href="https://arxiv.org/pdf/2303.09030.pdf">[Paper]</a>
<a href="./resources/bibtex/ICCV_2023_LSKNet.txt">[BibTex]</a>
<a href="https://github.com/zcablii/LSKNet">[Code]</a><img
src="https://img.shields.io/github/stars/zcablii/LSKNet?style=social"/>
<br>
<a href="https://paperswithcode.com/sota/object-detection-in-aerial-images-on-dota-1?p=large-selective-kernel-network-for-remote"><img
src="https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/large-selective-kernel-network-for-remote/object-detection-in-aerial-images-on-dota-1"/></a>
<a href="https://paperswithcode.com/sota/object-detection-in-aerial-images-on-hrsc2016?p=large-selective-kernel-network-for-remote"><img
src="https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/large-selective-kernel-network-for-remote/object-detection-in-aerial-images-on-hrsc2016"/></a>
<br>
<alert>
LSKNet can dynamically adjust its large spatial receptive field to better model the ranging context of various categories of objects in remote sensing scenarios
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/AAAI_2023_CTKD.png"
title="Curriculum Temperature for Knowledge Distillation">
<div><strong>Curriculum Temperature for Knowledge Distillation</strong><br>
Zheng Li, <strong>Xiang Li</strong>#, Lingfeng Yang, Borui Zhao, Renjie Song, Lei Luo, Jun Li, Jian Yang#<br>
in AAAI, 2023<br>
<a href="https://arxiv.org/pdf/2211.16231.pdf">[Paper]</a>
<a href="./resources/bibtex/AAAI_2023_CTKD.txt">[BibTex]</a>
<a href="https://github.com/zhengli97/CTKD">[Code]</a><img
src="https://img.shields.io/github/stars/zhengli97/CTKD?style=social"/>
<br>
<alert>
CTKD organizes the distillation task from easy to hard through a dynamic and learnable temperature. The temperature is learned during the student’s training process with a reversed gradient that aims to maximize the distillation loss in an adversarial manner.
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/NeurIPS_2022_RM.png"
title="RecursiveMix: Mixed Learning with History">
<div><strong>RecursiveMix: Mixed Learning with History</strong><br>
Lingfeng Yang*, <strong>Xiang Li</strong>*, Borui Zhao, Renjie Song, Jian Yang#<br>
in NeurIPS (<strong>Spotlight</strong>), 2022<br>
<a href="https://openreview.net/pdf?id=NjP18IbKKlX">[Paper]</a>
<a href="./resources/bibtex/NeurIPS_2022_RM.txt">[BibTex]</a>
<a href="https://github.com/implus/RecursiveMix-pytorch">[Code]</a><img
src="https://img.shields.io/github/stars/implus/RecursiveMix-pytorch?style=social"/>
<br>
<alert>
RecursiveMix is a simple but effective data augmentation technique that first leverages the historical input-prediction-label triplets.
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/NeurIPS_2022_DTG.png"
title="DTG-SSOD: Dense Teacher Guidance for Semi-Supervised Object Detection">
<div><strong>DTG-SSOD: Dense Teacher Guidance for Semi-Supervised Object Detection</strong><br>
Gang Li, <strong>Xiang Li</strong>#, Yujie Wang, Yichao Wu, Ding Liang, Shanshan Zhang#<br>
in NeurIPS, 2022<br>
<a href="https://openreview.net/pdf?id=0-uBrFiOVf">[Paper]</a>
<a href="./resources/bibtex/NeurIPS_2022_DTG.txt">[BibTex]</a>
<a href="https://github.com/ligang-cs/DTG-SSOD">[Code(to be released)]</a><img
src="https://img.shields.io/github/stars/ligang-cs/DTG-SSOD?style=social"/>
<br>
<alert>
DTG-SSOD explores a novel “dense-to-dense” paradigm, instead of the traditional “sparse-to-dense” paradigm, for effective semi-supervised object detection.
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/ECCV_2022_PseCo.png"
title="PseCo: Pseudo Labeling and Consistency Training for Semi-Supervised Object Detection">
<div><strong>PseCo: Pseudo Labeling and Consistency Training for Semi-Supervised Object Detection</strong><br>
Gang Li, <strong>Xiang Li</strong>#, Yujie Wang, Yichao Wu, Ding Liang, Shanshan Zhang#<br>
in ECCV, 2022<br>
<a href="https://arxiv.org/pdf/2203.16317.pdf">[Paper]</a>
<a href="./resources/bibtex/ECCV_2022_PseCo.txt">[BibTex]</a>
<a href="https://github.com/ligang-cs/PseCo">[Code]</a><img
src="https://img.shields.io/github/stars/ligang-cs/PseCo?style=social"/>
<a href="https://zhuanlan.zhihu.com/p/544346080">[Blog(Chinese)]</a>
<a href="https://www.bilibili.com/video/BV1DP411j7kg/">[Video(Chinese)]</a>
<br>
<alert>
PseCo delves into two key techniques of semi-supervised learning (e.g., pseudo labeling and consistency training) for SSOD, and integrate object detection properties into them.
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/arXiv_2022_UMMAE.png"
title="Uniform Masking: Enabling MAE Pre-training for Pyramid-based Vision Transformers with Locality">
<div><strong>Uniform Masking: Enabling MAE Pre-training for Pyramid-based Vision Transformers with Locality</strong><br>
<strong>Xiang Li</strong>, Wenhai Wang, Lingfeng Yang, Jian Yang#<br>
in arXiv, 2022<br>
<a href="https://arxiv.org/pdf/2205.10063.pdf">[Paper]</a>
<a href="./resources/bibtex/arXiv_2022_UMMAE.txt">[BibTex]</a>
<a href="https://github.com/implus/UM-MAE">[Code]</a><img
src="https://img.shields.io/github/stars/implus/UM-MAE?style=social"/>
<a href="https://zhuanlan.zhihu.com/p/520228061">[Blog(Chinese)]</a>
<br>
<alert>
UM-MAE is an efficient and general technique that supports MAE-style MIM Pre-training for popular Pyramid-based Vision Transformers (e.g., PVT, Swin).
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/CVPR_2022_DynamicMLP.png"
title="Dynamic MLP for Fine-Grained Image Classification by Leveraging Geographical and Temporal Information">
<div><strong>Dynamic MLP for Fine-Grained Image Classification by Leveraging Geographical and Temporal Information</strong><br>
Lingfeng Yang, <strong>Xiang Li</strong>#, Renjie Song, Borui Zhao, Juntian Tao, Shihao Zhou, Jiajun Liang, Jian Yang#<br>
in CVPR (<strong>Oral</strong>), 2022<br>
<!-- <a href="https://arxiv.org/pdf/2106.13797.pdf">[Paper]</a> -->
<a href="https://arxiv.org/pdf/2203.03253.pdf">[Paper]</a>
<a href="./resources/bibtex/dynamicMLP.txt">[BibTex]</a>
<a href="https://github.com/ylingfeng/DynamicMLP">[Code]</a><img
src="https://img.shields.io/github/stars/ylingfeng/DynamicMLP?style=social"/>
<br>
<alert>
A very simple and effective approach for fine-grained recognition tasks using auxiliary knowledge like geographical/temporal information. We achieve SOTA results and take third place in the iNaturalist challenge at FGVC8 (CVPR21 workshop)
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/AAAI_2022_KD.png"
title="Knowledge Distillation for Object Detection via Rank Mimicking and Prediction-guided Feature Imitation">
<div><strong>Knowledge Distillation for Object Detection via Rank Mimicking and Prediction-guided Feature Imitation</strong><br>
Gang Li*, <strong>Xiang Li</strong>*, Yujie Wang, Shanshan Zhang#, Yichao Wu, Ding Liang
in AAAI, 2022<br>
<!-- <a href="https://arxiv.org/pdf/2106.13797.pdf">[Paper]</a> -->
<a href="https://arxiv.org/pdf/2112.04840.pdf">[Paper]</a>
<a href="./resources/bibtex/kd_rm_pfi.txt">[BibTex]</a>
<br>
<alert>Rank Mimicking and Prediction-guided Feature Imitation for knowledge Distillation of Dense Object Detection, A Simple and Effective Approach!</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/arXiv_2021_PVTv2.png"
title="PVTv2: Improved Baselines with Pyramid Vision Transformer">
<div><strong>PVTv2: Improved Baselines with Pyramid Vision Transformer</strong><br>
Wenhai Wang, Enze Xie, <strong>Xiang Li</strong>, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping
Luo, Ling Shao<br>
Technical Report, 2021<br>
<!-- <a href="https://arxiv.org/pdf/2106.13797.pdf">[Paper]</a> -->
<a href="./resources/reports/PVTv2_Improved_Baselines_with_Pyramid_Vision_Transformer.pdf">[Paper]</a>
<a href="https://github.com/whai362/PVT">[Code]</a><img
src="https://img.shields.io/github/stars/whai362/PVT?style=social"/>
<a href="https://zhuanlan.zhihu.com/p/353222035">[中文解读]</a>
<a href="./resources/reports/wangwenhai_vision_transformer.pdf">[Report]</a>
<a href="https://www.techbeat.net/talk-info?id=562">[Talk]</a>
<a href="./resources/bibtex/arXiv_2021_PVTv2.txt">[BibTex]</a>
<br>
<alert>A better PVT.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/ICCV_2021_PVT.png"
title="Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions">
<div><strong>Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions</strong><br>
Wenhai Wang, Enze Xie, <strong>Xiang Li</strong>, Deng-Ping Fan#, Kaitao Song, Ding Liang, Tong Lu#, Ping
Luo, Ling Shao<br>
in ICCV, 2021 (<strong>oral presentation</strong>)<br>
<a href="https://arxiv.org/pdf/2102.12122.pdf">[Paper]</a>
<!-- [<a href="./resources/posters/ICCV_2019_PAN.pdf">Poster</a>]-->
<a href="https://github.com/whai362/PVT">[Code]</a><img
src="https://img.shields.io/github/stars/whai362/PVT?style=social"/>
<a href="https://zhuanlan.zhihu.com/p/353222035">[中文解读]</a>
<a href="./resources/reports/wangwenhai_vision_transformer.pdf">[Report]</a>
<a href="https://www.techbeat.net/talk-info?id=562">[Talk]</a>
<a href="./resources/bibtex/ICCV_2021_PVT.txt">[BibTex]</a>
<br>
<alert>A pure Transformer backbone for dense prediction, such as object detection and semantic segmentation.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/TPAMI_2021_PAN++.png"
title="PAN++: Towards Efficient and Accurate End-to-End Spotting of Arbitrarily-Shaped Text">
<div><strong>PAN++: Towards Efficient and Accurate End-to-End Spotting of Arbitrarily-Shaped Text</strong><br>
Wenhai Wang*, Enze Xie*, <strong>Xiang Li</strong>, Xuebo Liu, Ding Liang, Zhibo Yang, Tong Lu#, Chunhua Shen<br>
TPAMI, 2021<br>
<a href="https://ieeexplore.ieee.org/document/9423611">[Paper]</a>
<a href="https://github.com/whai362/pan_pp.pytorch">[Code]</a><img
src="https://img.shields.io/github/stars/whai362/pan_pp.pytorch?style=social"/>
<a href="./resources/bibtex/TPAMI_2021_PAN++.txt">[BibTex]</a>
<br>
<alert>We extend PSENet (CVPR'19) and PAN (ICCV'19) to a text spotting system.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/CVPR_2021_GFLv2.png"
title="Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection">
<div><strong>Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object
Detection</strong><br>
<strong>Xiang Li</strong>*, Wenhai Wang, Xiaolin Hu, Jun Li, Jinhui Tang, Jian Yang<br>
in CVPR, 2021<br>
<a href="https://arxiv.org/pdf/2011.12885.pdf">[Paper]</a>
<a href="https://github.com/implus/GFocalV2">[Code]</a><img
src="https://img.shields.io/github/stars/implus/GFocalV2?style=social"/>
<a href="./resources/bibtex/CVPR_2021_GFLv2.txt">[BibTex]</a>
<br>
<alert>The improved version of GFocal!
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/GFocal.png"
title="Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection">
<div><strong>Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection</strong><br>
<strong>Xiang Li</strong>*, Wenhai Wang, Lijun Wu, Shuo Chen, Xiaolin Hu, Jun Li, Jinhui Tang, Jian Yang<br>
in NeurIPS, 2020<br>
[<a href="https://arxiv.org/pdf/2006.04388.pdf">Paper</a>]
[<a href="https://github.com/implus/GFocal">Code</a>]<img src="https://img.shields.io/github/stars/implus/GFocal?style=social"/>
<br>
<alert>We propose the generalized focal loss for learning the improved representations of dense object detector. GFocal is
officially included in [<a href="https://github.com/open-mmlab/mmdetection">MMDetection</a>], and is an important part of the
[<a href="https://dy.163.com/article/FLF2LGTP0511ABV6.html">winning solution</a>] in GigaVision contest (object detection and tracking tracks)
hosted in ECCV 2020 workshop (winner: DeepBlueAI team).
</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/CVPR_2019_SKNet.png"
title="Selective kernel networks">
<div><strong>Selective kernel networks</strong><br>
<strong>Xiang Li</strong>*, Wenhai Wang, Xiaolin Hu, Jian Yang<br>
in CVPR, 2019<br>
[<a href="http://openaccess.thecvf.com/content_CVPR_2019/papers/Li_Selective_Kernel_Networks_CVPR_2019_paper.pdf">Paper</a>]
[<a href="./resources/bibtex/CVPR_2019_SKNet.txt">BibTex</a>]
[<a href="https://github.com/implus/PytorchInsight">Code</a>]<img src="https://img.shields.io/github/stars/implus/PytorchInsight?style=social"/>
<br>
<alert>We propose a selective kernel mechanism for convolution.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/CVPR_2019_DIS_DROPOUT_BN.png"
title="Understanding the disharmony between dropout and batch normalization by variance shift">
<div><strong>Understanding the disharmony between dropout and batch normalization by variance shift</strong><br>
<strong>Xiang Li</strong>*, Shuo Chen, Xiaolin Hu, Jian Yang<br>
in CVPR, 2019<br>
[<a href="http://openaccess.thecvf.com/content_CVPR_2019/papers/Li_Understanding_the_Disharmony_Between_Dropout_and_Batch_Normalization_by_Variance_CVPR_2019_paper.pdf">Paper</a>]
[<a href="./resources/bibtex/CVPR_2019_DIS_DROPOUT_BN.txt">BibTex</a>]
<!--[<a href="">Code</a>]-->
<br>
<alert>We explore and address the disharmony between dropout and batch normalization.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/AAAI_2020_DIS_WN_WD.png"
title="Understanding the disharmony between weight normalization family and weight decay">
<div><strong>Understanding the disharmony between weight normalization family and weight decay</strong><br>
<strong>Xiang Li</strong>*, Shuo Chen, Jian Yang<br>
in AAAI, 2020<br>
[<a href="https://www.aaai.org/Papers/AAAI/2020GB/AAAI-LiX.1379.pdf">Paper</a>]
<!--[<a href="">Code</a>]-->
<br>
<alert>We explore and address the disharmony between weight normalization family and weight decay.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/NeurIPS_2016_LightRNN.png"
title="LightRNN: Memory and computation-efficient recurrent neural networks">
<div><strong>LightRNN: Memory and computation-efficient recurrent neural networks</strong><br>
<strong>Xiang Li</strong>*, Tao Qin, Jian Yang, Tie-Yan Liu<br>
in NeurIPS, 2016<br>
[<a href="https://papers.nips.cc/paper/6512-lightrnn-memory-and-computation-efficient-recurrent-neural-networks.pdf">Paper</a>]
[<a href="./resources/bibtex/NeurIPS_2016_LightRNN.txt">BibTex</a>]
<!--[<a href="">Code</a>]-->
<br>
<alert>We propose a memory and computation-efficient recurrent neural networks for language model.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/CVPR_2018_STCGAN.png"
title="Stacked conditional generative adversarial networks for jointly learning shadow detection and shadow removal">
<div><strong>Stacked conditional generative adversarial networks for jointly learning shadow detection and shadow removal</strong><br>
Jifeng Wang*, <strong>Xiang Li</strong>*, Jian Yang<br>
in CVPR, 2018<br>
[<a href="http://openaccess.thecvf.com/content_cvpr_2018/papers/Wang_Stacked_Conditional_Generative_CVPR_2018_paper.pdf">Paper</a>]
[<a href="./resources/bibtex/CVPR_2018_STCGAN.txt">BibTex</a>]
[<a href="https://github.com/DeepInsight-PCALab/ST-CGAN">Dataset</a>]<img src="https://img.shields.io/github/stars/DeepInsight-PCALab/ST-CGAN?style=social"/>
<br>
<alert>We release a new dataset for jointly shadow detection and removal.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/CVPR_2019_PSENet.png"
title="Shape Robust Text Detection with Progressive Scale Expansion Network">
<div><strong>Shape Robust Text Detection with Progressive Scale Expansion Network</strong><br>
Wenhai Wang*, Enze Xie*, <strong>Xiang Li</strong>*, Wenbo Hou, Tong Lu, Gang Yu, Shuai Shao<br>
in CVPR, 2019<br>
[<a href="http://openaccess.thecvf.com/content_CVPR_2019/papers/Wang_Shape_Robust_Text_Detection_With_Progressive_Scale_Expansion_Network_CVPR_2019_paper.pdf">Paper</a>]
[<a href="./resources/posters/CVPR_2019_PSENet.pdf">Poster</a>]
[<a href="./resources/bibtex/CVPR_2019_PSENet.txt">BibTex</a>]
[<a href="https://github.com/whai362/PSENet">Code</a>]<img src="https://img.shields.io/github/stars/whai362/PSENet?style=social"/>
<br>
<alert>We proposed a segmentation-based text detector that can precisely detect text instances with arbitrary shapes.</alert>
</div>
<div class="spanner"></div>
</div>
<div class="paper"><img class="paper" src="./resources/paper_icon/IJCAI_2018_MixNet.png"
title="Mixed Link Networks">
<div><strong>Mixed Link Networks</strong><br>
Wenhai Wang*, <strong>Xiang Li</strong>*, Jian Yang, Tong Lu<br>
in IJCAI, 2018<br>
[<a href="https://www.ijcai.org/Proceedings/2018/0391.pdf">Paper</a>]
[<a href="./resources/posters/IJCAI_2018_MixNet.pdf">Poster</a>]
[<a href="./resources/bibtex/IJCAI_2018_MixNet.txt">BibTex</a>]
[<a href="https://github.com/DeepInsight-PCALab/MixNet">Code</a>]<img src="https://img.shields.io/github/stars/DeepInsight-PCALab/MixNet?style=social"/>
<br>
<alert>We proposed an parameter-efficient convolutional neural networks for image classification. </alert>
</div>
<div class="spanner"></div>
</div>
</div>
</div>
<div style="clear: both;">
<div class="section">
<h2>Review Services</h2>
<div class="paper">
<strong>Journal Reviewer</strong><br>
IEEE Transactions on Neural Networks and Learning Systems (TNNLS)<br>
IEEE Transactions on Image Processing (TIP)<br>
IEEE Transactions on Multimedia (TMM)<br>
International Journal of Computer Vision (IJCV)<br>
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)<br>
<br>
<strong>Conference Reviewer</strong><br>
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020, 2021, 2022, 2023<br>
AAAI Conference on Artificial Intelligence (AAAI), 2019, 2020, 2021, 2022, 2023<br>
European Conference on Computer Vision (ECCV), 2022<br>
</div>
</div>
</div>
<div style='width:600px;height:300px;margin:0 auto'>
<!--<a href="https://clustrmaps.com/site/1b7cl" title="Visit tracker"><img src="//www.clustrmaps.com/map_v2.png?d=wHZzsZcsBCB6HxOqmaDqDJX5oLT_yvrQ5HQx9agcvJo&cl=ffffff"></a>-->
<!--<script type="text/javascript" id="clustrmaps" src="//cdn.clustrmaps.com/map_v2.js?d=wHZzsZcsBCB6HxOqmaDqDJX5oLT_yvrQ5HQx9agcvJo&cl=ffffff&w=a"></script>-->
<!--<script type="text/javascript" id="clustrmaps" src="//cdn.clustrmaps.com/map_v2.js?d=ItPniLpvAlUyBKssdgFtKwBDg8lP1ao3ju4dmDzw6uA&cl=ffffff&w=a"></script>-->
<script type='text/javascript' id='clustrmaps' src='//cdn.clustrmaps.com/map_v2.js?cl=ffffff&w=a&t=tt&d=ItPniLpvAlUyBKssdgFtKwBDg8lP1ao3ju4dmDzw6uA&co=2d78ad&cmo=3acc3a&cmn=ff5353&ct=ffffff'></script>
</div>
</body>
</html>