-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
1705 lines (1579 loc) · 107 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
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!doctype html>
<html class="no-js" lang="en">
<head>
<meta charset="utf-8">
<meta http-equiv="x-ua-compatible" content="ie=edge">
<title>Siyuan Qi @ UCLA</title>
<meta name="description" content="">
<meta name="viewport" content="width=device-width, initial-scale=1">
<link rel="shortcut icon" href="img/Q.jpg"/>
<!-- Place favicon.ico in the root directory -->
<link href="https://fonts.googleapis.com/css?family=Libre+Baskerville|Merriweather|Merriweather+Sans"
rel="stylesheet">
<link rel="stylesheet" href="css/animate.min.css">
<link rel="stylesheet" href="css/normalize.css">
<link rel="stylesheet" href="css/bootstrap.min.css">
<link rel="stylesheet" href="css/bootflat.min.css">
<link rel="stylesheet" href="css/font-awesome/css/font-awesome.css">
<link rel="stylesheet" href="css/academicons/css/academicons.css">
<link rel="stylesheet" href="css/main.css">
<link rel="stylesheet" href="css/scrolling-nav.css">
<script src="js/vendor/modernizr-2.8.3.min.js"></script>
<meta name="google-site-verification" content="TZK0p3WighxuhGYfOu7bIxXGFLn0IAi0J8xF57zCMTY" />
</head>
<body class="animated fadeIn" id="page-top" data-spy="scroll" data-target=".navbar-fixed-top">
<!--[if lte IE 9]>
<p class="browserupgrade">You are using an <strong>outdated</strong> browser. Please <a href="http://browsehappy.com/">upgrade
your browser</a> to improve your experience and security.</p>
<![endif]-->
<!-- Add your site or application content here -->
<!-- Navigation -->
<nav class="navbar navbar-default navbar-fixed-top" role="navigation"
style="font-family: 'Merriweather Sans', sans-serif;">
<div class="container">
<div class="col-lg-8">
<div class="navbar-header page-scroll">
<button type="button" class="navbar-toggle" data-toggle="collapse" data-target=".navbar-ex1-collapse">
<span class="sr-only">Toggle navigation</span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
<span class="icon-bar"></span>
</button>
<a class="navbar-brand page-scroll" href="#page-top">Siyuan Qi</a>
</div>
<!-- Collect the nav links, forms, and other content for toggling -->
<div class="collapse navbar-collapse navbar-ex1-collapse">
<ul class="nav navbar-nav">
<!-- Hidden li included to remove active class from about link when scrolled up past about section -->
<li class="hidden">
<a class="page-scroll" href="#page-top"></a>
</li>
<!--<li>-->
<!--<a class="page-scroll" href="#intro">Home</a>-->
<!--</li>-->
<li>
<a class="page-scroll" href="#about">About</a>
</li>
<li>
<a class="page-scroll" href="#news">News</a>
</li>
<li>
<a class="page-scroll" href="#projects">Projects</a>
</li>
<li>
<a class="page-scroll" href="#publications">Publications</a>
</li>
<li>
<a class="page-scroll" href="#more">More</a>
</li>
</ul>
</div>
<!-- /.navbar-collapse -->
</div>
<div class="col-lg-4">
<ul class="top-contact-info">
<li><a target="_self" href="mailto:[email protected]"><i class="fa fa-envelope"></i></a></li>
<li><a target="_blank" href="https://scholar.google.com/citations?hl=en&user=ePclJR4AAAAJ"><i
class="ai ai-google-scholar"></i></a></li>
<li><a target="_blank" href="https://www.linkedin.com/in/siyuanqi"><i class="fa fa-linkedin"></i></a>
</li>
<li><a target="_blank" href="https://github.com/SiyuanQi"><i class="fa fa-github"></i></a></li>
<li><a target="_blank" href="https://soundcloud.com/siyuanqi"><i class="fa fa-soundcloud"></i></a></li>
</ul>
</div>
</div>
<!-- /.container -->
</nav>
<!-- Intro Section -->
<section id="intro" class="intro-section">
<div class="intro-profile">
<div class="intro-avatar">
<div class="section-title">
<img class="img-circle" src="img/avatar.jpg" alt="" height="150">
</div>
</div>
<div class="intro-info">
<div class="section-title">
<h3>Siyuan Qi</h3>
<h5>PhD @ UCLA CS</h5>
</div>
<!-- <a class="btn btn-default page-scroll" href="#about">Click Me to Scroll Down!</a> -->
</div>
</div>
</section>
<!-- About Section -->
<section id="about" class="about-section">
<div class="container">
<div class="row">
<div class="col-lg-1"></div>
<div class="col-lg-10">
<div class="section-title"><h3>About Me</h3></div>
<p>I graduated from the <a href="http://www.cs.ucla.edu/">Computer Science
Department</a> at the <a href="http://www.ucla.edu/">University of California, Los Angeles</a> in
2019. During my Ph.D. study I did computer vision research in the <a href="http://vcla.stat.ucla.edu/" target="_blank">Center for Vision,
Cognition, Learning, and Autonomy</a> advised by <a href="http://www.stat.ucla.edu/~sczhu/" target="_blank">Professor Song-Chun Zhu</a>.</p>
<p>I am currently working at Google. Please refer to <a
href="https://scholar.google.com/citations?user=ePclJR4AAAAJ" target="_blank">my Google Scholar
page</a> for a more up-to-date publication list.</p>
<p>My research interests include Computer Vision, Machine Learning, and Cognitive Science.</p>
<blockquote>
<p>We who cut mere stones must always be envisioning cathedrals.</p>
<footer>Quarry worker's creed</footer>
</blockquote>
</div>
<div class="col-lg-1"></div>
</div>
</div>
</section>
<!-- News Section -->
<section id="news" class="news-section">
<div class="container">
<div class="row">
<div class="col-lg-1"></div>
<div class="col-lg-10">
<div class="section-title"><h2>News</h2></div>
</div>
<div class="col-lg-1"></div>
</div>
<div class="row">
<div class="col-md-12">
<div class="timeline">
<dl>
<dt data-toggle="collapse" href="#news-2018" style="cursor: pointer">2018</dt>
<div id="news-2018" class="panel-collapse collapse in">
<dd class="pos-right clearfix">
<div class="circ"></div>
<div class="time">2018 Sep</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">One paper accepted at <a target="_blank"
href="https://neurips.cc/Conferences/2018">NeurIPS
2018</a></h4>
<p>Cooperative Holistic 3D Scene Understanding from a Single RGB Image.</p>
<p>
[<a target="_blank"
href="publications/neurips2018cooperative/neurips2018cooperative.pdf">paper</a>]
[<a target="_blank"
href="publications/neurips2018cooperative/neurips2018cooperative_supplementary.pdf">supplementary</a>]
<!--[<a target="_blank" href="https://github.com/SiyuanQi/generalized-earley-parser">code</a>]-->
</p>
<p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Machine Learning</span>
<span class="label label-normal">Computer Vision</span>
</p>
</div>
</div>
</dd>
<dd class="pos-left clearfix">
<div class="circ"></div>
<div class="time">2018 Jul</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">Two papers accepted at <a target="_blank"
href="https://eccv2018.org/">ECCV
2018</a></h4>
<p>Learning Human-Object Interactions by Graph Parsing Neural Networks.</p>
<p>
[<a target="_blank"
href="publications/eccv2018gpnn/eccv2018gpnn.pdf">paper</a>]
[<a target="_blank" href="https://github.com/SiyuanQi/gpnn">code</a>]
</p>
<p>Holistic 3D Scene Parsing and Reconstruction from a Single RGB Image.</p>
<p>
[<a target="_blank" href="publications/eccv2018scene/eccv2018scene.pdf">paper</a>]
[<a target="_blank"
href="publications/eccv2018scene/eccv2018scene_supplementary.pdf">supplementary</a>]
[<a target="_blank"
href="https://github.com/thusiyuan/holistic_scene_parsing">code</a>]
</p>
<p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Computer Vision</span>
</p>
</div>
</div>
</dd>
<dd class="pos-right clearfix">
<div class="circ"></div>
<div class="time">2018 May</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">One paper accepted at <a target="_blank"
href="https://icml.cc/Conferences/2018">ICML
2018</a></h4>
<p>Generalized Earley Parser: Bridging Symbolic Grammars and Sequence Data for
Future Prediction.</p>
<p>
[<a target="_blank" href="publications/icml2018earley/icml2018earley.pdf">paper</a>]
[<a target="_blank"
href="publications/icml2018earley/icml2018earley_supplementary.pdf">supplementary</a>]
[<a target="_blank"
href="https://github.com/SiyuanQi/generalized-earley-parser">code</a>]
</p>
<p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Machine Learning</span>
<span class="label label-normal">Computer Vision</span>
</p>
</div>
</div>
</dd>
<dd class="pos-left clearfix">
<div class="circ"></div>
<div class="time">2018 May</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">One <a target="_blank"
href="https://link.springer.com/journal/11263">IJCV</a>
paper accepted.</h4>
<p>Configurable 3D Scene Synthesis and 2D Image Rendering with Per-Pixel Ground
Truth Using Stochastic Grammars.</p>
<p>
[<a target="_blank"
href="publications/ijcv2018synthesis/ijcv2018synthesis.pdf">paper</a>]
[<a target="_blank" href="https://vimeo.com/211226594">demo</a>]
[<a target="_blank"
href="https://github.com/SiyuanQi/human-centric-scene-synthesis">code</a>]
</p>
<p>
<span class="label label-primary">Journal</span>
<span class="label label-normal">Computer Vision</span>
<span class="label label-normal">Computer Graphics</span>
</p>
</div>
</div>
</dd>
<dd class="pos-right clearfix">
<div class="circ"></div>
<div class="time">2018 Feb</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">One paper accepted at <a target="_blank"
href="http://cvpr2018.thecvf.com/">CVPR
2018</a></h4>
<p>Human-centric Indoor Scene Synthesis Using Stochastic Grammar.</p>
<p>
[<a target="_blank"
href="publications/cvpr2018synthesis/cvpr2018synthesis.pdf">paper</a>]
[<a target="_blank"
href="publications/cvpr2018synthesis/cvpr2018synthesis_supplementary.pdf">supplementary</a>]
[<a target="_blank"
href="https://github.com/SiyuanQi/human-centric-scene-synthesis">code</a>]
[<a target="_blank"
href="http://www.yzhu.io/projects/cvpr18_scenesynthesis/index.html">project</a>]
</p>
<p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Computer Vision</span>
<span class="label label-normal">Computer Graphics</span>
</p>
</div>
</div>
</dd>
<dd class="pos-left clearfix">
<div class="circ"></div>
<div class="time">2018 Jan</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">Two papers accepted at <a target="_blank"
href="https://icra2018.org/">ICRA
2018</a></h4>
<p>Intent-aware Multi-agent Reinforcement Learning.</p>
<p>
[<a target="_blank" href="publications/icra2018intent/icra2018intent.pdf">paper</a>]
[<a target="_blank" href="https://youtu.be/UtYbynV9sK0">demo</a>]
[<a target="_blank" href="https://github.com/SiyuanQi/intentMARL">code</a>]
</p>
<p>Unsupervised Learning of Hierarchical Models for Hand-Object Interactions
Using Tactile Glove.</p>
<p>
[<a target="_blank"
href="publications/icra2018gloveaction/icra2018gloveaction.pdf">paper</a>]
[<a target="_blank" href="https://vimeo.com/259416784">demo</a>]
[<a target="_blank"
href="https://github.com/xuxie1031/UnsupervisedGloveAction">code</a>]
[<a target="_blank"
href="http://www.yzhu.io/projects/icra18_gloveaction/index.html">project</a>]
</p>
<p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Robotics</span>
</p>
</div>
</div>
</dd>
</div>
<dt data-toggle="collapse" href="#news-2017" style="cursor: pointer">2017</dt>
<div id="news-2017" class="panel-collapse collapse in">
<dd class="pos-right clearfix">
<div class="circ"></div>
<div class="time">2017 Jul</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">One paper accepted at <a target="_blank"
href="http://iccv2017.thecvf.com/">ICCV
2017</a></h4>
<p>Predicting Human Activities Using Stochastic Grammar.</p>
<p>
[<a target="_blank"
href="publications/iccv2017prediction/iccv2017prediction.pdf">paper</a>]
[<a target="_blank" href="https://youtu.be/z2gwj8AnA_c">demo</a>]
[<a target="_blank"
href="https://github.com/SiyuanQi/grammar-activity-prediction">code</a>]
</p>
<p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Computer Vision</span>
</p>
</div>
</div>
</dd>
<dd class="pos-left clearfix">
<div class="circ"></div>
<div class="time">2017 Jun</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">One paper accepted at <a target="_blank"
href="https://www.iros2017.org/">IROS
2017</a></h4>
<p>Feeling the Force: Integrating Force and Pose for Fluent Discovery through
Imitation Learning to Open Medicine Bottles.</p>
<p>
[<a target="_blank"
href="publications/iros2017openbottle/iros2017openbottle.pdf">paper</a>]
[<a target="_blank" href="https://vimeo.com/227504384">demo</a>]
[<a target="_blank"
href="https://github.com/xiaozhuchacha/OpenBottle">code</a>]
[<a target="_blank"
href="http://www.yzhu.io/projects/iros17_openbottle/index.html">project</a>]
</p>
<p>
<!--<span class="label label-success">Oral</span> -->
<span class="label label-primary">Conference</span>
<span class="label label-normal">Robotics</span>
</p>
</div>
</div>
</dd>
<dd class="pos-right clearfix">
<div class="circ"></div>
<div class="time">2017 Apr</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-user fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">Invited talk at <a target="_blank"
href="http://www.virtualrealityla.com/">VRLA
Expo 2017</a></h4>
<p><b>[Invited talk]</b> I presented our work on "Examining Human Physical
Judgments Across Virtual Gravity Fields" in VRLA 2017.</p>
<p>
<span class="label label-primary">Invited Talk</span>
<span class="label label-normal">Virtual Reality</span>
<span class="label label-normal">Cognitive Science</span>
</p>
</div>
</div>
</dd>
</div>
<dt data-toggle="collapse" href="#news-2016" style="cursor: pointer">2016</dt>
<div id="news-2016" class="panel-collapse collapse in">
<dd class="pos-left clearfix">
<div class="circ"></div>
<div class="time">2016 Nov</div>
<div class="events">
<div class="pull-left">
<p class="events-object img-rounded"><i class="fa fa-file fa-lg"></i></p>
</div>
<div class="events-body">
<h4 class="events-heading">One paper accepted at <a target="_blank"
href="http://www.ieeevr.org/2017/">IEEE
Virtual Reality 2017</a></h4>
<p><b>[Oral]</b> The Martian: Examining Human Physical Judgments Across Virtual
Gravity Fields.</p>
<p>Accepted to <a target="_blank"
href="https://www.computer.org/web/tvcg">TVCG</a></p>
<p>
[<a target="_blank" href="publications/tvcg2016gravity/tvcg2016gravity.pdf">paper</a>]
[<a target="_blank" href="https://vimeo.com/195425617">demo</a>]
[<a target="_blank"
href="http://www.yzhu.io/projects/tvcg16_gravity/index.html">project</a>]
</p>
<p>
<span class="label label-success">Oral</span>
<span class="label label-primary">Journal</span>
<span class="label label-normal">Virtual Reality</span>
<span class="label label-normal">Cognitive Science</span>
</p>
</div>
</div>
</dd>
</div>
</dl>
</div>
</div>
</div>
</div>
</section>
<!-- Projects Section -->
<section id="projects" class="projects-section">
<div class="container">
<div class="row">
<div class="col-lg-1"></div>
<div class="col-lg-10">
<div class="section-title"><h2>Projects</h2></div>
<hr>
</div>
<div class="col-lg-1"></div>
</div>
<div class="row">
<div class="col-lg-1"></div>
<div class="col-lg-10">
<!--Put the contents here-->
<!--------------- Human Prediction --------------->
<div class="panel panel-default">
<div class="panel-body">
<ul class="media-list">
<li class="media">
<a class="pull-left" target="_blank" href=""><img
class="events-object img-thumbnail img-rounded"
src="projects/thumbnails/prediction.gif" width="150" height="150"></a>
<div class="media-body">
<h4 class="media-heading">Human Activity Prediction</h4>
<p> This project aims to predict future human activities from partially
observed RGB-D videos. Human activity prediction is generally difficult
due to its non-Markovian property and the rich context between human and
environments. We use a stochastic grammar model to capture the
compositional/hierarchical structure of events, integrating human
actions, objects, and their affordances. </p>
<span class="label label-primary">Computer Vision</span>
<span class="label label-primary">Robotics</span>
</div>
</li>
</ul>
</div>
</div>
<!--------------- Scene Synthesis --------------->
<div class="panel panel-default">
<div class="panel-body">
<ul class="media-list">
<li class="media">
<a class="pull-left" target="_blank" href=""><img
class="events-object img-thumbnail img-rounded"
src="projects/thumbnails/scenesynthesis.gif" width="150" height="150"></a>
<div class="media-body">
<h4 class="media-heading">Indoor Scene Synthesis by Stochastic Grammar</h4>
<p>This project studies how to realistically synthesis indoor scene layouts
using stochastic grammar. We present a novel human-centric method to
sample 3D room layouts and synthesis photo-realistic images using
physics-based rendering. We use object affordance and human activity
planning to model indoor scenes, which contains functional grouping
relations and supporting relations between furniture and objects. An
attributed spatial And-Or graph (S-AOG) is proposed to model indoor
scenes. The S-AOG is a stochastic context sensitive grammar, in which
the terminal nodes are object entities including room, furniture and
supported objects. </p>
<span class="label label-primary">Computer Vision</span>
<span class="label label-primary">Computer Graphics</span>
</div>
</li>
</ul>
</div>
</div>
</div>
</div>
</div>
</section>
<!-- Publications Section -->
<section id="publications" class="publications-section">
<div class="container">
<div class="row">
<div class="col-lg-1"></div>
<div class="col-lg-10">
<div class="section-title"><h2>Publications</h2></div>
<hr>
</div>
<div class="col-lg-1"></div>
</div>
<div class="row">
<div class="col-lg-1"></div>
<div class="col-lg-10">
<!--------------- Cooperative Scene Parsing --------------->
<div class="panel panel-default">
<!--<div class="panel-heading">-->
<!--<h3 class="panel-title">2016</h3>-->
<!--</div>-->
<div class="panel-body" data-toggle="collapse" href="#pub-abstract-neurips2018cooperative"
style="cursor: pointer">
<ul class="media-list">
<li class="media">
<a class="pull-left" target="_blank" href=""><img
class="events-object img-thumbnail img-rounded"
src="publications/thumbnails/neurips2018cooperative.png" width="150"
height="150"></a>
<div class="media-body">
<h4 class="media-heading">Cooperative Holistic Scene Understanding: Unifying
3D Object, Layout, and Camera Pose Estimation</h4>
<p>Siyuan Huang, <b>Siyuan Qi</b>, Yinxue Xiao, Yixin Zhu, Ying Nian Wu, Song-Chun
Zhu.</p>
<p>neurips 2018, Montreal, Canada</p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Machine Learning</span>
<span class="label label-normal">Computer Vision</span>
</div>
</li>
<li>
<div align="justify" id="pub-abstract-neurips2018cooperative"
class="panel-collapse collapse">
<hr>
<h5>Abstract</h5>
Holistic 3D indoor scene understanding refers to jointly recovering the i) object
bounding boxes, ii) room layout, and iii) camera pose, all in 3D. The existing
methods either are ineffective or only tackle the problem partially. In this paper,
we propose an end-to-end model that simultaneously solves all three tasks in
realtime
given only a single RGB image. The essence of the proposed method is to
improve the prediction by i) parametrizing the targets (e.g., 3D boxes) instead of
directly estimating the targets, and ii) cooperative training across different
modules
in contrast to training these modules individually. Specifically, we parametrize
the 3D object bounding boxes by the predictions from several modules, i.e., 3D
camera pose and object attributes. The proposed method provides two major
advantages: i) The parametrization helps maintain the consistency between the
2D image and the 3D world, thus largely reducing the prediction variances in
3D coordinates. ii) Constraints can be imposed on the parametrization to train
different modules simultaneously. We call these constraints "cooperative losses" as
they enable the joint training and inference. We employ three cooperative losses
for 3D bounding boxes, 2D projections, and physical constraints to estimate a
geometrically consistent and physically plausible 3D scene. Experiments on the
SUN RGB-D dataset shows that the proposed method significantly outperforms
prior approaches on 3D object detection, 3D layout estimation, 3D camera pose
estimation, and holistic scene understanding.
</div>
</li>
</ul>
</div>
<ul class="list-group">
<li class="list-group-item">
<span class="pull-right">
<a target="_blank" data-toggle="collapse"
title="bibtex" href="#pub-bibtex-neurips2018cooperative">
<i class="fa fa-pencil fa-lg"></i>
</a>
<a target="_blank" data-toggle="tooltip"
title="Paper"
href="publications/neurips2018cooperative/neurips2018cooperative.pdf">
<i class="fa fa-file-pdf-o fa-lg"></i>
</a>  
<a target="_blank" data-toggle="tooltip"
title="Supplementary"
href="publications/neurips2018cooperative/neurips2018cooperative_supplementary.pdf">
<i class="fa fa-file-pdf-o fa-lg"></i>
</a>  
</span>
</li>
</ul>
<div align="justify" id="pub-bibtex-neurips2018cooperative" class="panel-collapse collapse">
<pre>
@inproceedings{huang2018cooperative,
title={Cooperative Holistic Scene Understanding: Unifying
3D Object, Layout, and Camera Pose Estimation},
author={Huang, Siyuan and Qi, Siyuan and Xiao, Yinxue and Zhu, Yixin and Wu, Ying Nian and Zhu, Song-Chun},
booktitle={Conference on Neural Information Processing Systems (NeurIPS)},
year={2018}
}</pre>
</div>
</div>
<!--------------- Graph Parsing Neural Network --------------->
<div class="panel panel-default">
<!--<div class="panel-heading">-->
<!--<h3 class="panel-title">2016</h3>-->
<!--</div>-->
<div class="panel-body" data-toggle="collapse" href="#pub-abstract-eccv2018gpnn"
style="cursor: pointer">
<ul class="media-list">
<li class="media">
<a class="pull-left" target="_blank" href=""><img
class="events-object img-thumbnail img-rounded"
src="publications/thumbnails/eccv2018gpnn.png" width="150" height="150"></a>
<div class="media-body">
<h4 class="media-heading">Learning Human-Object Interactions by Graph Parsing Neural
Networks</h4>
<p><b>Siyuan Qi*</b>, Wenguan Wang*, Baoxiong Jia, Jianbing Shen, Song-Chun Zhu.</p>
<p>ECCV 2018, Munich, Germany</p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Computer Vision</span>
</div>
</li>
<li>
<div align="justify" id="pub-abstract-eccv2018gpnn" class="panel-collapse collapse">
<hr>
<h5>Abstract</h5>
This paper addresses the task of detecting and recognizing human-object interactions
(HOI) in images and videos.
We introduce the Graph Parsing Neural Network (GPNN), a framework that incorporates
structural knowledge while being differentiable end-to-end.
For a given scene, GPNN infers a parse graph that includes i) the HOI graph
structure represented by an adjacency matrix, and ii) the node labels.
Within a message passing inference framework, GPNN iteratively computes the
adjacency matrices and node labels.
We extensively evaluate our model on three HOI detection benchmarks on images and
videos: HICO-DET, V-COCO, and CAD-120 datasets.
Our approach significantly outperforms state-of-art methods, verifying that GPNN is
scalable to large datasets and applies to spatial-temporal settings.
</div>
</li>
</ul>
</div>
<ul class="list-group">
<li class="list-group-item">
<span class="pull-right">
<a target="_blank" data-toggle="collapse"
title="bibtex" href="#pub-bibtex-eccv2018gpnn">
<i class="fa fa-pencil fa-lg"></i>
</a>
<a target="_blank" data-toggle="tooltip"
title="Paper" href="publications/eccv2018gpnn/eccv2018gpnn.pdf">
<i class="fa fa-file-pdf-o fa-lg"></i>
</a>  
<a target="_blank" data-toggle="tooltip"
title="Code" href="https://github.com/SiyuanQi/gpnn">
<i class="fa fa-code fa-lg"></i>
</a>
</span>
</li>
</ul>
<div align="justify" id="pub-bibtex-eccv2018gpnn" class="panel-collapse collapse">
<pre>
@inproceedings{qi2018learning,
title={Learning Human-Object Interactions by Graph Parsing Neural Networks},
author={Qi, Siyuan and Wang, Wenguan and Jia, Baoxiong and Shen, Jianbing and Zhu, Song-Chun},
booktitle={European Conference on Computer Vision (ECCV)},
year={2018}
}</pre>
</div>
</div>
<!--------------- Holistic Scene Grammar --------------->
<div class="panel panel-default">
<!--<div class="panel-heading">-->
<!--<h3 class="panel-title">2016</h3>-->
<!--</div>-->
<div class="panel-body" data-toggle="collapse" href="#pub-abstract-eccv2018scene"
style="cursor: pointer">
<ul class="media-list">
<li class="media">
<a class="pull-left" target="_blank" href=""><img
class="events-object img-thumbnail img-rounded"
src="publications/thumbnails/eccv2018scene.png" width="150" height="150"></a>
<div class="media-body">
<h4 class="media-heading">Holistic 3D Scene Parsing and Reconstruction from a Single
RGB Image</h4>
<p>Siyuan Huang, <b>Siyuan Qi</b>, Yixin Zhu, Yinxue Xiao, Yuanlu Xu, Song-Chun Zhu.
</p>
<p>ECCV 2018, Munich, Germany</p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Computer Vision</span>
</div>
</li>
<li>
<div align="justify" id="pub-abstract-eccv2018scene" class="panel-collapse collapse">
<hr>
<h5>Abstract</h5>
We propose a computational framework to jointly parse a
single RGB image and reconstruct a holistic 3D configuration composed
by a set of CAD models using a stochastic grammar model. Specifically,
we introduce a Holistic Scene Grammar (HSG) to represent the 3D scene
structure, which characterizes a joint distribution over the functional and
geometric space of indoor scenes. The proposed HSG captures three essential
and often latent dimensions of the indoor scenes: i) latent human
context, describing the affordance and the functionality of a room arrangement,
ii) geometric constraints over the scene configurations, and
iii) physical constraints that guarantee physically plausible parsing and
reconstruction. We solve this joint parsing and reconstruction problem
in an analysis-by-synthesis fashion, seeking to minimize the differences
between the input image and the rendered images generated by our 3D
representation, over the space of depth, surface normal, and object segmentation
map. The optimal configuration, represented by a parse graph,
is inferred using Markov chain Monte Carlo (MCMC), which efficiently
traverses through the non-differentiable solution space, jointly optimizing
object localization, 3D layout, and hidden human context. Experimental
results demonstrate that the proposed algorithm improves the generalization
ability and significantly outperforms prior methods on 3D layout
estimation, 3D object detection, and holistic scene understanding.
</div>
</li>
</ul>
</div>
<ul class="list-group">
<li class="list-group-item">
<span class="pull-right">
<a target="_blank" data-toggle="collapse"
title="bibtex" href="#pub-bibtex-eccv2018scene">
<i class="fa fa-pencil fa-lg"></i>
</a>
<a target="_blank" data-toggle="tooltip"
title="Paper" href="publications/eccv2018scene/eccv2018scene.pdf">
<i class="fa fa-file-pdf-o fa-lg"></i>
</a>  
<a target="_blank" data-toggle="tooltip"
title="Supplementary"
href="publications/eccv2018scene/eccv2018scene_supplementary.pdf">
<i class="fa fa-file-pdf-o fa-lg"></i>
</a>  
<a target="_blank" data-toggle="tooltip"
title="Code" href="https://github.com/thusiyuan/holistic_scene_parsing">
<i class="fa fa-code fa-lg"></i>
</a>
</span>
</li>
</ul>
<div align="justify" id="pub-bibtex-eccv2018scene" class="panel-collapse collapse">
<pre>
@inproceedings{huang2018holistic,
title={Holistic 3D Scene Parsing and Reconstruction from a Single RGB Image},
author={Huang, Siyuan and Qi, Siyuan and Zhu, Yixin and Xiao, Yinxue and Xu, Yuanlu and Zhu, Song-Chun},
booktitle={European Conference on Computer Vision (ECCV)},
year={2018}
}</pre>
</div>
</div>
<!--------------- Generalized Earley Parser --------------->
<div class="panel panel-default">
<!--<div class="panel-heading">-->
<!--<h3 class="panel-title">2016</h3>-->
<!--</div>-->
<div class="panel-body" data-toggle="collapse" href="#pub-abstract-icml2018earley"
style="cursor: pointer">
<ul class="media-list">
<li class="media">
<a class="pull-left" target="_blank" href=""><img
class="events-object img-thumbnail img-rounded"
src="publications/thumbnails/icml2018earley.jpg" width="150" height="150"></a>
<div class="media-body">
<h4 class="media-heading">Generalized Earley Parser: Bridging Symbolic Grammars and
Sequence Data for Future Prediction</h4>
<p><b>Siyuan Qi</b>, Baoxiong Jia, Song-Chun Zhu.</p>
<p>ICML 2018, Stockholm, Sweden</p>
<span class="label label-primary">Conference</span>
<span class="label label-normal">Machine Learning</span>
<span class="label label-normal">Computer Vision</span>
</div>
</li>
<li>
<div align="justify" id="pub-abstract-icml2018earley" class="panel-collapse collapse">
<hr>
<h5>Abstract</h5>
Future predictions on sequence data (e.g., videos or audios) require the algorithms
to capture non-Markovian and compositional properties of high-level semantics.
Context-free grammars are natural choices to capture such properties, but
traditional grammar parsers (e.g., Earley parser) only take symbolic sentences as
input.
In this paper, we generalize the Earley parser to parse sequence data which is
neither segmented nor labeled.
This generalized Earley parser integrates a grammar parser with a classifier to find
the optimal segmentation and labels, and makes top-down future predictions
accordingly.
Experiments show that our method significantly outperforms other approaches for
future human activity prediction.
</div>
</li>
</ul>
</div>
<ul class="list-group">
<li class="list-group-item">
<span class="pull-right">
<a target="_blank" data-toggle="collapse"
title="bibtex" href="#pub-bibtex-icml2018earley">
<i class="fa fa-pencil fa-lg"></i>
</a>
<a target="_blank" data-toggle="tooltip"
title="Paper" href="publications/icml2018earley/icml2018earley.pdf">
<i class="fa fa-file-pdf-o fa-lg"></i>
</a>  
<a target="_blank" data-toggle="tooltip"
title="Supplementary"
href="publications/icml2018earley/icml2018earley_supplementary.pdf">
<i class="fa fa-file-pdf-o fa-lg"></i>
</a>  
<a target="_blank" data-toggle="tooltip"
title="Code" href="https://github.com/SiyuanQi/generalized-earley-parser">
<i class="fa fa-code fa-lg"></i>
</a>
</span>
</li>
</ul>
<div align="justify" id="pub-bibtex-icml2018earley" class="panel-collapse collapse">
<pre>
@inproceedings{qi2018generalized,
title={Generalized Earley Parser: Bridging Symbolic Grammars and Sequence Data for Future Prediction},
author={Qi, Siyuan and Jia, Baoxiong and Zhu, Song-Chun},
booktitle={International Conference on Machine Learning (ICML)},
year={2018}
}</pre>
</div>
</div>
<!--------------- IJCV Scene Synthesis --------------->
<div class="panel panel-default">
<!--<div class="panel-heading">-->
<!--<h3 class="panel-title">2016</h3>-->
<!--</div>-->
<div class="panel-body" data-toggle="collapse" href="#pub-abstract-ijcv2018synthesis"
style="cursor: pointer">
<ul class="media-list">
<li class="media">
<a class="pull-left" target="_blank" href=""><img
class="events-object img-thumbnail img-rounded"
src="publications/thumbnails/ijcv2018synthesis.gif" width="150" height="150"></a>
<div class="media-body">
<h4 class="media-heading">Configurable 3D Scene Synthesis and 2D Image Rendering
with Per-Pixel Ground Truth Using Stochastic Grammars</h4>
<p>Chenfanfu Jiang*, <b>Siyuan Qi*</b>, Yixin Zhu*, Siyuan Huang*,
Jenny Lin, Lap-Fai Yu, Demetri Terzopoulos, Song-Chun Zhu.</p>
<p>IJCV 2018</p>
<span class="label label-primary">Journal</span>
<span class="label label-normal">Computer Vision</span>
<span class="label label-normal">Computer Graphics</span>
</div>
</li>
<li>
<div align="justify" id="pub-abstract-ijcv2018synthesis"
class="panel-collapse collapse">
<hr>
<h5>Abstract</h5>
We propose a systematic learning-based approach to the generation of massive
quantities of synthetic 3D scenes and numerous photorealistic 2D images thereof,
with associated ground truth information, for the purposes of training,
benchmarking, and diagnosing learning-based computer vision and robotics algorithms.
In particular, we devise a learning-based pipeline of algorithms capable of
automatically generating and rendering a potentially infinite variety of indoor
scenes by using a stochastic grammar, represented as an attributed Spatial And-Or
Graph, in conjunction with state-of-the-art physics-based rendering. Our pipeline is
capable of synthesizing scene layouts with high diversity, and it is configurable in
that it enables the precise customization and control of important attributes of the
generated scenes. It renders photorealistic RGB images of the generated scenes while
automatically synthesizing detailed, per-pixel ground truth data, including visible
surface depth and normal, object identity, and material information (detailed to
object parts), as well as environments (e.g., illumination and camera viewpoints).
We demonstrate the value of our dataset, by improving performance in certain
machine-learning-based scene understanding tasks--e.g., depth and surface normal
prediction, semantic segmentation, reconstruction, etc.---and by providing
benchmarks for and diagnostics of trained models by modifying object attributes and
scene properties in a controllable manner.
</div>
</li>
</ul>
</div>
<ul class="list-group">
<li class="list-group-item">
<span class="pull-right">
<a target="_blank" data-toggle="collapse"
title="bibtex" href="#pub-bibtex-ijcv2018synthesis">
<i class="fa fa-pencil fa-lg"></i>
</a>
<a target="_blank" data-toggle="tooltip"
title="Paper" href="publications/ijcv2018synthesis/ijcv2018synthesis.pdf">
<i class="fa fa-file-pdf-o fa-lg"></i>
</a>
<a target="_blank" data-toggle="tooltip"
title="Demo" href="https://vimeo.com/211226594">
<i class="fa fa-file-video-o fa-lg"></i>
</a>
<a target="_blank" data-toggle="tooltip"
title="Project page"
href="http://www.yzhu.io/projects/arxiv_scenesynthesis/index.html">
<i class="fa fa-expand fa-lg"></i>
</a>
</span>
</li>
</ul>
<div align="justify" id="pub-bibtex-ijcv2018synthesis" class="panel-collapse collapse">
<pre>
@article{jiang2018configurable,
title={Configurable 3D Scene Synthesis and 2D Image Rendering with Per-Pixel Ground Truth Using Stochastic Grammars},
author={Jiang, Chenfanfu and Qi, Siyuan and Zhu, Yixin and Huang, Siyuan and Lin, Jenny and Yu, Lap-Fai and Terzopoulos, Demetri, Zhu, Song-Chun},
journal = {International Journal of Computer Vision (IJCV)},
year={2018}
}</pre>
</div>
</div>
<!--------------- Human-centric Indoor Scene Synthesis --------------->
<div class="panel panel-default">
<!--<div class="panel-heading">-->
<!--<h3 class="panel-title">2016</h3>-->
<!--</div>-->
<div class="panel-body" data-toggle="collapse" href="#pub-abstract-cvpr2018synthesis"
style="cursor: pointer">
<ul class="media-list">
<li class="media">
<a class="pull-left" target="_blank" href=""><img
class="events-object img-thumbnail img-rounded"
src="publications/thumbnails/cvpr2018synthesis.gif" width="150" height="150"></a>
<div class="media-body">