en
0.25
0.5
0.75
1.25
1.5
1.75
2
PanoContext: A Whole-room 3D Context Model for Panoramic Scene Understanding
Published on Oct 29, 20148072 Views
The field-of-view of standard cameras is very small, which is one of the main reasons that contextual information is not as useful as it should be for object detection. To overcome this limitation, we
Related categories
Chapter list
PanoContext: A Whole-room 3D Context Model for Panoramic Scene Understanding00:00
Context is important - 100:09
Context is important - 200:16
Context is important - 300:17
Context is important - 400:21
Context models00:29
What’s the problem in context?00:53
Game: What is this object?01:14
What is this object? - 101:25
What is this object? - 201:30
What is this object? - 301:31
What is this object? - 401:36
What is this object? - 501:36
What is this object? - 601:50
What is this object? - 701:52
What is this object? - 801:58
What is this object? - 901:59
What is this object? - 1002:00
What is this object? - 1102:00
What is this object? - 1202:07
Picture02:46
Small FOV, you will miss a lot!02:52
How does large FOV help? - 103:01
How does large FOV help? - 203:46
PanoContext - 103:59
PanoContext - 204:16
PanoContext - 304:26
PanoContext - 404:43
PanoContext - 504:54
PanoContext - 604:54
PanoContext - 704:55
PanoContext - 805:03
PanoContext - 905:14
PanoContext - 1005:15
PanoContext - 1105:20
PanoContext - 1205:35
PanoContext - 1305:40
PanoContext - 1405:44
PanoContext - 1505:46
PanoContext - 1605:52
PanoContext - 1706:00
Context model ≥ Object detector06:13
Algorithm - 106:35
Algorithm - 206:39
Algorithm - 306:58
Algorithm - 407:08
Room layout hypothesis - 107:15
Room layout hypothesis - 207:22
Room layout hypothesis - 307:24
Room layout hypothesis - 407:26
Room layout hypothesis - 507:30
Room layout hypothesis - 607:37
Room layout hypothesis - 707:39
Room layout hypothesis - 807:40
Room layout hypothesis - 907:53
Algorithm - 507:55
Cuboid detection07:59
From 2D to 3D08:08
Semantic classification - 108:17
Semantic classification - 208:23
Semantic classification - 308:29
Semantic classification - 408:31
Semantic classification - 508:32
Semantic classification - 608:34
2/3D annotation on panorama - 108:52
2/3D annotation on panorama - 209:11
Annotated panorama dataset09:17
Semantic classification09:24
Hypotheses for some categories - 109:29
Hypotheses for some categories - 209:34
Hypotheses for some categories - 309:35
Algorithm - 609:37
Data-driven sampling - 109:45
Data-driven sampling - 310:21
Data-driven sampling - 410:34
Data-driven sampling - 510:37
Data-driven sampling - 610:41
Data-driven sampling - 710:54
Data-driven sampling - 810:58
Data-driven sampling - 910:59
Data-driven sampling - 1011:02
Data-driven sampling - 1111:04
Data-driven sampling - 1211:05
Data-driven sampling - 1311:05
Data-driven sampling - 1411:06
Data-driven sampling - 1511:07
Data-driven sampling - 1611:07
Data-driven sampling - 1711:08
Algorithm - 711:12
Algorithm - 811:21
Holistic ranking - 111:23
Holistic ranking - 211:30
Holistic ranking - 311:34
Holistic feature - 111:41
Holistic feature - 211:52
Holistic feature - 311:57
Holistic feature - 412:24
Holistic feature - 512:30
Final output12:35
Living room12:42
Automatic Recognition Results12:46
Analysis12:51
How does 3D context help? - 112:56
How does 3D context help? - 213:16
How does 3D context help? - 313:28
Context vs. Appearance - 113:41
Context vs. Appearance - 314:00
Is larger FOV helpful? - 114:12
Is larger FOV helpful? - 214:29
PanoContext - 1814:38
Data and code available: http://panocontext.cs.princeton.edu15:03
Context vs. Appearance - 215:23
Data-driven sampling - 216:24