0.25
0.5
0.75
1.25
1.5
1.75
2
Linking People in Videos with “Their” Names Using Coreference Resolution
Published on Oct 29, 20143491 Views
Natural language descriptions of videos provide a potentially rich and vast source of supervision. However, the highly-varied nature of language presents a major barrier to its effective use. What is
Related categories
Chapter list
Linking people in videos with “their” names using coreference resolution00:00
Video Data00:08
Video Data + Text Data - 100:18
Video Data + Text Data - 200:26
Video Data + Text Data - 300:36
Video Data + Text Data - 400:42
Video Data + Text Data - 500:50
Text to Video00:58
Unidirectional Models: Text to Video - 101:17
Unidirectional Models: Text to Video - 201:23
Unidirectional Models: Text to Video - 301:29
Unidirectional Models: Text to Video - 401:40
Temporal ordering of people - 101:50
Temporal ordering of people - 202:00
Temporal ordering of people - 302:09
Coreference Resolution in NLP - 102:20
Coreference Resolution in NLP - 202:30
Coreference Resolution in NLP - 302:39
Coreference Resolution in NLP - 402:55
Coreference Resolution in NLP - 503:02
Coreference Resolution in NLP - 603:12
Coreference Resolution in NLP - 703:27
Coreference Resolution in NLP - 803:35
Coreference Resolution in NLP - 903:41
Coreference Resolution in NLP - 1003:46
Bidirectional model03:56
Linking people with coreference resolution04:03
Linking people with “their” names - 104:16
Problem Setup (Input) - 104:21
Problem Setup (Input) - 204:29
Problem Setup (Input) - 304:36
Problem Setup (Input) - 404:49
Problem Setup (Output) - 104:55
Problem Setup (Output) - 205:01
Problem Setup (Output) - 305:05
Linking people with “their” names - 205:11
Our Bidirectional Model - 105:16
Our Bidirectional Model - 205:22
Our Bidirectional Model - 305:26
Our Bidirectional Model - 405:31
Our Bidirectional Model: Track name assignment - 105:36
Our Bidirectional Model: Track name assignment - 205:40
Our Bidirectional Model: Track name assignment - 305:46
Our Bidirectional Model: Track name assignment - 405:56
Our Bidirectional Model: Track name assignment - 505:59
Our Bidirectional Model: Track name assignment - 606:10
Our Bidirectional Model: Track name assignment - 706:14
Our Bidirectional Model: Track name assignment - 806:36
Our Bidirectional Model: Track name assignment - 906:42
Our Bidirectional Model: Track name assignment - 1006:57
Our Bidirectional Model: Track name assignment - 1107:02
Our Bidirectional Model - 507:07
Our Bidirectional Model - 607:14
Our Bidirectional Model: Mention name assignment - 107:19
Our Bidirectional Model: Mention name assignment - 207:24
Our Bidirectional Model: Mention name assignment - 307:38
Our Bidirectional Model: Mention name assignment - 407:52
Our Bidirectional Model: Mention name assignment - 508:09
Our Bidirectional Model: Mention name assignment - 608:18
Our Bidirectional Model: Mention name assignment - 708:26
Our Bidirectional Model: Mention name assignment - 808:33
Our Bidirectional Model: Mention name assignment - 908:38
Our Bidirectional Model: Mention name assignment - 1008:46
Our Bidirectional Model: Mention name assignment - 1108:54
Our Bidirectional Model: Mention name assignment - 1209:05
Our Bidirectional Model: Mention name assignment - 1309:11
Our Bidirectional Model: Mention name assignment - 1409:21
Our Bidirectional Model - 709:26
Our Bidirectional Model - 809:35
Our Bidirectional Model: Alignment - 109:40
Our Bidirectional Model: Alignment - 209:47
Our Bidirectional Model: Alignment - 310:08
Our Bidirectional Model: Alignment - 410:17
Our Bidirectional Model: Alignment - 510:24
Our Bidirectional Model: Alignment - 610:32
Our Bidirectional Model - 910:44
Optimization - 110:48
Optimization - 211:00
Optimization - 311:06
Optimization - 411:10
Linking people with “their” names - 311:16
Dataset - 111:19
Dataset - 211:24
Dataset - 311:31
Dataset - 411:41
Dataset - 511:46
Dataset - 611:48
Experiments: Name Assignment to Tracks - 111:53
Experiments: Name Assignment to Tracks - 212:05
Experiments: Name Assignment to Tracks - 312:23
Experiments: Name Assignment to Tracks - 412:36
Dataset - 712:44
Experiments: Name Assignment to Mentions - 112:49
Experiments: Name Assignment to Mentions - 213:01
Experiments: Name Assignment to Mentions - 313:15
Experiments: Benefit of bidirectional model - 113:26
Experiments: Benefit of bidirectional model - 213:39
Experiments: Benefit of bidirectional model - 314:03
Experiments: Benefit of bidirectional model - 414:14
Error Analysis: Name Assignment to Mentions - 114:18
Error Analysis: Name Assignment to Mentions - 214:33
Error Analysis: Name Assignment to Mentions - 314:40
Linking people with “their” names - 414:52
Summary14:54
Thank you15:08