A Unified Framework for Multi-Target Tracking and Collective Activity Recognition thumbnail
slide-image
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

A Unified Framework for Multi-Target Tracking and Collective Activity Recognition

Published on Nov 12, 20127056 Views

We present a coherent, discriminative framework for simultaneously tracking multiple people and estimating their collective activities. Instead of treating the two problems separately, our model is g

Related categories

Chapter list

A Unified Framework for Multi-Target Tracking and Collective Activity Recognition00:00
Our Goal (1)00:04
Our Goal (2)00:06
Our Goal (3)00:10
Our Goal (4)00:14
Our Goal (5)00:21
Our Goal (6)00:36
Our Goal (7)00:51
Background (1)01:00
Background (2)01:06
Background (3)01:18
Contributions (1)01:27
Contributions (2)01:43
Contributions (3)01:55
Contributions (4)02:09
Contributions (5)02:40
Outline02:56
Joint Model02:59
Hierarchical Activity Model (1)03:02
Hierarchical Activity Model (2)03:19
Hierarchical Activity Model (3)03:48
Hierarchical Activity Model (4)03:58
Hierarchical Activity Model (5)04:18
Hierarchical Activity Model (6)04:50
Collective-Interaction Potential (1)05:03
Collective-Interaction Potential (2)05:43
Collective-Observation Potential 05:56
Activity Transition Potential06:10
Trajectory Estimation06:22
Tracklet Association Problem (1)06:37
Tracklet Association Problem (2)06:49
Tracklet Association Model (1)07:08
Tracklet Association Model (2)07:44
Tracklet Association Model (3)08:13
Tracklet Association Model (4)08:23
Inference/Training method08:28
Inference (1)08:33
Inference (2)08:48
Training09:15
Experimental evaluation09:23
Experiments (1)09:29
Experiments (2)09:48
Classification Results10:00
Target Association (1)10:29
Target Association (2)10:50
Target Association (3)10:59
Target Association (4)11:10
Example Classification Result (1)11:15
Example Classification Result (2)11:32
Example Classification Result (3)11:54
Example Classification Result (4)12:03
Example Classification Result (5)12:10
Conclusion (1)12:46
Conclusion (2)12:53
Conclusion (3)13:01
Thanks to13:07