Cross-View Action Recognition via a Transferable Dictionary Pair 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

Cross-View Action Recognition via a Transferable Dictionary Pair

Published on Oct 09, 20124320 Views

Discriminative appearance features are effective for recognizing actions in a fixed view, but generalize poorly to changes in viewpoint. We present a method for viewinvariant action recognition base

Related categories

Chapter list

Cross-View Action Recognition via a Transferable Dictionary Pair00:00
Cross-view Action Recognition00:18
Our Framework02:01
Our Framework-203:38
Our Framework-304:48
Our Framework-406:06
Review of Dictionary Learning06:24
Unsupervised Transferable Dictionary Pair Learning07:16
Supervised dictionary pair learning08:04
Extension: multi-view action recognition10:51
Experiment: IXMAS multi-view dataset11:43
Experiment Results (1)12:44
Experiment Results (2)14:35
Experiment Results (3)14:55
Experiment Results (4)15:43
Experiment Results (5)16:11
Conclusions17:10
Thank You!17:40