0.25
0.5
0.75
1.25
1.5
1.75
2
Visual Tracking by Sampling Tree-Structured Graphical Models
Published on Oct 29, 20143610 Views
Probabilistic tracking algorithms typically rely on graphical models based on the first-order Markov assumption. Although such linear structure models are simple and reasonable, it is not appropriate
Related categories
Chapter list
Visual Tracking by Sampling Tree-Structured Graphical Models00:00
Goal of Visual Tracking00:13
Conventional Tracking Approaches00:22
Orderless Tracking [HongICCV2013] - 100:45
OrderlessTracking [HongICCV2013] - 201:09
Our Approach - 101:35
Tracking on Tree-Structure01:50
Challenges - 102:35
Our Approach - 202:47
Sampling Tree Structure by MCMC03:23
Challenges - 204:08
Proposing A New Tree - 104:31
Proposing A New Tree - 204:57
Validating A New Tree - 105:23
Validating A New Tree - 206:09
Tracking on Tree Structure - 106:43
Tracking on Tree Structure - 207:18
Identified Tree Structure - 108:01
Identified Tree Structure - 208:20
Computational Complexity - 108:30
Computational Complexity - 209:07
Computational Complexity - 309:30
Hierarchical Approach09:45
Key Frame Selection [HongICCV2013]10:14
Tree Extension by Manifold Alignment - 110:54
Tree Extension by Manifold Alignment - 211:13
Tree Extension by Manifold Alignment - 311:49
Tree Extension by Manifold Alignment - 412:14
Qualitative Results - 112:37
Qualitative Results - 213:20
Quantitative Results13:46
Summary14:13
Advertisement14:41