Tracking Human Under Occlusion Based on Adaptive Multiple Kernels With Projected Gradients.
Chun-Te ChuJenq-Neng HwangHung-I PaiKung-Ming LanPublished in: IEEE Trans. Multim. (2013)
Keyphrases
- partially occluded
- occlusion reasoning
- real time
- multiple objects
- particle filter
- occlusion handling
- object tracking
- occlusion detection
- visual tracking
- human behavior
- multiple targets
- cluttered background
- human walking
- linear predictors
- tracking objects
- human operators
- kernel methods
- human activities
- appearance model
- motion model
- linear combination
- video sequences