Multiple pedestrian tracking based on couple-states Markov chain with semantic topic learning for video surveillance.
Peng ZhangLiang WangWei HuangLei XieGuang ChenPublished in: Soft Comput. (2015)
Keyphrases
- video surveillance
- markov chain
- pedestrian tracking
- transition probabilities
- learning algorithm
- background subtraction
- motion detection
- computer vision
- transition matrix
- event detection
- moving objects
- prior knowledge
- multi target tracking
- reinforcement learning
- high dimensional data
- object tracking
- supervised learning
- multi class
- state space
- pedestrian detection
- infrared imagery
- multiscale