Real-time outdoor video surveillance with robust foreground extraction and object tracking via multi-state transition management.
Bang Jun LeiLi-Qun XuPublished in: Pattern Recognit. Lett. (2006)
Keyphrases
- video surveillance
- object tracking
- state transition
- surveillance system
- dynamic background
- video surveillance systems
- moving object detection
- moving objects
- surveillance videos
- partial occlusion
- illumination variations
- image sequences
- traffic monitoring
- background subtraction
- foreground objects
- static camera
- particle filter
- multiple cameras
- background modeling
- mean shift
- motion detection
- robust tracking
- appearance model
- video sequences
- visual surveillance
- computer vision
- foreground segmentation
- state space
- input output
- hidden markov models
- multi camera
- black box
- visual tracking
- surveillance cameras
- background model
- sparse representation
- probability distribution
- data association
- shadow detection
- bayesian networks
- camera motion
- higher order