Robust Video Object Segmentation Based on K-Means Background Clustering and Watershed in Ill-Conditioned Surveillance Systems.
Tse-Wei ChenShou-Chieh HsuShao-Yi ChienPublished in: ICME (2007)
Keyphrases
- k means
- surveillance system
- clustering algorithm
- clustering method
- background subtraction
- hierarchical clustering
- video surveillance
- global motion estimation
- region growing
- motion detection
- spectral clustering
- computer vision algorithms
- foreground objects
- cluster centers
- dynamic background
- ptz camera
- background model
- multiscale
- multi camera
- omni directional
- foreground detection
- segmentation method
- expectation maximization
- computer vision
- clustering quality
- multi class