Improved Background Mixture Models for Video Surveillance Applications.
Chris PoppeGaëtan MartensPeter LambertRik Van de WallePublished in: ACCV (1) (2007)
Keyphrases
- mixture model
- gaussian mixture model
- em algorithm
- video surveillance
- generative model
- probabilistic model
- density estimation
- video sequences
- model selection
- human activities
- unsupervised learning
- background subtraction
- video content
- finite mixtures
- mixture modeling
- background modeling
- video frames
- video data
- expectation maximization
- video retrieval
- maximum likelihood
- moving object detection
- foreground objects
- language model
- probability density function
- surveillance videos
- automatic model selection
- supervised learning
- text mining
- model based clustering
- background model
- moving camera
- mixture of gaussians
- video database
- video clips
- dirichlet distribution
- mixture components
- image segmentation
- key frames
- object tracking