Vehicle detector training with labels derived from background subtraction algorithms in video surveillance.
Sebastian CygertAndrzej CzyzewskiPublished in: SPA (2018)
Keyphrases
- video surveillance
- background subtraction
- traffic monitoring
- moving objects
- motion detection
- surveillance system
- foreground objects
- object detection
- image sequences
- shadow detection
- moving object detection
- background model
- visual surveillance
- background modeling
- dynamic scenes
- foreground detection
- detection of moving objects
- appearance model
- gaussian mixture model
- computer vision
- event detection
- training set
- human detection
- object tracking
- surveillance videos
- pedestrian detection
- training data
- static camera
- markov random field
- gait recognition
- object recognition
- reinforcement learning
- bayesian networks
- similarity measure
- activity analysis
- learning algorithm