Semantic segmentation of trajectories with improved agent models for pedestrian behavior analysis.
Toru TamakiDaisuke OgawaBisser RaytchevKazufumi KanedaPublished in: CoRR (2019)
Keyphrases
- behavior analysis
- semantic segmentation
- agent model
- conditional random fields
- superpixels
- surveillance system
- weakly supervised
- collective decision making
- pascal voc
- object detection
- anomaly detection
- scene classification
- object categories
- pedestrian detection
- machine learning
- multi class
- mobile devices
- video sequences
- bayesian networks
- training data
- image sequences