AutoTrajectory: Label-Free Trajectory Extraction and Prediction from Videos Using Dynamic Points.
Yuexin MaXinge ZhuXinjing ChengRuigang YangJiming LiuDinesh ManochaPublished in: ECCV (13) (2020)
Keyphrases
- prediction accuracy
- video sequences
- prediction model
- neural network
- point sets
- video surveillance
- information extraction
- space time
- action recognition
- multi label
- human activities
- convex hull
- text categorization
- video content
- human actions
- data points
- prediction error
- moving camera
- video database
- automatic extraction
- motion trajectories
- location prediction