Multi-Object Model-Free Tracking with Joint Appearance and Motion Inference.
Chongyu LiuRui YaoSeyed Hamid RezatofighiIan D. ReidQinfeng ShiPublished in: DICTA (2017)
Keyphrases
- multi object
- model free
- object detection and tracking
- motion trajectories
- reinforcement learning
- motion model
- appearance model
- multiple objects
- motion analysis
- function approximation
- human perception
- motion patterns
- image sequences
- motion estimation
- object tracking
- joint angles
- optical flow
- human motion
- raw data
- camera motion
- space time
- spatio temporal
- feature extraction
- visual tracking
- point correspondences
- video data