Online object tracking based on BLSTM-RNN with contextual-sequential labeling.
Xiangzeng ZhouLei XiePeng ZhangYanning ZhangPublished in: J. Ambient Intell. Humaniz. Comput. (2017)
Keyphrases
- object tracking
- particle filter
- video sequences
- mean shift
- recurrent neural networks
- appearance model
- moving objects
- image sequences
- kalman filter
- appearance modeling
- computer vision
- nearest neighbor
- video surveillance
- video surveillance systems
- partial occlusion
- tracking accuracy
- robust object tracking
- consistent labeling
- scene segmentation
- kernel density approximation
- data association
- multi camera
- multiple object tracking
- particle filtering
- multiple cameras
- visual object tracking
- occlusion handling
- visual tracking
- object location
- tracking objects
- robust tracking
- pattern recognition
- image segmentation
- sparse representation
- tracked objects
- object tracking algorithm
- three dimensional
- neural network