LaSOT: A High-quality Large-scale Single Object Tracking Benchmark.
Heng FanHexin BaiLiting LinFan YangPeng ChuGe DengSijia Yu HarshitMingzhen HuangJuehuan LiuYong XuChunyuan LiaoLin YuanHaibin LingPublished in: Int. J. Comput. Vis. (2021)
Keyphrases
- object tracking
- high quality
- particle filter
- mean shift
- appearance model
- image sequences
- video sequences
- video surveillance
- computer vision
- kalman filter
- robust object tracking
- visual tracking
- appearance modeling
- particle filtering
- sparse representation
- video surveillance systems
- moving objects
- multiple object tracking
- data association
- visual object tracking
- multi camera
- partial occlusion
- multiple cameras
- robust tracking
- multiple targets
- object appearance
- tracking accuracy
- object location
- scene segmentation
- depth map
- multi view
- higher order
- image processing