TAP-Vid: A Benchmark for Tracking Any Point in a Video.
Carl DoerschAnkush GuptaLarisa MarkeevaAdrià RecasensLucas SmairaYusuf AytarJoão CarreiraAndrew ZissermanYi YangPublished in: CoRR (2022)
Keyphrases
- object segmentation and tracking
- face detection and tracking
- real time
- person detection
- video surveillance
- video images
- surveillance videos
- successive frames
- video data
- video content
- moving camera
- video streams
- video sequences
- object detection and tracking
- video tracking
- object tracking
- particle filter
- text detection
- video frames
- multimedia
- single frame
- video database
- low frame rate
- video dataset
- object motion
- real world
- articulated human motion
- kalman filter
- video analysis
- visual tracking
- moving target
- image sequences
- objects in video sequences
- real time face tracking
- point tracking
- temporal continuity
- stationary camera
- input video
- dynamic background
- temporal consistency
- robust tracking
- image frames
- event detection
- live video
- multimedia data
- background subtraction
- video retrieval
- spatial and temporal
- foreground background segmentation
- video clips
- appearance model
- motion trajectories
- visual data
- moving objects
- video shots