BURST: A Benchmark for Unifying Object Recognition, Segmentation and Tracking in Video.
Ali AtharJonathon LuitenPaul VoigtlaenderTarasha KhuranaAchal DaveBastian LeibeDeva RamananPublished in: WACV (2023)
Keyphrases
- object segmentation and tracking
- object recognition
- text detection
- objects in video sequences
- foreground background segmentation
- motion segmentation
- video segmentation
- video scene
- multiple frames
- face detection and tracking
- single frame
- video surveillance
- object detection and tracking
- real time
- motion cues
- video tracking
- video sequences
- segmentation method
- computer vision
- video images
- object recognition and tracking
- video frames
- object detection
- person detection
- level set
- shape prior
- berkeley segmentation dataset
- video streams
- object matching
- object segmentation
- segmentation algorithm
- object tracking
- visual tracking
- image segmentation
- surveillance videos
- motion estimation
- automatic initialization
- image sequences
- multimedia
- temporal continuity
- image features
- graph cuts
- video content
- template matching
- video data
- moving camera
- temporal segmentation
- video analysis
- multiple objects
- figure ground segmentation
- motion model
- moving objects
- video objects
- medical images
- particle filter
- appearance model
- background subtraction
- partial occlusion
- superpixels
- video shots
- complex background