Object-Agnostic Transformers for Video Referring Segmentation.
Xu YangHao WangDe XieCheng DengDacheng TaoPublished in: IEEE Trans. Image Process. (2022)
Keyphrases
- object segmentation
- video objects
- video segmentation
- multiple objects
- video scene
- video sequences
- object segmentation and tracking
- segmentation accuracy
- video analysis
- video content
- video data
- segmentation algorithm
- pixel level
- bounding box
- segmentation method
- level set
- object motion
- image segmentation
- shape prior
- motion cues
- spatial and temporal
- multimedia
- moving objects
- background subtraction
- foreground background segmentation
- signed distance
- video surveillance
- video frames
- object tracking
- image regions
- text detection
- multiscale
- object shape
- final segmentation
- video tracking
- image segments
- temporal segmentation
- medical images
- space time
- video streams
- target object
- object model
- region growing
- foreground and background
- object contours
- key frames
- foreground objects
- background pixels
- object classes
- point trajectories
- video clips