DeVIS: Making Deformable Transformers Work for Video Instance Segmentation.
Adrià CaellesTim MeinhardtGuillem BrasóLaura Leal-TaixéPublished in: CoRR (2022)
Keyphrases
- video segmentation
- video sequences
- image segmentation
- object segmentation and tracking
- video data
- multimedia
- segmentation algorithm
- objects in video sequences
- level set
- foreground background segmentation
- segmentation method
- shape prior
- video frames
- video streams
- video content
- image analysis
- text detection
- fully automatic
- deformable models
- space time
- layered representation
- medical images
- temporal coherence
- single frame
- segmentation accuracy
- multiscale
- energy function
- video objects
- object segmentation
- motion segmentation
- region growing