Lift, Splat, Map: Lifting Foundation Masks for Label-Free Semantic Scene Completion.
Arthur ZhangRainier HeijneJoydeep BiswasPublished in: CoRR (2024)
Keyphrases
- video sequences
- d scene
- semantic labels
- single image
- semantic web
- cognitive map
- natural language
- image sequences
- semantic space
- wavelet transform
- multiple images
- label assignment
- scene recognition
- semantic similarity
- scene classification
- high level
- semantic context
- line features
- object recognition
- maximum a posteriori
- scene analysis
- scene understanding
- domain specific
- semantic information
- real scenes
- input image
- low level
- scene interpretation
- dynamic scenes
- spatial relations
- multiscale