MonoScene: Monocular 3D Semantic Scene Completion.
Anh-Quan CaoRaoul de CharettePublished in: CVPR (2022)
Keyphrases
- image sequences
- stereo camera
- depth cues
- d scene
- relative depth
- semantic context
- road vehicles
- ego motion
- video sequences
- complex scenes
- three dimensional
- multiple images
- natural language
- pose estimation
- moving camera
- scene structure
- projective geometry
- high level
- dynamic scenes
- single image
- input image
- optical flow
- moving objects
- computer vision
- semantic information
- apparent motion
- independently moving objects
- static scene
- outdoor scenes
- image retrieval
- real scenes
- semantic similarity
- field of view