VoxFormer: Sparse Voxel Transformer for Camera-Based 3D Semantic Scene Completion.
Yiming LiZhiding YuChristopher B. ChoyChaowei XiaoJosé M. ÁlvarezSanja FidlerChen FengAnima AnandkumarPublished in: CVPR (2023)
Keyphrases
- real scenes
- camera images
- scene structure
- scene geometry
- multiple images
- autocalibration
- d scene
- three dimensional
- fish eye
- moving camera
- stereo camera
- ground plane
- active illumination
- acquired images
- video images
- camera positions
- camera tracking
- imaging process
- video rate
- light conditions
- uncalibrated cameras
- structure from motion
- image capture
- camera views
- object motion
- single image
- field of view
- white balance
- vanishing points
- depth map
- depth estimation
- virtual camera
- video camera
- point features
- camera calibration
- single shot
- multiple cameras
- camera motion
- camera parameters
- time of flight
- camera movement
- epipolar geometry
- optical axis
- defocused images
- camera viewpoint
- response function
- ego motion
- live video
- vision system
- camera pose
- single camera
- video sequences
- image correspondences
- digital camera
- moving objects
- hand held
- line features
- viewing direction
- relative position
- multi camera
- stereo images
- multiple views
- fault diagnosis
- input image
- image sequences