URFormer: Unified Representation LiDAR-Camera 3D Object Detection with Transformer.
Guoxin ZhangJun XieLin LiuZhepeng WangKuihe YangZiying SongPublished in: PRCV (3) (2023)
Keyphrases
- d objects
- invariant recognition
- multi view
- lidar data
- object representation
- viewpoint
- object recognition
- pose estimation
- geometric models
- multiple views
- partial matching
- physically plausible
- camera calibration
- range data
- shape descriptors
- viewing direction
- cad models
- camera parameters
- three dimensional objects
- topological information
- control points
- point cloud
- three dimensional
- line drawings
- structure from motion
- pose normalization
- recognition and pose estimation
- image contours
- stereo camera
- object classes
- vision system
- surface properties
- d mesh
- object views
- object detection
- high resolution