CramNet: Camera-Radar Fusion with Ray-Constrained Cross-Attention for Robust 3D Object Detection.
Jyh-Jing HwangHenrik KretzschmarJoshua ManelaSean RaffertyNicholas Armstrong-CrewsTiffany ChenDragomir AnguelovPublished in: CoRR (2022)
Keyphrases
- d objects
- multi view
- pose estimation
- viewpoint
- multiple views
- object recognition
- three dimensional
- range data
- shape descriptors
- cad models
- focal length
- object detection
- vision system
- pose normalization
- viewing direction
- camera parameters
- d mesh
- object features
- image contours
- partial occlusion
- multiple images
- three dimensional objects
- object classes
- topological information
- control points
- automatic target recognition
- line drawings