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: ECCV (38) (2022)
Keyphrases
- d objects
- viewpoint
- multi view
- object recognition
- pose estimation
- multiple views
- three dimensional
- range data
- image contours
- shape descriptors
- line drawings
- viewing direction
- cad models
- object detection
- d mesh
- texture mapping
- ray tracing
- multiple images
- focal length
- structure from motion
- automatic target recognition
- planar objects
- vision system
- three dimensional objects
- pose normalization
- integral imaging
- ground plane
- camera parameters
- computer vision
- partial occlusion
- camera motion