PV-RCNN++: semantical point-voxel feature interaction for 3D object detection.
Peng WuLipeng GuXuefeng YanHaoran XieFu Lee WangGary ChengMingqiang WeiPublished in: Vis. Comput. (2023)
Keyphrases
- d objects
- point distribution
- three dimensional
- viewpoint
- control points
- object features
- object recognition
- multi view
- surface points
- pose estimation
- shape descriptors
- multiple views
- range data
- d mesh
- feature vectors
- minimum bounding
- cad models
- physically plausible
- line drawings
- geometric invariants
- detection method
- object detection
- image features
- real world objects
- object classes
- three dimensional objects
- image contours
- object views