PV-RCNN++: Point-Voxel Feature Set Abstraction With Local Vector Representation for 3D Object Detection.
Shaoshuai ShiLi JiangJiajun DengZhe WangChaoxu GuoJianping ShiXiaogang WangHongsheng LiPublished in: Int. J. Comput. Vis. (2023)
Keyphrases
- d objects
- feature set
- vector representation
- three dimensional
- viewpoint
- feature selection
- feature vectors
- multiple views
- feature extraction
- classification accuracy
- multi view
- minimum bounding
- fixed length
- object recognition
- pose estimation
- shape descriptors
- feature space
- principal component analysis
- semantic description
- similarity measure
- co occurrence
- distance measure
- image features