Pillar in Pillar: Multi-Scale and Dynamic Feature Extraction for 3D Object Detection in Point Clouds.
Yonglin TianLichao HuangXuesong LiYuan LiZilei WangFei-Yue WangPublished in: CoRR (2019)
Keyphrases
- d objects
- point cloud
- feature extraction
- multiscale
- real world objects
- cad model
- viewpoint
- range data
- object recognition
- surface reconstruction
- structure from motion
- cad models
- pose estimation
- multi view
- three dimensional
- line drawings
- object classes
- multiple views
- point sets
- shape descriptors
- object detection
- feature vectors
- control points
- face recognition
- image processing
- surface normals
- training set
- image segmentation