PVTransformer: Point-to-Voxel Transformer for Scalable 3D Object Detection.
Zhaoqi LengPei SunTong HeDragomir AnguelovMingxing TanPublished in: CoRR (2024)
Keyphrases
- d objects
- three dimensional
- object recognition
- viewpoint
- control points
- pose estimation
- surface points
- multi view
- cad models
- multiple views
- range data
- shape descriptors
- detection method
- object detection
- minimum bounding
- line drawings
- point distribution
- real world objects
- integral imaging
- physically plausible
- pose normalization
- image contours
- object features
- viewing direction
- object classes
- object views
- geometric invariants
- structure from motion