PQ-Transformer: Jointly Parsing 3D Objects and Layouts from Point Clouds.
Xiaoxue ChenHao ZhaoGuyue ZhouYa-Qin ZhangPublished in: CoRR (2021)
Keyphrases
- point cloud
- d objects
- structure from motion
- object recognition
- surface reconstruction
- multi view
- cad models
- viewpoint
- range data
- three dimensional
- pose estimation
- multiple views
- stereo camera
- shape descriptors
- line drawings
- point sets
- d mesh
- photorealistic
- real world objects
- urban scenes
- cad model
- surface normals
- machine learning
- object surface
- computer vision
- surface patches