Leveraging SE(3) Equivariance for Self-supervised Category-Level Object Pose Estimation from Point Clouds.
Xiaolong LiYijia WengLi YiLeonidas J. GuibasA. Lynn AbbottShuran SongHe WangPublished in: NeurIPS (2021)
Keyphrases
- pose estimation
- point cloud
- category level
- d objects
- object detection
- object recognition
- computer vision
- structure from motion
- object categories
- point sets
- human body
- feature points
- object classes
- depth images
- face recognition
- viewpoint
- object model
- object class
- multi view
- moving objects
- range data
- pairwise
- shape matching