MSF: Motion-guided Sequential Fusion for Efficient 3D Object Detection from Point Cloud Sequences.
Chenhang HeRuihuang LiYabin ZhangShuai LiLei ZhangPublished in: CVPR (2023)
Keyphrases
- d objects
- point cloud
- real world objects
- viewpoint
- multi view
- physically plausible
- range data
- object recognition
- pose estimation
- point sets
- cad model
- shape descriptors
- multiple views
- three dimensional
- surface reconstruction
- laser scanner
- optical flow
- image sequences
- structure from motion
- d mesh
- object classes
- point correspondences
- computer aided design
- camera motion