MsSVT: Mixed-scale Sparse Voxel Transformer for 3D Object Detection on Point Clouds.
Shaocong DongLihe DingHaiyang WangTingfa XuXinli XuJie WangZiyang BianYing WangJianan LiPublished in: NeurIPS (2022)
Keyphrases
- knn
- distance function
- d objects
- point cloud
- point sets
- three dimensional
- real world objects
- cad model
- object recognition
- surface reconstruction
- cad models
- multi view
- viewpoint
- line drawings
- multiple views
- range data
- structure from motion
- pose estimation
- shape descriptors
- high dimensional
- control points
- surface patches
- visual hull
- d mesh
- object classes
- stereo camera
- object surface
- photorealistic
- object matching
- computer graphics
- machine learning