Self-supervised Learning of Rotation-invariant 3D Point Set Features using Transformer and its Self-distillation.
Takahiko FuruyaZhoujie ChenRyutarou OhbuchiZhenzhong KuangPublished in: CoRR (2023)
Keyphrases
- rotation invariant
- point sets
- learning algorithm
- texture analysis
- feature vectors
- feature extraction
- feature set
- steerable filters
- data sets
- rotation invariance
- zernike moments
- distance function
- supervised learning
- image processing
- computer vision
- nearest neighbor
- classification accuracy
- image analysis
- image reconstruction
- viewpoint
- pairwise
- texture classification
- object recognition
- video sequences
- shape context
- scale invariance