MixCon3D: Synergizing Multi-View and Cross-Modal Contrastive Learning for Enhancing 3D Representation.
Yipeng GaoZeyu WangWei-Shi ZhengCihang XieYuyin ZhouPublished in: CoRR (2023)
Keyphrases
- multi view
- multi view learning
- single view
- cross modal
- perceptual information
- learning process
- three dimensional
- multiple views
- active learning
- image classification
- semi supervised
- learning algorithm
- learning tasks
- visual recognition
- d objects
- statistical learning
- feature vectors
- training data
- similarity measure
- feature selection