EDNets: Deep Feature Learning for Document Image Classification Based on Multi-view Encoder-Decoder Neural Networks.
Akrem SellamiSalvatore TabbonePublished in: ICDAR (4) (2021)
Keyphrases
- multi view
- image classification
- multi view learning
- image features
- three dimensional
- co training
- d objects
- multiple views
- reinforcement learning
- learning process
- low complexity
- learning tasks
- depth map
- supervised learning
- motion estimation
- active learning
- learning algorithm
- statistical learning
- view synthesis
- multiple viewpoints
- multi view reconstruction
- multi view face detection