Learning representation hierarchies by sharing visual features: a computational investigation of Persian character recognition with unsupervised deep learning.
Zahra SadeghiAlberto TestolinPublished in: Cogn. Process. (2017)
Keyphrases
- visual features
- deep learning
- character recognition
- unsupervised learning
- deep architectures
- supervised learning
- visual information
- image classification
- weakly supervised
- image retrieval
- learning algorithm
- keywords
- machine vision
- unsupervised feature learning
- active learning
- object recognition
- image collections
- training data
- deep belief networks
- machine learning
- image representation
- spatial information
- low level features
- semi supervised
- image annotation
- low level
- reinforcement learning
- knn