Cutting the Error by Half: Investigation of Very Deep CNN and Advanced Training Strategies for Document Image Classification.
Muhammad Zeshan AfzalAndreas KölschSheraz AhmedMarcus LiwickiPublished in: CoRR (2017)
Keyphrases
- image classification
- cellular neural networks
- bag of words
- error rate
- feature extraction
- deep architectures
- document clustering
- image representation
- classifier training
- visual features
- web documents
- information retrieval
- training set
- retrieval strategies
- document classification
- total error rate
- deep learning
- training process
- error bounds
- supervised learning
- document retrieval
- document images
- generalization error
- vector space model
- structured documents
- text mining
- sparse representation
- information retrieval systems
- training examples