Evaluation of Deep Convolutional Nets for Document Image Classification and Retrieval

TL;DR

This paper demonstrates that deep convolutional neural networks significantly improve document image classification and retrieval, outperforming traditional methods, and introduces a new large dataset for training and benchmarking.

Contribution

It shows CNN features are superior to hand-crafted features for document analysis and provides a large labeled dataset for future research.

Findings

CNN features outperform hand-crafted features in document classification

Features are robust to compression and transfer well across domains

Region-specific feature learning is unnecessary with sufficient data

Abstract

This paper presents a new state-of-the-art for document image classification and retrieval, using features learned by deep convolutional neural networks (CNNs). In object and scene analysis, deep neural nets are capable of learning a hierarchical chain of abstraction from pixel inputs to concise and descriptive representations. The current work explores this capacity in the realm of document analysis, and confirms that this representation strategy is superior to a variety of popular hand-crafted alternatives. Experiments also show that (i) features extracted from CNNs are robust to compression, (ii) CNNs trained on non-document images transfer well to document analysis tasks, and (iii) enforcing region-specific feature-learning is unnecessary given sufficient training data. This work also makes available a new labelled subset of the IIT-CDIP collection, containing 400,000 document images across 16 categories, useful for training new CNNs for document analysis.