Domain Agnostic Few-Shot Learning For Document Intelligence

TL;DR

This paper introduces a domain-agnostic few-shot learning approach tailored for classifying semi-structured document images, addressing the challenge of domain shift where existing methods often fail.

Contribution

The work presents a novel few-shot learning method specifically designed for semi-structured document classification across different domains, filling a gap in current research.

Findings

Consistent improvements over existing methods in few-shot domain-shift scenarios

Effective classification of semi-structured document images with limited samples

Demonstrated robustness across diverse document types

Abstract

Few-shot learning aims to generalize to novel classes with only a few samples with class labels. Research in few-shot learning has borrowed techniques from transfer learning, metric learning, meta-learning, and Bayesian methods. These methods also aim to train models from limited training samples, and while encouraging performance has been achieved, they often fail to generalize to novel domains. Many of the existing meta-learning methods rely on training data for which the base classes are sampled from the same domain as the novel classes used for meta-testing. However, in many applications in the industry, such as document classification, collecting large samples of data for meta-learning is infeasible or impossible. While research in the field of the cross-domain few-shot learning exists, it is mostly limited to computer vision. To our knowledge, no work yet exists that examines the use of few-shot learning for classification of semi-structured documents (scans of paper documents) generated as part of a business workflow (forms, letters, bills, etc.). Here the domain shift is significant, going from natural images to the semi-structured documents of interest. In this work, we address the problem of few-shot document image classification under domain shift. We evaluate our work by extensive comparisons with existing methods. Experimental results demonstrate that the proposed method shows consistent improvements on the few-shot classification performance under domain shift.