DocReader: Bounding-Box Free Training of a Document Information Extraction Model

TL;DR

DocReader is an end-to-end neural network for document information extraction that trains solely on images and target values, eliminating the need for bounding-box annotations and enabling leveraging existing data.

Contribution

It introduces a bounding-box free training method for document information extraction, simplifying data annotation and facilitating continual learning.

Findings

Outperforms methods requiring bounding-box annotations.

Can leverage existing historical extraction data.

Supports continual learning during deployment.

Abstract

Information extraction from documents is a ubiquitous first step in many business applications. During this step, the entries of various fields must first be read from the images of scanned documents before being further processed and inserted into the corresponding databases. While many different methods have been developed over the past years in order to automate the above extraction step, they all share the requirement of bounding-box or text segment annotations of their training documents. In this work we present DocReader, an end-to-end neural-network-based information extraction solution which can be trained using solely the images and the target values that need to be read. The DocReader can thus leverage existing historical extraction data, completely eliminating the need for any additional annotations beyond what is naturally available in existing human-operated service centres. We demonstrate that the DocReader can reach and surpass other methods which require bounding-boxes for training, as well as provide a clear path for continual learning during its deployment in production.