TS-Net: OCR Trained to Switch Between Text Transcription Styles

TL;DR

TS-Net introduces a Transcription Style Block that enables OCR systems to adapt to various transcription styles automatically, improving accuracy and flexibility across different datasets and styles.

Contribution

The paper presents a novel Transcription Style Block (TSB) that allows OCR networks to switch between transcription styles without explicit rules, enhancing adaptability and accuracy.

Findings

TSB learns different transcription styles in artificial data.

TSB improves OCR accuracy on real-world data.

TSB can adapt to new styles with few examples.

Abstract

Users of OCR systems, from different institutions and scientific disciplines, prefer and produce different transcription styles. This presents a problem for training of consistent text recognition neural networks on real-world data. We propose to extend existing text recognition networks with a Transcription Style Block (TSB) which can learn from data to switch between multiple transcription styles without any explicit knowledge of transcription rules. TSB is an adaptive instance normalization conditioned by identifiers representing consistently transcribed documents (e.g. single document, documents by a single transcriber, or an institution). We show that TSB is able to learn completely different transcription styles in controlled experiments on artificial data, it improves text recognition accuracy on large-scale real-world data, and it learns semantically meaningful transcription style embedding. We also show how TSB can efficiently adapt to transcription styles of new documents from transcriptions of only a few text lines.