TS-Net: OCR Trained to Switch Between Text Transcription Styles
This addresses the issue of varied transcription preferences across institutions and disciplines, offering an incremental improvement for OCR training.
The paper tackles the problem of inconsistent transcription styles in OCR systems by introducing a Transcription Style Block (TSB) that learns to switch between styles, improving text recognition accuracy on real-world data.
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.