DELINE8K: A Synthetic Data Pipeline for the Semantic Segmentation of Historical Documents

TL;DR

This paper introduces DELINE8K, a comprehensive synthetic dataset for semantic segmentation of historical documents, addressing limitations of existing datasets and improving performance on the NAFSS benchmark.

Contribution

We developed DELINE8K, the most extensive synthetic dataset for document segmentation, combining diverse text types and backgrounds to enhance model training.

Findings

Deline8K outperforms previous datasets on NAFSS benchmark.

Synthetic data improves segmentation accuracy for historical documents.

The dataset is publicly available for further research.

Abstract

Document semantic segmentation is a promising avenue that can facilitate document analysis tasks, including optical character recognition (OCR), form classification, and document editing. Although several synthetic datasets have been developed to distinguish handwriting from printed text, they fall short in class variety and document diversity. We demonstrate the limitations of training on existing datasets when solving the National Archives Form Semantic Segmentation dataset (NAFSS), a dataset which we introduce. To address these limitations, we propose the most comprehensive document semantic segmentation synthesis pipeline to date, incorporating preprinted text, handwriting, and document backgrounds from over 10 sources to create the Document Element Layer INtegration Ensemble 8K, or DELINE8K dataset. Our customized dataset exhibits superior performance on the NAFSS benchmark, demonstrating it as a promising tool in further research. The DELINE8K dataset is available at https://github.com/Tahlor/deline8k.