Handwritten Text Recognition: A Survey

TL;DR

This survey reviews the evolution of Handwritten Text Recognition (HTR) models from early heuristic methods to advanced neural networks, highlighting recent progress, datasets, challenges, and future research directions in the field.

Contribution

It provides a comprehensive categorization of HTR models, a unified framework for research methodologies, and insights into recent benchmarks and datasets, guiding future advancements.

Findings

Deep learning models have significantly improved recognition accuracy.

Recent end-to-end document-level approaches enable comprehensive understanding.

Benchmark datasets are crucial for evaluating and comparing HTR systems.

Abstract

Handwritten Text Recognition (HTR) has become an essential field within pattern recognition and machine learning, with applications spanning historical document preservation to modern data entry and accessibility solutions. The complexity of HTR lies in the high variability of handwriting, which makes it challenging to develop robust recognition systems. This survey examines the evolution of HTR models, tracing their progression from early heuristic-based approaches to contemporary state-of-the-art neural models, which leverage deep learning techniques. The scope of the field has also expanded, with models initially capable of recognizing only word-level content progressing to recent end-to-end document-level approaches. Our paper categorizes existing work into two primary levels of recognition: (1) \emph{up to line-level}, encompassing word and line recognition, and (2) \emph{beyond line-level}, addressing paragraph- and document-level challenges. We provide a unified framework that examines research methodologies, recent advances in benchmarking, key datasets in the field, and a discussion of the results reported in the literature. Finally, we identify pressing research challenges and outline promising future directions, aiming to equip researchers and practitioners with a roadmap for advancing the field.