Beyond the Pipeline: Analyzing Key Factors in End-to-End Deep Learning for Historical Writer Identification

TL;DR

This paper analyzes factors affecting end-to-end deep learning models for historical writer identification, highlighting challenges and proposing a simpler setup that performs comparably to complex systems.

Contribution

It provides a comprehensive analysis of various configurations and identifies key challenges, offering insights to improve end-to-end HWI systems.

Findings

Most configurations perform poorly due to weak visual feature capture.

Inconsistent patch representations and content noise affect performance.

A simpler end-to-end setup can achieve results comparable to top systems.

Abstract

This paper investigates various factors that influence the performance of end-to-end deep learning approaches for historical writer identification (HWI), a task that remains challenging due to the diversity of handwriting styles, document degradation, and the limited number of labelled samples per writer. These conditions often make accurate recognition difficult, even for human experts. Traditional HWI methods typically rely on handcrafted image processing and clustering techniques, which tend to perform well on small and carefully curated datasets. In contrast, end-to-end pipelines aim to automate the process by learning features directly from document images. However, our experiments show that many of these models struggle to generalise in more realistic, document-level settings, especially under zero-shot scenarios where writers in the test set are not present in the training data. We explore different combinations of pre-processing methods, backbone architectures, and post-processing strategies, including text segmentation, patch sampling, and feature aggregation. The results suggest that most configurations perform poorly due to weak capture of low-level visual features, inconsistent patch representations, and high sensitivity to content noise. Still, we identify one end-to-end setup that achieves results comparable to the top-performing system, despite using a simpler design. These findings point to key challenges in building robust end-to-end systems and offer insight into design choices that improve performance in historical document writer identification.