Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

An Interpretable Deep Learning Approach for Morphological Script Type Analysis

Malamatenia Vlachou-Efstathiou, Ioannis Siglidis, Dominique Stutzmann, Mathieu Aubry

TL;DR

This paper tackles the challenge of translating qualitative palaeographic script typologies into objective, quantitative analysis. It adapts the Learnable Typewriter to learn aligned character prototypes from manuscripts, enabling both qualitative visual comparisons and quantitative graph-based analyses across documents. By introducing prototype filtering, failure-case detection, and variability measures, the approach provides interpretable tools that can complement Derolez’s Textualis Formata taxonomy and illuminate Northern vs. Southern Textualis morphologies. The case study demonstrates how these prototypes and associated graphs can reveal stylistic distinctions, intra-class variation, and late-script deviations, offering a practical pathway to bridge traditional palaeography and data-driven methods.

Abstract

Defining script types and establishing classification criteria for medieval handwriting is a central aspect of palaeographical analysis. However, existing typologies often encounter methodological challenges, such as descriptive limitations and subjective criteria. We propose an interpretable deep learning-based approach to morphological script type analysis, which enables systematic and objective analysis and contributes to bridging the gap between qualitative observations and quantitative measurements. More precisely, we adapt a deep instance segmentation method to learn comparable character prototypes, representative of letter morphology, and provide qualitative and quantitative tools for their comparison and analysis. We demonstrate our approach by applying it to the Textualis Formata script type and its two subtypes formalized by A. Derolez: Northern and Southern Textualis

An Interpretable Deep Learning Approach for Morphological Script Type Analysis

TL;DR

This paper tackles the challenge of translating qualitative palaeographic script typologies into objective, quantitative analysis. It adapts the Learnable Typewriter to learn aligned character prototypes from manuscripts, enabling both qualitative visual comparisons and quantitative graph-based analyses across documents. By introducing prototype filtering, failure-case detection, and variability measures, the approach provides interpretable tools that can complement Derolez’s Textualis Formata taxonomy and illuminate Northern vs. Southern Textualis morphologies. The case study demonstrates how these prototypes and associated graphs can reveal stylistic distinctions, intra-class variation, and late-script deviations, offering a practical pathway to bridge traditional palaeography and data-driven methods.

Abstract

Defining script types and establishing classification criteria for medieval handwriting is a central aspect of palaeographical analysis. However, existing typologies often encounter methodological challenges, such as descriptive limitations and subjective criteria. We propose an interpretable deep learning-based approach to morphological script type analysis, which enables systematic and objective analysis and contributes to bridging the gap between qualitative observations and quantitative measurements. More precisely, we adapt a deep instance segmentation method to learn comparable character prototypes, representative of letter morphology, and provide qualitative and quantitative tools for their comparison and analysis. We demonstrate our approach by applying it to the Textualis Formata script type and its two subtypes formalized by A. Derolez: Northern and Southern Textualis
Paper Structure (32 sections, 2 equations, 7 figures, 1 table)

This paper contains 32 sections, 2 equations, 7 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related work
  3. Quantitative methods in palaeography.
  4. Prototype-based approaches in document analysis.
  5. Approach
  6. Learning comparable character prototypes
  7. The Learnable Typewriter.
  8. Finetuning character prototypes.
  9. Character prototype filtering.
  10. Automatic identification of failure cases.
  11. Character prototype comparison for palaeographic analysis
  12. Visual comparison.
  13. Character and document comparison graphs.
  14. Quantifying character variability.
  15. Experiments
  16. ...and 17 more sections

Figures (7)

  • Figure 1: The Learnable Typewriter Model learns to reconstruct text lines using a set of learned character prototypes. We demonstrate how the character prototypes can be used for palaeographic analysis.
  • Figure 2: Prototype filtering and failure case identification. We use a mask M defined from the reference prototype $\textbf{R}$ to remove artifacts from finetuned prototypes $\textbf{P}$, yieling a filtered prototype $\textbf{F}$. We compute an error $e$ associated to the filtering to automatically identify potential failure cases.
  • Figure 3: Comparison graphs. The markers correspond to different document prototypes and their coordinates to their distance to the Northern and Southern Textualis prototypes. See text for details.
  • Figure 4: Our filtered prototypes on type, sub-type and document level. The highlighted prototypes are the ones for which filtering had a significant impact (see Section \ref{['sec:ltw']} and \ref{['sec:res']} for details).
  • Figure 5: Character comparison graphs.
  • ...and 2 more figures