Transformer-Based Vector Font Classification Using Different Font Formats: TrueType versus PostScript

TL;DR

This study compares TrueType and PostScript font formats in Transformer-based vector font classification, revealing that PostScript outlines yield better performance and emphasizing the importance of outline format choice in vector graphic deep learning.

Contribution

It demonstrates that PostScript outlines outperform TrueType in Transformer-based font classification, providing new insights into font representation choices for improved model performance.

Findings

PostScript outlines outperform TrueType in classification accuracy.

Information aggregation is key in Transformer-based vector font processing.

Guidance for selecting font outline formats in future vector graphic research.

Abstract

Modern fonts adopt vector-based formats, which ensure scalability without loss of quality. While many deep learning studies on fonts focus on bitmap formats, deep learning for vector fonts remains underexplored. In studies involving deep learning for vector fonts, the choice of font representation has often been made conventionally. However, the font representation format is one of the factors that can influence the computational performance of machine learning models in font-related tasks. Here we show that font representations based on PostScript outlines outperform those based on TrueType outlines in Transformer-based vector font classification. TrueType outlines represent character shapes as sequences of points and their associated flags, whereas PostScript outlines represent them as sequences of commands. In previous research, PostScript outlines have been predominantly used when fonts are treated as part of vector graphics, while TrueType outlines are mainly employed when focusing on fonts alone. Whether to use PostScript or TrueType outlines has been mainly determined by file format specifications and precedent settings in previous studies, rather than performance considerations. To date, few studies have compared which outline format provides better embedding representations. Our findings suggest that information aggregation is crucial in Transformer-based deep learning for vector graphics, as in tokenization in language models and patch division in bitmap-based image recognition models. This insight provides valuable guidance for selecting outline formats in future research on vector graphics.