Font Shape-to-Impression Translation

TL;DR

This paper introduces a Transformer-based method for analyzing and translating font shapes into impressions, effectively capturing local part correlations to understand how font features influence perceived impressions.

Contribution

It presents a novel Transformer architecture for part-based font impression analysis and translation, outperforming existing methods in accuracy.

Findings

Transformer-based approaches estimate font impressions more accurately.

The method reveals key local parts responsible for specific impressions.

Approach works for both classification and translation tasks.

Abstract

Different fonts have different impressions, such as elegant, scary, and cool. This paper tackles part-based shape-impression analysis based on the Transformer architecture, which is able to handle the correlation among local parts by its self-attention mechanism. This ability will reveal how combinations of local parts realize a specific impression of a font. The versatility of Transformer allows us to realize two very different approaches for the analysis, i.e., multi-label classification and translation. A quantitative evaluation shows that our Transformer-based approaches estimate the font impressions from a set of local parts more accurately than other approaches. A qualitative evaluation then indicates the important local parts for a specific impression.