# Which Parts Determine the Impression of the Font?

arXiv: https://arxiv.org/abs/2103.14216 · HTML: https://mlsanity.com/paper/2103.14216/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 24 references that are in the map.
- First posted on arXiv 26 Mar 2021.

## Builds on (most cited first, 9 of 9)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Deep Sets](https://mlsanity.com/paper/1703.06114/) (2017)
- [Multi-Content GAN for Few-Shot Font Style Transfer](https://mlsanity.com/paper/1712.00516/) (2017)
- [SIFT Meets CNN: A Decade Survey of Instance Retrieval](https://mlsanity.com/paper/1608.01807/) (2016)
- [Few-shot Compositional Font Generation with Dual Memory](https://mlsanity.com/paper/2005.10510/) (2020)
- [Large-scale Tag-based Font Retrieval with Generative Feature Learning](https://mlsanity.com/paper/1909.02072/) (2019)
- [A Deep Factorization of Style and Structure in Fonts](https://mlsanity.com/paper/1910.00748/) (2019)
- [Attribute2Font: Creating Fonts You Want From Attributes](https://mlsanity.com/paper/2005.07865/) (2020)
- [Serif or Sans: Visual Font Analytics on Book Covers and Online Advertisements](https://mlsanity.com/paper/1906.10269/) (2019)

## Built on by (most cited first, 7 of 7)

- [Font Generation with Missing Impression Labels](https://mlsanity.com/paper/2203.10348/) (2022)
- [Font Shape-to-Impression Translation](https://mlsanity.com/paper/2203.05808/) (2022)
- [Impression-CLIP: Contrastive Shape-Impression Embedding for Fonts](https://mlsanity.com/paper/2402.16350/) (2024)
- [TrueType Transformer: Character and Font Style Recognition in Outline Format](https://mlsanity.com/paper/2203.05338/) (2022)
- [Local Style Awareness of Font Images](https://mlsanity.com/paper/2310.06337/) (2023)
- [Font Impression Estimation in the Wild](https://mlsanity.com/paper/2402.15236/) (2024)
- [Few-Part-Shot Font Generation](https://mlsanity.com/paper/2509.10006/) (2025)

## Abstract (arXiv)

Various fonts give different impressions, such as legible, rough, and comic-text.This paper aims to analyze the correlation between the local shapes, or parts, and the impression of fonts. By focusing on local shapes instead of the whole letter shape, we can realize letter-shape independent and more general analysis. The analysis is performed by newly combining SIFT and DeepSets, to extract an arbitrary number of essential parts from a particular font and aggregate them to infer the font impressions by nonlinear regression. Our qualitative and quantitative analyses prove that (1)fonts with similar parts have similar impressions, (2)many impressions, such as legible and rough, largely depend on specific parts, (3)several impressions are very irrelevant to parts.