Syed Talal Wasim ; Romain Collaud ; Lara Défayes ; Nicolas Henchoz ; Mathieu Salzmann et al. - Toward Automatic Typography Analysis: Serif Classification and Font Similarities

jdmdh:10230 - Journal of Data Mining & Digital Humanities, February 6, 2024, Historical Documents and automatic text recognition - https://doi.org/10.46298/jdmdh.10230
Toward Automatic Typography Analysis: Serif Classification and Font SimilaritiesArticle

Authors: Syed Talal Wasim ORCID1; Romain Collaud ORCID2; Lara Défayes ORCID2; Nicolas Henchoz ORCID2; Mathieu Slazmann ORCID1; Delphine Ribes Lemay ORCID2

Whether a document is of historical or contemporary significance, typography plays a crucial role in its composition. From the early days of modern printing, typographic techniques have evolved and transformed, resulting in changes to the features of typography. By analyzing these features, we can gain insights into specific time periods, geographical locations, and messages conveyed through typography. Therefore, in this paper, we aim to investigate the feasibility of training a model to classify serif typeswithout knowledge of the font and character. We also investigate how to train a vectorial-based image model able to group together fonts with similar features. Specifically, we compare the use of state-of-theart image classification methods, such as the EfficientNet-B2 and the Vision Transformer Base model with different patch sizes, and the state-of-the-art fine-grained image classification method, TransFG, on the serif classification task. We also evaluate the use of the DeepSVG model to learn to group fonts with similar features. Our investigation reveals that fine-grained image classification methods are better suited for the serif classification tasks and that leveraging the character labels helps to learn more meaningful font similarities.This repository contains: - Paper published in the Journal of data mining and digital humanities:WasimEtAl_Toward_Automatic_Typography_Analysis__Serif_Classification_and_Font_Similarities.pdf - Two datasets: The first for serif classification  consisting of 126666 training and 2914 font-independent testing images in raster format.  The second dataset for svg based similarity learning consists of 124010 training and 2914  font-independent testing images. The images have been categorized into sans-serif,  linear-serif, slab-serif, and triangular serif by a designer at the EPFL+ECAL Labfont_serif_dataset.zipfont_svg_dataset.zip - model weights:serif_B_16_long_checkpoint.bin: refers to TransFG ViT B/16 Backendserif_B_32_long_checkpoint.bin: refers to TransFG ViT B/32 BackendViT-B_16.npz ViT-B_32.npz are publicly available at: https://github.com/TACJu/TransFGand were used for training TransFG ViT B/16 Backend and TransFG ViT B/32 Backend - code:https://github.com/TalalWasim/GEST-Serif 


Volume: Historical Documents and automatic text recognition
Section: Project presentations
Published on: February 6, 2024
Accepted on: November 6, 2023
Submitted on: October 31, 2022
Keywords: Machine learning,Digital humanities,Typography,Classification models

Files

Name Size
ViT-B_32.npz
md5: 9f43c05aac4a905b3ca9bdac814d4c32
400.01 MB
ViT-B_16.npz
md5: d9715d3bf39e807337a3c3e117756738
393.69 MB
serif_B_32_long_checkpoint.bin
md5: 82bfc61f48cfa6a23d4a5455e5d29f87
337.15 MB
serif_B_16_long_checkpoint.bin
md5: f5a53cf74b003c90e409b9a765679538
330.82 MB
font_serif_dataset.zip
md5: f87971f593457b7da225d313b44354d9
183.75 MB
font_svg_dataset.zip
md5: 13ef7c1d68cd9e4e33c68542ba691f3a
70.53 MB
WasimEtAl_Toward_Automatic_Typography_Analysis__Serif_Classification_and_Font_Similarities.pdf
md5: 2fa9a6f625403d7f673092cc6a9f1ed9
4.75 MB
Readme.txt
md5: 343b6a033f7a7fa912fbf69c02818de3
1.08 KB

Consultation statistics

This page has been seen 235 times.
This article's PDF has been downloaded 100 times.