Large-Scale Tag-Based Font Retrieval With Generative Feature Learning

Abstract

Font selection is one of the most important steps in a design workflow. Traditional methods rely on ordered lists which require significant domain knowledge and are often difficult to use even for trained professionals. In this paper, we address the problem of large-scale tag-based font retrieval which aims to bring semantics to the font selection process and enable people without expert knowledge to use fonts effectively. We collect a large-scale font tagging dataset of high-quality professional fonts. The dataset contains nearly 20,000 fonts, 2,000 tags, and hundreds of thousands of font-tag relations. We propose a novel generative feature learning algorithm that leverages the unique characteristics of fonts. The key idea is that font images are synthetic and can therefore be controlled by the learning algorithm. We design an integrated rendering and learning process so that the visual feature from one image can be used to reconstruct another image with different text. The resulting feature captures important font design details while is robust to nuisance factors such as text. We propose a novel attention mechanism to re-weight the visual feature for joint visual-text modeling. We combine the feature and the attention mechanism in a novel recognition-retrieval model. Experimental results show that our method significantly outperforms the state-of-the-art for the important problem of large-scale tag-based font retrieval.

AdaptiFont: Increasing Individuals’ Reading Speed with a Generative Font Model and Bayesian Optimization

“…Our data driven approach to obtain a generative font space is most closely related to previous approaches of unsupervised learning of fonts [17] and recent systems based on Generative Adversarial Networks (GAN), e.g. [6,19,20,62,68]. Diferently from [17], which used a polyline representation of letters, we used a pixel based representation as the current study renders text only to a monitor up to a size of 40 pt.…”

“…Another recent approach for font generation employs GANs, see e.g. [6,19,20,62,68]. Even though this technique gives good results at the letter and word level, current implementations sufer from problems in alignment and kerning in continuous texts.…”

Shared Latent Space of Font Shapes and Their Noisy Impressions

“…O'Donovan et al [15] have published a dataset with 200 different fonts, each of which is annotated with the degrees of 37 attributes. Chen et al [4] published a far larger collection of fonts tagged with impression words and proposed a font retrieval method using word queries.…”

“…3 MyFonts dataset [4] As the font dataset with impression words, we employ the dataset published by Chen et al, [4]. This dataset, hereafter called the MyFonts dataset, comprises 18,815 fonts collected at MyFonts.com.…”

“…It is known that different font shapes give different impressions. A recent paper [4] provides a large font dataset, where many people attach impression words to individual fonts. Fig.…”