One-Shot Multilingual Font Generation Via ViT

TL;DR

This paper presents a ViT-based model for multilingual font generation that handles complex logographic and alphabetic scripts, utilizing pretraining and retrieval guidance to produce high-quality, adaptable fonts for unseen characters.

Contribution

Introduces a novel ViT-based framework with retrieval-augmented guidance for scalable, high-quality multilingual font generation, addressing limitations of prior methods.

Findings

Effective font generation across multiple languages.

High-quality results for unseen and user-crafted characters.

Enhanced scalability and adaptability demonstrated.

Abstract

Font design poses unique challenges for logographic languages like Chinese, Japanese, and Korean (CJK), where thousands of unique characters must be individually crafted. This paper introduces a novel Vision Transformer (ViT)-based model for multi-language font generation, effectively addressing the complexities of both logographic and alphabetic scripts. By leveraging ViT and pretraining with a strong visual pretext task (Masked Autoencoding, MAE), our model eliminates the need for complex design components in prior frameworks while achieving comprehensive results with enhanced generalizability. Remarkably, it can generate high-quality fonts across multiple languages for unseen, unknown, and even user-crafted characters. Additionally, we integrate a Retrieval-Augmented Guidance (RAG) module to dynamically retrieve and adapt style references, improving scalability and real-world applicability. We evaluated our approach in various font generation tasks, demonstrating its effectiveness, adaptability, and scalability.