Handwritten Text Generation from Visual Archetypes

TL;DR

This paper introduces a Transformer-based model for few-shot handwritten text generation that effectively captures writer style and rare characters by using visual archetypes and pre-training on synthetic data, outperforming existing methods.

Contribution

The work presents a novel representation of textual content as visual archetypes and leverages pre-training to improve style and character generalization in handwritten text synthesis.

Findings

Effective generation of unseen styles and rare characters.

Outperforms existing approaches in fidelity of generated handwriting.

Robust representation of writer style through synthetic pre-training.

Abstract

Generating synthetic images of handwritten text in a writer-specific style is a challenging task, especially in the case of unseen styles and new words, and even more when these latter contain characters that are rarely encountered during training. While emulating a writer's style has been recently addressed by generative models, the generalization towards rare characters has been disregarded. In this work, we devise a Transformer-based model for Few-Shot styled handwritten text generation and focus on obtaining a robust and informative representation of both the text and the style. In particular, we propose a novel representation of the textual content as a sequence of dense vectors obtained from images of symbols written as standard GNU Unifont glyphs, which can be considered their visual archetypes. This strategy is more suitable for generating characters that, despite having been seen rarely during training, possibly share visual details with the frequently observed ones. As for the style, we obtain a robust representation of unseen writers' calligraphy by exploiting specific pre-training on a large synthetic dataset. Quantitative and qualitative results demonstrate the effectiveness of our proposal in generating words in unseen styles and with rare characters more faithfully than existing approaches relying on independent one-hot encodings of the characters.