InkDiffuser: High-Fidelity One-shot Chinese Calligraphy via Differentiable Morphological Optimization

TL;DR

InkDiffuser is a diffusion-based framework that significantly improves the realism and fidelity of one-shot Chinese calligraphy synthesis by enhancing high-frequency details and explicitly modeling ink morphology.

Contribution

It introduces a high-frequency enhancement mechanism and a differentiable ink structure loss to improve stroke rendering and ink morphology in calligraphy generation.

Findings

Outperforms existing few-shot font generation methods in structural consistency.

Produces more realistic ink rendering effects from a single reference glyph.

Demonstrates superior detail fidelity and visual authenticity across styles.

Abstract

Current Chinese calligraphy generation methods suffer from poor stroke rendering and unrealistic ink morphology, resulting in outputs with limited visual fidelity and artistic fluidity. To address this problem, we propose \textbf{InkDiffuser}, a diffusion-based generative framework for one-shot Chinese calligraphy synthesis. To guarantee high-fidelity rendering, we introduce two core contributions: a high-frequency enhancement mechanism and a Differentiable Ink Structure (DIS) loss that explicitly regularizes ink morphology. Inspired by the observation that high-frequency information in individual samples typically carries contour details, we enhance content extraction by explicitly fusing high-frequency representations for more accurate font structure. Furthermore, we propose a differentiable ink structure loss that integrates differentiable morphological operations into the diffusion process. By allowing the model to learn an explicit decomposition of ink-trace structures, DIS facilitates fine-grained refinement of stroke contours and delivers significantly improved visual realism in the generated calligraphy. Extensive experiments on various calligraphic styles and complex characters demonstrate that InkDiffuser can generate superior calligraphy fonts with realistic ink rendering effects from only a single reference glyph and outperform existing few-shot font generation approaches in structural consistency, detail fidelity, and visual authenticity. The code is available at the following address: https://github.com/JingVIPLab/InkDiffuser.