Synthetic Craquelure Generation for Unsupervised Painting Restoration

TL;DR

This paper introduces a fully annotation-free, synthetic craquelure generation framework combined with a specialized detection and inpainting pipeline to improve digital painting restoration, especially for fine crack patterns.

Contribution

It presents a novel synthetic craquelure generator and a crack detection method that outperforms existing models in zero-shot painting restoration tasks.

Findings

Outperforms state-of-the-art restoration models in zero-shot settings

Faithfully preserves original brushwork during inpainting

Effective synthetic data generation for crack detection

Abstract

Cultural heritage preservation increasingly demands non-invasive digital methods for painting restoration, yet identifying and restoring fine craquelure patterns from complex brushstrokes remains challenging due to scarce pixel-level annotations. We propose a fully annotation-free framework driven by a domain-specific synthetic craquelure generator, which simulates realistic branching and tapered fissure geometry using B\'ezier trajectories. Our approach couples a classical morphological detector with a learning-based refinement module: a SegFormer backbone adapted via Low-Rank Adaptation (LoRA). Uniquely, we employ a detector-guided strategy, injecting the morphological map as an input spatial prior, while a masked hybrid loss and logit adjustment constrain the training to focus specifically on refining candidate crack regions. The refined masks subsequently guide an Anisotropic Diffusion inpainting stage to reconstruct missing content. Experimental results demonstrate that our pipeline significantly outperforms state-of-the-art photographic restoration models in zero-shot settings, while faithfully preserving the original paint brushwork.