High-Fidelity Mural Restoration via a Unified Hybrid Mask-Aware Transformer

TL;DR

This paper introduces HMAT, a hybrid transformer framework that effectively restores degraded murals by combining local texture modeling, long-range structural inference, and mask-guided generative processes.

Contribution

The paper proposes a novel unified hybrid transformer architecture with mask-aware modules and a teacher-forcing decoder for high-fidelity mural restoration.

Findings

HMAT achieves competitive performance on DHMural and Nine-Colored Deer datasets.

The method produces more structurally coherent restorations than state-of-the-art approaches.

Experimental results validate the effectiveness of the hybrid mask-aware transformer design.

Abstract

Ancient murals are valuable cultural artifacts, but many have suffered severe degradation due to environmental exposure, material aging, and human activity. Restoring these artworks is challenging because it requires both reconstructing large missing structures and strictly preserving authentic, undamaged regions. This paper presents the Hybrid Mask-Aware Transformer (HMAT), a unified framework for high-fidelity mural restoration. HMAT integrates Mask-Aware Dynamic Filtering for robust local texture modeling with a Transformer bottleneck for long-range structural inference. To further address the diverse morphology of degradation, we introduce a mask-conditional style fusion module that dynamically guides the generative process. In addition, a Teacher-Forcing Decoder with hard-gated skip connections is designed to enforce fidelity in valid regions and focus reconstruction on missing areas. We evaluate HMAT on the DHMural dataset and a curated Nine-Colored Deer dataset under varying degradation levels. Experimental results demonstrate that the proposed method achieves competitive performance compared to state-of-the-art approaches, while producing more structurally coherent and visually faithful restorations. These findings suggest that HMAT provides an effective solution for the digital restoration of cultural heritage murals.