What question did this study set out to answer?

This research aims to enhance digital restoration of Dunhuang murals using a novel emotion-driven framework.

July 11, 2026npj Heritage ScienceOpen Access

Emotion-driven multimodal generation of Dunhuang murals with spatial motif priors for interactive digital heritage

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates improved digital restoration of Dunhuang murals, highlighting cultural fidelity and interactivity.

This research aims to enhance digital restoration of Dunhuang murals using a novel emotion-driven framework.
Proposed an emotion-driven multimodal generative framework for mural synthesis.
Used 2D-FFT and CNNs to extract spatial motif priors from the murals.
Implemented an asynchronous latent-space partitioning strategy to enable localized editing.
Achieved improved emotion–style alignment and cultural-symbol fidelity in digital restorations.
Results evaluated using CLIP similarity, CSA, PCI, and FID metrics, with recorded seed-level verification for transparency.

Yuan et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark