Research map / Image generation / Image restoration
Building Bridge Across the Time: Disruption and Restoration of Murals In the Wild
Huiyang Shao, Qianqian Xu, Peisong Wen, Peifeng Gao, Zhiyong Yang, Qingming Huang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train baselines on the proposed physically simulated mural dataset instead of the previous masked dataset (ablation) | baseline restorers · real damaged murals | PSNR/FID: SwinIR PSNR 13.51 -> 17.83; ADF FID 57.13 -> 47.59 | Improved |
| add the damage attention map (DAM) module to the guided-diffusion backbone (ablation) | ADF · mural benchmark | PSNR/SSIM/IS/FID | Improved |
| combine heatmap loss Lh with DAM (ablation) | ADF · mural benchmark | FID | Improved |
| add edge loss Ls (ablation) | ADF · mural benchmark | edge sharpness (qualitative) | Improved |
| add estimation loss Le (ablation) | ADF · mural benchmark | performance and generation speed | Improved |
| transplant DAM and losses to a conditional GAN backbone (ablation) | conditional GAN · mural restoration (qualitative) | restoration quality | Improved |
| use ADF instead of strongest baselines on benchmark | mural restoration · mural benchmark | FID: 37.23 -> 35.88 | Improved |
| vary degradation: higher cracking and peeling in training data (ablation) | mural simulation dataset · mural restoration | repainting ability | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Restormer — Restormer: Efficient Transformer for High-Resolution Image Restoration (CVPR 2022) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Learning Where and What to Restore for Composite Image Restoration (NeurIPS 2026 · similarity 0.84)
- Visual-Instructed Degradation Diffusion for All-in-One Image Restoration (CVPR 2025 · similarity 0.83)
- Selective Hourglass Mapping for Universal Image Restoration Based on Diffusion Model (CVPR 2024 · similarity 0.82)
- Prune and Repaint: Content-Aware Image Retargeting for any Ratio (NeurIPS 2024 · similarity 0.82)
- HD-Painter: High-Resolution and Prompt-Faithful Text-Guided Image Inpainting with Diffusion Models (ICLR 2025 · similarity 0.81)
- ReFIR: Grounding Large Restoration Models with Retrieval Augmentation (NeurIPS 2024 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
