Research map / Image generation / Image restoration
P-PatchDiff: Progressive Patch Diffusion Models for Low-Light Image Enhancement
Ruoyu Guo, Haonan Zhong, Maurice Pagnucco, Yang Song
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 |
|---|---|---|---|
| add multi-patch alignment (brightness proxy g) (ablation) | P-PatchDiff · LOL-v2-Real | PSNR: 27.04 -> 28.04 | Improved |
| add multi-patch alignment (brightness proxy g) (ablation) | P-PatchDiff · LOL-v1 | PSNR: 26.38 -> 27.18 | Improved |
| add multi-patch alignment (brightness proxy g) (ablation) | P-PatchDiff · LOL-v2-Syn | PSNR: 26.76 -> 28.15 | Improved |
| add multi-patch alignment (brightness proxy g) (ablation) | P-PatchDiff · LOL-v2-Syn | FID: 23.48 -> 23.90 | Hurt |
| add multi-patch alignment (ablation) | P-PatchDiff · LOL-v2-Real | sampling time: 2.24 -> 2.45 | No clear effect |
| use progressive patch instead of fixed patch (n=5) (ablation) | P-PatchDiff · LOL-v2-Real | sampling time (s): 28.48 -> 2.45 | Improved |
| use progressive patch instead of multi-scale patch (n=5) (ablation) | P-PatchDiff · LOL-v2-Real | PSNR: 26.46 -> 28.04 | Improved |
| use progressive patch instead of fixed patch with p=192 (ablation) | P-PatchDiff · LOL-v2-Real | PSNR: 26.54 -> 28.04 | Improved |
| use progressive stride with progressive patch (ablation) | P-PatchDiff · LOL-v2-Real | PSNR: 21.83 -> 28.04 | Improved |
| use progressive stride with fixed patch (ablation) | fixed patch · LOL-v2-Real | PSNR: 26.54 -> 26.35 | No clear effect |
| increase number of scales n from 5 to 6 (ablation) | P-PatchDiff · LOL-v2-Real | PSNR: 28.04 -> 27.85 | No clear effect |
| increase number of scales n from 5 to 7 (ablation) | P-PatchDiff · LOL-v2-Real | PSNR: 28.04 -> 27.92 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- Restormer — Restormer: Efficient Transformer for High-Resolution Image Restoration (CVPR 2022) · better on 1 of 2 shared results
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
- Multi-Domain Multi-Scale Diffusion Model for Low-Light Image Enhancement (AAAI 2024 · similarity 0.90)
- Patched Denoising Diffusion Models For High-Resolution Image Synthesis (ICLR 2024 · similarity 0.89)
- AGLLDiff: Guiding Diffusion Models Towards Unsupervised Training-free Real-world Low-light Image Enhancement (AAAI 2025 · similarity 0.89)
- Diff-Shadow: Global-guided Diffusion Model for Shadow Removal (AAAI 2025 · similarity 0.88)
- Fourier Priors-Guided Diffusion for Zero-Shot Joint Low-Light Enhancement and Deblurring (CVPR 2024 · similarity 0.88)
- PHR-DIFF: Portrait Highlights Removal via Patch-aware Diffusion Model (AAAI 2025 · similarity 0.87)
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.
