Research map / Image generation / Image restoration
Diff-Retinex: Rethinking Low-light Image Enhancement with A Generative Diffusion Model
Xunpeng Yi, Han Xu, Hao Zhang, Linfeng Tang, Jiayi Ma
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 |
|---|---|---|---|
| restore reflectance with diffusion model instead of BM3D or CNN (ablation) | Retinex-based enhancement · LOL reflectance restoration | FID: 111.29 (BM3D), 171.45 (CNN) -> 61.33 | Improved |
| use Diff-Retinex instead of LLFormer on pixel-wise fidelity | LOL · LOL | PSNR: 23.66 -> 21.98 | Hurt |
| use Diff-Retinex instead of URetinex on perceptual metrics | LOL · LOL | FID: 59.00 -> 47.85 | Improved |
| use Diff-Retinex instead of LLFormer on lightness order error | LOL · LOL | LOE: 176.61 -> 191.56 | Hurt |
| use Diff-Retinex on VE-LOL-L versus best baselines | VE-LOL-L · VE-LOL-L | FID: 48.36 -> 47.75 | Improved |
Later papers that compared with this method
2 later accepted papers report Diff-Retinex as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Reti-Diff: Illumination Degradation Image Restoration with Retinex-based Latent Diffusion Model (ICLR 2025 · better on 1 of 1 shared result)
- RAVE: Residual Vector Embedding for CLIP-Guided Backlit Image Enhancement (ECCV 2024 · better on 2 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
- Diff-Retinex++: Retinex-Driven Reinforced Diffusion Model for Low-Light Image Enhancement (TPAMI 2025 · similarity 0.96)
- LightenDiffusion: Unsupervised Low-Light Image Enhancement with Latent-Retinex Diffusion Models (ECCV 2024 · similarity 0.93)
- Reti-Diff: Illumination Degradation Image Restoration with Retinex-based Latent Diffusion Model (ICLR 2025 · similarity 0.90)
- IniRetinex: Rethinking Retinex-type Low-Light Image Enhancer via Initialization Perspective (AAAI 2025 · similarity 0.90)
- Zero-Shot Low-Light Image Enhancement via Latent Diffusion Models (AAAI 2025 · similarity 0.88)
- AGLLDiff: Guiding Diffusion Models Towards Unsupervised Training-free Real-world Low-light Image Enhancement (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.
