Research map / Image generation / Image restoration
DI-Retinex: Digital-Imaging Retinex Model for Low-Light Image Enhancement
Shangquan Sun, Wenqi Ren, Jingyang Peng, Fenglong Song, Xiaochun Cao
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 the offset beta (ablation) | DI-Retinex model · LOL-v1 | PSNR: 10.44 -> 21.54 | Improved |
| add variance suppression loss (ablation) | DI-Retinex model · LOL-v1 | PSNR: 20.15 -> 21.54 | Improved |
| add reverse degradation loss (ablation) | DI-Retinex model · LOL-v1 | PSNR: 17.62 -> 21.54 | Improved |
| add mask function M in reverse degradation loss (ablation) | DI-Retinex model · LOL-v1 | PSNR: 21.52 -> 21.54 | Improved |
| use mapping g instead of g1 (ablation) | DI-Retinex model · LOL-v1 | PSNR: 17.43 -> 21.54 | Improved |
| use mapping g instead of g2 (ablation) | DI-Retinex model · LOL-v1 | PSNR: 20.34 -> 21.54 | Improved |
| use g3 (log of g) instead of g (ablation) | DI-Retinex model · LOL-v1 | PSNR: 21.54 -> 21.62; FPS 954 -> 831 | No clear effect |
| predict matrix b and matrix c instead of scalar b and c (ablation) | DI-Retinex model · LOL-v1 | PSNR: 14.29 -> 21.54 | Improved |
| predict matrix c instead of scalar c (with matrix b) (ablation) | DI-Retinex model · LOL-v1 | PSNR: 17.78 -> 21.54 | Improved |
| predict matrix b instead of scalar b (with matrix c) (ablation) | DI-Retinex model · LOL-v1 | PSNR: 19.40 -> 21.54 | Improved |
| full-size model instead of small (channels 64 to 4 internal, 6 to 2 output) | DI-Retinex model · LOL-v1 | PSNR: 18.448 -> 21.542 | Improved |
| full-size model instead of small | DI-Retinex model · LOL-v2 | PSNR: 20.631 -> 21.362 | Improved |
Closest papers by meaning
- Diff-Retinex++: Retinex-Driven Reinforced Diffusion Model for Low-Light Image Enhancement (TPAMI 2025 · similarity 0.92)
- Diff-Retinex: Rethinking Low-light Image Enhancement with A Generative Diffusion Model (ICCV 2023 · similarity 0.91)
- IniRetinex: Rethinking Retinex-type Low-Light Image Enhancer via Initialization Perspective (AAAI 2025 · similarity 0.90)
- Reti-Diff: Illumination Degradation Image Restoration with Retinex-based Latent Diffusion Model (ICLR 2025 · similarity 0.88)
- MambaLLIE: Implicit Retinex-Aware Low Light Enhancement with Global-then-Local State Space (NeurIPS 2024 · similarity 0.86)
- Interpretable Unsupervised Joint Denoising and Enhancement for Real-World low-light Scenarios (ICLR 2025 · similarity 0.85)
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.
