Research map / Image generation / Image restoration
LEDNet: Joint Low-Light Enhancement and Deblurring in the Dark
Shangchen Zhou, Chongyi Li, Chen Change Loy
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 PPM (ablation) | LEDNet · LOL-Blur test | PSNR: 21.85 -> 25.74 | Improved |
| add PPM (ablation) | LEDNet · LOL-Blur test | SSIM: 0.781 -> 0.850 | Improved |
| add CurveNLU (ablation) | LEDNet · LOL-Blur test | PSNR: 25.20 -> 25.74 | Improved |
| add CurveNLU (ablation) | LEDNet · LOL-Blur test | SSIM: 0.823 -> 0.850 | Improved |
| use FASC connection instead of concatenation (ablation) | LEDNet · LOL-Blur test | PSNR: 25.31 -> 25.74 | Improved |
| use FASC connection instead of concatenation (ablation) | LEDNet · LOL-Blur test | SSIM: 0.826 -> 0.850 | Improved |
| add enhancement loss L_en (ablation) | LEDNet · LOL-Blur test | PSNR: 24.05 -> 25.74 | Improved |
| add enhancement loss L_en (ablation) | LEDNet · LOL-Blur test | SSIM: 0.784 -> 0.850 | Improved |
| raise CurveNLU order from 3 to 4 (ablation) | LEDNet · LOL-Blur test | PSNR: 25.74 -> 25.77 | No clear effect |
| raise CurveNLU order from 3 to 4 (ablation) | LEDNet · LOL-Blur test | SSIM: 0.850 -> 0.850 | No clear effect |
| lower CurveNLU order from 3 to 2 (ablation) | LEDNet · LOL-Blur test | PSNR: 25.74 -> 25.48 | Hurt |
| train on LOL-Blur with LEDNet instead of MIMO-UNet | LEDNet · LOL-Blur test | PSNR: 22.41 -> 25.74 | Improved |
Later papers that compared with this method
5 later accepted papers report LEDNet as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- PASDiff: Physics-Aware Semantic Guidance for Joint Real-world Low-Light Face Enhancement and Restoration (ECCV 2026 · better on 5 of 5 shared results)
- Improving Image-to-Image Translation via a Rectified Flow Reformulation (ECCV 2026 · better on 1 of 1 shared result)
- DarkIR: Robust Low-Light Image Restoration (CVPR 2025 · better on 0 of 1 shared result)
- VQCNIR: Clearer Night Image Restoration with Vector-Quantized Codebook (AAAI 2024 · better on 6 of 6 shared results)
- Fourier Priors-Guided Diffusion for Zero-Shot Joint Low-Light Enhancement and Deblurring (CVPR 2024 · better on 2 of 4 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
- Interpretable Unsupervised Joint Denoising and Enhancement for Real-World low-light Scenarios (ICLR 2025 · similarity 0.88)
- DarkIR: Robust Low-Light Image Restoration (CVPR 2025 · similarity 0.88)
- NightHaze: Nighttime Image Dehazing via Self-Prior Learning (AAAI 2025 · similarity 0.87)
- Unveiling Details in the Dark: Simultaneous Brightening and Zooming for Low-Light Image Enhancement (AAAI 2024 · similarity 0.86)
- VQCNIR: Clearer Night Image Restoration with Vector-Quantized Codebook (AAAI 2024 · similarity 0.86)
- Enhancing Low-Light Images: A Synthetic Data Perspective on Practical and Generalizable Solutions (AAAI 2025 · similarity 0.86)
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.
