Research map / Image generation / Image restoration
Global Structure-Aware Diffusion Process for Low-light Image Enhancement
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 rank-based global structure-aware regularization (hierarchical clustering) to the diffusion baseline (ablation) | diffusion LLIE baseline · LOLv1 | PSNR: 26.02 -> 27.70 | Helped |
| use rank-based regularization instead of L1 loss (ablation) | GSAD · LOLv1 | PSNR: 26.45 -> 27.70 | Helped |
| use rank-based regularization instead of SSIM loss (ablation) | GSAD · LOLv1 | PSNR: 26.23 -> 27.70 | Helped |
| use hierarchical clustering instead of K-Means for patch grouping (ablation) | GSAD · LOLv1 | PSNR: 27.02 -> 27.70 | Helped |
| use hierarchical clustering instead of Gaussian mixture model (ablation) | GSAD · LOLv1 | LPIPS: 0.095 -> 0.092 | Helped |
| remove adaptive regularization schedule from rank-based module (ablation) | GSAD · LOLv1 | PSNR/SSIM/LPIPS | Hurt |
| add uncertainty-guided regularization alone (ablation) | diffusion baseline · LOLv1 | PSNR/SSIM vs LPIPS | Mixed |
| GSAD vs SNR-Aware | GSAD · LOLv1 | PSNR: 26.716 -> 27.839 | Helped |
| GSAD inference cost vs SNR-Aware | GSAD · LOLv1 | inference time (s): 0.09 -> 0.55 | Hurt |
Closest papers by meaning
- Multi-Domain Multi-Scale Diffusion Model for Low-Light Image Enhancement (AAAI 2024 · similarity 0.90)
- Efficient Diffusion as Low Light Enhancer (CVPR 2025 · similarity 0.89)
- Zero-Shot Low-Light Image Enhancement via Latent Diffusion Models (AAAI 2025 · similarity 0.89)
- Fourier Priors-Guided Diffusion for Zero-Shot Joint Low-Light Enhancement and Deblurring (CVPR 2024 · similarity 0.88)
- AGLLDiff: Guiding Diffusion Models Towards Unsupervised Training-free Real-world Low-light Image Enhancement (AAAI 2025 · similarity 0.88)
- A Variational Perspective on Solving Inverse Problems with Diffusion Models (ICLR 2024 · similarity 0.88)
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.