Research map / Image generation / Image restoration
Learning to Restore Low-Light Images via Decomposition-and-Enhancement
Ke Xu, Xin Yang, Baocai Yin, Rynson W.H. Lau
Check 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 |
|---|---|---|---|
| use frequency-based decomposition and two-stage enhancement | low-light image restoration · sRGB test set | PSNR: 21.55 -> 22.13 | Improved |
| use frequency-based decomposition and two-stage enhancement | low-light image restoration · RAW test set | PSNR: 29.13 -> 29.56 | Improved |
| add the ACE module for frequency-aware feature selection (ablation) | decomposition-and-enhancement model · sRGB test set | PSNR: 21.34 -> 22.13 | Improved |
| use ACE rather than a non-local block (ablation) | decomposition-and-enhancement model · sRGB test set | PSNR: 21.49 -> 22.13 | Improved |
| add the CDT module (ablation) | decomposition-and-enhancement model · sRGB test set | PSNR: 21.47 -> 22.13 | Improved |
| use the CDT contrast-aware map rather than the ACE contrast-aware map (ablation) | decomposition-and-enhancement model · sRGB test set | PSNR: 21.84 -> 22.13 | Improved |
| add perceptual loss (ablation) | decomposition-and-enhancement model · sRGB test set | PSNR: 22.03 -> 22.13 | Improved |
| use a two-stage rather than single-stage enhancement model (ablation) | decomposition-and-enhancement model · sRGB test set | PSNR: 21.63 -> 22.13 | Improved |
| supervise the first stage with the low-frequency ground truth rather than the full ground-truth image (ablation) | two-stage decomposition-and-enhancement model · sRGB test set | PSNR: 21.76 -> 22.13 | Improved |
Closest papers by meaning
- Enhancing Low Light Videos by Exploring High Sensitivity Camera Noise (ICCV 2019 · similarity 0.89)
- Interpretable Unsupervised Joint Denoising and Enhancement for Real-World low-light Scenarios (ICLR 2025 · similarity 0.89)
- Contrastive Learning for Low-Light Raw Denoising (Student Abstract) (AAAI 2024 · similarity 0.89)
- ZERO-IG: Zero-Shot Illumination-Guided Joint Denoising and Adaptive Enhancement for Low-Light Images (CVPR 2024 · similarity 0.89)
- Zero-Shot Low-Light Image Enhancement via Latent Diffusion Models (AAAI 2025 · similarity 0.88)
- DPLUT: Unsupervised Low-light Image Enhancement with Lookup Tables and Diffusion Priors (AAAI 2025 · 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.
