Research map / Image generation / Image restoration
Towards General Low-Light Raw Noise Synthesis and Modeling
Feng Zhang, Bin Xu, Zhiqiang Li, Xinran Liu, Qingbo Lu, Changxin Gao, Nong Sang
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 |
|---|---|---|---|
| separate noise into physics-based signal-dependent and learned signal-independent parts (ablation) | noise synthesis model · SID (x300) | PSNR: 33.05 -> 36.03 | Improved |
| add pre-trained Denoise-Net for image domain alignment (ablation) | noise synthesis model · SID (x300) | PSNR: 35.66 -> 36.03 | Improved |
| replace vanilla transformer discriminator with Fourier transformer discriminator (ablation) | noise synthesis model · SID (x300) | PSNR: 33.22 -> 36.03 | Improved |
| replace vanilla transformer discriminator with Fourier transformer discriminator (ablation) | noise synthesis model · SID (x100) | SSIM: 0.931 -> 0.956 | Improved |
| generative learned noise model instead of ELD physics-based noise model | denoiser training data · LRD (-1EV) | PSNR: 38.31 -> 38.89 | Improved |
| synthetic noise model vs ELD on ELD x100 | denoiser training data · ELD (x100) | PSNR: 45.06 -> 44.95 | Hurt |
| train on synthetic noise instead of real paired data at highest ratio | denoiser training data · SID (x300) | PSNR: 36.55 -> 36.03 | Hurt |
| train on synthetic noise instead of real paired data on SID x100 | denoiser training data · SID (x100) | PSNR: 41.39 -> 41.95 | Improved |
Closest papers by meaning
- 2-Shots in the Dark: Low-Light Denoising with Minimal Data Acquisition (CVPR 2026 · similarity 0.92)
- Rethinking Noise Synthesis and Modeling in Raw Denoising (ICCV 2021 · similarity 0.91)
- A Physics-Based Noise Formation Model for Extreme Low-Light Raw Denoising (CVPR 2020 · similarity 0.91)
- Noise Modeling in One Hour: Minimizing Preparation Efforts for Self-supervised Low-Light RAW Image Denoising (CVPR 2025 · similarity 0.91)
- Calibration-Free Raw Image Denoising via Fine-Grained Noise Estimation (TPAMI 2025 · similarity 0.91)
- Dark Noise Diffusion: Noise Synthesis for Low-Light Image Denoising (TPAMI 2025 · similarity 0.91)
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.
