Research map / Image generation / GANs
Image Blind Denoising With Generative Adversarial Network Based Noise Modeling
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 |
|---|---|---|---|
| Use GAN-generated noise samples to augment extracted noise blocks when training the denoising CNN (ablation) | GCBD · BSD68, synthetic Gaussian and mixture noise | PSNR (dB): 0.34 dB and 0.91 dB lower | Helped |
| Use GAN-based noise modeling and generated training data rather than the blind Gaussian DnCNN-B model | GCBD · BSD68, mixture noise s = 15 and s = 25 | PSNR (dB): 40.75 -> 42.00; 37.54 -> 39.87 | Helped |
| Use GAN-based noise modeling and generated training data rather than BM3D | GCBD · DND | PSNR (dB): 34.61 -> 35.58 | Helped |
| Train with approximate noise data modeled by GAN instead of accurate synthetic noise data | GCBD · BSD68, mixture noise s = 15 and s = 25 | PSNR (dB): 42.42 dB and 40.21 dB; 42.46 dB and 40.22 dB | No clear effect |
| Use GAN rather than pixel-based GMM for noise modeling | GCBD · Real-world noise sample modeling | noise sample similarity to observed noise | Helped |
| Apply GCBD rather than DnCNN-B | BSD68 Gaussian blind denoising · Gaussian noise σ = 15 and σ = 25 | PSNR (dB): 31.61 -> 31.59; 29.16 -> 29.15 | No clear effect |
Closest papers by meaning
- GAN2GAN: Generative Noise Learning for Blind Denoising with Single Noisy Images (ICLR 2021 · similarity 0.93)
- End-to-End Unpaired Image Denoising with Conditional Adversarial Networks (AAAI 2020 · similarity 0.91)
- sRGB Real Noise Modeling via Noise-Aware Sampling with Normalizing Flows (ICLR 2024 · similarity 0.88)
- Robust Image Denoising through Adversarial Frequency Mixup (CVPR 2024 · similarity 0.87)
- Is Noise Conditioning Necessary for Denoising Generative Models? (ICML 2025 · similarity 0.86)
- LAN: Learning to Adapt Noise for Image Denoising (CVPR 2024 · 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.