Research map / Image generation / Image restoration
Enhanced Latent Space Blind Model for Real Image Denoising via Alternative Optimization
Chao Ren, Yizhong Pan, Jie Huang
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 latent space encoding/decoding (ablation) | ScaoedNet · SIDD validation | PSNR (dB)/SSIM: 39.19 -> 39.48 | Improved |
| add self-correction scheme (ablation) | ScaoedNet · SIDD validation | PSNR (dB)/SSIM: 39.22 -> 39.48 | Improved |
| add guidance constraint (G-Module and DEM) (ablation) | ScaoedNet · SIDD validation | PSNR (dB)/SSIM: 0.13dB/0.0015 gain | Improved |
| add FM2A attention module (ablation) | ScaoedNet · SIDD validation | PSNR (dB)/SSIM: 39.28 -> 39.48 | Improved |
| add SSIM loss (ablation) | ScaoedNet · SIDD validation | PSNR (dB)/SSIM: 39.46 -> 39.48 | No clear effect |
| add NFAE layer to noise estimation network (ablation) | uini-Net · BSD68/Kodak24 noise map | L1 distance: 5.97 -> 5.80 | Improved |
| add NFAE layer to noise estimation network (ablation) | uini-Net · BSD68/Kodak24 noise map | PSNR (dB): 32.53 -> 33.16 | Improved |
| increase iteration stages K from 1 to 5 | ScaoedNet · SIDD validation | PSNR (dB): 39.24 -> 39.48 | Improved |
| increase K from 5 to 6 | ScaoedNet · SIDD validation | PSNR (dB): 39.48 -> 39.48 | No clear effect |
| increase FM2ARB number T from 4 to 16 | ScaoedNet · SIDD validation | PSNR (dB): 39.10 -> 39.48 | Improved |
| increase FM2ARB number T from 16 to 20 | ScaoedNet · SIDD validation | PSNR (dB): 39.48 -> 39.49 | No clear effect |
| apply self-ensemble at test time | ScaoedNet · DnD benchmark | PSNR (dB): 40.12 -> 40.17 | Improved |
Closest papers by meaning
- Unpaired Learning of Deep Image Denoising (ECCV 2020 · similarity 0.89)
- Variational Denoising Network: Toward Blind Noise Modeling and Removal (NeurIPS 2019 · similarity 0.88)
- Real Image Denoising With Feature Attention (ICCV 2019 · similarity 0.87)
- Toward Convolutional Blind Denoising of Real Photographs (CVPR 2019 · similarity 0.87)
- An Unsupervised Deep Learning Approach for Real-World Image Denoising (ICLR 2021 · similarity 0.87)
- APR-RD: Complemental Two Steps for Self-Supervised Real Image Denoising (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.
