Research map / Label-efficient and robust learning / Self-supervised and contrastive
CVF-SID: Cyclic Multi-Variate Function for Self-Supervised Image Denoising by Disentangling Noise From Image
Reyhaneh Neshatavar, Mohsen Yavartanoo, Sanghyun Son, Kyoung Mu Lee
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 |
|---|---|---|---|
| train directly on the evaluation dataset rather than SIDD-Medium | CVF-SID · SIDD validation | PSNR: 34.51 -> 34.67 | Improved |
| double-train by applying two successive models instead of one | CVF-SID · SIDD validation | PSNR: 34.67 -> 34.81 | Improved |
| add identity loss (ablation) | CVF-SID · SIDD validation | PSNR: 33.10 -> 34.24 | Improved |
| add zero loss after identity loss (ablation) | CVF-SID · SIDD validation | PSNR: 34.24 -> 34.29 | Improved |
| add regularization loss after zero loss (ablation) | CVF-SID · SIDD validation | PSNR: 34.29 -> 34.43 | Improved |
| add augmentation set A (ablation) | CVF-SID · SIDD validation | PSNR: 34.43 -> 34.48 | Improved |
| add augmentation set B to set A (ablation) | CVF-SID · SIDD validation | PSNR: 34.48 -> 34.67 | Improved |
| vary γ from 0.25 through 1.50 (ablation) | CVF-SID · SIDD validation | PSNR: 34.45; 34.46; 34.67; 34.66 | Mixed |
| cascade three models rather than two (ablation) | CVF-SID · SIDD validation | PSNR: 34.81 -> 34.84 | Improved |
| cascade four models rather than three (ablation) | CVF-SID · SIDD validation | PSNR: 34.84 -> 34.84 | No clear effect |
| increase augmentation loss weight from 0.1 to 1 (ablation) | CVF-SID · SIDD validation | PSNR: 34.67 -> 34.55 | Hurt |
Closest papers by meaning
- IDR: Self-Supervised Image Denoising via Iterative Data Refinement (CVPR 2022 · similarity 0.92)
- Iterative Denoiser and Noise Estimator for Self-Supervised Image Denoising (ICCV 2023 · similarity 0.91)
- Noise2Score: Tweedie’s Approach to Self-Supervised Image Denoising without Clean Images (NeurIPS 2021 · similarity 0.91)
- Noise2Same: Optimizing A Self-Supervised Bound for Image Denoising (NeurIPS 2020 · similarity 0.90)
- Convexity-Aware Noise Calibration: A Self-Supervised Framework for Noise-Level-Unknown Image Denoising (CVPR 2026 · similarity 0.90)
- Positive2Negative: Breaking the Information-Lossy Barrier in Self-Supervised Single Image Denoising (CVPR 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.
