Research map / Image generation / Image restoration
One-Pot Multi-frame Denoising
Lujia Jin, Qing Guo, Zhao Shi, Lei Zhu, Qian Chen, Qiushi Ren, Yanye Lu
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 |
|---|---|---|---|
| replace N2N pairwise supervision with OPD-AL mutual supervision | N2N · Synthetic Gaussian noise | PSNR: 27.48 -> 28.36 | Improved |
| replace N2N with OPD-AL | N2N · Synthetic Gaussian noise | SSIM: 0.787 -> 0.807 | Improved |
| replace N2N with OPD-AL | N2N · Synthetic Gaussian noise | RMSE: 0.048 -> 0.039 | Improved |
| replace N2N with OPD-RC | N2N · Synthetic Poisson noise | PSNR: 27.28 -> 28.22 | Improved |
| replace N2N with OPD-RC | N2N · OCT speckle noise | PSNR: 28.07 -> 30.69 | Improved |
| use unsupervised OPD-RC instead of supervised N2C | N2C · Synthetic Gaussian noise | PSNR: 28.04 -> 28.15 | Improved |
| use supervised multi-frame BPN instead of unsupervised OPD-AL | OPD-AL · Synthetic Gaussian noise | PSNR: 28.36 -> 33.84 | Hurt |
| use kernel-prediction supervised BPN instead of OPD-RC | OPD-RC · OCT speckle noise | PSNR: 30.69 -> 29.00 | Hurt |
| use self-supervised N2S instead of OPD-RC | OPD-RC · OCT speckle noise | PSNR: 30.69 -> 22.23 | Hurt |
| use alienation loss (OPD-AL) instead of random coupling (OPD-RC) (ablation) | OPD · Synthetic Gaussian noise | PSNR: 28.15 -> 28.36 | Mixed |
| use alienation loss (OPD-AL) instead of random coupling (OPD-RC) (ablation) | OPD · OCT speckle noise | PSNR: 30.69 -> 30.40 | Hurt |
Closest papers by meaning
- IDR: Self-Supervised Image Denoising via Iterative Data Refinement (CVPR 2022 · similarity 0.88)
- Unsupervised Image Restoration Using Partially Linear Denoisers (TPAMI 2021 · similarity 0.87)
- Unsupervised Image Denoising with Score Function (NeurIPS 2023 · similarity 0.87)
- CVF-SID: Cyclic Multi-Variate Function for Self-Supervised Image Denoising by Disentangling Noise From Image (CVPR 2022 · similarity 0.87)
- Noise2Score: Tweedie’s Approach to Self-Supervised Image Denoising without Clean Images (NeurIPS 2021 · similarity 0.87)
- Zero-Shot Noise2Mean: Gap Minimization for Efficient Denoising from a Single Noisy Image (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.
