Research map / Image generation / Image restoration
Elucidating the Design Space of Arbitrary-Noise-Based Diffusion Models
Xingyu Qiu, Mengying Yang, Xinghua Ma, Dong Liang, Fanding Li, Gongning Luo, Wei Wang, Kuanquan Wang, Shuo Li
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 |
|---|---|---|---|
| start reverse process from degraded image with arbitrary-noise diffusion (EDA) instead of Gaussian-noise EDM (Refusion) | MRI bias field correction · HCP MRI | PSNR: 34.67 -> 38.02 | Improved |
| use EDA with 5 steps instead of 100-step Refusion | MRI bias field correction · HCP MRI | sec/slice: 9.665 -> 0.182 | Improved |
| use EDA instead of Refusion (image-domain) | CT metal artifact reduction · DeepLesion average | PSNR: 38.15 -> 38.67 | Improved |
| use EDA instead of Refusion on large-metal case | CT metal artifact reduction · DeepLesion large metal | PSNR: 33.66 -> 32.77 | Hurt |
| use image-domain EDA instead of dual-domain DICDNet | CT metal artifact reduction · DeepLesion average | PSNR: 41.83 -> 38.67 | Hurt |
| use EDA instead of ShadowFormer | shadow removal · ISTD all image | PSNR: 31.81 -> 32.01 | Improved |
| use EDA instead of ShadowFormer in shadow region | shadow removal · ISTD shadow region | PSNR: 37.99 -> 37.80 | No clear effect |
| use SDE-based EDA instead of ODE-based MeanFlow | shadow removal · ISTD all image | RMSE (LAB): 9.77 -> 4.14 | Improved |
Closest papers by meaning
- Denoising as Adaptation: Noise-Space Domain Adaptation for Image Restoration (ICLR 2025 · similarity 0.88)
- Residual Denoising Diffusion Models (CVPR 2024 · similarity 0.88)
- Resfusion: Denoising Diffusion Probabilistic Models for Image Restoration Based on Prior Residual Noise (NeurIPS 2024 · similarity 0.88)
- Diffusion Priors for Variational Likelihood Estimation and Image Denoising (NeurIPS 2024 · similarity 0.87)
- A Variational Perspective on Solving Inverse Problems with Diffusion Models (ICLR 2024 · similarity 0.87)
- An Edit Friendly DDPM Noise Space: Inversion and Manipulations (CVPR 2024 · similarity 0.87)
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.
