wyhuai/DDNM
[ICLR 2023 Oral] Zero-Shot Image Restoration Using Denoising Diffusion Null-Space Model
Leverages a null-space decomposition approach with pre-trained diffusion models (guided-diffusion, SDEdit) to solve inverse problems by sampling only in the null-space of degradation operators, eliminating need for task-specific training. Offers both SVD-based and simplified versions—the latter enables flexible custom degradation definitions—with configurable time-travel sampling parameters for quality-speed tradeoffs across restoration tasks.
1,331 stars. No commits in the last 6 months.
Stars
1,331
Forks
104
Language
Python
License
MIT
Category
Last pushed
Apr 25, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/wyhuai/DDNM"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
guyyariv/DyPE
Official implementation for "DyPE: Dynamic Position Extrapolation for Ultra High Resolution Diffusion".
devzhk/InverseBench
InverseBench (ICLR 2025 spotlight)
bahjat-kawar/ddrm
[NeurIPS 2022] Denoising Diffusion Restoration Models -- Official Code Repository
yuanzhi-zhu/DiffPIR
"Denoising Diffusion Models for Plug-and-Play Image Restoration", Yuanzhi Zhu, Kai Zhang,...
Algolzw/image-restoration-sde
Image Restoration with Mean-Reverting Stochastic Differential Equations, ICML 2023. Winning...