Image Restoration with Mean-Reverting Stochastic Differential Equations
Authors:
- Ziwei LuoZLZiwei Luo
- Fredrik K. GustafssonFGFredrik K. Gustafsson
- Zheng ZhaoZZZheng Zhao
- Jens SjölundJSJens Sjölund
- Thomas B. SchönTSThomas B. Schön
Preprint
Abstract
This paper presents a stochastic differential equation (SDE) approach for general-purpose image restoration. The key construction consists in a mean-reverting SDE that transforms a high-quality image into a degraded counterpart as a mean state with fixed Gaussian noise. Then, by simulating the corresponding reverse-time SDE, we are able to restore the origin of the low-quality image without relying on any task-specific prior knowledge. Crucially, the proposed mean-reverting SDE has a closed-form solution, allowing us to compute the ground truth time-dependent score and learn it with a neural network. Moreover, we propose a maximum likelihood objective to learn an optimal reverse trajectory that stabilizes the training and improves the restoration results. The experiments show that our proposed method achieves highly competitive performance in quantitative comparisons on image deraining, deblurring, and denoising, setting a new state-of-the-art on two deraining datasets. Finally, the general applicability of our approach is further demonstrated via qualitative results on image super-resolution, inpainting, and dehazing. Code is available at https://github.com/Algolzw/image-restoration-sde.Keywords
All Related Versions
This publication has 18 references indexed in Scilit:
- Speech Enhancement and Dereverberation With Diffusion-Based Generative ModelsIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023
- SRDiff: Single image super-resolution with diffusion probabilistic modelsNeurocomputing, 2022
- Image Restoration: From Sparse and Low-Rank Priors to Deep Priors [Lecture Notes]IEEE Signal Processing Magazine, 2017
- Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image DenoisingIEEE Transactions on Image Processing, 2017
- Waterloo Exploration Database: New Challenges for Image Quality Assessment ModelsIEEE Transactions on Image Processing, 2016
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004
- Digital image restorationIEEE Signal Processing Magazine, 1997
- Exact numerical simulation of the Ornstein-Uhlenbeck process and its integralPhysical Review E, 1996
- Digital image restoration: A surveyComputer, 1974
- The Application of Constrained Least Squares Estimation to Image Restoration by Digital ComputerIEEE Transactions on Computers, 1973