Image Restoration with Mean-Reverting Stochastic Differential Equations
Jul 24, 2023
Speakers
Fredrik K Gustafsson
Speaker · 0 followers
Zheng Zhao
Speaker · 0 followers
Jens Sjölund
Speaker · 0 followers
Thomas Schön
Speaker · 0 followers
About
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 which stabilizes the training and improves the restoration results. In the experiments, we 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.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…
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Toward Fast Grad-Based Value Estimation
Watch later
Generalization and Corruption Resistance via Distributionally Robust Optimization
Watch later
Brauer's Group Equivariant Neural Networks
Watch later
Diffusion Models Are Minimax Optimal Distribution Estimators
Watch later
Kazusato Oko, …
Neural signature kernels as infinite-width-depth-limits of controlled ResNets
Watch later
Differentially Private Optimization on Large Model at Small Cost
Watch later
Zhiqi Bu, …