# On learning optimized reaction diffusion processes for effective image restoration

arXiv: https://arxiv.org/abs/1503.05768 · HTML: https://mlsanity.com/paper/1503.05768/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Mar 2015.

## Built on by (most cited first, 25 of 25)

- [Physics-based Noise Modeling for Extreme Low-light Photography](https://mlsanity.com/paper/2108.02158/) (2021)
- [Mobile Computational Photography: A Tour](https://mlsanity.com/paper/2102.09000/) (2021)
- [Encoding physics to learn reaction-diffusion processes](https://mlsanity.com/paper/2106.04781/) (2021)
- [DWDN: Deep Wiener Deconvolution Network for Non-Blind Image Deblurring](https://mlsanity.com/paper/2103.09962/) (2021)
- [Learning to Generate Realistic Noisy Images via Pixel-level Noise-aware Adversarial Training](https://mlsanity.com/paper/2204.02844/) (2022)
- [Recovery Analysis for Plug-and-Play Priors using the Restricted Eigenvalue Condition](https://mlsanity.com/paper/2106.03668/) (2021)
- [Towards General Low-Light Raw Noise Synthesis and Modeling](https://mlsanity.com/paper/2307.16508/) (2023)
- [SGD-Net: Efficient Model-Based Deep Learning with Theoretical Guarantees](https://mlsanity.com/paper/2101.09379/) (2021)
- [Deformation-Compensated Learning for Image Reconstruction without Ground Truth](https://mlsanity.com/paper/2107.05533/) (2021)
- [Hard Encoding of Physics for Learning Spatiotemporal Dynamics](https://mlsanity.com/paper/2105.00557/) (2021)
- [Super Neurons](https://mlsanity.com/paper/2109.01594/) (2021)
- [ELMformer: Efficient Raw Image Restoration with a Locally Multiplicative Transformer](https://mlsanity.com/paper/2208.14704/) (2022)
- [Recurrent Inference Machines as inverse problem solvers for MR relaxometry](https://mlsanity.com/paper/2106.07379/) (2021)
- [Assessment of Data Consistency through Cascades of Independently Recurrent Inference Machines for fast and robust accelerated MRI reconstruction](https://mlsanity.com/paper/2111.15498/) (2021)
- [RDGCL: Reaction-Diffusion Graph Contrastive Learning for Recommendation](https://mlsanity.com/paper/2312.16563/) (2023)
- [How to Best Combine Demosaicing and Denoising?](https://mlsanity.com/paper/2408.06684/) (2024)
- [LW-ISP: A Lightweight Model with ISP and Deep Learning](https://mlsanity.com/paper/2210.03904/) (2022)
- [Neural Stochastic Differential Equations for Robust and Explainable Analysis of Electromagnetic Unintended Radiated Emissions](https://mlsanity.com/paper/2309.15386/) (2023)
- [Image Segmentation via Variational Model Based Tailored UNet: A Deep Variational Framework](https://mlsanity.com/paper/2505.05806/) (2025)
- [Deep Iteration Assisted by Multi-level Obey-pixel Network Discriminator (DIAMOND) for Medical Image Recovery](https://mlsanity.com/paper/2102.06102/) (2021)
- [A Note on Machine Learning Approach for Computational Imaging](https://mlsanity.com/paper/2202.11883/) (2022)
- [Deep Convolutional Framelet Denoising for Panoramic by Mixed Wavelet Integration](https://mlsanity.com/paper/2302.10306/) (2023)
- [Inverse Evolution Layers: Physics-informed Regularizers for Deep Neural Networks](https://mlsanity.com/paper/2307.07344/) (2023)
- [Improving Robustness of Spectrogram Classifiers with Neural Stochastic Differential Equations](https://mlsanity.com/paper/2409.01532/) (2024)
- [Human Knowledge Integrated Multi-modal Learning for Single Source Domain Generalization](https://mlsanity.com/paper/2603.12369/) (2026)

## Abstract (arXiv)

For several decades, image restoration remains an active research topic in low-level computer vision and hence new approaches are constantly emerging. However, many recently proposed algorithms achieve state-of-the-art performance only at the expense of very high computation time, which clearly limits their practical relevance. In this work, we propose a simple but effective approach with both high computational efficiency and high restoration quality. We extend conventional nonlinear reaction diffusion models by several parametrized linear filters as well as several parametrized influence functions. We propose to train the parameters of the filters and the influence functions through a loss based approach. Experiments show that our trained nonlinear reaction diffusion models largely benefit from the training of the parameters and finally lead to the best reported performance on common test datasets for image restoration. Due to their structural simplicity, our trained models are highly efficient and are also well-suited for parallel computation on GPUs.