Learning Generic Diffusion Processes for Image Restoration

TL;DR

This paper introduces a generic diffusion process based on TNRD models that can handle multiple Gaussian denoising tasks and transfer to non-blind deconvolution, achieving efficient training and competitive results.

Contribution

The paper proposes a shared diffusion term across multiple TNRD models for various denoising problems, enabling transfer to deconvolution and improving training efficiency.

Findings

Shared diffusion term improves denoising performance.

Transferred diffusion prior benefits non-blind deconvolution.

Achieves competitive results with efficient training.

Abstract

Image restoration problems are typical ill-posed problems where the regularization term plays an important role. The regularization term learned via generative approaches is easy to transfer to various image restoration, but offers inferior restoration quality compared with that learned via discriminative approaches. On the contrary, the regularization term learned via discriminative approaches are usually trained for a specific image restoration problem, and fail in the problem for which it is not trained. To address this issue, we propose a generic diffusion process (genericDP) to handle multiple Gaussian denoising problems based on the Trainable Non-linear Reaction Diffusion (TNRD) models. Instead of one model, which consists of a diffusion and a reaction term, for one Gaussian denoising problem in TNRD, we enforce multiple TNRD models to share one diffusion term. The trained genericDP model can provide both promising denoising performance and high training efficiency compared with the original TNRD models. We also transfer the trained diffusion term to non-blind deconvolution which is unseen in the training phase. Experiment results show that the trained diffusion term for multiple Gaussian denoising can be transferred to image non-blind deconvolution as an image prior and provide competitive performance.