Deep Mean-Shift Priors for Image Restoration

TL;DR

This paper introduces a Gaussian-smoothed natural image prior based on mean-shift vectors, enabling effective noise-blind image restoration through a Bayesian framework and autoencoder-learned gradients.

Contribution

It presents a novel image prior that directly models the smoothed image distribution and leverages autoencoders to learn mean-shift vectors for restoration tasks.

Findings

Achieves competitive results in noise-blind deblurring.

Effective in super-resolution and demosaicing.

Demonstrates the utility of mean-shift priors in image restoration.

Abstract

In this paper we introduce a natural image prior that directly represents a Gaussian-smoothed version of the natural image distribution. We include our prior in a formulation of image restoration as a Bayes estimator that also allows us to solve noise-blind image restoration problems. We show that the gradient of our prior corresponds to the mean-shift vector on the natural image distribution. In addition, we learn the mean-shift vector field using denoising autoencoders, and use it in a gradient descent approach to perform Bayes risk minimization. We demonstrate competitive results for noise-blind deblurring, super-resolution, and demosaicing.