Image Restoration using Autoencoding Priors

TL;DR

This paper introduces a novel image restoration method leveraging denoising autoencoders as priors, enabling state-of-the-art results across multiple tasks without task-specific training.

Contribution

The authors propose using denoising autoencoders as universal priors for various image restoration tasks, eliminating the need for separate training for each task.

Findings

Achieved state-of-the-art results in non-blind deconvolution.

Achieved state-of-the-art results in super-resolution.

Unified approach applicable to multiple restoration tasks.

Abstract

We propose to leverage denoising autoencoder networks as priors to address image restoration problems. We build on the key observation that the output of an optimal denoising autoencoder is a local mean of the true data density, and the autoencoder error (the difference between the output and input of the trained autoencoder) is a mean shift vector. We use the magnitude of this mean shift vector, that is, the distance to the local mean, as the negative log likelihood of our natural image prior. For image restoration, we maximize the likelihood using gradient descent by backpropagating the autoencoder error. A key advantage of our approach is that we do not need to train separate networks for different image restoration tasks, such as non-blind deconvolution with different kernels, or super-resolution at different magnification factors. We demonstrate state of the art results for non-blind deconvolution and super-resolution using the same autoencoding prior.