Blind Image Deconvolution using Deep Generative Priors

TL;DR

This paper introduces a novel blind image deblurring method using deep generative priors for images and kernels, demonstrating effective results even with large blurs and noise, and exploring untrained networks as priors.

Contribution

It presents a new approach employing separate generative models for images and kernels with an alternating gradient descent scheme, and extends to untrained networks as priors.

Findings

Effective deblurring on heavily blurred images.

Augmentation with classical priors improves performance.

Untrained structured networks can serve as image priors.

Abstract

This paper proposes a novel approach to regularize the \textit{ill-posed} and \textit{non-linear} blind image deconvolution (blind deblurring) using deep generative networks as priors. We employ two separate generative models --- one trained to produce sharp images while the other trained to generate blur kernels from lower-dimensional parameters. To deblur, we propose an alternating gradient descent scheme operating in the latent lower-dimensional space of each of the pretrained generative models. Our experiments show promising deblurring results on images even under large blurs, and heavy noise. To address the shortcomings of generative models such as mode collapse, we augment our generative priors with classical image priors and report improved performance on complex image datasets. The deblurring performance depends on how well the range of the generator spans the image class. Interestingly, our experiments show that even an untrained structured (convolutional) generative networks acts as an image prior in the image deblurring context allowing us to extend our results to more diverse natural image datasets.