Generative Latent Kernel Modeling for Blind Motion Deblurring

TL;DR

This paper introduces a deep generative kernel model to improve blind motion deblurring by providing better kernel initialization, reducing sensitivity to initial guesses, and achieving state-of-the-art results on benchmarks.

Contribution

A novel framework using a GAN-based kernel generator and initializer to encode kernel priors and improve BMD performance, adaptable to non-uniform motion deblurring.

Findings

Enhanced kernel initialization reduces sensitivity in BMD.

Achieved state-of-the-art results on benchmark datasets.

Plug-and-play integration with existing BMD methods.

Abstract

Deep prior-based approaches have demonstrated remarkable success in blind motion deblurring (BMD) recently. These methods, however, are often limited by the high non-convexity of the underlying optimization process in BMD, which leads to extreme sensitivity to the initial blur kernel. To address this issue, we propose a novel framework for BMD that leverages a deep generative model to encode the kernel prior and induce a better initialization for the blur kernel. Specifically, we pre-train a kernel generator based on a generative adversarial network (GAN) to aptly characterize the kernel's prior distribution, as well as a kernel initializer to provide a well-informed and high-quality starting point for kernel estimation. By combining these two components, we constrain the BMD solution within a compact latent kernel manifold, thus alleviating the aforementioned sensitivity for kernel initialization. Notably, the kernel generator and initializer are designed to be easily integrated with existing BMD methods in a plug-and-play manner, enhancing their overall performance. Furthermore, we extend our approach to tackle blind non-uniform motion deblurring without the need for additional priors, achieving state-of-the-art performance on challenging benchmark datasets. The source code is available at https://github.com/dch0319/GLKM-Deblur.