Gaussian Kernel Mixture Network for Single Image Defocus Deblurring

TL;DR

This paper introduces GKMNet, a deep learning model that effectively removes spatially variant defocus blur from single images by using a novel Gaussian kernel mixture model, improving accuracy and efficiency.

Contribution

The paper proposes a new pixel-wise Gaussian kernel mixture model and a deep neural network that unrolls fixed-point iterations for superior defocus deblurring performance.

Findings

GKMNet outperforms existing methods in accuracy.

GKMNet has lower model complexity.

GKMNet is more computationally efficient.

Abstract

Defocus blur is one kind of blur effects often seen in images, which is challenging to remove due to its spatially variant amount. This paper presents an end-to-end deep learning approach for removing defocus blur from a single image, so as to have an all-in-focus image for consequent vision tasks. First, a pixel-wise Gaussian kernel mixture (GKM) model is proposed for representing spatially variant defocus blur kernels in an efficient linear parametric form, with higher accuracy than existing models. Then, a deep neural network called GKMNet is developed by unrolling a fixed-point iteration of the GKM-based deblurring. The GKMNet is built on a lightweight scale-recurrent architecture, with a scale-recurrent attention module for estimating the mixing coefficients in GKM for defocus deblurring. Extensive experiments show that the GKMNet not only noticeably outperforms existing defocus deblurring methods, but also has its advantages in terms of model complexity and computational efficiency.