Single Image Defocus Deblurring Using Kernel-Sharing Parallel Atrous Convolutions

TL;DR

This paper introduces a novel deep learning method using kernel-sharing parallel atrous convolutions for single image defocus deblurring, effectively handling spatially varying blur with fewer parameters and achieving state-of-the-art results.

Contribution

The paper proposes a kernel-sharing parallel atrous convolutional (KPAC) block that exploits inverse kernel properties for improved defocus deblurring with fewer parameters.

Findings

Achieves state-of-the-art deblurring performance.

Uses fewer parameters than previous methods.

Effectively handles spatially varying defocus blur.

Abstract

This paper proposes a novel deep learning approach for single image defocus deblurring based on inverse kernels. In a defocused image, the blur shapes are similar among pixels although the blur sizes can spatially vary. To utilize the property with inverse kernels, we exploit the observation that when only the size of a defocus blur changes while keeping the shape, the shape of the corresponding inverse kernel remains the same and only the scale changes. Based on the observation, we propose a kernel-sharing parallel atrous convolutional (KPAC) block specifically designed by incorporating the property of inverse kernels for single image defocus deblurring. To effectively simulate the invariant shapes of inverse kernels with different scales, KPAC shares the same convolutional weights among multiple atrous convolution layers. To efficiently simulate the varying scales of inverse kernels, KPAC consists of only a few atrous convolution layers with different dilations and learns per-pixel scale attentions to aggregate the outputs of the layers. KPAC also utilizes the shape attention to combine the outputs of multiple convolution filters in each atrous convolution layer, to deal with defocus blur with a slightly varying shape. We demonstrate that our approach achieves state-of-the-art performance with a much smaller number of parameters than previous methods.