Iterative Residual Image Deconvolution

TL;DR

This paper introduces an iterative residual deconvolution algorithm and a novel CNN architecture, CRCNet, that effectively improves image deblurring by leveraging residual components and achieving superior results.

Contribution

The paper presents a new iterative residual deconvolution method and a residual CNN architecture, CRCNet, which enhances image deblurring performance and interpretability.

Findings

CRCNet outperforms state-of-the-art methods in quantitative metrics.

CRCNet recovers more visually plausible texture details.

The iterative residual approach provides insights into MMSE solutions for deblurring.

Abstract

Image deblurring, a.k.a. image deconvolution, recovers a clear image from pixel superposition caused by blur degradation. Few deep convolutional neural networks (CNN) succeed in addressing this task. In this paper, we first demonstrate that the minimum-mean-square-error (MMSE) solution to image deblurring can be interestingly unfolded into a series of residual components. Based on this analysis, we propose a novel iterative residual deconvolution (IRD) algorithm. Further, IRD motivates us to take one step forward to design an explicable and effective CNN architecture for image deconvolution. Specifically, a sequence of residual CNN units are deployed, whose intermediate outputs are then concatenated and integrated, resulting in concatenated residual convolutional network (CRCNet). The experimental results demonstrate that proposed CRCNet not only achieves better quantitative metrics but also recovers more visually plausible texture details compared with state-of-the-art methods.