Single Frame Deblurring with Laplacian Filters

TL;DR

This paper introduces a novel single-frame blind deblurring method using Laplacian filters and a Residual Dense Network, achieving significant improvements in image quality over existing deep learning approaches.

Contribution

It presents a new deblurring technique combining Laplacian filters with a Residual Dense Network, enhancing image restoration performance.

Findings

Significant improvement in image quality objectively

Better subjective visual results

Outperforms state-of-the-art DNN methods

Abstract

Blind single image deblurring has been a challenge over many decades due to the ill-posed nature of the problem. In this paper, we propose a single-frame blind deblurring solution with the aid of Laplacian filters. Utilized Residual Dense Network has proven its strengths in superresolution task, thus we selected it as a baseline architecture. We evaluated the proposed solution with state-of-art DNN methods on a benchmark dataset. The proposed method shows significant improvement in image quality measured objectively and subjectively.