A Comprehensive Survey on Deep Neural Image Deblurring

TL;DR

This paper provides a comprehensive review of recent deep neural network methods for image deblurring, highlighting architectures, performance, challenges, and future research directions.

Contribution

It offers an extensive survey of deep learning approaches in image deblurring, detailing architectures, datasets, and identifying research gaps.

Findings

Deep neural networks significantly improve image deblurring quality.

Various architectures have been developed with distinct strengths.

Current challenges include dataset limitations and model generalization issues.

Abstract

Image deblurring tries to eliminate degradation elements of an image causing blurriness and improve the quality of an image for better texture and object visualization. Traditionally, prior-based optimization approaches predominated in image deblurring, but deep neural networks recently brought a major breakthrough in the field. In this paper, we comprehensively review the recent progress of the deep neural architectures in both blind and non-blind image deblurring. We outline the most popular deep neural network structures used in deblurring applications, describe their strengths and novelties, summarize performance metrics, and introduce broadly used datasets. In addition, we discuss the current challenges and research gaps in this domain and suggest potential research directions for future works.