Deep Image Deblurring: A Survey

Image deblurring is a classic problem in low-level computer vision with the aim to recover sharp image from blurred input image. Advances deep learning have led significant progress solving this problem, and large number of networks been proposed. This paper presents comprehensive timely survey recently published deep-learning based approaches, aiming serve community as useful literature review. We start by discussing common causes blur, introduce benchmark datasets performance metrics, summarize different formulations. Next, we present taxonomy methods using convolutional neural (CNN) on architecture, loss function, application, offering detailed review comparison. In addition, discuss some domain-specific applications including face images, text, stereo pairs. conclude key challenges future research directions.