Skip to main content

AS seminar: Deep Learning in Zero-shot Blind Image Deconvolution


The aim of blind image deconvolution is to recover a sharp image from a blurred one. Assuming that there is no other data than the one blurred image, the problem is highly ill-posed.  Many methods were proposed, yet there is none that would be 100% reliable. Approaches based on Bayesian models that attempt to describe image statistics using priors played major role in zero-shot blind image deblurring until a deep image prior (DIP) and subsequently a DIP framework for blind image deconvolution were proposed. In this seminar, these approaches will be described and the reasons of their success will be discussed.

Submitted by neuner on