Deep Image Prior

Michael Scherbela, 02. Nov 2022

Deep convolutional networks have become a popular tool for image generation and restoration. Generally, their excellent performance is imputed to their ability to learn realistic image priors from a large number of example images. In this paper, we show that, on the contrary, the structure of a generator network is sufficient to capture a great deal of low-level image statistics prior to any learning.

Deep Image Prior, Ulyanov et al. 2017