NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:114
Title:Blind Super-Resolution Kernel Estimation using an Internal-GAN


		
The paper proposes a method for blind super-resolutions by estimating the kernel with a GAN. The method is based on zero-shot learning: it assumes unknown SR kernel, and thus estimates the kernel in a blind manner at test time. The method improves restoration quality by a large margin with the aid of the accurately estimated SR kernel. The paper is well written. Reviewers agreed since the beginning on the acceptance and are satisfied by the rebuttal. Thus the area chair agrees with an acceptance.