NoGAN: Deblurring Images without  Adversarial Training

dc.contributor.author	Vei, Roman
dc.date.accessioned	2024-02-15T08:20:37Z
dc.date.available	2024-02-15T08:20:37Z
dc.date.issued	2020
dc.identifier.citation	Vei, Roman. NoGAN: Deblurring Images without Adversarial Training / Vei, Roman; Supervisor: Orest Kupyn; Ukrainian Catholic University, Department of Computer Sciences. – Lviv: 2020. – 34 p.	uk
dc.identifier.uri	https://er.ucu.edu.ua/handle/1/4491
dc.language.iso	en	uk
dc.title	NoGAN: Deblurring Images without Adversarial Training	uk
dc.type	Preprint	uk
dc.status	Публікується вперше	uk
dc.description.abstracten	In this paper, we systematically study generative adversarial networks (GANs) for single image motion deblurring. Firstly, we compare adversarial loss functions, dis- criminators, and other training configurations to find optimal setup. Secondly, we train the blurred image classification model and use it as a pretrained discriminator in the GAN setup. We train GANs in two ways: with frozen and unfrozen discrimi- nator weights.	uk