Noise Model Effect upon the GAN-Synthesized Images

Autor:	V. O. Yachnaya, V. R. Lutsiv, E. N. Yablokov, M. A. Mikhalkova
Rok vydání:	2020
Předmět:	Noise symbols.namesake Materials science Artificial neural network Gaussian noise symbols Salt-and-pepper noise Biological system Generative adversarial network Image contrast
Zdroj:	2020 Wave Electronics and its Application in Information and Telecommunication Systems (WECONF).
DOI:	10.1109/weconf48837.2020.9131529
Popis:	The article is devoted to studying of the images synthesized by Generative Adversarial Networks (GAN). The investigated networks are stimulated with the noise-like vectors of random samples fed into their inputs. The influence of changes of input noise parameters upon the properties of synthesized images is of primary interest in this research.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::0677ba16fd6592222c754367f2d431d0 https://doi.org/10.1109/weconf48837.2020.9131529 Zobrazit plný text záznamu