Generative adversarial networks via a composite annealing of noise and diffusion
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Nakamura, Kensuke | - |
dc.contributor.author | Korman, Simon | - |
dc.contributor.author | Hong, Byung-Woo | - |
dc.date.accessioned | 2023-11-15T02:41:28Z | - |
dc.date.available | 2023-11-15T02:41:28Z | - |
dc.date.issued | 2024-02 | - |
dc.identifier.issn | 0031-3203 | - |
dc.identifier.issn | 1873-5142 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/cau/handle/2019.sw.cau/68581 | - |
dc.description.abstract | Generative adversarial network (GAN) is a framework for generating fake data using a set of real examples. However, GAN is unstable in the training stage. In order to stabilize GANs, the noise injection has been used to enlarge the overlap of the real and fake distributions at the cost of increasing variance. The diffusion process (or data smoothing in its spatial domain) removes fine details in order to capture the structure and important patterns in data but it suppresses the capability of GANs to learn high-frequency information in the training procedure. Based on these observations, we propose a data representation for the GAN training, called noisy scale-space (NSS), that recursively applies the smoothing with a balanced noise to data in order to replace the high-frequency information by random data, leading to a coarse-to-fine training of GANs. We experiment with NSS using DCGAN and StyleGAN2 based on benchmark datasets in which the NSS-based GANs outperforms the state-of-the-arts in most cases. © 2023 Elsevier Ltd | - |
dc.language | 영어 | - |
dc.language.iso | ENG | - |
dc.publisher | Elsevier Ltd | - |
dc.title | Generative adversarial networks via a composite annealing of noise and diffusion | - |
dc.type | Article | - |
dc.identifier.doi | 10.1016/j.patcog.2023.110034 | - |
dc.identifier.bibliographicCitation | Pattern Recognition, v.146 | - |
dc.description.isOpenAccess | N | - |
dc.identifier.wosid | 001102957500001 | - |
dc.identifier.scopusid | 2-s2.0-85173876524 | - |
dc.citation.title | Pattern Recognition | - |
dc.citation.volume | 146 | - |
dc.type.docType | Article | - |
dc.publisher.location | 영국 | - |
dc.subject.keywordAuthor | Coarse-to-fine training | - |
dc.subject.keywordAuthor | Generative adversarial networks | - |
dc.subject.keywordAuthor | Noise injection | - |
dc.subject.keywordAuthor | Optimization | - |
dc.subject.keywordAuthor | Scale-space | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalResearchArea | Engineering | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Artificial Intelligence | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.description.journalRegisteredClass | scopus | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
84, Heukseok-ro, Dongjak-gu, Seoul, Republic of Korea (06974)02-820-6194
COPYRIGHT 2019 Chung-Ang University All Rights Reserved.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.