DC Field | Value | Language |
---|---|---|
dc.contributor.author | KIM, SEONGTAE | - |
dc.contributor.author | KANG, KYOUNGKOOK | - |
dc.contributor.author | KIM, GEON UNG | - |
dc.contributor.author | BAEK, SEUNG HWAN | - |
dc.contributor.author | CHO, SUNGHYUN | - |
dc.date.accessioned | 2022-11-29T04:40:16Z | - |
dc.date.available | 2022-11-29T04:40:16Z | - |
dc.date.created | 2022-11-24 | - |
dc.date.issued | 2022-12-06 | - |
dc.identifier.uri | https://oasis.postech.ac.kr/handle/2014.oak/114431 | - |
dc.description.abstract | © 2022 ACM.Few-shot domain adaptation to multiple domains aims to learn a complex image distribution across multiple domains from a few training images. A naïve solution here is to train a separate model for each domain using few-shot domain adaptation methods. Unfortunately, this approach mandates linearly-scaled computational resources both in memory and computation time and, more importantly, such separate models cannot exploit the shared knowledge between target domains. In this paper, we propose DynaGAN, a novel few-shot domain-adaptation method for multiple target domains. DynaGAN has an adaptation module, which is a hyper-network that dynamically adapts a pretrained GAN model into the multiple target domains. Hence, we can fully exploit the shared knowledge across target domains and avoid the linearly-scaled computational requirements. As it is still computationally challenging to adapt a large-size GAN model, we design our adaptation module to be lightweight using the rank-1 tensor decomposition. Lastly, we propose a contrastive-adaptation loss suitable for multi-domain few-shot adaptation. We validate the effectiveness of our method through extensive qualitative and quantitative evaluations. | - |
dc.language | English | - |
dc.publisher | ACM | - |
dc.relation.isPartOf | ACM SIGGRAPH ASIA 2022 | - |
dc.relation.isPartOf | Proceedings - SIGGRAPH Asia 2022 Conference Papers | - |
dc.title | DynaGAN: Dynamic Few-shot Adaptation of GANs to Multiple Domains | - |
dc.type | Conference | - |
dc.type.rims | CONF | - |
dc.identifier.bibliographicCitation | ACM SIGGRAPH ASIA 2022 | - |
dc.citation.conferenceDate | 2022-12-06 | - |
dc.citation.conferencePlace | KO | - |
dc.citation.conferencePlace | 대구 | - |
dc.citation.title | ACM SIGGRAPH ASIA 2022 | - |
dc.contributor.affiliatedAuthor | KIM, SEONGTAE | - |
dc.contributor.affiliatedAuthor | KANG, KYOUNGKOOK | - |
dc.contributor.affiliatedAuthor | KIM, GEON UNG | - |
dc.contributor.affiliatedAuthor | BAEK, SEUNG HWAN | - |
dc.contributor.affiliatedAuthor | CHO, SUNGHYUN | - |
dc.identifier.scopusid | 2-s2.0-85143978693 | - |
dc.description.journalClass | 1 | - |
dc.description.journalClass | 1 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.
library@postech.ac.kr Tel: 054-279-2548
Copyrights © by 2017 Pohang University of Science ad Technology All right reserved.