Multi-Domain Image-to-Image Translation via a Unified Circular Framework

doi:10.1109/TIP.2020.3037528

CORC > 自动化研究所 > 中国科学院自动化研究所 > 智能感知与计算研究中心

	Multi-Domain Image-to-Image Translation via a Unified Circular Framework
	Wang, Yuxi 1,2; Zhang, Zhaoxiang 1,2; Hao, Wangli 1,2; Song, Chunfeng 1,2
刊名	IEEE TRANSACTIONS ON IMAGE PROCESSING
	2021
卷号	30 页码:670-684
关键词	Task analysis Semantics Visualization Generative adversarial networks Generators Feature extraction Meteorology Image-to-image transfer multiple domain pairs sharing knowledge module GANs
ISSN号	1057-7149
DOI	10.1109/TIP.2020.3037528
通讯作者	Zhang, Zhaoxiang(zhaoxiang.zhang@ia.ac.cn)
英文摘要	The image-to-image translation aims to learn the corresponding information between the source and target domains. Several state-of-the-art works have made significant progress based on generative adversarial networks (GANs). However, most existing one-to-one translation methods ignore the correlations among different domain pairs. We argue that there is common information among different domain pairs and it is vital to multiple domain pairs translation. In this paper, we propose a unified circular framework for multiple domain pairs translation, leveraging a shared knowledge module across numerous domains. One selected translation pair can benefit from the complementary information from other pairs, and the sharing knowledge is conducive to mutual learning between domains. Moreover, absolute consistency loss is proposed and applied in the corresponding feature maps to ensure intra-domain consistency. Furthermore, our model can be trained in an end-to-end manner. Extensive experiments demonstrate the effectiveness of our approach on several complex translation scenarios, such as Thermal IR switching, weather changing, and semantic transfer tasks.
资助项目	Major Project for New Generation of AI[2018AAA0100400] ; National Natural Science Foundation of China[61836014] ; National Natural Science Foundation of China[61761146004] ; National Natural Science Foundation of China[61773375]
WOS关键词	ADVERSARIAL NETWORKS
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000597161500005
资助机构	Major Project for New Generation of AI ; National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/42692]
专题	自动化研究所_智能感知与计算研究中心
通讯作者	Zhang, Zhaoxiang
作者单位	1.Chinese Acad Sci, Natl Lab Pattern Recognit, Ctr Res Intelligent Percept & Comp, Inst Automat, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Wang, Yuxi,Zhang, Zhaoxiang,Hao, Wangli,et al. Multi-Domain Image-to-Image Translation via a Unified Circular Framework[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2021,30:670-684.
APA	Wang, Yuxi,Zhang, Zhaoxiang,Hao, Wangli,&Song, Chunfeng.(2021).Multi-Domain Image-to-Image Translation via a Unified Circular Framework.IEEE TRANSACTIONS ON IMAGE PROCESSING,30,670-684.
MLA	Wang, Yuxi,et al."Multi-Domain Image-to-Image Translation via a Unified Circular Framework".IEEE TRANSACTIONS ON IMAGE PROCESSING 30(2021):670-684.