Publication Type
Journal Article
Version
publishedVersion
Publication Date
2-2023
Abstract
PersonImageSynthesisaimsattransferringtheappearanceofthesourcepersonimageintoatargetpose. Existingmethods cannot handle largeposevariations and therefore suffer fromtwocritical problems: (1)synthesisdistortionduetotheentanglementofposeandappearanceinformationamongdifferentbody componentsand(2)failureinpreservingoriginalsemantics(e.g.,thesameoutfit).Inthisarticle,weexplicitly addressthesetwoproblemsbyproposingaPose-andAttribute-consistentPersonImageSynthesisNetwork (PAC-GAN).Toreduceposeandappearancematchingambiguity,weproposeacomponent-wisetransferring modelconsistingoftwostages.Theformerstagefocusesonlyonsynthesizingtargetposes,whilethelatter renderstargetappearancesbyexplicitlytransferringtheappearanceinformationfromthesourceimageto thetargetimageinacomponent-wisemanner. Inthisway,source-targetmatchingambiguityiseliminated duetothecomponent-wisedisentanglementofposeandappearancesynthesis.Second,tomaintainattribute consistency,werepresenttheinputimageasanattributevectorandimposeahigh-levelsemanticconstraint usingthisvectortoregularizethetargetsynthesis.ExtensiveexperimentalresultsontheDeepFashiondataset demonstratethesuperiorityofourmethodoverthestateoftheart,especiallyformaintainingposeandattributeconsistenciesunderlargeposevariations.
Keywords
Computing methodologies, Artificial intelligence, Computer vision, Computer graphics, Image manipulation, Image processing
Discipline
Artificial Intelligence and Robotics
Research Areas
Information Systems and Management
Publication
ACM Transactions on Multimedia Computing, Communications and Applications
Volume
19
Issue
2
First Page
1
Last Page
21
ISSN
1551-6857
Identifier
10.1145/3554739
Publisher
Association for Computing Machinery (ACM)
Citation
XU, Cheng; CHEN, Zejun; MAI, Jiajie; XU, Xuemiao; and HE, Shengfeng.
Pose- and Attribute-consistent Person Image Synthesis. (2023). ACM Transactions on Multimedia Computing, Communications and Applications. 19, (2), 1-21.
Available at: https://ink.library.smu.edu.sg/sis_research/8369
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Additional URL
https://doi.org/10.1145/3554739