Please use this identifier to cite or link to this item: http://drsr.daiict.ac.in//handle/123456789/959
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorJat, P.M.
dc.contributor.authorKapadiya, Ashik Kumar
dc.date.accessioned2020-09-22T05:43:38Z
dc.date.available2023-02-17T05:43:38Z
dc.date.issued2020
dc.identifier.citationKapadiya, Ashik Kumar (2020). Avatar retargeting. Dhirubhai Ambani Institute of Information and Communication Technology. v, 9 p. (Acc.No: T00880)
dc.identifier.urihttp://drsr.daiict.ac.in//handle/123456789/959
dc.description.abstractHuman Motion analysis has variety of applications including animation and content creation. In order to extract human motion and apply it to other character with different body structure and limb sizes, we require to have disentanglement between set of latent components for motion, view-angle and structure of a person. After receiving latent representations, we then combine motion with different skeletons and decode a retargeted temporal sequence. Finally, using video to video synthesis retargeted skeleton is transformed to retargeted video of an avatar.
dc.publisherDhirubhai Ambani Institute of Information and Communication Technology
dc.subjectMotion retargeting
dc.subjectHuman pose estimation
dc.subjectAuto-encoder
dc.subjectVideo to video synthesis (vid2vid)
dc.classification.ddc006.6 KAP
dc.titleAvatar retargeting
dc.typeDissertation
dc.degreeM. Tech
dc.student.id201811052
dc.accession.numberT00880
Appears in Collections:M Tech Dissertations

Files in This Item:
File Description SizeFormat 
201811052.pdf
  Restricted Access
342.86 kBAdobe PDFView/Open Request a copy


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.