File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Fast neural style transfer for motion data

TitleFast neural style transfer for motion data
Authors
Holden, DanielHabibie, IkhsanulKusajima, IkuoKomura, Taku
Keywordsdeep learning
motion capture
style transfer
machine learning
computer graphics
Issue Date2017
Citation
IEEE Computer Graphics and Applications, 2017, v. 37, n. 4, p. 42-49 How to Cite?
DOI: http://dx.doi.org/10.1109/MCG.2017.3271464
Abstract© 1981-2012 IEEE. Automating motion style transfer can help save animators time by allowing them to produce a single set of motions, which can then be automatically adapted for use with different characters. The proposed fast, efficient technique for performing neural style transfer of human motion data uses a feed-forward neural network trained on a large motion database. The proposed framework can transform the style of motion thousands of times faster than previous approaches that use optimization.
Persistent Identifierhttp://hdl.handle.net/10722/288877
ISSN
0272-1716
2021 Impact Factor: 1.909
2020 SCImago Journal Rankings: 0.349
ISI Accession Number ID
WOS:000411626600007

 

DC FieldValueLanguage
dc.contributor.authorHolden, Daniel-
dc.contributor.authorHabibie, Ikhsanul-
dc.contributor.authorKusajima, Ikuo-
dc.contributor.authorKomura, Taku-
dc.date.accessioned2020-10-12T08:06:06Z-
dc.date.available2020-10-12T08:06:06Z-
dc.date.issued2017-
dc.identifier.citationIEEE Computer Graphics and Applications, 2017, v. 37, n. 4, p. 42-49-
dc.identifier.issn0272-1716-
dc.identifier.urihttp://hdl.handle.net/10722/288877-
dc.description.abstract© 1981-2012 IEEE. Automating motion style transfer can help save animators time by allowing them to produce a single set of motions, which can then be automatically adapted for use with different characters. The proposed fast, efficient technique for performing neural style transfer of human motion data uses a feed-forward neural network trained on a large motion database. The proposed framework can transform the style of motion thousands of times faster than previous approaches that use optimization.-
dc.languageeng-
dc.relation.ispartofIEEE Computer Graphics and Applications-
dc.subjectdeep learning-
dc.subjectmotion capture-
dc.subjectstyle transfer-
dc.subjectmachine learning-
dc.subjectcomputer graphics-
dc.titleFast neural style transfer for motion data-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/MCG.2017.3271464-
dc.identifier.pmid28829292-
dc.identifier.scopuseid_2-s2.0-85028884994-
dc.identifier.volume37-
dc.identifier.issue4-
dc.identifier.spage42-
dc.identifier.epage49-
dc.identifier.isiWOS:000411626600007-
dc.identifier.issnl0272-1716-

Export via OAI-PMH Interface in XML Formats

OR

Export to Other Non-XML Formats