Image-based object reconstruction using run-length representation

Cited 1 time in webofscience Cited 0 time in scopus
  • Hit : 208
  • Download : 0
DC FieldValueLanguage
dc.contributor.authorHwang, Sung Sooko
dc.contributor.authorKim, Hee Dongko
dc.contributor.authorJang, Tae Youngko
dc.contributor.authorYoo, Jisungko
dc.contributor.authorKim, Sujungko
dc.contributor.authorPaeng, Kyung Hyunko
dc.contributor.authorKim, Seong Daeko
dc.date.accessioned2017-03-28T06:51:28Z-
dc.date.available2017-03-28T06:51:28Z-
dc.date.created2016-11-23-
dc.date.created2016-11-23-
dc.date.issued2017-02-
dc.identifier.citationSIGNAL PROCESSING-IMAGE COMMUNICATION, v.51, pp.1 - 12-
dc.identifier.issn0923-5965-
dc.identifier.urihttp://hdl.handle.net/10203/220881-
dc.description.abstractThis paper presents an image-based object reconstruction with a low memory footprint using run-length representation. While conventional volume-based approaches, which utilize voxels as primitives, are intuitive and easy to manipulate 3D data, they require a large amount of memory and computation during the reconstruction process. To overcome these burdens, this paper uses 3D runs to represent a 3D object and reconstructs each 3D run from multi-view silhouettes with a small amount of memory. The proposed geometry reconstruction is also computationally inexpensive, as it processes multiple voxels simultaneously. And for the compatibility with the conventional data formats, generation of polygonal 3D meshes from the reconstructed 3D runs is proposed as well. Lastly, texture mapping is proposed to additionally reduce the amount of memory for object reconstruction. The proposed reconstruction scheme has been simulated using various types of multi view datasets. The results show that the proposed method performs object reconstruction with a smaller amount of memory and computation than voxel-based approaches.-
dc.languageEnglish-
dc.publisherELSEVIER SCIENCE BV-
dc.subjectFREE-VIEWPOINT VIDEO-
dc.subjectREAL-TIME-
dc.subject3D VIDEO-
dc.subjectSHAPE RECONSTRUCTION-
dc.subjectVIRTUALIZED REALITY-
dc.subjectHUMAN ACTORS-
dc.subjectCAPTURE-
dc.subjectTRACKING-
dc.subjectCAMERAS-
dc.titleImage-based object reconstruction using run-length representation-
dc.typeArticle-
dc.identifier.wosid000392773300001-
dc.identifier.scopusid2-s2.0-84996590190-
dc.type.rimsART-
dc.citation.volume51-
dc.citation.beginningpage1-
dc.citation.endingpage12-
dc.citation.publicationnameSIGNAL PROCESSING-IMAGE COMMUNICATION-
dc.identifier.doi10.1016/j.image.2016.11.002-
dc.contributor.localauthorKim, Seong Dae-
dc.description.isOpenAccessN-
dc.type.journalArticleArticle-
dc.subject.keywordAuthor3D object reconstruction-
dc.subject.keywordAuthorMulti-view images-
dc.subject.keywordAuthorRun-length representation-
dc.subject.keywordAuthorVisual hull computation-
dc.subject.keywordAuthorMesh generation-
dc.subject.keywordAuthorTexture mapping-
dc.subject.keywordPlusFREE-VIEWPOINT VIDEO-
dc.subject.keywordPlusREAL-TIME-
dc.subject.keywordPlus3D VIDEO-
dc.subject.keywordPlusSHAPE RECONSTRUCTION-
dc.subject.keywordPlusVIRTUALIZED REALITY-
dc.subject.keywordPlusHUMAN ACTORS-
dc.subject.keywordPlusCAPTURE-
dc.subject.keywordPlusTRACKING-
dc.subject.keywordPlusCAMERAS-
Appears in Collection
EE-Journal Papers(저널논문)
Files in This Item
There are no files associated with this item.
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 1 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0