3D human pose search using oriented cylinders

Simple item page
dc.citation.epage22en_US
dc.citation.spage16en_US
dc.contributor.authorPehlivan, Selenen_US
dc.contributor.authorDuygulu, Pınaren_US
dc.coverage.spatialKyoto, Japan
dc.date.accessioned2016-02-08T12:26:33Z
dc.date.available2016-02-08T12:26:33Z
dc.date.issued2009-09-10en_US
dc.departmentDepartment of Computer Engineeringen_US
dc.descriptionDate of Conference: 27 Sept.-4 Oct. 2009
dc.descriptionConference name: IEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops 2009
dc.description.abstractIn this study, we present a representation based on a new 3D search technique for volumetric human poses which is then used to recognize actions in three dimensional video sequences. We generate a set of cylinder like 3D kernels in various sizes and orientations. These kernels are searched over 3D volumes to find high response regions. The distribution of these responses are then used to represent a 3D pose. We use the proposed representation for (i) pose retrieval using Nearest Neighbor (NN) based classification and Support Vector Machine (SVM) based classification methods, and for (ii) action recognition on a set of actions using Dynamic Time Warping (DTW) and Hidden Markov Model (HMM) based classification methods. Evaluations on IXMAS dataset supports the effectiveness of such a robust pose representation. ©2009 IEEE.en_US
dc.identifier.doi10.1109/ICCVW.2009.5457722en_US
dc.identifier.urihttp://hdl.handle.net/11693/28664
dc.language.isoEnglishen_US
dc.publisherIEEE
dc.relation.isversionofhttp://dx.doi.org/10.1109/ICCVW.2009.5457722en_US
dc.source.titleIEEE 12th International Conference on Computer Vision Workshops, ICCV Workshops 2009en_US
dc.subjectAction recognitionen_US
dc.subjectClassification methodsen_US
dc.subjectData setsen_US
dc.subjectDynamic time warpingen_US
dc.subjectHigh responseen_US
dc.subjectHuman poseen_US
dc.subjectNearest neighborsen_US
dc.subjectSearch techniqueen_US
dc.subjectVideo sequencesen_US
dc.subjectComputer visionen_US
dc.subjectCylinders (shapes)en_US
dc.subjectHidden Markov modelsen_US
dc.subjectVideo recordingen_US
dc.subjectThree dimensionalen_US
dc.title3D human pose search using oriented cylindersen_US
dc.typeConference Paperen_US

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
3D human pose search using oriented cylinders.pdf
Size:
376.53 KB
Format:
Adobe Portable Document Format
Description:
Full printable version
Download

Collections

Department of Computer Engineering