Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/55348
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Zhou, H. | en |
dc.contributor.author | Wang, L. | en |
dc.contributor.author | Suter, D. | en |
dc.date.issued | 2008 | en |
dc.identifier.citation | Proceedings of the 19th International Conference on Pattern Recognition (ICPR), Tampa, Florida, USA., 2008: pp.1-4 | en |
dc.identifier.isbn | 9781424421749 | en |
dc.identifier.issn | 1051-4651 | en |
dc.identifier.uri | http://hdl.handle.net/2440/55348 | - |
dc.description.abstract | This paper investigates the applicability of Gaussian processes (GP) classification for recognition of articulated and deformable human motions from image sequences. Using tensor subspace analysis (TSA), space-time human silhouettes (extracted from motion videos) are transformed to low-dimensional multivariate time series, based on which structure-based statistical features are calculated to summarize the motion properties. GP classification is then used to learn and predict motion categories. Experimental results on two real-world state-of-the-art datasets show that the proposed approach is effective, and outperforms support vector machine (SVM). | en |
dc.description.statementofresponsibility | Hang Zhou, Liang Wang and David Suter | en |
dc.description.uri | http://dx.doi.org/10.1109/ICPR.2008.4761140 | en |
dc.language.iso | en | en |
dc.publisher | IEEE | en |
dc.relation.ispartofseries | International Conference on Pattern Recognition | en |
dc.title | Human motion recognition using gaussian processes classification | en |
dc.type | Conference paper | en |
dc.contributor.conference | International Conference on Pattern Recognition (19th : 2008 : Florida) | en |
dc.publisher.place | Online | en |
pubs.publication-status | Published | en |
dc.identifier.orcid | Suter, D. [0000-0001-6306-3023] | en |
Appears in Collections: | Aurora harvest 5 Computer Science publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.