Action Recognition by Multiple Features and Hyper-sphere Multi-class SVM

UTSePress Research/Manakin Repository

Search UTSePress Research


Advanced Search

Browse

My Account

Show simple item record

dc.contributor.author Liu, Jia en_US
dc.contributor.author Yang, Jie en_US
dc.contributor.author Zhang, Yi en_US
dc.contributor.author He, Sean en_US
dc.contributor.editor Mejdat Aetin, Kim Boyer and Seong-Whan Lee - ICPR 2010 Technical Program Chairs en_US
dc.date.accessioned 2012-02-02T11:08:07Z
dc.date.available 2012-02-02T11:08:07Z
dc.date.issued 2010 en_US
dc.identifier 2009007669 en_US
dc.identifier.citation Liu Jia et al. 2010, 'Action Recognition by Multiple Features and Hyper-sphere Multi-class SVM', , IEEE Computer Society, Istanbul Turkey, , pp. 3744-3747. en_US
dc.identifier.issn 978-1-4244-7542-1 en_US
dc.identifier.other E1 en_US
dc.identifier.uri http://hdl.handle.net/10453/16277
dc.description.abstract In this paper we propose a novel framework for action recognition based on multiple features for improve action recognition in videos. The fusion of multiple features is important for recognizing actions as often a single features based representation is not enough to capture the imaging variations (view-point, illumination etc.) and attributes of individuals (size, age, gender etc). Hence, we use two kinds of features: i) a quantized vocabulary of local spatio-temporal (ST) volumes (cuboids and 2-D SIFT), and ii) the higher order statistical models of interest points, which aims to capture the global information of the actor. We construct video presentation in terms of local space time features and global features and integrate such representations with hper-sphere multi-class SVM. Experiments on publicly available datasets show that our proposed approach is effective. An additional experiment shows that using both local and global features provides a richer representation of human action when compared to the use of a single feature type. en_US
dc.language English en_US
dc.publisher IEEE Computer Society en_US
dc.relation.isbasedon http://dx.doi.org/10.1109/ICPR.2010.912 en_US
dc.title Action Recognition by Multiple Features and Hyper-sphere Multi-class SVM en_US
dc.parent Proceedings: 2010 20th International Conference Pattern Recognition (ICPR 2010) en_US
dc.journal.volume en_US
dc.journal.number en_US
dc.publocation Istanbul Turkey en_US
dc.identifier.startpage 3744 en_US
dc.identifier.endpage 3747 en_US
dc.cauo.name FEIT.School of Computing and Communications en_US
dc.conference Verified OK en_US
dc.for 080100 en_US
dc.personcode 0000063898 en_US
dc.personcode 0000027379 en_US
dc.personcode 0000063899 en_US
dc.personcode 990421 en_US
dc.percentage 100 en_US
dc.classification.name Artificial Intelligence and Image Processing en_US
dc.classification.type FOR-08 en_US
dc.edition en_US
dc.custom International Conference Pattern Recognition en_US
dc.date.activity 20100823 en_US
dc.location.activity Istanbul Turkey en_US
dc.description.keywords human action recognition; multiple features; hyper-sphere Multi-class SVM en_US
dc.staffid 990421 en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record