Show simple item record

dc.contributor.authorŞener, Fadimeen_US
dc.contributor.authorBas, C.en_US
dc.contributor.authorIkizler-Cinbis, N.en_US
dc.coverage.spatialFlorence, Italyen_US
dc.date.accessioned2016-02-08T12:12:29Z
dc.date.available2016-02-08T12:12:29Z
dc.date.issued2012en_US
dc.identifier.issn0302-9743
dc.identifier.urihttp://hdl.handle.net/11693/28150
dc.descriptionDate of Conference: October 7-13, 2012en_US
dc.descriptionConference name: Conference name: ECCV: European Conference on Computer Vision ECCV 2012. Workshops and Demonstrationsen_US
dc.description.abstractWe propose a multi-cue based approach for recognizing human actions in still images, where relevant object regions are discovered and utilized in a weakly supervised manner. Our approach does not require any explicitly trained object detector or part/attribute annotation. Instead, a multiple instance learning approach is used over sets of object hypotheses in order to represent objects relevant to the actions. We test our method on the extensive Stanford 40 Actions dataset [1] and achieve significant performance gain compared to the state-of-the-art. Our results show that using multiple object hypotheses within multiple instance learning is effective for human action recognition in still images and such an object representation is suitable for using in conjunction with other visual features. © 2012 Springer-Verlag.en_US
dc.language.isoEnglishen_US
dc.source.titleComputer Vision – ECCV 2012. Workshops and Demonstrationsen_US
dc.relation.isversionofhttp://dx.doi.org/10.1007/978-3-642-33885-4_27en_US
dc.relation.isversionofhttps://doi.org/10.1007/978-3-642-33885-4en_US
dc.subjectData setsen_US
dc.subjectHuman actionsen_US
dc.subjectHuman-action recognitionen_US
dc.subjectMultiple featuresen_US
dc.subjectMultiple instance learningen_US
dc.subjectMultiple objectsen_US
dc.subjectObject detectorsen_US
dc.subjectObject regionen_US
dc.subjectObject representationsen_US
dc.subjectPerformance Gainen_US
dc.subjectStanforden_US
dc.subjectStill imagesen_US
dc.subjectVisual featureen_US
dc.subjectGesture recognitionen_US
dc.subjectLearning systemsen_US
dc.subjectStatistical testsen_US
dc.subjectComputer visionen_US
dc.titleOn recognizing actions in still images via multiple featuresen_US
dc.typeConference Paperen_US
dc.departmentDepartment of Computer Engineeringen_US
dc.citation.spage263en_US
dc.citation.epage272en_US
dc.citation.volumeNumber3en_US
dc.identifier.doi10.1007/978-3-642-33885-4_27en_US
dc.identifier.doi10.1007/978-3-642-33885-4en_US
dc.publisherSpringer, Berlin, Heidelbergen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record