Recognizing actions from still images

dc.contributor.authorİkizler, Nazlıen_US
dc.contributor.authorCinbiş, R .Gökberken_US
dc.contributor.authorPehlivan, Selenen_US
dc.contributor.authorDuygulu, Pınaren_US
dc.coverage.spatialTampa, FL, USA
dc.date.accessioned2016-02-08T11:35:55Z
dc.date.available2016-02-08T11:35:55Z
dc.date.issued2008-12en_US
dc.departmentDepartment of Computer Engineeringen_US
dc.descriptionConference name: 19th International Conference on Pattern Recognition, 2008
dc.descriptionDate of Conference: 8-11 Dec. 2008
dc.description.abstractIn this paper, we approach the problem of under- standing human actions from still images. Our method involves representing the pose with a spatial and ori- entational histogramming of rectangular regions on a parse probability map. We use LDA to obtain a more compact and discriminative feature representation and binary SVMs for classification. Our results over a new dataset collected for this problem show that by using a rectangle histogramming approach, we can discriminate actions to a great extent. We also show how we can use this approach in an unsupervised setting. To our best knowledge, this is one of the first studies that try to recognize actions within still images. © 2008 IEEE.en_US
dc.description.provenanceMade available in DSpace on 2016-02-08T11:35:55Z (GMT). No. of bitstreams: 1 bilkent-research-paper.pdf: 70227 bytes, checksum: 26e812c6f5156f83f0e77b261a471b5a (MD5) Previous issue date: 2008en
dc.identifier.doi10.1109/ICPR.2008.4761663en_US
dc.identifier.urihttp://hdl.handle.net/11693/26790en_US
dc.language.isoEnglishen_US
dc.publisherIEEEen_US
dc.relation.isversionofhttps://doi.org/10.1109/ICPR.2008.4761663
dc.source.title19th International Conference on Pattern Recognition, 2008en_US
dc.subjectData setsen_US
dc.subjectDiscriminative featuresen_US
dc.subjectHistogrammingen_US
dc.subjectHuman actionsen_US
dc.subjectProbability mapsen_US
dc.subjectRectangular regionsen_US
dc.subjectStill imagesen_US
dc.subjectPattern recognitionen_US
dc.titleRecognizing actions from still imagesen_US
dc.typeConference Paperen_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Recognizing actions from still images.pdf
Size:
649.08 KB
Format:
Adobe Portable Document Format
Description:
Full printable version