On recognizing actions in still images via multiple features
dc.citation.epage | 272 | en_US |
dc.citation.spage | 263 | en_US |
dc.citation.volumeNumber | 3 | en_US |
dc.contributor.author | Şener, Fadime | en_US |
dc.contributor.author | Bas, C. | en_US |
dc.contributor.author | Ikizler-Cinbis, N. | en_US |
dc.coverage.spatial | Florence, Italy | en_US |
dc.date.accessioned | 2016-02-08T12:12:29Z | |
dc.date.available | 2016-02-08T12:12:29Z | |
dc.date.issued | 2012 | en_US |
dc.department | Department of Computer Engineering | en_US |
dc.description | Date of Conference: October 7-13, 2012 | en_US |
dc.description | Conference name: Conference name: ECCV: European Conference on Computer Vision ECCV 2012. Workshops and Demonstrations | en_US |
dc.description.abstract | We 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.description.provenance | Made available in DSpace on 2016-02-08T12:12:29Z (GMT). No. of bitstreams: 1 bilkent-research-paper.pdf: 70227 bytes, checksum: 26e812c6f5156f83f0e77b261a471b5a (MD5) Previous issue date: 2012 | en |
dc.identifier.doi | 10.1007/978-3-642-33885-4_27 | en_US |
dc.identifier.doi | 10.1007/978-3-642-33885-4 | en_US |
dc.identifier.issn | 0302-9743 | en_US |
dc.identifier.uri | http://hdl.handle.net/11693/28150 | en_US |
dc.language.iso | English | en_US |
dc.publisher | Springer, Berlin, Heidelberg | en_US |
dc.relation.isversionof | http://dx.doi.org/10.1007/978-3-642-33885-4_27 | en_US |
dc.relation.isversionof | https://doi.org/10.1007/978-3-642-33885-4 | en_US |
dc.source.title | Computer Vision – ECCV 2012. Workshops and Demonstrations | en_US |
dc.subject | Data sets | en_US |
dc.subject | Human actions | en_US |
dc.subject | Human-action recognition | en_US |
dc.subject | Multiple features | en_US |
dc.subject | Multiple instance learning | en_US |
dc.subject | Multiple objects | en_US |
dc.subject | Object detectors | en_US |
dc.subject | Object region | en_US |
dc.subject | Object representations | en_US |
dc.subject | Performance Gain | en_US |
dc.subject | Stanford | en_US |
dc.subject | Still images | en_US |
dc.subject | Visual feature | en_US |
dc.subject | Gesture recognition | en_US |
dc.subject | Learning systems | en_US |
dc.subject | Statistical tests | en_US |
dc.subject | Computer vision | en_US |
dc.title | On recognizing actions in still images via multiple features | en_US |
dc.type | Conference Paper | en_US |
Files
Original bundle
1 - 1 of 1
Loading...
- Name:
- On recognizing actions in still images via multiple features.pdf
- Size:
- 1019.18 KB
- Format:
- Adobe Portable Document Format
- Description:
- Full printable version