Action Recognition Using Discriminative Structured Trajectory Groups

Handle URI:
http://hdl.handle.net/10754/556158
Title:
Action Recognition Using Discriminative Structured Trajectory Groups
Authors:
Atmosukarto, Indriyati; Ahuja, Narendra; Ghanem, Bernard ( 0000-0002-5534-587X )
Abstract:
In this paper, we develop a novel framework for action recognition in videos. The framework is based on automatically learning the discriminative trajectory groups that are relevant to an action. Different from previous approaches, our method does not require complex computation for graph matching or complex latent models to localize the parts. We model a video as a structured bag of trajectory groups with latent class variables. We model action recognition problem in a weakly supervised setting and learn discriminative trajectory groups by employing multiple instance learning (MIL) based Support Vector Machine (SVM) using pre-computed kernels. The kernels depend on the spatio-temporal relationship between the extracted trajectory groups and their associated features. We demonstrate both quantitatively and qualitatively that the classification performance of our proposed method is superior to baselines and several state-of-the-art approaches on three challenging standard benchmark datasets.
KAUST Department:
Image and Video Understanding Lab
Citation:
Indriyati Atmosukarto, Narendra Ahuja, Bernard Ghanem "Action Recognition using Discriminative Structured Trajectory Groups" Winter Conference on Applications of Computer Vision (WACV 2015)​​​
Publisher:
IEEE
Journal:
Applications of Computer Vision (WACV), 2015 IEEE Winter Conference on
Conference/Event name:
2015 15th IEEE Winter Conference on Applications of Computer Vision, WACV 2015
Issue Date:
6-Jan-2015
DOI:
10.1109/WACV.2015.124
Type:
Conference Paper
Additional Links:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7045978; http://vcc.kaust.edu.sa/Documents/B.%20Ghanem/papers/action_recognition_wacv2015.pdf
Appears in Collections:
Conference Papers

Full metadata record

DC FieldValue Language
dc.contributor.authorAtmosukarto, Indriyatien
dc.contributor.authorAhuja, Narendraen
dc.contributor.authorGhanem, Bernarden
dc.date.accessioned2015-06-02T13:38:49Zen
dc.date.available2015-06-02T13:38:49Zen
dc.date.issued2015-01-06en
dc.identifier.citationIndriyati Atmosukarto, Narendra Ahuja, Bernard Ghanem "Action Recognition using Discriminative Structured Trajectory Groups" Winter Conference on Applications of Computer Vision (WACV 2015)​​​en
dc.identifier.doi10.1109/WACV.2015.124en
dc.identifier.urihttp://hdl.handle.net/10754/556158en
dc.description.abstractIn this paper, we develop a novel framework for action recognition in videos. The framework is based on automatically learning the discriminative trajectory groups that are relevant to an action. Different from previous approaches, our method does not require complex computation for graph matching or complex latent models to localize the parts. We model a video as a structured bag of trajectory groups with latent class variables. We model action recognition problem in a weakly supervised setting and learn discriminative trajectory groups by employing multiple instance learning (MIL) based Support Vector Machine (SVM) using pre-computed kernels. The kernels depend on the spatio-temporal relationship between the extracted trajectory groups and their associated features. We demonstrate both quantitatively and qualitatively that the classification performance of our proposed method is superior to baselines and several state-of-the-art approaches on three challenging standard benchmark datasets.en
dc.publisherIEEEen
dc.relation.urlhttp://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7045978en
dc.relation.urlhttp://vcc.kaust.edu.sa/Documents/B.%20Ghanem/papers/action_recognition_wacv2015.pdfen
dc.rights(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.en
dc.subjectAction Recognitionen
dc.titleAction Recognition Using Discriminative Structured Trajectory Groupsen
dc.typeConference Paperen
dc.contributor.departmentImage and Video Understanding Laben
dc.identifier.journalApplications of Computer Vision (WACV), 2015 IEEE Winter Conference onen
dc.conference.date5 January 2015 through 9 January 2015en
dc.conference.name2015 15th IEEE Winter Conference on Applications of Computer Vision, WACV 2015en
dc.eprint.versionPost-printen
dc.contributor.institutionAdvanced Digital Sciences Center of Illinois, Singaporeen
dc.contributor.institutionUniversity of Illinois at Urbana Champaign, USAen
kaust.authorGhanem, Bernarden
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.