Action recognition using bag of features extracted from a beam of trajectories

Thanh Phuong Nguyen, Antoine Manzanera

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

A new spatio temporal descriptor is proposed for action recognition. The action is modelled from a beam of trajectories obtained using semi dense point tracking on the video sequence. We detect the dominant points of these trajectories as points of local extremum curvature and extract their corresponding feature vectors, to form a dictionary of atomic action elements. The high density of these informative and invariant elements allows effective statistical action description. Then, human action recognition is performed using a bag of feature model with SVM classifier. Experimentations show promising results on several well-known datasets.

Original languageEnglish
Title of host publication2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings
PublisherIEEE Computer Society
Pages4354-4357
Number of pages4
ISBN (Print)9781479923410
DOIs
Publication statusPublished - 1 Jan 2013
Event2013 20th IEEE International Conference on Image Processing, ICIP 2013 - Melbourne, VIC, Australia
Duration: 15 Sept 201318 Sept 2013

Publication series

Name2013 IEEE International Conference on Image Processing, ICIP 2013 - Proceedings

Conference

Conference2013 20th IEEE International Conference on Image Processing, ICIP 2013
Country/TerritoryAustralia
CityMelbourne, VIC
Period15/09/1318/09/13

Keywords

  • action recognition
  • bag of features
  • dominant point
  • semi dense point tracking
  • spatio temporal feature
  • trajectory beam

Fingerprint

Dive into the research topics of 'Action recognition using bag of features extracted from a beam of trajectories'. Together they form a unique fingerprint.

Cite this