Video characterization based on activity clustering

Nikolaos Kourous, Alexandros Iosifidis, Anastasios Tefas, Nikos Nikolaidis, Ioannis Pitas

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

189 Downloads (Pure)

Abstract

In this paper, we propose an efficient method for video characterization based on activity information. We employ a state-of-the-art video representation in order to learn human activity concepts, i.e., video groups formed by videos depicting
similar human activities. In order to exploit the enriched visual information that is available in multi-view settings, we propose the use of the circular shift invariance property of the coefficients of the Discrete Fourier Transform (DFT) that leads to a view-independent multi-view action representation. In the test phase, in order to assign a test video to one (or multiple) activity groups, we perform temporal video segmentation in order to determine shorter videos depicting simple actions. Experimental results on the i3DPost multi-view action database and a new multi-view action database denote the effectiveness of the proposed approach.
Original languageEnglish
Title of host publication2014 International Conference on Electrical and Computer Engineering (ICECE 2014)
Subtitle of host publicationProceedings of a meeting held 20-22 December 2014, Dhaka, Bangladesh
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages266-269
Number of pages4
ISBN (Electronic)9781479941674
ISBN (Print)9781479941650
DOIs
Publication statusPublished - Mar 2015
EventInternational Conference on Electrical and Computer Engineering (ICECE) - Dhaka, Bangladesh
Duration: 20 Dec 201422 Dec 2014

Conference

ConferenceInternational Conference on Electrical and Computer Engineering (ICECE)
CountryBangladesh
CityDhaka
Period20/12/1422/12/14

Keywords

  • Video characterization
  • Activity clustering
  • Multi-camera setup

Fingerprint

Dive into the research topics of 'Video characterization based on activity clustering'. Together they form a unique fingerprint.

Cite this