Video characterization based on activity clustering

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

doi:10.1109/ICECE.2014.7026876

Video characterization based on activity clustering

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

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

6 Citations (Scopus)

Abstract

In this paper, we propose a method for video characterization based on activity description 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 2 multi-view action databases denote the effectiveness of the proposed approach.

Original language	English
Title of host publication	8th International Conference on Electrical and Computer Engineering: Advancing Technology for a Better Tomorrow, ICECE 2014
Publisher	IEEE
Pages	266-269
Number of pages	4
ISBN (Print)	9781479941667
DOIs	https://doi.org/10.1109/ICECE.2014.7026876
Publication status	Published - 28 Jan 2015
Publication type	A4 Article in conference proceedings
Event	8th International Conference on Electrical and Computer Engineering, ICECE 2014 - Dhaka, Bangladesh Duration: 20 Dec 2014 → 22 Dec 2014

Conference

Conference	8th International Conference on Electrical and Computer Engineering, ICECE 2014
Country/Territory	Bangladesh
City	Dhaka
Period	20/12/14 → 22/12/14

Keywords

Activity clustering
Multi-camera setup
Temporal video segmentation
Video characterization

ASJC Scopus subject areas

Control and Systems Engineering
Electrical and Electronic Engineering

Access to Document

10.1109/ICECE.2014.7026876

Cite this

@inproceedings{f8d06128a1554e1fac3f8bf829f438bf,

title = "Video characterization based on activity clustering",

abstract = "In this paper, we propose a method for video characterization based on activity description 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 2 multi-view action databases denote the effectiveness of the proposed approach.",

keywords = "Activity clustering, Multi-camera setup, Temporal video segmentation, Video characterization",

author = "Nikolaos Kourous and Alexandros Iosifidis and Anastasios Tefas and Nikos Nikolaidis and Ioannis Pitas",

year = "2015",

month = jan,

day = "28",

doi = "10.1109/ICECE.2014.7026876",

language = "English",

isbn = "9781479941667",

pages = "266--269",

booktitle = "8th International Conference on Electrical and Computer Engineering: Advancing Technology for a Better Tomorrow, ICECE 2014",

publisher = "IEEE",

address = "United States",

note = "8th International Conference on Electrical and Computer Engineering, ICECE 2014 ; Conference date: 20-12-2014 Through 22-12-2014",

}

Kourous, N, Iosifidis, A, Tefas, A, Nikolaidis, N & Pitas, I 2015, Video characterization based on activity clustering. in 8th International Conference on Electrical and Computer Engineering: Advancing Technology for a Better Tomorrow, ICECE 2014. IEEE, pp. 266-269, 8th International Conference on Electrical and Computer Engineering, ICECE 2014, Dhaka, Bangladesh, 20/12/14. https://doi.org/10.1109/ICECE.2014.7026876

Video characterization based on activity clustering. / Kourous, Nikolaos; Iosifidis, Alexandros; Tefas, Anastasios et al.
8th International Conference on Electrical and Computer Engineering: Advancing Technology for a Better Tomorrow, ICECE 2014. IEEE, 2015. p. 266-269.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Video characterization based on activity clustering

AU - Kourous, Nikolaos

AU - Iosifidis, Alexandros

AU - Tefas, Anastasios

AU - Nikolaidis, Nikos

AU - Pitas, Ioannis

PY - 2015/1/28

Y1 - 2015/1/28

N2 - In this paper, we propose a method for video characterization based on activity description 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 2 multi-view action databases denote the effectiveness of the proposed approach.

AB - In this paper, we propose a method for video characterization based on activity description 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 2 multi-view action databases denote the effectiveness of the proposed approach.

KW - Activity clustering

KW - Multi-camera setup

KW - Temporal video segmentation

KW - Video characterization

U2 - 10.1109/ICECE.2014.7026876

DO - 10.1109/ICECE.2014.7026876

M3 - Conference contribution

SN - 9781479941667

SP - 266

EP - 269

BT - 8th International Conference on Electrical and Computer Engineering: Advancing Technology for a Better Tomorrow, ICECE 2014

PB - IEEE

T2 - 8th International Conference on Electrical and Computer Engineering, ICECE 2014

Y2 - 20 December 2014 through 22 December 2014

ER -

Video characterization based on activity clustering

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this