Real time violence detection in video

Vicente Machaca Arceda; Karla Fernández Fabián; Juan Carlos Gutíerrez

Real time violence detection in video

Vicente Machaca Arceda, Karla Fernández Fabián, Juan Carlos Gutíerrez

Research output: Contribution to conference › Paper › peer-review

1 Scopus citations

Abstract

We propose a Violent Flow (ViF) variation using Horn-Schunck instead of Iterative Reweighted Least Squares (IRLS) as optical flow algorithm and Support Vector Machine (SVM) as a classifier, we evaluated the performance in different datasets. In Hockey dataset Horn-Schunck had a better performance, but in Movies and Crowded datasets IRLS was better. We also evaluated the computational cost processing two frames, Horn-Schunck had a very low cost of 0.25 seconds against 7.80 seconds of IRLS enabling its use in real time.

Original language	English
State	Published - 2016
Event	International Conference on Pattern Recognition Systems, ICPRS 2016 - Talca, Chile Duration: 20 Apr 2016 → 22 Apr 2016

Conference

Conference	International Conference on Pattern Recognition Systems, ICPRS 2016
Country/Territory	Chile
City	Talca
Period	20/04/16 → 22/04/16

Keywords

MoSIFT
SIFT
STIP
ViF

Cite this

@conference{201f2125469f4c35a6708afdebc89b60,

title = "Real time violence detection in video",

abstract = "We propose a Violent Flow (ViF) variation using Horn-Schunck instead of Iterative Reweighted Least Squares (IRLS) as optical flow algorithm and Support Vector Machine (SVM) as a classifier, we evaluated the performance in different datasets. In Hockey dataset Horn-Schunck had a better performance, but in Movies and Crowded datasets IRLS was better. We also evaluated the computational cost processing two frames, Horn-Schunck had a very low cost of 0.25 seconds against 7.80 seconds of IRLS enabling its use in real time.",

keywords = "MoSIFT, SIFT, STIP, ViF",

author = "Arceda, {Vicente Machaca} and Fabi{\'a}n, {Karla Fern{\'a}ndez} and Gut{\'i}errez, {Juan Carlos}",

year = "2016",

language = "Ingl{\'e}s",

note = "null ; Conference date: 20-04-2016 Through 22-04-2016",

}

TY - CONF

T1 - Real time violence detection in video

AU - Arceda, Vicente Machaca

AU - Fabián, Karla Fernández

AU - Gutíerrez, Juan Carlos

PY - 2016

Y1 - 2016

N2 - We propose a Violent Flow (ViF) variation using Horn-Schunck instead of Iterative Reweighted Least Squares (IRLS) as optical flow algorithm and Support Vector Machine (SVM) as a classifier, we evaluated the performance in different datasets. In Hockey dataset Horn-Schunck had a better performance, but in Movies and Crowded datasets IRLS was better. We also evaluated the computational cost processing two frames, Horn-Schunck had a very low cost of 0.25 seconds against 7.80 seconds of IRLS enabling its use in real time.

AB - We propose a Violent Flow (ViF) variation using Horn-Schunck instead of Iterative Reweighted Least Squares (IRLS) as optical flow algorithm and Support Vector Machine (SVM) as a classifier, we evaluated the performance in different datasets. In Hockey dataset Horn-Schunck had a better performance, but in Movies and Crowded datasets IRLS was better. We also evaluated the computational cost processing two frames, Horn-Schunck had a very low cost of 0.25 seconds against 7.80 seconds of IRLS enabling its use in real time.

KW - MoSIFT

KW - SIFT

KW - STIP

KW - ViF

UR - http://www.scopus.com/inward/record.url?scp=85006856446&partnerID=8YFLogxK

M3 - Artículo

AN - SCOPUS:85006856446

Y2 - 20 April 2016 through 22 April 2016

ER -

Real time violence detection in video

Abstract

Conference

Keywords

Other files and links

Fingerprint

Cite this