Fast Face Detection in Violent Video Scenes

V. E. Machaca Arceda; K. M. Fernández Fabián; P. C. Laguna Laura; J. J. Rivera Tito; J. C. Gutiérrez Cáceres

doi:10.1016/j.entcs.2016.12.002

Fast Face Detection in Violent Video Scenes

V. E. Machaca Arceda, K. M. Fernández Fabián, P. C. Laguna Laura, J. J. Rivera Tito, J. C. Gutiérrez Cáceres

Research output: Contribution to journal › Article › peer-review

12 Scopus citations

Abstract

In this work we aim to detect faces in violence scenes, in order to help the security control. We used the Violent Flow (ViF) descriptor with Horn-Schunck proposed in [V. Machaca Arceda and K. Fernańdez Fabián and J.C. Gutiérrez, Real Time Violence Detection in Video, “IET Conference Proceedings”, Institution of Engineering and Technology. (2016)] for violence scenes detection at first stage. Then we applied the non-adaptive interpolation super resolution algorithm to improve the video quality and finally we fire a Kanade-Lucas-Tomasi (KLT) face detector. In order to get a very low time processing, we paralleled the super resolution and face detector algorithms with CUDA. For the experiments we used the Boss Dataset and also we built a violence dataset, taking scenes from surveillance cameras. We have promising results detecting faces in this environment, because of the benefits of our proposal.

Original language	English
Pages (from-to)	5-26
Number of pages	22
Journal	Electronic Notes in Theoretical Computer Science
Volume	329
DOIs	https://doi.org/10.1016/j.entcs.2016.12.002
State	Published - 9 Dec 2016

Bibliographical note

Publisher Copyright:
© 2016 The Author(s)

Keywords

Face detection
GPU parallel computing
Horn-Schunck
Lower computational time
Optic Flow
Resolution enhancement
Super-Resolution
ViF
Video

Access to Document

10.1016/j.entcs.2016.12.002

Cite this

@article{1b688aba359a466a96a66e16c4ad90f4,

title = "Fast Face Detection in Violent Video Scenes",

abstract = "In this work we aim to detect faces in violence scenes, in order to help the security control. We used the Violent Flow (ViF) descriptor with Horn-Schunck proposed in [V. Machaca Arceda and K. Ferna{\'n}dez Fabi{\'a}n and J.C. Guti{\'e}rrez, Real Time Violence Detection in Video, “IET Conference Proceedings”, Institution of Engineering and Technology. (2016)] for violence scenes detection at first stage. Then we applied the non-adaptive interpolation super resolution algorithm to improve the video quality and finally we fire a Kanade-Lucas-Tomasi (KLT) face detector. In order to get a very low time processing, we paralleled the super resolution and face detector algorithms with CUDA. For the experiments we used the Boss Dataset and also we built a violence dataset, taking scenes from surveillance cameras. We have promising results detecting faces in this environment, because of the benefits of our proposal.",

keywords = "Face detection, GPU parallel computing, Horn-Schunck, Lower computational time, Optic Flow, Resolution enhancement, Super-Resolution, ViF, Video",

author = "{Machaca Arceda}, {V. E.} and {Fern{\'a}ndez Fabi{\'a}n}, {K. M.} and {Laguna Laura}, {P. C.} and {Rivera Tito}, {J. J.} and {Guti{\'e}rrez C{\'a}ceres}, {J. C.}",

note = "Publisher Copyright: {\textcopyright} 2016 The Author(s)",

year = "2016",

month = dec,

day = "9",

doi = "10.1016/j.entcs.2016.12.002",

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

volume = "329",

pages = "5--26",

journal = "Electronic Notes in Theoretical Computer Science",

issn = "1571-0661",

publisher = "Elsevier",

}

TY - JOUR

T1 - Fast Face Detection in Violent Video Scenes

AU - Machaca Arceda, V. E.

AU - Fernández Fabián, K. M.

AU - Laguna Laura, P. C.

AU - Rivera Tito, J. J.

AU - Gutiérrez Cáceres, J. C.

PY - 2016/12/9

Y1 - 2016/12/9

N2 - In this work we aim to detect faces in violence scenes, in order to help the security control. We used the Violent Flow (ViF) descriptor with Horn-Schunck proposed in [V. Machaca Arceda and K. Fernańdez Fabián and J.C. Gutiérrez, Real Time Violence Detection in Video, “IET Conference Proceedings”, Institution of Engineering and Technology. (2016)] for violence scenes detection at first stage. Then we applied the non-adaptive interpolation super resolution algorithm to improve the video quality and finally we fire a Kanade-Lucas-Tomasi (KLT) face detector. In order to get a very low time processing, we paralleled the super resolution and face detector algorithms with CUDA. For the experiments we used the Boss Dataset and also we built a violence dataset, taking scenes from surveillance cameras. We have promising results detecting faces in this environment, because of the benefits of our proposal.

AB - In this work we aim to detect faces in violence scenes, in order to help the security control. We used the Violent Flow (ViF) descriptor with Horn-Schunck proposed in [V. Machaca Arceda and K. Fernańdez Fabián and J.C. Gutiérrez, Real Time Violence Detection in Video, “IET Conference Proceedings”, Institution of Engineering and Technology. (2016)] for violence scenes detection at first stage. Then we applied the non-adaptive interpolation super resolution algorithm to improve the video quality and finally we fire a Kanade-Lucas-Tomasi (KLT) face detector. In order to get a very low time processing, we paralleled the super resolution and face detector algorithms with CUDA. For the experiments we used the Boss Dataset and also we built a violence dataset, taking scenes from surveillance cameras. We have promising results detecting faces in this environment, because of the benefits of our proposal.

KW - Face detection

KW - GPU parallel computing

KW - Horn-Schunck

KW - Lower computational time

KW - Optic Flow

KW - Resolution enhancement

KW - Super-Resolution

KW - ViF

KW - Video

UR - http://www.scopus.com/inward/record.url?scp=85003823257&partnerID=8YFLogxK

U2 - 10.1016/j.entcs.2016.12.002

DO - 10.1016/j.entcs.2016.12.002

M3 - Artículo

AN - SCOPUS:85003823257

VL - 329

SP - 5

EP - 26

JO - Electronic Notes in Theoretical Computer Science

JF - Electronic Notes in Theoretical Computer Science

SN - 1571-0661

ER -

Fast Face Detection in Violent Video Scenes

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this