Publication detail

Robust Real-Time Face Tracking System

PŘINOSIL, J. SMÉKAL, Z.

Original Title

Robust Real-Time Face Tracking System

English Title

Robust Real-Time Face Tracking System

Type

conference paper

Language

en

Original Abstract

This paper deals with the proposal of robust system for face tracking in real time, where several techniques are used for face detection in each video frame. At the first, the initial face position is searched using modified Viola-Jones object detector, and after that the position is continuously updated using template matching algorithm, simple Viola-Jones detector, skin color mask and CamShift algorithm. The system allows tracking faces of any size with arbitrary head movement (translation, head rotation). Furthermore, it is computationally fast, thus an eventual face processing algorithm can be easily performed.

English abstract

This paper deals with the proposal of robust system for face tracking in real time, where several techniques are used for face detection in each video frame. At the first, the initial face position is searched using modified Viola-Jones object detector, and after that the position is continuously updated using template matching algorithm, simple Viola-Jones detector, skin color mask and CamShift algorithm. The system allows tracking faces of any size with arbitrary head movement (translation, head rotation). Furthermore, it is computationally fast, thus an eventual face processing algorithm can be easily performed.

Keywords

Face tracking, Viola-Jones detector, skin color model, CamShift algorithm

RIV year

2009

Released

27.08.2009

Publisher

ASSZISZTENCIA Congress Bureau

Location

Dunakiliti

ISBN

978-963-06-7716-5

Book

32nd International Conference on Telecommunications and Signal Processing - TSP'2009

Pages from

1

Pages to

4

Pages count

4

BibTex


@inproceedings{BUT33272,
  author="Jiří {Přinosil} and Zdeněk {Smékal}",
  title="Robust Real-Time Face Tracking System",
  annote="This paper deals with the proposal of robust system for face tracking in real time, where several techniques are used for face detection in each video frame. At the first, the initial face position is searched using modified Viola-Jones object detector, and after that the position is continuously updated using template matching algorithm, simple Viola-Jones detector, skin color mask and CamShift algorithm. The system allows tracking faces of any size with arbitrary head movement (translation, head rotation). Furthermore, it is computationally fast, thus an eventual face processing algorithm can be easily performed.",
  address="ASSZISZTENCIA Congress Bureau",
  booktitle="32nd International Conference on Telecommunications and Signal Processing - TSP'2009",
  chapter="33272",
  howpublished="electronic, physical medium",
  institution="ASSZISZTENCIA Congress Bureau",
  year="2009",
  month="august",
  pages="1--4",
  publisher="ASSZISZTENCIA Congress Bureau",
  type="conference paper"
}