Robust face tracking based on region correspondence and its application for person based indexing system

Hanjin Ryu, Myunghoon Kim, Vietcuong Dinh, Seongsoo Chun, Sanghoon Sull

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

In this paper, a face tracking-approach is proposed based on templats matching for applying a video indexing application. Typically, this method is simpler and faster than other methods, but the main drawback is the poor performance in the case of face scaling changes and template drifts. To overcome these problems, the information about the facial features is incorporated into a template matching process. First, the face template is represented by two projection histograms of the face region and matching methods are used to determine the candidate face region. The matching is done this way as it is faster to match two 1-dimensional projection histograms than it is to match one 2-dimensional image. Next, the facial features are extracted from the candidate face region and a dissimilarity measure, based on the proximity of the facial features, is used to verify the face region. Finally, an anthropometric model, based on geometrical relationship between facial features, is used to refine the face region. The template to determine the face region in the next frame is dynamically updated using the refined face region. Thus, the proposed method can adapt to scale any changes with less computational cost. Experimental results are provided to demonstrate the efficiency of the proposed method.

Original languageEnglish
Pages (from-to)2861-2873
Number of pages13
JournalInternational Journal of Innovative Computing, Information and Control
Volume4
Issue number11
Publication statusPublished - 2008 Nov 1

Fingerprint

Face Tracking
Template matching
Indexing
Person
Correspondence
Face
Costs
Template
Histogram
Projection
Video Indexing
Dissimilarity Measure
Template Matching
Proximity
Computational Cost
Scaling
Model-based
Verify

Keywords

  • Face tracking
  • Projection histogram
  • Template matching
  • Video indexing

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Information Systems
  • Software
  • Theoretical Computer Science

Cite this

Robust face tracking based on region correspondence and its application for person based indexing system. / Ryu, Hanjin; Kim, Myunghoon; Dinh, Vietcuong; Chun, Seongsoo; Sull, Sanghoon.

In: International Journal of Innovative Computing, Information and Control, Vol. 4, No. 11, 01.11.2008, p. 2861-2873.

Research output: Contribution to journalArticle

@article{d7abd8099540480882ccc73511ab5d44,
title = "Robust face tracking based on region correspondence and its application for person based indexing system",
abstract = "In this paper, a face tracking-approach is proposed based on templats matching for applying a video indexing application. Typically, this method is simpler and faster than other methods, but the main drawback is the poor performance in the case of face scaling changes and template drifts. To overcome these problems, the information about the facial features is incorporated into a template matching process. First, the face template is represented by two projection histograms of the face region and matching methods are used to determine the candidate face region. The matching is done this way as it is faster to match two 1-dimensional projection histograms than it is to match one 2-dimensional image. Next, the facial features are extracted from the candidate face region and a dissimilarity measure, based on the proximity of the facial features, is used to verify the face region. Finally, an anthropometric model, based on geometrical relationship between facial features, is used to refine the face region. The template to determine the face region in the next frame is dynamically updated using the refined face region. Thus, the proposed method can adapt to scale any changes with less computational cost. Experimental results are provided to demonstrate the efficiency of the proposed method.",
keywords = "Face tracking, Projection histogram, Template matching, Video indexing",
author = "Hanjin Ryu and Myunghoon Kim and Vietcuong Dinh and Seongsoo Chun and Sanghoon Sull",
year = "2008",
month = "11",
day = "1",
language = "English",
volume = "4",
pages = "2861--2873",
journal = "International Journal of Innovative Computing, Information and Control",
issn = "1349-4198",
publisher = "IJICIC Editorial Office",
number = "11",

}

TY - JOUR

T1 - Robust face tracking based on region correspondence and its application for person based indexing system

AU - Ryu, Hanjin

AU - Kim, Myunghoon

AU - Dinh, Vietcuong

AU - Chun, Seongsoo

AU - Sull, Sanghoon

PY - 2008/11/1

Y1 - 2008/11/1

N2 - In this paper, a face tracking-approach is proposed based on templats matching for applying a video indexing application. Typically, this method is simpler and faster than other methods, but the main drawback is the poor performance in the case of face scaling changes and template drifts. To overcome these problems, the information about the facial features is incorporated into a template matching process. First, the face template is represented by two projection histograms of the face region and matching methods are used to determine the candidate face region. The matching is done this way as it is faster to match two 1-dimensional projection histograms than it is to match one 2-dimensional image. Next, the facial features are extracted from the candidate face region and a dissimilarity measure, based on the proximity of the facial features, is used to verify the face region. Finally, an anthropometric model, based on geometrical relationship between facial features, is used to refine the face region. The template to determine the face region in the next frame is dynamically updated using the refined face region. Thus, the proposed method can adapt to scale any changes with less computational cost. Experimental results are provided to demonstrate the efficiency of the proposed method.

AB - In this paper, a face tracking-approach is proposed based on templats matching for applying a video indexing application. Typically, this method is simpler and faster than other methods, but the main drawback is the poor performance in the case of face scaling changes and template drifts. To overcome these problems, the information about the facial features is incorporated into a template matching process. First, the face template is represented by two projection histograms of the face region and matching methods are used to determine the candidate face region. The matching is done this way as it is faster to match two 1-dimensional projection histograms than it is to match one 2-dimensional image. Next, the facial features are extracted from the candidate face region and a dissimilarity measure, based on the proximity of the facial features, is used to verify the face region. Finally, an anthropometric model, based on geometrical relationship between facial features, is used to refine the face region. The template to determine the face region in the next frame is dynamically updated using the refined face region. Thus, the proposed method can adapt to scale any changes with less computational cost. Experimental results are provided to demonstrate the efficiency of the proposed method.

KW - Face tracking

KW - Projection histogram

KW - Template matching

KW - Video indexing

UR - http://www.scopus.com/inward/record.url?scp=63549134035&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=63549134035&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:63549134035

VL - 4

SP - 2861

EP - 2873

JO - International Journal of Innovative Computing, Information and Control

JF - International Journal of Innovative Computing, Information and Control

SN - 1349-4198

IS - 11

ER -