Face detection based on support vector machines

Dihua Xi, Seong Whan Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)


Face detection is a key problem in building an automatic face system such as face recognition and authentication. A number of approaches have been proposed for face detection. Recently, a novel statistical machine learning method, support vector machine, has been employed. Generally, the current SVM-based methods can be divided into two categories: component-based and whole face-based. It is difficult for the component-based method to extract the small face due to no enough information for each component exists. On the other hand, the whole face-based method is too much computationally expensive to build an effective system. In this paper we present a fast system named wavelet-SVM method to extract a wide range scales of faces from grey-scale images or color images with a preprocessing using a TSL color model. The system is not only accurate and effective, but also largely speeds the system up by applying a TSL B-G color model and multiresolution wavelet decomposition.

Original languageEnglish
Title of host publicationPattern Recognition with Support Vector Machines - First International Workshop, SVM 2002 Niagara Falls, Canada, August 10, 2002 Proceedings
EditorsSeong-Whan Lee, Alessandro Verri
PublisherSpringer Verlag
Number of pages18
ISBN (Print)354044016X
Publication statusPublished - 2002
Event1st International Workshop on Pattern Recognition with Support Vector Machines, SVM 2002 - Niagara Falls, Canada
Duration: 2002 Aug 102002 Aug 10

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Other1st International Workshop on Pattern Recognition with Support Vector Machines, SVM 2002
CityNiagara Falls

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science


Dive into the research topics of 'Face detection based on support vector machines'. Together they form a unique fingerprint.

Cite this