Face detection based on support vector machines

Dihua Xi, Seong Whan Lee

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

    2 Citations (Scopus)

    Abstract

    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
    Pages370-387
    Number of pages18
    ISBN (Print)354044016X
    DOIs
    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)
    Volume2388
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Other

    Other1st International Workshop on Pattern Recognition with Support Vector Machines, SVM 2002
    Country/TerritoryCanada
    CityNiagara Falls
    Period02/8/1002/8/10

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • General Computer Science

    Fingerprint

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

    Cite this