Extracting contours and features from frontal face images

Taro Yokoyama, Katsuaki Tanaka, Kensuke Hisatomi, Yasushi Yagi, Masahiko Yachida, Fumio Hara, Shuji Hashimoto

    Research output: Contribution to journalArticle

    4 Citations (Scopus)

    Abstract

    The aim of this research is to locate a face in an input image, then extract the contour lines of the face and the facial features, such as the eyes, mouth, eyebrows and nose in a standard office environment. In this paper we describe a system for extracting the facial features required to reconstruct a 3D model of a face from a 2D image. The technique for extracting the eye and mouth contour lines is based on the combination of a quadratic approximation of the edges and an active contour model. The technique for face outline contour line relys on the hybridization of a symmetrical active contour model and an elastic contour model with a standard facial shape model. We applied this system to 46 faces and compared the results with manually detected contours: 90% of the eye and mouth contours and 80% of the jaw contours and pupil positions matched the hand-prepared contours closely.

    Original languageEnglish
    Pages (from-to)1605-1614
    Number of pages10
    JournalKyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers
    Volume53
    Issue number11
    Publication statusPublished - 1999

    ASJC Scopus subject areas

    • Computer Vision and Pattern Recognition
    • Electronic, Optical and Magnetic Materials

    Cite this

    Yokoyama, T., Tanaka, K., Hisatomi, K., Yagi, Y., Yachida, M., Hara, F., & Hashimoto, S. (1999). Extracting contours and features from frontal face images. Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers, 53(11), 1605-1614.