Physical facial model based on 3D-CT data for facial image analysis and synthesis

Yoshimitsu Aoki, Shuji Hashimoto

    研究成果: Conference contribution

    10 被引用数 (Scopus)

    抄録

    In this paper, we have proposed a physics-based facial model based on anatomical data and knowledge, which has a three dimensional structure for facial image analysis and synthesis. The skin and the skull models are constructed from 3D-CT scanned data, and the facial muscles were placed between them just like as a real human face. In regards to the skin and muscle models, spring frames are used to simulate the elastic dynamics of a real facial skin and muscles. Facial Modification can be flexibly realized by two factors: the expansion and contraction of non-linear springs which simulate a skin and muscles, and the motion of the jaw part of the skull. Using our system, we can represent realistic facial animation and predict facial modifications under given conditions for muscles and the skull. Furthermore, our method is effective for analyzing relationship between modification on the facial surface and the motion of muscles and the skull.

    本文言語English
    ホスト出版物のタイトルProceedings - 3rd IEEE International Conference on Automatic Face and Gesture Recognition, FG 1998
    出版社IEEE Computer Society
    ページ448-453
    ページ数6
    ISBN(印刷版)0818683449, 9780818683442
    DOI
    出版ステータスPublished - 1998
    イベント3rd IEEE International Conference on Automatic Face and Gesture Recognition, FG 1998 - Nara
    継続期間: 1998 4 141998 4 16

    Other

    Other3rd IEEE International Conference on Automatic Face and Gesture Recognition, FG 1998
    CityNara
    Period98/4/1498/4/16

    ASJC Scopus subject areas

    • Computer Vision and Pattern Recognition

    フィンガープリント 「Physical facial model based on 3D-CT data for facial image analysis and synthesis」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

    引用スタイル