Title :
Face animation parameters extraction and driving
Author :
Wang, Ke ; Yin, Baocai ; Guo, Jintong ; Ma, Shuyan
Author_Institution :
Multimedia & Intelligent Software Technol. Lab., Beijing Univ. of Technol., China
Abstract :
A new method combining the CANDIDE model with the Lucas-Kanade algorithm is proposed to track face features and extract face animation parameters. To adapt a general mesh model to a specific face scan data, a face model is made. Combining the parameter model and muscle model, the face animation is created finally by driving face animation parameters. The experimental results show the face animation parameter extraction and driving are accurate.
Keywords :
computer animation; feature extraction; parameter estimation; tracking; CANDIDE model; Lucas-Kanade algorithm; face animation parameter driving; face feature tracking; face model; face scan data; general mesh model; muscle model; parameter extraction; parameter model; Active appearance model; Electronic mail; Face; Facial animation; MPEG 4 Standard; Muscles; Optimization methods; Parameter extraction; Shape; Tracking;
Conference_Titel :
Communications and Information Technology, 2004. ISCIT 2004. IEEE International Symposium on
Print_ISBN :
0-7803-8593-4
DOI :
10.1109/ISCIT.2004.1413918
