4) Discussion and Evaluation of the Results: Compared to the previous facial animation techniques, our technique has many advantages.
Accuracy and Robustness. The approach presented in this paper uses 103 markers so that movement of any part of the face can be recorded accurately, ensuring expressive facial emotion. To construct Phoneme-Independent Expression Eigenspace (PIEES) in the modeling part, dimensionally reduction has been performed on the original data to reduce the size. In our approach, EM Algorithms for PCA has been chosen to reduce the dimensions of the original data. After the reduction of dimensions on the data, EMPCA resulted in
negligible error of 1-2 percent variation compared with the original data. With that, accuracy and robustness are ensured in our approach.