Automatic Facial Expression Recognition Using SVM Based on AAMs

An automatic facial expression recognition method is proposed to effectively recognize facial expression without any region unrelated to facial region. Support Vector Machine (SVM) is applied to recognize facial expression by Gabor features extracting using Gabor wavelet transformation after separate facial region from images Based on Active Appearance Models (AAMs), which reduce influence of illumination and pose. The feasibility and effectiveness of this system are verified by multiple experiments, and satisfied results are achieved.