A Novel Real-Time Facial Expression Capturing Method
Yong Sheng Wang · Applied Mechanics and Materials · 2013
In this paper, we propose a new facial expression capturing method in real-time mode. Our proposed method main includes two processes: 1) Online process and 2) Offline process. The offline process adopt human face database to build 2D face model and 3D shape model, and then train an expression feature model. Afterwards, online process extracts feature points from human face images, and then obtains facial expression by SVM classifier which is trained from offline process. The main creativity of our method lies in that we propose an effective face detection approach and propose an optimal evaluation method to facial expression recognition. Experimental results show that our approach can capture facial expression precisely in real-time mode.