3D Facial Expression Recognition Based on Encoded Templates
Yiding Wang, Xiaolei Ma · 2012
In this paper, we propose a method for 3D facial expression recognition. The algorithm is composed of three steps. The first step is to extract the region of interested 3D face, some data preprocessing works, including face location, point cloud rotation and uniform distribution of points cloud, have been done in this step. Otherwise, the second step is feature extraction, novel features are extracted by calculating the normal vectors in 21 facial blocks. Six expressions' encoded templates are developed by finding the common attribution of the same expression. In the final step, the feature matching method is used to identify the facial expressions. The average recognition rate is above 83%, and for some certain expressions (happiness, anger), the recognition rate is nearly 100%. Experimental results have shown the effectiveness of the proposed algorithm. All experiments are based on Beihang IRIP 3DFACE database.