3D and 2D face recognition based on image segmentation

In this paper we propose a framework for 3D Face Recognition System (3DFRS) using segmentation by grouping of regions of facial images before and after fusion of two modalities (color and depth images). Firstly, the detection of face region is based on the localization of nose tip and integral projection curves. Then, the features resulting from Principle Component Analyses (PCA) followed by Enhanced Fisher Model (EFM) are extracted. Finally, the classification process is performed with two methods, distance measurement L3 and Support Vector Machine (SVM). Experiments are performed on the CASIA3D face database which contains 123 persons under varying illumination and expression. We have tried to examine all the variants associated with our algorithms in order to optimize the maximum our recognition system. The promising results of the experimental evaluation show that our proposed approach achieves a high recognition performance.