In this paper, we present a framework for analyzing faces, with the specific goals of matching, comparing, and averaging their shapes. Here we handle variations of facial expression, pose variations, and occlusions between a gallery and probe scans. The radial curves are drawn from the nose tips and filled to the occluded part to form the shape of full facial surfaces. This representation seems natural for measuring facial deformations and is robust to challenges such as large facial expressions, large pose variations, missing parts, and partial occlusions due to glasses, hair, and so on. In this, we consider ORL data set for handling different types of challenges, like SVD is used to the estimation of missing facial parts. Here we using MATLAB for implementing our project.