3D facial image analysis for human identification

摘要

Presents a new method based on 3D facial section analysis forhuman face identification. The range data on different curves ofintersection, such as, vertical, horizontal, and circular, with the 3Dface are used as the distinguishing surface features. These are robustagainst changes in makeup and lighting as compared to the 2D approach,and more efficient for computation and data storage than other 3Dapproaches. The performance of this method for human face identificationis evaluated through several matching experiments. It has been observedthat the curves of intersection crossing the facial central areavertically and contain the features, like, nose and mouth, have themajor distinctiveness. Those crossing near the inner corners of the eyesand a part of the nose are also effective for human face identification.But it seems that such an effectiveness requires the accurate extractionof the sections against their locational variations