Statistical Shape Analysis for 3D Facial Images

Abstract — Recently, the genetic association of human facial morphological variation attracts substantial attention. This study proposes a general framework for analyzing facial morphology variation using scanned 3D landmarks, and explores the phenotype features of facial morphology for identifying population root of Japanese archipelago. After registration for the dense 3D facial points, we investigate both PCA and Mean Hyperplane for exploring the facial morphological variations. Then, in order to reduce the in-population variance of statistical features, we normalize them firstly, and explore the identification of population using the combined phenotype features. Experiments show that our proposed strategy can give promising identification performances between the Mainland Japanese and the Ryukyuan. Keywords-facial morphology; gene; statistical learning; mean hyperplane; normalization I