Co klu Gauss Karõs Modeli Tabanlõ Y ¨ uz ¨ Oznitelikleri Bulma Algoritmasõ Multi-stream Gaussian Mixture Model based Facial Feature Localization
Kenichi Kumatani, Hazım Kemal Ekenel, Rainer Stiefelhagen · 2008
This paper presents a new facial feature localization system which estimates positions of eyes, nose and mouth corners si- multaneously. In contrast to conventional systems, we use the multi-stream Gaussian mixture model (GMM) framework in or- der to represent structural and appearance information of facial features. We construct a GMM for the region of each facial fea- ture, where the principal component analysis is used to extract each facial feature. We also build a GMM which represents the structural information of a face, relative positions of facial fea- tures. Those models are combined based on the multi-stream framework. It can reduce the computation time to search re- gion of interest (ROI). We demonstrate the effectiveness of our algorithm through experiments on the BioID Face Database.