A Realistic Human Face Modeling
This paper presents a novel approach to produce a realistic 3D face model from multi-view face images. To extract facial region and facial feature points from color image a new nonparametric skin color model that is proposed. Conventionally used parametric skin color models for face detection have lack of robustness for varying lighting conditions and need extra work. To resolve the limitation of current skin color model, we exploit the Hue-Tint chrominance components and represent the skin chrominance distribution as a linear function. Thus, the facial color distribution is simply described as a combination of the maximum and minimum values of Hue and Tint components. Moreover, the minimal facial feature positions detected by the proposed skin model are adjusted by using edge information of the detected facial region along with the proportions of the face. To produce the realistic face model, we adopt log RBF(Radial-Based Function) to deform the generic face model according to the detected facial feature points from face images. The experiments show that the proposed approach efficiently detects facial feature points and produces a realistic 3D face model.