Facial feature point localization algorithm combining improved ASM and AAM
Xiaolong Ren · Journal of Optoelectronics·laser · 2011
Active shape models(ASM) and active appearance model(AAM) are the primary methods of facial feature point localization.In this paper,an improved facial feature localization algorithm that combines ASM and AAM is proposed to overcome the shortcomings of the traditional ASM and AAM.ASM is used to locate the outer points,and the AAM is used to locate the inner points,which significantly improves the accuracy.Besides,the 2D contour is used on some key feature points to improve the accuracy of locating.The position information of face,eyes,nose and mouth is used to initialize the average face shape.The multi-scaled ASM model is used to accelerate the convergence rate.Experimental results show that our proposed method can achieve higher accuracy than the traditional ASM,AAM and the improved IASM,PAAM method.