A Hybrid Algorithm of Facial Feature Point Location Based on Improved MR-ASM and AAM

Accurate and robust location of feature point is a difficult and complicated issue in face recognition. This paper proposes a facial feature point location algorithm based on improved multi-resolution-active shape models (MR-ASM) and active appearance models (AAM). There are three main achievements of our algorithm: 1, we unify MR-ASM and AAM together to improve facial feature point location. 2, AAM is improved so that it has the same implementation framework as ASM. 3, the MR-ASM is adjusted to figure out more robust parameters. Experimental results prove that our algorithm is more accurate than traditional MR-ASM