Abstract
Active Shape Model (ASM) is a powerful satistical tool for image interpretation, especially in face alignment. In this paper, an improved ASM framework, GentleBoost based SIFT-ASM is proposed. Local appearances of landmarks are originally represented by SIFT (Scale-Invariant Feature Transform) descriptors, and GentleBoost classifiers are applied to model and search the SIFT features instead of the unnecessary assumption of Gaussian distribution. Experimental results show that SIFT-ASM significantly outperforms the original ASM in aligning and localizing facial features.
