Efficient facial features warping using BSM (Bayesian Shape Model)

초록

This paper proposes an efficient method for warping facial features. The existing methods use points, which are standard for facial features warping, without the reason or calculation. And existing methods have difficulties for facial feature warping. We estimate the standard points by using BSM(Bayeian Shape Model). From the experiment results for the various image, the proposed algorithm shows more natural results than the conventional algorithm and is more efficient than ASM(Active shape model). © 2008 Springer-Verlag Berlin Heidelberg.