Facial Feature Localization Algorithm Based on A Modified Active Shape Model

A modified active shape model is proposed after deeply researching the original algorithm. The proposed algorithm includes the following four aspects. Firstly, edge information is strengthened before being marked in training images. Secondly, this paper proposes to extract 2D gradient feature on the highest level and the higher level of multi-resolution pyramid images, and use Gabor wavelet transformation to extract the lowest level's 2D texture feature. Thirdly, pyramid images are got by wavelet transformation. Finally, the length of 2D profile searching rectangle is changed according to different pyramid levels. Experimental results show great performance improvements both in localization accuracy and in computational speed are made by the proposed algorithm.