New Hausdorff Distance Based Approach For Face Localization

n improved approach of face localization based on the Hausdorff Distance(HD) measure is proposed in this work. First, we modify the similarity measure in the template-based matching. The common form of HD only computes the average of the mismatch of each point in the point set A to the point set B. We improved it by considering the average and the standard deviation together. The new measure is more robust against the variance of poses, expressions and identities. Then we also apply the HD measure to compute the symmetry of the test image, which is an important characteristic of human faces. This symmetry measure is combined with the template-based matching during face localization process. The experimental results demonstrate that both improvements can enhance the performance of face localization.