Efficiently comparing face images using a modified Hausdorff distance

Comparing face images using the Hausdorff distance is one of the face matching and fast screening techniques. As a fast screening technique, the computational efficiency is a key issue. An efficient method for Hausdorff distance-based face matching and fast screening is proposed. The method utilises dominant points, instead of edge maps, as features for measuring similarity. A new formulation of Hausdorff distance is designed for significance-based dominant point matching. The experiments demonstrate that the proposed approach significantly improves the computational speed and storage economy. It provides a very efficient way to use the Hausdorff distance measure for face matching and fast screening.