A Face Detector Based on Hausdorff Distance

Face detector is used very often as a previous step in some biometric applications, like face recognition or advanced human computer interaction,, which usually needs the face detection step to be efficient and robust enough to achieve better results. In this paper Hausdorff distance is used to search interested region, which is more similar to a face. A further verification step is taken to detect some false positives. Experiments show that the proposed face detector can perform accurate face detection even with complex background.