Statistical model for human face detection using multi-resolutionfeatures

摘要

A novel method for face detection is presented, which is robust toface, illumination, and pose variations. This method is based onprocessing of image features constructed using multi-resolutiontechniques, and using statistical models of these multi-resolutionfeatures to detect the presence of faces in an image. Simulation resultsshow that the new method significantly outperforms traditional methodsbased on normalized correlation of images with an average face