Wavelet packet analysis for face recognition

Content-based indexing methods are of great interest for image and video retrieval in audio-visual archives, such as in the Esprit project DiVAN that we are currently developing. Detecting and recognizing human faces automatically in video data provide users with powerful tools for performing queries. In this article, a new scheme for face recognition using a wavelet packet decomposition is presented. Each face is described by a subset of band ltered images containing wavelet coe cients. These coe cients characterize the face texture and a set of simple statistical measures allows us to form compact and meaningful feature vectors. Then, an e cient and reliable probabilistic metric derived from the Bhattacharrya distance is used in order to classify the face feature vectors into person classes.