Active authentication using facial attributes
The focus of this chapter is on face-based biometric authentication methods. These methods first use the camera sensor images to detect the face of the users. Next, they extract low-level features from the face images, and then apply their algorithm to the extracted features to authenticate the user. These algorithms have access to some model of the enrolled user for comparison. In [11], Hadid et al. use Haar cascade and Adaboost of [12] to detect face components and then use Local Binary Pattern (LBP) histograms [13] and nearest-neighbor thresholding for authentication. In [7], Fathy et al. extract two types of intensity features from the full face and facial parts and compare four still image-based verification algorithms with four image set-based methods. The common pitfall of most of these algorithms is that they are very sensitive to changes in the low-level feature domain. They are sensitive in the sense that if two face images are under the same pose and lighting condition, they can perform well, but in unconstrained settings they become very inaccurate.
Active authentication using facial attributes, Page 1 of 2
< Previous page Next page > /docserver/preview/fulltext/books/sc/pbse003e/PBSE003E_ch5-1.gif /docserver/preview/fulltext/books/sc/pbse003e/PBSE003E_ch5-2.gif