Face pose estimating system based on eigen space analysis

In this paper, we propose a new system for estimating face pose from a facial image. In this system, input facial image is compared with database of images of various face pose, then the matched image provides the face pose. The database of images includes not only various face pose but also various illumination conditions, so that the face pose estimating system can be used under various illumination condition. For collecting such various facial images, they are generated by computer, rather than taking real images. Eigen space method is used for searching the matched image with input facial image. Since various illumination images are collected in the database of facial images, the extracted principle eigen vectors mostly depend on the face pose. By performing the matching process in the eigen space, matched image with the input facial image can be found. The pose of the matched image is most closest to the input face. The matching process is also fast because it is performed in small dimensional space spanned by only selected eigen vectors. The proposed pose estimating system can continuously track the face pose of different person under different light condition.