Face Recognition Based on Eigeneyes 1
Paulo Quintiliano, Antonio N. Santa-Rosa · 2003
Face recognition is a subject of emergent research that offers great challenges, mainly in adverse conditions. This paper proposes still a larger challenge: to perform face recognition from fragments of face images with approximately 20% of the face, based on eigeneyes techniques. This approach can work with partly occluded or nonideal illuminated images as well as in the cases when a person is disguised or wear a scarf, sun glasses, or mask. Even working with fragments of image, we achieved the recognition rate of 87%. Images were extracted from the Yale Face Database. 1 1