A scheme for pose-independent face recognition is presented. An ``unwrapped'' texture map is constructed from a video sequence using a texture-from-motion approach, which is shown to be quite accurate. Simple lighting normalization methods improve robustness to directional and/or varying lighting conditions. Recognition of single frames against calculated unwrapped textures is carried out using principal component analysis. The system is typically better than 90% correct in its identifications.
Home Contents Author index Keyword index