Face recognition under varying poses: the role of texture and shape

Although remarkably robust, face recognition is not perfectly invariant to pose and viewpoint changes. It has long been known that both profile and full-face views result in poorer recognition performance than a 3/4 view. However, little data exist which investigate this phenomenon in detail. The present work provides such data using a high angular resolution and a large range of poses. Since there are inconsistencies in the literature concerning these issues, we emphasize the different roles of the learning view and the testing view in the recognition experiment. We also emphasize the roles of information contained in the texture and in the shape of a face. Our stimuli were generated from laser-scanned head models and contained either the natural texture or only Lambertian shading and no texture. The results of our same/different face recognition experiments are: (1) only the learning view but not the testing view affects recognition performance. (2) For textured faces the optimal learning view is closer to the full-face view than for the shaded faces. (3) For shaded faces, we find a significantly better recognition performance for the symmetric view. The results can be interpreted in terms of different strategies to recover invariants from texture and from shading.