Facial Shape Spaces from Surface Normals
In this paper, we draw on ideas from the field of statistical shape analysis to construct shape-spaces that span facial expressions and gender, and use the resulting shape-model to perform face recognition under varying expression and gender. Our novel contribution is to show how to construct shape-spaces over fields of surface normals rather than Cartesian landmark points. According to this model face needle-maps (or fields of surface normals) are points in a high-dimensional manifold referred to as a shape-space. We compute geodesic distances to compare the similarity between faces and gender difference.
KeywordsShape-Space Surface Normals Geodesic distance
Unable to display preview. Download preview PDF.
- 2.Dryden, I.L., Mardia, K.V.: Statistical Shape Analysis. Wiley Series in Probability and Statistics (1998)Google Scholar
- 8.Pennec, X.: Probabilities and statistics on Riemannian manifolds: basic tools for geometric measurements. In: Proc. IEEE Workshop on Nonlinear Signal and Image Processing (1999)Google Scholar
- 13.Ziezold, H.: On expected figures and a strong law of large numbers for random elements in quasi-metric spaces. In: Trans. 7th Praque Conf. Information Theory, Stat. Dec. Func, Random Processes, vol. A, pp. 491–510 (1977)Google Scholar