K-means Principal Geodesic Analysis on Riemannian Manifolds

  • Youshan ZhangEmail author
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 1069)


Principal geodesic analysis (PGA) has been proposed for data dimensionality reduction on manifolds. However, a single PGA model is limited to data with a single modality. In this paper, we are the first to propose a generalized K-means-PGA model on manifolds. This model can analyze multi-mode nonlinear data, which applies to more manifolds (Sphere, Kendall’s shape and Grassmannian). To show the applicability of our model, we apply our model to spherical, Kendall’s shape and Grassmannian manifolds. Our K-means-PGA model offers correct geometry geodesic recovery than K-means-PCA model. We also show shape variations from different clusters of human corpus callosum and mandible data. To demonstrate the efficiency of our model, we perform face recognition using ORL dataset. Our model has a higher accuracy (99.5%) than K-means-PCA model (95%) in Euclidean space.


K-means Principal Geodesic Analysis Manifolds 


  1. 1.
    Turk, M., Pentland, A.: Eigenfaces for recognition. J. Cogn. Neurosci. 3(1), 71–86 (1991)CrossRefGoogle Scholar
  2. 2.
    Timothy, F.C., Gareth, J.E., Christopher, J.T.: Active appearance models. IEEE Trans. Pattern Anal. Mach. Intell. 23(6), 681–685 (2001)CrossRefGoogle Scholar
  3. 3.
    Blanz, V., Vetter, T.: Face recognition based on fitting a 3D morphable model. IEEE Trans. Pattern Anal. Mach. Intell. 25(9), 1063–1074 (2003)CrossRefGoogle Scholar
  4. 4.
    Basri, R., Jacobs, D.: Lambertian reflectance and linear subspaces. In: Proceedings Eighth IEEE International Conference on Computer Vision 2001. ICCV 2001, vol. 2, pp. 383–390. IEEE (2001)Google Scholar
  5. 5.
    Peter, N.B., David, J.K.: What is the set of images of an object under all possible illumination conditions? Int. J. Comput. Vision 28(3), 245–260 (1998)MathSciNetCrossRefGoogle Scholar
  6. 6.
    William, A.P.S., Edwin, R.H.: Recovering facial shape using a statistical model of surface normal direction. IEEE Trans. Pattern Anal. Mach. Intell. 28(12), 1914–1930 (2006)CrossRefGoogle Scholar
  7. 7.
    William, A.P.S., Edwin, R.H.: Face recognition using 2.5D shape information. In: 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 1407–1414. IEEE (2006)Google Scholar
  8. 8.
    Yin, X., Liu, X.: Multi-task convolutional neural network for pose-invariant face recognition. IEEE Trans. Image Process. 27(2), 964–975 (2018)MathSciNetCrossRefGoogle Scholar
  9. 9.
    Ding, C., Tao, D.: Trunk-branch ensemble convolutional neural networks for video-based face recognition. IEEE Trans. Pattern Anal. Mach. Intell. 40(4), 1002–1014 (2018)CrossRefGoogle Scholar
  10. 10.
    Fletcher, P.T., Lu, C., Pizer, S.M., Joshi, S.: Principal geodesic analysis for the study of nonlinear statistics of shape. IEEE Trans. Med. Imaging 23(8), 995–1005 (2004)CrossRefGoogle Scholar
  11. 11.
    Dickens, M.P., Smith, W.A.P., Wu, J., Hancock, E.R.: Face recognition using principal geodesic analysis and manifold learning. In: Iberian Conference on Pattern Recognition and Image Analysis, pp. 426–434. Springer (2007)Google Scholar
  12. 12.
    Pennec, X.: Probabilities and statistics on Riemannian manifolds: a geometric approach. PhD thesis, INRIA (2004)Google Scholar
  13. 13.
    Jing, W., William, A.P.S., Edwin, R.H.: Weighted principal geodesic analysis for facial gender classification. In: Iberoamerican Congress on Pattern Recognition, pp. 331–339. Springer (2007)Google Scholar
  14. 14.
    Perdigão do Carmo, M.: Riemannian Geometry. Birkhauser (1992)Google Scholar
  15. 15.
    Gallier, J.: Notes on differential geometry and lie groups (2012)Google Scholar
  16. 16.
    Pennec, X.: Statistical computing on manifolds: from Riemannian geometry to computational anatomy. In: Emerging Trends in Visual Computing, pp. 347–386. Springer (2009)Google Scholar
  17. 17.
    Zhang, M., Fletcher, P.T.: Probabilistic principal geodesic analysis. In: Advances in Neural Information Processing Systems, pp. 1178–1186 (2013)Google Scholar
  18. 18.
    David, G.K.: Shape manifolds, procrustean metrics, and complex projective spaces. Bull. Lond. Math. Soc. 16(2), 81–121 (1984)MathSciNetCrossRefGoogle Scholar
  19. 19.
    John, C.G.: Generalized procrustes analysis. Psychometrika 40(1), 33–51 (1975)MathSciNetCrossRefGoogle Scholar
  20. 20.
    Kyle, A.G., Anuj, S., Xiuwen, L., Paul Van, D.: Efficient algorithms for inferences on Grassmann manifolds. In: 2003 IEEE Workshop on Statistical Signal Processing, pp. 315–318. IEEE (2003)Google Scholar
  21. 21.
    Edelman, A., Arias, T.A., Smith, S.T.: The geometry of algorithms with orthogonality constraints. SIAM J. Matrix Anal. Appl. 20(2), 303–353 (1998)MathSciNetCrossRefGoogle Scholar
  22. 22.
    Absil, P.-A., Mahony, R., Sepulchre, R.: Riemannian geometry of Grassmann manifolds with a view on algorithmic computation. Acta Applicandae Mathematica 80(2), 199–220 (2004)MathSciNetCrossRefGoogle Scholar
  23. 23.
    Moo, K.C., Anqi, Q., Seongho, S., Houri, K.V.: Unified heat Kernel regression for diffusion, Kernel smoothing and wavelets on manifolds and its application to mandible growth modeling in CT images. Med. Image Anal. 22(1), 63–76 (2015)CrossRefGoogle Scholar
  24. 24.
    Guodong, G., Stan, Z.L., Kapluk, C.: Face recognition by support vector machines. In: Proceedings Fourth IEEE International Conference on Automatic Face and Gesture Recognition 2000, pp. 196–201. IEEE (2000)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Computer Science and EngineeringLehigh UniversityBethlehemUSA

Personalised recommendations