Human Motion Synthesis by Motion Manifold Learning and Motion Primitive Segmentation

  • Chan-Su Lee
  • Ahmed Elgammal
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4069)


We propose motion manifold learning and motion primitive segmentation framework for human motion synthesis from motion-captured data. High dimensional motion capture date are represented using a low dimensional representation by topology preserving network, which maps similar motion instances to the neighborhood points on the low dimensional motion manifold. Nonlinear manifold learning between a low dimensional manifold representation and high dimensional motion data provides a generative model to synthesize new motion sequence by controlling trajectory on the low dimensional motion manifold. We segment motion primitives by analyzing low dimensional representation of body poses through motion from motion captured data. Clustering techniques like k-means algorithms are used to find motion primitives after dimensionality reduction. Motion dynamics in training sequences can be described by transition characteristics of motion primitives. The transition matrix represents the temporal dynamics of the motion with Markovian assumption. We can generate new motion sequences by perturbing the temporal dynamics.


Gaussian Mixture Model Human Motion Motion Sequence Motion Primitive Motion Capture Data 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Barbic, J., Safonova, A., Pan, J.-Y., Faloutsos, C., Hodgins, J.K., Pollard, N.S.: Segmenting motion capture data into distinct behaviors. In: Proc. of Graphics Interface (2004)Google Scholar
  2. 2.
    Bettinger, F., Cootes, T.F.: A model of facial behaviour. In: Proc. of FGR, pp. 123–128 (2004)Google Scholar
  3. 3.
    Bowden, R.: Learning statistical models of human motion. In: Proc. of IEEE Workshop on Human Modeling, Analysis & Synthesis (2000)Google Scholar
  4. 4.
    Brand, M., Hertzmann, A.: Style machines. In: Proc. of SIGGRAPH, pp. 183–192 (2000)Google Scholar
  5. 5.
    Brand, M., Kettnaker, V.: Discovery and segmentation of activities in video. IEEE Trans. on PAMI 22(8) (2000)Google Scholar
  6. 6.
    Elgammal, A., Lee, C.-S.: Separating style and content on a nonlinear manifold. In: Proc. CVPR, vol. 1, pp. 478–485 (2004)Google Scholar
  7. 7.
    Galata, A., Johnson, N., Hogg, D.: Learning variable-length markov models of behavior. Computer Vision and Image Understanding 81, 398–413 (2001)MATHCrossRefGoogle Scholar
  8. 8.
    Grochow, K., Martin, S.L., Hertzmann, A., Popovic, Z.: Style-based inverse kinematics. ACM Trans. Graph. 23(3), 522–531 (2004)CrossRefGoogle Scholar
  9. 9.
    Kovar, L., Gleicher, M.: Flexible automatic motion blending with registration curves. In: Proc. of SCA, pp. 214–224 (2003)Google Scholar
  10. 10.
    Mussa-Ivaldi, F., Bizzi, E.: Motor learning through the combination of primitives. Philosopical Transactions of the Royal Society of London Seris B, Biological Science 355, 1755–1769 (2000)CrossRefGoogle Scholar
  11. 11.
    Park, S., Aggarwal, J.K.: Recognition of two-person interactions using a hierarchical bayesian network. In: Proc. of Workshop on Video surveillance, pp. 65–76. ACM Press, New York (2003)Google Scholar
  12. 12.
    Poggio, T., Girosi, F.: Networks for approximation and learning. Proc. IEEE 78(9), 1481–1497 (1990)CrossRefGoogle Scholar
  13. 13.
    Ritter, H., Martinetz, T., Schulten, K.: Nueral Computation and Self-Organizing Maps. Addison-Wesley, Reading (1991)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Chan-Su Lee
    • 1
  • Ahmed Elgammal
    • 1
  1. 1.Rutgers UniversityPiscatawayUSA

Personalised recommendations