Advertisement

Nonparametric Density Estimation for Human Pose Tracking

  • Thomas Brox
  • Bodo Rosenhahn
  • Uwe G. Kersting
  • Daniel Cremers
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4174)

Abstract

The present paper considers the supplement of prior knowledge about joint angle configurations in the scope of 3-D human pose tracking. Training samples obtained from an industrial marker based tracking system are used for a nonparametric Parzen density estimation in the 12-dimensional joint configuration space. These learned probability densities constrain the image-driven joint angle estimates by drawing solutions towards familiar configurations. This prevents the method from producing unrealistic pose estimates due to unreliable image cues. Experiments on sequences with a human leg model reveal a considerably increased robustness, particularly in the presence of disturbed images and occlusions.

Keywords

Training Sample Joint Angle Rigid Body Motion Kinematic Chain Point Correspondence 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Agarwal, A., Triggs, B.: Recovering 3D human pose from monocular images. IEEE Transactions on Pattern Analysis and Machine Intelligence 28(1), 44–58 (2006)CrossRefGoogle Scholar
  2. 2.
    Bregler, C., Malik, J., Pullen, K.: Twist based acquisition and tracking of animal and human kinematics. International Journal of Computer Vision 56(3), 179–194 (2004)CrossRefGoogle Scholar
  3. 3.
    Brox, T., Rosenhahn, B., Weickert, J.: Three-Dimensional Shape Knowledge for Joint Image Segmentation and Pose Estimation. In: Kropatsch, W.G., Sablatnig, R., Hanbury, A. (eds.) DAGM 2005. LNCS, vol. 3663, pp. 109–116. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  4. 4.
    Cremers, D., Osher, S., Soatto, S.: Kernel density estimation and intrinsic alignment for shape priors in level set segmentation. In: International Journal of Computer Vision (to appear)Google Scholar
  5. 5.
    Kadir, T., Brady, M.: Unsupervised non-parametric region segmentation using level sets. In: Proc. Ninth IEEE International Conference on Computer Vision, vol. 2, pp. 1267–1274 (2003)Google Scholar
  6. 6.
    Kim, J., Fisher, J., Yezzi, A., Cetin, M., Willsky, A.: A nonparametric statistical method for image segmentation using information theory and curve evolution. IEEE Transactions on Image Processing 14(10), 1486–1502 (2005)CrossRefMathSciNetGoogle Scholar
  7. 7.
    Murray, R., Li, Z., Sastry, S.: Mathematical Introduction to Robotic Manipulation. CRC Press, Baton Rouge (1994)MATHGoogle Scholar
  8. 8.
    Parzen, E.: On the estimation of a probability density function and the mode. Annals of Mathematical Statistics 33, 1065–1076 (1962)MATHCrossRefMathSciNetGoogle Scholar
  9. 9.
    Rosenblatt, F.: Remarks on some nonparametric estimates of a density function. Annals of Mathematical Statistics 27, 832–837 (1956)MATHCrossRefMathSciNetGoogle Scholar
  10. 10.
    Rosenhahn, B., Brox, T., Kersting, U., Smith, A., Gurney, J., Klette, R.: A system for marker-less motion capture. Künstliche Intelligenz (1), 45–51 (2006)Google Scholar
  11. 11.
    Rosenhahn, B., Sommer, G.: Pose Estimation of Free-Form Objects. In: Pajdla, T., Matas, J(G.) (eds.) ECCV 2004. LNCS, vol. 3021, pp. 414–427. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  12. 12.
    Rousson, M., Brox, T., Deriche, R.: Active unsupervised texture segmentation on a diffusion based feature space. In: Proc. International Conference on Computer Vision and Pattern Recognition, Madison, WI, pp. 699–704 (June 2003)Google Scholar
  13. 13.
    Shakhnarovich, G., Viola, P., Darell, T.: Fast pose estimation with parameter sensitive hashing. In: Proc. International Conference on Computer Vision, Nice, France, pp. 750–757 (October 2003)Google Scholar
  14. 14.
    Silverman, B.W.: Density Estimation for Statistics and Data Analysis. Chapman and Hall, New York (1986)MATHGoogle Scholar
  15. 15.
    Sminchisescu, C., Jepson, A.: Generative modeling for continuous non-linearly embedded visual inference. In: Proc. International Conference on Machine Learning (2004)Google Scholar
  16. 16.
    Sminchisescu, C., Triggs, B.: Estimating articulated human motion with covariance scaled sampling. International Journal of Robotics Research 22(6), 371–391 (2003)CrossRefGoogle Scholar
  17. 17.
    Zhang, J., Collins, R., Liu, Y.: Representation and matching of articulated shapes. In: Proc. International Conference on Computer Vision and Pattern Recognition, pp. 342–349 (June 2004)Google Scholar
  18. 18.
    Zhang, J., Collins, R., Liu, Y.: Bayesian body localization using mixture of nonlinear shape models. In: Proc. International Conference on Computer Vision, Beijing, China, pp. 725–732 (October 2005)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Thomas Brox
    • 1
  • Bodo Rosenhahn
    • 2
  • Uwe G. Kersting
    • 3
  • Daniel Cremers
    • 1
  1. 1.CVPR GroupUniversity of BonnBonnGermany
  2. 2.MPI for Computer ScienceSaarbrückenGermany
  3. 3.Department of Sport and Exercise ScienceThe University of AucklandNew Zealand

Personalised recommendations