3D Hand Tracking in a Stochastic Approximation Setting

  • Desmond Chik
  • Jochen Trumpf
  • Nicol N. Schraudolph
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4814)


This paper introduces a hand tracking system with a theoretical proof of convergence. The tracking system follows a model-based approach and uses image-based cues, namely silhouettes and colour constancy. We show that, with the exception of a small set of parameter configurations, the cost function of our tracker has a well-behaved unique minimum. The convergence proof for the tracker relies on the convergence theory in stochastic approximation. We demonstrate that our tracker meets the sufficient conditions for stochastic approximation to hold locally. Experimental results on synthetic images generated from real hand motions show the feasibility of this approach.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Erol, A., Bebis, G.N., Nicolescu, M., Boyle, R.D., Twombly, X.: A review on vision-based full DOF hand motion estimation. In: Vision for Human-Computer Interaction, vol. III, p. 75 (2005)Google Scholar
  2. 2.
    Wang, L., Hu, W.M., Tan, T.N.: Recent developments in human motion analysis. Pattern Recognition 36(3), 585–601 (2003)CrossRefGoogle Scholar
  3. 3.
    Sminchisescu, C., Triggs, B.: Covariance scaled sampling for monocular 3D body tracking. In: CVPR, vol. I, pp. 447–454 (2001)Google Scholar
  4. 4.
    Sudderth, E.B., Mandel, M.I., Freeman, W.T., Willsky, A.S.: Visual hand tracking using nonparametric belief propagation. In: Workshop on Generative Model Based Vision, 189 (2004)Google Scholar
  5. 5.
    Kehl, R., Gool, L.J.V.: Markerless tracking of complex human motions from multiple views. Computer Vision and Image Understanding 103(2-3), 190–209 (2006)CrossRefGoogle Scholar
  6. 6.
    Carranza, J., Theobalt, C., Magnor, M., Seidel, H.: Freeviewpoint video of human actors. In: ACM Transactions on Graphics. ACM SIGGRAPH, pp. 569–577. ACM Press, New York (2003)Google Scholar
  7. 7.
    Theobalt, C., Carranza, J., Magnor, M.A., Seidel, H.P.: Combining 3d flow fields with silhouette-based human motion capture for immersive video. Graph. Models 66(6), 333–351 (2004)CrossRefGoogle Scholar
  8. 8.
    Sundaresan, A., Chellappa, R.: Multi-camera tracking of articulated human motion using motion and shape cues. In: Narayanan, P.J., Nayar, S.K., Shum, H-Y. (eds.) ACCV 2006. LNCS, vol. 3852, pp. 131–140. Springer, Heidelberg (2006)CrossRefGoogle Scholar
  9. 9.
    Lu, S., Metaxas, D., Samaras, D., Oliensis, J.: Using multiple cues for hand tracking and model refinement. In: IEEE Computer Vision and Pattern Recognition or CVPR, vol. II, pp. 443–450. IEEE Computer Society Press, Los Alamitos (2003)Google Scholar
  10. 10.
    Balan, A.O., Sigal, L., Black, M.J.: A quantitative evaluation of video-based 3D person tracking. In: International Workshop on Performance Evaluation of Tracking and Surveillance, pp. 349–356 (2005)Google Scholar
  11. 11.
    Lewis, J.P., Cordner, M., Fong, N.: Pose space deformation: a unified approach to shape interpolation and skeleton-driven deformation. In: SIGGRAPH 2000, New York, USA, pp. 165–172 (2000)Google Scholar
  12. 12.
    Borgefors, G.: Distance transformations in digital images. Comput. Vision Graph. Image Process. 34(3), 344–371 (1986)CrossRefGoogle Scholar
  13. 13.
    Hartley, R.I., Zisserman, A.: Multiple View Geometry in Computer Vision, 2nd edn. Cambridge University Press, Cambridge (2004)MATHGoogle Scholar
  14. 14.
    Robbins, H., Monro, S.: A stochastic approximation method. Annals of Mathematical Statistics 22, 400–407 (1951)MathSciNetGoogle Scholar
  15. 15.
    Blum, J.R.: Multidimensional stochastic approximation methods. Annals of Mathematical Statistics 25, 737–744 (1954)MathSciNetGoogle Scholar
  16. 16.
    Schraudolph, N.N.: Local gain adaptation in stochastic gradient descent. In: ICANN, Edinburgh, Scotland, pp. 569–574. IEE, London (1999)Google Scholar
  17. 17.
    Schraudolph, N.N.: Fast curvature matrix-vector products for second-order gradient descent. Neural Computation 14(7), 1723–1738 (2002)MATHCrossRefGoogle Scholar
  18. 18.
    Bray, M., Koller-Meier, E., Müller, P., Schraudolph, N.N., Gool, L.V.: Stochastic Optimization for High-Dimensional Tracking in Dense Range Maps. IEE Proceedings Vision, Image & Signal Processing 152(4), 501–512 (2005)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Desmond Chik
    • 1
    • 2
  • Jochen Trumpf
    • 1
    • 2
  • Nicol N. Schraudolph
    • 1
    • 2
  1. 1.Research School of Information Sciences and Engineering, Australian National University, Canberra ACT 0200Australia
  2. 2.Statistical Machine Learning, NICTA, Locked Bag 8001, Canberra ACT 2601Australia

Personalised recommendations