Upper-Body Pose Estimation Using Geodesic Distances and Skin-Color

  • Sebastian Handrich
  • Ayoub Al-Hamadi
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8192)

Abstract

We propose a real-time capable method for human pose estimation from depth and color images that does not need any pre-trained pose classifiers. The pose estimation focuses on the upper body, as it is the relevant part for a subsequent gesture and posture recognition and therefore the basis for a real human-machine-interaction. Using a graph-based representation of the 3D point cloud, we compute geodesic distances between body parts. The geodesic distances are independent of pose and allow the robust determination of anatomical landmarks which serve as input to a skeleton fitting process using inverse kinematics. In case of degenerated graphs, landmarks are tracked locally with a meanshift algorithm based on skin color probability.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Jaeggli, T., Koller-Meier, E., Gool, L.: Learning generative models for multi-activity body pose estimation. IJCV 83(2), 121–134 (2009)CrossRefGoogle Scholar
  2. 2.
    Le Ly, D., Saxena, A., Lipson, H.: Pose estimation from a single depth image for arbitrary kinematic skeletons. CoRR, vol. abs/1106.5341 (2011)Google Scholar
  3. 3.
    Pons-Moll, G., Baak, A., Helten, T., Muller, M., Seidel, H.-P., Rosenhahn, B.: Multisensor-fusion for 3d full-body human motion capture. In: CVPR, pp. 663–670 (2010)Google Scholar
  4. 4.
    Srinivasan, K., Porkumaran, K., Sainarayanan, G.: Skin colour segmentation based 2d and 3d human pose modelling using discrete wavelet transform. Pattern Recognit. Image Anal. 21(4), 740–753 (2011)CrossRefGoogle Scholar
  5. 5.
    Liang, Q., Miao, Z.: Markerless human pose estimation using image features and extremal contour. In: ISPACS, pp. 1–4 (2010)Google Scholar
  6. 6.
    Chen, D.C.Y., Fookes, C.B.: Labelled silhouettes for human pose estimation. In: Int. C. on Inform. Science, Signal Proc. a their App. (2010)Google Scholar
  7. 7.
    Wang, Y., Qian, G.: Robust human pose recognition using unlabelled markers. Appl. of Comp. Vision, 1–7 (2008)Google Scholar
  8. 8.
    Soutschek, S., Penne, J., Hornegger, J., Kornhuber, J.: 3-d gesture-based scene navigation in medical imaging applications using time-of-flight cameras. In: CVPR Workshops, pp. 1–6 (2008)Google Scholar
  9. 9.
    Hu, R.Z.-L., Hartfiel, A., Tung, J., Fakih, A., Hoey, J., Poupart, P.: 3d pose tracking of walker users’ lower limb with a structured-light camera on a moving platform. In: CVPRW, pp. 29–36 (2011)Google Scholar
  10. 10.
    Schwarz, L.A., Mkhitaryan, A., Mateus, D., Navab, N.: Estimating human 3d pose from time-of-flight images based on geodesic distances and optical flow. In: IEEE Automatic Face Gesture Recog. a. WS, pp. 700–706 (2011)Google Scholar
  11. 11.
    Jones, M.J., Rehg, J.M.: Statistical color models with application to skin detection. Int. J. Comput. Vision 46(1), 81–96 (2002)CrossRefMATHGoogle Scholar
  12. 12.
    Wang, L.-C.T., Chen, C.C.: A combined optimization method for solving the inverse kinematics problems of mechanical manipulators. Robotics and Automation 7(4), 489–499 (1991)MathSciNetCrossRefGoogle Scholar
  13. 13.
    Handrich, S., Al-Hamadi, A.: Multi hypotheses based object tracking in hci environments. In: ICIP, Orlando, USA, pp. 1981–1984 (2012)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Sebastian Handrich
    • 1
  • Ayoub Al-Hamadi
    • 1
  1. 1.Institute of Information Technology and CommunicationsOtto-von-Guericke-University MagdeburgGermany

Personalised recommendations