REPRESENTATION AND RECOGNITION OF HUMAN ACTIONS - A New Approach based on an Optimal Control Motor Model

Sumitra Ganesh, Ruzena Bajcsy

2008

Abstract

We present a novel approach to the problem of representation and recognition of human actions, that uses an optimal control based model to connect the high-level goals of a human subject to the low-level movement trajectories captured by a computer vision system. These models quantify the high-level goals as a performance criterion or cost function which the human sensorimotor system optimizes by picking the control strategy that achieves the best possible performance. We show that the human body can be modeled as a hybrid linear system that can operate in one of several possible modes, where each mode corresponds to a particular high-level goal or cost function. The problem of action recognition, then is to infer the current mode of the system from observations of the movement trajectory. We demonstrate our approach on 3D visual data of human arm motion.

References

  1. Blom, H. A. P. and Bar-Shalom, Y. (1988). The interacting multiple model algorithm for systems with markovian switching coefficients. Automatic Control, IEEE Transactions on, 33(8):780-783.
  2. ference on Computer Vision and Pattern Recognition (CVPR), pages pp 568-674.
  3. de Freitas, N. (2002). Rao-blackwellised particle filtering for fault diagnosis. Aerospace Conference Proceedings, 2002. IEEE, 4.
  4. Del Vecchio, D., Murray, R., and Perona, P. (2003). Decomposition of human motion into dynamics based primitives with ap- plication to drawing tasks. Automatica, 39(12):2085-2098.
  5. Fod, A., Mataric, M., and Jenkins, O. (2002). Automated Derivation of Primitives for Movement Classification. Autonomous Robots, 12(1):39-54.
  6. Harris, C. and Wolpert, D. (1998). Signal-dependent noise determines motor planning. Nature, 394(6695):780- 4.
  7. Lewis, F. and Syrmos, V. (1995). Optimal Control. WileyInterscience.
  8. Li, W. and Todorov, E. (2004). Iterative linear-quadratic regulator design for nonlinear biological movement systems. First International Conference on Informatics in Control, Automation and Robotics, 1:222-229.
  9. Lien, J.-M., Kurillo, G., and Bajcsy, R. (2007). Skeletonbased data compression for multi-camera teleimmersion system. In Proceedings of the International Symposium on Visual Computing, Lake Tahoe, Nevada/California,Nov 2007, to appear.
  10. McGinnity, S. and Irwin, G. (2000). Multiple model bootstrap filter for maneuvering target tracking. Aerospace and Electronic Systems, IEEE Transactions on, 36(3):1006-1012.
  11. Murray, R., Sastry, S., and Li, Z. (1994). A Mathematical Introduction to Robotic Manipulation. CRC Press.
  12. Nori, F. and Frezza, R. (2005). Control of a manipulator with a minimum number of motion primitives. Proceedings of the 2005 IEEE International Conference on Robotics and Automation, 2005., pages 2344- 2349.
  13. Oliver, N., Garg, A., and Horvitz, E. (2004). Layered representations for learning and inferring office activity from multiple sensory channels. Computer Vision and Image Understanding, 96(2):163-180.
  14. Park, S. and Aggarwal, J. (2004). A hierarchical Bayesian network for event recognition of human actions and interactions. Multimedia Systems, 10(2):164-179.
  15. Pitt, M. K. and Shephard, N. (2001). Auxiliary variable based particle filters. In book Sequential Monte Carlo Methods in Practice, Arnaud Doucet - Nando de Freitas - Neil Gordon (eds). Springer-Verlag, 2001.
  16. Safonova, A., Hodgins, J., and Pollard, N. (2004). Synthesizing physically realistic human motion in lowdimensional, behavior-specific spaces. ACM Transactions on Graphics (TOG), 23(3):514-521.
  17. Scott, S. (2004). Optimal feedback control and the neural basis of volitional motor control. Nature Reviews Neuroscience, 5(7):532-546.
  18. Todorov, E. (2004). Optimality principles in sensorimotor control. Nature Neuroscience, 2004:907-915.
  19. Weinland, D., Ronfard, R., and Boyer, E. (2006). Automatic Discovery of Action Taxonomies from Multiple Views. Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition-Volume 2, pages 1639-1645.
Download


Paper Citation


in Harvard Style

Ganesh S. and Bajcsy R. (2008). REPRESENTATION AND RECOGNITION OF HUMAN ACTIONS - A New Approach based on an Optimal Control Motor Model . In Proceedings of the Third International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2008) ISBN 978-989-8111-21-0, pages 99-104. DOI: 10.5220/0001086700990104


in Bibtex Style

@conference{visapp08,
author={Sumitra Ganesh and Ruzena Bajcsy},
title={REPRESENTATION AND RECOGNITION OF HUMAN ACTIONS - A New Approach based on an Optimal Control Motor Model},
booktitle={Proceedings of the Third International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2008)},
year={2008},
pages={99-104},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001086700990104},
isbn={978-989-8111-21-0},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Third International Conference on Computer Vision Theory and Applications - Volume 1: VISAPP, (VISIGRAPP 2008)
TI - REPRESENTATION AND RECOGNITION OF HUMAN ACTIONS - A New Approach based on an Optimal Control Motor Model
SN - 978-989-8111-21-0
AU - Ganesh S.
AU - Bajcsy R.
PY - 2008
SP - 99
EP - 104
DO - 10.5220/0001086700990104