Advertisement

A Human Action Recognition Algorithm Based on Semi-supervised Kmeans Clustering

  • Hejin Yuan
  • Cuiru Wang
Chapter
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6758)

Abstract

This paper proposes a new method of semi-supervised human action recognition. In our approach, the motion energy image(MEI) and motion history image(MHI) are firstly used as the feature representation of the human action. Then, the constrained semi-supervised kmeans clustering algorithm is utilized to predict the class label of unlabeled training example. Meanwhile the average motion energy and history images are calculated as the recognition model for each category action. The category of the observed action is determined according to the correlation coefficients between its feature images and the pre-established average templates. The experiments on Weizmann dataset demonstrate that our method is effective and the average recognition accuracy can reach above 90% even when only using very small number of labeled action sequences.

Keywords

human action recognition semi-supervised learning kmeans clustering 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Du, Y., Chen, F., Xu, W., et al.: A Survey On The Vision-based Human Motion Recognition. J. Acta Electrionica Sinica. 35, 377–387 (2007)Google Scholar
  2. 2.
    Gu, J., Ding, X., Wang, S.: A Survey of Activity Analysis Algorithms. Journal of Image and Graphics 14, 976–990 (2009)Google Scholar
  3. 3.
    Popple, R.: A Survey On Vision-based Human Action Recognition. J. Image and Vision Computing. 28, 976–990 (2010)CrossRefGoogle Scholar
  4. 4.
    Yamato, J., Ohya, J., Ishii, K.: Recognizing Human Action In Time-sequential Images Using Hidden Markov Model. In: IEEE International Conference on Computer Vision and Pattern Recognition, pp. 379–385. IEEE Press, New York (1992)Google Scholar
  5. 5.
    Bobick, A.F., Davis, J.W.: The Recognition of Human Movement Using Temporal Templates. J. IEEE Trans. on Pattern Analysis and Machine Intelligence 25, 257–267 (2001)CrossRefGoogle Scholar
  6. 6.
    Wang, L., Suter, D.: Informative Shape Representations for Human Action Recognition. In: International Conference on Pattern Recognition, pp. 1266–1269. IEEE Press, Hong Kong (2006)Google Scholar
  7. 7.
    Weinland, D., Boyer, E.: Action Recognition Using Exemplar-based Embedding. In: IEEE International Conference on Computer Vision and Pattern Recognition, pp. 1–7. IEEE Press, Anchorage (2008)Google Scholar
  8. 8.
  9. 9.
    Basu, S., Banerjee, A., Mooney, R.: Semi-supervised Clustering By Seeding. In: International Conference on Machine Learning, pp. 19–26. IEEE Press, Sydney (2002)Google Scholar
  10. 10.
    Ali, S., Basharat, A., Shah, M.: Chaotic Invariants for Human Action Recognition. In: International Conference on Computer Vision, pp. 1–8. IEEE Press, Janeiro (2007)Google Scholar
  11. 11.
    Wang, L., Sute, D.: Recognizing Human Activities from Silhouettes: Motion Subspace and Factorial Discriminative Graphical Mode. In: IEEE International Conference on Computer Vision and Pattern Recognition, pp. 1–7. IEEE Press, Minneapolis (2007)Google Scholar
  12. 12.
    Niebles, J.C., Li, F.F.: A Hierarchical Model of Shape and Appearance for Human Action Classification. In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 1–8. IEEE Press, Minneapolis (2007)Google Scholar
  13. 13.
    Gorelick, L., Blank, M., Shechtman, E., et al.: Actions as Space-time Shapes. J. IEEE Trans on Pattern Analysis and Machine Intelligence 29, 2247–2253 (2007)CrossRefGoogle Scholar
  14. 14.
    Liu, C., Yuen, P.C.: Human Action Recognition Using Boosted Eigen Actions. J. Image and Vision Computing 28, 825–835 (2010)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Hejin Yuan
    • 1
  • Cuiru Wang
    • 1
  1. 1.Department of ComputerNorth China Electric Power UniversityBaodingChina

Personalised recommendations