Deeply Optimized Hough Transform: Application to Action Segmentation

  • Adrien Chan-Hon-Tong
  • Catherine Achard
  • Laurent Lucat
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8156)

Abstract

Hough-like methods like Implicit Shape Model (ISM) and Hough forest have been successfully applied in multiple computer vision fields like object detection, tracking, skeleton extraction or human action detection. However, these methods are known to generate false positives. To handle this issue, several works like Max-Margin Hough Transform (MMHT) or Implicit Shape Kernel (ISK) have reported significant performance improvements by adding discriminative parameters to the generative ones introduced by ISM. In this paper, we offer to use only discriminative parameters that are globally optimized according to all the variables of the Hough transform. To this end, we abstract the common vote process of all Hough methods into linear equations, leading to a training formulation that can be solved using linear programming solvers. Our new Hough Transform significantly outperforms the previous ones on HoneyBee and TUM datasets, two public databases of action and behaviour segmentation.

Keywords

Hough Transform Learning Action Segmentation 

References

  1. 1.
    Ballard, D.: Generalizing the Hough transform to detect arbitrary shapes. Pattern Recognition (1981)Google Scholar
  2. 2.
    Chakrabartty, S., Cauwenberghs, G.: Gini support vector machine: Quadratic entropy based robust multi-class probability regression. Journal of Machine Learning (2007)Google Scholar
  3. 3.
    Chan-Hon-Tong, A., Ballas, N., Achard, C., Delezoide, B., Lucat, L., Sayd, P., Coise Prêteux, F.: Skeleton point trajectories for human daily activity recognition. In: International Conference on Computer Vision Theory and Application (2013)Google Scholar
  4. 4.
    Cortes, C., Vapnik, V.: Support-vector networks. Journal of Machine Learning (1995)Google Scholar
  5. 5.
    Gall, J., Lempitsky, V.: Class-specific Hough forests for object detection. In: Internationnal Conference on Computer Vision and Pattern Recognition (2009)Google Scholar
  6. 6.
    Gall, J., Yao, A., Razavi, N., Van Gool, L., Lempitsky, V.: Hough forests for object detection, tracking, and action recognition. Pattern Analysis and Machine Intelligence (2011)Google Scholar
  7. 7.
    Girshick, R., Shotton, J., Kohli, P., Criminisi, A., Fitzgibbon, A.: Efficient regression of general-activity human poses from depth images. In: Internationnal Conference on Computer Vision and Pattern Recognition (2011)Google Scholar
  8. 8.
    Hoai, M., Lan, Z., De la Torre, F.: Joint segmentation and classification of human actions in video. In: Internationnal Conference on Computer Vision and Pattern Recognition (2011)Google Scholar
  9. 9.
    Hough, P.V.: Method and means for recognizing complex patterns. Tech. rep. (1962)Google Scholar
  10. 10.
    Karmarkar, N.: A new polynomial-time algorithm for linear programming. In: Theory of Computing (1984)Google Scholar
  11. 11.
    Leibe, B., Leonardis, A., Schiele, B.: Combined object categorization and segmentation with an implicit shape model. In: Workshop on Statistical Learning in Computer Vision (2004)Google Scholar
  12. 12.
    Sreedevi, M., Jeno Paul, P.: An efficient image segmentation using Hough transformation. Asian Journal of Information Technology (2011)Google Scholar
  13. 13.
    Maji, S., Malik, J.: Object detection using a max-margin Hough transform. In: Internationnal Conference on Computer Vision and Pattern Recognition (2009)Google Scholar
  14. 14.
    Oh, S., Rehg, J., Balch, T., Dellaert, F.: Learning and inferring motion patterns using parametric segmental switching linear dynamic systems. Journal of Computer Vision (2008)Google Scholar
  15. 15.
    Sun, J., Wu, X., Yan, S., Cheong, L., Chua, T., Li, J.: Hierarchical spatio-temporal context modeling for action recognition. In: Internationnal Conference on Computer Vision and Pattern Recognition (2009)Google Scholar
  16. 16.
    Tenorth, M., Bandouch, J., Beetz, M.: The tum kitchen data set of everyday manipulation activities for motion tracking and action recognition. In: International Conference on Computer Vision Workshops (2009)Google Scholar
  17. 17.
    Wang, H., Kläser, A., Schmid, C., Cheng-Lin, L.: Action Recognition by Dense Trajectories. In: Internationnal Conference on Computer Vision and Pattern Recognition (2011)Google Scholar
  18. 18.
    Wohlhart, P., Schulter, S., Kostinger, M., Roth, P., Bischof, H.: Discriminative Hough forests for object detection. In: British Machine Vision Conference (2012)Google Scholar
  19. 19.
    Yao, A., Gall, J., Van Gool, L.: A Hough transform-based voting framework for action recognition. In: Internationnal Conference on Computer Vision and Pattern Recognition (2010)Google Scholar
  20. 20.
    Yao, A., Gall, J., Fanelli, G., Van Gool, L.: Does human action recognition benefit from pose estimation? In: British Machine Vision Conference (2011)Google Scholar
  21. 21.
    Zhang, Y., Chen, T.: Implicit shape kernel for discriminative learning of the Hough transform detector. In: British Machine Vision Conference (2010)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Adrien Chan-Hon-Tong
    • 1
  • Catherine Achard
    • 2
  • Laurent Lucat
    • 1
  1. 1.LIST, DIASI, Laboratoire Vision et Ingénierie des ContenusCEAFrance
  2. 2.Institut des Systèmes Intelligents et RobotiqueUPMCFrance

Personalised recommendations