Skip to main content

Advertisement

SpringerLink
Log in

Multi-view transition HMMs based view-invariant human action recognition method

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

View-invariant human action recognition is a challenging research topic in computer vision. Hidden Markov Models(HMM) and their extensions have been widely used for view-invariant action recognition. However those methods are usually according to a large parameter space, requiring amounts of training data and with low classification accuracies for real application. A novel graphical structure based on HMM with multi-view transition is proposed to model the human action with viewpoint changing. The model consists of multiple sub action models, which correspond to the traditional HMM utilized to model the human action in a particular rotation viewpoint space. In the training process, the novel model can be built by connecting the sub action models between adjacent viewpoint spaces. In the recognition process, action with unknown viewpoint is recognized by using improved forward algorithm. The proposed model can not only simplify the model training process by decomposing the parameter space into multiple sub-spaces, but also improve the performance the algorithm by constraining the possible viewpoint changing. Experiment results on IXMAS dataset demonstrated that the proposed model obtains better performance than other recent view-invariant action recognition method.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

References

  1. Ahmad M., Lee S (2006) Hmm-based human action recognition using multiview image sequences. In: the 18th IEEE int. conf. pattern recognition (ICPR), pp 263–266

  2. Ashraf N, Sun C, Foroosh H (2014) View invariant action recognition using projective depth. Int J Comput Vis Image Underst 123:41–52

    Article  Google Scholar 

  3. Bregonzio M, Gong S, Xiang T (2009) Recognising action as clouds of space-time interest points. In: the 27th IEEE conf. computer vision and pattern recognition (CVPR). USA, pp 1948–1955

  4. Holte M, Moeslund T (2011) Human action recognition using multiple views: a comparative perspective on recent developments. In: the 2011 joint ACM workshop on human gesture and behavior understanding, pp 47–52

  5. Ji X, Liu H (2010) Advances in view-invariant human motion analysis: a review. IEEE Trans Syst Man Cybern Part C 1:13–24

    MathSciNet  Google Scholar 

  6. Ji X, Wang C, Li Y (2014) A view-invariant action recognition based on multi-view space hidden markov models. Int J Humanoid Rob 11(1):1–17

    Google Scholar 

  7. Junejo I, Dexter E, Laptev I, Perez P (2008) Cross-view action recognition from temporal self-similarities. In: the 10th European conf. computer vision. France, pp 293–306

  8. Lee AR, Suk HI, Lee SW (2014) View-invariant 3D action recognition using spatiotemporal self-similarities from depth camera. In: the 22nd int. conf. pattern recognition (ICPR). Sweden, pp 501–505

  9. Lim C, Vats E, Chan C (2015) Fuzzy human motion analysis: a review. Pattern Recogn 48(5):1773–1796

    Article  Google Scholar 

  10. Liu J, Shah M, Kuipers B, Savarese S (2011) Cross-view action recognition via view knowledge transfer. In: the 29th IEEE conf. computer vision and pattern recognition (CVPR). USA, pp 3209–3216

  11. Lv F, Nevatia R (2006) Recognition and segmentation of 3-d human action using hmm and multi-class adaboost. In: the 9th European conf. computer vision (ECCV). Austria, pp 359–372

  12. Lv F, Nevatia R (2007) Single view human action recognition using key pose matching and Viterbi path searching. In: the 25th IEEE int. conf. computer vision and pattern recognition (CVPR). USA, pp 1–8

  13. Natarajan P, Nevatia R (2008) View and scale invariant action recognition using multiview shape-flow models. In: the 26th IEEE int. conf. computer vision and pattern recognition (CVPR), pp 1–8

  14. Niebles JC, Wang H, Feifei L (2008) Unsupervised learnling of human action categories using spatial-temporal words. Int J Comput Vis 79:299–318

    Article  Google Scholar 

  15. Peng B, Qian G, Rajko S (2008) View-invariant full-body gesture recognition from video. In: the 19th int. conf. pattern recognition (ICPR). USA, pp 1–5

  16. Peursum P, Venkatesh S, West G (2007) Tracking as recognition for articulated full body human motion analysis. In: the 25th IEEE int. conf. computer vision and pattern recognition (CVPR). USA, pp 1–8

  17. Poppe R (2010) A survey on vision-based human action recognition. Image Vis Comput 28(6):976–990

    Article  Google Scholar 

  18. Rogez G, Guerrero JJ, Martinez J, Orrite C (2006) Viewpoint independent human motion analysis in man-made environments. In: the 17th British machine vision conference (BWVC). UK, pp 659–668

  19. Singh VK, Nevatia R (2011) Simultaneous tracking and action recognition for single actor human actions. IEEE Trans Vis Comput Graph 27(12):1115–1123

    Google Scholar 

  20. Tran D, Sorokin A (2008) Human activity recognition with metric learning. In: the 10th European conf. computer vision (ECCV). France, pp 548–561

  21. Weinland D, Ronfard R, Boyer E (2006) Free viewpoint action recognition using motion history volumes. Comput Vis Image Underst 104(2):249–257

    Article  Google Scholar 

  22. Weinland D, Ozuysal M, Fua P (2010) Making action recognition robust to occlusions and viewpoint changes. In: the 10th European conf. computer vision. Greece, pp 635–648

  23. Wu X, Xu D, Duan L, Luo J (2011) Action recognition using context and appearance distribution features. In: IEEE int. conf. on computer vision and pattern recognition (CVPR). USA, pp 489–496

  24. Wu D, Shao L (2014) Multi-max-margin support vector machine for multi-source human action recognition. Neurocomputing 127:98–103

    Article  Google Scholar 

  25. Yan Y, Ricci E, Subramanian R, Liu GW, Sebe N (2014) Multitask linear discriminant analysis for view invariant action recognition. IEEE Trans Image Process 23(12):5599–5611

    Article  MathSciNet  Google Scholar 

  26. Yilmaz A, Shah M (2005) Actions sketch: a novel action representation. In: the 23th IEEE int. conf. computer vision and pattern recognition (CVPR). USA, pp 984–989

Download references

Acknowledgments

The Project supported by the National Natural Science Foundation of China No. 61103123 and the Program for Liaoning Excellent Talents in University (No. LJQ2014018).

Author information

Authors and Affiliations

  1. School of Automation, Shenyang Aerospace University, Shenyang, China

    Xiaofei Ji, Ce Wang & Changhui Wang

  2. School of Computing, University of Portsmouth, Portsmouth, UK

    Zhaojie Ju

Authors
  1. Xiaofei Ji
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Zhaojie Ju
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ce Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Changhui Wang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Xiaofei Ji.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ji, X., Ju, Z., Wang, C. et al. Multi-view transition HMMs based view-invariant human action recognition method. Multimed Tools Appl 75, 11847–11864 (2016). https://doi.org/10.1007/s11042-015-2661-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-015-2661-y

Keywords

Search

Navigation