A Co-Training Based Semi-Supervised Human Action Recognition Algorithm

Abstract:

A novel semi-supervised algorithm based on co-training is proposed in this paper. In the method, the motion energy history image are used as the different feature representation of human action; then the co-training based semi-supervised learning algorithm is utilized to predict the category of unlabeled training examples. And the average motion energy and history images are calculated as the recognition model for each category action. When recognition, the observed action is firstly classified through its correlation coefficients to the prior established templates respectively; then its final category is determined according to the consistency between the classification results of motion energy and motion history images. The experiments on Weizmann dataset demonstrate that our method is effective for human action recognition.

• Periodical: Advanced Materials Research (Volume 267)
• Pages: 1065-1070
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.267.1065
• Online since: June 2011
• Authors: He Jin Yuan, Cui Ru Wang, Jun Liu
• Keywords: Co-Training, Human Action Recognition, Semi-Supervised Learning

References

[1] Dj.M. Maric, P.F. Meier and S.K. Estreicher: Mater. Sci. Forum Vol. 83-87 (1992), p.119 Y. Du, F. Chen, W. Xu, et al: A Survey on the vision-based human motion recognition. Acta Electrionica Sinica, Vol. 35(2007), p.84.

[2] J. Gu, X. Ding, S. Wang: A survey of activity analysis algorithms. Journal of Image and Graphics, Vol. 14(2009), p.377.

[3] R. Popple: A Survey on Vision-Based on Human Action Recognition. Image and Vision Computing, Vol. 28 (2010)6, p.976.

DOI: 10.1016/j.imavis.2009.11.014

[4] J. Yamato, J. Ohya, K. Ishii: Recognizing Human Action in Time-Sequential Images Using Hidden Markov Model. in: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. IEEE Computer Society Publications (1992).

DOI: 10.1109/cvpr.1992.223161

[5] A. F. Bobick, J. W. Davis: The Recognition of Human Movement Using Temporal Templates. IEEE Trans on Pattern Analysis and Machine Intelligence, Vol. 25(2001)3, p.257.

DOI: 10.1109/34.910878

[6] L. Wang, D. Suter: Informative Shape Representations for Human Action Recognition. in: Proceedings of International Conference on Pattern Recognition. IEEE Computer Society Publications (2006).

DOI: 10.1109/icpr.2006.711

[7] H. Meng, N Pears: Descriptive Temporal Templates Features for Visual Motion Recognition. Pattern Recognition Letters, Vol. 30(2009), p.1049.

DOI: 10.1016/j.patrec.2009.03.003

[8] M. Ahmad, S. Lee: Variable Silhouette Energy Image Representations for Recognizing Human Actions. Image and Vision Computing, Vol. 28(2010), p.814.

DOI: 10.1016/j.imavis.2009.09.018

[9] D. Weinland, E. Boyer: Action Recognition Using Exemplar-Based Embedding. in: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society Publications (2008).

DOI: 10.1109/cvpr.2008.4587731

[10] http: /www. wisdom. weizmann. ac. il/~vision/SpaceTimeActions. html.

[11] A. Bluma, T. Mitehell: Combining Labeled and Unlabeled Data with Co-training. in: Proeeedings of the1lth Annual Conference on Leaming Theory (1998).

[12] S. Ali, A. Basharat, M. Shah: Chaotic Invariants for Human Action Recognition. in: Proceedings of International Conference on Computer Vision. IEEE Computer Society Press(2007).

DOI: 10.1109/iccv.2007.4409046

[13] L. Wang, D. Suter: Recognizing Human Activities From Silhouettes: Motion Subspace and . (1)nalysisard brackets (meV)hina Electric Power University, ����������������������������������������������������������������Factorial Discriminative Graphical Mode. in: Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. IEEE Computer Society Press(2007).

DOI: 10.1109/cvpr.2007.383298