Multi-Channel MKL for Video Human Action Recognition

Abstract:

Human action recognition in videos plays an important role in the field of computer vision and image understanding. A novel method of multi-channel bag of visual words and multiple kernel learning is proposed in this paper. The videos are described by multi-channel bag of visual words, and a multiple kernel learning classifier is used for action classification, in which each kernel function of the classifier corresponds to a video channel in order to avoid the noise interference from other channels. The proposed approach improves the ability in distinguishing easily confused actions. Experiments on KTH show that the presented method achieves remarkable performance on the average recognition rate, and obtains comparable recognition rate with state-of-the-art methods.

• Periodical: Applied Mechanics and Materials (Volumes 599-601)
• Pages: 1571-1574
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.599-601.1571
• Online since: August 2014
• Authors: Jia Ding*, Yang Yi, Ze Min Qiu, Jun Shi Liu
• Keywords: BoVW, Multiple Kernel Learning, Video Human Action Recognition

References

[1] I. Laptev, M. Marszalek, C. Schmid C, et al., Learning realistic human actions from movies, IEEE Conference on Computer Vision and Pattern Recognition, 2008, p.1–8.

DOI: 10.1109/cvpr.2008.4587756

[2] J. Sun, X. Wu, S. Yan, et al., Hierarchical spatio-temporal context modeling for action recognition, IEEE Conference on Computer Vision and Pattern Recognition, 2009, p.2004–(2011).

DOI: 10.1109/cvpr.2009.5206721

[3] S. Vishwanathan, Z. Sun, N. Theera-Ampornpunt N, et al., Multiple kernel learning and the SMO algorithm, Advances in Neural Information Processing Systems 23, 2010, p.2361–2369.

[4] B. Li, O. I. Camps, M. Sznaier, Cross-view activity recognition using Hankelets, IEEE Conference on Computer Vision and Pattern Recognition, 2012, p.1362–1369.

DOI: 10.1109/cvpr.2012.6247822

[5] S. Bhattacharya, R. Sukthankar, R. Jin, et al., A probabilistic representation for efficient large scale visual recognition tasks, IEEE Conference on Computer Vision and Pattern Recognition, 2011, p.2593–2600.

DOI: 10.1109/cvpr.2011.5995746

[6] A. Kovashka, K. Grauman, Learning a hierarchy of discriminative space-time neighborhood features for human action recognition, IEEE Conference on Computer Vision and Pattern Recognition, 2010, p.2046–(2053).

DOI: 10.1109/cvpr.2010.5539881

[7] J. Wang, Z. Chen, Y. Wu, Action recognition with multiscale spatio-temporal contexts, IEEE Conference on Computer Vision and Pattern Recognition, 2011, p.3185–3192.

DOI: 10.1109/cvpr.2011.5995493

[8] B. Chakraborty, M. B. Holte, T. B. Moeslund, et al., A selective spatio-temporal interest point detector for human action recognition in complex scenes, IEEE International Conference on Computer Vision, 2011, p.1776–1783.

DOI: 10.1109/iccv.2011.6126443

[9] L. Cao, Z. Liu, T. Huang, Cross-dataset action detection, IEEE Conference on Computer Vision and Pattern Recognition, 2010, p.1998–(2005).

DOI: 10.1109/cvpr.2010.5539875

[10] M. Kaâniche, F. Bremond F, Gesture recognition by learning local motion signatures, IEEE Conference on Computer Vision and Pattern Recognition, 2010, p.2745–2752.

DOI: 10.1109/cvpr.2010.5539999

[11] Z. Jiang, Z. Lin, L. S. Davis, Recognizing human actions by learning and matching shape-motion prototype trees, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(3): 533–547.

DOI: 10.1109/tpami.2011.147