Abstract
This paper presents a new classification method for single person’s motion, which is represented by ℜ transform descriptor and classified by Linear-chain Conditional Random Fields. What it solves is that the global features are described and the states in models are less dependent on adjacent ones. We extract binary silhouette and segment them by cycle after creating the background model. Then the low-level features are described by ℜ transform and principal vectors are determined by Principal Component Analysis. We utilize linear-chain conditional random fields to train and classify cycle sequences, and demonstrate the usability. Compared with others, our approach is simple in motion representation and independent between adjacent frames, as the advantages of ℜ transform descriptor lying in computational complexity, geometric invariance and classification performance, and linear-chain conditional random fields in states independency. So the video surveillance based on these is practicable in (but not limited to) many scenarios where the background is known.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Aggarwal, J.K., Cai, Q.: Human motion analysis: A review. Computer Vision and Image Understanding 73(3), 90–102 (1999)
Moeslund, T.B., Hilton, A., Krger, V.: A survey of advances in vision-based human motion capture and analysis. Computer Vision and Image Understanding 104(2-3), 90–126 (2006)
Turaga, P., Chellappa, R., Subrahmanian, V.S., Udrea, O.: Machine recognition of human activities: A survey. IEEE Trans. Circuits Syst. Video Technol. 18(11), 1473–1488 (2008)
Poppe, R.: A survey on vision-based human action recognition. Image and Vision Computing 28(6), 976–990 (2010)
Lafferty, J., McCallum, A., Pereira, F.: Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In: Proc. of Internat. Conf. on Machine Learning (2001)
Sminchisescu, C., Kanaujia, A., Metaxas, D.: Conditional models for contextual human motion recognition. Computer Vision and Image Understanding 104(2-3), 210–220 (2006)
Huang, T., Shi, C., Li, F.: Discriminative random fields for behavior modeling. In: Proc. of 2009 WRI World Congress on Computer Science and Information Engineering, pp. 17–21 (2009)
Deans, S.R.: Applications of the Radon Transform. Wiley Interscience Publications, Chichester (1983)
Tabbone, S., Wendling, L., Salmon, J.-P.: A new shape descriptor defined on the Radon transform. Computer Vision and Image Understanding 102(1), 42–51 (2006)
Wang, Y., Huang, K., Tan, T.: Human activity recognition based on Transform. In: Proc. of IEEE Comput. Soc. Conf. Comput. Vision Pattern Recognit., pp. 3722–3729 (2007)
Zhang, H., Liu, Z.: Automated gait recognition using weighted DTW distance. Journal of Image and Graphics 15(5), 830–836 (2010)
Zhang, H., Liu, Z., Zhao, H., Cheng, G.: Key-frame based human activity recognition. In: WASE Global Congress on Science Engineering (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wei, Q., Zhang, H., Zhao, H., Liu, Z. (2010). Human Motion Classification Using ℜ Transform. In: Zhu, R., Zhang, Y., Liu, B., Liu, C. (eds) Information Computing and Applications. ICICA 2010. Communications in Computer and Information Science, vol 106. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-16339-5_22
Download citation
DOI: https://doi.org/10.1007/978-3-642-16339-5_22
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-16338-8
Online ISBN: 978-3-642-16339-5
eBook Packages: Computer ScienceComputer Science (R0)