Study on the Generation Model of Weighted Visual Codebook for Action Recognition

Abstract:

The action recognition based on local spatial-temporal feature has attracted more and more attentions. The codebook used in the recognition is usually generated by k-means and the code words are considered to have the same importance. But in fact, it isn’t true. So in this paper, we propose two strategies to measure the importance of different code words and improve the HIK-Kmeans algorithm .The experiment result shows that our improved algorithm increases the accuracy of action recognition.

• Periodical: Applied Mechanics and Materials (Volumes 333-335)
• Pages: 944-947
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.333-335.944
• Online since: July 2013
• Authors: Xian Ming Lin, Shao Zi Li, Dong Lin Cao, Wei Yuan Zhuang
• Keywords: Action Recognition, HIK Clustering, Weighted Visual Codebook

References

[1] J. K. Aggarwal and M. S. Ryoo1. Human Activity Analysis: A Review. [C] ACM Computing Surveys (CSUR), 43(3), April (2011).

[2] I. Laptev. On space-time interest points. [J] International Journal of Computer Vision, 2005, 64(2): 107-123.

[3] P. Dollar, V. Rabaud, G. Cottrell, et al. Behavior recognition via sparse spatio-temporal features. [C] IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance, (2005).

DOI: 10.1109/vspets.2005.1570899

[4] A. Oikonomopoulos, I. Patras, and M. Pantic. Spatio-temporal salient points for visual recognitoin of human actions. [J] IEEE Trans. Systems, Man and Cybernetics, Part B, 36(3): 710-719, (2006).

DOI: 10.1109/tsmcb.2005.861864

[5] P. Scovanner,S. Ali, and M. Shah. A 3-dimensional SIFT descriptor and its application to action recognition. [C] In ACM International Conference on Multimedia, (2007).

DOI: 10.1145/1291233.1291311

[6] G. Willems, T. Tuytelaars and L. V. Gool. An efficient dense and scale-invariant spatio-temporal interest point detector. [C] European Conference on Computer Vision, (2008).

DOI: 10.1007/978-3-540-88688-4_48

[7] I. Laptev, B. Caputo, C. Schuldt, et al. Local velocity-adapted motion events for spatio-temporal recognition. [J] Computer Vision and Image Understanding, 2007, 108: 207-229.

DOI: 10.1016/j.cviu.2006.11.023

[8] A. Klaser, M. Marszalek and C. Schmid. A spatio-temporal descriptor based on 3D-gradients. [C] British Machine Vision Conference, (2008).

DOI: 10.5244/c.22.99

[9] C. Schuldt, I. Laptev and B. Caputo. Recognizing human actions: A local SVM approach. [C] Internation Conference on Pattern Recognition, (2004).

DOI: 10.1109/icpr.2004.1334462

[10] S. Maji, A. C. Berg, and J. Malik. Classification using intersection kernel support vector machines is efficient. In CVPR, (2008).

DOI: 10.1109/cvpr.2008.4587630

[11] I. Laptev, M. Marszalek, C. Schmid, et al. Learning realistic human actions from movies. [C] Computer Vision and Pattern Recognition, (2008).

DOI: 10.1109/cvpr.2008.4587756

[12] M. J. Swain and D. H. Ballard. Color indexing. [J] International Journal of Computer Vision, 7(1): 11–32, (1991).

[13] F. Odone, A. Barla, and A. Verri. Building kernels from binary strings for image matching. [J] IEEE Trans. Image Processing, 14(2): 169–180, (2005).

DOI: 10.1109/tip.2004.840701

[14] Jianxin Wu , James M. Rehg. Beyond the Euclidean distance: Creating effective visual codebooks using the histogram intersection kernel. [C] 2009 IEEE 12th International Conference on Computer Vision. pp.630-637 , (2009).

DOI: 10.1109/iccv.2009.5459178

[15] F. Jurie and B. Triggs. Creating efficient codebooks for visual recognition. [C] In ICCV, volume 1, pages 604–610, (2005).

DOI: 10.1109/iccv.2005.66