A Fast Convergent Adaptive-K Mixture-Of-Gaussian Model for Video Object Segmentation

Abstract:

It is a key step to extract moving objects from background for computer vision applications. GMM based methods are the most commonly used technique for background subtraction in video sequence. However, how to establish efficient and precision background model with fast convergence rate is a Research-Worthy problem. In this paper, an effective scheme is proposed to accelerate the convergence rate of Adaptive-K Gaussian Mixture Model (AKGMM). The AKGMM algorithm alters the dimension of the parameter space at each pixel based on the changing frequency of pixel value. The number of GMM reflects the complexity of pattern at the pixel. An improved learning method is proposed for Gaussian Mixture Model. An adaptive learning rate is calculated for each Gaussian at every frame for speeding up the convergence without compromising model stability. Experimental results demonstrated that the proposed method gets a faster convergence while maintaining good robustness against complex environment compared to a conventional method.

• Periodical: Advanced Materials Research (Volumes 694-697)
• Pages: 1919-1924
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.694-697.1919
• Online since: May 2013
• Authors: Hao Zhou, Yun Gao, Guo Wu Yuan, Xue Jie Zhang
• Keywords: Adaptive-K Gaussian Mixture Model, Background Modelling, Fast Convergent, Motion Detection, Online EM

References

[1] R. Cucchiara, C. Grana, M. Piccardi, and A. Prati. Detecting moving objects, ghosts, and shadows in video streams. IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), 25(10):1337–1342, 2003.

DOI: 10.1109/tpami.2003.1233909

[2] Wren, C.R., Azarbayejani, A., Darrell, T., Pentland, A., 1997. Pfinder:Real-time tracking of the human body. IEEE Trans. Pattern Anal.Mach. Intell. 19 (7), 780-785

DOI: 10.1109/34.598236

[3] M.Harville, G.Gordon,and J.Woodfill, "Foreground Segmentation Using Adaptive Mixture Models in Color and Depth," Proc, ICCV Workshop Detection and Recognition of Events in Video, July 2001.

DOI: 10.1109/event.2001.938860

[4] Lee, D.-S., 2005. Effective Gaussian Mixture learning for video background subtraction. IEEE Trans. Pattern Anal. Mach. Intell. 27 (5),827–832.

DOI: 10.1109/tpami.2005.102

[5] Z. Zivkovic and F. V. der Heijden. Efficient adaptive density estimation per image pixel for the task of background subtraction. Pat. Rec. Letters, 27(7):773–780, 2006.

DOI: 10.1016/j.patrec.2005.11.005

[6] B.Han, D.Comaniciu et.al "Sequential Kernel Density Approximation through Mode Propagation: Applications to Background Modeling". Asian Conference on Computer Vision 2004, page818~823, (2004)

[7] Shimada, A.  Arita, D.  Taniguchi, R. Dynamic Control of Adaptive Mixture-of-Gaussians Background Model Proceedings of the IEEE International Conference on Video and Signal Based Surveillance,2006, AVSS'06

DOI: 10.1109/avss.2006.44

[8] J.Mike McHugh, et al. Foreground-Adaptive Background Subtraction. IEEE Signal Processing Letter, VOL. 16, No. 5, May (2009)

[9] Hao Zhou Xuejie Zhang Yun Gao Pengfei Yu. Video Background Subtraction Using Improved Adaptive-K Gaussian Mixture Model. 2010 3rd International Conference on Advanced Computer Theory and Engineering. (ICACTE 2010) pp: 363-366. August 20-22. 2010. Chengdu, China.

DOI: 10.1109/icacte.2010.5579536