Multimodal Mean Adaptive Background Modeling
The availability of low-cost, portable imagers and new embedded
computing platforms makes video surveillance possible in new
environments. However, situations in which a portable, embedded video
surveillance system is most useful (e.g., monitoring outdoor and/or
busy scenes) also pose the greatest challenges. Real-world scenes are
characterized by changing illumination and shadows, multimodal
features (such as rippling waves and rustling leaves), and frequent,
multilevel occlusions. To extract foreground in these dynamic visual
environments, adaptive multimodal background models are
frequently used that maintain historical scene information to improve
accuracy. These methods are problematic in real-time embedded
environments where limited computation and storage restrict the amount
of historical data that can be processed and stored.

Results of MM Background Subtraction on 3 Video Sequences:
Waving Trees and
Bootstrapping (from the
Wallflower benchmark images) and an Outdoor Sequence.
See our ECVW07 paper for comparison with other backgrounding techniques.
We have developed a new adaptive technique, multimodal mean
(MM), which balances accuracy, performance, and efficiency to meet
embedded system requirements. This algorithm delivers comparable
accuracy of the best alternative (Mixture of Gaussians) with a 6X
improvement in execution time and an 18% reduction in required storage
on an
eBox-2300 Thin Client VESA PC running Windows Embedded CE 6.0.

eBox-2300 Thin Client VESA PC
Publications:
-
B. Valentine, S. Apewokin, L. M. Wills and S. Wills, An Efficient, Chromatic Clustering-Based Background Model for Embedded Vision Platforms, to appear Journal of Computer Vision and Image Understanding, Elsevier, 2010.
-
S. Apewokin, B. Valentine, J. Choi, L. M. Wills and S. Wills, Real-Time Adaptive Background Modeling for Multicore Embedded Systems, Journal of Signal Processing Systems for Signal, Image, and Video Technology, Springer, online January 2009.
-
S. Apewokin, B. Valentine, D. Forsthoefel, S. Wills, L. M. Wills and A. Gentile, Embedded Real-Time Surveillance Using Multimodal Mean Background Modeling, Embedded Computer Vision, B. Kisacanin and S. Bhattacharyya and S. Chai(eds.), Springer, pp. 163-175, peer-reviewed, January 2009.
-
B. Valentine, J. Choi, S. Apewokin, L. M. Wills and S. Wills, Bypassing BigBackground: An Efficient Hybrid Background Modelling Algorithm for Embedded Video Surveillance, Proceedings of the Second ACM/IEEE International Conference on Distributed Smart Cameras (ICDSC-08), pp. 1-8, Stanford University, CA, September 2008.
-
J. Choi, S. Apewokin, B. Valentine, S. Wills and L. M. Wills, Edge Noise Removal in Multimodal Background Modeling Techniques, Proceedings of the SPIE - Image Processing: Machine Vision Applications, Vol. 6813, No. 1, Kurt S. Niel and David Fofi(eds.), pp. 1-7, San Jose, California, March 2008.
-
S. Apewokin, B. Valentine, S. Wills, L. M. Wills and A. Gentile, Multimodal Mean Adaptive Backgrounding for Embedded Real-Time Video Surveillance, Proceedings of the Embedded Computer Vision Workshop (ECVW07), pp. 1-6, Minneapolis, Minnesota, held in conjunction with the 2007 IEEE Conference on Computer Vision and Pattern Recognition (CVPR-2007), June 2007.