![](bg6.png)
the segmentation errors due to the Kalman filter ac-
tion. In fact the segmentation module outputs at 8 fps
while the tracking module can work even at a lower
data rate, thus a number of errors can be corrected.
As future work we intend to improve the integra-
tion between the two different types of background
and to add a series of zigbee sensors to the scene in
order to merge information coming from two different
sources: the stereo cameras and the zigbee sensors.
ACKNOWLEDGEMENTS
The authors thank Susan Rush for helping out in the
design and organization of the data acquisition exer-
cises.
REFERENCES
Bahadori, S., Iocchi, L., Leone, G. R., Nardi, D., and Scoz-
zafava, L. (2007). Real-time people localization and
tracking through fixed stereo vision. Applied Intelli-
gence, 26(2):83–97.
Darrell, T., Demirdjian, D., Checka, N., and Felzenszwalb,
P. F. (2001). Plan-view trajectory estimation with
dense stereo background models. In ICCV, pages
628–635.
Elgammal, A. M., Harwood, D., and Davis, L. S. (2000).
Non-parametric model for background subtraction. In
ECCV ’00: Proceedings of the 6th European Con-
ference on Computer Vision-Part II, pages 751–767,
London, UK. Springer-Verlag.
Friedman, N. and Russell, S. (1997). Image segmentation
in video sequences: A probabilistic approach. pages
175–181.
Gilbert, A. and Bowden, R. (2007). Multi person tracking
within crowded scenes. In Workshop on Human Mo-
tion, pages 166–179.
Halevi, G. and Weinshall, D. (1999). Motion of distur-
bances: Detection and tracking of multi-body non-
rigid motion. Machine Vision and Applications,
11:122–137.
Haritaoglu, I., Cutler, R., Harwood, D., and Davis, L. S.
(2001). Backpack: Detection of people carrying ob-
jects using silhouettes. Computer Vision and Image
Understanding: CVIU, 81(3):385–397.
Haritaoglu, I., Harwood, D., and Davis, L. S. (1999). Hy-
dra: Multiple people detection and tracking using sil-
houettes. In ICIAP ’99: Proceedings of the 10th In-
ternational Conference on Image Analysis and Pro-
cessing, pages 280–285, Washington, DC, USA. IEEE
Computer Society.
Heikkil¨a, J. and Silven, O. (1999). A real-time system for
monitoring of cyclists and pedestrians. Proc. Second
IEEE International Workshop on Visual Surveillance,
June 26, Fort Collins, Colorado, USA, 74-81.
Hu, W., Tieniu, T., Liang, W., and Maybank, S. (2004). A
survey on visual surveillance of object motion and be-
haviors. Systems, Man and Cybernetics, Part C, IEEE
Transactions on, 34(3):334–352.
Iocchi, L. and Bolles, R. C. (2005). Integrating plan-view
tracking and color-based person models for multiple
people tracking. In ICIP (3), pages 872–875.
Khan, S. M. and Shah, M. (2006). A multiview approach
to tracking people in crowded scenes using a planar
homography constraint. In ECCV (4), pages 133–146.
Konolige, K. (1997). Small vision systems: hardware and
implementation. In Eighth International Symposium
on Robotics Research, Hayama, Japan.
Stauffer, C. and Grimson, W. E. L. (2000). Learning pat-
terns of activity using real-time tracking. IEEE Trans.
Pattern Anal. Mach. Intell., 22(8):747–757.
Tsai, Y.-T., Shih, H.-C., and Huang, C.-L. (2006). Mul-
tiple human objects tracking in crowded scenes. In
ICPR ’06: Proceedings of the 18th International Con-
ference on Pattern Recognition, pages 51–54, Wash-
ington, DC, USA. IEEE Computer Society.
A NOVEL SEGMENTATION METHOD FOR CROWDED SCENES
489