Distributed Smart Cameras and Distributed Computer Vision - Signal Processing Systems - page 469

8 Summary

Distributed algorithms offer a new approach to computer vision. Nodes in a
distributed system communicate explicitly through mechanisms that have non-zero
cost in delay and energy. This constraint encourages us to consider algorithms that
refine the representation of a scene in stages with predictable ways of combining
data within a stage. Distributed algorithms are, in general, more difficult to develop
and prove correct, but they also provide advantages such as robustness. Distributed
signal processing may require new statistical representations of signals that can
be efficiently shared over the network while preserving the useful properties of
the underling signal. We also need a better understanding of the accuracy of
the approximations underlying distributed signal processing algorithms—how do
termination conditions affect the accuracy of the estimate, for example.

Acknowledgements This work was supported in part by the National Science Foundation under

grant 0720536.

References

1. de Agapito, L., Hartley, R., Hayman, E.: Linear self-calibration of a rotating and zooming cam-
era. In: Computer Vision and Pattern Recognition, 1999. IEEE Computer Society Conference
on., vol. 1, pp. 2 vol. (xxiii+637+663) (1999). DOI 10.1109/CVPR.1999.786911
2. den Bergh, M.V., Koller-Meier, E., Kehl, R., Gool, L.V.: Real-time 3d body pose estimation.
In: H. Aghajan, A. Cavallaro (eds.) Multi-Camera Networks: Principles and Applications,
chap. 14. Academic Press (2009)
3. Bimbo, A.D., Dini, F., Pernici, F., Grifoni, A.: Pan-tilt-zoom camera networks. In: H. Aghajan,
A. Cavallaro (eds.) Multi-Camera Networks: Principles and Applications, chap. 8. Academic
Press (2009)
4. Boykov, V., Huttenlocher, D.: Adaptive bayesian recognition in tracking rigid objects. In:
Proceedings, IEEE Conference on Computer Vision and Pattern Recognition, pp. 697-704.
IEEE (2000)
5. Bramberger, M., Quaritsch, M., Winkler, T., Rinner, B., Schwabach, H.: Integrating multi-
camera tracking into a dynamic task allocation system for smart cameras. In: Proceedings
of the IEEE Conference on Advanced Video and Signal Based Surveillance (AVSS 2005),
pp. 474-479. IEEE (2005)
6. Candy, J.V.: Boostrap particle filtering. IEEE Signal Processing Magazine 73 , 73-85 (2007)
7. Coates, M.: Distributed particle filters for sensor networks. In: Information Processing in
Sensor Networks, 2004. IPSN 2004. Third International Symposium on, pp. 99-107. IEEE
(2004)
8. Collins, R.T., Lipton, A.J., Fujiyoshi, H., Kanade, T.: Algorithms for cooperative multisensory
surveillance. Proceedings of the IEEE 89 (10), 1456-1477 (2001)
9. Fleuret, F., Berclaz, J., Lengagne, R., Fua, P.: Multicamera people tracking with a probabilistic
occupancy map. Pattern Analysis and Machine Intelligence, IEEE Transactions on 30 (2),
267-282 (2008). DOI 10.1109/TPAMI.2007.1174
10. Hartley, R.I., Zisserman, A.: Multiple View Geometry in Computer Vision, second edn.
Cambridge University Press, ISBN: 0521540518 (2004)