Generalized Multi-sensor Planning
Vision systems for various tasks are increasingly being deployed. Although significant effort has gone into improving the algorithms for such tasks, there has been relatively little work on determining optimal sensor configurations. This paper addresses this need. We specifically address and enhance the state-of-the-art in the analysis of scenarios where there are dynamically occuring objects capable of occluding each other. The visibility constraints for such scenarios are analyzed in a multi-camera setting. Also analyzed are other static constraints such as image resolution and field-of-view, and algorithmic requirements such as stereo reconstruction, face detection and background appearance. Theoretical analysis with the proper integration of such visibility and static constraints leads to a generic framework for sensor planning, which can then be customized for a particular task. Our analysis can be applied to a variety of applications, especially those involving randomly occuring objects, and include surveillance and industrial automation. Several examples illustrate the wide applicability of the approach.
KeywordsStatic Constraint Quality Function Face Detection Soft Constraint Stereo Match
Unable to display preview. Download preview PDF.
- 2.Cameron, A., Durrant-Whyte, H.F.: A bayesian approach to optimal sensor placement. IJRR 9(5), 70–88 (1990)Google Scholar
- 3.Chen, X., Davis, J.: Camera placement considering occlusion for robust motion capture. Technical Report CS-TR-2000-07, Stanford University (December 2000)Google Scholar
- 7.González-Banos, H., Latombe, J.C.: A randomized art-gallery algorithm for sensor placement. In: SCG, Medford, MA (June 2001)Google Scholar
- 8.Hager, G., Mintz, M.: Computational methods for task-directed sensor data fusion and sensor planning. IJRR 10(4), 285–313 (1991)Google Scholar
- 10.Kang, S.B., Seitz, S.M., Sloan, P.P.: Visual tunnel analysis for visibility prediction and camera planning. In: CVPR, Hilton Head, SC, June 2000, pp. II:195–202 (2000)Google Scholar
- 16.Mittal, A., Davis, L.S.: A general method for sensor planning in multi-sensor systems: Extension to random occlusion. IJCV (submitted, 2005)Google Scholar
- 17.Miura, J., Ikeuchi, K.: Task-oriented generation of visual sensing strategies. In: ICCV, Boston, MA, pp. 1106–1113 (1995)Google Scholar
- 19.Rahimi, A., Dunagan, B., Darrell, T.J.: Simultaneous calibration and tracking with a network of non-overlapping sensors. In: CVPR, pp. I:187–194 (2004)Google Scholar
- 21.Shang, Y.: Global Search Methods for Solving Nonlinear Optimization Problems. PhD thesis, University of Illinois at Urbana-Champaign (1997)Google Scholar
- 22.Spletzer, J., Taylor, C.J.: A framework for sensor planning and control with applications to vision guided multi-robot systems. In: CVPR, Kauai, Hawaii (2001)Google Scholar
- 25.Ye, Y., Tsotsos, J.K.: Sensor planning for 3d object search. CVIU 73(2), 145–168 (1999)Google Scholar
- 26.Yi, S.K., Haralick, R.M., Shapiro, L.G.: Optimal sensor and light-source positioning for machine vision. CVIU 61(1), 122–137 (1995)Google Scholar