Global Pose Estimation of Multiple Cameras with Particle Filters
Though image processing algorithms are sophisticated and provided as software libraries, it is still difficult to assure that complicated programs can work in various situations. In this paper, we propose a novel global pose estimation method for network cameras to actualize auto-calibration. This method uses native information from images. The sets of partial information are integrated with particle filters. Though some kinds of limitation still exist in the method, we can verify that the particle filters can deal with the nonlinearity of estimation with the experiment.
KeywordsParticle Filter Translation Vector Image Processing Algorithm Multiple Camera Rotational Error
Unable to display preview. Download preview PDF.
- 1.Bradski, G., Kaehler, A. (eds.): Learning OpenCV: Computer Vision With the Opencv Library. SOreilly (2008)Google Scholar
- 5.Ueda, R., et al.: Expansion Resetting for Recovery from Fatal Error in Monte Carlo Localization – Comparison with Sensor Resetting Methods. In: Proc. of IROS, pp. 2481–2486 (2004)Google Scholar
- 6.Lenser, S., Veloso, M.: Sensor resetting localization for poorly modelled robots. In: Proc. of IEEE ICRA, pp. 1225–1232 (2000)Google Scholar