Camera calibration and navigation in networks of rotating cameras

Abstract

Camera calibration is one of the basic problems concerning ntelligent video analysis in networks of multiple cameras with changeable an and tilt (PT). Traditional calibration methods give satisfactory esults, but are human labour intensive. In this paper we introduce a ethod of camera calibration and navigation based on continuous tracking, hich requires minimal human involvement. After the initial precalibration, t allows the camera pose to be calculated recursively in real ime on the basis of the current and previous camera images and the previous ose. The method is suitable if multiple coplanar points are shared etween views from neighbouring cameras, which is often the case in the ideo surveillance systems.