Direct estimation of the stereo geometry from monocular normal flows

Abstract

The increasing use of active vision systems makes it necessary to determine the relative geometry between the cameras in the system at arbitrary time. There has been some work on on-line estimation of the relative camera geometry parameters. However, many of them are based on epipolar geometry, motion correspondences, or even presence of some calibration reference objects in the scene. In this paper, we describe a method that allows the relative geometry of two cameras be estimated without assuming that their visual fields picture the same object, nor that motion correspondences in each camera are fully estimated beforehand. The method starts from monocular normal flows in the two cameras and estimates the relative geometry parameters without evening accessing the full optical flows. Experimental results are shown to illustrate the performance of the method. © Springer-Verlag Berlin Heidelberg 2006.