|
Abstract:This paper deals with matching between points of interest detected from two images at different resolutions taken using a camera equipped with a motorized zoom. The camera is supposed to be stationary or to be slightly displaced between the two images. In this context, the methods based on photometric data are inefficient, that is why we present a matching algorithm based on measures that are independent on the intrinsic parameters of the camera. The matching can be used in different applications such as video-surveillance, active vision or visual servoing. |