Notes:
The stereo information can be used in the following way to find the correspondence between the camera view and the CT-data. We have one left and one right videostream of the scene, and by using stereo algorithms ideally a depthmap like the one being seen here can be extracted. Using the genereated depthmap we try to find it's position on the CT data. When this surface registration process finds the matching position of the depthmap on the CT data we can deduce the camera parameters and thus the location of the hidden object on the cameraview can be found.