STRICOM, together with the School of Electrical Engineering and Computer Science (SEECS) and the Institute for Simulation & Training (IST) at the University of Central Florida are developing a system to allow virtual objects to be placed in live images in real time. The proposed approach is simulation driven in that it will use a geometric database of the site of the live scene to drive a simulator which will be used to predict the location of the synthetic object in the real scene at each instant of time. The research we are conducting involves object identification in the real world scene using registered overlays, registration of the real world view with the synthetic view of the virtual terrain data, placement of the virtual object with the synthetic terrain and then the natural view using simulation, and finally realistic integration of the synthetic object into the live scene. New techniques are being developed to determine the occlusion of virtual objects based on their relation to terrain features in the live scene. This paper describes the whole process used in the project, discusses the basic algorithms and presents novel techniques used for recognition and placement of the objects.