A role for 3D modelling in controlling webeams and interpreting real-time video

This paper reflects on the Valhalla Project, funded by the EC, its integration of GIS generated VRML with real-time Video, and on potential further development of this approach. The VRML model is used to interpret video images through hyperlinked descriptive information. The GIS contains information about the location of objects in the scene, and is used to build the hyperlinked VRML model, which is slaved to the camera control data to offer matching field of view and focus. It is also used to set up and run server based spatial analytical searches to identify, and retrieve archived video clips and historic images or photographs on the basis of the objects that are imaged within them. The models are useful in enabling an overview of the area, not otherwise readily achieved, and may be used to request real-time video tours to be scheduled, as one solution to web cam control contention.