ZIRST – 655 avenue de l’Europe

This paper presents an evolutionary approach on Immersive TV, an exciting new concept of future TV entertainment, where TV watchers get the impression of being present at a real event. Starting from a short-term solution which can be introduced with available hardware at reasonable costs, the paper particularly discusses thinkable extensions towards a longterm solution, called IVVV (Interactive Virtual View Video), targeting at a highly interactive TV system. The idea is based on an implicit 3D representation of the video scene supporting head motion parallax viewing, as well as user-guided navigation with a virtual camera. The concept distinguishes between accurate offline 3D analysis of static scene parts and online real-time processing of dynamic video objects. Static and dynamic 3D data are merged at the receiver and used, together with the transmitted video sequences, to create the desired intermediate views by means of image-based rendering.

Related Work Other structures have been proposed to model visibility information in computer vision and computer graphics (e.g. aspect graphs [6] and visibility skeleton [2]), but they need a rigid 3D model as input and are not optimized for the same uses. In contrast to these, our strucinria-00590116,