International Conference on Image Processing (ICIP 97), vol. 1, pp. 444-447, Santa Barbara, October 1997. In this paper a model-based multi-view image generation system for video conferencing is presented. The system assumes that a 3-D model of the person in front of the camera is available. It extracts texture from a speaking person sequenceimages and maps it to the static 3-D model during the video conference session. Since only the incrementally updated texture information is transmitted during the whole session, the bandwidth requirement is very small. Based on the experimental results one can concludethat the proposed system is very promising for practical applications.
Chun-Jen Tsai, Aggelos K. Katsaggelos, Peter Eiser