Abstract
This paper describes the model-based video coding scheme which combines computer vision and computer graphics techniques. Human face modeling system is used as an example. In this system, a generic 3D face model is used for both facial motion analysis and expression synthesis. The motion trajectories of the feature points are derived using computer vision techniques and computer graphics techniques are used to reconstruct the original motions using the 3D face model and the motion parameters. The potential application of this system in model-based video compression is also discussed in this paper. Some preliminary experimental results show promising future of this approach.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
