Human face texture generation based on MPEG-4 standard

摘要

This paper is about a new method of generation texture for 3D face model. The 3D model is used as a signing avatar to help deaf people to interact easily. This approach is based on the MPEG-4 standard for facial detection points from two orthogonal photos representing the front and side view of a human head. The two photos are cut, combined and finally deformed corresponding to an UV-map to finally generate a texture which can be mapped on the 3D head.