A new method for generating and animating a 3-D model of a person's face is proposed. The method involves novel algorithms for 2-D to 3-D construction under perspective projection model, real-time mesh deformation using a lower-resolution control mesh, and texture image creation that involves texture blending in 3-D. The resulting face models can be readily used in 3-D games, mobile messaging, e-learning applications such as lip-reading and personalized training, 3-D movies and 3-D TV programs, and for providing human-computer interaction (HCI). User interfaces with personalized face models could be used to guide people in accessing knowledge and information on the Internet or to facilitate the usage of computers and software. Thus, it is believed that the utilization of personalized 3-D face models will help bring down the barrier between computers and people.
展开▼
