The Modeling Of Virtual Actor's Facial Expression Based On GA Algorithm

As the profound research of Artificial Neural Network, the method of Artificial Neural Network has been successfully applied in many fields. The parallelity of information management, distributed information storage, self-organizaion and self-adaption, strong study and association function and its fault tolerance embody the main characteristic of Neural Network.Three-dimentional computer graphics (3DCG) animated movies are a modern form of entertainment. The animated directors need better and better outputs of virtual actor facial expression. They also need computers help themselves to get satisfactory virtual actor facial expression.Genetic Algorithm emulates biology evolution which based on the natural choice. It is a random method and it simulates the biology evolution process. It discards traditional search mode. Genetic Algorithm randomly search the target space by manual evolution.Interactive Evolutionary Computation bases on Evolutionary Computation. It integrates Evolutionary Computation with subjective human evaluation. It extends algorithm and has stronger description ability than algorithm.This paper illustrates a neuro-fuzzy method for training virtual actors in three dimensional computer graphical movies according to the effect that the human directors expect. Virtual actor facial expression modeling is an important and practical topic. In this paper, fuzzy neural network is used to accommodate director expert knowledge such that appropriate virtual actor facial expression can be produced. In order to speed up the convergence of the learning algorithm, Genetic Algorithm and Interactive Evolutionary Computation are respectively incorporated into the fuzzy neural network. Our experimental results indicate the success of our approach here.