Cellular Genetic Algorithm for Supporting Creative Design of 3D Animation Modelings

In order to generate a variety of innovative 3D animation modelings automatically, an improved cellular genetic algorithm was proposed in this paper. ACIS rule expressions were used to scale the existing 3D animation modelings unevenly in the x-axis, y-axis and z-axis direction separately in this algorithm, and tree coding was adopted for the generation and evolution of ACIS rule expressions. Due to the difficulty that there was no uniform evaluation criteria determining the creative design, human-computer interaction was adopted which used expert knowledge to determine the fitness function value of individuals. An innovation system of 3D animation modelings is developed, and the experiments in the system show that the improved cellular genetic algorithm can generate a series of 3D animation creative modelings.