Deep Learning Network for Computer-Aided Fractal Art Pattern Design

Abstract

The current conventional computer-aided fractal art pattern design method mainly designs the mechanical parameters of the pattern samples to build a three-dimensional fractal simulation image, which leads to the poor effect of the generated pattern due to the lack of design of the pattern light source and texture. In this regard, a computer-aided fractal art pattern design method with deep learning network is proposed. An image transformation network structure is built, the light sources and textures of fractal patterns are designed using OpenGL software, and the Connett and Martin fractal number sets are used to generate fractal patterns. In the experiments, the generation performance of the proposed method is verified. The analysis of the experimental results shows that the proposed method generates fractal art patterns with low FID values and has a more excellent pattern generation performance.