At present, traditional design is facing challenges, and AI brings new opportunities to the design field. This article focuses on the application of AI (Artificial Intelligence) visual expression characteristics in modern design. Firstly, this article analyzes the characteristics of precision, efficiency and innovation of AI visual expression, expounds the requirements of modern design in various fields, and then explains in detail the specific applications in plane, product, interior and other designs. In this article, an algorithm model based on the combination of GAN (Generative Adversarial Network) and CNN (Convolutional Neural Network) is adopted. Training and testing on a data set covering 10000 images. The experimental results show that the SSIM (Structural Similarity Index) of the generated image is stable at about 0.8 after 50 rounds of training, and the display of the generated image hash distance is highly innovative, and it takes only 10 minutes to generate 100 design images on average, which is much faster than the traditional design method. The research shows that the algorithm model can effectively improve the application efficiency of AI in modern design and provide strong support for the innovation and development of modern design.
Guohong Guan (Sun,) studied this question.