Artificial Intelligence Augments Design Creativity: A Typeface Family Design Experiment

Abstract

This paper attempts to explore a way to augment design creativity with artificial intelligence through a typeface family design experiment. At the beginning, the article briefly reviews the theory of creative generation, and based on the commonality of these theories, the design creative model of “Inspiration - Creation - Reflection - Re-Creation - Re-Reflection” is summarized. Then, the hypothesis of artificial intelligence augment design creativity based on this model was proposed. In order to prove the feasibility of this hypothesis, we have carried out an experiment of typeface design for Chinese characters. The problem to be solved in this experiment is the contradiction between traditional Chinese character design method and large-scale diversification of information needs. We propose to design a Chinese character typeface family instead of designing a typeface which provides diverse character experience based on the concept of “one character with thousands of forms”. In order to obtain the typeface family, we adopt Generative Adversarial Networks and select different typical Chinese typefaces as data for model training, finally generate the diverse typefaces. This design method is completely different from the traditional typeface design method and it has produced a new typeface design form. In general, the designer initiated the button of design creativity according to the design problem. The artificial intelligence continuously gives the result of the learning while the designer continuously responds to that. The interaction between designers and the artificial intelligence brings out a new type of Human-machine collaborative relationship, that is: artificial intelligence augments design creativity.