A New Art Design Method Based on AIGC: Analysis from the Perspective of Creation Efficiency

Mengzhen Xiong,Jian Rao

DOI: https://doi.org/10.1109/ICID60307.2023.10396804

2023-10-20

Abstract:With the development of "AI-generated content(AI GC)" technology, the field of art and design is witnessing a new transformation. Along with applying AI technologies such as generative adversarial networks, diffusion models, and transformer architecture in the visual text filed, art design has transformed from the traditional personal inspiration-centered creation model in to an AI large model-driven creation model. Starting from the efficiency of art design creation, this paper proposes a new "text-image" iterative art creation method based on prompt word engineering. It introduces the specific process and steps of art creation through AIGC tools. At the same time, questionnaires, semi-structured interviews, and other methods to quantitatively analyze the creative methods of teachers and design students. However, the results of the AIGC tools in deep cultural semantic understanding are poor and must be further supplemented manually. Given the above research, this paper further elaborates on the subsequent development direction of AIGC tools, which should take dynamic real-time interaction as the core. Multi-modal domain information should be integrated and embedded into the creation process, which will realize the dynamic creation and improve the quality of works.

Art,Computer Science