생성형 AI 기반 부조 창작 방법에 대한 타당성 연구 = A Feasibility Study on Relief Creation Methods Based on Generative AI

다국어 초록 (Multilingual Abstract)

This study explores the application of Generative Artificial Intelligence (Gen AI) technology in the creation of 3D reliefs. The research highlights the potential of AI in the realm of art, especially in the context of relief art creation. The methodology combines experimental and empirical analyses with the goal of developing a streamlined artistic process that enables artists and creators to easily transform creative ideas into 3D relief artworks. The experimental segment involves generating 2D images from text prompts, converting these into 3D relief models, and employing 3D printing technology to physically fabricate the models. Utilizing tools such as Midjourney and Stable Diffusion, combined with ZBrush software, the study generated 2D images containing depth information from text prompts and further transformed these into 3D models. The experiment validated the effectiveness and feasibility of this method. In the empirical research segment, Exploratory Factor Analysis (EFA) and Analytic Hierarchy Process (AHP) were employed to evaluate the impact factors of AI in 3D relief art creation and to determine the relative importance of various factors. This analysis revealed the perspectives and priorities of experts on different aspects of 3D relief art creation, highlighting elements such as artistic expression, technical efficiency, innovative integration, and creative generation. This research demonstrates the feasibility of a Gen AI-based method for creating 3D relief art, simplifying traditional creative processes, and offering new possibilities for the cross-fertilization of art and technology. The findings emphasize the importance of technical efficiency and artistic expression in realizing the digital transformation of art creation.