A product appearance design method based on artificial intelligence generated content

Abstract

This study proposes a product appearance design method that integrates advanced Artificial Intelligence Generated Content (AIGC) tools such as ChatGPT and Midjourney, aimed at assisting the design process and improving design efficiency. The proposed method first utilizes ChatGPT's natural language processing capabilities to extract Kansei engineering vocabulary that captures target product attributes. These terms are then used to generate detailed prompts, which are fed into Midjourney to create preliminary conceptual designs. Given AIGC's ability to generate a vast array of design proposals, close collaboration between designers and AIGC tools is essential to ensuring that the final design meets innovative objectives and practical requirements. To verify the feasibility and effectiveness of the design proposals generated by the proposed approach, this study employs the Intuitionistic Fuzzy Set (IFS) method that combines subjective and objective assessments, along with eye-tracking technology. Through IFS, experts select the designs that most closely align with the Kansei engineering vocabulary. Meanwhile, eye-tracking experiments provide objective physiological data to quantify user preferences and visual appeal for the design proposals. The application of AIGC technology highlights the transformative potential of artificial intelligence in product design, demonstrating its capacity as a powerful tool that effectively enhances human creativity and significantly improves design efficiency and quality.