AiFashion: Multi-Modal and Multi-Dimensional Large Model Based on Self-Trained Customer Digital-Twin for Fashion Design and Manufacturing

This paper presents the concept design and tech-nical roadmap of a novel AI -driven fashion design and clothing manufacturing platform, called AiFashion. The vision of AiFashion is enabling every customer to wear personalized and customized clothes at low costs and with shorter waiting times. By integrating advanced information techniques such as Digital Twins, Generative AI, and Large Language Models. AiFashion can revolutionize the entire lifecycle of fashion designing and clothing manufacturing. Compared with existing AI -assisted fashion tools, AiFashion is characterized by its self-training nature, customization, interactivity, and lifecycle orientation in the garment industry. The first module of AiFashion, Chat Customer, effectively captures multi-modal customer profiles, including shallow body parameters and underlying preferences. The second module, Chat Tailor is a fashion designer based on Generative AI, which automatically generates the most suitable clothes satisfying multi-dimensional demands such as occasion, weather, and budget. To efficiently produce customized and highly distinctive clothing designed by Chat Tailor in clothing factories, the third module, Chat Manufacturing, leverages its absorbed professional knowledge in task planning and resource scheduling by chatting with managers, supervisors, and operators in factories. This paper gives brief implementation plans for each module in AiFashion.