Minjung Park

Despite growing interest in artificial intelligence (AI) in higher education, research on its pedagogical integration in fashion design education remains limited, with most studies focusing on the technical aspects of AI tools rather than their instructional roles or educational implications. This study explores how AI tools can be effectively integrated into fashion design education by examining student challenges, AI-supported learning competencies, and instructional considerations from the perspectives of instructors. Semi-structured interviews were conducted with 10 university-level instructors from South Korea and the United States who have experience teaching fashion design-related courses. Thematic analysis is employed to examine the educational roles of AI in supporting students across the various design process stages. Findings identify four primary areas where students commonly face difficulties: initial design planning, ideation and creative exploration, technical garment construction, and communication and feedback adaptation. This study presents a pedagogical framework that illustrates how AI tools, including large language models, image generators, simulations, and feedback systems, can enhance visualization, creative thinking, and student engagement. This framework aligns student challenges, AI tool functionalities, and instructional strategies across four stages of the design process—problem definition, ideation and refinement, prototype development, and evaluation—offering practical guidance for the intentional integration of AI in fashion design education that promotes learner autonomy and human–AI collaboration.

Purpose: This study explores how VR interactivity influences user perceptions and behavioural intentions by stimulating mental imagery. Grounded in flow theory and TAM, it positions mental imagery as a key mechanism linking interactivity to sensory and behavioural outcomes. Design/methodology/approach: A convenience sample of U.S. undergraduates used Google Cardboard to explore a virtual store. Findings: VR interactivity influenced mental imagery and sensory app experience, which in turn influenced perceived enjoyment and usefulness. Enjoyment affected attitudes and behavioural intent, while usefulness influenced intent but not attitudes. Mental imagery mediated the relationship between interactivity and sensory app experiences. Originality/value: This study highlights VR’s underexplored potential to evoke mental imagery through interactivity. Conceptualizing mental imagery as a second-order construct, the findings show that VR interactivity promotes mental imagery, which in turn enhances sensory engagement and behavioural intentions.

Generative AI and LLMs Recent advancements in generative AI technology, utilizing deep learning and machine learning algorithms, have garnered significant attention for their ability to learn patterns from input data and generate new data while preserving the characteristics of the original dataset (Baidoo-Anu & Ansah, 2023). Particularly, LLMs have demonstrated impressive capabilities across a wide range of complex and economically valuable tasks, challenging previously held beliefs about the necessity of human cognition for these tasks (Naveed et al., 2023). These LLMs specialize in text generation, producing natural-sounding sentences based on provided topics or ideas, with advantages including understanding complex language contexts and delivering sentences with high accuracy, naturalness, and style (Ozdemir, 2023). LLMs show significant promise in storytelling (Yuan et al., 2022). Incorporating user preferences into LLMs offers a considerable advantage in ensuring accurate outputs. Particularly, accurately reflecting the characteristics of the fashion domain necessitates close collaboration between experts in the fashion field and IT professionals. This underscores the importance of integrated technological development through academia-industry collaboration.

Fashion image datasets, in which each fashion image has a label indicating its design attributes and styles, have contributed to the achievement of various machine learning techniques in the fashion industry. Computer vision studies have investigated labeling categories (such as fashion items, colors, materials, details, and styles) to create fashion image datasets for supervised learning. Although a considerable number of fashion image datasets has been developed, different style classification criteria exist because of a lack of understanding concerning fashion style. Since fashion styles reflect various design attributes, multiple styles can often be included in a single outfit. Thus, this study aims to build a Hybrid Style Framework to develop a fashion image dataset that can be efficiently applied to supervised learning. We conducted focus group interviews with six fashion experts to determine fashion style categories with which to classify hybrid styles in fashion images. We developed 1,206,931K-fashion image datasets and analyzed the hybrid style convergence. Finally, we applied the datasets to the machine learning model and verified the accuracy of the computer’s ability to recognize style. Overall, this study concludes that the Hybrid Style Framework and developed K-fashion image datasets are helpful, as they can be applied to data-driven fashion services to offer personalized fashion design solutions.