ITAA - KSCT Joint Symposium

Evaluating Midjourney’s Interpretation of Fundamental Fashion Design Elements: A Structured Prompt-Based Analysis

---

Abstract

This study examines Midjourney’s capacity to interpret and visualize fundamental fashion design elements using a structured, prompt-based evaluation method. Although generative AI tools are increasingly used for rapid ideation in creative fields, their ability to accurately interpret detailed design descriptors remains unclear. Five key fashion design elements—style, color, fabric, prints, and silhouette—were tested through 3–5 controlled text-only prompts per category, with only one variable adjusted at a time. Results reveal significant inconsistencies across elements: style strongly influenced model gender presentation; color interpretation varied widely, especially with complex descriptors; fabric was represented more reliably when visually distinct; prints were frequently misinterpreted as environments; and silhouette accuracy depended heavily on garment type. Findings highlight both the creative potential and the limitations of Midjourney in precision-driven fashion development. Recommendations emphasize strategic prompt design and continued human oversight when integrating AI into early-stage fashion ideation.

Keywords: Midjourney; generative AI; fashion design; prompt engineering; design elements; visual interpretation

How to Cite:

Chen, L. & Zhang, L., (2025) “Evaluating Midjourney’s Interpretation of Fundamental Fashion Design Elements: A Structured Prompt-Based Analysis”, International Textile and Apparel Association Annual Conference Proceedings 82(1). doi: https://doi.org/10.31274/itaa.21476