AI-driven computational creativity in fashion design: a review
Xiaopei Wu, Li Li
Abstract
The emergence of text-to-image generative AI tools has garnered significant attention, particularly in creative design applications. This review article explores the field of AI-driven computational creativity, which has witnessed advancements in computational methods, ranging from traditional programming-based techniques to machine learning algorithms, and now to deep learning models. These deep learning models, including the recent text-to-image generative AI tools, have demonstrated impressive capabilities in creative content generation. While previous studies have examined the application of AI in the fashion industry, this review aims to provide a unique perspective. First, it presents AI-driven creativity within the framework of computational creativity, offering historical context. Second, it focuses specifically on the creative design applications in the fashion industry, rather than other aspects such as retail or supply chain. Lastly, it evaluates the outcomes of these studies from the perspective of industry fashion designers, considering the creative and practical value, instead of solely focusing on technical and theoretical performance from a computer science standpoint. By incorporating these distinct perspectives, this review contributes to the understanding of AI applications in fashion design and highlights their relevance in the creative domain.