New study explores artificial intelligence in fashion

The use of artificial intelligence (AI) in the fashion industry has increased significantly in recent years. AI is being used for tasks such as personalizing fashion recommendations for customers, optimizing supply chain management, automating processes, and reducing waste by improving sustainability.

However, the creative process of fashion design remains largely human-driven, and not much research exists in the area of ​​using AI in fashion design. Additionally, research is typically done with data scientists who build the AI ​​platform and are involved in the technical aspects of the process. However, the other side of the equation, the designer himself, is less bound by research.

To explore the practical applicability of AI models to implement creative designs and collaborate with human designers, Associate Professor Yoon Kyung Lee of Busan National University in South Korea conducted an in-depth study. her research thinking and creativity.

“At a time when AI is deeply ingrained in our lives, this research instead began by looking at what humans do better than AI,” says Professor Lee, who is behind the study. explain the motives. “Is there an effective collaboration between humans and AI for creative design?”

Professor Lee started by generating new textile designs using Deep Convolutional Generative Adversarial Networks (DC-GANs) and cycle GANs. Outputs from these models were compared to similar designs created by design students.

Although the designs are similar, I found that the biggest difference is the uniqueness and originality of the human design based on the person’s experience. However, using his AI for repetitive tasks increases designer efficiency and frees up time to focus on more challenging creative tasks.

AI-generated designs can also be used as learning tools for those without fashion expertise who want to explore their creativity. These people can create designs with the help of AI. Professor Lee therefore proposes a collaborative human-AI network that integrates GANs and human creativity to generate designs.

He has also defined and studied various elements of complex systems involved in human-AI co-design. She also established a human AI model where designers collaborate with her AI to produce novel design her ideas. The model is built so that as designers share their creative processes and ideas with others, the system interconnects and evolves, thereby improving designs.

The fashion industry can leverage this to anticipate changes in the fashion industry and provide recommendations and co-creation services. Setting goals, variables, and limits is part of the designer’s job in a human-AI co-design environment. Therefore, their work should not only be limited to visual aspects, but also cover different areas.

“In the future, everyone will be able to become creators and designers with the help of AI models. So far, only professional fashion designers can design and showcase clothes. But in the future, everyone will be able to do it: design the clothes they want and show their creativity,” Professor Lee concludes.

Provided by Pusan ​​National University