Study suggests human-AI collaboration in fashion design – WWD

As the buzz around ChatGPT continues, academics and communications experts say the AI-powered technology has uses for marketing and content creation, but is far from replacing human creativity.

This rating was also found by Korean researchers, who studied AI in fashion and textile design. The study compared human and AI-generated designs. And while similar, humans have an advantage, the report says, while concluding that AI can help designers in the creative process. And it could open the door for non-professionals to create their own fashion designs.

In a study from Pusan ​​National University, researchers explored the use of “collaborative artificial intelligence models” in fashion design. They found that AI has “a wide range of trendy applications, from increasing process efficiency and reducing waste to improving the overall functioning of industry,” they said. the report’s authors said in a statement, adding that while creative processes aren’t often automated, AI can help in the creative design process. AI-powered creative technology can also play a role in teaching fashion design.

The study noted that AI is used in personalization, such as apps or websites that offer fashion recommendations. AI is also used to optimize the supply chain and to automate various manufacturing, logistics and fulfillment functions.

The research was conducted by Yoon Kyung Lee, assistant professor at Pusan ​​National University. Published in the December edition of the academic journal ScienceDirect, it was titled “How Complex Systems Engage in Fashion Design Creation: Using Artificial Intelligence.”

The impetus behind the research was to see if AI can augment the creative process. In a statement last week, Lee said “at a time when AI is so deeply ingrained in our lives, this study instead started by looking at what a human can do better than AI. Could there be have effective collaboration between humans and AI for creative design purposes?”

The first step was to generate new textile designs that used “deep convolutional generative adversarial networks (DC-GANs) and cyclic GANs,” the report notes. “The results of these models were compared to similar designs produced by design students.”

Lee said the comparison between the human-created and AI-generated designs showed that while the two designs were similar, “the biggest difference was the uniqueness and originality seen in the human-made designs. , which came from the person’s experiences. However, using AI in repetitive tasks can improve designers’ efficiency and free up their time to focus on more challenging creative work. Lee said it’s worth the pain of creating a “human-AI collaboration” model that integrates GANs with human creativity.

The human-AI model would include sharing design ideas with other designers, creating an interconnected and scalable creative platform.

Lee said that in the future, “everyone can be a creator or designer using AI models. Until now, only professional fashion designers have been able to design and put into value of clothes. But in the future, it will be possible for anyone to design the clothes they want and show their creativity.

Researchers from Pusan ​​National University in Korea conducted an extensive study exploring the use of collaborative AI models to create new designs and engage complex systems. This encourages human-AI collaborative design, which increases efficiency and improves sustainability. Graphic courtesy of the university.

Add a Comment

Your email address will not be published. Required fields are marked *