Artificial Intelligence Empowers Designers In IBM, Tommy Hilfiger And FIT Collaboration

Young fashion buyers today demand a customization more than ever. According to an IBM study, 52% of women in Generation Z would like to see tools that allow them to personalize their products.

This coincides with an increasing expectation of speed in the delivery of the product. While several fast-fashion retailers can take the product to the shelves in weeks, most clothing items take six to 12 months to develop.

Technology is impacting throughout the supply chain to advance, even in the creative process itself. Artificial intelligence (AI), for example, which incorporates artificial vision, natural language understanding and deep learning, is being used to generate key ideas about trends both to streamline the initial design process and to better predict the demand for hyper-localized products. .

IBM has partnered with Tommy Hilfiger and the Fashion Technology Institute (FIT) Infor Design and Tech Lab in a project called Reimagine Retail to demonstrate this. The goal is to show how Artificial Intelligence capabilities can give retailers a speed advantage and equip the next generation of retail leaders with new skills using AI in design, according to Steve Laughlin, general manager of IBM Global Consumer Industries. .

To do so, FIT students had access to the artificial intelligence capabilities of IBM Research, including computer vision, natural language comprehension, and in-depth learning techniques specifically trained with fashion data.

Those tools were applied to 15,000 images of Tommy Hilfiger products, some 600,000 images of publicly available clues, and almost 100,000 patterns of fabric sites. Then they brought with them key silhouettes, colors and impressions and novel patterns that could be used as informed inspiration for student designs.

"The machine learning analysis gave us information about the colors, silhouettes and impressions of Tommy Hilfiger that we could not begin to consume or understand with the human mind." This allowed FIT Fashion Design students to be inspired by the trends of American fashion or popular and marry the 'DNA', if you will, of the Tommy Hilfiger brand in those dimensions to create completely new design concepts, "explains Michael Ferraro, executive director of FIT's Infor Design and Tech Lab.

"Then, a key impact [of the project] was a new inspiration." Another important impact for the consumer is the opportunities to personalize and personalize the clothes, without losing the style of what they love about Tommy Hilfiger, "he adds. Cognitive impression generator tools and personality analysis provided some great opportunities to provide consumers. In addition to exploring how AI could affect fashion design decision making, we also explored how tools such as listening on social networks and speech recognition can create a more personalized shopping experience based on an interactive 'intelligent' supply chain strategy that optimizes waste and minimizes environmental impact. "

The resulting 3D digital designs were presented to Tommy Hilfiger and IBM executives, who chose a checkered technology jacket from FIT senior Grace McCarty. In designing his appearance, McCarty said he was inspired by AI's views on the style and silhouettes of Tommy's brands, as well as popular and trend colors, and AI generated new patterns.

The design also incorporated a special thread integrated into a removable futuristic plaid panel with IBM's Watson's Tone Analyzer, which responds almost in real time to the feeling in a client's social media accounts.

"As a brand, we are always pushing the boundaries of what is possible through innovation and disruption, and these young designers really embody this spirit by showing the successful integration of fashion, technology and science," wrote the director of Tommy Hilfiger brand, Avery Baker, in a blog post for IBM.

One of the great conversations about the use of artificial intelligence in design is if it marks the death of creativity. If AI is used to exactly shape the new collections, is there a need for a designer in the first place? Both Ferraro and Laughlin believe that this project demonstrates how it can be used as a means to increase human processes, instead of replacing them.

"AI can help design teams improve and reduce overall delivery times, and expand their creative discovery by analyzing and memorizing thousands of images and videos using computer vision." These designers can also more easily find out how They can integrate trend colors, key patterns and styles, "explains Laughlin. It's about reducing a time-consuming and resource-intensive manual process, or exploiting that research element by providing access to much wider sources than ever, he said.

"It is the fundamental difference between the systems that improve and scale the human experience, instead of those that try to replicate all the human intelligence, the AI ​​inspired us and breathed new life into the design process", adds Ferraro.

IBM previously worked with Australian designer Jason Grech in the Cognitive Couture collection for Melbourne Fashion Week. Similarly, thousands of fashion images of catwalks were analyzed, as well as social media content to help inform the design process in a data-driven manner.

Laughlin emphasizes that making trend prediction more accessible and bringing products to market faster means less lost sales. "For consumers, when a desired item is not immediately available in the online store, the sale will not occur," he explains.

"The slow process of [fashion companies] hinders the ability of a brand to synchronize with the expectations of constantly evolving consumers, product trends and external market forces ... In the future of fashion with AI , designers could obtain internal and external data sources so that they can make their designs more informed and relevant, and can also give them the ability to design elements that personalize and personalize certain markets or consumers. "