AI hits the catwalk with Hong Kong-UK fashion collab

In the halls of London’s V&A Museum, renowned for showcasing fashion history with couture collections from icons such as Dior and Alexander McQueen, AI-designed clothes made their entrance this week.

Thursday evening’s ‘Culture x AI’ event, brought to London from Hong Kong, displayed an amalgamation of designs from five fashion creatives, artist Kan Tai Keung’s ink-wash paintings, and an AI tool called ‘AiDa’ yielded 20 designs on display.

Monochromatic garments with ruffles, mirrors, and traces of the ink-wash artistry adorned the catwalk.

The aim: “To show the power of collaboration,” according to Jeanne Tan, professor at the School of Fashion and Textiles at the Hong Kong Polytechnic University, and head of the project.

At a time when AI is often seen as a threat to creative jobs, Keung claims that the project demonstrates how technology should be embraced for good.

He explained that in the same vein as the traditional Chinese wisdom of ‘using a brush instead of being manipulated by it’ he added: “We must face the challenges of any new intelligent technology in the future with an absolutely open attitude, to create a better future.”

AiDa

The platform behind AiDa, the Laboratory for Artificial Intelligence in Design (AiDLabs), was established by the Hong Kong Polytechnic University (PolyU) and the UK’s Royal College of Art.

As part of it, Calvin Wong, a professor at Hong Kong PolyU, and his team developed AiDa (the Artificial Intelligence Based Interactive Design Assistant’) in a mission for “transformation research,” says Wong.

The tool was initially created in 2020 and is now in its third version.

Adds Wong: “We are not using AI to replace designers. AI is already happening and it’s unavoidable, even for the fashion industry.”

“We need to embrace AI in the conventional design process,” he adds. “So AiDa is treated as an assistant to facilitate designers.”

For AiDa to produce its designs, users input a mood board for reference and can ask it to design anything from a complete collection to a single item.

The AI tool creates a visual of the clothing item, which the designers can then take to a pattern maker to create a blueprint of their design.

“AiDa is focusing on the inspiration process at the beginning,” says Wong.

The backbone of the model is trained on 500,000 images, embedded with labels and annotations that clearly represent key design elements. Then, it has included style recognition, print recognition, and generation.

Users of the web-based tool can upload sketches they’ve already created to tweak or create designs completely from scratch – if they don’t like the output, they can press ‘redesign.’

According to Wong, the process has cut time spent at this stage by 70%.

Designers will be able to use AiDa on either a $50 a month subscription or an annual $500 dollar subscription – and a five-day free trial is available.

He explains that the aim is to help designers, either experienced or new, to gain inspiration using the technology.

Young designers interested in learning about the technology have until the end of August to apply for one of ten places in its mentorship programme.

Security-wise, Wong ensured that the AI does not reuse any designs put into the tool, or any of the outputs to ensure that the user’s intellectual property stayed secure.

Culture x AI

The five young designers who took part in the show this week – Derek Chan, Wilson Choi, Tak Lee, Sophia XinLi, and Aries Spin – each chose a phase of Kan Tai Keung’s art to cast into the AI and create items based on its output.

“This is a very meaningful project for me” says Keung, “because before embarking on a design career I worked as a tailor for over ten years.”

When creating the designs together, Keung explains that they discussed his inspiration behind his own art at the time and encouraged the designers to embrace the new technology while ensuring that their own personalities remained in the final product.

“I told them, don’t just use the design the AI produces, add your own thoughts.”