Innovative project assesses how AI can improve online bra fitting

Improving how online shoppers submit potentially sensitive personal information is at the heart of a new project involving the University of Huddersfield and an innovative fashion tech company.

Brarista is an online bra-fitting service that has spun out of University College London (UCL) and is led by entrepreneur Bella Trang Ngo. She has ambitions that the company will help solve what she calls “one of the biggest problems of female well-being.”

The University’s Professor Petko Kusev and Professor David Peebles have begun a 12-month long project with Brarista boosted by a grant of £50,000 from Future Fashion Factory (FFF), an industry-led collaborative research programme.

“We wanted to provide a better way of getting professionally-fitted bras for consumers,” says Bella. “Having been a professional bra fitter myself by eyesight, I learned that the market is flawed. The most popular fitting method is with a tape measure, but that has been shown to be only 30 percent accurate.

“That method does not consider shape, or inconsistency of fit across different products. Because of that, bra consumers experience a lot of physical discomfort, self-confidence issues and at times health issues.

“We believe we can do it better, and we are using artificial intelligence (AI). The idea is that with a few photos of the upper body, non-naked, you can get your correct bra size and best fit across different products.”

Huddersfield to help evaluate Brarista system

The University will support Brarista in designing and conducting experiments and online surveys, as well as analysing their data, to help develop fitting algorithms that will aid Brarista deliver an improved user experience.

“Brarista wanted people with expertise in experimental design and consumer evaluation, as they were looking at discovering how potential customers might accept this service,” says Professor Peebles.

“They will give us a prototype of the system which we will evaluate by getting participants to use it, interact with the system and then tell us what they think of it. It’s an interesting and unique use of AI.”

Professor Peebles was involved in a report published in 2021 that highlighted how AI and VR could be used to improve healthcare training and education, while Professor Kusev’s expertise in behavioural science, experimental psychology and AI will also be essential for achieving Brarista’s goals.

Taking the idea elsewhere in fashion industry

“They are a very ambitious start-up company that are developing algorithms for the fashion industry, which is quite novel,” says Professor Kusev. “The software can learn, adapt and provide decision support to their customers for online shopping. The idea is to extend this to all areas of fashion retail, so that it could be possible to use it with everything that needs measurements.”

Bella, who graduated with from UCL an MSc in Entrepreneurship in 2018, is delighted with how Huddersfield’s experience across different sectors is helping her start-up, based in London’s Kings Cross.

“We’ve been so pleased with the support from the academics on this project. It is really interesting how they are all male, but they are aware of the problem and what is required.

“We will work together on designing and refining the user experience for our software. The hypothesis is that, as it is a new technology used around a very sensitive matter, we are asking consumers to do something that requires privacy. As a result, trust issues are quite significant.

“It is a very human-centric product, so the idea is that with the University’s help we are going to leverage their expertise in consumer behaviours to help to design better products.”

/Uni Public Release. This material from the originating organization/author(s) may be of a point-in-time nature, edited for clarity, style and length. The views and opinions expressed are those of the author(s).View in full here.