News Release

New 'virtual' tape measure could give online clothes shoppers the perfect fit

Business Announcement

Engineering and Physical Sciences Research Council

A ground-breaking web-based system that takes unprecedentedly detailed measurements of the body could revolutionise online clothes shopping.

The shopper would simply download software which, in conjunction with their webcam or smartphone, works like a 'virtual' tape measure, taking accurate waist, hip, chest and other measurements and advising the user on which size garment to buy whenever they visit the website of a participating retailer.

Taking multiple measurements of the body quickly, easily and accurately, the system aims to ensure the best possible fit and so save retailers and shoppers millions of pounds a year in return postage costs, as well as eliminating the hassle involved in sending back clothes that are the wrong size or fit.

With funding from the Engineering and Physical Sciences Research Council (EPSRC), the software is currently being developed by London College of Fashion and computer vision experts at the University of Surrey, in collaboration with body-mapping specialists Bodymetrics and digital creative agency Guided.

Body scanning is already starting to make a mark in the clothing retail sector. But because the new system takes measurements at a number of different points on the body and combines these with a person's overall proportions to build up a detailed 3D image, it offers much greater precision than anything else available in-store or online.

Moreover, most online shoppers currently buy clothes simply on the basis of waist size, for instance, or small/medium/large categorisation, whose accuracy is inevitably limited and often depends on the shopper's subjective perception of their own body size.

The new system avoids these problems. Once they find the item they are interested in, the shopper simply clicks their mouse to activate the software, stands in front of their webcam or smartphone in their underwear, takes a photo, types in their height and lets the software do the rest. The photo remains entirely confidential and is not transmitted over the internet in any way. The height measurement gives the software the starting point for ascertaining the body size of the shopper.

To inform customers, a logo (and possibly a pop-up on the computer screen) would appear on the websites of participating retailers. As slight variations often exist in the proportions of clothes with the same label size but produced by different manufacturers, retailers would also supply detailed information about the size of all their individual garments. The software would take this into account when recommending the size of a particular item for a particular shopper.

"The potential benefits for the fashion industry and for shoppers are huge," says Philip Delamore from London College of Fashion. "Currently, it's common for online shoppers to order two or three different sizes of the same item of clothing at the same time, as they're unsure which one will fit best."

Professor Adrian Hilton from the University of Surrey says: "It's unrealistic to expect online clothes shoppers to have the time or inclination to take a series of highly accurate body measurements of themselves. The new system makes it all very easy."

Plans to take the new system to market are now in hand, with a launch anticipated within two years.

###

Notes for Editors

The 18-month project 'Body Shape Recognition for Online Fashion' is receiving total EPSRC funding of around £350,000.

The research builds on previous work by the University of Surrey in developing 3D body-shape templates from single-view images. This work has been used commercially in creating animated representations of people for games such as the Sims.

30-60% of clothes bought online are currently estimated to be returned to the retailer.

EPSRC is the main UK government agency for funding research and training in engineering and the physical sciences, investing more than £800 million a year in a broad range of subjects – from mathematics to materials science, and from information technology to structural engineering. www.epsrc.ac.uk

The Centre for Vision, Speech and Signal Processing (CVSSP), University of Surrey, directed by Professor Adrian Hilton, is one of the largest research centres in the UK focusing on signal processing and interpretation. It has over 120 members comprising academic and support staff, postdoctoral research fellows, visitors and PhD students. The Centre's expertise and activities span Computer Vision, Digital Signal Processing, Machine Learning and Artificial Intelligence, Computer Graphics and Human Computer Interaction, Medical Image Analysis and Multimedia Communication. www.surrey.ac.uk/cvssp

London College of Fashion's rich heritage and responsiveness to changes in design practice have positioned it as a leading global provider of fashion education, research and consultancy. The College's work is centred on the development of ideas: its staff and students use fashion alongside historical and cultural practice to challenge social, political and ethical agendas. This, combined with its forward-thinking business and management portfolio and its relationship with the global fashion and lifestyle industries, is the underpinning of its mission to "Fashion the Future". www.fashion.arts.ac.uk

For more information, contact:

Professor Adrian Hilton, University of Surrey, tel: 01483 683956, e-mail: a.hilton@surrey.ac.uk

Howard Wheeler, Media Relations Officer, University of Surrey, tel: 01483 686141, e-mail: h.wheeler@surrey.ac.uk

Images are available from the EPSRC Press Office. Contact tel: 01793 444404, e-mail: pressoffice@epsrc.ac.uk

Image details:

Dress.jpg: The new process could revolutionise online clothes shopping for both the retailer and the consumer (Image credit: Jane Reck).

Hourglass.jpg: The new system builds up a detailed image of body measurements, making it much easier for the shopper to order the correct size for their body dimensions.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.