One of the greatest challenges in purchasing of clothing is to find a proper fit. This problem is compounded in online shopping of clothing because online retailers are, so far, unable to provide the shoppers with a way to try on clothing. The traditional tape measurement techniques for custom fit clothing are inconvenient, time consuming and ambiguous for non-skilled consumers. In addition, using 3D body scanners for taking anthropometric body measurements only suites larger retailers and shopping centers and not online shoppers. With readily available imaging devices such as digital cameras, webcams and cell phone cameras, almost anyone can now take photos with reasonable quality. The Anthropometric Topography Observation & Measurement (ATOM) project uses photogrammetry and geometric modeling techniques to extract anatomic measurements of the customers from their photographs within quarter-inch accuracy. This can help to address the fitting problem of the online garment purchasing.

As illustrated in the above figure, in the ATOM project, two reference photos are acquired from the front and side views of the person in pre-specified postures. Then, the contours of the person are extracted automatically from these two photographs. An accurate 3D avatar of the person is constructed by matching a generic template to the resulting contours and their feature points. Final measurements are automatically calculated from the resulting 3D model. Different stages of this process can be enhanced by extracting some metrics from a 3D database of real human bodies (e.g. through scanning) as well as using standard anthropometric proportions. In addition, by putting clothing on this 3D model, it can be visualized and animated in a virtual fitting room. Extracted body measurements and the clothed human model can then be exported to CAD/CAM programs for automatic cutting of the garments. The accessibility and the ease of use of this approach combined with its high accuracy helps to mass-customize apparel products online.

Links

Team

  • Mark Sherlock, MSc student
  • Javad Sadeghi, PhD
  • Faramarz Samavati, Research lead.

Start-up (Industrial Partner)

Qoture http://www.qoture.com, Led by Oliver Leung

Achievements

  • Plug and Play Winner (competition for entrepreneurs to present their business ideas to a panel of judges). Received a $25,000 investment along with 3-months of office space and mentorship in Calgary and the equivalent in Silicon Valley.

Related news:

Support and Funding

  • NSERC (Natural Sciences and Engineering Research Council of Canada)
  • GRAND NCE: Graphics, Animation, and New Media
  • University of Calgary