Replicating the way our eyes and brains work together to recognize patterns, the app analyzes the image, searches for items with similar characteristics in color, shape or design, and returns images of the similar items to the user. "This system models what goes on in our own neurons to mimic the same level of accuracy our own brains offer," Cortexica's VP of business development Steve Semenzato told Information Week.
[ Is the U.K. poised to become the next technology hub? Read Britain: The New 'Silicon Island'? ]
"What Cortexica's Find Similar software does is broaden the choices available to the consumer, serving up similar and often better items available elsewhere," CEO Iain McCready explained. "Most shoppers have experienced that deep feeling of frustration after hunting endlessly and aimlessly for an item of clothing that they've admired. They've also had the experience of seeing an expensive item and wondering whether they might be able to find something similar [that's] more affordable."
Find Similar for Fashion uses parallel probabilistic computation, a neural net learning system that finds visual key points of interest tied to patterns in images and video. It also lets shoppers take visual clues from sources such as wallpaper, color swatches, or items worn by others.
As part of Imperial College's Imperial Innovations program, which seeks to develop commercial products from intellectual property developed in British higher education, Cortexica already supplies pattern-matching software to eBay Motors. That application analyzes photographs of cars to help buyers find similar models available on eBay. Cortexica has also developed systems to help U.K. retail firms, including Tesco, search for wines.
Several leading U.K. fashion retailers are currently testing the software, according to Cortexica. It will be integrated into websites and mobile phone-based apps ahead of a full launch in the fall.
To be effective, business technology pros gather information and interact with peers in a variety of ways. InformationWeek and its parent company, UBM Tech, are looking to discover what information you want and how you like to receive it, as well as your feelings on interactive communities, online content and live events. The results will help our editors develop products and services that best meet your needs. Take this survey and tell us how you like your tech content: Digital, live, opinionated? Tell us and enter to win a 32-GB Google Nexus 7 tablet.
The article is from http://www.informationweek.com/development/tools/uk-startup-offers-neural-net-based-fashi/240157476
没有评论:
发表评论