acm-header
Sign In

Communications of the ACM

ACM TechNews

China Develops Facial Recognition Payment System With Near-Perfect Accuracy


A representation of how a facial recognition system works.

The China Academy of Science says the facial recognition payment application it plans to unveil next year offers near-perfect accuracy.

Credit: TheHackerNews.com

The Chinese Academy of Science (CAS) plans to unveil a face-recognition payment application with near-perfect accuracy next year.

The app is designed to link with a bank account or credit card, and takes only one second to identify the payer, according to CAS Chongqing Green Technology Research Institute director Zhou Xi.

Consumers will be able to use the app for online shopping, but the research institute also is working to develop similar face-recognition devices for physical payments made at brick-and-mortar retailers.

The app will make use of the world's largest database of Asian faces, which has more than 50 million records.

Zhou says the app's data-collecting technology can simultaneously gather facial information from 91 different angles. He also notes the system scored the highest accuracy, 99.8 percent, according to a Carnegie Mellon University international standard, surpassing its predecessor's high of 97.6 percent.

The facial-recognition system already has been used at border controls and attendance machines.

Zhou says the algorithm will be very reliable, and can identify scanned faces as either real ones or a held-up photo or video. The system also features online learning abilities, and will automatically study the subject's shifting characteristics, thus preventing weight changes from affecting the results, for example.

From ZDNet
View Full Article

 

Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account