The National Institute of Standards and Technology is updating its Face Recognition Vendor Test to include new sets of algorithms developed by private sector participants, Fedscoop reported Friday.
NIST, who launched the FRVT effort to establish facial recognition technology standards and evaluate industry-developed algorithms, expects to report about a factor of 3 reduction in error rates for mugshot identification in the coming months. That’s on top of the factor of 20 percent reduction in error rates between 2013 and 2018.
Mei Lee Ngan, computer scientist at NIST, told attendees at a Security Industry Association event that Microsoft, whose technology has been able to process front-facing pictures only, now has the capacity to match side photos and mugshots with a 91 percent accuracy. The Department of Homeland Security’s Biometric Technology Center conducted a similar effort called Biometric Technology Rally which made use of real-life scenarios and volunteers to test participants’ facial recognition capabilities.
“There’s a percentage of [algorithms] that are really good and very capable, but there is this long tail of facial recognition algorithm developers that just don’t do that good of a job,” Ngan said. “So the technology hasn’t been commoditized yet.”