“…until now there was no recognized standard for the testing of 3D face matching.”
FaceTec is once again blazing a trail in the field of face biometrics, this time by leading the charge into the standardized evaluation of 3D face matching technology.
It’s working with A.C.C.S. Ltd., a UKAS-accredited testing body that has so far specialized in certifying compliance with the UK’s Digital Identity and Attributes Trust Framework. Now, in contractual relationship with FaceTec, the agency is going to perform real-world N:N testing of 3D face matching algorithms.
This latest testing news was announced alongside a remarkable performance benchmark: FaceTec’s new neural network model achieved a False Accept Rate (FAR) of 1-in-125 million.
The A.C.C.S. collaboration is an important development because, as FaceTec explained in a statement announcing the collaboration, until now there was no recognized standard for the testing of 3D face matching.
The National Institute of Standards and Technology’s testing programs are widely considered the industry standard in the evaluation of biometric technologies, of course. But NIST’s face matching protocols are based on standards that were instituted in 1994, and they are limited to assessing 2D data.
The kinds of 3D face matching systems used by the likes of Apple and FaceTec are a different kettle of fish, but until now, the companies had to be taken at their word with respect to claims about their accuracy.
“Our algorithms utilize the unique shape of the user’s 3D Face, not just 2D face data, and the results are exceptional,” asserted FaceTec CTO Josh Rose. “When we compare them, our latest 3D face matching algorithm is approximately 8,600% better than the #1 NIST FRVT 2D algorithm.”
Soon, A.C.C.S. will be in a position to validate that claim independently.
“We are committed to the certification of the latest, state-of-the-art biometrics, age estimation, and identity verification systems,” explained the firm’s CEO, Tony Allen, “and to that end, we are offering the first testing framework for 3D Face Matching accuracy, which will enable next-generation biometric modalities, like FaceTec’s, to be evaluated at scale.”
FaceTec’s engagement with A.C.C.S. offers the latest example of the company’s leadership in the industry, which has also been vividly illustrated through its Spoof Bounty Program, which offers cash payouts to any hackers who can successfully trick its facial recognition platform into a false positive. FaceTec recently doubled the program’s total potential payout, raising it to $200,000.
May 3, 2022 – by Alex Perala