FaceTec blows away top NIST FRVT facial recognition algorithm in 3D self-testing
FaceTec‘s ZoOm 3D matching algorithm has scored in its own test an accuracy rate far better than the top algorithm in NIST’s biometric facial recognition vendor test (FRVT). The FaceTec test was created to simulate as closely as possible the 1:1 portion of the 2D NIST test by false acceptance rate (FAR). The test result of 1/4,200,000 FAR at less than 1 percent false rejection rate (FRR) is 668 percent better than the top algorithm in latest NIST’s FRVT results, and the best result ever in this category of the test, according to the company announcement.
The FRVT does not currently have the ability to test 3D systems, like FaceTec’s ZoOm 3D algorithm or Apple’s Face ID, according to FaceTec. The company has published a white paper on its test results and methodology. The white paper indicates much better performance than Face ID and the OpenFace DNN.
“Our ZoOm algorithms fully utilize 3D FaceMap data, and the results are exceptional,” says FaceTec CTO Josh Rose. “When we compared their performance, the operating point (combination of FAR & FRR) for our 3D face matching algorithm was at least 668 percent better than the operating point reported by the number one NIST algorithm.”
FaceTec also provides its proprietary liveness detection with its face biometric solution, ZoOm. It is the first face biometric to have been certified by iBeta Quality Assurance for both Level 1 and Level 2 in the Presentation Attack Detection (PAD) test, both of which it passed with 100 percent success.
“ZoOm’s 3D FaceMaps are created with standard 2D cameras, but contain much more data than a flattened photo ever can,” comments FaceTec CEO Kevin Alan Tussy “Think of copper phone lines used for both dial-up and DSL. Just as the latter achieves vastly better results, ZoOm utilizes the same ubiquitous camera hardware to outperform 2D face matching in every way.”