Ever AI Facial Recognition Gets Top Ranking in FRVT 1:1 Testing for Wild Images

“This independent evaluation establishes us as the de facto leader of mission-critical face recognition in the US.” – Doug Aley, CEO, Ever AI

Ever AI is celebrating an important win in the latest round of facial recognition testing conducted by the National Institute of Standards and Technology (NIST): The firm’s technology tied for first in the world in the Wild Images category for 1:1 matching.

It’s a considerable achievement, with NIST’s Face Recognition Vendor Test (FRVT) being the world’s premier forum for the evaluation of such technologies. FRVT 1:1 Ongoing’s Wild Images category concerns the capability of a facial recognition system to match faces from images in which subjects are noncooperative – often under poor lighting or resolution conditions, and with partial occlusion or awkward angles that can make face scanning more difficult.

Commenting on the performance in a statement, Ever AI CEO Doug Aley said NIST’s “independent evaluation establishes us as the de facto leader of mission-critical face recognition in the US.” Aley added that Ever AI’s flexibility in deployments, its training across diverse datasets (to reduce potential bias), and its custom models for customers, together with the technology’s accuracy, make it “the best option for authentication, access control, security and surveillance applications.”

Ever ID’s strong performance in this evaluation comes after other recent demonstrations of Ever AI’s technological sophistication. The company attained a score of 99.85 percent in the University of Massachusetts’ Labeled Faces in the Wild (LFW) evaluation, and a score of 99.04 percent against the University of Washington’s MegaFace benchmark.