CyberLink’s FaceMe 1:N biometric accuracy among top-ranked by NIST

CyberLink Corp. announced its ranking as a global top ten vendor after its FaceMe AI facial recognition engine earned high marks during the latest Facial Recognition Vendor Test (FRVT). The National Institute of Standards and Technology (NIST)-administered FRVT is the most rigorous assessment for facial recognition precision within the industry.

CyberLink’s automated face recognition algorithm FaceMe promises flexible edge computing architecture optimized across devices and platforms, enabling a variety of use cases for smart banking, security, retail, and smart cities, the company says. Driven by deep learning algorithms, FaceMe brings reliable, high-precision, and real-time facial recognition crucial to AIoT applications.

The newest FRVT 1:N Identification report ranked the FaceMe algorithm behind those of nine other global vendors overall, after it scored a true acceptance rate (TAR) of 98.11 percent on its identity recognition against a collection of 1.6 million webcam and Visa photos. The company also notes that only three vendors from outside of China and Russia are ranked higher on the overall leaderboard.

CyberLink states FaceMe delivers efficient processing and higher levels of security than cloud-based solutions, and can run on low-powered CPUs for cost-efficient facial recognition on IoT and AIoT devices. The solution can also be integrated with high-end servers, workstations, and GPU-equipped personal computers.

“NIST’s independent evaluation establishes CyberLink as one of the de facto leaders in facial recognition research and development globally,” comments Dr. Jau Huang, CEO of CyberLink. “To date, FaceMe has been applied to a range of applications across multiple IoT/AIoT scenarios in the hospitality, retail, law enforcement, and financial industries. With the rising adoption of facial recognition in banks and fintech, we are ready to partner with industry leaders and create a safer ecosystem by deploying solutions such as electronic Know Your Customer (eKYC) to protect against identity fraud.”

Article Topics

accuracy  |  algorithms  |  biometric identification  |  biometric testing  |  biometrics  |  biometrics at the edge  |  CyberLink  |  Face Recognition Vendor Test (FRVT)  |  facial recognition  |  NIST