Perfect score at ISO-conforming iBeta’s liveness detection test cements FaceMe technology’s leadership
TAIPEI, TAIWAN — November 15, 2021 —CyberLink Corp. (5203.TW), a pioneer in AI and facial recognition technologies, announced today that the anti-spoofing capabilities of its FaceMe® facial recognition solution achieved a True Rejection Rate of 100%, obtaining a perfect score in iBeta’s industrial standard PAD (Presentation Attack Detection) test. FaceMe® detected all impersonation and substitution attempts from the test’s photos and videos. iBeta is an internationally recognized independent third-party testing and quality assurance organization.
In passing iBeta’s industrial standard PAD (Presentation Attack Detection) test, FaceMe is now compliant with ISO ISO-IEC 30107-3 (International Organization for Standardization). iBeta is one of the world’s few agencies accredited by NIST NVLAP (U.S. National Institute of Standards and Technology National Voluntary Laboratory Accreditation Program) for biometric testing. Per ISO-IEC 30107-3, iBeta performed Level 1 testing of FaceMe® on both Android and iOS devices. It was unable to spoof the technology with any presentation attacks (PAs), concluding with an Attack Presentation Classification Error Rate (APCER) of 0%.
Adding to CyberLink’s long list of achievements, the iBeta compliance confirms FaceMe®’s trustworthiness and its ability to accurately protect people’s identity, even when under spoofing attacks. The compliance with ISO provides users of mobile devices powered by FaceMe® the assurance that their identity, the private data stored on the device and the access it grants to their online information are securely protected.
“Preventing spoofing attacks is critical to the adoption of facial biometric technology in areas such as fintech and access control, where iron-clad identity protection is essential, yet it is one of its toughest challenges,” said Dr. Jau Huang, CEO of CyberLink. "Adding to FaceMe®’s acclaimed accuracy, consistently recognized by NIST and other independent parties, we are honored to receive iBeta’s Presentation Attack Detection compliance, which provides the assurance of its reliability in preventing unauthorized access through the use of spoofing techniques.”
FaceMe® is optimized to run across hardware configurations, from high-end workstations to low-power chipsets frequently used in IoT and AIoT devices. It is the most versatile and adaptable offering on the market today. FaceMe® provides solution builders and system integrators a fast, reliable, extremely precise and flexible facial recognition technology that can be deployed across a number of scenarios, including security, access control, public safety, fintech, smart retail and home protection.
CyberLink FaceMe® iBeta Presentation Attack Detection Report:
About CyberLink
Founded in 1996, CyberLink Corp. (5203.TW) is a leading provider of digital creative editing software and artificial intelligence (AI) solutions, inclusive of facial recognition. A driver of innovation, CyberLink has embraced the transformative power of AI, harnessing it to address the demands of consumer, commercial and education markets through a wide range of solutions. From the award-winning Director Suite’s PowerDirector, PhotoDirector and Promeo, to the acclaimed AI facial recognition engine, FaceMe®, CyberLink is committed to providing exceptional experiences. Empowering creators and marketers of all skill levels, Director Suite, elevates the content creation process, while FaceMe® delivers the fast and accurate facial recognition that is key to AIoT applications for security and surveillance.
To learn more about Director Suite, visit the CyberLink website: https://www.cyberlink.com/
For more information about CyberLink’s AI facial recognition engine, visit the FaceMe® website: https://www.cyberlink.com/faceme
####
Press Contacts
All companies and product names mentioned herein are for identification purposes only and are the sole property of their respective owners. Copyright © 2024 CyberLink Corp. All rights reserved.
