CyberLink’s face biometrics solution passes iBeta ISO liveness detection test
CyberLink has announced that its FaceMe facial recognition technology obtained a perfect score in an iBeta Quality Assurance liveness detection test, confirming the solution is complaint with ISO/IEC 30107-3.
An announcement from the company reveals that during the test, FaceMe earned a 100 percent True Rejection Rate in iBeta’s industrial standard Presentation Attack Detection (PAD) test, as it was able to successfully detect all biometric impersonation or substitution attempts from the photos and videos used.
Based on the ISO/IEC 30107-3 standard, iBeta, an independent third-party testing and quality assurance organization, performed Level 1 testing of FaceMe on both Android and iOS devices, and was not able to spoof the biometric technology with any presentation attacks, resulting in an Attack Presentation Classification Error Rate (APCER) of 0 percent.
The compliance also gives users of mobile devices that run on FaceMe technology safety and privacy assurances for their private data stored on their device.
The iBeta certification, CyberLink notes, comes to add to the list of FaceMe’s achievements which underscore the solution’s reliability and its ability to accurately protect people’s identity, even when they come under spoofing attacks.
Dr. Jau Huang, CEO of CyberLink, said the company is honored to receive the iBeta PAD compliance confirmation. “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.”
Huang also says the successful PAD test, combined with CyberLink’s accuracy as recognized by NIST, provides strong assurance against unauthorized access.
CyberLink recently sealed a deal for FaceMe to be integrated into FaceScan’s fever screening kiosks.
The many masks of Mastercard
A researcher who works for Mastercard Cyber & Intelligence has a mask custom-made by an artist who worked on the film ‘Avatar’ to use in biometric spoof attacks tests as the company seeks to improve authentication and security systems.
An article from the company describes how Rajat Maheshwari and his team attempt to anticipate attacks on biometric systems, and whether they will be successful. They also sometimes discover other effects on biometric systems, as when they realized that a repeatedly failed palm scan was due to a number written in the subject’s hand in invisible ink by a club the person had attended the night before.
Mastercard attempts to ensure that all biometric systems it works with have false match rates (FMRs) of 0.01 percent or lower, and false non-match rates (FNMRs) of 3 percent or lower.
biometric liveness detection | biometric testing | biometrics | CyberLink | face biometrics | facial recognition | iBeta | ISO standards | Mastercard | presentation attack detection | spoof detection