3D face recognition tech firm FaceTec, has announced that its ZoOm 3D Face Login is now the first – and only – biometric to achieve a Level 1 rating in the NIST/NVLAP-certified iBeta Presentation Attack Detection (PAD) Certification test.
The test determines the effectiveness of anti-spoofing Liveness Detection technology in compliance with the ISO 30107-3 global standard.
Over 1,500 spoof sessions utilizing hundreds of high resolution photos and videos were attempted over six days of testing.
No spoof attempts were able to fool the system and ZoOm achieved a perfect 100% anti-spoofing “score”.
Further adding to the challenge, the ISO 30107-3 standard requires the test subjects be “fully cooperative”, and provide iBeta with “any and all” biometric data requested.
This testing method ensures the authenticator is strong enough to prevent complicit user fraud and thwart phishing attacks.
The iBeta ZoOm 3D Face Login SDK PAD Certification Test Report Executive Summary can be viewed here.
What is ZoOm’s 3D FaceMap
ZoOm’s patented process scans a 3D face with a 2D camera to ensure the real user is present and that they match their previous enrollment.
Measurable perspective distortion occurs when a camera is moved closer to a 3D human face, therefor ZoOm captures much more human signal than photos or videos because the FaceMap includes data from unZoOmed, transitional and fully ZoOmed frames.
The ability to reliably determine 3D depth and human liveness makes ZoOm vastly superior to fingerprint, palm, iris and 2D Face.
(For all the rocket science behind ZoOm that makes it so incredibly secure, it’s as easy to use as taking a selfie. Please watch our short ZoOm explainer video to see just how simple it is to keep your personal information safe! Courtesy of FaceTec and YouTube. Posted on Nov 8, 2017.)
When addressing biometric privacy concerns, it is critical to understand the significance of Liveness Detection.
When a biometric authenticator is truly spoof-proof against digital media, 2D representations of a user won’t fool the system, so bad actors can’t take readily available photos from Facebook or videos from YouTube and use them to impersonate the user.
When a biometric system demands the real 3D user be present at the time of authentication, no photo or video will suffice… 2D Spoof attempts are futile and stolen user media is useless.
ZoOm detects the difference between “likeness” and “liveness” using Artificial Intelligence, and requires dozens of real human characteristics be observed concurrently.
These characteristics cannot be captured and reproduced without measurable generation loss and thus cannot be phished or shared like passwords.
(Learn More. See how easy it is for you to create a realistic 3D animated spoof from a 2D photograph using CrazyTalk software. Courtesy of FaceTec and YouTube. Posted on Jun 19, 2018.)