Liveness Checks: The Future of Online Identity Verification
July 2, 2020
Online KYC has come a long way – from sending printed paper proof that was manually examined, we have reached a point where KYC onboarding happens in a matter of few clicks using the camera of your phone, pictures, and videos. Whilst this has been a brilliant first step for online KYC, it’s proving to have its shortcomings.
Passive liveness checks have been popular amongst online businesses as they have significantly reduced operational costs, abandonment rates, and onboarding time. Where’s the problem, you might ask? Online fraudsters and scammers have targeted the new technology and they have slowly yet steadily picked it apart.
For the most part, liveness technology is structured around 2D projections that analyze images as flat. Whilst that type of tech can still detect movement, it can’t scan and process for depth and micromotions.
That immediately opens the door to several copycat techniques that are used to bypass the test: masks, printed images, pre-recorded videos, and avatars. What’s the solution? An active liveness check.
Active Liveness Check
Active liveness detection mechanisms require the user to follow a more demanding process. These mechanisms usually consist of mirroring a certain movement on camera, smiling tilting your head, or replicating a gesture with your hand.
Implementing these mechanisms has proven to be bittersweet for companies. On one hand, there’s no doubt that they raise the bar and quality of the identity verification process. On the other hand, they impede the speed and efficacy of the onboarding process, increasing abandonment rates.
ID R&D’s whitepaper titled “The Important Role Of Liveness Detection In Face Biometric Authentication”, companies report abandonment rates as high as 50% for active liveness. The difficulty for companies does not stop there.
Companies have to also consider the industry standard and regulation. The ISO/IEC 30107-3:2017 dealing with information technology — biometric presentation attack detection — testing and reporting, has been coined as tough and challenging to pass. The ISO/IEC 30107-3:2017 is not the only benchmark around.
The NIST contest known as FRVT 1:1 is one of the most reputable and credible benchmarks in the biometric and face recognition industries. Being ranked in their database is a hefty goal for companies, making the conundrum between quality onboarding and commercial success even tougher to crack.
Where Do Liveness Checks Go From Here?
As you might have realized, there are two camps in the liveness check world: the commercial side and the technology side. From a commercial standpoint, companies want onboarding to be fast, easy, and user-friendly, converting customers.
The technology side of the equation is a bit more cynical, focusing on the security of information, trying to make the onboarding process as robust and bullet-proof as possible. In an ideal scenario, a provider should be able to combine the best of both worlds and give you a mixture of passive and active liveness checks for optimal security and user experience.
For more information on our technology, feel free to contact a member of our team.