Neurotechnology, a provider of deep learning-based solutions and high-precision biometric identification technologies announced the release of a new version of their Face Verification system.
Face Verification 13.1 includes a multi-platform SDK and a web service designed for the integration of high-quality face capture, secure facial authentication, and robust face liveness detection into mobile and web applications.
It includes a small-footprint version of Neurotechnology’s facial recognition algorithm which ranked among the top results in the NIST FRVT 1:1 evaluation for use in digital identity onboarding, payment, banking, telecommunications, and other face recognition applications on personal devices.
One of the main features of the Face Verification system is the ability for the user to enrol his own face image in compliance with the ICAO recommendations as listed in the ISO 19794-5 standard. The latest version enhances this functionality with several new capabilities, including detecting if a person wears a hat/cap or if they wear glasses.
In such cases, Face Verification may generate a warning that can be used to mark, in the onboarding process, the pictures that do not conform to the expected face quality requirements set in the application.
The new version of Face Verification also adds the ability to estimate the age of the person performing the enrolling process and validate if the applicant conforms to public or private age regulatory requirements.
“We have extensive and proven experience with identity validation using our facial recognition algorithms in large-scale projects with databases of millions of people,” said Antonello Mincone, Business Development Director for Neurotechnology. “We have included some of the same robust technology in our new Face Verification system, enriching the capabilities for managing the self-enrollment process in both public and private sector digital onboarding applications.”
The web service component of Face Verification 13.1 also supports GPU capabilities for both the decoding of video streams and for the inference engine used in the various steps of face recognition. Through GPU usage, the new version of Face Verification dramatically increases the number of concurrent requests that are manageable by a single server.
