Patent granted to ID R&D for way to increase liveness detection speed and accuracy
ID R&D has been granted a patent for a method of detecting spoof attacks against onboarding and biometric authentication systems that the company says increases the speed and accuracy of its IDLive Face liveness software.
‘System and Method for Face Spoofing Attack Detection’ was granted by the U.S. Patent and Trademark Organization. ID R&D says it furthers the competitive advantages of its passive liveness technology by lowering the abandonment rates associated with user friction added by spoof detection, and increasing customer satisfaction.
The patent describes a method in which the most important areas of a facial image are highlighted during analysis. The process is analogous to the recent innovation in AI in which visual transformers are used to focus the attention of neural networks on the most important data during training, according to the company announcement.
“Rapid advancements in artificial intelligence bring new challenges to securing digital onboarding and authentication — particularly from deepfakes — and we expect this trend to persist,” comments Alexey Khitrov, CEO and co-founder of ID R&D, a Mitek company. “Fortunately, we are also working hard to research and develop new ways to leverage AI to its full potential to counter AI-powered fraud.”
ID R&D points out that identity theft continues to be the most common among fraud types reported to the U.S. Federal Trade Commission.
The company also introduced IDLive Face Plus earlier this year to provide protection against other types of attacks, like digital video injection.