Aware Outperforms in NIST’s FATE Presentation Attack Program

Aware, Inc. says it has attained the top ranking in security for the “TYPE 4” presentation attack in the National Institute of Standards and Technology (NIST)’s Face Analysis Technology Evaluation (FATE) program.

FATE is one of two branches of NIST’s evaluation programs for face biometrics technology, after the institution split its previous FRVT program into two tracks at the end of summer. The bifurcation is aimed at delivering a more nuanced evaluation of face biometrics technologies, segregating them based on their core functionalities.

The first track, dubbed the “Face Recognition Technology Evaluation” (FRTE), is tailored towards algorithms engaged in the processing and analysis of images. On the other hand, the “Face Analysis Technology Evaluation” (FATE) is designed to delve into identity verification specifics. This strategic split is intended to draw a clear demarcation between facial recognition technologies and other face biometrics technologies that aren’t utilized for biometric matching, like age estimation tools.

The NIST FATE program evaluates various PAD technologies against nine different types of presentation attacks, although the specifics of each type are not fully disclosed. For instance, while the report details that “PA Type 3” corresponds to a “flexible silicone mask” and “PA Type 8” to a “photo print/replay attack,” there’s no available information on the nature of TYPE 4 PA and several other types.

This lack of full disclosure stems from NIST’s methodology to encourage broad PAD effectiveness and deter tuning to specific attacks, a strategy aimed at fostering a comprehensive understanding of PAD technologies across a spectrum of potential real-world threats.

In any case, Aware’s success in outperforming the competition against at least one type of Presentation Attack speaks to the sophistication of its technology. The evaluation program had 82 algorithms competing in the impersonation detection task, and 27 in the evasion detection task.

“Aware strongly believes in building systems in tune with the operational conditions we encounter in our daily lives,” said Aware CTO Mohamed Lazzouni. “As such, providing an optimal experience to balance security and convenience is critical. This is why Aware participated in both impersonation and evasion tasks and performed exceptionally well in both.”

–

October 12, 2023 – by the FindBiometrics Editorial Team