Aware Ranks Top Performer in NIST Face Analysis Technology Evaluation (FATE) Benchmarking Test While Optimizing Demographic Parity - Seite 2

Aware achieved the rank #1 in security for both impersonation and evasion detection for the TYPE 4 PA.

NIST did not disclose all the PAIs used in the evaluation to encourage broad PAD effectiveness and to discourage tuning to specific attacks. NIST explained the rationale behind the choice to not disclose the evaluation method because it “reflects the operational reality that attackers do not advertise their methods.” Detection metrics for each PAI are reported but most are without description of the PAI. “PAs used to evaluate impersonation detection were also used to test evasion detection (by definition, impersonation is also a form of evasion).”

While accuracy and convenience are both important to a PAD solution, they are not sufficient. Speed and fairness are equally critical and Aware performed exceptionally well. Eliminating racial, gender, and other biases to achieve optimal demographic parity has been Aware’s focus for years in both face recognition and presentation attack detection. The results show that Aware continues to lead in this area.

“Aware applauds NIST for both the rigor and the thought leadership demonstrated in designing this evaluation,” says Dr. Mohamed Lazzouni, Chief Technology Officer at Aware. “Aware strongly believes in building systems in tune with the operational conditions we encounter in our daily lives. 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.”

Aware continues to demonstrate that the use of deep learning and sophisticated fusion strategies are critical building blocks for biometric artificial intelligence (AI) systems to advance current algorithms and build next generation ones. Aware’s core algorithms, as well as the systems built on them, use facial recognition and presentation attack detection deployed around the world. Aware has leveraged the operational scenarios to teach the AI-based algorithms to continue to improve performance and adapt to emerging threats. Aware does so without compromising customer experience or fairness.

“Aware’s performance would not have been possible without the continuous support of our customers,” continues Lazzouni. “They’ve provided invaluable feedback throughout the years and assisted us in making Aware one of the leading PAD systems on the market. We will continue to demonstrate that the use of deep learning and sophisticated fusion strategies are critical building blocks for biometric AI systems to advance current algorithms and build next generation ones.”