News in Brief

NIST Advances Face Biometrics Evaluation Algorithms

The US National Institute of Standards and Technology (NIST) has released its ‘Face Analysis Technology Evaluation (FATE) Part 11 report’ ¹ on the assessment of face image quality vectors, a critical facet in advancing face biometrics technology.

The report observes substantial improvements in image quality assessment algorithms submitted for evaluation. These algorithms play a pivotal role in detecting specific defects that could significantly impact the success of face biometrics matches. Of particular significance were contributions from Secunet Security Networks (secunet), coupled with modified measures aligning with ISO/IEC standards, which proved instrumental.

Submissions were entered by Digidata, Neurotechnology, Fraunhofer IGD, IDEMIA and Seamfix, along with two from secunet.

What stands out in the report is the acknowledgment that all 13 submitted algorithms, including the two from secunet, exhibited varying degrees of success in measuring a diverse spectrum of quality- related parameters. In particular, some algorithms demonstrated a marked reduction in false non-match rates (FNMR) when used for discarding lower quality images.