STCon edges in among established facial recognition accuracy leaders in NIST 1:N

A handful of new facial recognition algorithms have been added to the NIST FRTE 1:N Identification this year, but most of the leading submissions by biometric performance are the same as they were at the end of last year.

The one exception is STCon, which sits third in the VISA-KIOSK category as of the mid-February update, and in the top five in three others.

The software submitted at the end of 2024 by QazSmartVision still has the lowest false negative identification rate (FNIR) across several categories. NEC, Cloudwalk, Megvii, Innovatrics, Paravision, Idemia, Toshiba and Panasonic all placed in the top 10 of the VISA-BORDER category, and each appears near the top of the list in multiple others as well.

Viettel Artificial Intelligence and GTEL OTS JSC each submitted algorithms in February. Brazil’s Protenet, STCon, RealSense, Identy and India’s Adiance submitted in January.

While several of these systems show strong accuracy, particularly RealSense and Viettel, their results indicate they are most likely appropriate for narrowly targeted or niche applications.

Article Topics

biometric testing  |  biometrics  |  Face Recognition Technology Evaluation (FRTE)  |  facial recognition  |  NIST  |  STCon