Algorithm development at Cognitec Systems GmbH (Cognitec) continues to engineer the optimal balance between speed and accuracy of the face matching processes.
NIST’s Face Recognition Vendor Test
The latest results of the U.S. NIST (National Institute of Standards and Technology) Face Recognition Vendor Test for identification tasks, show the Cognitec algorithm in the best position of all algorithms tested, when relating the template generation time to the false negative identification rate (miss rate), for mug shot databases.
The identification test addresses the largest market for face recognition applications
The identification test addresses the largest market for face recognition applications, including the detection of duplicates in image databases and fraud detection, during passport and driver’s licence applications. These tests apply a very high matching threshold, where only 0.3 % (3/1000) of probes, without a mate in the gallery, produces a false hit, one of the most difficult face recognition tasks.
Accuracy advances in face matching algorithms
“We are proud to also see significant accuracy advances, in comparison to the algorithm that we submitted to the last test in early 2021,” said Dr. Thorsten Thies, the Director of Algorithm Development at Cognitec Systems GmbH.
Dr. Thorsten Thies adds, “For the test, with 12 million mug shot images, Cognitec achieved rank 24 out of 165 algorithms, with a match rate of 98.5 %. In the test, with 1.6 million mug shots, with a 99.4 % match rate, we ranked 26 out of 299 algorithms. These results show remarkable performance consistency, regardless of database size.”
Cognitec-005 face matching algorithm
The September 2021 test evaluates face matching algorithms from 85 different vendors. Cognitec Systems GmbH submitted a new algorithm, with a revised face finder, called Cognitec-005 algorithm, in the report.