Independent Vendor Tests
Cognitec Systems continues to advance the accuracy and speed of its face matching algorithm, and participates in renowned vendor tests for independent performance validation. Cognitec submitted its current algorithm B12 to the third round of the NIST Face Recognition Vendor Test (FRVT) 1:N in 2018, with results to be published soon.
FRVT 2019: Cognitec announced its successful participation in the NIST FRVT: Identification, confirming the solid market position of our matching technology, and the achievements of our algorithm research and development. The September 2019 results, comparing 203 algorithms from 51 commercial developers and one university, show Cognitec’s high performance in accuracy tests for large galleries of frontal mugshots and of unconstrained images (Wild).
FRVT 2018: Cognitec reported remarkable leaderboard positions in the Ongoing FRVT 1:1, ahead of all face recognition companies with established market presence and products. The November 2018 results, comparing 100 algorithms from 57 developers, show Cognitec’s high performance for accuracy tests with visa photos. In addition, Cognitec‘s template generation was the fastest among the highest ranking algorithms with false non-match rates lower than 0.01, at a false match rate of 0.0001.
FRVT 2013: Test results on the performance of automated age estimation algorithms confirmed Cognitec Systems’ leading position in the face recognition market. The test compared nine algorithms submitted by six participants, five companies and one university, and applied them to seven million images. Results show that Cognitec’s algorithm performs with the highest accuracy for all age groups. Most notably, the algorithm shows superior performance “in the youth and senior age groups, leading the next most accurate algorithm in 5-year accuracy by 30% and 16%, respectively,” according to the report.
MBE 2010: In the latest NIST test, conducted in 2010, seven companies and three research institutes participated in the MBE-STILL evaluation, comparing the performance of face recognition algorithms on still images. Cognitec again delivered excellent performance in all the tests. In the open-set identification test, the most relevant for challenging identification applications, Cognitec's algorithm had the best performance at the lowest false alarm levels, which corresponds to the operational setting with a multi-million gallery and limited resources for visual inspection.
FRVT 2006: The Face Recognition Vendor Test 2006, again conducted by NIST, confirmed the outstanding performance of Cognitec's technology. Of 22 companies and research institutes, only Cognitec and 10 other participants could complete the large-scale tests, and only Cognitec and one other company could deliver measurements in four
different test scenarios for still images and 2D/3D data.
The results also demonstrated that Cognitec exceeded the goal specified by the preceding Face Recognition Grand Challenge, or FRGC, which was to prove increased recognition performance by an order of magnitude as compared to the FRVT 2002. Cognitec again delivered top performance in the majority of the tests, especially with regard to common real-world scenarios (low resolution imagery).
FRVT 2002: At the Face Recognition Vendor Test 2002, conducted by the National Institute of Standards and Technology (NIST), Cognitec's face recognition software exhibited the best performance among all participants for large-scale verification, identification and watch list scenarios.Disclaimer: FRVT and MBE results do not constitute endorsement of any particular system by the U.S. Government. Download complete test results at face.nist.gov.