IDEMIA has announced that its facial recognition algorithm 1:N came top among 114 tested systems in NIST’s latest FRVT1. FRVT test results are acknowledged to be the gold standard of the global security industry. The results reconfirm the company’s ongoing 40-plus year history as an industry leader in biometrics and illustrate its place as the leader in biometric technology and innovation.
The test results reported facial recognition technology performance in one-to-many (1:N) scenarios, comparing one face image against a large database of face images to return one or more matching images. IDEMIA achieved the best accuracy score of 99.88% correct matches out of 12 million face images.
One of the important aspects of AI-based automated facial recognition is to teach its various algorithms not only to be accurate but equally important fast and optimized for fairness. It is all about striking the right balance when dealing with large volumes of face images. IDEMIA’s facial recognition solutions work with or without a mask and with the best trade-off between speed and accuracy, demographic parity, and can process face profile images. FRVT evaluated IDEMIA’s core algorithms underlying all its systems using facial recognition which addresses access control, public security and border control needs, proving that IDEMIA has one of the leading identification systems on the market.
IDEMIA is truly committed to the highest level of technology to provide fast, accurate, and fair solutions. These FRVT results come on the heels of other high-ranking NIST benchmarks already recently achieved, further demonstrating IDEMIA’s technology leadership.
The following awards speak to IDEMIA’s innovation and best-in-class solutions in security and public safety:
- FACE: #1 in Fairness the latest NIST Face Recognition Vendor Test (FRVT) in 2022;
- FACE acquisition: #1 in the US Department of Homeland Security annual’s biometric technologies rally in 2022;
- FINGERPRINT: #1 in mFIT Challenge in 2022.
- FINGERPRINT: #1 on most of data sets in the latest PFT III benchmark (Proprietary Fingerprint Template) for our AFIS-class algorithms in 2022.
- IDEMIA’s algorithms achieve the most accurate performance for fingerprint and palmprint datasets, with 7-60% better accuracy than any other algorithm at NIST’s ELFT test in 2022.
“These results are further evidence that IDEMIA is committed to the highest level of technology for its AI solutions in line with its mission to make the world safer through developing state-of-the-art identity technologies. We are very proud to outperform the market in line with our long-standing and ongoing 40-plus-year pledge to provide law enforcement agencies with the very best systems on the market,” says Jean-Christophe Fondeur, IDEMIA’s Chief Technology Officer.
