With #1 ranking, Paravision outpaces global leaders in AI, highlighting computer vision excellence focused on ethical use

Paravision, the U.S.-based leader in mission-critical computer vision, today announced that it has ranked as the most accurate face recognition vendor in the world in NIST’s January 2022 1:N FRVT Report. The test benchmarks Paravision’s upcoming 5th Generation face recognition technology and shows a 70% reduction in error rates compared to the company’s previous submission. 

The NIST FRVT reports are extensive and test many aspects of how face recognition algorithms perform in challenging use cases. In its primary 1:N Identification benchmark, NIST evaluates face recognition accuracy in a Visa / Border test, with high-quality enrollment images compared against a large dataset of images taken of people in challenging travel environments often characterized by varying pose angles, low spatial resolution, and low contrast. This reflects a common identification scenario in a secure environment, such as an airport. 

In this Visa / Border test, Paravision ranks #1 globally for accuracy with an error rate that is 18% lower than the second-ranked vendor, delivering a False Negative Identification Rate of 0.22% at a False Positive Identification Rate of 0.3% across a dataset of 1,600,000 images.


Paravision also ranks #1 globally in NIST’s leaderboard of face recognition at self-service immigration kiosks. This 1:N identification test compares high-quality enrollment images against a database of 1,600,000 kiosk photos characterized by a pitch-down pose often exceeding 45 degrees, poorly cropped images of short or tall people, and heavy background illumination. 

Since its first submission to NIST FRVT 1:N in 2018, Paravision has shown rapid and continuous improvements in accuracy. Over the last 3 years, Paravision’s error rates have decreased by more than 95%, including a 70% reduction in just the last year.

Accurate face recognition is vital for security, efficiency, and enhanced user experiences across so many industries and applications. It comes down to trust – trust that the technology works for everyone, is consistent, reliable and accurate. That’s why the NIST FRVT reports are so important. They are the independent gold standard when it comes to transparency in performance.
Charlie Rice, Paravision CTO

Paravision’s computer vision technology is deployed through its partners for use cases involving access control, identity verification, aviation, travel and borders, digital payments, government programs, and national security. Paravision’s ethics committee reviews its potential deployments to ensure alignment with the company’s AI Principles, which state its AI must be ethically trained and conscientiously sold.

Paravision is now independently recognized as the most accurate face recognition platform in the world and the only U.S. company in the top ten. Given the scale of legacy providers and the massive funding and resources going specifically to Chinese AI firms, the ranking is a huge testament to our approach. It’s an inflection point for our industry – trusted, ethical and accurate face recognition from a truly partner-focused company.
Doug Aley, Paravision CEO

Paravision’s 5th generation face recognition technology is planned for official release in Spring 2022. As well as the dramatic improvements in speed and accuracy, the new product set will include enhanced support for mobile devices, Edge AI and other toolkits to accelerate the deployment of face recognition in a range of applications.