Paravision scores leading combined error rate in DHS S&T’s liveness detection demo

Paravision demonstrated the leading accuracy and convenience of its new presentation attack detection (PAD) software when it scored the lowest combined error rate in the recent benchmark by the U.S. Department of Homeland Security’s Science and Technology Directorate (S&T), according to a company announcement.

The company has identified itself as vendor “P8” in anonymized Remote Identity Validation Technology Demonstration (RIVTD) Track 3: Liveness Detection, one of 15 participating passive PAD vendors.

The test found Paravision’s Liveness scored a bona fide presentation classification error rate (BPCER) of 0.3 percent and an attack presentation classification error rate (APCER) of 3.3 percent. The combined 3.6 percent error rate was far lower than the second-lowest combined error rate of 8.5 percent, according to the announcement. Paravision’s software also worked in less than a second, by average run time, which was faster than any active PAD systems.

DHS S&T also found that passive PAD systems including Paravision’s delivered consistent performance across demographic groups, while most active systems showed a higher BPCER for older users.

“Paravision’s performance in the DHS RIVTD Track 3 proves that security and convenience are not mutually exclusive,” said Joey Pritikin, Chief Product Officer at Paravision. “With the latest approaches to Identity AI technology, new user experiences and security paradigms are made possible — offering rapid, reliable, and user-friendly fraud prevention.”

The company launched its facial liveness detection software commercially just over a year ago, after it passed a Level 2 assessment by iBeta for compliance to the ISO/IEC 30107 standard.

Article Topics

biometric liveness detection  |  biometric testing  |  biometrics  |  DHS  |  DHS S&T  |  face biometrics  |  Paravision  |  presentation attack detection  |  Remote Identity Validation Technology Demonstration (RIVTD)