The National Institute of Standards says that facial recognition technology can be better than human identification with accuracy rates approaching 99 percent. Iris scanning, however, could improve.The report, dubbed the Face Recognition Vendor Test (FRVT) 2006 and the Iris Challenge Evaluation (ICE) 2006, evaluates a host of technology vendors, but doesn't rank them based on performance. GCN reported on the effort earlier this week.
From the report released in late March:
"The FRVT 2006 results from controlled still images and 3D images document an order-of-magnitude improvement in recognition performance over the FRVT 2002. This order-of-magnitude improvement was one of the goals of the preceding technology development effort, the Face Recognition Grand Challenge (FRGC). The FRVT 2006 and the ICE 2006 compared recognition performance from very-high resolution still face images, 3D face images, and single-iris images. On the FRVT 2006 and the ICE 2006 datasets, recognition performance was comparable for all three biometrics. In an experiment comparing human and algorithm performance, the best-performing face recognition algorithms were more accurate than humans."
The key takeaways seem to be:
Algorithm speeds to crunch the facial and iris data vary widely by vendor. Facial algorithms work faster. Iris algorithms show big disparities. One algorithm took 6 hours to complete NIST's tests. Another took about 300 hours.
Standard hardware was used to assess performance. The NIST report noted:
“For the ICE 2006 submissions, analysis was restricted to algorithms that could complete the large-scale iris experiments in three weeks of processing time on a single Intel Pentium 4 3.6GHz 660 processor. For the FRVT 2006 submissions, the large-scale requirements translated into restricting the analysis to fully automated systems; i.e., algorithms that do not need eye coordinates or other auxiliary meta-data.”
Organizations participating in the tests included: Animetrics, Carnegie Mellon University, Cognitec Systems GmbH, Diamond Information Systems, Geometrix, Guardia, Identix, Neven Vision, New Jersey Institute of Technology (NJIT), Nivis, LLC, Old Dominion University, Panvista Limited, Peking University, Center for Information Science, PeopleSpot, Rafael Armament Development Authority Ltd., SAGEM SA, Samsung Advanced Institute of Technology (SAIT), Tsinghua University, Tili Technology Limited, Toshiba Corporation, University of Houston and Viisage.