Face recognition vendor test (FRVT) part 7 :

One-to-many biometric search systems are discussed in their role of positive and negative identification - the former refers to the expectation that person in a probe sample is present in the database (as in access to an office) while the latter presumes the person is not (as in compulsive gamblers entering a casino). The distinction is useful because the application differ in their tolerance for false negatives and false positives. This report addresses the positive use of one-to-many facial recognition in airport transit settings in which travelers' faces are matched against galleries of individuals expected to be present.

We primarily consider the case where face recognition serves double-duty for access control (to an aircraft) and facilitation (of recording a visa-holder's exit). This report summarizes three NIST activities: First, to describe the biometric aspects of the immigration exit application and factors that are expected to affect its performance; second, to document results from running offline simulations in which recent accurate face recognition algorithms are applied to actual ENTRY and EXIT images with the goals of establishing a methodology, estimating accuracy, and exposing some factors that will affect those estimates; third to consider whether the use of face recognition at other airport touchpoints airport where higher populations are expected.