Ongoing Face Recognition Vendor Test (FRVT) Part 2: Identification
Patrick J. Grother, Mei L. Ngan, Kayee K. Hanaoka
This report documents performance of face recognition algorithms applied to the one-to-many identification of faces appearing in portrait images. The primary dataset is comprised of 26.6 million reasonably well controlled live photos of 12.3 million individuals. Three smaller datasets containing more unconstrained photos are used also. The report will be useful for the comparison of algorithms, and for assessing absolute capability of face recognition with portrait images.