Publisher's Synopsis
This document and associated dataset is an update to the Multiple Encounter Dataset I (MEDS-I), originally published by the National Institute of Standards and Technology (NIST) in May 20101. The MEDS is a test corpus organized from an extract of submission files of deceased persons with prior multiple encounters. This paper describes the basic properties of the images and some relevant image quality characteristics that pertain to collection practices and the calibration and evaluation of face recognition technology.