NIST Study Shows Accuracy Improvement for Masked Face Scanning

Software developers “have made demonstrable progress at recognizing masked faces” using face-scanning technology, the National Institute of Standards and Technology reported Tuesday (see 2007270065). The highest scoring technology made errors between 2.4 and 5% of the time on masked faces, "comparable to where the technology was in 2017 on nonmasked photos,” said NIST study author Mei Ngan. Error rates dropped by as much as a “factor of 10 between their pre- and post-COVID algorithms,” she said. NIST tested technology from Canon, Intel, Panasonic, Samsung and dozens of others.