By Planet Biometrics
A Russian and Chinese firm both scored highest in a face-recognition competition run by the US National Institute of Standards and Technology (NIST) and the Intelligence Advanced Research Projects Activity (IARPA).
The winner of the face-verification task was Russia’s Ntech whose FindFace product can match a person’s face correctly 99.9 per cent of the time.
The face-search task was won by Shanghai start-up Yitu Tech.
NTechlab won in the categories of “identification speed” and “verification accuracy”—the latter refers to an ability to verify that two photos are of the same person.
(Learn More. Courtesy of Fox News, FindFace and YouTube. Posted on Dec 28, 2016)
Yitu, won in the “identification accuracy” category for its software’s ability to match a face to a specific identity.
IARPA launched the challenge so it could assess software’s ability to quickly and passively identify faces in the “wild,” including from public security footage.
An internal IARPA research program called Janus is developing its own facial recognition technology.
The majority of the 16 companies participating in the challenge were international, program manager Chris Boehnen told Nextgov.
And while he’d expected participation from commercial tech companies in the United States, Silicon Valley stalwarts like Google, Facebook and Amazon “tend to be reluctant to participate.”
NIST plans to release a report on the solutions presented.
Original post http://www.planetbiometrics.com/article-details/i/6527/
(Learn More about the US National Institute of Standards and Technology (NIST), courtesy of NIST and YouTube)
Face Recognition Vendor Test (FRVT)
The FRVT aims at measurement of the performance of automated face recognition technologies applied to a wide range of civil, law enforcement and homeland security applications including verification of visa images, de-duplication of passports, recognition across photojournalism images, and identification of child exploitation victims.
In all cases the input image will contain one face only.
NIST will report the dependence of performance on the properties of the images and the subjects.
In its initial form, FRVT has one assessment track, for face verification.
Find out More…