IARPA Announces Facial Recognition Fusion Challenge

“The challenge is open to domestic and international interests from industry and academia, and will offer cash prizes to the most effective solutions…”

IARPA Announces Facial Recognition Fusion Challenge

The Intelligence Advanced Research Projects Activity (IARPA), a division of the Office of the Director of National Intelligence, has announced another industry challenge in a bid to push the state of the art of biometric identification. Dubbed “FOFRA”, for “Fusion of Face Recognition Algorithms”, the challenge is aimed at combing multiple facial recognition algorithms on the same input imagery.

In a statement announcing the challenge, IARPA program manger Chris Boehnen explained that the fusion of multiple facial recognition algorithms has been shown to improve accuracy. “IARPA is in a unique position to provide data for researchers to develop fusion methods that should further improve performance,” he said. “These fusion schemes could allow users to leverage multiple algorithms without overhauling or re-designing the core technologies.”

The challenge is open to domestic and international interests from industry and academia, and will offer cash prizes to the most effective solutions from a purse totaling $70,000. The registration period will run until August 6th, and winners will be announced in October.

The FOFRA challenge arrives a little over a year after the announcement of IARPA’s Face Recognition Prize Challenge, which was aimed at refining facial recognition software for “non-cooperative subjects.” IARPA has organized FOFRA in partnership with the National Institute of Standards and Technology (NIST), which itself issued an industry challenge last year concerning full-fingerprint scanning, nail bed to nail bed.

July 13, 2018 – by Alex Perala