US Intelligence Wants To Radically Advance Facial Recognition Software

coondoggie writes "Identifying people from video streams or boatloads of images can be a daunting task for humans and computers. But a 4-year development program set to start in April 2014 known as Janus aims to develop software and algorithms that erase those problems and could radically alter the facial recognition world as we know it. Funded by the Office of the Director of National Intelligence's 'high-risk, high-payoff research' group, Intelligence Advanced Research Projects Activity (IARPA) Janus 'seeks to improve face recognition performance using representations developed from real-world video and images instead of from calibrated and constrained collections.'"

Comments

Isn't going to help I expect, but..

[rusty0101]

Get Poser, or something similar, and start replacing the face picks of all your contacts with pics of poser models asses selected for a best match to the contact's ass. Remember to find an appropriate image for companies and agencies. I'm thinking a Hydra would be appropriate for the NSA, Medusa for the FBI, Mantis for the CIA, etc.

Bonus points for doing r/g stereo of the images, or 3d if the phone supports that directly.

Re:Not much need to worry

[Anonymous Coward]

Every picture on Facebook is scanned by Interpol by facial recognition software. Yes even the guy in the background that didn't know you took the picture. Interpol you say? Yep, that's why this whole fake outrage over the NSA is all bunk. They all work together, and have been for decades, spying on each other then sharing the information to bypass privacy laws in each country.

Re:Not much need to worry

[sifi]

Absolutely true. Even when they are 'accurate' they are of limited usefulness.

Assume that it's 99.9% accurate for a given success rate (wildly optimistic) That is for every 1000 faces you show it 1 is incorrectly flagged as matching.

Suppose that you have a list of 100 people you are 'interested' in. If the system is in an airport with 200,000 people per day - you are going to get 20,000 incorrect matches a day.