Army Tracking Plan: Drones That Never Forget a Face
Perhaps the idea of spy drones already makes your nervous. Maybe you’re uncomfortable with the notion of an unblinking, robotic eye in the sky that can watch your every move. If so, you may want to click away now. Because if the Army has its way, drones won’t just be able to look at what you do. They’ll be able to recognize your face — and track you, based on how you look. If the military machines assemble enough information, they might just be able to peer into your heart.
The Pentagon has tried all sort of tricks to keep tabs on its foes as they move around: tiny transmitters, lingering scents, even “human thermal fingerprints.” The military calls the effort “Tagging, Tracking, and Locating,” or “TTL.” And, as the strategy in places like Afghanistan has shifted from rebuilding societies to taking out individual insurgents, TTL has become increasingly central to the American effort. Hundreds of millions of dollars have been devoted to it.
The current technologies have their limits, however. Transmitters can be discovered, and discarded. Scents eventually waft away. Even the tagged can get lost in a crowd.
But there are some things that can’t be so easily discarded. Like the shape of your face. Or the feelings you keep inside. That’s why the Army just handed out a half-dozen contracts to firms to find faces from above, track targets, and even spot “adversarial intent.”
“If this works out, we’ll have the ability to track people persistently across wide areas,” says Tim Faltemier, the lead biometrics researcher at Progeny Systems Corporation, which recently won one of the Army contracts. “A guy can go under a bridge or inside a house. But when he comes out, we’ll know it was the same guy that went in.”
Progeny just started work on their drone-mounted, “Long Range, Non-cooperative, Biometric Tagging, Tracking and Location” system.
The company is one several firms that has developed algorithms for the military that use two-dimensional images to construct a 3D model of a face. It’s not an easy trick to pull off — even with the proper lighting, and even with a willing subject. Building a model of someone on the run is harder. Constructing a model using the bobbing, weaving, flying, relatively low-resolution cameras on small unmanned aerial vehicles is tougher still.
Read the full article at: wired.com