If you're on a mission to avert facial-recognition technology, Berlin-based artist Adam Harvey has you covered - literally.
What will it take? A complex pattern that can be printed on clothing or fabric intentionally exploits how facial-recognition algorithms work, turning the technology against itself.
As part of a new project called HyperFace, Harvey developed a pattern to fool computer vision from identifying the wearer's face. The shapes in the pattern, of which there could be hundreds or thousands, will register as "false faces," distracting from the wearer's actual face and reducing the computer's ability to accurately match it.
The pattern works by "overloading an algorithm with what it wants, oversaturating an area with faces to divert the gaze of the computer vision algorithm," Harvey told The Guardian.
Harvey created the camouflage pattern as a textile print, but how it could be used is a bit of a mystery. It seems like a cape, hat or hood would be an ideal use, but maybe Harvey has something else in mind. He's scheduled to show off the print January 16 at the Sundance Film Festival for the interaction studio Hyphen Labs and their NeuroSpeculative AfroFeminism virtual reality project.
In the future, the pattern could be applied to more than just textiles. It could be found on architecture or in urban environments, according to Harvey.