CV Dazzle Is A Makeup System That Kills Face Detection

The world we are living in is slowly turning into one huge surveillance camera with Apple, Google, Amazon and the NSA collecting every bit of info they can about our lives.

Aside from the information we are giving away for “free” they can collect info from street cams, ATM cams and any other cam that they can have access to. Of course, a key factor in that footage collection is the possibility to detect and tag faces and match them with other information about a person.

This is where Adam Harvey steps into the scene. As a long time privacy protector (and the new owner of Privacy Gift Shop, Harvey designed a set of makeup patterns designed to kill any face detection thrown at whomever is wearing it. Is it worth the weird looks you’d get on the street? Depends on how much you have to hide.

CV Dazzle Is A Makeup System That Kills Face Detection

Basically, face detection algorithms work by trying to pattern-match parts of the face, but “dazzle patterns” inspired by WWII camouflage techniques make the lives of those algorithms extremely hard:

CV Dazzle^™ is form of expressive interference that takes the form of makeup and hair styling (or other modifications). The name is derived from CV, a common abbreviation for computer vision, and Dazzle a type of camouflage used during WWI. Dazzle camouflage was originally used to break apart the gestalt image of warships, making it hard to discern their directionality, size, and orientation. Likewise, the goal of CV Dazzle is to break apart the gestalt of a face, or object, and make it undetectable to computer vision algorithms, in particular face detection.

And because face detection is the first step in any automated facial recognition system, CV Dazzle can be used in any environment where such systems are in use, including social network platforms like Facebook.

In the two examples below you can see how a face recognition software goes over two faces: one unmasked face, which the software easily recognizes, followed by a CV Dazzled face which the software fails to recognize (SW is OpenCV is this case).