Some time ago a start-up company, scraped billions of online photos to build a tool for the police that could lead them from a face to a Facebook account, revealing a person’s identity. Researchers have now developed algorithms and tools to foil those systems.
University of Chicago’s SAND Lab has now developed Fawkes, an algorithm and accompanying software tool that gives individuals the ability to limit how their own images can be used to track them. The Software comes for Windows and Mac platforms and can be installed locally on your computer. Fawkes takes your personal images and makes tiny, pixel-level changes that are invisible to the human eye, using a process called image cloaking. These cloaked photos can be used normally, sharing them on social media, or for printing and other uses you would use any photos.
The clocked photos however prevent peoples who tries to use these photos to build a facial recognition model as they will teach the model an highly distorted version of what makes you look like you. The cloak effect is not easily detectable by humans or machines and will not cause errors in model training. However, when someone tries to identify you by presenting an unaltered, “uncloaked” image of you (e.g. a photo taken in public) to the model, the model will fail to recognize you.
The research paper is available here. Another relevant article can be found at Gizmodo.
