Facebook Profiles and Face Recognition Help Identify People on The Street
The power of the Internet to disseminate information is awe-inspiring and frightening at the same time. And nowhere is this inertia felt the strongest than in the social media networks.
Facebook has had its share of backlash over face recognition — a feature that connects a photo of a person’s face with their Facebook profile, making it easier to tag people in photos — but researchers from Heinz College at Carnegie Mellon University recently proved that Facebook’s vast photo archive can be used to identify people on the street, too.
The authors of the study titled “Faces of Facebook: Privacy in the Age of Augmented Reality” — Alessandro Acquisti, Ralph Gross and Fred Stutzman — demonstrated it at the Black Hat technical security conference, which was July 30 to August 4 in Las Vegas. They used publicly available data — photos from Facebook profiles of students — and then used face recognition technology to recognize these students as they look into a web camera.
The results? Using a database of 25,000 photos taken from Facebook profiles, the authors’ face recognition software correctly identified 31 percent of the students after fewer than three (on average) quick comparisons. In another test, the authors took photos from 277,978 Facebook profiles and compared them to nearly 6,000 profiles from an unnamed dating website, managing to identify approximately 10 percent of the site’s members.
The study raises important questions about our privacy. Online, you can make steps to hide your real identity such as changing your name, but as the authors of the study note, “It is much harder … to change someone’s face.” Based on the results of the study, it’s not hard to imagine someone creating a simple software/hardware combination which could identify people simply as they walk through a street or peer into a store’s window.
“Our focus, however, was on examining whether the convergence of publicly available Web 2.0 data, cheap cloud computing, data mining, and off-the-shelf face recognition is bringing us closer to a world where anyone may run face recognition on anyone else, online and offline — and then infer additional, sensitive data about the target subject, starting merely from one anonymous piece of information about her: the face,” the study concludes.