Facial Recognition App Can Match Strangers to Their Facebook Profile

By October 01, 2011
Woman's faced digitized

This new tech has matched pedestrians to online information about them in studies. Powered by the cloud, the app has even its own researchers worried about its uses.

Cloud-power is coming to all sorts of applications, including the Kindle Fire's Silk web browser. A less well known app is being developed by Carnegie Mellon University's Heinz College and Google that can match a photograph of a person to actual identity. Using facial recognition software PittPatt, as well as the amount of personal data available on the cloud, a research team created a series of studies designed to test the integration between the tech and the resources. While facial recognition (FR) has been around since the late 1960's, according to the Atlantic today this system is faster and more efficient. FR has usually been limited to discrete databases, such as criminal databases when used by law enforcement. But drawing from the entirety of the Internet, including social networks, dating sites, gives results - the new mobile app can take less than a minute. PittPatt and services like it find public images and make matches from them - dating site images are matched to Facebook profiles, even if they use a pseudonym, and also match "pedestrians on a North American college campus with their online identities."

Researchers Alessandro Acquisti, Ralph Gross and Fred Stutzman are looking at a future of augmented reality where someone on the street can be immediately associated with a LinkedIn profile. In their third experiment, the researchers predicted intersts and Social Security numbers od some participants by combining the FR tech with algorithms they had developed previously that predict SSNs from publicly available data. This was meant to show the scope of what is possible to predict about a person, "...to show that it is possible to start from an anonymous face in the street, and end up with very sensitive information about that person, in a process of data "accretion." In the context of our experiment, it is this blending of online and offline data - made possible by the convergence of face recognition, social networks, data mining, and cloud computing - that we refer to as augmented reality."

PittPatt began post 9/11 and has been funded by DARPA, then was purchased by Google in July. Acquisti said in Marketwatch that " the prospect of selling his new app or making it available to the public 'horrifies him.'"

Legal mentions © L’Atelier BNP Paribas