Blog Archives

Facial Recognition – Part Duex


Bloodhound Betty posted a few days ago about Facial Recognition Mishaps and the pros and cons of false positives.

Today I stumbled upon an article from Fast Company about some work by Carnegie Mellon researchers using commercially available facial recognition software, cloud computing resources and open data feeds to identify people and possibly get access to their social security numbers.

“A person’s face is the veritable link between her offline and online identities,” said Acquisti, associate professor of information technology and public policy at the Heinz College and a Carnegie Mellon CyLab researcher. “When we share tagged photos of ourselves online, it becomes possible for others to link our face to our names in situations where we would normally expect anonymity.”

Image via The Black Box Office (http://theblackboxoffice.com)

As I read the article, I am thinking of Minority Report and the retina recognition and pushing ads to people. Kind of like Google Ad Words following you and pushing advertisements to you on digital billboards. I don’t think it is too far off, I was using Google Plus the other day looking through a photo album posted by a friend, and the software was able to identify facial regions of a person and was asking me to tag the picture with the person’s name.

I can only imagine what information might be able to be tied together if a face becomes the primary key to your personal data.

“Ultimately, all this access is going to force us to reconsider our notions of privacy,” Acquisti said.  “It may also affect how we interact with each other. Through natural evolution, human beings have evolved mechanisms to assign and manage trust in face-to-face interactions. Will we rely on our instincts or on our devices, when mobile phones can predict personal and sensitive information about a person?”

Now that social media sites are allowing geotagged photos, this information can now tie you to a place as well as identify who you associate with or what your preferences are.

The full results of the study will be released today at Black Hat today (8/4/11).

Facial Recognition Mishaps


Here’s Looking at You, and You, and You … – IEEE Spectrum.

This article points out some of the false positives that have come from the wide spread use of facial recognition software used by law enforcement agencies and for those who are falsely recognized it can cause all kinds of problems that will take a lot of time and effort to straighten out.  While facial recognition is really amazing it’s not exact and anyone who has used the Faces function of iPhoto knows it doesn’t come without it’s quirks.  So using it against DMV data, as was done here, resulted in several false positives.  Although they do other research into the findings to determine the validity of the ‘hits’ there are still some problems that need to be addressed but in the meantime have all your identity related documents in order in case you have to prove who you are.

Of course any new technology is going to have it’s down side, but there are a couple of areas where the value of facial recognition, even with false positives, outweigh the problems.  When applied to missing persons and child pornography facial recognition can offer some help especially when the tips stops coming in and any piece of information could offer some help.   I don’t think it’ll be long until someone is located based on facial recognition tools used against social media and file sharing sites.  Sure, there will be false positives but in cases like these there are just as many leads that go nowhere from eyewitness testimony, which has proven to be largely inaccurate.  Hmmm, that’s a good question, what would be more accurate, facial recognition software or an eyewitness?   My money is on the software every time.