Over a million developers have joined DZone.

The Quiet Creep of Facial Recognition

DZone 's Guide to

The Quiet Creep of Facial Recognition

· Big Data Zone ·
Free Resource

If you don’t follow Alistair Croll of Solve for Interesting, you should. In a piece published last week, You’ll Be Tagged, Croll makes the point that common photo tagging technology like Facebook allows for a remarkable decrease in personal privacy.

The face in the background of the photo today may not matter to those who ‘hold’ the photo, but any face has the potential to be matched to an individual with remarkable consistency at any point in time, for as long as the picture exists. In fact, the more times someone is tagged in photos, the better the facial recognition algorithm gets at identifying individuals.

Place of zero privacy

Alistair Croll - You'll Be Tagged

Alistair Croll – You’ll Be Tagged

Today’s facial recognition software lies at an interesting intersection of three concepts: How long things are stored for, pictures in which a person appears, and people who can recognize faces in a photo (and tag them). This creates a physical timeline and map for any individual’s appearance (consider that most photos are geo-tagged on our smartphones) that takes away any shred of what we currently consider personal privacy.

Video ups the ante

When you add that information to the rapidly increasing ability to do the same with video, the center of Croll’s venn diagram begins to grow larger and larger. Perhaps most scary, this issue doesn’t apply to things captured today. It applies to to anything ever captured at a resolution that allows for facial recognition. Old photos and videos can be ‘re-run’ to add to the database.

There’s a point coming where all of this data is reasonably dangerous in the hands of the wrong individuals, be it a government that wants to track its people or a anyone wishing to manipulate others for personal gain.

Marketing and facial recognition

Of course, us marketers will say it is an amazing tool to help serve contextual offers without being overbearing or impersonal. I like that sound of that much better.

In today’s Venture Beat, they cover facial recognition in Beauty, Finance and Medicine. Facial recognition software can be used to create a ‘virtual mannequin’ that takes the form of the individual but shows the products or services being offered, or can even create aging and other techniques that individualize a message for someone who isn’t thinking ahead to retirement.

Within this context, marketers are trying new experimental methods, with some interesting results and lessons. These innovative experiments span multiple industries and often combine elements of mobile, local and/or social, but they often share an element of personalization (i.e., making the marketing message about the individual, their friends, their photos, their location, etc.). One way of achieving a personalized marketing message has been through the use of facial recognition technology.

What facial recognition software will certainly do is redefine how we see ourselves and what we’re willing to tolerate in lost privacy. Interesting times.


Published at DZone with permission of

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}