A couple of weeks back, we ran some disturbing conclusions by Carnegie-Mellon researchers that privacy is reaching the point of no return with the advent of smartphones employing facial recognition apps. Add another dynamic to the mix: location-based marketing pitches that size up a person’s sex, age and dimensions.
The Los Angeles Times reports that some companies are adopting or already have facial recognition technology in place to provide the ultimate customization for digital ads:
“Once the stuff of science fiction and high-tech crime fighting, facial recognition technology has become one of the newest tools in marketing, even though privacy concerns abound. The Venetian resort, hotel and casino in Las Vegas has started using it on digital displays to tailor suggestions for restaurants, clubs and entertainment to passersby. Kraft Foods Inc. and Adidas say they are planning to experiment with it as early as this year to push their products.”
Some companies are even getting creative with employment of facial recognition technology. A group of bar owners in Chicago now employ mounted cameras to “keep tabs on the male/female ratio and age mixes of their crowds. Patrons planning a night out can use mobile apps to get a real-time check of a venue’s vibe.”
We know personalization is the holy grail of advertising and marketing. And facial recognition has been successfully deployed in good things, such as crime-fighting and terrorist-spotting. Security and access to facilities or systems may be another positive application of technology. Recognition technologies can enhance the overall experience of interacting with computers.
But does employing the technology to recognize and sell to people in public spaces take it a step too far?
For a good overview of the ethical, social and security implications of facial recognition technology, New York University’s Center for Catastrophe for Preparedness and Response issued a report back in 2009 that concludes that it may be some time before the technology can randomly pick faces out of crowds. A database of image galleries is necessary, and the report recommends that subjects have the ability to “opt in.” As the paper concludes:
“It might be that this is ultimately an unattainable goal, especially for larger populations. Not because the technology is not good enough but because there is not enough information (or variation) in faces to discriminate over large populations—i.e. with large populations it will create many biometric doubles that then need to be sorted out using another biometric.”
The below YouTube video, produced by TweakTown, offers a look at some facial recognition or gesture-controlled systems, that have some interesting applications, such as creating avatars or controlling screen positions. The final segment demonstrates how a TV can sense who is in the room watching, and deliver appropriately targeted ads. Hmm.