Facial recognition – a powerful ad tool or privacy nightmare?


Powered by Guardian.co.ukThis article titled “Facial recognition – a powerful ad tool or privacy nightmare?” was written by Sean Hargrave, for theguardian.com on Wednesday 17th August 2016 11.30 UTC

Whenever the future of advertising is discussed, thoughts often turn to the sci-fi image of a stalked Tom Cruise in Minority Report being recognised and served intrusive, personalised advertising as he tries to escape a futuristic city.

Hence, recent claims by Russian app FindFace that it could identify people in public by their profile picture on a Russian social media site were met with a combination of awe and concern. Such facial recognition could enable digital screens to tailor messages to each individual as they pass, meaning an end to brands wasting budget advertising the wrong products to the wrong people.

Facial recognition is already following celebrities’ pictures around online and providing advertising space next to them. Computer vision firm GumGum claims to be able to programme any face into its recognition system and then identifies it when it subsequently appears on the internet, even if there is no caption giving away the person’s identity. Typical customers are sponsors seeking to check up on what exposure a celebrity is earning them.

“We’re currently working with P&G to recognise the athletes they’re using in their Thank You Mum Olympics campaign,” says GumGum managing director Ed Preedy. “We not only count how many pictures are posted of the athletes they work with, so they can see how much coverage they’ve attracted, but we can also let them serve an advert next to each picture.”

The VIP treatment

Retailers are starting to see face recognition as a means of distinguishing their VIP customer loyalty schemes. At Axis, which develops facial recognition technology, business development manager Andy Martin points out facial recognition is already being used to enable advertisers to count how many people have viewed their outdoor advertising. The results are even split by gender and age. In a well-known outdoor campaign for domestic abuse charity Women’s Aid, a model’s bruises disappeared from her face as more people watched the screen, getting across the message that if people take notice, domestic violence could be halted.

“Where I think we’ll start to move on and actually see people individually recognised is customer loyalty schemes,” says Martin.

“We’re already working behind the scenes with some well-known retailers who are interested in spotting their loyal shoppers when they walk through the door. It would allow a sales assistant to be alerted and offer some a truly personalised shopping experience.”

While customer loyalty might be the first opportunity, individual adverts, targeted through facial recognition, could be arriving sooner than you think. Robin Arnold, director at Digital Media Technologies, which consults on outdoor campaigns, has used camera technology in projects to enable digital posters to change when they are being watched. A recent example was a J20 advert in which an on-screen predator played a game of What’s the time Mr Wolf? with onlookers. He has also worked on campaigns where the technology was used to subtly personalise car adverts when an advertising screen’s camera spotted a driver of a rival brand.

“The technology used number plate recognition software to find out what type of car [was] waiting at the lights and then changed the message on the digital screen,” he explains.

“So, when the client was Renault they would spot a Citroën car and say something like, ‘Hey, Citroën driver, check out this Renault’. BMW used it to promote its X5 model when the system spotted people driving the equivalent Mercedes model.”

Mood and gender targeting

Will the next step in targeted advertising see brands recognise each person’s face and serve them a personalised message? Chris Pelekanou, commercial director at outdoor advertising company Clear Channel, reveals that experiments have been taking place to determine if adverts can be altered according to an onlooker’s mood. One recent public campaign from Plan International used a camera to deliver men and women different messages to underline the limited education choices for women in some countries. However, recognising each person would probably be a step too far.

A passerby looks at the M&C Saatchi Artificial Intelligence poster in Oxford Street London.
A passerby looks at the M&C Saatchi Artificial Intelligence poster in Oxford Street London. Photograph: Laurence Gibson

“You can’t do something just because the technology is there, you have to be led by the consumer and not the technology,” he says. “If we did start recognising people en masse then I think outdoor would have the same problem that digital display is having with people getting fed up and install ad blockers.”

Social offers

However, Andy Pringle, head of performance media at digital marketing agency Performics suggests that we are entering a period where we should think of marketing as “add-vertising”. Give a customer something extra and they may well give permission to be recognised.

“It’s the kind of thing you can imagine being used in customer loyalty but also social media,” he says. “You can imagine brands asking people to give permission to be recognised in return for offers while they’re out and about. Say, there’s a guy waiting for a bus for ages in front of digital screen running a beer campaign. If that person likes that brand on Facebook you can foresee either the screen saying “hi” and giving him or her a voucher code for a free beer or triggering a voucher to be delivered to their Facebook inbox.”

Too many faces, too many laws

That kind of facial recognition is going to be far more likely than the sci-fi image of all advertising being individually tailored, according to Scott Ross, CTO at advertising agency DigitasLBi. He suggests those who imagine facial recognition happening en masse should consider exactly how much computing power would be needed.

“Just on an escalator at a tube station you’d need to recognise a hundred people at the same time,” he says.

“And each of those faces would have to be retrieved from a database of millions in the split second it takes them to pass from one screen to another. It’s a possibility but highly unlikely.”

As one would imagine, it’s not only a technological issue but a legal one. Nick Johnson, partner at Osborne Clarke and member of its advertising law team, points out that the law is moving towards tougher opt-in rules as the EU general data protection regulation moves from approval to becoming law in May 2018.

“The new data privacy laws are very specific about any biometrics belonging to a new, distinct type of data deemed as extra sensitive, alongside health and religion,” he says.

“It means brands will need very clear, explicit opt-in permission in which they make it very clear exactly what will be done with the data and offer very clear choices to consumers to opt out. So, it’s a very sensitive area.”

En masse, then, facial recognition looks to be a way off and will possibly never be accepted because of privacy concerns. However, as a special feature for customer loyalty schemes, recognition systems are already being tested behind closed doors. If you’re a high flyer, expect to be asked to be recognised soon: but if you are just one of the crowd, expect your face to be counted, but not recognised, next time you stop to stare at an ad on a digital screen.

To get weekly news analysis, job alerts and event notifications direct to your inbox, sign up free for Media & Tech Network membership.

All Guardian Media & Tech Network content is editorially independent except for pieces labelled “Paid for by” – find out more here.

guardian.co.uk © Guardian News & Media Limited 2010

Published via the Guardian News Feed plugin for WordPress.

Neville Hobson

Social Strategist, Communicator, Writer, and Podcaster with a curiosity for tech and how people use it. Believer in an Internet for everyone. Early adopter (and leaver) and experimenter with social media. Occasional test pilot of shiny new objects. Avid tea drinker.

Comments are closed.