Facial Recognition

Imagine the scenario. You walk into a clothes shop. You browse the rails and select a few items to try on. You decide on a couple of items and simply walk out. You’ve not interacted with anyone or anything else, and yet your current account / credit card is charged for the items the moment you leave the store. 

Sound far fetched? It’s not. It’s not even the future. Most of the technology needed for this to happen is available now.  

Facial recognition is here to stay. Imagine being the store owner and being notified when your most valuable customers walk into the store. Imagine being able to match a customer’s browsing history to their in-store experience. Imagine being able to re-deploy cashiers to offer a more personalised and enhanced shopping experience. Imagine faster payments and no queues. Imagine instant age verification. 

Walmart has already patented the technology to read customer’s faces at the checkout to see whether they are smiling or not in order to improve customer experience; Amazon Go are trialling cashierless stores, using advanced algorithms and heavy surveillance to enable customers to ‘pay by face’; and several retailers are using facial recognition to help with loyalty programmes and prevent shoplifting.

Paul Wilks who runs a Budgens supermarket in Aylesbury, got into the news last year when, exasperated by shoplifting by repeat offenders, he installed a facial recognition system. As you walk into his store your face is compared to a database of known shoplifters and if your face matches, the store manager gets a discreet alert within 3 seconds. 

Shoplifting in the UK is estimated at around £700m each year with only about 10% of cases being reported. One of the problems is that shoplifting below £200 is now considered a less significant ‘summary offence.’ Combine this with and an overstretched police force directing resources to more serious crimes, and it’s no wonder shop owners are taking their own measures to reduce theft.

Facial recognition algorithms work by giving them a large number of images and telling them which ones are the same person. The algorithms analyse the images to pick out the features that define an individual. This is done by generating a 3D image based on gradients of light and dark rather than the length of a nose or the distance between the eyes. Using high quality cameras in good light and viewing the subject face on, the algorithms can have a success rate of up to 99%.

It won’t be long before facial recognition software, advanced in-store surveillance and algorithms are merged to create a powerful combination. As a consumer, there is a possibility that soon you will be able to walk into a store and pay with your face. Mobile phones with Android Pay already allow users to confirm financial transactions using facial recognition software. It’s a small step from there to purchasing items on the high street with a smile. 

Paying by face seems a bit scary, but facial recognition software is so sophisticated these days that paying by face is more secure than paying by credit card and, in some parts of China, it is already a reality. So if the technology is available, could we be ‘paying by face’ on the UK high street soon? This very much depends on whether customers see the benefits outweighing their concerns regarding the storage and use of their personal data. 

Retail needs to do something different. It’s due a seismic change in how it operates and how it can battle e-commerce. Facial recognition might be what it needs to make a difference.