Sightcorp logo

Facial Recognition in Retail

Everything about Facial Recognition in Retail

     

Facial Recognition in Retail

At first glance, the retail industry might not be a place where you would expect to find a huge demand for facial recognition technology. But if you take a closer look, you will see that there is plenty of potential for this technology to change the world of retail.

How facial recognition can transform the retail industry

 

Facial Recognition in Retail

Let’s have a look at the various ways in which face recognition can be used within retail:

  • Point of sale / self-service checkout: To help prevent identity fraud, and to ensure compliance with the strong customer authentication requirements set out in PSD2 in the EU, facial recognition can be used as part of payment verification during the checkout process. 
  • Loyalty programs: Retailers usually have comprehensive data on their loyalty program members, along with permission to use this data for personalized marketing. This makes it easy to use facial recognition to make specialized offers to these customers in-store. One example would be where a kiosk in a fast food outlet presents a customer with a customized display of their favorite dishes when they go up to order.
  • Frictionless shopping: Amazon [link to Facial Recognition Software Companies article], for example, has already started using facial recognition to introduce frictionless shopping in their Amazon Go stores by allowing shoppers to pay for their purchases by scanning their faces. Since the goods are tagged with RFID chips, and people’s faces are attached to their payment methods in a database, it’s not necessary for customers to manually scan the goods or to present their credit cards to make the payment.
  • Security: Facial recognition technology helps retailers flag known shoplifters as soon as they enter the store. If the system identifies someone as a known shoplifter, it sends out an alert, and – for example – an employee can be instructed to address the identified person by name and to offer them customer service. This is often a good deterrent, since the person then knows that they have been noticed.
  • Personalized shopping experience: Facial recognition makes it easy to identify VIP customers and to make them special offers in store.
  • Enhanced customer service: When integrated with other systems, facial recognition can provide access to a wide range of data on individual customers. The type of data includes how frequently a customer visits the store, when they last made a purchase, what their frequently purchased items are, and so on. With this information, store assistants can provide more efficient customer service and more personalized assistance to each customer.
  • Employee tracking: Facial recognition technology can be used to replace traditional time and attendance systems, decreasing time fraud and increasing efficiency.

 

Important factors to consider

When implementing facial recognition technology in a retail environment, you need to keep in mind how it will affect your customers and how the technology will be perceived by them. Here are some of the factors that you need to pay close attention to: 

Informing customers: If you’re using facial recognition technology in your store, you should have signs up in front of the entrance informing customers of this fact before they enter the store. This way, customers can make an informed choice.

Opt-in vs opt-out: Will customers be enrolled in your facial image database unless they opt out? Or will they only be enrolled if they opt in? Unless you are using facial recognition for security reasons, the recommendation is that you use an opt-in system. And in the case of using facial recognition for security, it is recommended to purge the non-matched faces on a regular and frequent basis.

Data security: You need to make sure that the data you’re collecting is secure, and that you’re able to remove data if requested to do so by a customer. You can also take measures such as:

  • Not storing facial images if you’re using them for payment verification purposes
  • If you do have to store facial images – storing them locally, and not in the cloud, so that you have complete control over the data
  • Regularly purging data that you no longer need

 

Sightcorp’s Facial Recognition Technology and Retail

Sightcorp’s main focus for facial recognition is on Know Your Customer (KYC), Access Control, and Payment Verification. The latter, in particular, is highly relevant in retail. If you’re looking for a face recognition solution in this context, please reach out to us so that we can discuss the options that are available for you.

FaceMatch

Our FaceMatch face recognition technology matches faces in images. Some other key features include that it is deep learning-based, that it is customizable in terms of confidence thresholds and the ability to retrain the models on customer data, that it provides best-in-class accuracy, and that it allows for data to be processed locally.

Currently, there is an SDK available, which allows you to develop your own face recognition solution, or to integrate face recognition into an existing solution. To develop the SDK into your own application, you will need C++ expertise.

Our web-based API service, which requires minimal development and use of hardware, is in the development stage and will be available soon.

Find out more about Sightcorp’s FaceMatch solution

Here are other articles you might find interesting:

 

   

Technical Specifications

The table below shows how FaceMatch SDK performs on the Labelled Faces in the Wild (LFW) dataset:

FPRTPRThreshold (Inverse of distance)
0.10.99900 ±0.002130.55448
0.010.99667 ±0.005370.59791
0.0010.99367 ±0.006050.62989

FPR = False Positive Rate
TPR = True Positive Rate

These results are an indication only and are based on the specific dataset Labelled Faces in the Wild. Customers can expect similar performance, with possible variations due to hardware and the availability of annotated data.