FB pixel

Biometrics for payments, not tracking customers in Amazon Just Walk Out stores

Palm payment acceptance ticks up
Biometrics for payments, not tracking customers in Amazon Just Walk Out stores

Amazon says the Just Walk Out store in the Seahawks’ Lumen Field NFL stadium launched last year is a rousing success. The AI system allows fans to enter a concession stand, grab what they want, and walk directly back to their seats without going through a checkout process. A Just Walk Out store in the stadium saw double sales when compared to the previous store in the same location.

Those entering the store can use a credit card at the entry gate or elect to scan their palm biometrics with an Amazon One device as their payment method. When a user picks up an item in the store, the system adds it to their virtual cart. If the user returns it to the shelf, it will automatically remove it. Customers are charged for what they take as they leave the store.

While the population at large is generally not comfortable with conducting biometric payments through hand recognition, the percent of U.S. adults who are at least “somewhat comfortable” with the method has increased from 22 percent in 2019 to 27 percent in 2023, says Civic Science.

Thirty-three percent of those who currently use mobile payment methods like Apple Pay or Venmo are at least somewhat comfortable with paying with their palms. Those who intend to use mobile payment apps are actually more comfortable with using a palm payment method, with 45 percent being at least somewhat comfortable with it.

The Just Walk Out system operates independently from Amazon One and doesn’t collect any biometric information to track the identity of shoppers, the company says. It instead links a customer with their payment information and assigns a temporary numeric code which is deleted once they leave the store.

“All we need to know is where that person is on the floor, and where their hands are in relation to the store’s merchandise,” Gérard Medioni, vice president and scientist at Amazon says.

The computer vision team used datasets from AI-generated synthetic images and video clips mimicking realistic shopping scenarios in various store formats, lighting, and crowds of shoppers. As a result, the system can recognize millions of actions “without making mistakes,” reads the announcement. The company also used synthetic data to train its Amazon One palm biometrics.

The system is available at over 70 Amazon-owned stores and over 85 third-party locations across the U.S., UK and Australia.

Amazon One is also set to be launched in all Whole Foods stores by the end of this year.

Related Posts

Article Topics

 |   |   |   |   | 

Latest Biometrics News


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Most Read This Week

Featured Company

Biometrics Insight, Opinion

Biometrics White Papers

Biometrics Events

Explaining Biometrics