Open for Business: The AI-powered Autonomous Store

Retail store technology continues to move forward at a fast pace. Cloudpick has rolled out a digital solution for brick and mortar stores that includes computer vision, AI, deep learning, and edge computing technology. Using the solution, the Singapore retail chain Cheers has opened an advanced, unmanned AI-powered autonomous store.

Image credit: Cloudpick

Convenience stores, fast-food chains, and quick-serve restaurants (QSRs) frequently have self-serve kiosks and self-checkout stands to accelerate purchasing and provide a faster experience for the busy buyer. While these solutions can be more efficient than traditional checkout lines, they still require human staff to assist customers. And without an attendant present, they can be more susceptible to theft.

The next phase of efficiency in self-serve retail technology is a model that eliminates checkout all together: autonomous stores. These stores include end-to-end technology installations that mirror the traditional shopping experience. They implement a combination of AI, computer vision, deep learning, and edge computing technology that allows customers who have registered with an app to select multiple products–then simply walk out.

The Intelligent Smart Store

One solution provider who has rolled out this technology is Cloudpick. An Intel® Solutions partner, Cloudpick’s Autonomous Store solution taps AI and computer vision technology to assist retailers in turning traditional brick-and-mortar stores into highly digitized, intelligent cashier-less smart stores.

“In this new type of store, we ask customers to register an account on a mobile app before entering the store,” explains Dr. Jeff Feng, CEO of Cloudpick. “After downloading the app, they add a payment method, just like they would do with other service apps. And after registering, they enter the store, take items, and leave. It's a really fast, frictionless checkout.”

Here’s a step-by-step of how Autonomous Store technology works:

• Shoppers open a store’s mobile application, which provides a QR code to the shopper to gain access at the store’s entry gates.
• The code is associated with a virtual shopping cart for each shopper inside the store.
• A vision system uses deep learning algorithms to identify the items a shopper selects and adds them to the cart.
• If a customer puts an item back on the shelf, it’s removed from the cart.
• At checkout, edge computing platforms correlate the data and transmit it to a cloud-based billing platform, in real-time.

Cloudpick states that its Autonomous Store solution can track more than 30 shoppers and their actions over an area of roughly 1,000 square feet.

Image credit: Cloudpick

Vision Systems Go to Work in the Store

For inventory and purchasing accuracy, Cloudpick Autonomous Store uses two types of AI in parallel. The first technology is image recognition, where the solution uses algorithms that identify various products in the store. Images of every SKU are labeled in a training database, and the resulting algorithms are deployed into the Cloudpick vision system using the Intel® Distribution of OpenVINO™ toolkit.

Separately, the system generates a virtual representation of everything in the retail space (including shoppers) and applies gesture recognition algorithms to determine whether consumers have picked or replaced certain items. These algorithms are also capable of identifying behaviors such as walking, standing, crouching, or running, which helps distinguish one shopper from another.

“The vision system is smart enough to re-create everything in the store,” explains Feng. “It basically re-creates every single person as a physical entity, so we’re able to tell customer A from customer B. And we’re able to distinguish them during the entire time they’re in the store.”

Most of these workloads are executed on edge computing platforms using 8th or 9th generation Intel® Core™ processors. By using edge intelligence, the Cloudpick solution can dramatically reduce the amount of data sent to the cloud.

On the privacy front, the QR codes are the only information used to track individuals within the virtual rendering of the store environment, which ensures individual customer privacy. “When a customer registers using the mobile app, that’s the only information we get. We’re not able to identify the person by name,” Feng says. The solution only registers that a customer has entered the store, and that they have individual payment information.

Three Cheers for Cashierless Stores

Cheers, a sub-brand of Singapore’s largest convenience store chain FairPrice, recently launched an AI-powered cashierless store with Cloudpick. The grab-and-go Cheers store sells a variety of drinks, prepackaged food, and snacks. Customers can choose to use facial recognition or scan the QR code from the Cheers SG App to enter the store. Customers grab items they want, and walk out. Virtual baskets are updated by the algorithms, and customers’ digital accounts are automatically charged as they exit the store.

In addition to fast payment, the Cloudpick Autonomous Store solution can enrich the customer’s shopping experience, using smart bar screens on the product shelves. The store owner can set up smart bar screens that can display product information or coupons. The bar screens can also be configured to play advertisements from competitive brands, providing a way for store owners to earn additional ad revenue. The Cloudpick vision and AI systems can identify the general demographic of a person standing in front of the screen and play an advertisement that it deems most relevant to that customer.

• Learn more about the Cloudpick Intel® IoT Market-Ready Solution.
• See the digital store operations technology from Cloudpick.