Why cameras are ideal for frictionless shopping

artificial intelligence
Computer vision can support autonomous stores.

Online shopping and tap-to-pay methods have slowly introduced shoppers to the future of tech-powered retail, prioritizing convenience and contactless efficiency.

Studies show that the global spending by retailers on AI solutions will reach $12 billion by 2023, due to the increased use of AI to offer personalized services and predict consumer behavior.

The pandemic only expedited AI deployments in retail with low-contact shopping, such as curbside pickup and delivery, due to safety concerns, and these new methods have since stayed around.

Now, autonomous retail is becoming mainstream, from grab-and-go style grocery stores to cashierless kiosks at sports stadiums and frictionless convenience stores on college campuses.

Grocery tech trial and error

AI-powered smart carts have garnered a lot of attention recently, as have touchless self-checkout systems, which typically rely on depth cameras to identify items from any angle and instantly ring them up in a single transaction.

While growing in popularity and producing higher accuracy than smart carts, no matter how fast the checkout process itself is, shoppers still don’t want to wait in a line. Autonomous stores have thus far reigned supreme, and incorporate two core technology components: computer vision and shelf sensors. Most deployments utilize both technologies together, but there may be advantages to leaving shelf sensors out of the equation.

Weighing the pros and cons

Most autonomous stores incorporate shelf sensors within their shelves that work in conjunction with their computer vision systems. Store shelves are equipped with a weighing system, which provides information about the product being taken off of a shelf or put back on.

However, the merging of both weight and visual components may overcomplicate a store’s operation, cost the retailer more and limits the shape, size and layout of the store. Camera-only solutions can reduce installation costs by 60%, and therefore also shorten the time needed to get a store up and running.

Computer vision with standard security cameras can do the job alone, providing flexibility, affordability, and scalability for autonomous stores.

Computer vision reigns supreme

Computer vision can be implemented to a diverse range of store types and products, including big grocery chains, mini stores, and even drink lanes. Instead of having to undergo reconstruction, cameras can adapt to the existing format of a store and work with any layout the operators want to create.

Cameras can also easily track difficult-to-weigh items such as draft beer, hot and ready foods, baked goods, fruits and vegetables, and merchandise.

To no great surprise, one technology system is cheaper than two, and when one system can do the job of both, it’s also cost-effective. The more sensors you have, whether they be cameras or weight sensors, the more machines are needed for computation.

When it comes to scale, given that computer vision can work with a store of any size or format, it’s typically the more feasible option for a small or unique store, allowing the technology to spread far and wide.

The not-so-distant future

In the coming years, more and more tech innovations will pop up and shift the way we shop. Among the innovations so far, autonomous retail is picking up steam. Retailers and technology providers alike are figuring out exactly what kind of equipment, tech and set-ups are needed to successfully automate the shopping experience.

The industry is already seeing customers excited about the expansion of autonomous retail, and is expecting widespread adoption.

João Diogo Falcão is CTO at AiFi

More Blog Posts in This Series

X
This ad will auto-close in 10 seconds