Recently I had the opportunity to spend quality time with a leading computer vision company headquartered in Ireland. As an advisor to this company, I have been lucky enough to see their transformation from a single application to a portfolio of retail visual innovative disruptive solutions that will enter the marketplace in the balance of this year.
Concurrently, I remembered where Gartner placed computer vision in the Hype Cycle of Artificial Intelligence, 2020.
Gartner has computer vision in the “Trough of Disillusionment” reaching the “Plateau of Productivity” in 2 to 5 years. Based on my retail industry observations over the last several months, powerful computer vision solutions are arriving much sooner to a store near you.
In 2021, the major milestone of over 1 billion CCTV cameras installed around the world will be reached. What started as a security technology to prevent crime, the video camera is now a major visual data gathering device that will transform all industries including retail.
All Those Cameras Have Data Eyes
Computer vision is a form of artificial intelligence (AI) that trains computers to interpret and understand the visual world. By combining digital images from camera feeds and videos with deep learning, machines can quickly learn to accurately identify and classify disparate objects.
Below are a few examples of the emerging retail applications.
The $33+ Billion Future of Computer Vision
The global computer vision market is expected to grow from $2.9 billion in 2018 to over $33 billion by 2025. According to a just published RIS News report, 10% of retailers have started a major upgrade of their computer vision solutions and nearly 17% more will deploy the technology in the next 12 to 24 months.
With the rise of e-commerce, consumers now expect a highly personalized shopping experience from any brand, not just digital-first ones. Computer vision enables retailers to offer a frictionless checkout experience, near-perfect inventory accuracy for omnichannel sales, and greater improvements in store operations, allowing store associates to focus on customer-oriented tasks.
The Future of Self-Checkout Is Already Here
As retail becomes more digitized, many essential tasks have become automated, providing greater accuracy and efficiency for customers and employees alike. Of these automated tasks, self-checkout is the most successful, as a faster checkout process offers convenience to customers and labor savings to retailers.
Visual artificial intelligence has arrived to secure the self-checkout process. New solutions are also emerging to replace the barcode with visual product identification engines.
We can’t discuss the future of self-checkout without mentioning Amazon Go, which similar to above example allows customers to simply pick up items they want to buy and walk out of the store, skipping over the checkout process completely. Nearly 30 of these fully automated stores are open in the United States. Amazon’s “Just Walk Out” technology uses computer vision-enabled cameras to follow shoppers’ movements throughout the store while shelf sensors detect when items have been removed or returned.
Fully Automated Inventory Management
In addition to self-checkout, today’s consumers also expect accurate information about product availability as they browse a store. Sixty-four percent of retailers are looking to deploy data-driven solutions like computer vision in the next two years to optimize inventory. By automating inventory cycle counts with computer vision, retailers can update their inventory system in real time to create an omnichannel retail experience.
Studies show that shoppers encounter out-of-stocks every 1 in 3 shopping trips, costing $1 trillion in annual sales.
In the near term, cameras will join an increased variety of connected sensors to dramatically improve inventory visibility.
Computer Vision AI-Powered Loss Prevention
Because computer vision essentially gives a computer “eyes,” there is huge potential for its application in retail loss prevention. Computer vision uses machine learning algorithms to observe people’s behavior, identify patterns, then make decisions based on this information. One of the most commonly imagined applications of computer vision in loss prevention is to use this technology to detect suspicious behavior associated with thieves.
Computer vision has already proven that it can reduce employee theft at checkout by addressing major challenges such as sweethearting, where cashiers don’t scan every item or ring them up at lower prices. By identifying every single item in the checkout area and associating it with a transaction, computer vision prevents ill-intentioned employees from attempting to steal merchandise.
Improved Customer Experience with Computer Vision
In addition to the convenience of having orders delivered right to your door, one of the biggest benefits of online shopping is the ability to research product reviews before making a purchase. According to TrustPilot’s 2020 study on the role of reviews in online trust, 89% of consumers read a review before making a purchase decision.
With computer vision, retailers can bring online reviews to the physical store. For example, a customer can scan an item’s barcode to pull up product information and reviews, right in the palm of their hand. These applications can also leverage augmented reality (AR) to overlay information about an item on a customer’s phone. By adding computer vision on the customer side of the retail experience, retailers can provide consumers with the powerful features of online shopping, like reading reviews and filtering products, during the brick-and-mortar shopping experience. Bridging the gap between physical retail stores and e-commerce creates a very effective omnichannel experience, which more retailers have adopted as customer expectations have evolved to embrace a more digitized retail landscape.
Understanding consumer preferences and delivering immersive experiences is critical to the post pandemic future of retail.
Retailers can leverage computer vision to make data-driven, personalized marketing decisions. Combined with machine learning to recognize different behaviors, computer vision can remember different shopping patterns and build targeted marketing messages using this information.
Data Collection for Store Layout Improvement
In addition to using data about customer behavior to make data-driven marketing decisions, retailers can also use computer vision to improve their store layouts. Computer vision-enabled cameras can track customers’ movements throughout a store and identify “hot spots” and inefficiencies to provide retailers with more insight into the performance of their store layout. According to McKinsey, COVID-19 has accelerated shopping behavior changes.
Shoppers are demanding increased convenience, more efficient store layouts, and frictionless checkouts. Consumer journeys need to be re-designed through emerging computer vision solutions to meet the “new normal” across harmonious retail channels.
While many emerging technologies still remain on the horizon, computer vision isn’t just a concept for the distant future anymore. Its current applications in retail show that this technology can transform the customer experience while optimizing retail operations in multiple areas, including inventory management, loss prevention, marketing, store layouts, and much more. Welcome to a more profitable computer vision optimized future of retail.
