Future Vision: 5 ways Vision AI will shape the future of retail
Editor's Note: This blogpost was first published in July 2020, and has been revamped and updated for accuracy and comprehensiveness.
We’ve said it: brick and mortar retail isn’t dead. No amount of online shopping can trump the experience of:
- The ability to interact with an item before buying. Think - checking the freshness of fruits & vegetables, trying out personal items such as clothes and cosmetics, and seeing the quality of an expensive piece of furniture before making a purchase (IKEA!).
- Customer service and asking questions directly to a sales assistant.
- Avoiding shipping wait time and costs, avoiding the complexity of returning unwanted items.
- Window shopping as retail therapy, the collective experience of spending quality time with loved ones by “grabbing a coffee then going shopping”.
To put simply, there is one thing physical stores have that online shopping can’t provide - human interaction. So how can retailers survive in the age of digital?
That is, by creating a personalised shopping experience and providing excellent customer service.
Vision AI technology for retail
Companies can transform their customer experience by using vision AI. While still in its infancy, savvy retailers are beginning to experiment with applications of this next-gen technology, which will help them evolve with customer’s changing shopping habits.
To put it simply, vision AI works by acquiring an image, processing the image, and then understanding or classifying it to build an artificial data system. Artificial Intelligence (AI) can then take actions based on the learned understanding of this data.
In retail, computer vision technology can be applied in the following ways:
Who hasn't heard of Amazon Go? This revolutionary technology allows for automated payments in its grab and go stores. No stopping at POS and scanning products. Customers can simply select their items and walk out of the store.
The Amazon Go uses a mix of technologies, with cameras installed to track customer movements, recognise when a product is selected and charge the customer’s account when they exit the store.
"The majority of sensing is from above. Cameras figure out which interactions you have with the shelves. Computer vision figures out which items are taken. Machine-learning algorithms also determine which item it is," says Dilip Kumar, vice president of technology for Amazon Go.
2. Customer data for business intelligence
All companies collect customer data which forms the foundation of any business decisions. Computer vision makes the process of collecting data simpler. Using facial recognition, computer vision can identify different customer demographics and create a persona based on their purchase patterns and in-store activity.
Just like analysing the performance of a website, computer vision allows retailers to create heat-maps for stores - seeing which displays draw people’s eyes and which ones don’t. This allows retailers to position products in the most convenient way.
3. In-store advertising
Similarly, for stores that use digital signage to display advertising, facial recognition technology is a game-changer. Now, content can be targeted based on who is viewing it. Vision analytics face detection allows retailers to recognise customers' faces and anonymously tag them to a demographic persona type.
Guided with these insights, marketers and store planners can strategize their content to make sure it achieves the highest ROI possible.
4. Compliance & theft prevention
Employee theft and shoplifting have cost retailers almost $100 billion globally. Traditional security cameras have blind spots. And when a store is particularly busy, the security staff is strained, leaving room for human errors.
Computer vision makes this process easy by teaching systems to monitor and recognise suspicious activity. Using video footage of previous thefts or other suspicious activities, cameras can be trained to detect these movements and immediately alert staff. This process can similarly be replicated to detect compliance errors - reducing the countless hours retailers spend to ensure that their stores comply with safety and merchandising standards.
5. Extra layer of protection & cleanliness
Not only can vision AI monitor human activity, it can also keep track of hand-to-surface interactions.
Surface areas in public places are touched by multitudes of people. Powerful vision AI cameras can continually monitor these interactions, presenting this information in the form of heatmaps. Once a threshold of hand-to-surface interactions is reached, a cleaning crew will be notified to perform cleaning and disinfection routine.
meldCX SAMi* (Surface Awareness Management Intelligence) offers this capability. This technology not only provides extra protection for customers, but also allows staff to focus their cleaning efforts and track compliance.
The challenge of applying computer vision to the physical world
Installing a computer vision-based system within a retail store can be a challenging endeavour. A retailer would need to set up cameras at the right points in the store. The computer vision algorithm behind the cameras would need to be trained on what to recognise, and what subsequent actions to be taken.
Imagine if you could do all this through a single-click install? With meldCX Viana, now you can. Find out more & book a demo here.
Not the solution you’re after? We are always open for POC ideas! To get in touch, email us at email@example.com
This article is an installment of our Future Vision series, where we tackle insights on Vision AI disruption across industries.
*meldCX has a pending provisional application, Australian Provisional Patent Application 2020901795. meldCX SAMi™ is our proprietary technology that utilizes touch-enabled devices and computer vision solutions to ensure cleaning compliance of surfaces.