Vispera - Image Recognition Solutions for Retail
INTEL UK

Vispera ShelfSight partnership with Intel is Helping Grocery Retailers to Cut Sales Losses with AI-Powered Computer Vision

An article featuring the new white paper “Intel and Vispera ShelfSight*: Effective On-Shelf Inventory Management for Grocers” is featured at intel.co.uk.

The article written by Libby Plummer states: 

“Nearly $1.1 trillion is lost by retailers worldwide each year as a result of out of stock products, overstocking, and shrinkage.1 If preferred products are out of stock, customers will take their business elsewhere. And with customer expectations at an all-time high, situations where products are out of stock could also damage customer loyalty. This is especially challenging for low margin grocery retailers losing out on high value items. Out of stock products can also incur other costs, such as the reduced impact of instore promotions and the time employees’ use up dealing with customer requests about missing items.

Forward thinking retail businesses are now turning to digital strategies to help understand their customers, streamline their operations, boost profits and remain competitive. Computer vision, which focuses on how computers acquire and understand visual cues, has surfaced as a key technology for retailers. When combined with Artificial Intelligence (AI) algorithms, it enables retailers to automate processes that previously would have been carried out manually.

“Computer vision has emerged as a great digital strategy, providing retailers with a vast amount of store data,” explains a new white paper from Intel and retail technology firm Vispera*. “Computer vision coupled with the advances in deep learning algorithms provides an excellent opportunity to automate many activities in the store that were previously done manually. Out of stock and other non-compliance issues can be automatically alerted by building intelligence at the edge.”

Making use of these technology trends, Vispera ShelfSight is a real time shelf monitoring and management system designed to analyse retail shelves for out of stock products. The platform uses IoT cameras to give businesses a digitised view of the store in real time. ShelfSight can recognise individual products, detect empty space, and also ensure compliance with planograms – detailed diagrams that show where specific products should be placed on retail shelves.

Powered by Intel technology, ShelfSight uses deep learning algorithms developed by Vispera to recognise individual products and also detect empty space, excess stock, and missing or misplaced facing. The hybrid system is located both on premise and in the cloud, sending notifications to store staff on mobile devices so that they can replenish stock quickly when needed and minimise any loss of sales. “Visual data captured has a direct impact on the retailer’s bottom line,” explains the white paper.

Along with stock information sent to store personnel, ShelfSight also sends all shelf data to the cloud for calculating various KPIs (key performance indicators) and other data analytics. As the system can also support third-party applications, it can be used as a versatile IoT platform for collecting other measurements such as shopper counts and shopper-product interaction using technology such as RFID tags and digital signage.

In testing economic times, retailers cannot afford to lose out due to stock being unavailable or misplaced. And Intel-powered technologies like Vispera’s ShelfSight can help them to ensure that this doesn’t happen.

*Other names and brands may be claimed as the property of others"

The White Paper is written by Dr. Ceyhun Burak Akgül, Vispera General Manager & CTO, Dr. Edem Yörük, Vispera Chief Scientist and Dr. Farhaan Mohideen, UK Business Unit Lead for Retail, Intel Corporation. 

Please download the white paper here.

Visit our ShelfSight page for more information.