As consumers’ retail experience expectations continue to rise, the importance of product availability and placement also grows. Out of stock, overstock, product placement non-compliance, and discrepancies between invoiced inventory and actual inventory (shrinkage) all contribute heavily to the $1.1 trillion in lost revenue for retailers globally — and the damage to customer loyalty and brand equity can be even more costly.1 Mitigating these losses requires a fundamental shift in inventory management: getting rid of slow, manual checks in favor of fast, automated recognition and response.
Tasking personnel with manually stocking and replenishing inventory — of fast-moving goods, in particular — is a time-consuming and error-prone process. The speed and level of detail required to efficiently identify low stock or missing items among hundreds of products is beyond the capabilities of most humans.
Fortunately for retailers, this is where computer vision excels. Powered by Cisco Hyperflex™ Edge and the Intel® Distribution of OpenVINO™ toolkit, Vispera ShelfSight* is an integrated, end-to-end smart shelving solution that delivers a digitized view of store inventory with over 99% recognition accuracy.2
“It’s a near real-time shelf monitoring and stock management system that uses fixed cameras and on-premises servers inside a physical store to constantly and autonomously track the contents of the shelves,” says Dr. Erdem Yörük, Chief Scientist at Vispera Information Technologies.
ShelfSight uses image acquisition units like shelf-facing smart cameras and surveillance systems to capture visual data inside the physical store. Then, rather than sending these images to a remote data center for processing, Cisco Hyperflex™ Edge enables on-premise, near real-time analysis that informs retail-centric KPIs, such as:
- Out-of-stock items
- Misshelved items, also known as planogram non-compliance
- Empty-space detection
- Display equipment recognition and condition checks
- Price tag detection and reading
- Customer tracking
Equipped with these analytics and insights, retailers are able to make more informed inventory and personnel decisions. They can help mitigate the risk of inaccurate shelf measurement and reduce staffing inefficiencies by delegating this otherwise laborious task from human employees to machines. Computer vision solutions and deep-learning algorithms deployed via OpenVINO™ pave the way for increased automation of tasks, while Cisco Hyperflex™ Edge provides a flexible edge computing solution that enables workload consolidation and evolves with a store’s inventory. For customers, the result is a better buying experience: obtaining the products they want, when and where they want them, with less friction in the process.
The data economy is booming, and this unique convergence of technologies from three IoT innovators – Intel, Vispera, and Cisco – is already revolutionizing the inventory management process. We’re excited to continue helping retailers gain data-driven insights that translate to real business success.
Intel© Corporation. Intel, the Intel logo, OpenVINO, and Intel Xeon are all trademarks of Intel Corporation or its subsidiaries in the U.S. and/or other countries.
*Other names and brands may be claimed as the property of others.