Image credit: WonderStore
The retail experience is wildly fluid these days. Stores are opening, then shutting, adding contactless purchasing, refusing returns, and offering deep discounts only to completely reverse policies, change shipping options, or decline sales incentives the next day.
None of this is fazing consumers, who remain committed to shopping even with the wild fluctuations. Between May and June 2020, total retail sales in the United States increased 7.5 percent. Clothing and clothing accessory sales were especially popular among consumers, growing 105.1 percent in the same time period, rebounding from declines in February and March.
Predicting what will happen in the coming months or year is next to impossible. Another stimulus check or a flattening of coronavirus infections could push sales higher—or not. Looking further ahead to that time when all retail doors are open, Joe Jensen, vice president in the Internet of Things (IoT) Group and general manager of the Retail, Banking, Hospitality and Education Group at Intel, says that in-store experiences will have to outshine the convenience of online shopping or risk losing their customers. (Listen to more of Jensen’s predictions in our recent IoT Integrator Wire podcast.)
Smarter Shopping Experiences
To level the retail playing field, WonderStore, an Italian computer vision software boutique, is providing brands a raft of valuable in-store data combined with business insight. WonderStore offers multiple retail solutions that raise the bar of in-store shopping experiences by leveraging data in ways similar to online retailers. Think about your most recent online shopping experience, how the retailer collected data from your clicks to offer a selection of similar products or a matching accessory. Online retailers are unmatched when it comes to collecting data from our shopping histories and using it to simplify and improve the purchasing process.
Brick-and-mortar stores rarely have that data readily available, but WonderStore offers a cloud-based service that provides those retailers with a different set of data. The WonderStore Brand Detector identifies clothing choices and displays that catch consumers’ attention. The platform relies on cameras with visual sensors that collect and analyze data captured at shop windows, entrances, and store shelves. Based on the recommendations and analytic insights, retailers can offer consumers better shopping experiences through relevant recommendations.
Image credit: WonderStore
Improving Appeal of Windows and Displays
Recognizing brands and making individual recommendations for multiple customers is no easy task. WonderStore does it with the help of the Intel® Distribution of OpenVINO™ toolkit and Intel® DevCloud for the Edge. Using computer vision algorithms and AI trained on models from the Intel® Distribution of OpenVINO toolkit, data is analyzed, aggregated, and anonymized before being sent to the cloud.
In real time, the results of the analysis power multiple smart shopping experiences in the store, as well as communicate data with sales associates and retail marketing teams. Retailers can do rapid A/B testing of shop windows or instore displays to increase revenue and improve targeting.
The results show WonderStore is making an impression within retail stores:
- Stores are making staffing decisions that put the right number of personnel on the floor when visitor flow is at its highest.
- WonderStore increases the shop window conversion rate by 16.7 percent by helping retailers understand which customers are looking at window displays and for how long.
- Sales are increasing from omnichannel campaigns that combine customer data with external applications for targeted email and social media engagements.
Using Deep Learning Models
For systems integrators, WonderStore is not only a smart store solution, it is also a great example of what’s possible with the Intel® DevCloud for the Edge and the Intel® Distribution of OpenVINO toolkit. To reduce time to market, WonderStore tapped into the deep ecosystem of Intel technologies.
- The pre-trained deep learning models included in the Intel® Distribution of OpenVINO toolkit rapidly improved the quality and performance of the visual recognition features.
- Developers can rapidly iterate on new models for inference from video or images, with scalable storage that easily accommodates large data sets, and leverage pretrained models to quickly develop new solutions, like the Brand Detector, on top of existing ones.
For the WonderStore product launch and release, the company’s developers created a GDPR-compliant smart store platform powered by Intel® processors, Intel® HD Graphics, Intel® VPUs, the Intel® Distribution of OpenVINO toolkit, and Intel® DevCloud for the Edge.
Find out more about how the two worked together:
- This solution brief explains more about how Intel® DevCloud for the Edge and WonderStore can change the in-store customer experience.
- Learn, build, and test computer vision solutions by signing up for Intel® DevCloud for the Edge at devcloud.intel.com/edge.
- Learn more about WonderStore.
- Checkout the Intel® IoT Market Ready Solution from WonderStore.
Note: Intel is committed to respecting human rights and avoiding complicity in human rights abuses. See Intel’s Global Human Rights Principles. Intel’s products and software are intended only to be used in applications that do not cause or contribute to a violation of an internationally recognized human right.