Computer vision in retail is a dynamically developing field. More and more retail and e-commerce companies utilize various computer vision solutions in order to support their customers and manage inventory. In this article, we are going to show you some of the real-life applications of computer vision in retail and e-commerce. Buckle up; we have some pretty amazing applications in store!
Computer vision in retail becomes more and more prevalent. We can say that it all started with… Google. On October 27, 2009, Google Images added a feature to its image search that allows you to find similar images. And next, in May 2011, Google introduced a sort-by-subject feature for a visual category scheme overview of a search query.
Thanks to these new features, you can upload a picture into Google Images and search for similar ones on the web. This is exactly how computer vision works. The Google algorithms scan the uploaded file, analyze it, and search for similar images. The new features introduced in 2009 and 2011 significantly improved the traditional Google Images search engine and opened a way for computer vision in retail and e-commerce.
Today, computer vision in retail is a global standard that helps in:
- Organizing the inventory and monitoring the way it’s presented on the website
- Finding the product a customer seeks
- Improving customer service and user experience (UX)
- Understanding what each customer is looking for
Computer vision in retail: Related technologies
Computer vision is not some standalone technology that works entirely on its own. Rather, it’s a set of various technologies and applications. For instance, when talking about computer vision in retail, we ought to mention Augmented Reality and Virtual Reality. Both AR and VR can be used to help you make an informed decision about your purchase.
The Augmented Reality algorithms place computer-generated images into a real image of your living room/bedroom/garage/garden/. These computer-generated 3D images are simply visualizations of products that can be purchased online. All you have to do is:
- Download the dedicated mobile app to your device and choose the item you like. Sometimes, you don’t even have to download anything. The AR feature is accessible directly from the online store’s website.
- Pick the product you want to buy and use the AR extension.
- Next, you should see a 3D virtual image of this product placed on your desk or in your kitchen.
So, even before you buy it, you can decide if it goes well with your interior. The same happens with clothes and cosmetics. This way, you can see what they are going to look like without leaving your house or purchasing them.
You may find it interesting – How To Maximize Profits In Retail Using Business Intelligence
COMPUTER VISION: YOUR NEW CUSTOMER ADVISOR?
Here’s yet another interesting application of computer vision in retail. Perhaps, in the near future, CV algorithms will be so advanced that they will help you find the perfect product or an accessory matching your new jacket. After all, even today, e-commerce search engines help you find the product you seek. We reckon that soon they will be fully-operational customer advisors, available 24/7.
The vast majority of the technologies we’ve mentioned so far are already available! Now, we are going to examine these applications and see what the current real-life applications of computer vision in retail and e-commerce are:
Computer vision in retail: The real-life examples
FIND IT ON EBAY & IMAGE SEARCH
In 2017, eBay introduced not one but two different computer vision features. Find It On eBay is a feature that’s available both via eBay app and mobile platform. It lets you share images from any social platform or web browser. All you have to do is “share” the image. EBay’s computer vision engine will find listings of the item in your image or if nothing can be found, similar products. And then, there’s the Image Search feature. With it, you can take a photo of a product you intend to purchase and put it into the eBay search bar. The search engine will show you products matching the one in your picture.
It’s also worth mentioning that computer vision is not the only technology that’s used by eBay to analyze images. When you upload images to run Find It On eBay and Image Search, eBay also uses a deep learning model called a convolutional neural network to process the images. As you know from our other blog posts, deep learning is the most complex AI-related technology, enabling much higher accuracy and a better understanding of the picture in question. The fact that eBay uses such an advanced tool just to analyze pictures means it’s the real deal, and they understand how important this feature is in modern market conditions.
SEPHORA VIRTUAL ARTIST
Here’s a tremendous example of Augmented Reality in action. Sephora is a global retail chain of personal care and beauty products. Now, they bring AR and Virtual Reality technology to the beauty industry. Virtual Artist is a Sephora’s makeup app that uses facial recognition to allow customers to try on their beauty products.
Sephora’s app scans your face, detects your eyes, lips, and cheeks for product placement. When this part is done, the app lets you try on makeup products entirely virtually. As a result, you can quickly see if that eyeliner or that lipstick looks good on you. You can even compare various products and pick the perfect match before you place an order.
If you want to see what Virtual Artist looks like in action, just download Sephora’s mobile app, pick the product you’re interested in, and choose the “Try This Shade” button. The VA will start automatically.
It’s an iOS mobile app used as a 3D floor planner. This app also uses AR and works with real products that can be purchased online. Thanks to Amikasa, you can, all on your own, design your dream room/apartment, experience your design thanks to the in-the-room mode, view real products, and purchase them online. Amikasa is a 2015 Webby Award Winner and now can be downloaded from the Apple Store for just 0.99 USD.
While we are on the home furnishing subject, here’s another interesting application of computer vision in retail. IKEA Place is a mobile app that lets you virtually place true-to-scale 3D models in your very own space. Moreover, this app enables you to use the visual search feature that helps you find products similar to yours in IKEA’s database. Just like Amikasa, this app is available only for Apple device users.
Yes, we mentioned them in the previous article. Here, however, we want to take a look at LoweBots from the computer vision in retail perspective. LoweBots act like customer advisors and assistants that are available in the Lowe’s home improvement stores. All you have to do is approach the closest LoweBot and type or tell what you are looking for. The LoweBot can show you where your product can be found or, if you need it, take you there!
Especially sellers will be interested in this use case. Scandit is a company that offers barcode scanning capabilities, extended by CV, OCR, and AR functions. Their technological development is based on scanning barcodes and QR codes using cameras built into mobile tools. For instance, Scandit allows warehouse managers and employees to identify which goods are in packages quickly. Additionally, thanks to AR technology, it gives access to broader information about a given product or package. Interestingly, Scandit’s capabilities go far beyond the retail industry::
- Airline staff can view passenger details from a boarding pass scan
- The healthcare companies and hospitals can scan to cross-check patients with medication
- Delivery drivers can see special instructions just by scanning a package
Furthermore, the OCR function enables you to verify prices, passports & IDs, IBAN numbers, VIN numbers, and LOT & REF codes. The scope of possible applications is, in fact, very broad.
Syte.ai is a Swiss-based AI company that operates primarily in the retail sector. They offer visual discovery solutions that comprise:
- Camera search: You can upload your photo or browse a gallery.
- In-Store solutions: Interactive discovery experiences and styling suggestions for customers
- Recommendation engine: this solution allows you to “Shop The Look” or find similar products and then purchase them.
- Discovery icon: If you have many products in your picture, but you’re particularly interested in just one, you can select it and find matching products.
According to Syte.ai, their computer vision products are utilized by Samsung, Sainsbury’s, Prada, C&A, River Island, and many more retail and e-commerce brands.
The number of computer vision in retail applications grows every year. In the near future, we can expect to see more advanced recommendation engines, personalized shopping experience, and more complex computer vision algorithms that will help you find the desired product, try it, and then place an order. Everything within several minutes, completely remotely.
If you are interested in computer vision in retail and would like to implement this technology in your retail/e-commerce company–feel free to contact us. Computer vision is one of these technologies that we use every day. We will gladly help you pick and implement the solution that’s fully tailored to your needs.