Computer Vision Startups Disrupting the Retail Industry
Online retail has been growing steadily for years with no sign of stopping. Especially amid movement restrictions induced by the COVID-19 pandemic, research shows that global online sales jumped to $26.7 trillion in 2020. With the rise of ecommerce, one thing is abundantly clear: brick-and-mortar retailers need to innovate if they want to stay competitive.
The use of AI technologies like computer vision is rapidly increasing in the retail industry. AI-enhanced retail holds the promise to eliminate operational inefficiencies and provide shoppers with frictionless in-store experiences. In this article, we’ve put together a list of the most innovative computer vision startups in the retail space.
The Future of AI in Retail
In recent years, an increasing number of retail companies have started to quietly transform physical stores. Walmart, for instance, began installing an array of sensors, cameras and processors to monitor inventory levels, perform automated product quality checks, and more.
Many industry giants have followed suit by employing similar approaches to drive in-store efficiency, better logistics, prevent theft, and more. Research shows that the retail AI market is growing fast—according to a recent RIS News report, only 3% of retailers were utilizing computer vision technology at the end of 2020, yet an additional 40% had plans to deploy new solutions within the next year.
Computer vision solutions present retailers with ample opportunity to boost operations as well as enhance the shopping experience for customers. Among the most promising applications of computer vision include inventory management, loss prevention, automated checkout, and behavioral analytics. From employee-free shops to in-store surveillance, here are several computer vision startups disrupting the retail industry.
Computer vision startups in retail
- RADAR: RADAR is a fully integrated hardware and software solution to automate inventory management using RFID and computer vision techniques. Their mission is to streamline inventory management via automated inventory counts, improved in-store replenishment and instantaneous customer stock checks.
- Trax: Singapore-based startup Trax provides an in-store solution that uses a combination of computer vision models and hardware to keep track of their inventory in real time. This solution ensures out-of-stock items are repurchased efficiently, while expired items are pulled off from the shelves. The company holds 23 patents on its technology and can analyze images from phones, in-store cameras, and grocery store robots.
- Standard.ai: Previously known as Standard Cognition, Standard.ai’s automated checkout solution is made to fit with retailers’ existing stores and technology. They boast an easy to install camera-first solution that doesn’t employ the use of turnstiles or gates. Standard doesn't use any facial recognition or biometrics, and all deployments are on-premise to ensure maximum performance and security for retailers and shoppers alike.
- Trigo: Using proprietary algorithms and affordable off-the-shelf sensor kits, Tel Aviv-based Trigo allows retailers to analyze anonymized shoppers’ movements and product choices in real time. The system automatically compiles selected items into a virtual shopping list, enabling shoppers to leave without going through a traditional checkout line.
- Accel Robotics: Accel Robotics provides checkout-free shopping experiences across existing and new store formats with its patented camera-based AI system. They recently launched Valet Market, a completely automated convenience storefront without cashiers or checkout kiosks.
- StopLift: With roots in MIT’s artificial intelligence labs, StopLift analyzes security video and POS data to distinguish between legitimate and fraudulent behavior at checkout. By applying advanced computer vision algorithms to existing camera feeds, StopLift’s ScanItAll system is capable of tracking items that pass through the checkout lane, associate them with POS, and flag suspicious activity as it happens.
- Vaak: Japanese startup Vaak provides a cloud-based computer vision system that monitors retail security camera footage for suspicious behavior. Already deployed in over 50 stores within Japan, VaakEye analyzes movement at more than 100 points across the body, automatically weighing behavior for suspiciousness. Once a customer reaches a certain threshold, the system sends an alert, along relevant video clips, to the appropriate staff member.
- Deep North: Deep North provides an analytics platform that builds real-time video intelligence for retailers based on video data from CCTV and other cameras that those retailers already use. Deep North’s proprietary technology captures parameters as daily entries and exits, customer occupancy, queue times, conversions and more.
- Advertima: Based on information captured by visual sensors, Advertima’s platform provides retailers with a real-time view of what’s going on in physical stores as shoppers move through the space. The platform claims to only process minimal anonymized data, without storing any recordings or personal information for future use.
- Aura Vision: Using existing security cameras, Aura Vision help retailers improve in-store experiences by tracking visitor demographics, movement and engagement from existing security cameras. The platform provides insight into window and product display performance, queue times, in-store promotions, and marketing campaigns.
Real-time visibility is essential to operating brick-and-mortar retail. That’s why more and more retailers are employing computer vision in an effort to increase operational efficiency, better the customer experience, and gain an edge over competitors.
Early adopters are already seeing great results—according to estimates from RBC Capital Markets analysts, cashierless Amazon Go stores bring in about 50% more revenue on average than typical convenience stores.
Visual data management for retail
Companies that build computer vision solutions for retail are constantly building and growing their ML training data sets. Today, most companies have to rely on internal tools or manual solutions like spreadsheets to do this.
SiaSearch helps retail companies to simplify and speed up this process with a lightweight API that simplifies data exploration, visualization and selection. As a result, companies can reduce annotation costs and increase model performance.