The team demonstrates how computer vision can help transform e-commerce online shopping
BOURNE END, United Kingdom – July 13, 2022 – Zebra Technologies Corporation (NASDAQ: ZBRA), an innovator in the front line of business with solutions and partners that provide a performance solution, today announced a team of Zebra AI researchers in the United Kingdom that took second place in one of the world’s most popular computer vision challenges. The team’s challenge rankings were announced in June at a CVPR RetailVision workshop designed to address the problems of online e-commerce searches.
“If a consumer logs into an e-commerce website and starts searching for clothing items, for example, the website will return a variety of results, some better than others,” he said. by Andrea Mirabile, Senior Manager, Computer Vision, Zebra Technologies, who leads the Zebra team.
Sponsored by Alibaba Group and Trax, the CVPR 2022 Challenge: Large Cross-Modal Product Retrieval brings together 650 research teams from around the world to work on a large multimodal retail dataset of nearly five million pairs. image caption of circa 100,000 products. Each team is tasked with finding top-K product candidates that fit a question such as ‘blue men’s turtleneck sweater.’
“From the consumer’s point of view, it’s a less optimistic experience, wasted time and effort, which could result in any order, ordering the wrong item, or having to look elsewhere,” Mirabile added. . “From the retailer’s perspective, it’s about the problem of how to better use words, images, and search functions that match what consumers are looking for when they shop and are able to make it more relevant. product recommendations. ”
Top-K is the most common measure of performance in machine learning and computer vision. In the context of limited and noise data, such as retail data with a mixed order of image quality and text caption, the use of a loss function (the function that calculates the distance between the current output of the an algorithm and the expected output) are designed. for the top-K classification can bring significant improvements.
“AI applications consist of speech, text, sound, and vision. Our challenge team and the broader Zebra global AI research team put them together, in the same way that one person uses their five senses to feel and analyze the world around them to inform the decision. -take action, “Mirabile explains.
Zebra’s computer vision researchers ranked second out of 650 teams in the prestigious global CVPR Computer Vision Challenge.
The challenge seeks to address the problems faced by consumers and retailers around e-commerce online search.
Zebra’s computer vision team pioneered multimodal AI techniques, combining multiple AI applications for sensing, analysis, and movement.
ABOUT ZEBRA TECHNOLOGIES
Zebra (NASDAQ: ZBRA) empowers organizations to grow the on-demand economy by making every front-line worker and asset on the edge visible, connected and fully optimized. With an ecosystem of more than 10,000 partners in more than 100 countries, Zebra serves customers of all sizes-including 94% of the Fortune 100-with an award-winning portfolio of hardware, software , services and solutions that digitize and automate workflows. Supply chains are more dynamic, customers and patients are better served, and workers are more engaged when they use Zebra’s innovations to help them understand, analyze and operate in real time. . Zebra recently expanded its industrial automation portfolio using its acquisition of Fetch Robotics and expanded machine vision and AI software capabilities with the acquisitions of Adaptive Vision, antuit.ai and Matrox Imaging. The Zebra at #25 on Newsweek’s inaugurallist of America’s Most Loved Workplaces and #79 on Forbes’ list of America’s 500 Best Midsize Employers. Learn more at www.zebra.com or sign up for news alerts. Follow Zebra’s Your Edge blog, LinkedIn, Twitter and Facebook, and check out our Story Hub: Zebra Perspectives.
+44 (0) 7408 864 597
Contact the Industry Analyst:
Facebook Facebook logo Sign up for Facebook to connect with Kasia Fahmy
ZEBRA and the Zebra styled head are trademarks of Zebra Technologies Corp., which are registered in many jurisdictions around the world. All other trademarks are the property of their respective owners. © 2022 Zebra Technologies Corp. and/or affiliates.