Amazon has announced the launch of Lens Live, a new feature in its shopping app that allows users to instantly scan products and view real-time matches through a swipeable carousel. The company says this tool is designed to make it easier for customers to find items they encounter on social media or while out in public.
The Lens Live feature integrates Amazon’s AI shopping assistant, Rufus, which provides product insights, summaries, and answers questions as customers browse. According to Amazon, “When customers with Lens Live open Amazon Lens, the Lens camera will instantly begin scanning products and show top matching items in a swipeable carousel at the bottom of the screen, allowing for quick comparisons. Customers will also now have the useful ability to tap an item in the camera view to focus in on a specific product, add items directly to their cart by tapping the + icon, and save to their wish lists by tapping the heart icon, all without leaving the camera view.”
To further assist shoppers, Amazon explains: “To help customers using Lens Live learn more about products they’re viewing, we’ve integrated our AI shopping assistant, Rufus, into the experience. While in the camera view, customers will now see suggested questions and quick summaries of what makes a product stand out. These conversational prompts and summaries appear under the product carousel, allowing customers to perform speedy research, quickly access key product insights, and get their questions answered.”
Lens Live is currently available for tens of millions of U.S. customers using iOS devices through the Amazon Shopping app. The company plans to expand access over the coming weeks.
The technology behind Lens Live relies on AWS-managed OpenSearch and SageMaker services for large-scale machine learning model deployment. It uses an on-device computer vision object detection model for real-time identification of products as users move their cameras across different scenes or focus on particular items. This approach aims to minimize customer interaction while maintaining accuracy.
Additionally, Lens Live uses deep learning visual embedding models that compare what is seen through a customer’s device against billions of products available on Amazon’s platform. This enables retrieval of exact or highly similar items from its catalog.
Customers who prefer traditional methods can still use existing options such as taking photos or uploading images within Amazon Lens.
