Google Upgrades Circle to Search with Multi-Object Recognition and Virtual Try-On
Google Enhances Circle to Search with Multi-Object Search

Google Unveils Major Enhancement to Circle to Search Feature

In a significant development for Android users, Google has announced a substantial update to its popular AI-powered tool, Circle to Search. This enhancement, revealed by Sameer Samat, President of the Android Ecosystem at Google, during the Galaxy Unpacked event, introduces the ability to search for multiple objects within a single image all at once. The update marks a leap forward in mobile search functionality, addressing a key limitation of the previous version.

First Devices to Receive the Update

The Samsung Galaxy S26 series, including the Galaxy S26 Ultra, Galaxy S26+, and Galaxy S26, will be the inaugural devices to benefit from this upgrade. Additionally, Google has confirmed that its own Pixel 10 series—comprising the Pixel 10, Pixel 10 Pro, Pixel 10 Pro XL, and Pixel 10 Pro Fold—will also receive the update. This rollout strategy highlights a collaborative effort between Google and Samsung to bring cutting-edge AI features to premium Android smartphones.

How the Updated Circle to Search Works

Since its launch, Circle to Search has processed billions of queries monthly, allowing users to identify items on their screens by circling, scribbling, or highlighting them to initiate a Google search. Previously, the feature was optimized for single-item searches, but the new update eliminates this constraint. Google explained in a blog post that while the tool has been transformative for questions like "What are those shoes?" or "Where is this hiking trail?", users often seek information about entire scenes rather than isolated elements.

The upgrade leverages the advanced reasoning and planning capabilities of Gemini 3, Google's latest AI model. Instead of relying on a single visual match, the model now employs a multi-step process. It identifies the most relevant parts of an image, conducts several searches concurrently, and cross-references the results to generate a comprehensive response covering all selected items. This sophisticated approach enhances accuracy and efficiency, making searches more intuitive and thorough.

Practical Applications and Use Cases

Google points to fashion as a prime example of the update's utility, noting that shopping-related queries are already among the top uses of Circle to Search. For instance, if a user encounters an outfit on social media, they can now circle the entire ensemble—including clothing, shoes, and accessories—and receive search results for every component in one seamless step. This eliminates the need for separate searches, streamlining the shopping experience and saving valuable time.

Introduction of Virtual Try-On Feature

In addition to multi-object recognition, Google has integrated a virtual try-on option directly within Circle to Search. In markets where this feature is available through standard Google Shopping, users of the Galaxy S26 series and Pixel 10 devices can now access it from within the Circle to Search interface. This integration allows for a more immersive and interactive shopping experience, enabling users to visualize products before making purchase decisions.

Availability and Future Rollout

The updated Circle to Search feature is currently accessible on the Samsung Galaxy S26 series and Google's Pixel 10 devices. Google has indicated plans for a broader rollout to other Android devices in the near future, ensuring that more users can benefit from these advanced capabilities. This expansion underscores Google's commitment to enhancing AI-driven tools across the Android ecosystem, fostering innovation and convenience for millions of users worldwide.

The announcement reflects Google's ongoing efforts to refine its AI technologies, making them more versatile and user-friendly. As Circle to Search continues to evolve, it is poised to become an even more indispensable tool for navigating the digital landscape, from everyday queries to complex multi-object searches.