Google Gemini adds image markup tools for targeted AI analysis

Google Gemini is developing image markup tools that will allow users to highlight specific portions of images to direct the AI’s attention during analysis. The feature, discovered in version 16.42.61.sa.arm64 of the Google app, represents a significant enhancement to Gemini’s existing image analysis capabilities by enabling more precise visual queries.

What you should know: The markup functionality lets users draw on images after uploading them from their gallery or capturing new photos with their camera.

• Users can highlight portions of images with circles or other markings to focus Gemini’s analysis on specific areas.
• The interface currently includes different color options, though their specific purpose isn’t yet clear.
• Google’s AI automatically recognizes highlighted areas as points of interest without requiring explicit instructions.

How it works: Once users mark up an image, they can direct Gemini to perform targeted analysis or operations on the highlighted regions.

• The system integrates with Nano Banana’s image editing capabilities, allowing users to quickly remove unwanted content from screenshots.
• Users can ask questions specifically about highlighted portions rather than the entire image.
• The tool maintains Gemini’s existing ease of use while adding precision to visual queries.

Why this matters: This enhancement addresses a key limitation in current AI image analysis by allowing users to specify exactly what they want the AI to focus on.

• Current image analysis tools often analyze entire images equally, which can lead to less relevant or diluted responses.
• The markup feature could significantly improve the accuracy and usefulness of AI-powered image analysis across various applications.
• It represents a step toward more intuitive human-AI interaction through visual communication.

The big picture: While the feature appears to use a somewhat generic interface rather than one specifically designed for Gemini communication, it demonstrates Google’s continued efforts to refine AI-human interaction.

• The company will likely further modify how users access and utilize the markup tools before public release.
• This development aligns with broader industry trends toward more sophisticated and user-directed AI capabilities.