Google has announced the launch of native image editing in the Gemini app. From now on, users can edit both images created with AI and those uploaded from their own devices. Key features include background changes, object replacement, adding new elements, and changing the image style. For example, you can upload your own photo and ask Gemini to show how a person would look with a different hair color.
Editing is performed using text prompts, allowing you to make step-by-step changes to the image while preserving the context of previous requests. This enables more precise and comprehensive results by integrating text and images in responses. The service can also create illustrated stories, for example, preparing a first draft of a fairy tale about dragons along with relevant illustrations.
To ensure transparency of image origins, all images created or edited with Gemini will contain an invisible digital watermark SynthID. Additionally, Google is testing the possibility of adding visible labeling to all images generated by the model.
The native image editing feature is being rolled out gradually starting today. Over the coming weeks, it will become available to users in most countries and will support more than forty-five languages.
