The upcoming update aims to solve this problem by introducing two new ways to fine-tune AI-generated images. The first method will enable Gemini to understand natural language prompts for modifying existing images. For instance, if you like the overall image but want to change the background, you can simply tell Gemini to do so without having to create a whole new image.
The second method is even more intuitive. You can use your finger or a stylus to circle the part of the image you want to change, and then tell Gemini what edits you want to make. This eliminates the need to describe the specific area you want to modify, as Gemini will automatically understand the circled part.
Examples of Gemini edits to its previously generated images | Images credit — Android Authority
While this feature is still under development and not yet fully functional, it’s a promising addition to Gemini’s capabilities. It suggests that Google is committed to enhancing the user experience by providing more control and flexibility over image generation.
I can see this update being particularly beneficial for creative professionals and those who frequently use AI-generated images. For example, for YouTube thumbnails or for blog article headers. It will streamline the editing process and allow for greater customization, ultimately saving users time and effort.