Google's Whisk AI experimental tool can mash-up images to generate unique outputs

 

Google unveiled a new experimental artificial intelligence (AI) tool on Monday that can combine photos to produce a unique result. Whisk is a delightful instrument with no other purpose than to do its intended job. The Mountain View-based tech behemoth has lately published many entertaining AI tools, like GenChess, which use the Imagen 3 AI model to produce unique chessboard pieces. Whisk demonstrates how AI may utilize pictures as a stimulus to make unique art.

Google's Whisk can 'Remix' input images
The new AI tool was revealed in a blog post by the tech giant. Whisk is presently only available in the United States and may be accessed through Google Labs, the company's portal for releasing experimental products developed using native AI models. Whisk, like all other tools, is experimental, and Google warns that it may not always operate as users expect.

AI picture generators are rather prevalent; nevertheless, the most of them only accept text or a combination of text and images as input. In summary, picture creation models require natural language inputs in order to determine what to make. However, Whisk differs from similar models in that users can only upload photos to trigger the model to generate outputs.

Whisk requires users to upload three images: one for the subject, scene, and style. Once inserted, the AI tool automatically interprets the visual information to produce a unique image that is a composite of all three input photographs. Users can also produce output by just adding two photographs, one for the person and one for the scene.

Google revealed that the Gemini model examines the photos and generates a comprehensive natural language prompt, which is then put into the Imagen 3 model. The prompt seeks to capture the essence of the photos rather than generating an objective blend of the inputs.

Whisk is an experimental model, therefore the generated visuals may differ from the user's expectations. Whisk allows users to enhance and change photos after they have been generated, giving them greater control over the end result. Users may quickly verify the underlying prompt created by Gemini and edit or add extra information to get the desired effect.

"We designed it for quick visual exploration, not pixel-perfect editing. "It's about exploring ideas in new and creative ways, allowing you to go through dozens of options and download the ones you like," Google explained.

#google #googlewhisk #googlewhiskai #newsmintora #googlenews