Whisk allows users to prompt with images rather than text, enabling the remixing of photos by altering their subject, scene, and style, enhancing creative possibilities.
Google's Imagen 3 model combines three images—a subject, a scene, and a style—to generate remixes, allowing intricate modifications and the option to add descriptive text prompts.
While promising, Whisk's results may vary due to its focus on key characteristics, meaning outcomes might not align with user expectations regarding physical attributes.
Currently an experimental tool available to U.S. users only, Whisk represents a new frontier in image editing, exploring the combination of visual elements with generated captions.
Collection
[
|
...
]