news 
Google Experiments with New Image Generator That Remixes Three Images into One Creation
In a bold move to expand the possibilities of artificial intelligence in visual content generation, Google has unveiled a new experimental tool called Whisk. This innovative technology allows users to prompt the system with images rather than text, enabling them to create unique, mixed-media creations by combining elements from three distinct images.
According to reports, Whisk employs Google’s cutting-edge image generation model, Imagen 3, to merge the subject, scene, and style of the input images. This means that users can select a photo of themselves as the subject, a futuristic landscape as the scene, and an anime-inspired style for the final product.
When utilizing this tool, users are able to provide detailed descriptions or “prompts” to further refine their desired outcome. For instance, one could specify that the subject in the resulting image is riding a flying bike. This level of control over the creative process is unprecedented and has the potential to revolutionize the field of computer-generated imagery.
However, Google warns users that Whisk’s results may not always meet their expectations. The system may modify certain characteristics of the input images, such as height, weight, hairstyle, or skin tone, which could lead to unexpected outcomes. Nonetheless, this technology represents a significant leap forward in AI-powered content creation and has immense potential for artistic expression.
As part of its ongoing efforts to push the boundaries of machine learning, Google has made Whisk available exclusively to users based in the United States. Those interested can access the tool at labs.google/whisk, where they can experiment with the capabilities of this innovative technology.
The introduction of Whisk highlights Google’s commitment to exploring new frontiers in AI research and its dedication to empowering creators and innovators through cutting-edge tools and technologies.
Source: techcrunch.com
