Google Whisk is a new way to create AI visuals using image prompts here's how to try it

Date:
Sun, 22 Dec 2024 09:00:00 +0000

Description:
Google's new AI tool uses images as inputs instead of text-based prompts here's how to take it for a spin.

FULL STORY ======================================================================Google Whisk uses images as inputs instead of text-based prompts It's built on Googles Imagen 3 generative AI model The experimental tool is free to try for users in the US

Googles new AI tool makes it easier to create and remix your visual concepts. Instead of asking you to describe whats in your minds eye, Whisk lets you input three image prompts: one for subject, one for scene and one for style. Whisk takes care of the rest, making it a more intuitive way to experiment with different ideas.

While most of the best AI image generators require you to write a detailed prompt, Whisk handles that behind the scenes. When you drop pictures into the web-based Whisk interface as inspiration, Googles Gemini model automatically analyzes them and writes a detailed caption for each. These are then fed into the Imagen 3 model, to create a matching image.

For example, you could drop in an image of a car as the subject and a photo
of a rural landscape for the scene. You could them add a watercolor as the style to see what Whisk creates. Hit the button and youll get a pair of
images based on your inputs.

From here, its easy to remix the images. The interface allows you to specify additional text-based details to tweak the outcomes. You can also easily drop in different source images or roll the dice if youre in need of inspiration. New results appear in pairs in the feed, making it an intuitive way to
ideate. You can also choose to refine images by revealing the text prompt and adding more details. Whisk it up

While Whisk is designed to eliminate the need for text-based prompts, Google includes the option to refine the written prompts because results wont always match up to the source material.

In a blog post about the experimental tool, Google explains that Whisk, captures your subjects essence, not an exact replica. Its only as effective
as Geminis analysis of the images you submit. While this is generally very impressive, it also isnt able to get inside your mind: you might expect Whisk to pull out one detail from an image, where it focuses on another.

The post explains further: Since Whisk extracts only a few key
characteristics from your image, it might generate images that differ from your expectations. For example, the generated subject might have a different height, weight, hairstyle or skin tone. We understand these features may be crucial for your project and Whisk may miss the mark, so we let you view and edit the underlying prompts at any time.

Even with these shortcomings, Whisk an interesting application of Googles existing AI tools. The underlying generative models are the same as if you were chatting with Gemini via its text interface. By relying on image inputs, though, Whisk is a more accessible and intuitive way for visual creators to play with their ideas.

Based on early feedback from digital creatives, Google refers to Whisk as a new type of creative tool which is intended for rapid visual exploration, not pixel-perfect edits. How to try Google Whisk

Google Whisk is currently only available to users in the US. If youre based there, you can try it out via your web browser at labs.google/whisk .

The experimental tool is completely free to play with. Data from your experience with Whisk will be fed back to Google to help refine and develop future AI products. You might also like... What is Google Gemini? Everything you need to know Best AI image generators I tested five AI art generators



======================================================================
Link to news story: https://www.techradar.com/computing/artificial-intelligence/google-whisk-is-a- new-way-to-create-ai-visuals-using-image-prompts-heres-how-to-try-it


--- Mystic BBS v1.12 A47 (Linux/64)
* Origin: tqwNet Technology News (1337:1/100)