Whisk, Google’s New Tool for Creating Custom Images | Ai art generator from text | Ai image generator from image free online from text | Image to image generator | Turtles AI

Google is testing Whisk, an experimental image generator that allows users to remix three photos into a single creation. Using the powerful Imagen 3 template, users can combine a chosen subject, environment and style to create unique works. Currently, the service is only available in the United States.

Key points:

• Whisk allows users to combine three images: subject, scene and style.
• The system uses Google’s Imagen 3 model to generate remixed images.
• Additional details can be entered to better define the desired result.
• The service is currently accessible only to users in the United States.

Google Labs, the tech giant’s experimental division, has launched an innovative image generator called Whisk, which aims to offer a unique experience in visual creation. This tool, which harnesses the power of the Imagen 3 model, aims to improve the way users interact with images, enabling them to remix and customize photographs in intuitive and creative ways. The distinguishing element of Whisk is the ability to upload three different images that will form a unique composition. Each image represents a specific aspect: one relates to the main subject, another to the environment in which it fits, and a third defines the visual style you want to adopt. For example, you can choose a personal photo as the subject, a futuristic landscape as the background, and an anime-inspired graphic style to achieve a striking and personalized result. The entire operation is done through a system that, from the uploaded images, automatically generates a descriptive caption that serves as a guide for creating the remix, making the process accessible even to those unfamiliar with the use of text descriptions. However, the experience can be further enhanced by the inclusion of textual prompts, which allow users to further refine the creation. In this regard, details such as “The subject is sailing on a flying boat” can be added to bring more imaginative and specific situations to life. Despite the ease of use, the results generated may not exactly reflect expectations in terms of the subject’s appearance, as the AI mainly focuses on the main aspects of each image, with possible variations in height, weight, hairstyle and skin tone. To ensure greater customization, users can view and modify the prompts used at any time, thus tailoring the result to their preferences. Currently, the experiment is only available to a limited number of users in the United States, accessible via labs.google/whisk.

Pending a future extension of global availability, Google’s Whisk project represents a new frontier in automatic image generation, combining computational power and creativity in a way never seen before.