Google has just introduced a new artificial intelligence called Whisk, which can generate images without the need for textual commands from users. This artificial intelligence can get ideas from other images provided by users to generate images.
Whisk is one of the latest products of Google’s artificial intelligence laboratory, which works like most similar services in the field of image production. The most important difference between Whisk and its competitors is that users can give Whisk images similar to what they want, and this service will produce the most similar image.
Getting to know the artificial intelligence capabilities of Google’s Whisk
Unlike most existing image-generating artificial intelligence models, Whisk does not depend on textual prompts to generate images. Instead, it gets ideas from images that users give it as examples. Also, if you don’t have a specific image in mind, you can click on the bald icon so that Google will fill the requests with its artificial intelligence images.
If users wish, they can still use the traditional method of text command and in the end, if they need more details in the image, they can create the necessary edits in the image by writing the text command.
At the end, Whisk’s artificial intelligence will produce the desired image along with the text command and you can also download the image. Of course, if you are not satisfied with the result, you can still edit the image.
In its blog post, Google emphasized that Whisk AI is designed for a quick visual exploration experience, not advanced editing. Google also says that this service may have errors in following users’ commands; For this reason, users are allowed to edit prompts.
Google has used the Imagen 3 model to create this service. The company has also introduced a new artificial intelligence called Veo 2, which is used for video production. According to Google, Veo 2 has an understanding of the unique language of cinematography.
RCO NEWS