Google launches new AI feature: now you can create images based on text

发布于 作者 量尺寸留下评论

Google officially announced that starting from Thursday (October 12), Google’s Search Generation Experience (SGE) will open a new feature that can create images directly from text prompts.

Microsoft has provided the image creation function for Bing chats as early as March this year, based on OpenAI’s DALL-E model.

Now, users can enter relevant queries into the Google search bar, and once the input is complete, SGE can create a number of images based on the prompts given, and the user can choose from them.

Google spokesman Craig Ewer said the tool is powered by the Imagen family of artificial intelligence models.

In one GIF, Google shows how the tool creates an image based on the prompt “Draw a capybara wearing a chef’s hat and making breakfast.”

And as shown in the animation, users can also edit and optimize the input text information after selecting the center image. In addition, users can also use Google Images to create images generated by Google’s artificial intelligence. When you browse Google Images search results, you may see a box prompting you to generate images based on your query – the generated images will appear in the sidebar.

Google product manager Hema Budaraju emphasized in the article that the company is working hard to ensure that its image generation tool is launched in a responsible manner and that the tool is not allowed to generate images that violate the company’s relevant policies. Budaraju said that each image produced with SGE will “have metadata tags and embedded watermarks to indicate that it was created by artificial intelligence”; and that Google will not let users create images related to celebrities. Google limits image generation to users 18 or older, meaning teenagers won’t be able to use it.


您的电子邮箱地址不会被公开。 必填项已用 * 标注