Canva launched AI-functions to turn the text into an image
CANVA graphic designers platform presented a function based on artificial intelligence to generate images at a text request.
Go to the more Tab and Select Text to Image. Describe The Image You Want to See and The Choose Your Style of Photo. Pic.Twitter.COM/0NMNFMWWRZ
– canva (@canva) november 8, 2022
Text to Image is based on a text converting model into an image with an open source code Stable Diffusion. The developers have added several additional safety filters and a custom interface.
To use the function, you need to describe the desired image and choose one of the styles. At the output, the model will give a net of four pictures and add them to the design canvas.
Also, users can complain about images with violence, nudity, inciting hatred and “biased or stereotypical” content.
The function is available for all plans, including free. Users can create up to 100 images per day.
Examples of the generated image (left), style selection (center) and form of a complaint about unacceptable content (right). Data: canva.
“We strive to present to the [community] this technology, because it is a developing area,” said the co -founder and director of Canva Cameron Adams in an interview with The Verge.
According to him, the tool is already used in many areas.
“One of my favorite ways is how students use it to visualize their stories […]. We also saw how he [tool] creates images for presentations, leaflets and t -shirts that they [users] can print through Canva, ”Adams said.
He also noted that the company is studying the issues of the author’s right to images generated by artificial intelligence:
“To some extent, AI products can be considered conscientious use, and this will differ in different countries of the world. We are closely following the situation “.
Canva launched a beta testing of the tool in September 2022. Now the function is available to all 100 million platform users https://gagarin.news/news/japan-and-sryptocurrency-startups-are-tax-benefits-forthcoming/.
Recall that in November Photo editor Photoroom learned to change the background in images using AI.
In October, Shutterstock expanded the partnership with Openai to sell stock images generated by the Dall-E 2 model.
Subscribe to FORKLOG news at Telegram: Forklog AI – all news from the world of AI!