Meta launched an independent AI image generator, which is free but only supports English prompts.

Thanks to CTOnews.com netizens Baa Yang, soft media users 1520111 clue delivery! CTOnews.com, December 7 (Xinhua)-- Meta has launched a new, independent AI image generator, Imagine with Meta, which allows users to create images through natural language descriptions.

According to reports, the new artificial image generator is supported by Meta's existing Emu image generation model and can create high-resolution images according to text prompts. It is currently free for English users in the United States (it is unknown whether to charge later), and each prompt generates four images.

Previously, the Meta image generation model faced controversy because of racially biased image stickers. In order to solve this problem, Meta said that invisible watermarks will be added to the images generated by Imagine with Meta. These watermarks will be generated by artificial intelligence models and can be detected by corresponding models to improve content transparency.

CTOnews.com noted that Meta last announced two AI-based image editing tools for Facebook and Instagram, "Emu Edit" and "Emu Video", which include photos and videos.

According to officials, Emu Edit can accept user instructions for various forms of editing, including regional and global editing, removing and adding backgrounds, adjusting colors and vector image conversion, or detecting and segmenting image elements.

Meta uses 10 million synthetic data sets to train Emu Edit, which claims to be the largest of its kind.

Meta said that Emu Edit uses visual tasks as instructions into the generated model, which in turn provides better control in video generation and editing. The researchers point out that the current image editing model usually overmodifies the image or undermodifies it, while the advantage of Emu Edit is that it can accurately follow the instructions.