The American company Adobe has officially demonstrated its new program – the artificial intelligence Firefly, which can create any image from a text description. This has been reported Web site Developer known for photo editor Photoshop.
It is stated that the artificial intelligence is in the beta testing phase and access to it can be obtained by invitation only.
Visually, Adobe’s Firefly is similar to the Midjourney and DALL-E generator AIs on the market. Work from these programs is already actively shared on social networks by users from all over the world.
To simplify text annotation, developers from Adobe built Firefly the ability to select lighting options, artistic styles, and aspect ratios before rendering an image. Also, Firefly added the ability to apply filters to an already finished image in the form of the Instagram social network (its owner, Meta, was considered extremist and banned in Russia).
Adobe has guaranteed that only non-copyrighted images are used for AI training. Adobe Stock, including the image library, was taken for development. However, the company has announced that it plans to let artists train Firefly on their own images in the future.
Previously using Midjourney neural network created frames From The Last of Us, if the action of the series took place in the USSR during the pigweed epidemic.
Source: Gazeta
Jackson Ruhl is a tech and sci-fi expert, who writes for “Social Bites”. He brings his readers the latest news and developments from the world of technology and science fiction.