During its Adobe Max event in London, Adobe introduced the latest Firefly Image generative AI (genAI) model, promising enhanced realism and improved control over generated outputs. The next-generation Firefly model will be integrated into Photoshop, bringing several new features to the image editor later this year.
Adobe Firefly consists of generative AI models used for creating and modifying content like photographic style images, illustrations, and fonts. It will soon be available in Premiere Pro for video editing and a music-generation algorithm is also in development. The Firefly app can be used independently or within Adobe’s Creative Cloud suite.
After more than a year since its launch, Firefly’s Image model is now in its third iteration, surpassing the previous versions with enhancements in image quality, especially for images featuring people. This results in more realistic outputs, improved lighting, subject positioning, and a wider range of expressions.
Adobe Firefly generated image of an artist in a studio.
Adobe
The latest model introduces the Structure Reference feature announced by Adobe last month, allowing users to apply the structure of a reference image for more accurate outputs. The Style Reference feature also helps in creating a consistent image style.
Users can anticipate a wider range of output styles for illustrations, photographic art, and vector art for iconography with Firefly Image 3. Adobe mentioned that the model will have a better understanding of user prompts, accurately reflecting longer and more complex inputs than before.
2024-04-26 01:00:03
Article from www.computerworld.com