Image: Adobe
Photoshop’s Firefly AI tools are now available on Photoshop for the Web, Adobe said Tuesday, along with a new Generative Match feature that allows AI to match the existing style of an image when generating new content. The latter is part of Firefy Image 2, Adobe’s first AI update.
Adobe’s announcement continues the drumbeat of new AI releases from Adobe, the most recent of which confirmed that it would offer Firefly as part of a credit plan. While Firefly is now available for free as a standalone tool, Firefly-powered generative AI art capabilities were added to Adobe Photoshop in May and to Adobe Express in August.
Now, Adobe’s Generative Fill functions and Generative Expand functions (inpainting and outpainting) have arrived on Photoshop for the Web, which means that they’re available to Chromebooks. Interestingly, though, Adobe is saying that the tools are available to the new Chromebook Plus devices that were announced in early October, without saying if they won’t work on older devices.
Adobe Express is the next tool on Adobe’s list to receive the AI treatment. Express will receive AI-powered Generative Fill and Text to Template, plus Translate and Drawing and Painting features.
Meet Firefly Image 2
The AI improvements include Firefly Image 2, which “generates higher-quality images and illustrations, enhances human rendering quality with improvements to skin, hair, eyes, hands, and body structure, offers better colors and improved dynamic range, and provides users greater ability to control output – ultimately helping produce creative visions faster.” It, too, is trained on Adobe’s own stock images to avoid copyright concerns.
The new Generative Match feature is a stylistic AI tool. When generating new AI content, users generally have a choice of styles. But previously, the AI generating the content wouldn’t understand the “big picture,” or literally how the new AI content would fit into the larger image. With Generative Match, Adobe is promising that this problem has been solved. Users can either upload their own photo to serve as a model (and must attest that it is original work) or choose from a list of selected images.
Adobe’s AdobeMax develoepr conference begins today, where the company may talk more about Project Stardust, its next-gen AI image editor that recognizes objects in a photo as actual objects that can be manipulated.
Author: Mark Hachman, Senior Editor
As PCWorld’s senior editor, Mark focuses on Microsoft news and chip technology, among other beats. He has formerly written for PCMag, BYTE, Slashdot, eWEEK, and ReadWrite.
Recent stories by Mark Hachman:
Microsoft’s Copilot AI is stealing one of Midjourney’s best featuresMore workers are using AI, but they’re ashamed to admit itPhotoshop took my favorite feature and massively boosted it with AI
