Photoshop’s desktop beta now includes new AI features. These tools can make backgrounds, improve details, and work with reference images.
Adobe has introduced new generative AI tools in its Photoshop software. These tools are designed to help users manage their creations better.
The standout feature is the Reference Image tool. This tool uses images uploaded by users to guide Adobe’s AI. It creates outputs that match the style and color of these images.
For instance, rather than continuously adjusting a text prompt such as “a blue vintage truck with flower decals,” users can now upload a picture for Photoshop to use as a reference.
“Prompting is a pain in the butt,” Ely Greenfield, chief technology officer for Digital Media at Adobe told The Verge.
“Why spend an hour trying to craft a three-paragraph prompt if you have an image that you’ve created that’s exactly the thing you want to reference? The saying ‘a picture is worth a thousand words’ applies here.”
Users need to own the rights to any images they use as references. Greenfield explained to the Verge that a notification about this requirement appears when the tool is first activated.
Additionally, Adobe is developing a “do not train” tag as part of its Content Authenticity Initiative to prevent certain images from being used as references. Images provided for reference will not be used to train the Firefly model.
Although users are responsible for image ownership, Adobe states that this referencing tool is still “safe for commercial use.” This is considered a significant benefit of Firefly compared to other generative AI models.
Other AI tools in the Photoshop beta app are Generate Background and Enhance Detail. Generate Background can create and replace background images for uses such as product photography. Enhance Detail improves the sharpness and clarity of images, making them look clearer.
Generate Similar is also offered, allowing users to take one of three images made by Photoshop’s Firefly tools and create content that looks alike.
Additionally, Generate Image lets users start with an empty page and make a whole image from a text description for the first time.
Adobe’s new third-generation Firefly model offers improved image generation quality over its earlier version. It is now available for anyone to test in a global beta through the Firefly web app, not just in Photoshop.
Adobe claims this latest model provides “unprecedented photorealistic quality” with enhanced lighting, positioning, and detail. Firefly Image 3 better understands lengthy, descriptive text prompts and produces clearer text in the images it creates.
Besides generative AI, Adobe is introducing new, regular tools to Photoshop that enhance the speed of creative tasks. These tools feature an Adjustment Brush, which allows users to alter parts of an image, like color changes, without damaging it.
There is also a new Adjustment Presets feature that quickly alters an image with filters, and an upgraded Font Browser. This browser provides instant access to more than 25,000 fonts in Adobe’s cloud directly within the Photoshop app.
What we think?
I think Adobe’s new AI tools in Photoshop will make creating images much easier and faster. The Reference Image tool sounds amazing because you can use a photo to guide the AI instead of typing a lot of details.
This will save time and help make exactly what I want. The other tools like Generate Background and Enhance Detail also seem really useful for making professional-looking photos quickly.
