Tools like DALL-E, which create images based on text you type, have recently generated buzz on social media. But now the design, photo, and video software juggernaut Adobe is delving into this intriguing AI technology branch, too.
As the company outlined at its 2022 Max conference, the new capabilities will show up as beta options in Photoshop, Lightroom, Premiere Pro, and other applications that are part of the company's Creative Cloud suite.
Adobe revealed 10 projects involving AI during the conference's Sneaks session:
- Project Clever Composites: Turns the once-tedious process of compositing images into backgrounds into a simple drag-and-drop process with auto adjustments for color and shadows.
- Project Instant Add: Simplifies the typically complex and time-consuming process of editing videos and adding VFX effects in post-production by using AI to automatically map graphics to the chosen element in the video.
- Project Magnetic Type: Lets anyone attach and instantly fuse any shape—even hand-drawn calligraphy—to a piece of live digital text.
- Project Vector Edge: Provides designers and their teams with the ability to visualize, make edits, and collaborate on 2D designs within 3D environments during the design process.
- Project Motion Mix: Enables creators to easily create high-quality looping human animations, complete with realistic human movements, all from a still image.
- Project Blink: Helps anyone find and quickly pull highlights out from video content, making editing videos as easy as editing text. Project Blink is now available in beta and you can sign up for beta access.
- Project Artistic Scenes: Transforms 3D scenes using styles from 2D artwork. Creators can quickly create high-quality stylized 3D content that will be essential for the future of VR and AR experiences.
- Project All of Me: Makes un-cropping images a breeze. Leveraging AI, Project All of Me generates un-cropped components of photos, edits out unwanted distractions, and can even modify the photo subject’s apparel.
- Project Beyond the Seen: Easily generates full 360-degree panoramas from a single 2D image to create realistic and fully immersive 3D environments.
- Project Made in the Shade: Allows users to move a person or object within a photo and cast realistic shadows, making 3D image editing easy and intuitive.
You can watch demos of the new tech on the Adobe Max Sneaks YouTube channel.
The most fascinating of these for me is Project All of Me, as well as those that create 3D scenes from 2D images, (though CyberLink did something like this several years ago, in the days when 3D video was a hot topic).
Though it's not quite fully generating an image from text as DALL-E and other do, All of Me doesn't simply reproduce a consistent background, but actually creates body parts and clothes that don't exist in the original image.
At the Adobe Max conference, company execs also mentioned a Backdrop Neural filter in beta for Photoshop, which will let you tell the software to, for example, "Take me to Thailand" and then generate an appropriate background, or to tell Lightroom to change a photo from day to night.
According to company press materials, "In Photoshop, Generative AI will let you add new objects to an image, create variations based on your existing image, or even add a new section to your composite just by describing what you want." It also mentions that Adobe Express will have AI generative features, though Canva already has a beta AI text-to-image generation tool.
Adobe announced a gaggle of other new features for its software, which you can read about in our Adobe Max 2022 news post.