RedShark News - Video technology news and analysis

Adobe’s latest AI tools are frankly jaw-dropping

Written by Andy Stout | Oct 10, 2023 4:00:00 PM

Adobe Max is currently underway in Los Angeles and the company has kicked it off with a slew of product enhancements, including some new AI tools that look genuinely astonishing.

One of the things that is becoming rapidly apparent with the use of AI in post-production and image manipulation tools is that we've only really been in the foothills so far. Adobe Max has just started in Los Angeles and some of the capabilities that we are seeing could even be comfortably referred to as second generation. Not bad, considering the first generation of generative AI tools are barely a year old yet.

Take Project Stardust as an example. This is Adobe’s new object-aware image editing engine technology and is based on the fact that Stardust can automatically turn any image into editable layers. And as you can see from the preview demo below (“a sneak of a sneak,” says the company), that means you can move or remove objects simply by clicking on them. Have a look.

Given that it can do things like that, AI is unsurprisingly front and centre of the Adobe Max announcements with a host of new Firefly innovations for Creative Cloud. These are based on the latest Adobe Firefly Image 2 Model, which Adobe says is the latest model of its generative AI for imaging designed to produce content safe for commercial use. Updates include new text-to-image capabilities now available in the Firefly web application.

As of Max, text-based editing improvements in Premiere Pro and Roto Brush in After Effects will be generally available. Premiere Pro (beta) users will be able to publish videos directly to Instagram, Facebook, YouTube, and TikTok. And Adobe will debut Share for Review (beta), where Premiere Pro’s “Share” button will now intuitively introduce Premiere users to Frame.io with the click of a button for faster content sharing and collaboration.

Illustrator introduces a new Firefly Vector Model that is the world’s first generative AI model for vector graphics, powering new Text to Vector Graphic (beta) capabilities. This should allow users to generate editable, high-quality vector graphics, quickly from simple text prompts. Updates to Retype (beta) enable users to quickly identify and edit fonts from static text; while Mockup (beta) lets them quickly transform images and graphics into realistic-looking product and branding mockups.

Adobe Express features a new Firefly Design Model that will power the instant generation of quality template designs with Text-to-Template capability (beta), the results of which will then be fully editable in Adobe Express. New Generative Fill is also coming.

A new Lens Blur feature in Lightroom will make it easier than ever to add an aesthetic blur effect to any part of a photograph without requiring a camera lens. A revamped mobile-optimised editing experience will streamline the Lightroom mobile toolbar to prioritise the popular features

And finally, Adobe Stock will unveil the newest addition to its growing marketplace: Video templates, a rich new video template collection that brings together Adobe Premiere Pro, After Effects, and Motion Graphics templates.