Adobe has previewed some of its upcoming tools for generating artificial intelligence movies, including a new feature that can generate movie clips from still images. This latest preview builds on the Firefly video model under development that the software giant showed off in April, which will power AI video and audio editing capabilities in Adobe Creative Cloud applications.
A new promotional trailer shows off footage from Firefly’s text-to-video feature that Adobe announced (but didn’t demo) earlier this year. The tool allows users to generate video clips using text descriptions and adjust the results using a variety of “camera controls” such as camera angle, movement and shooting distance. The Firefly video model also demonstrates image-to-video functionality, which allows clips to be generated using specific reference images. Adobe says this can be useful for creating additional B-roll footage or patching gaps in your production timeline.
If the example footage is any indication of the final release, the generated video quality looks on par with what we’ve seen from OpenAI’s Sora model so far, which Adobe is also “exploring” as a third-party integration for its Premiere Pro video software. However, Alexandru Costin, vice president of generated artificial intelligence at Adobe, said that the duration will be limited edge The maximum length of videos produced by the text-to-video and image-to-video functions is five seconds.
One advantage that Adobe’s own model may have for Sora is its promise that Firefly is “commercially safe” because it’s trained on openly licensed, public domain and AdAdobe Stock content, which could alleviate some concerns about copyright infringement.
Text-to-video and image-to-video features will initially be available in beta as standalone Firefly apps later this year. Adobe said the new Firefly video model will eventually be integrated into its Creative Cloud, Experience Cloud and Adobe Express applications.
The company also showed off some additional clips for Premiere Pro’s upcoming “Generative Extend” feature, which can extend the length of existing movie clips, similar to Photoshop’s Generative Expand tool for image backgrounds. Adobe said that version will also be released “later this year” and the exact date is unknown.
