Adobe has offered a preview of its next-generation AI video tools, announcing a new feature that can create video clips from still images. This latest development comes as part of a model that will support AI video and audio editing features in Adobe Creative Cloud apps, building on the Firefly video model that the software giant introduced in April.
The new trailer features footage created with Firefly’s text-to-video capabilities, which Adobe announced earlier this year but has yet to fully demonstrate. The tool lets users create video clips using text descriptions and adjust the results with various “camera controls.”
These controls simulate camera angles, movement, and shooting distance, changing the visual aesthetics of the video. Additionally, the Firefly video model demonstrated the ability to create video from footage, which allows clips to be produced using referenced footage. Adobe says this innovation can be particularly useful for creating additional B-roll footage or filling in gaps in production processes.
Adobe to limit AI video length
If the demo videos are anything to go by, the quality of the videos being created appears to be similar to what we’ve seen so far with OpenAI’s Sora model. Adobe also said it’s “looking into” the Sora model as a third-party integration for its Premiere Pro video software. However, text-to-video and image-to-video content will be limited to five seconds in length, Adobe’s Vice President of Generative AI Alexandru Costin told The Verge.
One advantage that Adobe’s Firefly model may have over Sora is its promise of commercial safety. Firefly is trained using openly licensed, public domain, and Adobe Stock content, providing a solution that could alleviate concerns about copyright infringement.
Text-to-video and image-to-video features will debut in beta as a standalone Firefly app later this year. Adobe said the new Firefly video model will eventually be integrated into Creative Cloud, Experience Cloud, and Adobe Express apps.
Adobe also shared images of the upcoming “Generative Extend” feature in Premiere Pro, which can extend the length of existing videos in a similar way to Photoshop’s background extender Generative Expand. Adobe said this new feature will also be available “later this year.”