Adobe will enhance its suite of applications for creative professionals with the limited release of a new AI-generated video creation and editing tool later this year. Named Adobe Firefly Video Model, the AI tool will be released in beta form and will join the Photoshop maker's existing lineup of Firefly image-generating applications that allow users to produce still images, designs and vector graphics.
The model will give Adobe a foothold in the growing market for AI video generation tools, which has been targeted by OpenAI's Sora, Stability AI's Stable Video Diffusion and other AI video apps from smaller startups.Adobe's generative AI tool can, based on a single cue Adobe's generative AI tool generates 5-second clips from a single cue and interprets text and image cues. Users can also specify the desired camera angle, pan, motion and zoom. The video model is trained on public domain or licensed content, not on any Adobe customer content.
The model trains them only on content from the Adobe Stock database, which contains 400 million images, illustrations, and videos that have been carefully curated to contain no intellectual property, trademarks, or recognizable characters.
Adobe will also introduce the Generative Extend tool in its Premiere Pro video editing software, which extends an existing clip by up to two seconds by generating appropriate inserts to fill in gaps in the clip.