- US software maker Adobe said it is allowing third-party-generated artificial intelligence tools to enter the video editing software side of the business.
- Adobe also stated that third-party generative AI plugins for third-party programs like OpenAI, Lika Labs, and Runway will continue to be used within Premiere Pro.
- Object addition, object removal, generative extension, and text-to-video will come to Premiere Pro this year.
Adobe has announced plans to add generative AI capabilities to Premiere Pro later this year. The new tools will be based on Adobe’s Firefly AI model and partnerships with some popular AI companies such as OpenAI, Pika Labs, and Runway.
Adobe’s Firefly AI model
The new Firefly video features will include the ability to add or remove objects from a video clip using natural language cues. Users will also be able to extend the length of video clips by prompting the AI model to generate additional semantically consistent content.
All videos generated by these tools will be automatically tagged with Adobe’s Content Credentials, providing transparency into which parts were generated by AI.
Also read: OpenAI voice-clone tool mimics your voice with 15-second sample
Three Firefly AI model features
Object addition, object removal, generative extension, and text-to-video will come to Premiere Pro this year.
Object addition and removal: simply select and track the objects, then replace them. This can be used to change the colour of clothes or even add previously non-existent images to a video clip. Also, for some instances when things in the background detract from a clip, but this tool will allow people to remove distracting objects.
Generative extend: seamlessly add frames to make clips longer, so it’s easier to perfectly adjust edit times and add smooth transitions.
Text to video: generate entirely new footage directly within Premiere Pro. Simply type text into a prompt or upload reference images.
Cooperation with other AI companies
Apart from the Firefly AI model, Adobe also said Monday that it is developing a way for users to generate and use video in Premiere Pro using third-party tools from OpenAI and startups Runway and Pika Labs.
OpenAI has demonstrated its Sora model, which generates realistic videos based on textual cues but has yet to make the technology public or give a timeline for when it will be available.
Deepa Subramaniam, Adobe’s vice president of product marketing for creative professional apps, said, “We’re excited to do is explore a world where you can have more choice beyond that through third-party models.”
