Instagram is working on a new artificial intelligence (AI) feature that will allow users to reimagine their videos. The AI-powered video editing tool is built on Meta’s Movie Gen AI model, an early research project that can generate videos from text prompts. On Instagram, instead of letting users generate videos from scratch, the feature will add enhancements to videos shot using cameras and featuring real humans. Users will be able to make changes to the videos such as changing outfit, background, and even their overall appearance.
The new feature was teased by Instagram Head Adam Mosseri in a Reel. In the minute-long video, he showcased some of the capabilities of the new AI feature as well as explained how it would work. Notably, the AI video editing tool is under development, and Mosseri highlighted that it could be shipped next year.
In the video, the tool could change Mosseri’s outfit, add a gold chain around his neck, and even add a hippopotamus playing in a swimming pool in the background. The tool was also shown to combine some of these effects to make the Instagram Head appear to be sitting beside a snowy mountain and in a desert landscape.
Captioning the video, Mosseri revealed that the AI video editing tool is powered by Meta’s Movie Gen AI model which was previewed in October. Notably, it is a multi-modal model that comes with four capabilities — video generation, personalised video generation, precise video editing, and audio generation. It can also combine video and audio generation to generate videos with synced audio. It is currently not available in the public domain.
“I’m super excited about Movie Gen, our early AI research model that will let you change nearly any aspect of your videos with a simple text prompt. Hoping to bring this to Instagram next year,” Mosseri wrote in the caption.
It appears that the under-development AI feature only borrows the precise video editing capability from Movie Gen. Meta described in the announcement post that the model can perform localised edits such as adding, removing, or replacing elements, and make global changes such as background or style modifications. The AI model also preserves original content and targets only the relevant pixels.