Facebook AI is Capable of Creating Videos From Text

AI does not stop showing us wonderful things. Meta, a company that includes networks such as Facebook, Instagram and WhatsApp, has just presented a new platform capable of creating videos through text. A twist on tools like Midjourney or DALL-E 2, designed to generate images from a brief explanation.

Meta’s tool, called Make-A-Video, represents “pretty amazing progress,” according to Meta Meta’s CEO, Mark Zuckerberg. “It is much more difficult to generate videos than photos because beyond correctly generating each pixel, the system also has to predict how they will change over time,” says the executive.

Generative AI research pushes creative expression forward by giving people tools to quickly and easily create new content. With just a few words or lines of text, Make-A-Video can bring imagination to life and create unique videos full of vivid colours, characters, and landscapes. The system can also create videos from images or take existing videos and create similar new ones.

“Make-A-Video solves this by adding a layer of unsupervised learning that allows the system to understand movement in the physical world and apply it to traditional text-to-image generation,” concludes the father of Facebook.

The technology company has shared a video where you can see different content created by its brand-new AI. Among them are recordings in which a teddy bear appears painting a painting, a dog flying through the sky dressed as a superhero or a flying saucer landing on the ground of Mars.

As with AIs designed to generate images from text, the goal is for the tool to become something useful for artists and designers.

At the moment, Make-A-Video is not available to the majority of users. Meta has made available to those interested an address from which to make a request to receive a notification when it is known to test it.

The technology company has presented its new Make-A-Video tool, which is currently unavailable to users.