The technological giant changed its name from Facebook to Meta and betting on artificial intelligence as a standard for the future, settling, above all, on the development of the metaverse.
Generating videos from text
Now, Meta has recently introduced a new type of artistic AI that can create videos from text prompts, with impressive results that do not go unnoticed for their originality and awe.
The artificial intelligence system called Make-A-Video uses existing images with captions to learn about the world and how it is described, and uses these untagged video clips to determine how the world is moving.
The tool combines static photos and gives them movement; that is, it uses image diffusion and identifies how realistic images work. Finally it generates 16 frames at a resolution of 64 x 64 pixels and another artificial intelligence rescales it to a resolution of 768 x 768 pixels.
What this AI does can only be described as amazing. Make-A-Video also allows you to transform still images and other videos into variants or extensions of them. The results are equally disturbing.
If you are wondering if you can already use this Meta tool, the quick answer is no, it is not publicly available (at the moment), but although the official release date of this tool is unknown, we are sure that we will be able to test it in the future. and be amazed (or scared) with the results when creating short videos.
If you want to be aware of everything, you can register on this list to find out about all the news of this AI.
Reference: META