Make-A-Video is an AI-powered video generation system that enables users to create high-quality videos from text prompts. It works by combining the recent advances in text-to-image generation with the ability to learn from unlabeled videos, allowing users to create custom videos with just a few words or lines of text.
Make-A-Video also provides users with the ability to create motion from a single image or fill-in the in-between motion of two images. Furthermore, the system provides users with the ability to create variations of their original video.
Make-A-Video has been tested and found to be 3x better at representing text input and 3x higher in quality when compared to the previous state of the art.
Meta AI is committed to developing responsible AI and ensuring the safe use of this state-of-the-art video technology, and is taking steps to reduce the potential for harmful content to be generated.
Make-A-Video is currently undergoing internal testing and development, but it is expected to be released to the public in the near future.