The Artificial Intelligence that generates an image from a text broke into the Internet in an overwhelming way so far in 2022. Sites like Midjourney either DALLE-2 They stand out as the most popular in this new field of technology.
However, the Artificial intelligence, which is nothing more than the automatic learning of a digital mechanism, is not only there to generate photographic or illustrative content. There is a program that creates a video from a text command.
Is named make-a-video and begins to gain popularity thanks to the onslaught of programs that create images with this method.
The most outstanding thing about Make-A-Video is that it belongs to Meta, a company run by Mark Zuckerberg and which brings together popular social networks such as Facebook, Instagram and the WhatsApp messaging service.
According to what colleagues at XatakaMake-An-Video is born from the Metaverse laboratories that Meta engineers want to develop.
We’re pleased to introduce Make-A-Video, our latest in #GenerativeAI research! With just a few words, this state-of-the-art AI system generates high-quality videos from text prompts.
Have an idea you want to see? Reply w/ your prompt using #MetaAI and we’ll share more results. pic.twitter.com/q8zjiwLBjb
— Meta AI (@MetaAI) September 29, 2022
Make-A-Video: Artificial Intelligence for videos
“It is much more difficult to generate videos than photos because, in addition to correctly generating each pixel, the system also has to predict how they will change over time,” said Mark Zuckerberg in an interview echoed by the aforementioned portal.
According to a Meta document, the Make-A-Video Artificial Intelligence collected data to strengthen its machine learning from two specific sites.
The first is WebVid-10M, from which he took 52,000 hours of videos, and the second HD-VILA-100M, from which he took 3.3 million audiovisual materials.
The site is not yet open to all users. But Meta, through this link that we share with youmotivates you to fill out this form and when everything is ready you will receive an email with access to start creating your own videos from a text.
Meta moves token: videos generated from text
Mark presented yesterday on Facebook Make-A-Video, an AI capable of generating videos based on a text request (Prompt).Draw like Dall-e and predict the movement of the images to generate “the movie”.
Brutal. pic.twitter.com/sfVEE3msyp
– Eduardo Tornos (@etornosinza) September 30, 2022