Tech

Meta’s new “Make-A-Video” AI helps turn video into text


For many months now, DALL-E stole attention for its surreal text-to-image capabilities, although it currently has a competitor, Metaof theMake a moviean AI that can create videos in addition to text prompts.
As the name implies, AI allows users to describe a scene and then create a short video that matches what they are told. Well, you won’t get some cinematic, artificial-looking videos with twisted animations, although it’s still a testament to what can be achieved with AI.
Not only text, but “Make-A-Video” can create new videos from older videos and also from an image or set of them.
In a Facebook post, Mark ZuckerbergThe CEO, Meta, describes this new AI work as “incredible progress”. He added, “It is much more difficult to create videos than photos because in addition to accurately generating each pixel, the system also has to predict how they will change over time.”
There are videos of “A teddy bear draws a portrait”, “Dancing robot in Times Square” and “Cat watching TV with remote in hand” and cat with human hand. Meanwhile, there are also some actual videos – “A close-up of canvas painting by an artist”, “A young couple walking in heavy rain” and “Drink water for the horse” – and these videos looks much more real.
Despite the “incredible progress”, the videos are shorter than 5 seconds and have no sound. Also, Meta doesn’t allow user to test model, it already provides templates for different prompts, but one can’t test it yourself, so these might be the best representations whose.

news7f

News7F: Update the world's latest breaking news online of the day, breaking news, politics, society today, international mainstream news .Updated news 24/7: Entertainment, Sports...at the World everyday world. Hot news, images, video clips that are updated quickly and reliably

Related Articles

Back to top button