A neural network has been developed in the United States that can create videos based on text descriptions

Photo: © Video screenshot

Read ren.tv in

Virtual reality in television and advertising will become indistinguishable from real filming. The developers assure that the new American neural network is capable of this.

A neural network has been developed in the United States that can create videos based on text descriptions

The program, based on artificial intelligence, can create videos up to a minute long based on text descriptions. Its main difference from its predecessors is the highest image quality. These frames are indistinguishable from real shooting, although they are all generated by a neural network. To do this, it was enough to just type, for example, “a stylish woman walks along the street of Tokyo.”

For now, only some artists and designers are allowed to test the system. But in the future it will be made publicly available.

Mishustin announced equal opportunities for domestic neural networks with ChatGPT

Source: Ren TV

Facebook
Pinterest
LinkedIn
Twitter
Email

Leave a Reply

Your email address will not be published. Required fields are marked *