2024-02-16 06:38:20
OpenAI recently released a new video generation model Sora, which allows users to generate videos using text. According to Mashable, Sora can generate 1-minute videos that are so realistic that they look like they were shot in real life. Netizens have polarized attitudes towards Sora. Some netizens expressed interest in trying Sora, while others expressed concern that the film production industry will decline.
Sora can generate movies with multiple characters. In addition, users can also set the movements of the video characters and the details of the video background. The Sora model can not only understand the user’s Prompt requirements, but also generate videos based on real-world physics.
▲Prompt: Tour of an art gallery with many beautiful works of art in different styles
OpenAI said that Sora still has weaknesses for the time being, and sometimes it may not understand logical relationships. For example, when generating a fragment of a person eating a cookie, the cookie may not show bite marks. Sora sometimes confuses the spatial details of user prompts, such as not distinguishing between left and right, and temporarily unable to comply with the camera movements requested by user prompts.
▲Prompt: Step-printing scene of a person running, cinematic film shot in 35mm.
related articles:
[Teaching]GPT Store officially opens ChatGPT AI Girlfriend + a photo to help you think regarding recipes GPT4All can use AI without Internet access[teaching]offline installation + VPN-free practical test [Teaching]ChatGPT voice conversation can understand Cantonese + direct voice answer
Receive the latest newsUnsubscribe from updates
1708092288
#OpenAI #Sora #releasesYouyimokaisuper #powerful #textgenerated #video