The OpenAI company today (Friday) unveiled "Sora", a groundbreaking artificial intelligence model that creates HD quality videos up to a minute long, based on simple text descriptions.
"Sura is able to produce complex scenes with multiple characters, specific movements and precise details of the subject and background," the company's website says.
"The model understands not only what the user is asking for in the description, but also how these things exist in the physical world."
Introducing Sora, our text-to-video model.
Sora can create videos of up to 60 seconds featuring highly detailed scenes, complex camera motion, and multiple characters with vibrant emotions.
https://t.co/7j2JN27M3W
Prompt: "Beautiful, snowy… pic.twitter.com/ruTEWn87vf
— OpenAI (@OpenAI) February 15, 2024
Among the videos presented on the OpenAI website, one stands out in which a young couple walks through the streets of snowy Tokyo, while cherry leaves and snowflakes float around them.
Another video shows woolly mammoths that look completely real, walking across a snowy field against a snowy mountain range.
In another video, you can see huge mammoths moving slowly through a snowy field, and even a segment of waves crashing on the cliffs from a drone's point of view.
This is how it looks:
View this post on Instagram
A post shared by Inon Ben Shushan (@inon_ben_shushan)
More examples:
Announcing Sora — our model which creates minute-long videos from a text prompt: https://t.co/SZ3OxPnxwz pic.twitter.com/0kzXTqK9bG
— Greg Brockman (@gdb) February 15, 2024
Prompt: "Animated scene features a close-up of a short fluffy monster kneeling beside a melting red candle. The art style is 3d and realistic, with a focus on lighting and texture. The mood of the painting is one of wonder and curiosity, as the monster gazes at the flame with… pic.twitter.com/aLMgJPI0y6
— OpenAI (@OpenAI) February 15, 2024
What impressed me the most about the OpenAI Sora videos is how close the physics is to reality: the way the waves break on the shore, the way the rocks emerge from the water.
The model doesn't know the fluid equations, yet it's almost impossible to tell that this is a simulation.
pic.twitter.com/6mEcUjucSD
— Luis Batalha ????
(@luismbat) February 16, 2024
Prompt: "A movie trailer featuring the adventures of the 30 year old space man wearing a red wool knitted motorcycle helmet, blue sky, salt desert, cinematic style, shot on 35mm film, vivid colors."
pic.twitter.com/0JzpwPUGPB
— OpenAI (@OpenAI) February 15, 2024
OpenAI, yazılan metinden detailed and realistic 1-minute videos üreten yapay zekama modeli 'Sora'yı düştı.
pic.twitter.com/wbxLAVERqP
— Teknoloji Elçilire (@teknoelci) February 16, 2024
The model can also be used to create a video from a static image, fill gaps in existing videos or extend them.
However, it must be admitted, some of the products show signs of being created using artificial intelligence, but all in all the results are quite impressive.
Despite the exciting potential, Sora will not be immediately accessible to the general public.
Instead, the company will make it available to a limited group of researchers and academics who will assess its potential risks and potential misuse.
Earlier this month, OpenAI announced the addition of watermarks to its text-to-image tool, DALL-E 3, but noted that they can be "easily removed."
Like its other AI products, OpenAI will have to deal with the consequences of fake photorealistic videos being mistaken for the real thing.
More on the same topic:
artificial intelligence