Live data from Hacker News

Sora: Creating video from text

openai.com

841–850 of 1001 posts

Re: Sora: Creating video from text

#844
What’s the connection between this and high end game engines (like unreal 5). I would expect 3d game engines to be used at least for training data and fine tuning. But perhaps also directly in the generation of the resulting videos?

For example this looks very much like something from a modern 3d engine:

https://twitter.com/OpenAI/status/1758192957386342435

Re: Sora: Creating video from text

#845

I think the implications go much further than just the image/video considerations. This model shows a very good (albeit not perfect) understanding of the physics of objects and relationships between them. The announcement mentions this several times. The OpenAI blog post lists "Archeologists discover a generic plastic chair in the desert, excavating and dusting it with great care." as one of the "failed" cases. But t…

Wouldn't having a good understanding of physics mean you know that a women doesn't slide down the road when she walks? Wouldn't it know that a woolly mammoth doesn't emit profuse amounts steam when walking on frozen snow? Wouldn't the model know that legs are solid objects in which other object cannot pass through? Maybe I'm missing the big picture here, but the above and all the weird spatial errors, like miniaturiz…

I agree, to me the most clear example is how the rocks in the sea vanish/transform after the wave: The generated frames are hyperreal for sure, but the represented space looks as consistent as a dream.

Re: Sora: Creating video from text

#847

I question how much anyone has really used these models if they actually think these systems can replace people. I’ve consistently failed to get professional results out of these things and the degree of work required to get professional results makes me think a new class of job will be created to get professional results out of these systems. That being said, there is value in these systems for casual use. For examp…

Many financiers are willing to trade quality with cost reduction.

Re: Sora: Creating video from text

#848

I think the implications go much further than just the image/video considerations. This model shows a very good (albeit not perfect) understanding of the physics of objects and relationships between them. The announcement mentions this several times. The OpenAI blog post lists "Archeologists discover a generic plastic chair in the desert, excavating and dusting it with great care." as one of the "failed" cases. But t…

Wouldn't having a good understanding of physics mean you know that a women doesn't slide down the road when she walks? Wouldn't it know that a woolly mammoth doesn't emit profuse amounts steam when walking on frozen snow? Wouldn't the model know that legs are solid objects in which other object cannot pass through? Maybe I'm missing the big picture here, but the above and all the weird spatial errors, like miniaturiz…

> Wouldn't having a good understanding of physics mean you know that a women doesn't slide down the road when she walks? Wouldn't it know that a woolly mammoth doesn't emit profuse amounts steam when walking on frozen snow? Wouldn't the model know that legs are solid objects in which other object cannot pass through?

This just hit me but humans do not have a good understanding of physics; or maybe most of humans have no understanding of physics. We just observe and recognize whether it's familiar or not.

AI will need to be, that being the case, way more powerful than a human mind. Maybe orders of magnitude more "neural networks" than a human brain has.

Re: Sora: Creating video from text

#849
This is the harbinger that announces that, as a technologist, the time has come for me to witness more and more things that I cannot understand how they work any more. The cycle has closed and I have now become my father.

Re: Sora: Creating video from text

#850

This is insane. But I'm impressed most of all by the quality of motion . I've quite simply never seen convincing computer-generated motion before . Just look at the way the wooly mammoths connect with the ground, and their lumbering mass feels real. Motion-capture works fine because that's real motion, but every time people try to animate humans and animals, even in big-budget CGI movies, it's always ultimately obvio…

Nah this still has the problem with connecting surfaces that never seems to look right in any CGI. It's actually interesting that it doesn't look right here as well considering they are completely different techniques.
Post reply on HN