How OpenAI's Sora Model Works • Hacker News Board

Re: How OpenAI's Sora Model Works

#21

Post by andy99 » Wed, Mar 20, 2024, 4:02 PM UTC

altruios wrote: post #18

andy99 wrote: post #17

Earlier quoted context omitted.

This is only a made-up issue for a few that are looking for something to criticize. Almost nobody cares, in the sense that appears to be meant here about "ownership" of the training data. Any this unfortunately hampers research and understanding of models because companies are reluctant to talk about training lest the trolls start jumping on. We're all worse off because of this.

What a short-sighted view. Having open access to the training data is how you prevent poisoning/biasing of the dataset. People complaining about bad data in the dataset improve the quality of the dataset. That's in addition to the benefit of creators being labeled in the dataset. Hiding the data from public view seems to only helps nefarious actors.

Pretty sure we're saying the same thing

Re: How OpenAI's Sora Model Works

#22

Post by bparsons » Wed, Mar 20, 2024, 4:04 PM UTC

andy99 wrote: post #17

ivraatiems wrote: post #11

A great write up. The elephant in the room, of course, is "where did Sora's dataset come from?"

This is only a made-up issue for a few that are looking for something to criticize. Almost nobody cares, in the sense that appears to be meant here about "ownership" of the training data. Any this unfortunately hampers research and understanding of models because companies are reluctant to talk about training lest the trolls start jumping on. We're all worse off because of this.

[flagged]

Re: How OpenAI's Sora Model Works

#23

Post by euW3EeBe » Wed, Mar 20, 2024, 4:26 PM UTC

stale2002 wrote: post #8

lvl102 wrote: post #3

What’s the status on companies building AI models to build actual 3D backend behind these generative videos. Anyone working on something similar? Imagine that’d be far more productive. For example, lookdev mlop is pretty low hanging fruit. Not sure why we don’t already have models from Autodesk, Epic or even Adobe (with resources ie A100/H100) where you upload an image/video and the model spits out workable 3D scaffo…

This is a good question, and the answer is that from a tech side it is surprisingly easier to solve the problem in the reverse direction. As in, making workable 3d models is harder than making video. And it is easier to make a 3d model by generating a video of the object instead. Why is that? I don't know. But that's the current state of the industry. 3D model generation is simply harder.

Probably also helps that there's way more image/video data to train on than 3D data.

Re: How OpenAI's Sora Model Works

#24

Post by altruios » Wed, Mar 20, 2024, 5:38 PM UTC

andy99 wrote: post #21

altruios wrote: post #18

Earlier quoted context omitted.

What a short-sighted view. Having open access to the training data is how you prevent poisoning/biasing of the dataset. People complaining about bad data in the dataset improve the quality of the dataset. That's in addition to the benefit of creators being labeled in the dataset. Hiding the data from public view seems to only helps nefarious actors.

Pretty sure we're saying the same thing

> Any this unfortunately hampers research and understanding of models because companies are reluctant to talk about training lest the trolls start jumping on

Respecting artists and being open about training data should go hand in hand. That companies feel the need to hide the training data from public scrutiny should immediately be suspect.

It seems like you are saying no one cares about copyright, I inform you that is not the case. I disagree with (most current forms of) copyright, but I do respect artists and their need to feed themselves. Proper attribution, and labeling and scrutiny of the dataset is imperative.

>This is only a made-up issue for a few that are looking for something to criticize. Almost nobody cares, in the sense that appears to be meant here about "ownership" of the training data

So it's not just 'trolls' that want the data to be open and labeled, is my point. If the companies are hurting artists ('s economic output), that should be examined and fixed (stopped and reattuned attention of said companies).

'trolling (bothering)' a company to be 'good (not against human interests)' isn't a bad thing.

Re: How OpenAI's Sora Model Works

#25

Post by Legend2440 » Wed, Mar 20, 2024, 6:48 PM UTC

huytersd wrote: post #4

lvl102 wrote: post #3

What’s the status on companies building AI models to build actual 3D backend behind these generative videos. Anyone working on something similar? Imagine that’d be far more productive. For example, lookdev mlop is pretty low hanging fruit. Not sure why we don’t already have models from Autodesk, Epic or even Adobe (with resources ie A100/H100) where you upload an image/video and the model spits out workable 3D scaffo…

If I’m not mistaken, Stability just released something like that a few days ago.

Yes, but it works by generating a video first and doing photogrammetry on it to produce a 3D model.

Re: How OpenAI's Sora Model Works

#26

Post by ivraatiems » Wed, Mar 20, 2024, 9:03 PM UTC

andy99 wrote: post #17

ivraatiems wrote: post #11

A great write up. The elephant in the room, of course, is "where did Sora's dataset come from?"

This is only a made-up issue for a few that are looking for something to criticize. Almost nobody cares, in the sense that appears to be meant here about "ownership" of the training data. Any this unfortunately hampers research and understanding of models because companies are reluctant to talk about training lest the trolls start jumping on. We're all worse off because of this.

Is it really so hard to imagine that someone might not like the idea of their work being used to train a machine to imitate their work with no compensation to them? And that machine is then instead used to benefit the shareholders of large corporations?

The position that this is a made-up issue when there are multiple large pending lawsuits about exactly this thing is pretty bizarre.

Re: How OpenAI's Sora Model Works

#27

Post by huytersd » Thu, Mar 21, 2024, 2:06 AM UTC

lvl102 wrote: post #9

huytersd wrote: post #4

Earlier quoted context omitted.

If I’m not mistaken, Stability just released something like that a few days ago.

Looks like I completely overlooked threestudio released last year. Thank you for pointing it out.

It’s something I’ve been interested in too. I do a bunch of CNC woodworking so would love the ability to atleast generate close enough 3D models I can then refine.

Re: How OpenAI's Sora Model Works

#28

Post by nunez » Thu, Mar 21, 2024, 2:43 AM UTC

> Total Nvidia H100 needed to support the creator community on TikTok & YouTube: 10.7M / 120 ≈ 89k

If the H100 is $40k worst case, that's one-time cost of $356M! I could definitely see the FAANGs throwing money at this.

Re: How OpenAI's Sora Model Works

#29

Post by htrp » Thu, Mar 21, 2024, 4:17 AM UTC

lvl102 wrote: post #3

What’s the status on companies building AI models to build actual 3D backend behind these generative videos. Anyone working on something similar? Imagine that’d be far more productive. For example, lookdev mlop is pretty low hanging fruit. Not sure why we don’t already have models from Autodesk, Epic or even Adobe (with resources ie A100/H100) where you upload an image/video and the model spits out workable 3D scaffo…

Nerfs and splatting