Something odd happens with that Tokyo woman's legs. First she skips a couple of times, then her feet change places.
Sora: Creating video from text
601–610 of 1001 posts
Re: Sora: Creating video from text
#602In the last few days I've been asking myself what would drive the next big leap in advertising efficiency after big data and conversion pixels. I think I have my answer now. This is going to disrupt the ad agency side of the business big time.
Re: Sora: Creating video from text
#603In less than a few hours Gemini 1.5 is old news. Sam is doing live demos on Twitter while Google just released a blog. Didn't think Google would be the first of the Facebook, Apple, Google and Microsoft to get disrupted.
This is a really silly take isn't it?
Re: Sora: Creating video from text
#604How far are we from just giving it a novel and effectively asking it to create a TV series from it
Re: Sora: Creating video from text
#605https://openai.com/sora?video=big-sur In this video, there's extremely consistent geometry as the camera moves, but the texture of the trees/shrubs on the top of the cliff on the left seems to remain very flat, reminiscent of low-poly geometry in games. I wonder if this is an artifact of the way videos are generated. Is the model separating scene geometry from camera? Maybe some sort of video-NeRF or Gaussian Splatti…
The model is essentially doing nothing but dreaming. I suspect that anything that looks like familiar 3D-rendering limitations is probably a result of the training dataset simply containing a lot of actual 3D-rendered content. We can't tell a model to dream everything except extra fingers, false perspective, and 3D-rendering compromises.
Technically we can, that's what negative prompting[1] is about. For whatever reason, OpenAI has never exposed this capability in its image models, so it remains an open source exclusive.
[1] https://stable-diffusion-art.com/how-to-use-negative-prompts...
Re: Sora: Creating video from text
#606Does anyone know how to handle the depression/doom one feels with these updates? Yes, it's a great technical achievement, but I just worry for the future. We don't have good social safety nets, and we aren't close to UBI. It's difficult for me to see that happen unless something drastic changes. I'm also afraid of one company just having so much power. How does anyone compete?
imagine being depressed at technological advances
we will finally free ourselves from mediocre humans being the bottleneck for everything
If you want a safety net move to Scandinavia
Re: Sora: Creating video from text
#607https://openai.com/sora?video=big-sur In this video, there's extremely consistent geometry as the camera moves, but the texture of the trees/shrubs on the top of the cliff on the left seems to remain very flat, reminiscent of low-poly geometry in games. I wonder if this is an artifact of the way videos are generated. Is the model separating scene geometry from camera? Maybe some sort of video-NeRF or Gaussian Splatti…
It's possible it was pre-trained on 3D renderings first, because it's easy to get almost infinite synthetic data that way, and after that they continued the training on real videos.
Re: Sora: Creating video from text
#608What real life problem does this solve?
Re: Sora: Creating video from text
#609I 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…
> very good... understanding of the physics of objects and relationships between them I am always torn here. A real physics engine has a better "understanding" but I suspect that word applies to neither Sora nor a physics engine: https://www.wikipedia.org/wiki/Chinese_room An understanding of physics would entail asking this generative network to invert gravity, change the density or energy output of something, or at…
Question is - how much do you need to understand something in order to mimick it?
The Chinese Room seems to however point to some sort of prewritten if-else type of algorithm type of situation. E.g. someone following scripted algorithmic procedures might not understand the content, but obviously this simplification is not the case with LLMs or this video generation, as the algorithmic scripting requires pre-written scripts.
Chinese Room seems to more refer to cases like "if someone tells me "xyz", then respond with "abc" - of course then you don't understand what xyz or abc mean, but it's not referring to neural networks training on ton of material to build this model representation of things.
Re: Sora: Creating video from text
#610What real life problem does this solve?
the same problem movies and tv solve, for one. entertainment?