271–280 of 342 posts
Earlier quoted context omitted.
This tech is cute but the only viable outcomes are going to be porn and mass produced slop that'll be uninteresting before it's even created. Why even bother?
Comments like this one are so predictable and incredulous. As if the current state of the art is the final form of this technology. This is just getting started. Big facepalm.
I got access to the preview, here's what it gave me for "A pelican riding a bicycle along a coastal path overlooking a harbor" - this video has all four versions shown: https://static.simonwillison.net/static/2024/pelicans-on-bic... Of the four two were a pelican riding a bicycle. One was a pelican just running along the road, one was a pelican perched on a stationary bicycle, and one had the pelican wearing a weird…
There's another important contender in the space: Hunyuan model from Tencent My company (Nim) is hosting Hunyuan model, so here's a quick test (first attempt) at "pelican riding a bycicle" via Hunyuan on Nim: https://nim.video/explore/OGs4EM3MIpW8 I think it's as good, if not better than Sora / Veo
Earlier quoted context omitted.
That is reality, that is nature. The natural world is filled with camouflaged animals and plants that prey on one another via their camouflage. This is evolution, and those unable to discriminate reality from fiction will be the causalities, as they always have since the dawn of life.
The naturalistic fallacy is weak at best, but this is one of the weirdest deployments of it I've encountered. It's not evolution, it's nothing like it. If it's kill or be killed, we should do away with medicine right? Only the strong survive. Why are we saving the weak? Sorry but this argument is beyond silly
Earlier quoted context omitted.
“Won” what exactly? I have no issues running stable diffusion locally. Since Llama3.3 came out it is my first stop for coding questions, and I’m only using closed models when llama3.3 has trouble. I think it’s fairly clear that between open weights and LLMs plateauing, the game will be who can build what on top of largely equivalent base models.
The quality for SD is no where near the clear leaders.
Earlier quoted context omitted.
There's another important contender in the space: Hunyuan model from Tencent My company (Nim) is hosting Hunyuan model, so here's a quick test (first attempt) at "pelican riding a bycicle" via Hunyuan on Nim: https://nim.video/explore/OGs4EM3MIpW8 I think it's as good, if not better than Sora / Veo
Hard to say about SORA but the video you shared is most definitely worse than Veo. The Pelican is doing some weird flying motion, motion blur is hiding a lack of detail, cycle is moving fast so background is blurred etc. I would even say SORA is better because I like the slow-motion and detail but it did do something very non physical. Veo is clearly the best in this example. It has high detail but also feels the mos…
Superficially impressive but what is the actual use case of the present state of the art? It makes 10-second demos, fine. But can a producer get a second shot of the same scene and the same characters, with visual continuity? Or a third, etc? In other words, can it be used to create a coherent movie --even a 60-second commercial -- with multiple shots having continuity of faces, backgrounds, and lighting? This quote…
Earlier quoted context omitted.
Does everyone have "legal" access to YouTube. In theory that should matter to something like Open(Closed)Ai. But who knows.
I mean, I have trained myself on Youtube. Why can't a silicon being train itself on Youtube as well?
Earlier quoted context omitted.
I mean, I have trained myself on Youtube. Why can't a silicon being train itself on Youtube as well?
Because silicon is a robot. A camcorder can't catch a flick with me in the theater even if I dress it up like a muppet.
Like, a blind person with vision restored by silicon eyes?
Do I not have rights to run whatever firmware I want on those eyes, because it's part of my body?
Okay, so what if that firmware could hypothetically save and train AI models?
Earlier quoted context omitted.
I mean, I have trained myself on Youtube. Why can't a silicon being train itself on Youtube as well?
The difference is that you didn't need to scrape millions of videos from YouTube with residential proxy network scrapers to train yourself.
If it just reduces to an issue of data efficiency, AI research will eventually get there though.
Earlier quoted context omitted.
Comments like this one are so predictable and incredulous. As if the current state of the art is the final form of this technology. This is just getting started. Big facepalm.
The most predictable comment is yours, especially since you completely missed the point of the original comment which had nothing to do with the video quality.