this is so unfair to actual animators 😭
Dreamina Octo took my one line sentence and turned it into a finished anime short. it handled the story, the storyboard, and the edit, all in one cavas.
same character, every frame, no slop.
full breakdown:
Acabo de estrenar un nuevo cortometraje hecho con IA.
Pero esta vez el experimento era distinto:
crear las escenas en 3D con Blender y usarlas para “dirigir” la IA plano a plano.
Se llama PIGMAN y está basado en un cómic que hizo mi padre en 1984.
Os dejo el corto en el siguiente tweet 👇
🚀 Experiment #5: Building a live Maya-to-AI pipeline. AI generates textures, skin, hair & environments from ONE rendered frame while preserving full animation control. Artists aren’t replaced—they’re amplified. 🎬⚡🤖 #AI#VFX#Maya
@hecubian_devil Yeah well said... This is Pandora's box opened. It's a machine designed to learn and get better. To think that you're going to put a cap on it or stop its progression it's futile. You're absolutely right. Is it good? Probably not... But trying to stop it? You can't...
@flurryfrenzy@RustlesJimothy@tenebr_ai Isn't he the customer in this scenario? Because you're charging him money... But either way here's another one for you ❤️ you could just fix the eyelashes
Force doesn’t stop at one object.
It transfers, it spreads, it affects everything around it.
That’s the difference between something looking animated…
and something feeling real.
How often are you thinking about force when you animate?
#animation#physics#animate
@andrewpprice What I find funny about this is that I've seen the haters bash on the images he provided before uploading this talk. And now that I've seen the talk they were bashing on the 3D render 🤣 because they couldn't tell which one was the 3D render and which one was AI generated
@BryanNHolt Yeah I agree... So many people are hating on the new technology because they don't want to accept that they have to change. That the reality is even if AI can replace even 1 shot that's enough to incorporate it into the pipeline and save on labor/render times/simulations
@andrewpprice Depth map huh? Interesting... And you could ask AI to generate single images of those objects to feed into meshy then all that you'd have to do from that point is fix the typology with quad draw... Very interesting...
@elonmusk so with the recent uproar with Americans and Japanese people. When are we going to throw a monster truck rally with burgers and hot dogs and fireworks over there? We can even play American songs like Dracula by Rob zombie.