Skip to content
Hallway
0:00 / 0:00

#stablediffusion#controlnet#dreambooth

Did We Just Change Animation Forever?

Corridor Crew

7.1M subscribers • 2 videos on Hallway

YouTube
  • Recommended by 1 person: 1 on Hacker News.
  • May be out of date. The tools, versions or practices it shows have likely moved on.

Intermediate. Assumes working knowledge.

3.4M views3 years ago1 recommendation

AI and machine learningGraphics and audioNeural networks and deep learningRendering

The Corridor Crew team explains how they used custom-trained Stable Diffusion models and ControlNet to produce a temporally consistent AI-animated short film.

ANYONE can make a cartoon with this groundbreaking technique. Want to learn how? We made a ONE-HOUR, CLICK-BY-CLICK TUTORIAL on corridordigital.com/ Watch the full ROCK PAPER SCISSORS anime on Corridor ► youtu.be/GVT3WUa-48Y This project has been a huge labor of love, and it is due to the amazing open-source community that we have this technology available to us. We hope that by sharing our discoveries and techniques that we can help push this technology forward for everyone. If you want to dip your toes into this tech, there are many amazing online communities ready to help teach you, including ours! With Your Support, We Can Make More! ► This project exists because of the amazing members of CorridorDigital, our INDEPENDENT STREAMING PLATFORM. All memberships begin with a 14-Day Free Trial and you can Cancel Anytime. Consider becoming a member yourself! corridordigital.com/learn-more Anime Rock, Paper, Scissors Shirt?! ► Available only until March 6th, we have a limited-edition Anime Rock Paper Scissors t-shirt and longsleeve design to celebrate this release. Check it out here! corridordigital.store/ Design by Kendrra Thoms kendrrathoms.com/ Gear Used ► Puget Systems Computers: bit.ly/PC_Puget_Workstations Lighting by Aputure: bit.ly/CORRIDOR_LIGHTS

Why it's here

1 person recommends this

What people say. Recommended as a groundbreaking demonstration of using Stable Diffusion and ControlNet to achieve temporal consistency in AI animation.

samspencon Hacker News • Recommends

On “Introducing ChatGPT and Whisper APIs”

I thought Midjourney was better as well, until I saw some recent videos from Corridor Crew on Youtube. For those who don't know, this is a VFX studio in LA that tries to keep at the cutting-edge of video production techniques and posts content to their Youtube channel, and they have a massive number of followers and several viral videos. They recently created a full 7-minute anime using Stable Diffusion with their own models and their existing video production gear, I'll post the links and let the results speak for themselves The actual 7-minute anime piece produced using SD: link Behind the scenes: "Did we change anime forever?" link to this video "VFX reveal before and after" link
View on Hacker News

sschuelleron Hacker News • Mentions

On “Diffusion with Offset Noise: Finetuning SD to generate very dark or light images”

The whole thing with SD is extremely interesting and difficult to keep up with as there are so many new things coming in almost daily. Just last week I saw ControlNet1 which ads a lot more control. Today I saw what corridor crew2 did to stabilize the randomness when you want to make videos. Very exciting. 1 link 2 link to this video
View on Hacker News

Kelamiron Hacker News • Mentions

On “Anime rock, paper, scissors [video]”

The characters here are processed with Stable Diffusion. This is an impressive progress in the field of video generation with help of AI, considering how consistent the generated characters look like across frames, in my opinion. Also here is another video link to this video "Did We Just Change Animation Forever?" where they demonstrated how it came to be.
View on Hacker News

yummypainton Hacker News • Mentions

On “Wonder Studio: this AI-powered tool might be a preview of the future of VFX”

This reminds me alot of this techniquefor turning live action into animation: link to this video It unsurprisingly received alot of pushback from people in the industry but I think it's really impressive.
View on Hacker News

dbspinon Hacker News • Mentions

On “If AI scaling is to be shut down, let it be for a coherent reason”

The issues you mention with Stable Diffusion are teething issues. Here's an example of a current technique to get temporally stable animations with Stable Diffusion - link Corridor digital show off a similar technique here - link to this video GPT4 with its new plugin architecture is rapidly becoming more capable. I think you're absolutely correct in the tendency of all LLMs to hallucinate is a major issue that we don't know how to correct even in principle. But there are techniques to moderate its impact on responses - triangulating using multiple models, citing sources etc. You could build an assistant like say this, today - link to make your appointments and answer the phone today, using technol…
View on Hacker News

somepersonon Hacker News • Mentions

On “Sora: Creating video from text”

> There was a video of some guys doing this (sorry can’t find it) to make an anime type animation. With consistent characters. They used videos of themselves running through their own models to make the characters. That was Corridor Crew: link to this video
View on Hacker News