ComfyUI Animatediff

#stablediffusion Animation in #comfyui
based on input video, using #animatediff #controlnet #ipadapter #reactor
We also see the advantage of using Templates inside ComfyUI
follow me @ / pixeleasel
workflow:
drive.google.com/file/d/1R_e5...

[@WiLDeveD]

very useful & informative video. Thanks. Please make more...

[@PixelEasel]

thx nice to know! will do more

[@francaleu7777]

thank you , very nice..

[@PixelEasel]

nice to know! thanks

[@drosselmir]

Just learned about the ReActor node, thank you! What if my source video is speaking, do you know of any setups that help with lip-sync?

[@PixelEasel]

you probably need to use controlnet to preserve the original lips movement

[@mick7727]

Curious, why use model sampling with lcm instead of lcm lora? Oh hang on, is lcm lora underneath the animatediff one?

[@PixelEasel]

indid there is lcm lora

[@Vanface-hn7tn]

Thank you for the tut. About Node IPAdapter Apply: With the changes mentioned by the author, this node should bebreplaced by IPAdapter Advanced? In affirmative case, what about "noise" field in IPAdaptet Apply? Thank you.

[@PixelEasel]

you can find an updated workflow in the description for using the noise , in the new ip, u should use image noise

[@ramn_]

thanks you for the video. How much vram we need to do this tutorial? I have gtx 1060 6gb

[@PixelEasel]

I'm not sure... 🤔 but you can try it