Mastering AI Animation: Use Auto-Mask, ControlNet and AnimateDiff Evolved! - Video To Video
HTML-код
- Опубликовано: 20 окт 2024
- Hello Dreamers! In this video, we explore the limitless possibilities of AnimateDiff animation mastery. It's not just about editing - it's about breaking boundaries, pushing creative limits, and turning ordinary videos into extraordinary visual journeys. Let's embark on this creative revolution together!
GPU USED: NVIDIA RTX 3080 10GB VRAM
** Links from the Video Tutorial **
ComfyUI-N-Suite: github.com/Nuk...
ComfyUI's ControlNet Auxiliary Preprocessors: github.com/Fan...
Tutorial on creating a simple yet efficent automatic person mask: • ComfyUI Tutorial - Aut...
ControlNet v1.1 Models: huggingface.co...
revAnimated Model: civitai.com/mo...
ReActor: github.com/Gou...
WORKFLOW: github.com/Nuk...
Face Used for ReActor: github.com/Nuk...
![Voices of the Future: How to Clone Any Voice with RVC and Make It Sing! - Xmas Video [1/3]](http://i.ytimg.com/vi/WQwOcNmkA9w/mqdefault.jpg)
![Voices of the Future: How to Clone Any Voice with RVC and Make It Sing! - Xmas Video [1/3]](/img/tr.png)
it says "invalid prompt: {'type': 'invalid_prompt', 'message': 'Cannot execute because a node is missing the class_type property.', 'details': "Node ID '#59'", 'extra_info': {}}" in the command window, and doesnt let me run the workflow, any help?
Love your channel, the content, and your vocal flow. Looking forward to trying this!
Wow! This is awesome. Thanks for sharing.
Great how you explain this workflow :)
which workflow did you use? didn't see it in the description
SORRY! I FORGOT TO ADD IT LOL
It's in the description now!
Amazing work. Well done
Thank you, I've learn a lot from you. 😇
Thanks for the tutorial.
Thank you so much, great video 👍
Nice video for beginners, highly suggested. Impressive result for a 10GB VRAM GPU.
However if you have more VRAM you can do a much better/complex workflow for a more fine results .
hello and thanks looks sick! I cannot understand yet why when I click queue prompt this error appears:
Error occurred when executing ADE_AnimateDiffLoaderWithContext:
module 'comfy.ops' has no attribute 'Linear'
Maybe you can help?
I have all models needed and installed all subdirectories models, facerestore etc.
EDIT:
I've uninstalled AnimateDiffEvolved and reinstalled it, it works now! Thanks
Hi, thanks for the information. I noticed that with the addition of the IPAdapter, the number of input frames is limited.
thanks for the information worked for me 👍
I noticed in your workflow nodes for the ipadapter. Why it is there but connected no where and if I wanted to add it for greater facial consistency where should I add it?
Is there cause i forgot it there xD , if you want to use it you can add it after the VAEDecoder
really cool thanks !
Hello, thanks again for another vid. I have one issue. "(IMPORT FAILED) ReActor Node for ComfyUI". I have tried un-installing the node, updating comfyui etc., then re-installing the node through manager to no avail. I've gone through the ReActor troubleshooter in the read.me but nothing. If anyone has a fix or advice short of opening an issue, please let me know. Thanks!
Usually the reason behind the "Import failed" error is in the looong log behind that!
Yes. It was just taking me a bit to figure why insight was not being found, I had tried to install the 3.11 while requiring the 3.10 file. Thank you again.
@@DennisKwasnycia 👍
That turned out very well. There's no "RIFLE" repo though, like you spelled it in the description, it's actually just RIFE with no L. RIFLE would have been a cool name, but there's nothing that starts with L in "Real-time Intermediate Flow Estimation".
You're right, also that link shouldn't be there, yesterday I was too tired and I made a lot of mistakes 😭
I've followed the tutorial but when i hit queue prompt nothing happens.... im new to comfy ui whats the issue ?
Hi, You should check if there are any errors in the console or if any nodes turn red
I had same issue has you with no errors in console. By just changing the positive prompt, the queue was working fine. My prompt dont accept chars like { or }.
Thanks for the tutorial. I'm getting "Error occurred when executing LoadVideo: chunk expects `chunks` to be greater than 0, got: 0" and I tried few different videos. Any idea?
uhm, they are mp4 videos? Batch size is set to zero?
Yes, they are mp4. Batch size was 5 since I used your flow. Now changed to zero and things are running. Thanks!@@DreamingAIChannel
@@YarivTawiliArt my flow has batch_size at zero, not 5 xD
Now seems like my Cuda is too new (12.1) for onnxruntime
@@YarivTawiliArt the portable version i'm using is working without any issues and is the official one using cuda 12.1,so you shouldn't have any problem with my workflow!
Please upload a workflow.
I forgot to add it to the description... sorry. It is there now!
Thank You
Might you know why I'm getting "RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0!"
Been using comfy for a while and this is the first time I've had this
That's weird, on what node is happening?
@@DreamingAIChannel happens at the ksampler. Im guessing it has to do with my amd card
@@DreamingAIChannel Once I turned off the lineart controlnet group it worked
Hey nice artwork! Just want to ask that are you interested in our new ai generation product? Maybe I can send you beta version to you~
Nice video, a suggestion is to make the background music less relaxing. Almost fell asleep before it ends.
Well sleep is good for your health :P
Nah relaxing music is perfect with his voice, it is so refreshing to have a calm person in the middle of too many screaming youtubers.