When proccess is done i have 6 black screen images and warning in PowerShell "RuntimeWarning: invalid value encountered in cast" =( Im using your config but RealVisXL_5.0_LIGHTNING not RealVisXL_5.0 How to fix it?
Organize Your PNG Files Place all PNG files in a single directory. Ensure the files are named in a way that reflects their order (e.g., 01.png, 02.png, ...). Step 2: Load and Stack the Images Using NumPy and Pillow: python Copy code import os import numpy as np from PIL import Image # Path to the directory containing PNG files directory = "path/to/your/png/files" # Sort and load files files = sorted([os.path.join(directory, f) for f in os.listdir(directory) if f.endswith('.png')]) # Load images and stack them into a 3D NumPy array volume = np.array([np.array(Image.open(file)) for file in files]) print("Volume shape:", volume.shape)
Image to image. my BiRefNet node only has the ckpt_name line. does not have the second use_gpu line below it. so i can not turn use_gpu to false. I followed your install instructions. what part should I review to solve this problem? Thanks.
@@ComfyUIworkflows No it's solved, so this custom node broke my comfyUI, I've had this problem while installing several other custom nodes. So I looked it up online because I was tired of reinstalling comfyUI. So I had to update the torch version and that solved the issue. Thanks for video although I do get the allocation error lol
@@AgustinCaniglia1992 You have a sequence of PNG images showing every side. It is possible to convert them into a 3D model. Alternatively, there is another method using a 3D node, which can also generate a 3D model. However, the output quality is not good, and it consumes a lot of VRAM.
It's not an AI voice. While editing, I removed all the silences in the video and also the 'um' words with just one click. If I were to do it manually, it would take a lot of time, which I don't have. Without the silences, the video's voice sounds too fast and unnatural, making it less engaging."
![Shoot EPIC MOVIES with this FREE AI Tool! [ComfyUI Tutorial + Free Workflow]](http://i.ytimg.com/vi/gHI6PjTkBF4/mqdefault.jpg)
![Shoot EPIC MOVIES with this FREE AI Tool! [ComfyUI Tutorial + Free Workflow]](/img/tr.png)
It is wonderful tutorial.you are fantastic！
When proccess is done i have 6 black screen images and warning in PowerShell "RuntimeWarning: invalid value encountered in cast" =(
Im using your config but RealVisXL_5.0_LIGHTNING not RealVisXL_5.0
How to fix it?
Change Lightning model to RealVisXL5.0 and nothing changes - i still have six black screen images when done.
Thanks !
Is there an ai that Can use the 6 images created to build a 3d model?
Organize Your PNG Files
Place all PNG files in a single directory.
Ensure the files are named in a way that reflects their order (e.g., 01.png, 02.png, ...).
Step 2: Load and Stack the Images
Using NumPy and Pillow:
python
Copy code
import os
import numpy as np
from PIL import Image
# Path to the directory containing PNG files
directory = "path/to/your/png/files"
# Sort and load files
files = sorted([os.path.join(directory, f) for f in os.listdir(directory) if f.endswith('.png')])
# Load images and stack them into a 3D NumPy array
volume = np.array([np.array(Image.open(file)) for file in files])
print("Volume shape:", volume.shape)
@@ComfyUIworkflows easier explanation plz
@@Bunoin next video, with the help of that you can generate fully 3d model in 1 click
Image to image. my BiRefNet node only has the ckpt_name line. does not have the second use_gpu line below it. so i can not turn use_gpu to false. I followed your install instructions. what part should I review to solve this problem? Thanks.
download www.mediafire.com/file/rhddod5pgyk9xxe/nodes.py/file
nodes.py file and replace with your Comfy UI MV Adapter folder nodes.py file
How to turn these frames into a short gif rotatating around the character ?
Convert a sequence of images to a video. You can search on Google; there are many websites available that can do this.
@ComfyUIworkflows 6 frames is not enough for a smooth rotation though.
Thanks to this tutorial and whatever custom nodes that come from this workflow, my comfyUI is broken
what error you getting?
@@ComfyUIworkflows No it's solved, so this custom node broke my comfyUI, I've had this problem while installing several other custom nodes. So I looked it up online because I was tired of reinstalling comfyUI. So I had to update the torch version and that solved the issue. Thanks for video although I do get the allocation error lol
Excerpt the 3D part is missing lol.
my next video generate 3D Model one click,
Where is the 3D?
use the lora
@@ComfyUIworkflows what lora? Do you even know what 3D actually is? This is not 3D.
@@AgustinCaniglia1992 You have a sequence of PNG images showing every side. It is possible to convert them into a 3D model. Alternatively, there is another method using a 3D node, which can also generate a 3D model. However, the output quality is not good, and it consumes a lot of VRAM.
@@ComfyUIworkflows every side or just 6 angles? That doesnt works for 3D. The other tool You mention might deserve the title on this video
@@AgustinCaniglia1992 use Microsoft/Trellis to convert image to 3d
Why do you use such a bad AI voice?
It's not an AI voice. While editing, I removed all the silences in the video and also the 'um' words with just one click. If I were to do it manually, it would take a lot of time, which I don't have. Without the silences, the video's voice sounds too fast and unnatural, making it less engaging."
I Got This Error Any One Help ME DiffusersModelMakeup
'down_blocks.1.attentions.0.transformer_blocks.0.attn1.to_q.weight'
which node
@@ComfyUIworkflows DiffusersModelMakeup
@@ComfyUIworkflowsDiffusersModelMakeup and output also black screen