i was trying to figure this out all day, works fine standalone. I dont know anything about python but i guess i ended up with an insightface folder in that venv folder and everything works now. Thanks.
Little friendly note, the thumbnail looks like saying "From video to image", but i think i understand you mean the opposite, This looks like a great tutorial that ill be checking out when the time comes! Keep it up! ✨
hello...thanks for video, please help me, show this error "PyTorch is not linked with support for mps devices" when execute queue.. over windows, with gforce 4070 graphics card
Thanks, But with ComfyUI installed with Pinokio we do not have the python_embeded directory, where do we paste the insigthface file? Can we better install insightface from the comfyui manager?
venv\scripts\Activate - Path cannot be found. There is no "venv" folder or file anywhere in my ComfyUI installation 😐 Now using LivePortrait with Pinokio, it's unofficial yet but installed fine and works like a charm.
Looking at their Github page, it seems to work on things that’s not photorealistic … like girls with pearl earring painting so I wonder how well it performs on character generated by sd. If it does work, video frames can be further used as training data for an original character LORA to help with consistency issue. Not what it’s intended for (paper from Kuaishou, dev of Kling) but interesting to try I think 🤔
Hi there, I'm trying to install it on ComfyUI by do not know where the stabilityai/sd-vae-ft-mse safetensor must be saved for it to appear in the related dropdown. Kindest regards.
Hi, I have it working now, but only through the CPU, no Cuda acceleration (my Pinokio terminal says that buffalo_l models use CPUexecutionProviders). Do you have any idea how to solve this ? Thanks a lot !!!
Hi, friend. The tutorial is excellent and I thank you for it. I just have a problem. When creating the .bat file in the python embedded folder it does not allow me to run it as administrator , that is, the option to run it does not appear, what can I do?
After hours, I give up. Every time I give Comfy a chance, this happens. Nothing but installation problems. I followed everything, step by step still nothing but weird errors that I search and can't find answers for. I now think there is something in with my system because even Pinokio installs also fail. Errors like: Prompt outputs failed validation ImageConcatMulti: - Return type mismatch between linked nodes: image_2, LP_OUT != IMAGE ImageResizeKJ: - Return type mismatch between linked nodes: get_image_size, LP_OUT != IMAGE and many more.
It's giving me an error: When loading the graph, the following node types were not found: • LivePortraitProcess • DownloadAndLoadLivePortraitModels Nodes that have failed to load will show as red on the graph.
Hi there, i have sent you an email with a job offer, but i think it might have reached your spam folder, since i haven't heard back from you. Hope to get in contact with you soon :)
📢 Last chance to get 40% OFF my AI Digital Model for Beginners COURSE: aiconomist.gumroad.com/l/ai-model-course
happy with the results!! I achieve it at the first attempt, thanks for the clear explanation and step by step guide
Great job!
really digging the elevator music, makes the frustrations with comfy ui bearable, soothing even
Thank you for your amazing tutorial! It was really helpful and easy to follow. I appreciate the effort you put into making it.
Glad you enjoyed it!
Thank you. Interesting!
i was trying to figure this out all day, works fine standalone. I dont know anything about python but i guess i ended up with an insightface folder in that venv folder and everything works now. Thanks.
I have a question, first I have to have a virtual environment and have installed some libraries such as torch, and LivePortrait?
Little friendly note, the thumbnail looks like saying "From video to image", but i think i understand you mean the opposite, This looks like a great tutorial that ill be checking out when the time comes! Keep it up! ✨
Muchas gracias dure muchas horas tratando de solucionarlo
Thanks, I've solved it completely, great tutorial!
Did you have issues with Insightface ?
Do i need to install LivePortrait to have AdvancedLivePortrait? Because Comfy says im missing some nodes and I can't fix it for an hour
hello...thanks for video, please help me, show this error "PyTorch is not linked with support for mps devices" when execute queue.. over windows, with gforce 4070 graphics card
this question may be stupid but i am new, can i run ComfyUI and use Live Portrait using CPU only? (no GPU)
Thanks, But with ComfyUI installed with Pinokio we do not have the python_embeded directory,
where do we paste the insigthface file? Can we better install insightface from the comfyui manager?
Did you find a workaround?
savior ! Thanks
venv\scripts\Activate - Path cannot be found. There is no "venv" folder or file anywhere in my ComfyUI installation 😐
Now using LivePortrait with Pinokio, it's unofficial yet but installed fine and works like a charm.
Looking at their Github page, it seems to work on things that’s not photorealistic … like girls with pearl earring painting so I wonder how well it performs on character generated by sd. If it does work, video frames can be further used as training data for an original character LORA to help with consistency issue. Not what it’s intended for (paper from Kuaishou, dev of Kling) but interesting to try I think 🤔
Just watch the video
I would love to see if my face could come alive like this!
Hi there, I'm trying to install it on ComfyUI by do not know where the stabilityai/sd-vae-ft-mse safetensor must be saved for it to appear in the related dropdown. Kindest regards.
That is some low resolution but I guess if you use Topaz Video AI you could up sample that.
Hi, I have it working now, but only through the CPU, no Cuda acceleration (my Pinokio terminal says that buffalo_l models use CPUexecutionProviders). Do you have any idea how to solve this ? Thanks a lot !!!
Hi, friend. The tutorial is excellent and I thank you for it. I just have a problem. When creating the .bat file in the python embedded folder it does not allow me to run it as administrator , that is, the option to run it does not appear, what can I do?
just run it without admin
hi, I also have a problem there, with ComfyUI installed with Pinokio I don't have any venv folder, so it semms I can properly install insightface....
You’re the best, thanks!
When loading the graph, the following node types were not found:
KeypointsToImage 🔗
LivePortraitCropper 🔗
Please help solve this problem.Thank you
Hey, is your course live? Also your website is quite confusing
Muchas gracias, saludos
Can the output video resolution be more then 512x512?
This is still new and didn't try yet, I made this video focusing on the installation process and how to run the workflow.
does it support amd based pc?
Can you make a Mac install version video please 👍👍👍🙂
I don't see "manager" in my comfyui interface.
After hours, I give up. Every time I give Comfy a chance, this happens. Nothing but installation problems. I followed everything, step by step still nothing but weird errors that I search and can't find answers for. I now think there is something in with my system because even Pinokio installs also fail. Errors like:
Prompt outputs failed validation
ImageConcatMulti:
- Return type mismatch between linked nodes: image_2, LP_OUT != IMAGE
ImageResizeKJ:
- Return type mismatch between linked nodes: get_image_size, LP_OUT != IMAGE and many more.
bro, make vid to vid version, please - it will be cool
It's giving me an error:
When loading the graph, the following node types were not found:
• LivePortraitProcess
• DownloadAndLoadLivePortraitModels
Nodes that have failed to load will show as red on the graph.
same here
the same here ! does anyone have a solution?
@06ju I manually installed insightface and updated. Fully close comfy after update amd it worked after that
i cannot load json file, no matter how much i try
You haven't explained the stitching and relative parameters
I got about 45 seconds into the video and since my web interface doesn't have a manager button I could not continue.
doesnt work for me, the face in the final video is covered with a big black box
Sir you also, I did not receive this workflow?
Workflow can be found in the link in the description or in LivePortraitKJ Github page.
@@Aiconomist Thanks, I've solved it. I would like to ask the blogger, can this full body photo and video be driven?
No workflow
Workflow can be found in the link in the description or in LivePortraitKJ Github page.
Sorry. 6:37
Hi there, i have sent you an email with a job offer, but i think it might have reached your spam folder, since i haven't heard back from you. Hope to get in contact with you soon :)
Riiiight?! Hope you don't mind my asking, have you had any luck in your search?
Hi! Your content is fantastic. May I have your email address for a collaboration. Thanks
Thank you! Sure you can email me here : info@aiconomist .cc