Install Deep Live Cam Locally for Realtime Face Swap
HTML-код
- Опубликовано: 11 сен 2024
- This video shows how to locally install Deep Live Cam on Windows and change the face in any video or in live webcam with single image for free.
🔥 Buy Me a Coffee to support the channel: ko-fi.com/fahd...
🔥 Get 50% Discount on any A6000 or A5000 GPU rental, use following link and coupon:
bit.ly/fahd-mirza
Coupon code: FahdMirza
▶ Become a Patron 🔥 - / fahdmirza
#deeplivecam #faceswap
PLEASE FOLLOW ME:
▶ LinkedIn: / fahdmirza
▶ RUclips: / @fahdmirza
▶ Blog: www.fahdmirza.com
RELATED VIDEOS:
▶ Code www.fahdmirza....
▶ Resource github.com/hac...
All rights reserved © 2021 Fahd Mirza
🔥 Deep Live Cam - How-To Use - ruclips.net/video/I5emQaYzDYU/видео.html
🔥 Flux WebUI - Run Flux-1 Dev and Flux-1 Schnell Locally in Gradio GUI - ruclips.net/video/EZZ0cEJwZLA/видео.html
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
🔥 Buy Me a Coffee to support the channel: ko-fi.com/fahdmirza
Can I use python 3.12?
Which I used for avatarify installation
I followed everything in your video, but when creating a video or using Live mode, the quality is just terrible, while yours looks great. What could be wrong? My computer specs:
GPU - Nvidia GeForce RTX 4060 TI
CPU -12th Gen Intel(R) Core(TM) i5-12400F 2.50 GHz
Memory - 32 GB
It is still too early for this type of project to work 90%.
make sure its running on GPU.
LOL bor 4060 TI when its requirement is only 2060💀💀💀
@@fahdmirza I have rtx 4080 and i9-14900HX. Still terrible quality. Even my webcam is of good quality yet the output is terrible
@@VortexFlickens Make sure its running on GPU.
error in : choco install git / See: Unable to resolve dependency: Unable to find a version of 'chocolatey-core.extension' that is compatible with 'git.install 2.46.0 constraint: chocolatey-core.extension (>= 1.3.3)', 'vcredist140 14.40.33810 constraint: chocolatey-core.extension (>= 1.3.3)'.
Chocolatey installed 0/1 packages. 1 packages failed.
See the log for details (C:\ProgramData\chocolatey\logs\chocolatey.log).
Failures
- git - Unable to resolve dependency: Unable to find a version of 'chocolatey-core.extension' that is compatible with 'git.install 2.46.0 constraint: chocolatey-core.extension (>= 1.3.3)', 'vcredist140 14.40.33810 constraint: chocolatey-core.extension (>= 1.3.3)'.
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
i get the following error PS C:\Windows\system32\Deep-Live-Cam> python run.py
Traceback (most recent call last):
File "C:\Windows\system32\Deep-Live-Cam
un.py", line 3, in
from modules import core
File "C:\Windows\system32\Deep-Live-Cam\modules\core.py", line 14, in
import torch
ModuleNotFoundError: No module named 'torch'
make sure you install all the pre-requisites like. All code is in the link in video description, thanks.
Greetings. I tried to install this program on my PC, but I did not get the final result, everything was done consistently as in the video can you help with the installation?
same error
I MAY have figured it out. When installing the PIP, leave it to install for ages. Theres a while there where it looks like its doing nothing, but if you exit early, the torch files dont install correctly.
@@kidinacagewhat do you mean by this?
The program opens but i get this message when i start the live button. any ideas?
INVALID_PROTOBUF : Load model from C:\Users\xxxxx\deep-live-cam\models\inswapper_128_fp16.onnx failed:Protobuf parsing failed.
Had that issue as well, needed to DL both model files again.
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
What do mean by "DL" @@nokoneffets405
Hi
Is it possible to route the live view to for example Teams?
I really want to show this as an awareness for my company in a video conference :)
Thank you very much, great work!
Got it up and running on my 4090, but the feed is super clunky and the face swap is just visual noise?
Same for me :(
Same for me
yeah you would need more VRAM
@@fahdmirza what u mean VRAM?
you need to install some cuda stuff for windows
I cant find the link to the command like you said
Thank you! You won a new sub
Awesome, thank you!
Wonderful work! Do you know what is the pixel size limit for the image and the target video?
Let me check the pixel size if I could find it.
i think it is 640 x 640
Does it work without GPU? How so? Is it limiting the quality of the video or just taking longer? IMO an option to choose either would be nice :).
Very cool project, keep up the good work!
cheers
Small correction, to check the version of CUDA which is installed "nvcc --version" is what I personally would use as nvidia-smi shows general information about the graphics drivers iirc.
nvidia-smi is fine.
I have an Error!
PS D:\Neiro\Deep-Live-Cam-main\Deep-Live-Cam> python run.py --execution-provider openvino
Traceback (most recent call last):
File "D:\Neiro\Deep-Live-Cam-main\Deep-Live-Cam
un.py", line 3, in
from modules import core
File "D:\Neiro\Deep-Live-Cam-main\Deep-Live-Cam\modules\core.py", line 14, in
import torch
File "C:\Users\ReynBighooves\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\__init__.py", line 148, in
raise err
OSError: [WinError 126] The specified module could not be found. Error loading "C:\Users\ReynBighooves\AppData\Local\Programs\Python\Python310\lib\site-packages\torch\lib\fbgemm.dll" or one of its dependencies.
you need to install all pre-requisites
your python version has to be 3.10. i had 3.12 installed and was getting the same error until i downgraded.
@@fahdmirza allright, buw how? I just download and install python 3.10 and i have this bug too, thx
error in
'Frame processor face_enhancer not found'
Please make sure to follow all the steps, especially the pre-requisites. Thanks.
I got the same error, did you find a fix?
Is there a bulk create function and what about chance background? And full body movement to another
Not at the moment.
Yup to bad this got errors bout to do what ur thinking n fuck hooly wood uo lolololol
So is it possible to do this without doing a video?
yes should be good
What are the systems requirement ?
Despite Running the CUDA version it is only utilizing my CPU and not the GPU. I believe it might be because I don't have that correct CUDA version installed? I have a newer version that I already had previously, should I downgrade it for this?
yes its version sensitive
how could i stream it in obs stream ?
please can you make a video on how to install the prerequisites
i receive the following error when trying to install the requirements: no module named "distutils"
You are not properly installing pre-reqs
your python version has to be 3.10. i had 3.12 installed and was getting the same error until i downgraded.
Do you think there will be possible in the future to do a full face replacing, including hair?
oh yes definitely
HeY fahad...thanks for putting up this video. I got this working as per your instructions....but it stops at the face enhancer - does not process that. I put the models in the model folder....any idea what to do?
Do you get any errors?
@@fahdmirza Actually once it's done part 1 and then moves onto processing...the progress just says at 0% and does not move forward. Without image enhancing, the end result is un-usable
Can you stream on Twitch and have this in live?
Yes should be fine.
bro, i have rtx 3060 and in all the tests I did, the face swapping was never good, very "BUGGED" you know? Why does this happen? I did everything as instructed in the repository
Hi, plz see this video for further info and any troubleshooting tips : ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=-zong7Sa__Zbs0Fj
Thanks for the Tutorial. Im facing a problem. As soon as i click to the Live Button, it goes live with my Elgato Capture Card. So i cant use my Cam. Any Ideas how to change that without deactivating my Capture Card?
may be disable the other card
note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for insightface
Failed to build insightface
ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (insightface)
please help
then when i run command python run.py --execution-provider cuda
it says
ModuleNotFoundError: No module named 'torch'
i have python version 3.10
pip 24.2
git version 2.46.0.windows.1
ffmpeg latest installed
running on latest vscode version too
i have the same "insightface" issue
I have the same problem
Same
same problem
how do you keep the face quality ? i tried mine even thought the selected face is on high resolution, it show low resolution on the output
GPU VRAM is the key
AttributeError: 'NoneType' object has no attribute 'shape'
I'm getting the same error too
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
@@fahdmirza ok now it shows another mistake, when i try to run Live cam it give another error: "CUDA driver version is insufficient for CUDA runtime version", does it work with AMD graphic card?
Interesting thing happens in my case.
Specs: Have a laptop with an intel processor and nVidia 3080 TI GPU.
Behaviour1: Starting the program with execution provider CUDA still seems to use only the processor, as it is under 100% usage while the dedicated GPU is at 0% usage.
Behaviour2: The program runs and finalizes the task, however the output is displaying the target with a dark / black square over the face.
Some suggestions:
In the video it would be nice if you'd show the terminal as well after starting the program to see if any warnings or errors come in the background. In my case when pressing start I see some error messages which have to do with:
"LoadLibrary failed with error 126 "" when trying to load site-packages\onnxruntime\capi\onnxruntime_providers_cuda.dll"
Moreover there you could also see the progress in percentages and live evolution.
Curious if any of you encountered the above.
Cheers!
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
i’m having the issue where torch isn’t being found but i look earlier in the code and see when i install requirements.txt it ignores torch and onnruntime?
Install it manually with pip
@@fahdmirza install what manually torch?
yes with pip install torch
Can't get it to run. A handful of issues like, "opencv-python 4.8.1.78 depends on numpy>=1.17.0; python_version >= "3.7"". I tried changing the requirements txt to remove the specific versions but can't get it working. Suggestions?
Its very sensitive about requirements and their versions, so please follow it to the dot. Check the code in my blog and link is in video description.
why is a black square drawn instead of replacing the face
It means its unable to use face you uploaded. Use a clearer photo as source
Sometimes this works:
pip uninstall onnxruntime onnxruntime-gpu
pip install onnxruntime-gpu==1.16.3
Hello! Thanks a lot, I installed everything, but when I turn on the webcam I get AttributeError: 'NoneType' object has no attribute 'shape' in Windows Powershell, also happens when I want to change the face on the video. Please help me
you just need to copy your photo/video and git folder into folder without unicode symbols, for example on C://myfolder
this might sound a bit odd, but do you know of any specific reason why i don't have the NSFW toggle?
thats weird, as it should be there. May be they are updating the interface or something as its quite actively evolving project. Or may be try refreshing your browser. Sometime , a local browser policy also disables radio buttons
My Powershell says that "pip" is not part of a Cmdlet, function or Script. What did i do wrong ?
Please watch this video ruclips.net/video/p7hjO2ZrfRU/видео.html
Friend, I'm afraid to install this program on my main computer, do you know if it's virus-free? thanks for the tutorial
use sandbox
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
Is there anyway to improve the lip movements? Thanks!
No, sorry
any ideas how to make live output shown in discord as a webcam ?
I think that would require writing a wrapper on top of it.
How@@fahdmirza
So this only works if you have an Nvida card? Cuda does not run without an Nvidia GPU.
Well anyways if you're planning for using this program seriously then you should consider having a good hardware so having an Nvidia GPU is pretty much a prerequisite for this.
agreed
I get an error: (all pre-requisites installed)
PS C:\Windows\system32\Deep-Live-Cam> python run.py
Traceback (most recent call last):
File "C:\Windows\system32\Deep-Live-Cam
un.py", line 3, in
from modules import core
File "C:\Windows\system32\Deep-Live-Cam\modules\core.py", line 14, in
import torch
ModuleNotFoundError: No module named 'torch'
pip install torch
Thanks.
Hello and thanks, it works maybe better than DeepFaceLive, but anyway I got this error "Frame processor face_enhancer not found" have any idea how to fix it? I'm thinking that with face enhancer fixed it will works even better!
It normally occurs due to inswapper_128_fp16.onnx file in models. When I changed it to another model file in a different resource (github.com/facefusion/facefusion-assets/releases/download/models/inswapper_128_fp16.onnx)
@@fahdmirza I have it
I had the same error, did you find a solution?
How do I start up the code again after turning on my system, My PC went off and I will like to restart the cam code since I already completed the installation and activation. can someone put me through on what to do?
Hi, plz see this video for further info and any troubleshooting tips : ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=-zong7Sa__Zbs0Fj
the process completes but the face gets only a black object!? how to solve it sir?
I answered this in another comment, plz check, thanks
it doesnt work for me, the output is just a copy of the targer video/ picture unchanged
I would suggest, cleanup everything and start from scratch and make sure to have a good GPU
Hello. My interface launches successfully, but in Windows PowerShell, the program stops with the error:
File "C:\python\lib\site-packages\insightface\model_zoo
etinaface.py", line 211, in detect
im_ratio = float(img.shape[0]) / img.shape[1]
AttributeError: 'NoneType' object has no attribute 'shape'
and no result is returned. Help me please
I would assume it's due to the program not being able to read frames from your webcam/ capture device since "img" is a NoneType.
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
Can someone help me? I managed to install it but when I put it live the camera doesn't open, there's a gray screen in the preview tab and the Powershell prompt says it's an error in the CPU, when I open it directly on the CPU It works very slow, like 2 fps, and it heats up the PC a lot. It should work better on my card, how do I make it work?
Please watch this video ruclips.net/video/p7hjO2ZrfRU/видео.html
Any chance for portable installation ?
Would you please elaborate what you mean?
@@fahdmirza I thinks he mean like AUTOMATIC1111 for ex. where you just run some .BAT and it installs all the reqs. and create the separate ENV.
Honestly wanna thank you for making this video, but given all the complicated stuff you have to do and the fact that I have a crappy laptop, this probably aint for me, which sucks because I had a pretty innocent use for this program (I wanted to use faceapp filter pictures I've taken and put it on my face in real time to mess around with them and see how they looked)
But alas, maybe another time.
Thank you.
anyone know how much HD space is needed to install those?
Its pretty light weight. I will check the space on my system and let you know, thanks
@@fahdmirza thank you
I‘m in China, and the SSL error is displayed in operation. Is it a network problem? Need a VPN?
whats the error
Please help ,i did everything to the very last end and when i run python run.py --execution-provider cuda , then i get this error .
Traceback (most recent call last):
File "run.py", line 3, in
from modules import core
File "C:\Windows\system32\Deep-Live-Cam\modules\core.py", line 20, in
import modules.ui as ui
File "C:\Windows\system32\Deep-Live-Cam\modules\ui.py", line 10, in
from modules.face_analyser import get_one_face
File "C:\Windows\system32\Deep-Live-Cam\modules\face_analyser.py", line 5, in
from modules.typing import Frame
File "C:\Windows\system32\Deep-Live-Cam\modules\typing.py", line 7, in
Frame = numpy.ndarray[Any, Any]
TypeError: Type subscription requires python >= 3.9
which python version are you using?
I'm having a long line of error while trying to install
Pip install -r requirements
The rest was smooth, but i don't know why I'm having the error.
Please I need your assistance
I'm using an nvidia pc
whats the error
@@fahdmirza note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for insightface
Failed to build insightface
ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (insightface)
I use a MacBook Pro 2021 M1 chip model .. will it work ? And how ?
Hi, plz see this video for further info and any troubleshooting tips : ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=-zong7Sa__Zbs0Fj
this is the new error i am having
Traceback (most recent call last):
File "C:\Users\chris\Deep-Live-Cam
un.py", line 3, in
from modules import core
File "C:\Users\chris\Deep-Live-Cam\modules\core.py", line 20, in
import modules.ui as ui
File "C:\Users\chris\Deep-Live-Cam\modules\ui.py", line 10, in
from modules.face_analyser import get_one_face
File "C:\Users\chris\Deep-Live-Cam\modules\face_analyser.py", line 2, in
import insightface
ModuleNotFoundError: No module named 'insightface'
Please install all the pre-requisites.
@@fahdmirza i have installed all?
@@zuczz you didnt run the requirement command through a powershell cd to the directory of the Deeplivecam.
As you can read for yourself, there is no module named insightface which is installed while installing prerequisites..
Whats the gpu requirement. What gpu r u using?
NVIDIA RTX A6000 , but any GPU around or above 24GB VRAM should be good. Ideally I would recommend 48GB VRAM.
Hi there.. Does anyone know the minimum requirements? and better GPU or CPU. I have a RTX 12GB and 16GB RAM.. thanks
I have observed that it struggles a lot any VRAM below 24GB. Ideally, I would suggest anything like 48GB VRAM especially if you are live. Thanks.
everything installed, but impossible to launch Deep-Live-Cam (base) PS C:\users\user> cd Deep-Live-Cam (base) PS C:\users\user\Deep-Live-Cam> python run.py Traceback (most recent call last): File "C:\users\user\Deep-Live-Cam
un.py", line 3, in from modules import core File "C:\users\user\Deep-Live -Cam\modules\core.py", line 14, in import torch File "C:\Users\User\anaconda3\Lib\site-packages\torch\__init__.py", line 148, in raise err OSError: [WinError 126] The specified module could not be found. Error loading "C:\Users\User\anaconda3\Lib\site-packages\torch\lib\fbgemm.dll" or one of its dependencies.
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
After I cloned git deep live on my local system i cant find any Folder in C:\Users\angel\Deep-Live-Cam\ \Deep-Live-Cam\ doesnt exist
Edit: I found it in the Sytem 32 Folder. Can i move it somewhere els because it wont allow me to save the link there because its the system 32 folder or would that break it?
No, and you would need admin rights on your system
i don't have envidia GPU I'm running it from my processor and it just shows black square on face, i tired face swapping on premade video but it doesn't show any video when i go to select it
Please use a clear photo for source
You should consider upgrading
note: This error originates from a subprocess, and is likely not a problem with pip.
error: subprocess-exited-with-error
× pip subprocess to install backend dependencies did not run successfully.
│ exit code: 2
╰─> See above for output.
note: This error originates from a subprocess, and is likely not a problem with pip
i got this erore help me please
This is not the main error, you need to install the packages which it is complaining about above this error.
so you are saying that I have no chance running this thing on my GTX 1660 super
You can try it out
error: subprocess-exited-with-error
× Building wheel for insightface (pyproject.toml) did not run successfully.
│ exit code: 1
╰─> [215 lines of output]
ERROR: Failed building wheel for insightface
Failed to build insightface
ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (insightface)
note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for insightface
got these, what should i do?
you need to install all pre-reqs
The live button only shows me a green square, I don't know how to make the live web cam work. Can someone help me? Thanks!
and does reloading the browser fix it?
@@fahdmirza Closing and reopening doesn't fix it, it's like opening the preview. I don't know if it has to do with any specific dependency right here
@@fahdmirza I use iriun webcam, I don't know if it's because of that, before having the application turned on it shows me the message to open the application, when I open it the rectangle appears completely green
Traceback (most recent call last):
File "C:\Windows\system32\Deep-Live-Cam
un.py", line 3, in
from modules import core
File "C:\Windows\system32\Deep-Live-Cam\modules\core.py", line 14, in
import torch
File "C:\Python310\lib\site-packages\torch\__init__.py", line 148, in
raise err
OSError: [WinError 126] The specified module could not be found. Error loading "C:\Python310\lib\site-packages\torch\lib\fbgemm.dll" or one of its dependencies.
HELP ME PLS!!
I would suggest, cleanup and then start from scratch and follow the steps I shared. Thanks.
@@fahdmirza I did it and its still the same problem, so, whats up? I have all versions same like you...
for some weird reason I get a black face no matter what type of picture I use, why is that?
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
do you have tutorials on how to run it in OBS live?
it will be same steps
@@fahdmirza how it will be same steps could you explain ?
does this work for mac os?
yes should be ok
got these, what should i do?
error: subprocess-exited-with-error
× Building wheel for insightface (pyproject.toml) did not run successfully.
│ exit code: 1
╰─> [215 lines of output]
ERROR: Failed building wheel for insightface
Failed to build insightface
ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (insightface)
note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for insightface
You need to install all pre-requisites with proper versions as I did
What computer configuration those this app work on
Hi, plz see this video for further info and any troubleshooting tips : ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=-zong7Sa__Zbs0Fj
When ll be test on 4090? U have freezing with enbled face enhancer on 4090 or not?I m looking for solution with face enhancer
No you need larger VRAM GPU
@@fahdmirza Can u show?How larger?I want to optimize GFPGAN
@@fahdmirza You dont have freezing with 4090?When you enable face enhancer?
@@fahdmirza I have 4070 ti super
Using a 5950x and a 3070 and its stuttering like hell :D Guess i need an upgrade for this
yep definitely need bigger GPU
How do you get pip you forgot to say that in the video and I'm stuck on a step because I don't have it
It mentions python as prerequisite and pip comes with it
@@fahdmirza at 7:43 in the video when I run the command it gives me the error "'pip' is not recognized as the name of a cmdlet, function, script file, or operable program." I made sure I was in the correct directory before I ran the program as well. I'm not sure why it's saying it.
@@foxivus7123 create a new path for pip exe directory in enviroment variables
i have this error
PS C:\Users\PC Gamer\Downloads\IACAM\Deep-Live-Cam> python run.py --execution-provider cuda
Traceback (most recent call last):
File "C:\Users\PC Gamer\Downloads\IACAM\Deep-Live-Cam
un.py", line 3, in
from modules import core
File "C:\Users\PC Gamer\Downloads\IACAM\Deep-Live-Cam\modules\core.py", line 20, in
import modules.ui as ui
File "C:\Users\PC Gamer\Downloads\IACAM\Deep-Live-Cam\modules\ui.py", line 5, in
import cv2
ModuleNotFoundError: No module named 'cv2'
i solved this error installing this codes.....
\/
pip install opencv-python
pip install insightface
but now, i have a new error.... " python run.py --execution-provider cuda
Frame processor face_enhancer not found"
Please see if you find solution to your error in this video:
🔥Common Errors Fixed in Live Web Cam - ruclips.net/video/p7hjO2ZrfRU/видео.htmlsi=pT0tvTd_BQAUJuTv
can you give us the requirements.txt manuallay so we can copy i get an error
Already shared all the code and link is in video description.
what to do if just a black square appears on your face?
I already answered that in another comment, plz check.
Can i use this on kali linux?
no
bro ,please give me your : pip install list,how the hell you meet the requirements for jax and onnxtime ,it NEEDS numpy above 24.0 version
You can just create a virtual environment with python 3.10, and then there should be no dependency issue
Agreed, and all the code is in link in video description
Where is the command to run the chocolatey?
Please see this video for troubleshooting deep live cam ruclips.net/video/p7hjO2ZrfRU/видео.html
Can this work on a nvidia gtx1650 4 gigabyte ram?
ERROR: Failed building wheel for insightface any help?
yes i had the same issue you need to make sure visual studio is installed correctly, make sure torch py is installed, id ask chat gpt to help you paste the errror messages from powershell into chat gpt
cheers
Please can I use python 3.12 instead of 3.10?
Please see this video for troubleshooting deep live cam ruclips.net/video/p7hjO2ZrfRU/видео.html
Is it available for MacBook?
shoudl be good
I have so many packages installed for Python and other shit that I have no space on C drive to install the dependencies so I just cancelled lol.
cheers
idk why i got virus or what, pay pal account jus pop up in my browser
its hacksider pay pal i think
Thats strange, I never got that. Must be issue with your own system.
@@fahdmirza I just open run-cuda.bat
What gpu do you have?
NVIDIA RTX A6000
What could possibly go wrong 🤔😂
cheers
Catfishing 2.0
Getting arrested for a crime I never committed 😂😂
@@curiositycloset2359 Hey any progress?
@@curiositycloset2359any progress?
How do I run this with a Jupyter/Vast AI GPU
same method
Anyone know if this is hella demanding while streaming on OBS ? id imagine it is.. Still gotta try this tho
Would you plz elaborate your question?
@@fahdmirza how demanding is this on your PC system ? I want to use this while streaming games. My PC lags when I do this with Nvidia broadcast green screen, I feel like it would be the same using this Deep Live software while playing games .
File "C:\Python310\lib\site-packages\torch\__init__.py", line 148, in
raise err
OSError: [WinError 126] The specified module could not be found. Error loading "C:\Python310\lib\site-packages\torch\lib\fbgemm.dll" or one of its dependencies.
i get this error, i dunno what to do
I would suggest cleanup installation and start from scratch and follow the code which I shared. Link is in video description
Can this work on Gtx 1050 4gb dedicated laptop?
you can try, it might be very slow though
wtf you just have an A10 ? it costs like a house here where i live.
You don't have to buy it. You can always rent for fraction of cost if you are keen to run it.
and how to uninstall everything later???
just use choco uninstall
Why I am getting really bad results?
may be its not running on your GPU
@@fahdmirza I had square problem with cpu, and solved it. Runs with twk xeons but slow. I think with gpu I had some errors version related. which are I guess tensorflow, torch etc. I will give it a try tomorrow for cuda power. Thank you video is very useful!
@@TolgaOzisik Sure, let us know how you go. thanks
@@TolgaOzisik, how did it go?
@@user-kl7eq4zy4z thanks for asking I managed to run with cpu. I will try with cuda soon.
I tried launching and this error pop up, I tried to install matplotlib but it couldn't uninstall the other version so I coulnd't upgrade it
ModuleNotFoundError: No module named 'matplotlib.pyplot'
plz share the commands you are using. I would suggest, cleanup and then start from scratch and follow the steps I shared. Thanks.
how do i with local video instead of live camera
Simply upload the video in target.
so you need to make sure you select video instead of the default image when you are in the folder with the video your uploading to target. the choice is on the bottom right of the file window
Can i run it on mac m2 ?
I haven't tried but should be fine.
why should I select NSFW? what does it do?
Its for Not Safe For Work, please search the channel for more explanation.
@@fahdmirza but it's just a face swap, what could there be nsfw?
please show how to hook it up to a livestream
You just have to click on Live button there. Thanks.
Cant find the face ? where is it
On your head. 👍
@@flyhyland hahaaha
Lol, very witty. Make sure the photo is clear which you are using.