Hi, it seems to point to either missing modules or the inswapper model. Can you have a look if one of the below links helps. Let me know if you get it resolved. github.com/hacksider/Deep-Live-Cam/issues/154 or check github.com/hacksider/Deep-Live-Cam/issues/219
Hi, you can open the anaconda prompt again. Activate the environment with conda activate dlc or whatever you called the environment. Make sure you navigate back to the project folder. From the project folder Deep-Live-Cam just launch the SW again. Check 9:26 into the video. For Nvidia you would run python run.py --execution-provider cuda. Thanks for reaching out, let me know if you come right.
@@stuffaboutstuff4045how do i navigate to the project folder, that’s what i don’t know how to do. I don’t understand how to launch SW, can you tell me the input i’ll use, i did exactly the same thing you did tho
please tell me. I did everything as in the video, but it gives me an error: attributeError: ‘NoneType’ object has no attribute ‘shape’. what could this mean?
@@stuffaboutstuff4045 Hi, figured out the error. it occurred when trying to start the face replacement process. the whole point is that I did not place the photos and videos I was using in the root folder with the dlc files.
Thank you 🙏 I have successfully installed and run without any issues. Which command line do I need to use when I want to re-run the program after switching off computer. ?
Hi, Pleasure glad the video assisted in getting you up and running. After reboot, open anaconda prompt again, make sure you activate the conda environment you created. Browse to your Deep-Live-Cam folder. Run the command you used for either CPU or GPU, if Nvidia you would launch with python run.py --execution-provider cuda. Thanks for reaching out, let me know if you are up and running.
Successfully install everything, it works, but when try to go live, all black in the screen Dont know whats going wrong here Anyonr face the same problems?
Hi, This seems to happen if the GPU acceleration not working properly or it might be the inswapper. It can also be another application holding onto the webcam and DLC cannot capture frames. Have a look if the below links help. I will also build the SW again and check if something changed on the repo. On the GitHub click on issues and search "webcam" You will see a couple of other fixes as well. First make sure your GPU acceleration is working for your GPU, then check inswapper model.. Let me know if you manage to resolve this... github.com/hacksider/Deep-Live-Cam/issues/663 github.com/hacksider/Deep-Live-Cam/issues/649 You can download the inswapper or replacement version from the project page direct. Scroll down to download models. github.com/hacksider/Deep-Live-Cam
After the cuda install, i failed in nvidia-smi, it said 'nvidia-smi' is not recoqnized as and internal ot external command, operable program or batch fule... Any solutions for this?
Finally i can pass the requirements.txt, found the solutions in youtube, using anaconda prompt, but now im stuck in : "cuda driver version is insufficient for cuda runtime version" Problems Is it because previously i cannot run nvidia-smi? Because when i try that nvida-smi cmd, it said nvidia -smi is not recognized as an internal or external command operable program or batch file
Hi, Do you have the Nivida drivers installed? nvidia-smi ships with the default driver. Let me know if you managed to resolve this. The drivers must be installed on the Windows machine OS you are using.
Ahhh , my laptop using AMD 😂😂😂 Thats why cant find NVIDIA So sorry about my own mistake So i think theres nothing i can do about it? Or can i do something to fix this?
Hi, glad you identified that. You can build it for your AMD. You need to get DirectML working. Check the below link, expand GPU Acceleration on the page. Follow the DirectML instructions for the AMD. Let me know if you get it working. github.com/hacksider/Deep-Live-Cam
Hi, can you confirm you installed ffmpeg? Make sure you installed all the required SW as from 1:58 in the video. Check the ffmpeg installation from 4:53 in the video. You can also check out the below link if the issue persist. Thanks for reaching out. github.com/hacksider/Deep-Live-Cam/issues/241
Hi, you need a relatively solid AMD or Nvidia Hardware GPU for most of the AI Tools to run with some form of respectable speed. Running of the CPU will also take forever to compile. Thanks for reaching out, if you need help let me know..
Thank you for your help, I was able to install and launch Deeplivecam on my cpu.👍 I opened the jpg format in Select a face and clicked live, but nothing showed up in Preview. What error is this??😢
Hi, did you copy your content into your Deep-Live-Cam project folder like I did in the video. Let me know if you were able to resolve and run your project. Thanks for reaching out.
I face the same problem here When i clicked live nothing show up in preview Do you mean for going live, we also need to put video( content)? Its not real time life ? Coz i want to try to use it in video call, so i want to swap the face in real time, i speak in the video call, but appear as other face in my friends screen
Hi, If you are still having issues with live webcam, Please make sure you have all the required SW and that each step in the video was followed. If you still have problems with webcam check if the below assists. Hi, This seems to happen if the GPU acceleration not working properly or it might be the inswapper. It can also be another application holding onto the webcam and DLC cannot capture frames. Have a look if the below links help. I will also build the SW again and check if something changed on the repo. On the GitHub click on issues and search "webcam" You will see a couple of other fixes as well. First make sure your GPU acceleration is working for your GPU, then check inswapper model.. Let me know if you manage to resolve this... github.com/hacksider/Deep-Live-Cam/issues/663 github.com/hacksider/Deep-Live-Cam/issues/649 You can download the inswapper or replacement version from the project page direct. Scroll down to download models. github.com/hacksider/Deep-Live-Cam
Hi, just checking if you managed to resolve this issue. If not can you confirm loading the required SW and building everything using Anaconda with the Python version etc. I use in the video? I have no Python installed on my Windows host directly and use Anaconda to supply Python inside the environment. Can you trace the steps again from about 2:00 onwards into the video. Let me know if you get it up and running. Thanks for reaching out.
Hi, sounds like a faulty Anaconda install. Are you running the full Anaconda install in Windows? You can have a look at the link included and see if it helps with your issue. Let me know if you are up and running. stackoverflow.com/questions/78347799/why-my-anaconda-keeps-showing-error-while-loading-conda-entry-point
Hi, you need Visual Studio with C++ Build Tools selected. Without it the build will fail on Windows. Let me know if you get it up and running. Thanks for reaching out.
Thanks so much for the detailed tutorial, i really appreciate. when i execute "python run.py --execution-provider cuda", the Deep Live Cam will pop out but when i click on the 'Start' or 'Live' Button i get the error messages below: Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'cudnn_conv_algo_search': 'EXHAUSTIVE', 'device_id': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'has_user_compute_stream': '0', 'gpu_external_alloc': '0', 'enable_cuda_graph': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0'}, 'CPUExecutionProvider': {}} find model: C:\Users\Administrator/.insightface\models\buffalo_l\1k3d68.onnx landmark_3d_68 ['None', 3, 192, 192] 0.0 1.0 Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'cudnn_conv_algo_search': 'EXHAUSTIVE', 'device_id': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'has_user_compute_stream': '0', 'gpu_external_alloc': '0', 'enable_cuda_graph': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0'}, 'CPUExecutionProvider': {}} But when i run the code using just "python run.py" well but it is hanging. please what could be the cause
Hi, can you confirm you installed ffmpeg? Make sure you installed all the required SW as from 1:58 in the video. Check the ffmpeg installation from 4:53 in the video. You can also check out the below link if the issue persist. Thanks for reaching out. github.com/hacksider/Deep-Live-Cam/issues/241
Hi, just checking if you managed to resolve. If not can you confirm loading the required SW especially MS Visio Studio Community with the selection I have in the video (2:00 minutes in)? Let me know if you managed to get this tool up and running. Thanks for reaching out.
@@stuffaboutstuff4045 thanks for checking in, I just logged into my RUclips account today. I already fixed that one and it's working but the real time deepfake function is not working
Hi, Please make sure you installed all the required SW first. If you have VS Studio etc. and problem persists please check if the below applies. Hi, This seems to happen if the GPU acceleration not working properly or it might be the inswapper. It can also be another application holding onto the webcam and DLC cannot capture frames. Have a look if the below links help. I will also build the SW again and check if something changed on the repo. On the GitHub click on issues and search "webcam" You will see a couple of other fixes as well. First make sure your GPU acceleration is working for your GPU, then check inswapper model.. Let me know if you manage to resolve this... github.com/hacksider/Deep-Live-Cam/issues/663 github.com/hacksider/Deep-Live-Cam/issues/649 You can download the inswapper or replacement version from the project page direct. Scroll down to download models. github.com/hacksider/Deep-Live-Cam
Hi, just checking if you managed to resolve this issue. Sounds like you have the Nvidia driver and the CUDA Toolkit installed with a mismatch between the two. You can try running it with only Gaming driver. It should give you enough support as is. If you already have CUDA Toolkit on the PC then you should stick with the driver it installs. Let me know if you managed to resolve this issue. Thanks for reaching out.
Hi, If you are still having issues with live webcam, Please make sure you have all the required SW and that each step in the video was followed. If you still have problems with webcam check if the below assists. Hi, This seems to happen if the GPU acceleration not working properly or it might be the inswapper. It can also be another application holding onto the webcam and DLC cannot capture frames. Have a look if the below links help. I will also build the SW again and check if something changed on the repo. On the GitHub click on issues and search "webcam" You will see a couple of other fixes as well. First make sure your GPU acceleration is working for your GPU, then check inswapper model.. Let me know if you manage to resolve this... github.com/hacksider/Deep-Live-Cam/issues/663 github.com/hacksider/Deep-Live-Cam/issues/649 You can download the inswapper or replacement version from the project page direct. Scroll down to download models. github.com/hacksider/Deep-Live-Cam
I LOVED Bloodsport. You are the man for picking that clip for this video.
😉Same here, it was a classic!
You are the BEST. Thanks for the detailed tutorial.....
Glad the video was helpful! Thanks for the feedback, much appreciated.
After running run.py it reads back “frame processor face_enhancer not found”help please
Hi, it seems to point to either missing modules or the inswapper model. Can you have a look if one of the below links helps. Let me know if you get it resolved.
github.com/hacksider/Deep-Live-Cam/issues/154 or check github.com/hacksider/Deep-Live-Cam/issues/219
Thank you it worked for me, i want to ask how do i open up the deep face live afresh when i want to use it again
I mean the codes to run it again
Hi, you can open the anaconda prompt again. Activate the environment with conda activate dlc or whatever you called the environment. Make sure you navigate back to the project folder. From the project folder Deep-Live-Cam just launch the SW again. Check 9:26 into the video. For Nvidia you would run python run.py --execution-provider cuda. Thanks for reaching out, let me know if you come right.
@@stuffaboutstuff4045how do i navigate to the project folder, that’s what i don’t know how to do. I don’t understand how to launch SW, can you tell me the input i’ll use, i did exactly the same thing you did tho
@@stuffaboutstuff4045thank you I finally got it
Pleasure, glad you are up and running.
please tell me. I did everything as in the video, but it gives me an error: attributeError: ‘NoneType’ object has no attribute ‘shape’. what could this mean?
Hi, after which command do you get this error?
@@stuffaboutstuff4045 Hi, figured out the error. it occurred when trying to start the face replacement process. the whole point is that I did not place the photos and videos I was using in the root folder with the dlc files.
Glad to hear you are up and running.
Thank you 🙏
I have successfully installed and run without any issues.
Which command line do I need to use when I want to re-run the program after switching off computer. ?
Hi, Pleasure glad the video assisted in getting you up and running. After reboot, open anaconda prompt again, make sure you activate the conda environment you created. Browse to your Deep-Live-Cam folder. Run the command you used for either CPU or GPU, if Nvidia you would launch with python run.py --execution-provider cuda. Thanks for reaching out, let me know if you are up and running.
Successfully install everything, it works, but when try to go live, all black in the screen
Dont know whats going wrong here
Anyonr face the same problems?
Hi, This seems to happen if the GPU acceleration not working properly or it might be the inswapper. It can also be another application holding onto the webcam and DLC cannot capture frames. Have a look if the below links help. I will also build the SW again and check if something changed on the repo. On the GitHub click on issues and search "webcam" You will see a couple of other fixes as well. First make sure your GPU acceleration is working for your GPU, then check inswapper model.. Let me know if you manage to resolve this...
github.com/hacksider/Deep-Live-Cam/issues/663
github.com/hacksider/Deep-Live-Cam/issues/649
You can download the inswapper or replacement version from the project page direct. Scroll down to download models.
github.com/hacksider/Deep-Live-Cam
After the cuda install, i failed in nvidia-smi, it said 'nvidia-smi' is not recoqnized as and internal ot external command, operable program or batch fule...
Any solutions for this?
Finally i can pass the requirements.txt, found the solutions in youtube, using anaconda prompt, but now im stuck in : "cuda driver version is insufficient for cuda runtime version" Problems
Is it because previously i cannot run nvidia-smi? Because when i try that nvida-smi cmd, it said nvidia -smi is not recognized as an internal or external command operable program or batch file
Hi, Do you have the Nivida drivers installed? nvidia-smi ships with the default driver. Let me know if you managed to resolve this. The drivers must be installed on the Windows machine OS you are using.
Ahhh , my laptop using AMD 😂😂😂
Thats why cant find NVIDIA
So sorry about my own mistake
So i think theres nothing i can do about it?
Or can i do something to fix this?
Hi, glad you identified that. You can build it for your AMD. You need to get DirectML working. Check the below link, expand GPU Acceleration on the page. Follow the DirectML instructions for the AMD. Let me know if you get it working.
github.com/hacksider/Deep-Live-Cam
the software work but when i start it up and try it out it says protobuf parsing failed
Hi, can you confirm you installed ffmpeg? Make sure you installed all the required SW as from 1:58 in the video. Check the ffmpeg installation from 4:53 in the video. You can also check out the below link if the issue persist. Thanks for reaching out. github.com/hacksider/Deep-Live-Cam/issues/241
same issue im facing too, any solution?
I don't understand I have a window 7 hp elitebook
Hi, you need a relatively solid AMD or Nvidia Hardware GPU for most of the AI Tools to run with some form of respectable speed. Running of the CPU will also take forever to compile. Thanks for reaching out, if you need help let me know..
Thank you for your help, I was able to install and launch Deeplivecam on my cpu.👍
I opened the jpg format in Select a face and clicked live, but nothing showed up in Preview. What error is this??😢
Hi, did you copy your content into your Deep-Live-Cam project folder like I did in the video. Let me know if you were able to resolve and run your project. Thanks for reaching out.
I face the same problem here
When i clicked live nothing show up in preview
Do you mean for going live, we also need to put video( content)?
Its not real time life ? Coz i want to try to use it in video call, so i want to swap the face in real time, i speak in the video call, but appear as other face in my friends screen
Hi, If you are still having issues with live webcam, Please make sure you have all the required SW and that each step in the video was followed. If you still have problems with webcam check if the below assists.
Hi, This seems to happen if the GPU acceleration not working properly or it might be the inswapper. It can also be another application holding onto the webcam and DLC cannot capture frames. Have a look if the below links help. I will also build the SW again and check if something changed on the repo. On the GitHub click on issues and search "webcam" You will see a couple of other fixes as well. First make sure your GPU acceleration is working for your GPU, then check inswapper model.. Let me know if you manage to resolve this...
github.com/hacksider/Deep-Live-Cam/issues/663
github.com/hacksider/Deep-Live-Cam/issues/649
You can download the inswapper or replacement version from the project page direct. Scroll down to download models.
github.com/hacksider/Deep-Live-Cam
As i enter the run Command this shows up
ModuleNotFoundError: No module named 'torch'
🤨
Hi, just checking if you managed to resolve this issue. If not can you confirm loading the required SW and building everything using Anaconda with the Python version etc. I use in the video? I have no Python installed on my Windows host directly and use Anaconda to supply Python inside the environment. Can you trace the steps again from about 2:00 onwards into the video. Let me know if you get it up and running. Thanks for reaching out.
'conda create -n dlc python=3.10.0" gives me the error:
"Error while loading conda entry point: conda-content-trust"
Hi, sounds like a faulty Anaconda install. Are you running the full Anaconda install in Windows? You can have a look at the link included and see if it helps with your issue. Let me know if you are up and running. stackoverflow.com/questions/78347799/why-my-anaconda-keeps-showing-error-while-loading-conda-entry-point
What's the role of visualstudio here? I plan to test it on my Linux system. The git repository as I observed has mostly python code only.
Hi, you need Visual Studio with C++ Build Tools selected. Without it the build will fail on Windows. Let me know if you get it up and running. Thanks for reaching out.
Sir, What's the GPU memory you used
Hi, I ran Deep-Live-Cam on my RTX 3060 with 12GB vram. Did you get yours up and running? Thanks for reaching out. Hope the video helped.
Thanks so much for the detailed tutorial, i really appreciate.
when i execute "python run.py --execution-provider cuda", the Deep Live Cam will pop out but when i click on the 'Start' or 'Live' Button i get the error messages below:
Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'cudnn_conv_algo_search': 'EXHAUSTIVE', 'device_id': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'has_user_compute_stream': '0', 'gpu_external_alloc': '0', 'enable_cuda_graph': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0'}, 'CPUExecutionProvider': {}}
find model: C:\Users\Administrator/.insightface\models\buffalo_l\1k3d68.onnx landmark_3d_68 ['None', 3, 192, 192] 0.0 1.0
Applied providers: ['CUDAExecutionProvider', 'CPUExecutionProvider'], with options: {'CUDAExecutionProvider': {'cudnn_conv_algo_search': 'EXHAUSTIVE', 'device_id': '0', 'cudnn_conv1d_pad_to_nc1d': '0', 'has_user_compute_stream': '0', 'gpu_external_alloc': '0', 'enable_cuda_graph': '0', 'gpu_mem_limit': '18446744073709551615', 'gpu_external_free': '0', 'gpu_external_empty_cache': '0', 'arena_extend_strategy': 'kNextPowerOfTwo', 'do_copy_in_default_stream': '1', 'cudnn_conv_use_max_workspace': '1', 'tunable_op_enable': '0', 'tunable_op_tuning_enable': '0', 'tunable_op_max_tuning_duration_ms': '0', 'enable_skip_layer_norm_strict_mode': '0'}, 'CPUExecutionProvider': {}}
But when i run the code using just "python run.py" well but it is hanging. please what could be the cause
onnxruntime error 7 please help
Hi, can you confirm you installed ffmpeg? Make sure you installed all the required SW as from 1:58 in the video. Check the ffmpeg installation from 4:53 in the video. You can also check out the below link if the issue persist. Thanks for reaching out. github.com/hacksider/Deep-Live-Cam/issues/241
Mine fail to build wheel for insightface
Hi, just checking if you managed to resolve. If not can you confirm loading the required SW especially MS Visio Studio Community with the selection I have in the video (2:00 minutes in)? Let me know if you managed to get this tool up and running. Thanks for reaching out.
@@stuffaboutstuff4045 thanks for checking in, I just logged into my RUclips account today. I already fixed that one and it's working but the real time deepfake function is not working
I have that that problem the days too
Hi, Please make sure you installed all the required SW first. If you have VS Studio etc. and problem persists please check if the below applies.
Hi, This seems to happen if the GPU acceleration not working properly or it might be the inswapper. It can also be another application holding onto the webcam and DLC cannot capture frames. Have a look if the below links help. I will also build the SW again and check if something changed on the repo. On the GitHub click on issues and search "webcam" You will see a couple of other fixes as well. First make sure your GPU acceleration is working for your GPU, then check inswapper model.. Let me know if you manage to resolve this...
github.com/hacksider/Deep-Live-Cam/issues/663
github.com/hacksider/Deep-Live-Cam/issues/649
You can download the inswapper or replacement version from the project page direct. Scroll down to download models.
github.com/hacksider/Deep-Live-Cam
"CudaCall CUDA failure 35: CUDA driver version is insufficient for CUDA runtime version"
Hi, just checking if you managed to resolve this issue. Sounds like you have the Nvidia driver and the CUDA Toolkit installed with a mismatch between the two. You can try running it with only Gaming driver. It should give you enough support as is. If you already have CUDA Toolkit on the PC then you should stick with the driver it installs. Let me know if you managed to resolve this issue. Thanks for reaching out.
Please do it live once..
😊That will be a bit longer video. Did you get yours up and running.
@@stuffaboutstuff4045 does it work for live though this install? thanks for the video!
@@stuffaboutstuff4045how to take cloud gpu
Hi, If you are still having issues with live webcam, Please make sure you have all the required SW and that each step in the video was followed. If you still have problems with webcam check if the below assists.
Hi, This seems to happen if the GPU acceleration not working properly or it might be the inswapper. It can also be another application holding onto the webcam and DLC cannot capture frames. Have a look if the below links help. I will also build the SW again and check if something changed on the repo. On the GitHub click on issues and search "webcam" You will see a couple of other fixes as well. First make sure your GPU acceleration is working for your GPU, then check inswapper model.. Let me know if you manage to resolve this...
github.com/hacksider/Deep-Live-Cam/issues/663
github.com/hacksider/Deep-Live-Cam/issues/649
You can download the inswapper or replacement version from the project page direct. Scroll down to download models.
github.com/hacksider/Deep-Live-Cam
@@stuffaboutstuff4045 i get this issue \models\inswapper_128_fp16.onnx failed:Protobuf parsing failed. any help?