search

SamKhoze / ComfyUI-DeepFuze

DeepFuze is a state-of-the-art deep learning tool that seamlessly integrates with ComfyUI to revolutionize facial transformations, lipsyncing, Face Swapping, Lipsync Translation, video generation, and voice cloning.
https://cogidigm.com/
Apache License 2.0
326 stars 40 forks source link

Error occurred when executing DeepFuzeFaceSwap #5

Open Zakutu opened 5 months ago

Zakutu commented 5 months ago

Error occurred when executing DeepFuzeFaceSwap:

C:\ComfyUI_windows_portable\ComfyUI\output\deepfuze\faceswap_17187334906389506.mp4 could not be loaded with cv.

File "C:\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) File "C:\ComfyUI_windows_portable\ComfyUI\execution.py", line 81, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) File "C:\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) File "C:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-DeepFuze\nodes.py", line 1018, in faceswampgenerate return load_video_cv(faceswap_filename,0,'Disabled',512,512,0,0,1) File "C:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-DeepFuze\nodes.py", line 543, in load_video_cv (width, height, fps, duration, total_frames, target_frame_time) = next(gen) File "C:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-DeepFuze\nodes.py", line 462, in cv_frame_generator raise ValueError(f"{video} could not be loaded with cv.")

SamKhoze commented 5 months ago

Are you using windows machine? currently it is not supporting cuda environment, We are working on it

iordcalin commented 5 months ago

Same problem for me as well.I'm using the official workflow and your own video /image assets

SamKhoze commented 5 months ago

@iordcalin The error message suggests most likely due to your installation. Make sure OpenCV is installed, go to your comfyui manager use the pip install tool type opencv-python , and click ok. I recommend deleting your DeepFuze folder and following the installation instructions for Windows or Mac. Additional steps are required to make this work because of the nature of Lipsync Generation and Voice Cloning. If the error remains the same please let us know.

SamKhoze commented 5 months ago

@iordcalin & @Zakutu you will get the cv load error If you have selected MPS or CUDA in your device options, please note that MPS is for MAC and CUDA will be released later. For now, you can use it on CPU, which will take a longer time if you apply face enhancer and frame enhancers."

iordcalin commented 5 months ago

@SamKhoze CPU is selected in the DeepFuze FaceSwap,is there another way to set the device options? I installed opencv and still getting this error...

FLYleoYBQ commented 5 months ago

same question，When the opencv installation is complete, a message indicating that cv cannot be loaded is displayed

Rex-Liu1027 commented 5 months ago

same question，When the opencv installation is complete, a message indicating that cv cannot be loaded is display

Hi,is it compatibled with Linux？thx~

SamKhoze commented 5 months ago

@Rex-Liu1027 Yes, follow the Mac installation via conda env. Once it's installed, I would appreciate it if you could document it and share with me. I can add it to the README.md. It will work with CPU, still I'm trying to find a way how to enable CUDA installation, the code is written for CUDA.

Bortus-AI commented 4 months ago

I'm on windows and having the same error using CPU. I get the CV error like the others. opencv-python is already installed

SamKhoze commented 4 months ago

@Bortus-AI Upgrade your opencv check the error section on the readme file

Bortus-AI commented 4 months ago

@SamKhoze already upgraded opencv and followed the error section. I tried on a fresh install of comfui as well and same issue

Bortus-AI commented 4 months ago

@SamKhoze redownloaded comfyui again but using different python version and pytorch and used the comfyui manager to install deepfuze but I get the following error

[WinError 267] The directory name is invalid

#######################################################################
[ComfyUI-Manager] Starting dependency installation/(de)activation for the extension

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'imageio_ffmpeg']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'cython>=0.29.30']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'
 [ERROR] Failed to execute install/(de)activation script: ['C:\\Users\\bpvar\\OneDrive\\Desktop\\ComfyUI\\custom_nodes\\ComfyUI-DeepFuze', 'C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'torch>=2.1']
 / invalid version number '2.3.0+cu118'
[SKIP] Downgrading pip package isn't allowed: soundfile (cur=0.12.1)
[SKIP] Downgrading pip package isn't allowed: scikit-learn (cur=1.5.1)
[SKIP] Downgrading pip package isn't allowed: numba (cur=0.60.0)
[SKIP] Downgrading pip package isn't allowed: inflect (cur=7.3.1)
[SKIP] Downgrading pip package isn't allowed: tqdm (cur=4.66.4)
[SKIP] Downgrading pip package isn't allowed: anyascii (cur=0.3.2)

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'pyyaml>=6.0']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'
 [ERROR] Failed to execute install/(de)activation script: ['C:\\Users\\bpvar\\OneDrive\\Desktop\\ComfyUI\\custom_nodes\\ComfyUI-DeepFuze', 'C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'fsspec>=2023.6.0 # <= 2023.9.1 makes aux tests fail']
 / invalid version number '2023.6.0 # <= 2023.9.1 makes aux tests fail'
[SKIP] Downgrading pip package isn't allowed: aiohttp (cur=3.9.5)
[SKIP] Downgrading pip package isn't allowed: packaging (cur=24.1)

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'flask>=2.0.1']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'
        [SKIP] Downgrading pip package isn't allowed: umap-learn (cur=0.5.6)
[ERROR] Failed to execute install/(de)activation script: ['C:\\Users\\bpvar\\OneDrive\\Desktop\\ComfyUI\\custom_nodes\\ComfyUI-DeepFuze', 'C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'pandas>=1.4,<2.0']
 / invalid version number '1.4,<2.0'
[SKIP] Downgrading pip package isn't allowed: matplotlib (cur=3.9.1)
[SKIP] Downgrading pip package isn't allowed: coqpit (cur=0.0.17)

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'hangul_romanize']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'gruut[de,es,fr]==2.2.3']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'
                    [SKIP] Downgrading pip package isn't allowed: g2pkk (cur=0.1.2)
[SKIP] Downgrading pip package isn't allowed: einops (cur=0.8.0)
[SKIP] Downgrading pip package isn't allowed: transformers (cur=4.42.3)

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'unidecode>=1.3.2']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', '-m', 'pip', 'install', 'spacy[ja]>=3']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'

## ComfyUI-Manager: EXECUTE => ['C:\\Users\\bpvar\\AppData\\Local\\Programs\\Python\\Python311\\python.exe', 'install.py']

## Execute install/(de)activation script for 'C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze'
 Conda is not activated

install/(de)activation script failed: C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze

[ComfyUI-Manager] Startup script completed.
#######################################################################
To see the GUI go to: http://127.0.0.1:8188
got prompt
1 6
video h265-mp4
{'enhancer': 'codeformer', 'faceswap_model': 'inswapper_128_fp16', 'frame_enhancer': 'clear_reality_x4', 'face_detector_model': 'retinaface', 'reference_face_index': 0, 'face_mask_padding_left': 0, 'face_mask_padding_right': 0, 'face_mask_padding_bottom': 0, 'face_mask_padding_top': 0, 'device': 'cuda', 'frame_rate': 25.0, 'loop_count': 0, 'filename_prefix': 'deepfuze', 'pingpong': False, 'save_output': True, 'source_images': ['10', 0], 'target_images': ['11', 0], 'format': 'video/h265-mp4', 'pix_fmt': 'yuv420p10le', 'crf': 22, 'save_metadata': ['save_metadata', 'BOOLEAN', {'default': True}]}
C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-VideoHelperSuite\videohelpersuite\..\video_formats\h265-mp4.json
pix_fmt
crf
save_metadata
Output images were not of valid resolution and have had padding applied
['python', './run.py', '--frame-processors', 'face_swapper', '-s', 'C:\\Users\\bpvar\\OneDrive\\Desktop\\ComfyUI\\output\\deepfuze_00004_.png', '-t', 'C:\\Users\\bpvar\\OneDrive\\Desktop\\ComfyUI\\output\\deepfuze_00004.mp4', '--face-detector-model', 'retinaface', '-o', 'C:\\Users\\bpvar\\OneDrive\\Desktop\\ComfyUI\\output\\deepfuze\\faceswap_17214119937331405.mp4', '--face-swapper-model', 'inswapper_128_fp16', '--reference-face-position', '0', '--face-mask-padding', '0', '0', '0', '0', '--headless', '--execution-providers', 'cuda']
!!! Exception during processing!!! [WinError 267] The directory name is invalid
Traceback (most recent call last):
  File "C:\Users\bpvar\OneDrive\Desktop\ComfyUI\execution.py", line 151, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\bpvar\OneDrive\Desktop\ComfyUI\execution.py", line 81, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\bpvar\OneDrive\Desktop\ComfyUI\execution.py", line 74, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\bpvar\OneDrive\Desktop\ComfyUI\custom_nodes\ComfyUI-DeepFuze\nodes.py", line 864, in faceswampgenerate
    result = subprocess.run(command,cwd="ComfyUI/custom_nodes/ComfyUI-DeepFuze",stdout=subprocess.PIPE)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\bpvar\AppData\Local\Programs\Python\Python311\Lib\subprocess.py", line 548, in run
    with Popen(*popenargs, **kwargs) as process:
         ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\bpvar\AppData\Local\Programs\Python\Python311\Lib\subprocess.py", line 1026, in __init__
    self._execute_child(args, executable, preexec_fn, close_fds,
  File "C:\Users\bpvar\AppData\Local\Programs\Python\Python311\Lib\subprocess.py", line 1538, in _execute_child
    hp, ht, pid, tid = _winapi.CreateProcess(executable, args,
                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
NotADirectoryError: [WinError 267] The directory name is invalid
Bortus-AI commented 4 months ago

happens with cpu or cuda

Bortus-AI commented 4 months ago

@SamKhoze already upgraded opencv and followed the error section. I tried on a fresh install of comfui as well and same issue

Anything I can try to fix this? Already downloaded your new commit from yesterday but no change

Bortus-AI commented 4 months ago

@SamKhoze interesting enough if I set the frame load cap to a specific number I get no errors with cpu or cuda.

Bortus-AI commented 4 months ago

Reference in new issue

@SamKhoze interesting enough if I set the frame load cap to a specific number I get no errors with cpu or cuda.

Its a bit hit and miss. Worked for one video but now the next 3 videos have the CV error again but then the 4th worked.