A video face swap tool using insightface
- Install CUDA Toolkit 11.8
- Download cuDNN for Cuda 11.x
- For windows go
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v{YourVersion}
, copy thebin
,include
,lib
folders from cuDNN to the CUDA folder, and addC:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v{YourVersion}\bin
to path
- clone this project
pip install -e .
.- change the
torch
andtorchvision
inrequirements.txt
as you need, thenpip install -r requirements.txt
- Setup a model path, e.g.
temp
(note: use absolute path!) - download
inswapper_128.onnx
into your model path, from huggingface or here - download buffalo_l.zip and unzip under
temp/models/buffalo_l
; you folder now looks like:temp/models/buffalo_l/*.onnx
- download GFPGANv1.4.pth into your model path
- download detection_Resnet50_Final.pth into your model path
gfpgan/weights
- download parsing_parsenet.pth into your model path
gfpgan/weights
either use ROOPIY_MODEL_PATH={PathToYourModel} roopiy
or roopiy --model-path={PathToYourModel}
to set the model path
use roopiy
to run. If it does not work, run python -m roopiy
check roopiy --help
for construction