vsdirty 0.5.0

VapourSynth wrappers for denoising, masking and edge fixing

vs-dirty

A collection of VapourSynth wrappers and utility functions focused on advanced denoising, masking, and edge fixing.

Follow the docstring for more information about the parameters.

Installation

You can install vsdirty via pip:

pip install vsdirty

If you want to enable GPU acceleration for neural networks, install the package with the appropriate extra based on your hardware:

• NVIDIA (TensorRT):

  pip install vsdirty[nvidia]
  

• AMD (DirectML / ORT):

  pip install vsdirty[amd]
  

• OpenCL (Generic):

  pip install vsdirty[cl]
  

Or build from source:

git clone https://github.com/PingWer/vs-dirty
cd vs-dirty
pip install .[nvidia] # or .[amd], .[cl]

Dependencies

This package relies on several external VapourSynth plugins. Most of them are automatically downloaded via pip, but ensure they are available in your VapourSynth environment:

Plugin	Used For
vapoursynth-edgemasks	High-performance convolution masks (Kroon, Sobel, Kirsch, Prewitt)
vapoursynth-cas	Contrast Adaptive Sharpening
vapoursynth-bm3d	High-quality block-matching denoising
vapoursynth-mvtools	Motion interpolation and analysis
vapoursynth-nlm-ispc	Non-local means denoising
vapoursynth-akarin	Fast mathematical expression evaluation (Expr)
libbore	Enhancing edge contours (dirtyfixer.bore)

Note: Currently, libbore is not available via pip. You will need to download and install it manually from its GitHub repository.

License

MIT License