The vs-openvino plugin provides optimized pure CPU runtime for some popular AI filters.
To build, you will need OpenVINO and its dependencies.
Only Model Optimizer
and Inference Engine
are required.
You can download official Intel releases:
Or, you can use our prebuilt Windows binary releases from AmusementClub, our release has the benefit of static linking support.
Sample cmake commands to build:
cmake -S . -B build -G Ninja -D CMAKE_BUILD_TYPE=Release
-D CMAKE_MSVC_RUNTIME_LIBRARY=MultiThreaded
-D InferenceEngine_DIR=openvino/runtime/cmake
-D VAPOURSYNTH_INCLUDE_DIRECTORY="path/to/vapoursynth/include"
cmake --build build
cmake --install build --prefix install
You should find vsov.dll
(or libvsov.so) under install/bin
. You will also need Intel TBB (you can get
tbb.dll
from OpenVINO release). On windows, tbb.dll
must be placed under vapoursynth/plugins/vsov/
directory for vsov.dll
to find.
Prototype: core.ov.Model(clip[] clips, string network_path[, int[] overlap = None, int[] tilesize = None, string device = "CPU", bint builtin = 0, string builtindir="models", bint fp16 = False, function config = None, bint path_is_serialization = False])
Arguments:
clip[] clips
: the input clips, only 32-bit floating point RGB or GRAY clips are supported. For model specific input requirements, please consult our wiki.string network_path
: the path to the network in ONNX format.int[] overlap
: some networks (e.g. CNN) support arbitrary input shape where other networks might only support fixed input shape and the input clip must be processed in tiles. Theoverlap
argument specifies the overlapping (horizontal and vertical, or both, in pixels) between adjacent tiles to minimize boundary issues. Please refer to network specific docs on the recommended overlapping size.int[] tilesize
: Even for CNN where arbitrary input sizes could be supported, sometimes the network does not work well for the entire range of input dimensions, and you have to limit the size of each tile. This parameter specify the tile size (horizontal and vertical, or both, including the overlapping). Please refer to network specific docs on the recommended tile size.string device
: Specifies the device to run the inference on. Currently"CPU"
and"GPU"
are supported."GPU"
requires Intel graphics (Broadwell+ processors with Gen8+ integrated GPUs or Xe discrete GPUs) with compatible graphics driver and compute runtime.bint builtin
: whether to load the model from the VS plugins directory, see alsobuiltindir
.string builtindir
: the model directory under VS plugins directory for builtin models, default "models".bint fp16
: whether to quantize model to fp16 for faster and memory efficient computation.function config
: plugin configuration parameters. It must be a callable object (e.g. a function) with no positional arguments, and returns the configuration parameter in a dictionarydict
. The dictionary must use stringstr
for its key andint
,float
orstr
for its values. Supported parameters: CPU, GPU (the prefixKEY_
has to be removed). Example:config = lambda: dict(CPU_THROUGHPUT_STREAMS=2)
bint path_is_serialization
: whether thenetwork_path
argument specifies an onnx serialization of typebytes
.
When overlap
and tilesize
are not specified, the filter will internally try to resize the network to fit the input clips. This might not always work (for example, the network might require the width to be divisible by 8), and the filter will error out in this case.
The general rule is to either:
- left out
overlap
,tilesize
at all and just process the input frame in one tile, or - set all three so that the frame is processed in
tilesize[0]
xtilesize[1]
tiles, and adjacent tiles will have an overlap ofoverlap[0]
xoverlap[1]
pixels on each direction. The overlapped region will be throw out so that only internal output pixels are used.