https://github.com/opencv/opencv/ https://github.com/opencv/opencv_contrib

go and checkout both repos to some version - e.g.: git checkout 4.9.0

sudo apt install pkg-config libgtk-3-dev

https://dev.to/naruaika/using-opencv-on-windows-subsystem-for-linux-1ako

and check version: nvidia-smi.exe

Possible arg for build:

• -D OPENCV_EXTRA_MODULES_PATH=~/opencv_with_contrib/opencv_contrib/modules
• -D CUDNN_LIBRARY=/usr/lib64/libcudnn.so.8
• -D CUDNN_INCLUDE_DIR=/usr/include
• -D PYTHON_EXECUTABLE=$HOME/Researches/venv/bin/python
• -D PYTHON3_NUMPY_INCLUDE_DIRS=$HOME/Researches/venv/lib64/python3.8/site-packages/numpy/core/include
• -D CMAKE_INSTALL_PREFIX=$HOME/Researches/venv

https://developer.nvidia.com/cuda-gpus

• RTX 20x0: 7.5
• RTX 30x0: 8.6

Set the right opencv_contrib path

cmake -D CMAKE_BUILD_TYPE=RELEASE -D OPENCV_GENERATE_PKGCONFIG=ON -D ENABLE_PRECOMPILED_HEADERS=OFF -D BUILD_opencv_legacy=OFF -D CUDA_ARCH_BIN=7.5 -D WITH_CUDA=ON -D WITH_CUDNN=ON -D OPENCV_DNN_CUDA=ON -D ENABLE_FAST_MATH=1 -D CUDA_FAST_MATH=1 -D WITH_CUBLAS=1 -D OPENCV_ENABLE_NONFREE=ON -D OPENCV_EXTRA_MODULES_PATH=../opencv_contrib/modules ../opencv make -j4 #increasing the number will make building faster. Maximum value can be found by running nproc.