Building OpenCV¶

Prerequisites¶

1. Install CUDA: https://docs.nvidia.com/cuda/cuda-installation-guide-linux/
2. Install cuDNN:
   

   $ sudo apt install libcudnn8 libcudnn8-dev
   

3. Install OpenCV prereqs:
   

   $ sudo apt install build-essential cmake git pkg-config libgtk-3-dev libavcodec-dev libavformat-dev libswscale-dev libv4l-dev libxvidcore-dev libx264-dev libjpeg-dev libpng-dev libtiff-dev gfortran openexr libatlas-base-dev python3-dev python3-numpy libtbb2 libtbb-dev libdc1394-22-dev libopenexr-dev libgstreamer-plugins-base1.0-dev libgstreamer1.0-dev libvtk7-dev libdc1394-dev
   

Build from source¶

1. From a working directory clone the contrib repository:
   

   $ git clone https://github.com/opencv/opencv_contrib.git
   

2. Download OpenCV release (currently 4.7.0) to working directory and untar:
   

   $ wget https://github.com/opencv/opencv/archive/refs/tags/4.7.0.tar.gz
   $ tar xf 4.7.0.tar.gz
   

3. Create build directory and enter it:
   

   $ mkdir biuld
   $ cd build
   

4. Configure build (NOTE: this will set the install directory to /opt/local, change as needed):
   

   $ cmake -D CMAKE_BUILD_TYPE=RELEASE -DCMAKE_INSTALL_PREFIX=/opt/local -DOPENCV_GENERATE_PKGCONFIG=ON -DOPENCV_EXTRA_MODULES_PATH=../..//opencv_contrib/modules -DBUILD_opencv_legacy=OFF -DWITH_CUDA=ON -DBUILD_TESTS=OFF -DBUILD_PERF_TESTS=OFF -DBUILD_EXAMPLES=OFF -DBUILD_opencv_apps=OFF ..
   

5. Build (this will take a while) then install:
   

   $ make -j32
   $ sudo make install
   

Miscellaneous¶

• To use the OpenCV Python bindings you need to set the PYTHONPATH to include the install directory under /opt/local:
  

  $ export PYTHONPATH="/opt/local/lib/python3.10/dist-packages"