rebrand opencv-python-headless as opencv-python
If you want to install OpenCV as a Python package, or include it as a dependency in another, (unless you want to build your own,) you'll first have to decide on a flavour (and top-level package name):
Select the correct package for your environment:
There are four different packages (see options 1, 2, 3 and 4 below) and you should SELECT ONLY ONE OF THEM. Do not install multiple different packages in the same environment. There is no plugin architecture: all the packages use the same namespace (
cv2). If you installed multiple different packages in the same environment, uninstall them all withpip uninstalland reinstall only one package.
a. Packages for standard desktop environments (Windows, macOS, almost any GNU/Linux distribution)
- Option 1 - Main modules package:
pip install opencv-python- Option 2 - Full package (contains both main modules and contrib/extra modules):
pip install opencv-contrib-python(check contrib/extra modules listing from OpenCV documentation)
b. Packages for server (headless) environments (such as Docker, cloud environments etc.), no GUI library dependencies
These packages are smaller than the two other packages above because they do not contain any GUI functionality (not compiled with Qt / other GUI components). This means that the packages avoid a heavy dependency chain to X11 libraries and you will have for example smaller Docker images as a result. You should always use these packages if you do not use
cv2.imshowet al. or you are using some other package (such as PyQt) than OpenCV to create your GUI.
- Option 3 - Headless main modules package:
pip install opencv-python-headless- Option 4 - Headless full package (contains both main modules and contrib/extra modules):
pip install opencv-contrib-python-headless(check contrib/extra modules listing from OpenCV documentation)
This creates a problem for dependent packages, though: Whatever OpenCV flavour you choose to depend on, your dependents will also rely on that particular choice. To make matters worse, pip does not consider these flavours as conflict when installed in parallel (despite all sharing the same submodule/subpackage name cv2).
So practically, if you depend on some OpenCV-dependent packages, but want to avoid the X11 dependencies, any of your dependencies could drag them back in if it chose to require opencv-python instead of opencv-python-headless.
If, however, you know for certain that none of them actually need X11, and might thus easily have been chosen headless, then this package is for you. It merely re-brands opencv-python-headless as opencv-python by requiring the former but providing the latter in an otherwise empty package.
You can install this package prior to other dependencies and thus block any subsequent installation of the actual prebuilt opencv-python. (But to make this work including a mirrored version number, opencv-python-headless needs to be installed early, too. Thus, installing this package will also install opencv-python-headless if not already present.)
# git clone and chdir, then pip:
pip install .
# or, pip via git:
pip install opencv-python@git+https://github.com/bertsky/wrap_opencv-python-headless