How to build cv2 binding

Question

I have followed the opencv doc for creating my own bindings.

I run the file gen2.py which generated some header files:

./
../
pyopencv_generated_enums.h
pyopencv_generated_funcs.h
pyopencv_generated_include.h
pyopencv_generated_modules_content.h
pyopencv_generated_modules.h
pyopencv_generated_types_content.h
pyopencv_generated_types.h
pyopencv_signatures.json

How should I build this? I tried running cmake CMakeLists.txt directly in /opencv/modules/python but some defines were not found.

and I tried on re-building running cmake CMakeLists.txt in /opencv/ , where I got:

  FATAL: In-source builds are not allowed.

         You should create a separate directory for build files.

I think both approaches were quite wrong, but I haven't found any doc explaining how to build using the generated headers.

I'm using opencv 4.5.5 which I cloned and built.

EDIT

I found this in /source/opencv/modules/python/bindings/CMakeLists.txt :

string(REPLACE ";" "\n" opencv_hdrs_ "${opencv_hdrs}")
file(WRITE "${CMAKE_CURRENT_BINARY_DIR}/headers.txt" "${opencv_hdrs_}")
add_custom_command(
    OUTPUT ${cv2_generated_files}
    COMMAND "${PYTHON_DEFAULT_EXECUTABLE}" "${PYTHON_SOURCE_DIR}/src2/gen2.py" "${CMAKE_CURRENT_BINARY_DIR}" "${CMAKE_CURRENT_BINARY_DIR}/headers.txt"
    DEPENDS "${PYTHON_SOURCE_DIR}/src2/gen2.py"
            "${PYTHON_SOURCE_DIR}/src2/hdr_parser.py"
            # not a real build dependency (file(WRITE) result): ${CMAKE_CURRENT_BINARY_DIR}/headers.txt
            ${opencv_hdrs}
    COMMENT "Generate files for Python bindings and documentation"
)

I guess I have to add the path to my headers in ${CMAKE_CURRENT_BINARY_DIR}/headers.txt just not sure how to get the value of CMAKE_CURRENT_BINARY_DIR

EDIT 2

As proposed by @berak, I tried CMAKE_EXTRA_MODULES_PATH=my/path this seems to get me really close to what I need.

It compiles my sources and generates the .so files as libopencv_xxx.so and saves them in my /usr/local/lib

(when running nm on it I see my class and method)

BUT! when I import cv2 in a script I don't achieve to load my module. I tried print(cv2.__dict__) and grep ed on it, but it's not there.

Any clue on how to add the compiled modules into my python-opencv installation?

Answer 1

There's not much documentation on how to create a bind (at least I only found this , which is very helpful but doesn't have all the information).

First, your module must use the following tree structure (I did everything with cmake):

.
src
└── modules
    └── your_module
        ├── CMakeLists.txt
        ├── include
        │   └── opencv2
        │       └── your_module_bind_lib.hpp
        └── src
            └── your_module_bind_lib.cpp

An example of my CMakeLists.txt (this of course may change according to your project):

set(the_description "yourModule LIB")
ocv_define_module(your_module opencv_imgproc WRAP python)
include_directories(${OpenCV_INCLUDE_DIRS})

Then I cloned OpenCV with

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

Make a build folder

mkdir wherever/build

And run the following command for cmake

cd wherever/build
cmake -DOPENCV_EXTRA_MODULES_PATH=/project_from_previous_tree/src/modules/ -D BUILD_EXAMPLES=OFF -D BUILD_opencv_apps=OFF -D BUILD_DOCS=OFF -D BUILD_PERF_TESTS=OFF -D BUILD_TESTS=OFF -D CMAKE_INSTALL_PREFIX=/usr/local/  your_opencv_location/opencv/
make -j8
make install

In that command the only "special" thing was DOPENCV_EXTRA_MODULES_PATH (thanks to @berak for the idea).

And then? what happens in python level?

Make sure your python installation actually points to the built opencv, or use something like:

import sys
sys.path.append('/usr/local/lib/python3.8/site-packages/cv2/python-3.8')
import cv2

where /usr/local/lib/python3.8/site-packages/cv2/python-3.8' is where the generated .so is located in my PC.

With this you will be able to use things like

cv2.my_cpp_function()
cv2.my_cpp_class.some_method()
...

Where my_cpp stuff was declared and defined in the module project following the doc and can "easily" have OpenCV typed objects as parameters.

Et voila, now you can use your c++ functions in python.

Notice that this solution does not do any modification to the opencv directory cloned from Git.

Answer 2

You're not supposed to run gen2.py manually, cmake will do this for you at some stage, internally.

since you seem to be on some linux, rather follow the build steps here , this will also generate python bindings, the gist of it is:

• make a special "build folder" and cd into it (dont try to build in the src folder, you'll never be able to clean anything up)

• run cmake <some args> path_to_src_folder (or better even - cmake-gui) , check if python3 shows up in the "To be build" section of the output

• run make install