libCBDetect 1.0.4

Python bindings for a C++ Checkerboard detector

Libcbdetect

• Unofficial implemention of libcbdetect in C++ based on https://github.com/ftdlyc/libcbdetect.

Libdetect is fully automatic sub-pixel checkerboard / chessboard / deltille pattern detection. The algorithm autmatically extracts corners to sub-pixel accuracy and combines them to (rectangular) checkerboards / chessboard-like.

---

Development

0. Setting up Windows environment

• Install Visual Studio Build Tools:

  1. Download the "Visual Studio Build Tools" installer from Microsoft (link)
  2. During installation, select "Desktop development with C++"

---

1. Installing local dependencies

1. OpenCV

1. Windows (prebuilt binaries)

   1. Download the latest Windows release ZIP from the OpenCV GitHub releases page.
   2. Extract it (e.g., to C:\).
   3. Set OPENCV_DIR in System > Environment Variables to the C:\opencv\build folder. (Integrating OpenCV with Visual Studio C++ Projects on Windows)

   [!TIP] Or use the following command in a PowerShell with administrator privileges to do everything at once

   $ver='4.11.0';$exe="$env:TEMP\opencv.exe";Invoke-WebRequest -Uri "https://github.com/opencv/opencv/releases/download/$ver/opencv-$ver-windows.exe" -OutFile $exe; Start-Process -FilePath $exe -ArgumentList "-y","-oC:\" -NoNewWindow -Wait; Remove-Item $exe; $base='C:\opencv'; $bin="$base\build\x64\vc16\bin"; $cfg="$base\build\x64\vc16\lib"; [Environment]::SetEnvironmentVariable('Path',[Environment]::GetEnvironmentVariable('Path','Machine')+';'+$bin,'Machine'); [Environment]::SetEnvironmentVariable('OpenCV_DIR',$cfg,'Machine'); $env:Path+=';'+$bin
   

2. Linux (Ubuntu/Debian)

   sudo apt update
   sudo apt install libopencv-dev -y
   

   This installs both runtime and development headers/libraries for OpenCV 4.x. (How to Install opencv in C++ on Linux?—GeeksforGeeks)

3. macOS (Homebrew)

   brew update
   brew install opencv
   

   After installation, headers are in /usr/local/include/opencv4, libs in /usr/local/lib. (Getting started with Opencv with C++ - Nikhil Ramesh)

2. pybind11

1. Windows

   pip install "pybind11[global]"
   

2. Linux (Ubuntu/Debian)

   sudo apt update
   sudo apt install -y python3-pybind11
   

3. macOS (Homebrew)

   brew install pybind11
   

---

2. Creating a Python virtual environment

[!NOTE] This environment will be used for building the project.

Linux & macOS (bash/zsh)

python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip wheel build pybind11 cibuildwheel

Windows (PowerShell)

py -3.10 -m venv .venv
.venv\Scripts\Activate.ps1
python -m pip install --upgrade pip wheel build pybind11 cibuildwheel

---

3. Building the package

3-a Local build

python -m build

scikit-build-core invokes CMake + pybind11 to compile the C++ extension for your Python, then packages:

dist/
  ├─ libCBDetect-0.1.0-cp313-cp313-macosx_15_0_arm64.whl
  └─ libCBDetect-0.1.0.tar.gz

Windows only: Repair the wheel with delvewheel to include all required DLLs:

pip install delvewheel
$wheel = Get-ChildItem -Path .\dist\*.whl | Sort-Object LastWriteTime -Descending | Select-Object -First 1
delvewheel repair $wheel.FullName -w .\dist\repaired\

Install:

pip install dist/checkerboard-*.whl --force-reinstall

Windows: Use the repaired wheel:

pip install dist\repaired\*.whl --force-reinstall

3-b Editable installation for rapid iteration

pip install -e .[dev]

Builds the extension once and leaves the Python files linked to your working copy, so every code change is picked up immediately.

---

Tips

If using CLion (on Windows), change the default toolchain in the CMake Profiles to Visual Studio.

---

Original CPP code:

https://github.com/ftdlyc/libcbdetect

Reference Papers

[1] Geiger, A., Moosmann, F., Car, Ö., & Schuster, B. (2012, May). Automatic camera and range sensor calibration using a single shot. In Robotics and Automation (ICRA), 2012 IEEE International Conference on (pp. 3936-3943). IEEE.
[2] Schönbein, M., Strauß, T., & Geiger, A. (2014, May). Calibrating and centering quasi-central catadioptric cameras. In Robotics and Automation (ICRA), 2014 IEEE International Conference on (pp. 4443-4450). IEEE.
[3] Placht, S., Fürsattel, P., Mengue, E. A., Hofmann, H., Schaller, C., Balda, M., & Angelopoulou, E. (2014, September). Rochade: Robust checkerboard advanced detection for camera calibration. In European Conference on Computer Vision (pp. 766-779). Springer, Cham.
[4] Ha, H., Perdoch, M., Alismail, H., Kweon, I. S., & Sheikh, Y. (2017, October). Deltille Grids for Geometric Camera Calibration. In 2017 IEEE International Conference on Computer Vision (ICCV) (pp. 5354-5362). IEEE.
[5] Duda, A., & Frese, U. (2018, September). Accurate Detection and Localization of Checkerboard Corners for Calibration. In British Machine Vision Conference (BMCV), 2018.
[6] Sels, S., Ribbens, B., Vanlanduit, S., & Penne, R. (2019). Camera Calibration Using Gray Code. In Sensors (Vol. 19, Issue 2, p. 246). MDPI AG. https://doi.org/10.3390/s19020246 [7] Hillen, M., De Boi, I., De Kerf, T., Sels, S., Cardenas De La Hoz, E., Gladines, J., Steenackers, G., Penne, R., & Vanlanduit, S. (2023). Enhanced Checkerboard Detection Using Gaussian Processes. Mathematics, 11(22), 4568. https://doi.org/10.3390/math11224568