========== bbox-utils

Utilities to easily convert between different bounding box formats (YOLO, XYWH, XYXY, etc.).

Description

You can find documentation for the project at here <https://bbox-utils.readthedocs.io/en/latest/>_.

2D Bounding Box Conversions |Colab Image Demo|

.. |Colab Image Demo| image:: https://colab.research.google.com/assets/colab-badge.svg :target: https://colab.research.google.com/drive/1NGiNhftz-hy671IHdOidkqgnaxaZ88Kn?usp=sharing

• List of points [top left, top right, bottom right, bottom left]
• XYWH: top left, width, height
• XYXY: top left, bottom right
• YOLO

3D Bounding Box Conversions |Colab PCD Demo|

.. |Colab PCD Demo| image:: https://colab.research.google.com/assets/colab-badge.svg :target: https://colab.research.google.com/drive/1NR9fR5hWHDtNcOyp8U0nPPjeuXR_auxd?usp=sharing

You can create a 3D bounding box with either:

• A center point, width, height, depth, and rotation
• The back-bottom-left point, width, height, depth, and rotation

You can convert between the two forms and also get a triangular polygon to use for plotting triangular meshes.

The majority of the 3D Bounding Box implementation comes from the bbox PyPI package <https://github.com/varunagrawal/bbox>_.

Visualizations You can use bbox-utils to visualize annotations within point clouds or images.

To use point clouds, you will need to install open3d <http://www.open3d.org/docs/release/getting_started.html> and plotly <https://plotly.com/python/getting-started/> with either::

pip3 install open3d plotly==4.14.3
pip install
# or
conda install -c open3d-admin open3d
conda install -c plotly plotly=4.14.3

At the time of writing this, open3d requires Python < 3.9

To use images, you will need to install OpenCV <https://opencv.org/>_::

pip3 install opencv-python
# or
conda install opencv -c conda-forge

Making Changes & Contributing

This project uses pre-commit_, please make sure to install it before making any changes::

pip install pre-commit
cd bbox-utils
pre-commit install

It is a good idea to update the hooks to the latest version::

pre-commit autoupdate

.. _pre-commit: http://pre-commit.com/

Note

This project has been set up using PyScaffold 4.0rc1. For details and usage information on PyScaffold see https://pyscaffold.org/.