Computer vision tools for computer vision engineers.
- Inprove documentation on installing on linux, apple silicon and windows
- Refactor bounding boxes to shapely geometry
First and foremost, we would like to express our deepest gratitude to Odd Industries for their incredible work on the first version of this library. Their dedication to coding and openly releasing the initial version has paved the way for the development of this project, which has since evolved and grown.
While the API for this release may not be backward compatible, the invaluable contributions from Odd Industries have helped shape the general concept of the OpenCV pipeline that we have embraced in this library. Their pioneering work has served as a solid foundation, enabling us to build upon and improve the project further.
As Entel Ocean we are happy to take this library to the furure
This package was created with Cookiecutter and the waynerv/cookiecutter-pypackage project template.
- Create a conda enviroment like:
conda create -n "cv310" python=3.10- Install this module
pip install -e .python setup.py sdist bdist_wheelpip install --upgrade cvutils-x.y.z-py3-none-any.whlcvutils testOdd Industries's cvutils had a many functionality with wide range of software engineering quality (documentation, abstraction, testing, updated requirements, etc). That same issue made a lot of work to release it open responsibly. At Entel Ocean, having access to previous code and engineer, we will be releasing only features that we are committed to use and support. If this repo opening takes traction, we'll open an issue where the community also could bring old features to present.