Label videos for computer vision projects using PyQT5, pims, and SQLite.
Basics functionality implemented including class labeling, pan & zoom, SQL storage, and export in darknet (YOLO) format.
An internal web app to share repositories among teams is in development.
Make sure conda-forge is appended to your conda config channels before moving forward with installation.
conda config --append channels conda-forge
Due to the complexity of the dependencies, PyAV is not always the easiest Python package to install. The most straight-foward install is via conda-forge:
conda install av -c conda-forge
pip install PyQt5
Currently using the development version of pims due to a compatability issue.
pip install git+https://github.com/soft-matter/pims.git
Writing more tests is on the todo list. To run the tests that are available, use the command below.
python -m tests.test
- Implement more testing
- Color change feature
- Learn to make better README files