Skip to content

About

What is Roboflow Trackers?

Roboflow Trackers is an open-source Python library that provides clean-room implementations of leading multi-object tracking (MOT) algorithms: SORT, ByteTrack, and OC-SORT. The library is designed to plug into any object detection model through the supervision library, giving you a single tracker.update(detections) call to add tracking to an existing detection pipeline.

Every algorithm is implemented from scratch following the original papers, with a shared interface and consistent parameter naming. The library ships with built-in evaluation tools for HOTA, IDF1, and MOTA metrics, a CLI for running trackers on video files, and dataset download utilities for standard MOT benchmarks.

Who Built It?

Roboflow Trackers is built and maintained by the Roboflow team. Roboflow builds tools and infrastructure for computer vision, from dataset management and model training to deployment and monitoring. The trackers library is part of the broader Roboflow open-source ecosystem that includes supervision, inference, and RF-DETR.

A full list of contributors is available on the GitHub contributors page.

Package Provenance

Roboflow Trackers is distributed as the trackers package on PyPI. The source code is hosted on GitHub under the Apache 2.0 license.

Detail Value
PyPI trackers
Repository roboflow/trackers
License Apache License 2.0
Python 3.10+

Support

If you run into a bug, have a feature request, or need help with integration:

Contributing

Contributions are welcome. To get started:

  1. Fork the repository and clone it locally.
  2. Install development dependencies with uv sync.
  3. Create a feature branch, make your changes, and add tests.
  4. Run pre-commit run --all-files to check formatting, linting, and type checks.
  5. Open a pull request against the develop branch.

See the repository's contributing guidelines for full details on code style, commit conventions, and the review process.

License

Roboflow Trackers is released under the Apache License 2.0. You are free to use, modify, and distribute the library in both commercial and non-commercial projects, subject to the terms of the license.

Citing Roboflow Trackers

If you use Roboflow Trackers in academic work or publications, please cite the library:

@software{roboflow_trackers,
  author  = {{Roboflow}},
  title   = {Roboflow Trackers},
  url     = {https://github.com/roboflow/trackers},
  year    = {2025},
  license = {Apache-2.0}
}

For citations of specific tracking algorithms, see the Reference section on each tracker page: SORT, ByteTrack, OC-SORT.

Learn More About Roboflow

Visit roboflow.com to learn about the full Roboflow platform for building, deploying, and managing computer vision applications.