Tracking objects in video recording can be tedious. Existing software can be expensive, the learning curve can be steep, necessitate a powerful computer, and the workflow can not be easily adapted when the project evolved.
FastTrack is a free (as freedom) desktop tracking software, easy to install, easy to use, and performant. The software can be downloaded for Linux, macOS, and Windows, and a public API is available to embed the core of the software in any C++/Qt project.
Two main features are implemented in the software:
- An automatic tracking algorithm that can detect and track objects (conserving the objects' identities across the video recording).
- A manual review of the tracking where errors can be corrected rapidly and easily to achieve 100% accuracy.
Stable builds of the software are available for Linux, Mac, and Windows on the download page and the GitHub repository. Continuous builds triggered after each modification are available on the continuous download page.
User documentation can be found here with video tutorials, examples, and an extended description of the software. Developer documentation can be found here. It contains information to set up a development environment and the API of FastTrack.
If you want to contribute to FastTrack, please follow the guidelines. Any contribution or idea is welcome.
If you need help, please contact firstname.lastname@example.org or open an issue or a discussion on GitHub.
You can cite FastTrack Gallois, Benjamin, and Raphaël Candelier. "FastTrack: An open-source software for tracking varying numbers of deformable objects." PLoS computational biology 17.2 (2021): e1008697..