Welcome to’s documentation! allows to track groups of up to 100 unmarked animals from videos recorded in laboratory conditions.

_images/fish_tracked.png _images/flies_tracked.png _images/14ants.png _images/mice.png _images/2fish.png

New release: v4

  • Works with Python 3.7 and Pytorch 1.10.0 and CUDA 10.2 or 11.3.

  • New horizontal GUI layout.

  • “Add setup points” feature allows to annotate groups of points in the frame that can be useful for analysis. These groups of points are stored together with the trajectories in the trajectories.npy and trajectories_wo_gaps.npy files.

  • Save trajectories as CSV files using the advanced parameters.

Check What’s new in v4 and join the users group to get announcements about new releases.

Start using

Check the Installation and requirements to find the best installation mode for your usage case.

Follow the instructions in the Quickstart to track the example video and get use to the workflow.

If you are unsure whether will work on your videos, check the Guidelines for good videos and the Gallery to see how our videos look like.

Our research using

Source code

The source code can be found at the Gitlab repository.

Check the code documentation Index for more information about different classes, functions and methods of


The data used in the article can be found in the Data section of this web page.


When using information from this web page please reference

Search in this webpage