Is there a way to add class labels to PyTrackObjects?

[cedarwarman]

Hi, I am using BayesianTracker to track pollen tube growth and pollen grains in microscopy images. It's working great, but I am wondering if there's a way to add a class label to a PyTrackObject? For example, I have a series of bounding boxes that I'm converting to PyTrackObjects using the from_dict() method. For pollen, the bounding boxes can have classes like "ungerminated", "germinated", and "burst". When the tracking happens the classes are ignored, which is the desired behavior for me. However, once I have the track for a specific pollen grain, I'd like to be able to see what classes it had over time (e.g. at what timepoint did the class change from ungerminated to germinated or from germinated to burst). Can the "features" parameter be used for this purpose? Thanks for your help!

[quantumjot]

When you create the PyTrackObject using from_dict() you can add extra properties in there. For example:

Then, when you track with these objects, the extra metadata should appear in the properties dictionary for each resulting track:

[cedarwarman]

This works, thank you! I had to add the replace_nan=False flag to the btrack.BayesianTrack() method to_napari() for the plotting to work, e.g.:

data, properties, graph = tracker.to_napari(replace_nan=False)