class_ids is not considered in the tracking

[CaterinaFabbri]

Hi, I tried the library and it seems that the tracker id can be shared between class_ids. If it's a multi object tracker, is it normal that the track id doesn't reset when it encounters different objects (ie with different class_ids)? Could you please clarify?

(
class_ids (numpy.ndarray or list): List of class_ids (int) corresponding to labels of the detected object. Default is None.

)

[adipandas]

Hello @CaterinaFabbri, yes it is not included in the current assignment cost functions.
Formally, if you check the class_ids and add penalty for assignment of one class to another, this could be addressed to some extent during track assignment optimization. I haven't implemented/tested that though. If you consider adding such a feature, you will have to update cost evaluations at below lines:

• See here for track assignment cost in Kalman Filter based trackers
• See here for track assignment cost in SORT.

I hope this helps.