Closed utility-aagrawal closed 8 months ago
At any given moment when you are iterating over the frames of your video, when you call the Tracker.update
method it should return a list with all the different objects that still exist.
So when you do
tracked_objects = tracker.update(detections)
you can know the amount of tracked objects at that instant by doing len(tracked_objects)
, and the coordinates of the bounding box of each tracked object in the current frame is just the TrackedObject.estimate
attribute of each TrackedObject
element in the tracked_objects
list.
for obj in tracked_objects:
print(f"bbox of object {obj.id}: {obj.estimate}")
I am not sure if this answered your question.
It does! Thanks so much, @aguscas !
Hi,
I am using norfair to track faces. I want to know how can I get how many distinct objects are there in my video, bbox coordinates for each object. What's the easiest way to fetch this information? I am looking at the code right now but wanted to post the question here in hope of getting a quick response :)
Thanks!