Closed AdityaKulshrestha closed 1 year ago
Sorry for the late reply. You referenced an external tutorial so we can't help much there. Maybe you can try checking the validity of your detections/cost matrix, the error message seems to imply that. I will close this issue, let me know if you need more specific help.
I have been following a tutorial from this website Deepsort I however I am facing the following error
File "D:\Drive D\LearningML\Work\Experience.Digital\PeopleCounter\test_video.py", line 32, in
tracks = object_tracker.update_tracks(detections, frame=image)
File "C:\Users\91876\anaconda3\lib\site-packages\deep_sort_realtime\deepsort_tracker.py", line 228, in update_tracks
self.tracker.update(detections, today=today)
File "C:\Users\91876\anaconda3\lib\site-packages\deep_sort_realtime\deep_sort\tracker.py", line 97, in update
matches, unmatched_tracks, unmatched_detections = self._match(detections)
File "C:\Users\91876\anaconda3\lib\site-packages\deep_sort_realtime\deep_sort\tracker.py", line 151, in _match
) = linear_assignment.matching_cascade(
File "C:\Users\91876\anaconda3\lib\site-packages\deep_sort_realtime\deep_sort\linear_assignment.py", line 147, in matching_cascade
matchesl, , unmatched_detections = min_cost_matching(
File "C:\Users\91876\anaconda3\lib\site-packages\deep_sort_realtime\deep_sort\linear_assignment.py", line 65, in min_cost_matching
indices = np.vstack(linear_sum_assignment(cost_matrix)).T
ValueError: matrix contains invalid numeric entries