Open Mr-Akbari opened 1 year ago
Any update on this? I was looking for the line where weighted sum of the Mahalanobis and Appearance based distance metric is taken. My video gets jammed periodically and the Mahalanobis distance metric thinks new object has entered. I want to give it less weight. Where did you find this line? Or did you write it yourself?
Hello Thanks for sharing your work. I want to concatenate iou distance and appearance distance in a weighted sum way, in first step association.
I done it by editing "tracker.py" :
# Associate confirmed tracks using appearance features. matches_a, unmatched_tracks_a, unmatched_detections = linear_assignment.matching_cascade( lambda * gated_metric + (1-lambda)* iou_matching.iou_cost, self.metric.matching_threshold, self.max_age, self.tracks, detections, confirmed_tracks)
But it is not working. I really appreciate it if you answer these questions. Thanks very much.
Any update on this? I was looking for the line where weighted sum of the Mahalanobis and Appearance based distance metric is taken. My video gets jammed periodically and the Mahalanobis distance metric thinks new object has entered. I want to give it less weight. Where did you find this line? Or did you write it yourself?
Hello Thanks for sharing your work. I want to concatenate iou distance and appearance distance in a weighted sum way, in first step association. I done it by editing "tracker.py" :
# Associate confirmed tracks using appearance features. matches_a, unmatched_tracks_a, unmatched_detections = linear_assignment.matching_cascade( lambda * gated_metric + (1-lambda)* iou_matching.iou_cost, self.metric.matching_threshold, self.max_age, self.tracks, detections, confirmed_tracks)
But it is not working. I really appreciate it if you answer these questions. Thanks very much.
I write it myself. Any update on this?
Hello Thanks for sharing your work. I want to concatenate iou distance and appearance distance in a weighted sum way, in first step association.
I done it by editing "tracker.py" :
# Associate confirmed tracks using appearance features. matches_a, unmatched_tracks_a, unmatched_detections = linear_assignment.matching_cascade( lambda * gated_metric + (1-lambda)* iou_matching.iou_cost, self.metric.matching_threshold, self.max_age, self.tracks, detections, confirmed_tracks)
But it is not working. I really appreciate it if you answer these questions. Thanks very much.