So I'm working on this project on which I should detect moving object throughout a street lamp view camera, and for that I need to save tracking id of each detected object in movement. I successfully programmed the detection and the movement part using yolo and using a threshold on velocity computed by an optical flow algorithm. The goal is to write in a csv file , the bbox coordinates, the tracking id of each detected object its corresponding velocity.
And my problem is I don't know how to extract tracking id out of each moving detected object and save it in a csv file at the same time of extracting the velocity and testing wether the object should be tracked or not. I can't infer deep sort and specify yolo, since that will remove the optical flow constraint.
Peac12
commented
2 years ago
Hello,
So I'm working on this project on which I should detect moving object throughout a street lamp view camera, and for that I need to save tracking id of each detected object in movement. I successfully programmed the detection and the movement part using yolo and using a threshold on velocity computed by an optical flow algorithm. The goal is to write in a csv file , the bbox coordinates, the tracking id of each detected object its corresponding velocity.
And my problem is I don't know how to extract tracking id out of each moving detected object and save it in a csv file at the same time of extracting the velocity and testing wether the object should be tracked or not. I can't infer deep sort and specify yolo, since that will remove the optical flow constraint.
Thank you in advance.