Where do you store the tracklets of each pedestrian ?

VisDrone / DroneCrowd

Drone-based Joint Density Map Estimation, Localization and Tracking with Space-Time Multi-Scale Attention Network

163 stars 32 forks source link

Where do you store the tracklets of each pedestrian ? #15

Open MounirB opened 2 years ago

MounirB commented 2 years ago

In mytest.py it seems to me that calc_trkpt helps you to find the next detection point using tracking. Are you storing tracklets somewhere to help you rebuild the complete track of each pedestrian ? How can we reproduce the tracking that is illustrated in the DroneCrowd README.md ?

MounirB commented 2 years ago

At this stage of my understanding, it seems that the the tracking information is used to localize pedestrians at the t moment relying on their t-1 moment positions ? Am I right ? I didn't find in your project how do you draw long tracks. In your paper, you said that you used Social LSTM and Min Cost Flow to compute these tracks. Do you use them in tandem or you compared both of the methods ?