Supervised and Unsupervised Detections for Multiple Object Tracking in Traffic Scenes: A Comparative Study

In this paper, we propose a multiple object tracker, called MF-Tracker, that\nintegrates multiple classical features (spatial distances and colours) and\nmodern features (detection labels and re-identification features) in its\ntracking framework. Since our tracker can work with detections coming either\nfrom unsupervised and supervised object detectors, we also investigated the\nimpact of supervised and unsupervised detection inputs in our method and for\ntracking road users in general. We also compared our results with existing\nmethods that were applied on the UA-Detrac and the UrbanTracker datasets.\nResults show that our proposed method is performing very well in both datasets\nwith different inputs (MOTA ranging from 0:3491 to 0:5805 for unsupervised\ninputs on the UrbanTracker dataset and an average MOTA of 0:7638 for supervised\ninputs on the UA Detrac dataset) under different circumstances. A well-trained\nsupervised object detector can give better results in challenging scenarios.\nHowever, in simpler scenarios, if good training data is not available,\nunsupervised method can perform well and can be a good alternative.\n