Joint Counting, Detection and Re-Identification for Multi-Object Tracking

TL;DR

CountingMOT is an end-to-end multi-object tracking framework that jointly models counting, detection, and re-identification, improving accuracy in crowded scenes by balancing detection and crowd density estimation.

Contribution

It introduces a novel joint modeling approach with mutual object-count constraints, enhancing tracking performance in crowded environments.

Findings

Achieves state-of-the-art MOTA scores on MOT16, MOT17, and MOT20 benchmarks.

Performs online and real-time tracking.

Effectively recovers missed detections and rejects false positives in crowded scenes.

Abstract

The recent trend in 2D multiple object tracking (MOT) is jointly solving detection and tracking, where object detection and appearance feature (or motion) are learned simultaneously. Despite competitive performance, in crowded scenes, joint detection and tracking usually fail to find accurate object associations due to missed or false detections. In this paper, we jointly model counting, detection and re-identification in an end-to-end framework, named CountingMOT, tailored for crowded scenes. By imposing mutual object-count constraints between detection and counting, the CountingMOT tries to find a balance between object detection and crowd density map estimation, which can help it to recover missed detections or reject false detections. Our approach is an attempt to bridge the gap of object detection, counting, and re-Identification. This is in contrast to prior MOT methods that either ignore the crowd density and thus are prone to failure in crowded scenes,or depend on local correlations to build a graphical relationship for matching targets. The proposed MOT tracker can perform online and real-time tracking, and achieves the state-of-the-art results on public benchmarks MOT16 (MOTA of 79.7), MOT17 (MOTA of 81.3%) and MOT20 (MOTA of 78.9%).