Tracking-by-Counting: Using Network Flows on Crowd Density Maps for Tracking Multiple Targets

TL;DR

This paper introduces a novel multi-object tracking approach called tracking-by-counting, which leverages crowd density maps and network flow optimization to improve tracking accuracy in crowded scenes where traditional detection-based methods struggle.

Contribution

The paper presents a new paradigm for multi-object tracking that jointly models detection, counting, and tracking using crowd density maps and network flow, addressing challenges in crowded environments.

Findings

Effective in crowded scenes with occlusions and high density

Outperforms traditional detection-based tracking methods

Validated on diverse benchmarks including people, cell, and fish tracking

Abstract

State-of-the-art multi-object tracking~(MOT) methods follow the tracking-by-detection paradigm, where object trajectories are obtained by associating per-frame outputs of object detectors. In crowded scenes, however, detectors often fail to obtain accurate detections due to heavy occlusions and high crowd density. In this paper, we propose a new MOT paradigm, tracking-by-counting, tailored for crowded scenes. Using crowd density maps, we jointly model detection, counting, and tracking of multiple targets as a network flow program, which simultaneously finds the global optimal detections and trajectories of multiple targets over the whole video. This is in contrast to prior MOT methods that either ignore the crowd density and thus are prone to errors in crowded scenes, or rely on a suboptimal two-step process using heuristic density-aware point-tracks for matching targets.Our approach yields promising results on public benchmarks of various domains including people tracking, cell tracking, and fish tracking.