Tracking-by-Counting: Using Network Flows on Crowd Density Maps for Tracking Multiple Targets
This addresses the challenge of accurate multi-object tracking in crowded scenes for applications such as surveillance and biological analysis, representing a new paradigm rather than an incremental improvement.
The paper tackles the problem of multi-object tracking in crowded scenes, where traditional detection-based methods fail due to occlusions and high density, by introducing a tracking-by-counting paradigm that uses crowd density maps and network flow optimization to jointly handle detection, counting, and tracking, achieving promising results on public benchmarks across domains like people, cell, and fish tracking.
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.