A time-weighted metric for sets of trajectories to assess multi-object tracking algorithms
This work provides an incremental improvement in metrics for multi-object tracking, addressing the need for more adaptable evaluation tools in computer vision and surveillance domains.
The paper tackles the problem of evaluating multi-object tracking algorithms by proposing a time-weighted metric for sets of trajectories, which extends an existing metric by including weights for costs at different time steps to increase flexibility for various applications and user preferences.
This paper proposes a metric for sets of trajectories to evaluate multi-object tracking algorithms that includes time-weighted costs for localisation errors of properly detected targets, for false targets, missed targets and track switches. The proposed metric extends the metric in [1] by including weights to the costs associated to different time steps. The time-weighted costs increase the flexibility of the metric [1] to fit more applications and user preferences. We first introduce a metric based on multi-dimensional assignments, and then its linear programming relaxation, which is computable in polynomial time and is also a metric. The metrics can also be extended to metrics on random finite sets of trajectories to evaluate and rank algorithms across different scenarios, each with a ground truth set of trajectories.