Abstract

This paper presents a novel introduction of online targetspecifific metric learning in track fragment (tracklet) association by network flflow optimization for long-term multiperson tracking. Different from other network flflow formulation, each node in our network represents a tracklet, and each edge represents the likelihood of neighboring tracklets belonging to the same trajectory as measured by our proposed affifinity score. In our method, targetspecifific similarity metrics are learned, which give rise to the appearance-based models used in the tracklet affifinity estimation. Trajectory-based tracklets are refifined by using the learned metrics to account for appearance consistency and to identify reliable tracklets. The metrics are then relearned using reliable tracklets for computing tracklet affifinity scores. Long-term trajectories are then obtained through network flflow optimization. Occlusions and missed detections are handled by a trajectory completion step. Our method is effective for long-term tracking even when the targets are spatially close or completely occluded by others. We validate our proposed framework on several public datasets and show that it outperforms several state of art methods