Heterogeneous Association Graph Fusion for Target Association in Multiple Object Tracking

TLDR

Tracking‑by‑detection is a common multi‑object tracking approach, yet detector failures caused by occlusion or pose often lead to tracking breakdown. The authors aim to mitigate these failures by constructing a heterogeneous association graph that fuses high‑level detections with low‑level image evidence, and by introducing adaptive weights to balance motion and appearance cues. They build track trees from the fused graph and solve them using a multiple‑hypotheses tracking framework enhanced with efficient pruning strategies. The resulting HAF tracker is less sensitive to parameter choices, easier to extend, and achieves state‑of‑the‑art performance on the MOT 2017 benchmark.

Abstract

Tracking-by-detection is one of the most popular approaches to tracking multiple objects in which the detector plays an important role. Sometimes, detector failures caused by occlusions or various poses are unavoidable and lead to tracking failure. To cope with this problem, we construct a heterogeneous association graph that fuses high-level detections and low-level image evidence for target association. Compared with other methods using low-level information, our proposed heterogeneous association fusion (HAF) tracker is less sensitive to particular parameters and is easier to extend and implement. We use the fused association graph to build track trees for HAF and solve them by the multiple hypotheses tracking framework, which has been proven to be competitive by introducing efficient pruning strategies. In addition, the novel idea of adaptive weights is proposed to analyze the contribution between motion and appearance. We also evaluated our results on the MOT challenge benchmarks and achieved state-of-the-art results on the MOT Challenge 2017.