SCGTracker: Spatio-temporal correlation and graph neural networks for multiple object tracking

Yajuan Zhang, Yongquan Liang, Jiaxu Leng, Zhihui Wang

Abstract: Highlights•Spatio-temporal correlation is modeled based on historical trajectories.•Undetected objects can be tracked with spatio-temporal correlation constraints.•Appearance and motion information are used to select graph nodes for aggregation.•Secondary matching enhances target association between detections and tracklets.