Robust Multi-Object Tracking Based On Higher-Order Graph

Multi-object tracking is the process of locating multiple moving objects of interest and forming the trajectory of the target motion in a given video sequence.It is produced in the analysis and application of the moving target image,and soon became one of the important contents of video analysis in the field of computer vision,which is essentially a time-space combined target state estimation based on the initial state of the captured target and the target characteristics obtained through feature extraction.Multi-object tracking has a wide range of applications in intelligent monitoring,behavior analysis and other fields.Most multi-object tracking algorithms are developed within the tracking-by-detection framework that compute the pairwise appearance similarities between detection responses or tracklets within a limited temporal window,and associate with each other,which causes less effective in handling long-term occlusions or distinguishing spatially close targets with similar appearance in crowded scenes.In this paper,we propose an algorithm that formulates the multi-object tracking task as one to construct a global higher-order graph,which consist of a group of vertices that are inter-connected with a set of hyperedges with high affinity values.The appearance and motion similarities among multiple tracklets across the spatio-temporal domain are considered globally by exploiting high-order similarities rather than pairwise ones,thereby facilitating distinguish spatially close targets with similar appearance.In addition,the algorithm segments the whole video sequence for Iterative processing,which is beneficial to improve the speed of computation and meet real-time demand.Extensive experiments on various challenging datasets of UA-DETRAC,result in state-of-the-art performance through qualitative and quantitative analysis.