Variational Tracking and Redetection for Closely-spaced Objects in Heavy Clutter

The non-homogeneous Poisson process (NHPP) is a widely used measurement model that allows for an object to generate multiple measurements over time. However, it can be difficult to track multiple objects efficiently and reliably under this NHPP model in scenarios with a high density of closely-spaced objects and heavy clutter. Therefore, based on the general coordinate ascent variational filtering framework, this paper presents a variational Bayes association-based NHPP tracker (VB-AbNHPP) that can efficiently track a fixed and known number of objects, perform data association, and learn object and clutter rates with a parallelisable implementation. Moreover, a variational localisation strategy is proposed, which enables rapid rediscovery of missed objects from a large surveillance area under extremely heavy clutter. This strategy is integrated into the VB-AbNHPP tracker, resulting in a robust methodology that can automatically detect and recover from track loss. This tracker demonstrates improved tracking performance compared with existing trackers in challenging scenarios, in terms of both accuracy and efficiency.