Visual Multiple-Object Tracking for Unknown Clutter Rate

摘要

In multi-object tracking applications, model parameter tuning is aprerequisite for reliable performance. In particular, it is difficult to knowstatistics of false measurements due to various sensing conditions and changesin the field of views. In this paper we are interested in designing amulti-object tracking algorithm that handles unknown false measurement rate.Recently proposed robust multi-Bernoulli filter is employed for clutterestimation while generalized labeled multi-Bernoulli filter is considered fortarget tracking. Performance evaluation with real videos demonstrates theeffectiveness of the tracking algorithm for real-world scenarios.