Spatio-temporal joint aberrance suppressed correlation filter for visual tracking

Abstract The discriminative correlation filter (DCF)-based tracking methods have achieved remarkable performance in visual tracking. However, the existing DCF paradigm still suffers from dilemmas such as boundary effect, degradation, and aberrance. To address these problems, we propose a spatio-temporal joint aberrance suppressed regularization (STAR) tracker under unified framework of response map. Specifically, dynamic regularizer is introduced into to alleviate effect simultaneously. Meanwhile, an exploited reduce interference background clutter. proposed STAR model effectively optimized using alternating direction method multipliers (ADMM). Finally, comprehensive experiments on TC128, OTB2013, OTB2015 UAV123 benchmarks demonstrate that achieves compelling compared with state-of-the-art (SOTA) trackers.