Context and saliency aware correlation filter for visual tracking

Visual tracking in complex scenarios is a big challenge in the computer vision community. Due to correlation filter (CF) recently have achieved excellent results both on accuracy and robustness in visual tracking, many researchers have focused on incorporating different features for better represent the tracking target. However, CF-based trackers have poor ability to handle problem in many complex scenes with challenges like deformation, motion blur and background clutters. To overcome these defects, we propose a context and saliency aware CF for visual tracking (CSCF). Context information around the target of interest is introduced into correlation filters to strengthen the discriminative ability of CF, which can reduce the boundary effect and the influence of the background. Then the saliency feature map of the target is combined with CF to strengthen the ability to extract targets of interest from complex background. Experimental results show that the proposed method shows competitive performance on OTB dataset and UAV dataset compared to several other CF trackers.