Correlation Filter of Multiple Candidates Match for Anti-Obscure Tracking in Unmanned Aerial Vehicle Scenario

Due to the complexity of Unmanned Aerial Vehicle (UAV) target tracking scenarios, drift caused by occlusion is common and has no suitable solution. In this paper, an occlusion-resistant algorithm based on correlated filter model proposed. First, instead traditional that uses single template matching locate target, we finding optimal match multiple candidates templates matching. Then, in order increase accuracy matching, use self-attentive mechanism for feature enhancement. We experiment our proposed datasets OTB100 UAV123, respectively, results show outperforms filtered model. addition, have also tested anti-occlusion performance some video sequences which occluded. The a certain resistance occlusion, especially UAV scenario.