PERCEPTION ENHANCED FRAME FOR VISUAL OBJECT TRACKING

Deep trackers which based on pre-trained network trained on object detection datasets, have also shown great potentials in visual object tracking. However, the gap between object detection and object tracking is non-negligible. And the fixed template with the initial target feature during tracking in some previous deep trackers greatly limit the performance of the trackers. Therefore, we propose a perception enhanced frame (PEF) to exploit the target-aware features which can better recognize the target from background and update the template features through response map. Our PEF network takes advantage of the fully connected network with mask loss to select target-aware feature channels, and updates the template to enhance the robustness, which enables our trackers to reduce the deep features, enhances the discriminative ability, and ensures the diversity of comparison template. Experimental results on three popular datasets show that our method get superior performance than the state-of-the-art trackers in terms of accuracy and speed.