To tackle problems arising from unexpected camera motions in unmanned aerial vehicles (UAVs), we propose a three-mode ensemble tracker where each mode specializes in distinctive situations. The proposed ensemble tracker is composed of appearance-based tracking mode, homography-based tracking mode, and momentum-based tracking mode. The appearance-based tracking mode tracks a moving object well when the UAV is nearly stopped, whereas the homography-based tracking mode shows good tracking performance under smooth UAV or object motion. The momentum-based tracking mode copes with large or abrupt motion of either the UAV or the object. We evaluate the proposed tracking scheme on a widely-used UAV123 benchmark dataset. The proposed motion-aware ensemble shows a 5.3% improvement in average precision compared to the baseline correlation filter tracker, which effectively employs deep features while achieving a tracking speed of at least 80fps in our experimental settings. In addition, the proposed method outperforms existing real-time correlation filter trackers.

为了解决无人飞行器（UAV）中意外的摄像机运动引起的问题，我们提出了一种三模式集成跟踪器，其中每种模式都专门针对不同的情况。提出的整体跟踪器由基于外观的跟踪模式，基于单应性的跟踪模式和基于动量的跟踪模式组成。当无人机接近停止时，基于外观的跟踪模式可以很好地跟踪运动的对象，而基于单应性的跟踪模式在平滑的无人机或对象运动下显示出良好的跟踪性能。基于动量的跟踪模式可以应对无人机或物体的大幅度运动或突然运动。我们在广泛使用的UAV123基准数据集上评估提出的跟踪方案。与基线相关滤波器跟踪器相比，拟议的运动感知集合显示出平均精度提高了5.3％，在我们的实验设置中，它有效地利用了深层功能，同时实现了至少80fps的跟踪速度。另外，所提出的方法优于现有的实时相关滤波器跟踪器。