arXiv 2103.129242021-03-24cs.CV

在空中视频上对深度跟踪器进行基准测试

Benchmarking Deep Trackers on Aerial Videos

Rochester Institute of Technology（罗切斯特理工学院）

Abu Md Niamul Taufique, Breton Minnehan, Andreas Savakis

备注：: 25 pages, 10 figures, 7 tables
发表信息：: Sensors 2020, 20(2), 547
DOI：: 10.3390/s20020547

摘要：

近年来，基于深度学习的视觉对象跟踪器在几个视觉对象跟踪基准测试中取得了最先进的性能。然而，大多数跟踪基准都集中在地面视频，而空中跟踪提出了一系列新的挑战。在本文中，我们在四个航空数据集上比较了十个基于深度学习技术的跟踪器。我们使用不同的方法选择性能最好的跟踪器，特别是通过检测跟踪，判别相关滤波器，暹罗网络和强化学习。在我们的实验中，我们使用具有航拍风格视频的OTB 2015数据集的子集;没有合成序列的UAV 123数据集;包含20个长序列的UAV 20 L数据集;以及DTB 70数据集作为我们的基准数据集。我们比较了不同的跟踪器在航空数据中遇到的不同跟踪情况下的优缺点。我们的研究结果表明，与标准地面视频相比，跟踪器在空中数据集中的表现明显较差。我们将这种效果归因于较小的目标尺寸、相机运动、相对于目标的显著相机旋转、视图外的运动以及以遮挡或类似外观的干扰物的形式出现在跟踪对象附近的混乱。

英文摘要：

In recent years, deep learning-based visual object trackers have achieved state-of-the-art performance on several visual object tracking benchmarks. However, most tracking benchmarks are focused on ground level videos, whereas aerial tracking presents a new set of challenges. In this paper, we compare ten trackers based on deep learning techniques on four aerial datasets. We choose top performing trackers utilizing different approaches, specifically tracking by detection, discriminative correlation filters, Siamese networks and reinforcement learning. In our experiments, we use a subset of OTB2015 dataset with aerial style videos; the UAV123 dataset without synthetic sequences; the UAV20L dataset, which contains 20 long sequences; and DTB70 dataset as our benchmark datasets. We compare the advantages and disadvantages of different trackers in different tracking situations encountered in aerial data. Our findings indicate that the trackers perform significantly worse in aerial datasets compared to standard ground level videos. We attribute this effect to smaller target size, camera motion, significant camera rotation with respect to the target, out of view movement, and clutter in the form of occlusions or similar looking distractors near tracked object.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

摘要：

英文摘要：

相关论文