论文标题
旨在解决无人机图像中长尾分布的挑战以进行对象检测
Towards Resolving the Challenge of Long-tail Distribution in UAV Images for Object Detection
论文作者
论文摘要
无人机图像中的对象检测的现有方法忽略了一个重要的挑战 - 无人机图像中的类分布不平衡 - 导致尾巴上的性能差。我们系统地研究了现有的长尾问题解决方案，并揭示了在自然图像数据集中有效的重新平衡方法不能琐碎地应用于无人机数据集。为此，我们重新考虑了无人机图像中的长尾对象检测，并提出了双抽样器和头部检测网络（DSHNET），这是旨在解决无人机图像中长尾分布的第一项工作。 DSHNET中的关键组件包括类别偏置的采样器（CBS）和双侧盒头（BBH），这些盒子的开发为以双路径的方式应对尾部和头类。如果没有铃铛和口哨声，DSHNET显着提高了在不同检测框架上的尾部类别的性能。此外，DSHNET在Visdrone和UAVDT数据集上的基础检测器和通用方法显着优于基本检测器和通用方法。与图像裁剪方法结合使用时，它实现了新的最先进的性能。代码可从https://github.com/we1pingyu/dshnet获得
Existing methods for object detection in UAV images ignored an important challenge - imbalanced class distribution in UAV images - which leads to poor performance on tail classes. We systematically investigate existing solutions to long-tail problems and unveil that re-balancing methods that are effective on natural image datasets cannot be trivially applied to UAV datasets. To this end, we rethink long-tailed object detection in UAV images and propose the Dual Sampler and Head detection Network (DSHNet), which is the first work that aims to resolve long-tail distribution in UAV images. The key components in DSHNet include Class-Biased Samplers (CBS) and Bilateral Box Heads (BBH), which are developed to cope with tail classes and head classes in a dual-path manner. Without bells and whistles, DSHNet significantly boosts the performance of tail classes on different detection frameworks. Moreover, DSHNet significantly outperforms base detectors and generic approaches for long-tail problems on VisDrone and UAVDT datasets. It achieves new state-of-the-art performance when combining with image cropping methods. Code is available at https://github.com/we1pingyu/DSHNet
