Towards Resolving the Challenge of Long-tail Distribution in UAV Images for Object Detection
This addresses the challenge of long-tail distribution in UAV images for object detection, which is incremental as it adapts existing long-tail methods to a specific domain.
The paper tackles the problem of imbalanced class distribution in UAV images for object detection, which leads to poor performance on tail classes, and proposes DSHNet, achieving new state-of-the-art performance on VisDrone and UAVDT datasets.
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