Boost UAV-Based Object Detection via Scale-Invariant Feature Disentanglement and Adversarial Learning

Detecting objects from uncrewed aerial vehicles (UAVs) are often hindered by a large number of small objects, resulting in low detection accuracy. To address this issue, mainstream approaches typically utilize multistage inferences. Despite their remarkable detecting accuracies, the real-time efficiency is sacrificed, making them less practical to handle real applications. To this end, we propose to improve the single-stage inference accuracy through learning scale-invariant features. Specifically, a scale-invariant feature disentangling (SIFD) module is designed to disentangle scale-related and scale-invariant features. Then, an adversarial feature learning (AFL) scheme is employed to enhance disentanglement. Finally, scale-invariant features are leveraged for robust UAV-based object detection (UAV-OD). Furthermore, we construct a multimodal UAV object detection dataset, State-Air, which incorporates annotated UAV state parameters. We apply our approach to three lightweight detection frameworks on two benchmark datasets. Extensive experiments demonstrate that our approach can effectively improve model accuracy and achieve state-of-the-art (SoTA) performance on three datasets. Our code and dataset are publicly available at: https://github.com/1e12Leon/SIFDAL