Abstract
The past few years have seen an increased interest in aerial image object detection due to its critical value to large-scale geo-scientific research like environmental studies, urban planning, and intelligence monitoring. However, the task is very challenging due to the birds-eye view perspective, complex backgrounds, large and various image sizes, different appearances of objects, and the scarcity of well-annotated datasets. Recent advances in computer vision have shown promise tackling the challenge. Specifically, Vision Transformer Detector (ViTDet) was proposed to extract multi-scale features for object detection. The empirical study shows that ViTDet's simple design achieves good performance on natural scene images and can be easily embedded into any detector architecture. To date, ViTDet's potential benefit to challenging aerial image object detection has not been explored. Therefore, in our study, 25 experiments were carried out to evaluate the effectiveness of ViTDet for aerial image object detection on three well-known datasets: Airbus Aircraft, RarePlanes, and Dataset of Object DeTection in Aerial images (DOTA). Our results show that ViTDet can consistently outperform its convolutional neural network counterparts on horizontal bounding box (HBB) object detection by a large margin (up to 17% on average precision) and that it achieves the competitive performance for oriented bounding box (OBB) object detection. Our results also establish a baseline for future research.
Abstract (translated)
过去几年中,由于对空中图像对象检测在环境研究、城市规划和情报监测等大规模地球科学研究的重要性日益关注,因此对空中图像对象检测的研究也变得越来越重要。然而,任务非常困难,因为的视角、复杂的背景、大型和多种图像大小、不同的物体外观以及缺乏标注的数据集的短缺。计算机视觉的进步表明,可以克服这种挑战。具体来说,视觉转换检测器(ViTDet)被提出用于提取多尺度特征来进行物体检测。经验研究表明,ViTDet的简单设计在自然场景图像中实现了良好的性能,并且可以轻松地嵌入到任何检测架构中。到目前为止,ViTDet对具有挑战性的空中图像对象检测的潜在益处尚未被探索。因此,在我们的研究中,进行了25次实验,以评估ViTDet在著名的数据集上对空中图像对象检测的有效性:Airbus aircraft、稀有飞机和空中图像对象检测数据集(DOTA)。我们的结果显示,ViTDet可以在水平边界框(HBB)物体检测方面 consistently outperform its卷积神经网络counterparts(平均精度高达17%),并且对于定向边界框(OBB)物体检测取得了竞争性能。我们的研究结果也建立了未来研究的基础。
URL
https://arxiv.org/abs/2301.12058