DARDet: A Dense Anchor-free Rotated Object Detector in Aerial Images
This work addresses the need for more robust and efficient object detection in aerial imagery, which is crucial for applications like surveillance and mapping, though it is incremental as it builds on existing anchor-free and rotated detection methods.
The paper tackles the problem of rotated object detection in aerial images, which is challenging due to variations in scale, rotation, aspect ratio, and dense arrangements, by proposing DARDet, a dense anchor-free detector that achieves state-of-the-art performance on datasets like DOTA, HRSC2016, and UCAS-AOD while maintaining high efficiency.
Rotated object detection in aerial images has received increasing attention for a wide range of applications. However, it is also a challenging task due to the huge variations of scale, rotation, aspect ratio, and densely arranged targets. Most existing methods heavily rely on a large number of pre-defined anchors with different scales, angles, and aspect ratios, and are optimized with a distance loss. Therefore, these methods are sensitive to anchor hyper-parameters and easily suffer from performance degradation caused by boundary discontinuity. To handle this problem, in this paper, we propose a dense anchor-free rotated object detector (DARDet) for rotated object detection in aerial images. Our DARDet directly predicts five parameters of rotated boxes at each foreground pixel of feature maps. We design a new alignment convolution module to extracts aligned features and introduce a PIoU loss for precise and stable regression. Our method achieves state-of-the-art performance on three commonly used aerial objects datasets (i.e., DOTA, HRSC2016, and UCAS-AOD) while keeping high efficiency. Code is available at https://github.com/zf020114/DARDet.