DARDet: A Dense Anchor-free Rotated Object Detector in Aerial Images

TL;DR

DARDet is a novel dense anchor-free rotated object detector that directly predicts rotated bounding boxes in aerial images, achieving state-of-the-art accuracy while maintaining high efficiency.

Contribution

It introduces a dense anchor-free approach with aligned feature extraction and a PIoU loss for improved rotated object detection in aerial images.

Findings

Achieves state-of-the-art results on DOTA, HRSC2016, and UCAS-AOD datasets.

Operates efficiently with high accuracy in detecting densely arranged rotated objects.

Outperforms existing anchor-based methods in robustness and precision.

Abstract

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.