HA-RDet: Hybrid Anchor Rotation Detector for Oriented Object Detection

TL;DR

HA-RDet introduces a hybrid approach combining anchor-based and anchor-free methods with orientation-aware convolution, achieving high accuracy in oriented object detection while reducing computational costs.

Contribution

It proposes a novel hybrid detection framework that uses a single preset anchor per location and refines it with orientation-aware convolution, improving efficiency and accuracy.

Findings

Achieves 75.41 mAP on DOTA-v1 dataset.

Reduces computational resources compared to traditional anchor-based methods.

Maintains competitive detection accuracy with fewer anchors.

Abstract

Oriented object detection in aerial images poses a significant challenge due to their varying sizes and orientations. Current state-of-the-art detectors typically rely on either two-stage or one-stage approaches, often employing Anchor-based strategies, which can result in computationally expensive operations due to the redundant number of generated anchors during training. In contrast, Anchor-free mechanisms offer faster processing but suffer from a reduction in the number of training samples, potentially impacting detection accuracy. To address these limitations, we propose the Hybrid-Anchor Rotation Detector (HA-RDet), which combines the advantages of both anchor-based and anchor-free schemes for oriented object detection. By utilizing only one preset anchor for each location on the feature maps and refining these anchors with our Orientation-Aware Convolution technique, HA-RDet achieves competitive accuracies, including 75.41 mAP on DOTA-v1, 65.3 mAP on DIOR-R, and 90.2 mAP on HRSC2016, against current anchor-based state-of-the-art methods, while significantly reducing computational resources.