Focus, Magnify, and Detect: An Adaptive Region-Aware Detector for Remote Sensing Object Detection
Remote Sensing · 09.11.2026 · 18 (21) · 3854
DOI: 10.3390/rs18213854 ↗Kurz gefasst
Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 09.11.2026. Verfasst von Xuehuai Shi und 4 weiteren Autor:innen. Die Arbeit nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Basic Research Program of Jiangsu, Jiangsu Province Youth Science and Technology Talent Support Project.
Abstract (Originalsprache)
Object detection in remote sensing images (RSIs) suffers severely from small-scale instances, which are inherently constrained by extremely low pixel occupancy, fragile spatial detail retention, and insufficient discriminative feature representation. Global scaling is an intuitive strategy to compensate for the spatial information loss of small targets, yet it inevitably brings significant computational complexity as convolutional networks grow quadratically with the elevation of spatial resolution. In order to alleviate this accuracy–efficiency dilemma, we propose a focus, magnify, detect network, named adaptive region-aware detector (ARADet), tailored for object detection in RSIs. In the focus stage, our region-aware attention (RAA) module generates a Gaussian-constrained attention heatmap to rapidly localize potential small-target regions from the regional attention map. In the magnify stage, we introduce a differentiable affine magnification mechanism with learnable scaling factors. For each focused region, the network adaptively predicts a scaling factor and constructs a center-aligned affine transformation matrix to achieve center-preserving regional magnification. Unlike global upsampling schemes, our module only enhances target-region features and fills them back into the original feature map, thus keeping the overall feature resolution unchanged. The final enhanced features are fed into the detection head for category classification and oriented bounding-box regression. Extensive experiments conducted on two standard aerial object detection benchmarks, including DOTA and DIOR-R, demonstrate that ARADet brings 1.18% mAP improvement over the baseline method, with merely 0.3726M additional parameters. Although the introduced feature amplification yields a moderate rise in FLOPs, ARADet achieves favorable accuracy gains with negligible parameter overhead and surpasses many state-of-the-art oriented detectors.
Förderung
- National Natural Science Foundation of China
- Basic Research Program of Jiangsu
- Jiangsu Province Youth Science and Technology Talent Support Project
- Jiangsu Provincial Innovation Support Program
- Jiangsu Provincial Key Research and Development Program
- Jiangsu Provincial Natural Science Foundation of China
- Nanjing University of Posts and Telecommunications
Zitieren
APA
Shi, X., Xie, R., Yu, K., Wei, Z., & Zheng, S. (2026). Focus, Magnify, and Detect: An Adaptive Region-Aware Detector for Remote Sensing Object Detection. Remote Sensing, 18(21), 3854. https://doi.org/10.3390/rs18213854
Vancouver
Shi X, Xie R, Yu K, Wei Z, Zheng S. Focus, Magnify, and Detect: An Adaptive Region-Aware Detector for Remote Sensing Object Detection. Remote Sensing. 2026;18(21):3854. doi:10.3390/rs18213854
BibTeX
@article{Shi2026,
author = {Shi, Xuehuai and Xie, Ruiqi and Yu, Kun and Wei, Zhihui and Zheng, Shangdong},
title = {Focus, Magnify, and Detect: An Adaptive Region-Aware Detector for Remote Sensing Object Detection},
journal = {Remote Sensing},
year = {2026},
volume = {18},
number = {21},
pages = {3854},
doi = {10.3390/rs18213854}
}
EUVIMED