RS-TinyNet: Stage-wise Feature Fusion Network for Detecting Tiny Objects in Remote Sensing Images

TL;DR

RS-TinyNet is a novel multi-stage feature fusion network designed to improve tiny object detection in remote sensing images by enhancing feature saliency and preserving structural details, outperforming existing methods.

Contribution

The paper introduces RS-TinyNet with two novel modules: tiny object saliency modeling and feature integrity reconstruction, advancing tiny object detection in remote sensing imagery.

Findings

RS-TinyNet achieves 4.0% higher AP than SOTA on AI-TOD dataset.

The model surpasses existing detectors by 6.5% AP75.

Experimental results validate the effectiveness of multi-stage feature fusion.

Abstract

Detecting tiny objects in remote sensing (RS) imagery has been a long-standing challenge due to their extremely limited spatial information, weak feature representations, and dense distributions across complex backgrounds. Despite numerous efforts devoted, mainstream detectors still underperform in such scenarios. To bridge this gap, we introduce RS-TinyNet, a multi-stage feature fusion and enhancement model explicitly tailored for RS tiny object detection in various RS scenarios. RS-TinyNet comes with two novel designs: tiny object saliency modeling and feature integrity reconstruction. Guided by these principles, we design three step-wise feature enhancement modules. Among them, the multi-dimensional collaborative attention (MDCA) module employs multi-dimensional attention to enhance the saliency of tiny objects. Additionally, the auxiliary reversible branch (ARB) and a progressive fusion detection head (PFDH) module are introduced to preserve information flow and fuse multi-level features to bridge semantic gaps and retain structural detail. Comprehensive experiments on public RS dataset AI-TOD show that our RS-TinyNet surpasses existing state-of-the-art (SOTA) detectors by 4.0% AP and 6.5% AP75. Evaluations on DIOR benchmark dataset further validate its superior detection performance in diverse RS scenarios. These results demonstrate that the proposed multi-stage feature fusion strategy offers an effective and practical solution for tiny object detection in complex RS environments.