AI 中文总结
SPEANet提出阶段特定先验提取与上下文调制，集成固定算子构建参数高效遥感检测骨干，在五个基准上实现精度-参数权衡，DOTA-v1.0达78.55% mAP。
AI 中文摘要
遥感目标检测（RSOD）需要紧凑的骨干网络，能够在极端尺度变化和背景杂波下保留微弱的几何线索。固定结构算子无需引入可学习的算子系数即可提供互补的轮廓和频率响应。然而，直接注入这些响应可能会放大与内容无关的纹理，而在整个层级中应用统一的算子设计可能无法很好地匹配各阶段特定的表示需求。我们提出了结构先验增强注意力网络（SPEANet），这是一种参数高效的RSOD骨干网络，通过阶段特定的先验提取和上下文条件响应调制来集成固定算子。SPEANet将平滑轮廓和多阶方向建模分配给浅层高分辨率特征，同时在更深阶段采用紧凑的近似-细节交互机制。在残差融合之前，学习到的空间门控调节生成的先验响应。在五个基准上的实验，以及在DOTA-v1.0上跨七个检测框架的评估，实现了良好的精度-参数权衡。使用Oriented R-CNN，SPEANet在DOTA-v1.0上达到78.55%的mAP，在DOTA-v1.5上达到72.24%的mAP，在DIOR-R上达到67.30%的mAP，总参数为23.0M，其中骨干网络参数为5.97M。
英文摘要
Remote sensing object detection (RSOD) requires compact backbones capable of preserving weak geometric cues under extreme scale variation and background clutter. Fixed structural operators provide complementary contour and frequency responses without introducing learnable operator coefficients. However, directly injecting these responses can amplify content-irrelevant textures, while applying a uniform operator design across the hierarchy may be poorly matched to stage-specific representation requirements. We propose the Structural Prior Enhanced Attention Network (SPEANet), a parameter-efficient RSOD backbone that integrates fixed operators through stage-specific prior extraction and context-conditioned response modulation. SPEANet assigns smoothed contour and multi-order directional modeling to shallow, high-resolution features, while employing a compact approximation-detail interaction mechanism in deeper stages. Learned spatial gates regulate the resulting prior responses before residual fusion. Experiments on five benchmarks, together with evaluations across seven detection frameworks on DOTA-v1.0, achieve a favorable accuracy-parameter trade-off. With Oriented R-CNN, SPEANet achieves 78.55\% mAP on DOTA-v1.0, 72.24\% mAP on DOTA-v1.5, and 67.30\% mAP on DIOR-R using 23.0M total parameters, including a 5.97M-parameter backbone.
