双流光谱解耦蒸馏用于遥感目标检测
Dual-Stream Spectral Decoupling Distillation for Remote Sensing Object Detection
AI总结：
DS2D2通过双流光谱解耦蒸馏提升遥感目标检测性能，解决混合特征和细微特征差异问题，实现更准确的检测效果。
摘要：
知识蒸馏是一种有效的且硬件友好的方法，在轻量化遥感目标检测中起着关键作用。然而，现有的蒸馏方法常常面临遥感图像（RSI）中混合特征的问题，并忽视了细微特征变化引起的差异，导致知识混淆。为了解决这些挑战，我们提出了一种架构无关的蒸馏方法，称为双流光谱解耦蒸馏（DS2D2），用于通用的遥感目标检测任务。具体而言，DS2D2集成了基于光谱分解的显式和隐式蒸馏。首先，应用一阶小波变换进行光谱分解，以保留RSI的关键空间特征。利用这种空间保留，设计了密度无关尺度权重（DISW），以解决RSI中密集和小目标检测的挑战。其次，我们展示了隐式知识隐藏在细微的学生-教师特征差异中，这些隐式知识在被检测头激活时显著影响预测。这种隐式知识通过全频和高频放大器提取，将特征差异映射到预测偏差。在DIOR和DOTA数据集上的大量实验验证了所提出方法的有效性。具体而言，在DIOR数据集上，DS2D2在RetinaNet上的AP50提升了4.2%，在Faster R-CNN上的AP50提升了3.8%，优于现有蒸馏方法。源代码将在https://github.com/PolarAid/DS2D2上提供。
英文摘要：
Knowledge distillation is an effective and hardware-friendly method, which plays a key role in lightweighting remote sensing object detection. However, existing distillation methods often encounter the issue of mixed features in remote sensing images (RSIs), and neglect the discrepancies caused by subtle feature variations, leading to entangled knowledge confusion. To address these challenges, we propose an architecture-agnostic distillation method named Dual-Stream Spectral Decoupling Distillation (DS2D2) for universal remote sensing object detection tasks. Specifically, DS2D2 integrates explicit and implicit distillation grounded in spectral decomposition. Firstly, the first-order wavelet transform is applied for spectral decomposition to preserve the critical spatial characteristics of RSIs. Leveraging this spatial preservation, a Density-Independent Scale Weight (DISW) is designed to address the challenges of dense and small object detection common in RSIs. Secondly, we show implicit knowledge hidden in subtle student-teacher feature discrepancies, which significantly influence predictions when activated by detection heads. This implicit knowledge is extracted via full-frequency and high-frequency amplifiers, which map feature differences to prediction deviations. Extensive experiments on DIOR and DOTA datasets validate the effectiveness of the proposed method. Specifically, on DIOR dataset, DS2D2 achieves improvements of 4.2% in AP50 for RetinaNet and 3.8% in AP50 for Faster R-CNN, outperforming existing distillation approaches. The source code will be available at https://github.com/PolarAid/DS2D2.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
