解决遥感增量目标检测中尺度变化与标注缺失问题，提升模型长期学习能力。
STAR-IOD: Scale-decoupled Topology Alignment with Pseudo-label Refinement for Remote Sensing Incremental Object Detection
- 通过解耦拓扑蒸馏对齐类别间结构关系，缓解尺度变化导致的表征偏差。
- 提出聚类驱动伪标签生成模块，动态区分真实目标与背景，解决旧类标注缺失问题。
- 在两个新构建的数据集上显著优于现有方法，适合遥感持续学习研究者使用。
遥感影像通常以连续数据流形式到达。传统检测器在学习新类别时容易遗忘旧类别，因此遥感增量目标检测（RS-IOD）研究意义重大。然而，现有方法大多忽视遥感场景中普遍存在的类内尺度变化，影响知识迁移与旧知识保留。此外，标注缺失导致模型将旧类实例误判为背景。为此，本文提出新框架STAR-IOD：首先引入子空间解耦拓扑蒸馏（STD）模块，显式对齐类别间拓扑关系，缓解尺度偏移引起的类内表征差异；其次设计聚类驱动伪标签生成器（CPG），利用K-Means动态确定类别专属阈值，准确区分真实目标与背景噪声，缓解旧类标注缺失问题。同时构建了两个新的遥感增量目标检测数据集DIOR-IOD和DOTA-IOD。大量实验表明，本方法在DIOR-IOD和DOTA-IOD上分别较最先进方法提升1.7%和2.1% mAP，有效缓解灾难性遗忘，同时保持基类与新类的强检测性能。代码与数据集已开源。
原文摘要 · Abstract (English)
Remote sensing imagery typically arrives in the form of continuous data streams. Traditional detectors often forget previously learned categories when learning new ones; therefore, research on Remote Sensing Incremental Object Detection (RS-IOD) is of great significance. However, existing methods largely overlook the intra-class scale variations prevalent in remote sensing scenes, which undermines the effectiveness of knowledge transfer and old knowledge preservation. Moreover, RS-IOD also suffers from missing annotations, which cause the model to misclassify old-class instances as background. To address these challenges, we propose a novel framework, STAR-IOD. First, we introduce a Subspace-decoupled Topology Distillation (STD) module to transfer structural knowledge, explicitly aligning inter-class topological relationships and mitigating intra-class representation discrepancies induced by scale shifts. Furthermore, we introduce the Clustering-driven Pseudo-label Generator (CPG), a plug-and-play module that leverages K-Means clustering to dynamically identify class-specific thresholds, thereby guaranteeing an accurate distinction between true positive targets and background noise and alleviating the issue of missing annotations for old classes. We also constructed two Remote Sensing Incremental Object Detection datasets, DIOR-IOD and DOTA-IOD to facilitate research on RS-IOD. Extensive experiments demonstrate that our method outperforms state-of-the-art approaches by 1.7% and 2.1% mAP on DIOR-IOD and DOTA-IOD, respectively, effectively alleviating catastrophic forgetting while preserving strong detection performance on both base and novel classes. The code and dataset are released at: https://github.com/zyt95579/STAR-IOD.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。