面向遥感图像细粒度目标检测的平衡层级对比学习与解耦查询
Balanced Hierarchical Contrastive Learning with Decoupled Queries for Fine-grained Object Detection in Remote Sensing Images
在 DETR 框架内把 object query 拆成分类集与定位集，并用带可学习类原型的层级对比损失平衡各层级类别梯度，可直接迁移到任何需要细粒度分类与定位解耦的检测器。
原文依据 separates DETR's object queries into classification and localization sets
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论细粒度遥感数据集常采用层级标签结构，以由粗到细的方式区分目标，每个目标在多个层级上被标注。然而，将这种语义层级嵌入表征学习空间以提升细粒度检测性能仍然具有挑战性。已有研究在不同层级上应用监督对比学习，将同一父类下的目标聚拢，同时区分同级子类。但这些研究忽略了两个关键问题：（1）标签层级上的数据分布不均衡导致高频类别主导学习过程；（2）类别间语义关系的学习会干扰与类别无关的定位。为解决这些问题，我们在检测 Transformer（DETR）框架内提出一种平衡层级对比损失，并结合解耦学习策略。所提出的损失引入可学习的类原型，并在每个层级上均衡不同类别所贡献的梯度，从而确保每个层级类别在每个小批次中对损失计算的贡献相等。解耦策略将 DETR 的目标查询分离为分类集与定位集，从而实现面向具体任务的特征提取与优化。在三个带层级标注的细粒度数据集上的实验表明，我们的方法优于最优方法。
英文原文摘要 1507 字符
Fine-grained remote sensing datasets often use hierarchical label structures to differentiate objects in a coarse-to-fine manner, with each object annotated across multiple levels. However, embedding this semantic hierarchy into the representation learning space to improve fine-grained detection performance remains challenging. Previous studies have applied supervised contrastive learning at different hierarchical levels to group objects under the same parent class while distinguishing sibling subcategories. Nevertheless, they overlook two critical issues: (1) imbalanced data distribution across the label hierarchy causes high-frequency classes to dominate the learning process, and (2) learning semantic relationships among categories interferes with class-agnostic localization. To address these issues, we propose a balanced hierarchical contrastive loss combined with a decoupled learning strategy within the detection transformer (DETR) framework. The proposed loss introduces learnable class prototypes and equilibrates gradients contributed by different classes at each hierarchical level, ensuring that each hierarchical class contributes equally to the loss computation in every mini-batch. The decoupled strategy separates DETR's object queries into classification and localization sets, enabling task-specific feature extraction and optimization. Experiments on three fine-grained datasets with hierarchical annotations demonstrate that our method outperforms state-of-the-art approaches.