发表机构
University of Electronic Science and Technology of China（电子科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出仅用距离窗口级标签训练雷达检测器的策略，通过孪生Transformer生成伪标签并施加位移等变一致性约束，无需单元级标签即可实现接近单元级标签训练的性能，且大幅降低标注成本。
AI 中文摘要
许多基于深度学习的雷达目标检测器依赖距离单元级标签进行训练，而这类标签获取成本高昂。为降低标注负担，本文提出一种仅使用距离窗口级标签的训练策略。具体而言，从同一窗口回波中随机裁剪两个子回波，使二者之间产生已知的相对位移。随后，利用孪生Transformer从其中一个子回波中识别目标显著响应并形成伪标签，再根据已知位移将这些伪标签映射到配对子回波上。此外，引入位移等变一致性（SEC）约束，以在位移补偿后对齐两个子回波的预测响应。通过利用每个已标注窗口内的位移先验，所提方法无需单元级标签即可学习可靠的检测。实验结果表明，该方法持续优于经典恒虚警率（CFAR）检测器，并达到接近使用距离单元级标签训练的性能，同时大幅降低标注成本。
英文摘要
Many deep learning based radar target detectors rely on range-cell level labels for training, which are expensive to obtain. To reduce the labeling burden, this paper presents a training strategy that uses only range-window level labels. Specifically, two sub-echoes are randomly cropped from the same window echo, resulting in a known relative shift between them. A siamese Transformer is then used to identify target-salient responses from one sub-echo and form pseudo labels, which are mapped to the paired sub-echo according to the known shift. In addition, a shift-equivariant consistency (SEC) constraint is imposed to align the predicted responses of the two sub-echoes after shift compensation. By exploiting shift priors within each labeled window, the proposed approach learns reliable detection without requiring cell-level labels. Experimental results show that it consistently surpasses classical constant false alarm rate (CFAR) detectors and achieves performance close to range-cell level labels training, while substantially reducing labeling cost.
CommentsIGARSS 2026 Oral
