通过抗噪声查询优化重新审视用于小目标检测的 DETR
Revisiting DETR for Small Object Detection via Noise-Resilient Query Optimization
AI总结：
针对小目标检测中 FPN 噪声敏感与标签分配查询质量不足的问题，提出 NRQO 范式，结合 NT-FPN 和 PS-RPN 提升特征与正查询质量，并在多个基准上超过现有基线。
摘要：
尽管基于 Transformer 的检测器在小目标检测（SOD）方面取得了进展，但近期研究表明，由于特征金字塔网络（FPN）固有的噪声敏感性以及现有标签分配策略中查询质量下降，这些检测器仍面临挑战。本文提出一种新颖的抗噪声查询优化（NRQO）范式，该范式创新性地融合了抗噪声特征金字塔网络（NT-FPN）和成组相似度区域提议网络（PS-RPN）。具体而言，NT-FPN 通过保持空间和语义信息完整性，缓解 FPN 特征融合过程中的噪声。与现有标签分配策略不同，PS-RPN 通过利用位置和形状相似度增强锚框与真实框匹配，生成足够数量的高质量正查询，且无需额外超参数。在多个基准上进行的大量实验一致表明，NRQO 优于最先进的基线方法。
英文摘要：
Despite advancements in Transformer-based detectors for small object detection (SOD), recent studies show that these detectors still face challenges due to inherent noise sensitivity in feature pyramid networks (FPN) and diminished query quality in existing label assignment strategies. In this paper, we propose a novel Noise-Resilient Query Optimization (NRQO) paradigm, which innovatively incorporates the Noise-Tolerance Feature Pyramid Network (NT-FPN) and the Pairwise-Similarity Region Proposal Network (PS-RPN). Specifically, NT-FPN mitigates noise during feature fusion in FPN by preserving spatial and semantic information integrity. Unlike existing label assignment strategies, PS-RPN generates a sufficient number of high-quality positive queries by enhancing anchor-ground truth matching through position and shape similarities, without the need for additional hyperparameters. Extensive experiments on multiple benchmarks consistently demonstrate the superiority of NRQO over state-of-the-art baselines.
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。
