针对小目标检测的三大瓶颈，提出频域迭代优化与动态特征聚合新方法。
DFIR-DETR: Frequency-Domain Iterative Refinement and Dynamic Feature Aggregation for Small Object Detection
- 通过频域迭代修正注意力分布，聚焦复杂区域
- 在NEU-DET和VisDrone上分别达到92.9%和51.6% mAP50
- 参数仅1170万，适合资源受限场景的小目标检测
复杂场景中的小目标检测面临神经网络设计的根本矛盾：主干网络注意力均匀分配计算资源而忽略内容差异，金字塔结构在上采样时放大激活值却无归一化补偿，瓶颈卷积通过累积空间滤波逐步模糊高频边缘信息。为此，我们基于RT-DETR基线的三项可度量缺陷，提出DFIR-DETR：针对忽略空间复杂性的均匀注意力、上采样后引发的归一化漂移、以及抑制小目标依赖的高频成分的空间卷积。在NEU-DET和VisDrone数据集上，仅用11.7M参数和47.2 GFLOPs，mAP50分别达到92.9%和51.6%，在两类差异显著的检测领域均实现稳定提升。
原文摘要 · Abstract (English)
Small object detection in complex scenes exposes a fundamental tension in neural network design: backbone attention distributes computation uniformly regardless of content, pyramid necks inflate activation magnitudes during upsampling without norm compensation, and bottleneck convolutions progressively smooth high-frequency edge components through accumulated spatial filtering. In response, we develop DFIR-DETR by tracing each proposed module back to a specific, measurable deficiency in the RT-DETR baseline: uniform attention that ignores spatial complexity, norm drift that destabilises upsampled features, and spatial convolutions that progressively suppress the high-frequency components small objects depend on. On NEU-DET and VisDrone, DFIR-DETR achieves 92.9% and 51.6% mAP50 with only 11.7M parameters and 47.2 GFLOPs, demonstrating consistent gains across two qualitatively different detection domains.
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。