arXiv 2504.114702025-04-11cs.CVcs.AI

SO-DETR：利用双域特征与知识蒸馏实现小目标检测

SO-DETR: Leveraging Dual-Domain Features and Knowledge Distillation for Small Object Detection

Academy for Engineering and Technology, Fudan University（复旦大学工程与技术学院）

Huaxiang Zhang, Hao Zhang, Aoran Mei, Zhongxue Gan, Guo-Niu Zhu

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对基于检测Transformer的方法在小目标检测中存在的低级特征融合不足、查询选择策略不匹配问题，提出SO-DETR模型，通过双域混合编码器、增强型查询选择机制及知识蒸馏策略，在相似计算需求下提升了小目标检测性能。

摘要：

基于检测Transformer的方法在通用目标检测中取得了显著进展，但有效检测小目标仍面临挑战。现有编码器难以高效融合低级特征，查询选择策略也未针对小目标优化。为此，本文提出高效模型Small Object Detection Transformer（SO-DETR），包含双域混合编码器、增强型查询选择机制和知识蒸馏策略三部分。双域混合编码器整合空间与频率域以有效融合多尺度特征，增强高分辨率特征表示且计算开销较低；增强型查询选择机制通过扩展IoU动态选择高分锚框优化查询初始化，改善查询资源分配；还结合轻量骨干网络与知识蒸馏策略。在VisDrone-2019-DET和UAVVaste数据集上的实验表明，SO-DETR在相似计算需求下优于现有方法。

英文摘要：

Detection Transformer-based methods have achieved significant advancements in general object detection. However, challenges remain in effectively detecting small objects. One key difficulty is that existing encoders struggle to efficiently fuse low-level features. Additionally, the query selection strategies are not effectively tailored for small objects. To address these challenges, this paper proposes an efficient model, Small Object Detection Transformer (SO-DETR). The model comprises three key components: a dual-domain hybrid encoder, an enhanced query selection mechanism, and a knowledge distillation strategy. The dual-domain hybrid encoder integrates spatial and frequency domains to fuse multi-scale features effectively. This approach enhances the representation of high-resolution features while maintaining relatively low computational overhead. The enhanced query selection mechanism optimizes query initialization by dynamically selecting high-scoring anchor boxes using expanded IoU, thereby improving the allocation of query resources. Furthermore, by incorporating a lightweight backbone network and implementing a knowledge distillation strategy, we develop an efficient detector for small objects. Experimental results on the VisDrone-2019-DET and UAVVaste datasets demonstrate that SO-DETR outperforms existing methods with similar computational demands. The project page is available at https://github.com/ValiantDiligent/SO_DETR.

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。