提出高效交通检测模型FlowDet，兼顾速度与精度。
FlowDet: Overcoming Perspective and Scale Challenges in Real-Time End-to-End Traffic Detection
- 采用解耦编码器优化，结合几何与尺度感知模块。
- 在Intersection-Flow-5k上提升精度1.5%~1.6%，降低63.2%计算量。
- 适合高密度、遮挡严重的实时交通监控场景。
端到端目标检测器为实时应用提供了免非极大值抑制的范式，但其高计算成本仍是主要障碍，尤其在交叉路口等复杂场景中。为此，我们提出FlowDet，一种基于DETR架构的高速检测器，采用解耦编码器优化策略。具体而言，FlowDet引入新型几何可变形单元（GDU）实现交通感知的几何建模，并设计尺度感知注意力（SAA）模块以应对极端尺度变化下的表征能力下降。为严格评估模型在严重遮挡和高密度物体环境中的性能，我们构建了Intersection-Flow-5k数据集，该数据集是当前任务的新挑战。在Intersection-Flow-5k上，FlowDet达到新最优表现：相比强基线RT-DETR，AP(test)提升1.5%，AP50(test)提升1.6%，同时减少63.2%的GFLOPs，推理速度提升16.2%。本工作展示了一条通往高效高精度真实世界感知系统的新路径。数据集已开源：https://github.com/AstronZh/Intersection-Flow-5K。
原文摘要 · Abstract (English)
End-to-end object detectors offer a promising NMS-free paradigm for real-time applications, yet their high computational cost remains a significant barrier, particularly for complex scenarios like intersection traffic monitoring. To address this challenge, we propose FlowDet, a high-speed detector featuring a decoupled encoder optimization strategy applied to the DETR architecture. Specifically, FlowDet employs a novel Geometric Deformable Unit (GDU) for traffic-aware geometric modeling and a Scale-Aware Attention (SAA) module to maintain high representational power across extreme scale variations. To rigorously evaluate the model's performance in environments with severe occlusion and high object density, we collected the Intersection-Flow-5k dataset, a new challenging scene for this task. Evaluated on Intersection-Flow-5k, FlowDet establishes a new state-of-the-art. Compared to the strong RT-DETR baseline, it improves AP(test) by 1.5% and AP50(test) by 1.6%, while simultaneously reducing GFLOPs by 63.2% and increasing inference speed by 16.2%. Our work demonstrates a new path towards building highly efficient and accurate detectors for demanding, real-world perception systems. The Intersection-Flow-5k dataset is available at https://github.com/AstronZh/Intersection-Flow-5K.
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。