轻量级网络实现高精度人流计数，适合实时边缘计算。
RepSFNet : A Single Fusion Network with Structural Reparameterization for Crowd Counting
- 用重参数大核视觉变压器提取多尺度特征
- 融合ASPP与感知上下文模块，提升密度适应性
- 无注意力机制设计，推理快34%，适合移动端
人流计数在密度变化大的场景中仍具挑战，主要源于尺度变化、遮挡以及现有模型计算成本高。为此，我们提出RepSFNet（重参数化单融合网络），一种轻量级架构，用于准确且实时的人流估计。RepSFNet采用具有大重参数化卷积核的RepLK-ViT骨干网络，高效提取多尺度特征；进一步集成结合空洞空间金字塔池化（ASPP）与上下文感知网络（CAN）的特征融合模块，实现鲁棒且密度自适应的上下文建模；采用拼接融合模块保留空间分辨率，生成高质量密度图。通过避免使用注意力机制和多分支结构，显著减少参数量与计算复杂度。训练目标融合均方误差与最优传输损失，同时提升计数准确性和空间分布一致性。在ShanghaiTech、NWPU和UCF-QNRF数据集上的实验表明，RepSFNet在保持竞争性精度的同时，推理延迟相比近期最先进方法最高降低34%，适用于实时与低功耗边缘计算场景。
原文摘要 · Abstract (English)
Crowd counting remains challenging in variable-density scenes due to scale variations, occlusions, and the high computational cost of existing models. To address these issues, we propose RepSFNet (Reparameterized Single Fusion Network), a lightweight architecture designed for accurate and real-time crowd estimation. RepSFNet leverages a RepLK-ViT backbone with large reparameterized kernels for efficient multi-scale feature extraction. It further integrates a Feature Fusion module combining Atrous Spatial Pyramid Pooling (ASPP) and Context-Aware Network (CAN) to achieve robust, density-adaptive context modeling. A Concatenate Fusion module is employed to preserve spatial resolution and generate high-quality density maps. By avoiding attention mechanisms and multi-branch designs, RepSFNet significantly reduces parameters and computational complexity. The training objective combines Mean Squared Error and Optimal Transport loss to improve both count accuracy and spatial distribution alignment. Experiments conducted on ShanghaiTech, NWPU, and UCF-QNRF datasets demonstrate that RepSFNet achieves competitive accuracy while reducing inference latency by up to 34 percent compared to recent state-of-the-art methods, making it suitable for real-time and low-power edge computing applications.
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。