Visible-infrared vehicle target detection aims to pinpoint the location and class of vehicles by fusing the favorable complementary information of visible–infrared image pairs. However, most of the detection methods cannot obtain ideal detection performance when visible–infrared image pairs are captured in low lighting environment. To solve this issue, we propose a global–local feature fusion network (GLFNet), which can adaptively integrate the saliency information from visible–infrared image pairs. Initially, a dual-stream ResNet-50 network is designed to extract cross-modal features from visible–infrared image pairs. Then, a global-local feature fusion (GLF) module is proposed to merge the multimodality features. Finally, the detection head utilizes the fused features of the deep interaction to get the detection results. Experiments on the DroneVehicle and LLVIP datasets show that the proposed method is increased by 7.4% and 1.2% compared with recently proposed methods, respectively.

中文翻译：

---

用于可见光-红外车辆检测的全局-局部特征融合网络

可见光-红外车辆目标检测旨在通过融合可见光-红外图像对的有利互补信息来精确定位车辆的位置和类别。然而，当在弱光环境下捕获可见光-红外图像对时，大多数检测方法无法获得理想的检测性能。为了解决这个问题，我们提出了一种全局-局部特征融合网络（GLFNet），它可以自适应地集成来自可见光-红外图像对的显着性信息。最初，双流 ResNet-50 网络设计用于从可见光-红外图像对中提取跨模态特征。然后，提出了全局局部特征融合（GLF）模块来合并多模态特征。最后，检测头利用深度交互的融合特征得到检测结果。在 DroneVehicle 和 LLVIP 数据集上的实验表明，与最近提出的方法相比，该方法分别提高了 7.4% 和 1.2%。