基于多分支和重参数化的孪生网络跟踪算法  

Siamese Network Tracking Algorithm Based on Diverse Branch Block and Reparameterization

作　　者：金铭[1] 唐宇[1] 韩勇[1] 刘帅[1] 闫锋刚[1] JIN Ming;TANG Yu;HAN Yong;LIU Shuai;YAN Feng Gang(Harbin Institute of Technology,Weihai 264200,China)

机构地区：[1]哈尔滨工业大学(威海),威海264200

出　　处：《遥测遥控》2024年第4期22-30,共9页Journal of Telemetry,Tracking and Command

基　　金：国家自然科学基金(61971158,62171150);泰山学者工程专项经费资助(tsqn202211087);国家自然基金面上项目资助(62071144);山东省自然科学基金项目(ZR2023MF091)。

摘　　要：针对孪生网络对尺度变化目标特征表达能力不足的问题,本文使用不同尺寸的卷积、池化分支和剪枝操作构成多分支结构,以提高特征的鲁棒性并保证孪生网络的平移不变性。针对多分支结构带来参数量过多的问题,本文在跟踪阶段将多分支结构重参数化为单一的卷积,有效减少跟踪阶段时间成本。实验结果表明:本文提出的算法相比于SiamFC,在OTB100数据集上,其精度、成功率和跟踪速度分别提高了5.1%、3%、30FPS,在UAV123和Temple-Color-128数据集上跟踪精度和成功率均有所提高。Aiming at the problem that the Siamese network has insufficient ability to express the features of scale-varying targets,a multi-branch structure is constructed by using convolution,pooling branches and pruning operations of different sizes to improve the robustness of features and ensure the translation invariance of the Siamese network.Aiming at the problem that the multi-branch structure brings too many parameters,the multi-branch structure is reparameterized into a single convolution in the tracking stage,which effectively reduces the time cost in the tracking stage.The experimental results show that compared with SiamFC,the accuracy,success rate and tracking speed of the proposed algorithm on the OTB100 datasets are improved by 5.1%,3%and 30 FPS,respectively.The tracking accuracy and success rate are improved on the UAV123 and Temple-Color-128 datasets.

关 键 词：视觉跟踪 孪生网络 特征提取 结构重参数化 

分 类 号：TN911.73[电子电信—通信与信息系统] TP391.4[电子电信—信息与通信工程]