计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (6A): 250400017-8.doi: 10.11896/jsjkx.250400017

• 图像处理&多媒体技术 • 上一篇    下一篇

基于RWM和多尺度注意力的车辆重识别

李亚龙1, 王海瑞1, 朱贵富2,3, 卢世宇1   

  1. 1 昆明理工大学信息工程与自动化学院 昆明 650504
    2 昆明理工大学信息化建设管理中心 昆明 650504
    3 昆明理工大学-曙光信息产业股份有限公司AI联合研究中心 昆明 650504
  • 出版日期:2026-06-16 发布日期:2026-06-12
  • 通讯作者: 朱贵富(zhuguifu@kust.edu.cn)
  • 作者简介:(20232204065@stu.kust.edu.cn)
  • 基金资助:
    国家自然科学基金(62462064,61863106)

Vehicle Re-identification Based on RWM and Multi-scale Attention

LI Yalong1, WANG Hairui1, ZHU Guifu2,3, LU Shiyu1   

  1. 1 School of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650504,China
    2 Information Construction Management Center,Kunming University of Science and Technology,Kunming 650504,China
    3 Kunming University of Science and Technology Shuguang Information Industry Co.,Ltd.,AI Joint Research Center,Kunming 650504,China
  • Published:2026-06-16 Online:2026-06-12
  • About author:LI Yalong,born in 2000,postergra-duate.His main research interests include computer vision and vehicle re-identification
    ZHU Guifu,born in 1984,supervisor,senior engineer.His main research interests include Intelligent diagnosis technology,education big data.
  • Supported by:
    National Natural Science Foundation of China(62462064,61863106).

PDF (PC)

38

摘要： 针对现有车辆重识别任务中样本类内差异性大和类间相似度高,导致关键特征提取、全局与局部特征融合不足的问题,提出了一种基于RWM和多尺度注意力的车辆重识别方法。首先,设计了一种区域加权映射(RWM),强化图像中关键区域的特征表示,有效减少背景信息的干扰;其次,在Transformer结构的自注意力机制基础上,引入了多尺度注意力模块(MAB),结合大核感受野和多尺度特性,实现对全局结构信息的有效建模,同时增强局部细节的表达能力,提高模型的区分能力;最后,构建混合损失函数,优化模型的特征学习过程,使不同类别的车辆特征更加可分,提升泛化能力。将所提方法在VeRi-776和VehicleID数据集上进行实验,CMC@1分别达到97.4%和85.8%,CMC@5分别达到98.9%和97.7%,结果表明所提方法能够提取更具判别力的车辆特征。

关键词: 车辆重识别, 区域加权映射, 多尺度注意力, 大核感受野

Abstract: This paper proposes a vehicle re-identification method based on RWM and multi-scale attention to address the pro-blems of large intra class differences and high inter class similarity in existing vehicle re-identification tasks,which lead to insufficient key feature extraction and fusion of global and local features.Firstly,a Region Weighted Mapping(RWM) is designed to enhance the feature representation of key regions in the image,effectively reducing the interference of background information.Se-condly,based on the self attention mechanism of the Transformer structure,a multi-scale attention module(MAB) is introduced,which combines the large kernel receptive field and multi-scale characteristics to effectively model global structural information,while enhancing the expression ability of local details and improving the model's discriminative ability.Finally,a mixed loss function is constructed to optimize the feature learning process of the model,making the features of different categories of vehicles more distinguishable and improving generalization ability.Experiments on the proposed method are conducted on the VeRi-776 and VehicleID datasets.The CMC@1 values reach 97.4% and 85.8% respectively,while the CMC@5 values reach 98.9% and 97.7% respectively.The results show that the proposed method can extract more discriminative vehicle features.

Key words: Vehicle re-identification, Regional weighted mapping, Multi-scale large attention, Large nuclear receptive field

中图分类号: 

  • TP391
[1] KHAN S D,ULLAH H.A survey of advances in vision-based vehicle re-identification[J].Computer Vision and Image Understanding,2019,182:50-63.
[2] LIU H,TIAN Y,YANG Y,et al.Deep relative distance learning:Tell the difference between similar vehicles[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:2167-2175.
[3] KHORRAMSHAHI P,KUMAR A,PERI N,et al.A dual-path model with adaptive attention for vehicle re-identification[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2019:6132-6141.
[4] LIU X,LIU W,MA H,et al.Large-scale vehicle re-identification in urban surveillance videos[C]//2016 IEEE International Conference on Multimedia and Expo(ICME).IEEE,2016:1-6.
[5] SUTHAHARAN S,SUTHAHARAN S.Support vector ma-chine[J].Machine learning models and algorithms for big data classification:thinking with examples for effective learning,2016:207-235.
[6] PISNER D A,SCHNYER D M.Support vector machine[M]//Machine learning.Academic Press,2020:101-121.
[7] WANG Y,LI Y,WANG G,et al.Multi-scale attention network for single image super-resolution[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2024:5950-5960.
[8] VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need[J].Advances in Neural Information Processing Systems,2017,30:5998-6008.
[9] HE S,LUO H,WANG P,et al.Transreid:Transformer-based object re-identification[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:15013-15022.
[10] SHEN F,XIE Y,ZHU J,et al.Git:Graph interactive transformer for vehicle re-identification[J].IEEE Transactions on Image Processing,2023,32:1039-1051.
[11] ZHANG X,LING Y,LI K,et al.Multimodality Adaptive Transformer and Mutual Learning for Unsupervised Domain Adaptation Vehicle Re-Identification[J].IEEE Transactions on Intelligent Transportation Systems,2024,25:20215-20226.
[12] VOITA E,TALBOT D,MOISEEV F,et al.Analyzing multi-head self-attention:Specialized heads do the heavy lifting,the rest can be pruned[J].arXiv:1905.09418,2019.
[13] KRUSE R,MOSTAGHIM S,BORGELT C,et al.Multi-layer perceptrons[M]//Computational intelligence:a methodological introduction.Cham:Springer International Publishing,2022:53-124.
[14] MAO A,MOHRI M,ZHONG Y.Cross-entropy loss functions:Theoretical analysis and applications[C]//International Conference on Machine Learning.PMLR,2023:23803-23828.
[15] ZHANG Z,WANG L,CHENG S.Composed query image retrieval based on triangle area triple loss function and combining CNN with transformer[J].Scientific Reports,2022,12(1):20800.
[16] EXPOSITO-ALONSO M,BOOKER T R,CZECH L,et al.Genetic diversity loss in the Anthropocene[J].Science,2022,377(6613):1431-1435.
[17] LIU H,TIAN Y,YANG Y,et al.Deep relative distance learning:Tell the difference between similar vehicles[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2016:2167-2175.
[18] WANG Z,TANG L,LIU X,et al.Orientation invariant feature embedding and spatial temporal regularization for vehicle re-identification[C]//Proceedings of the IEEE international conference on computer vision.2017:379-387.
[19] LOU Y,BAI Y,LIU J,et al.Embedding adversarial learning for vehicle re-identification[J].IEEE Transactions on Image Processing,2019,28(8):3794-3807.
[20] LIU X,ZHANG S,HUANG Q,et al.Ram:a region-aware deep model for vehicle re-identification[C]//2018 IEEE International Conference on Multimedia and Expo(ICME).IEEE,2018:1-6.
[21] HE B,LI J,ZHAO Y,et al.Part-regularized near-duplicate vehicle re-identification[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR).2019:3997-4005.
[22] ZHANG X,ZHANG R,CAO J,et al.Part-guided attentionlearning for vehicle instance retrieval[J].IEEE Transactions on Intelligent Transportation Systems,2020,23(4):3048-3060.
[23] LEE S,WOO T,LEE S H.Multi-attention-based soft partition network for vehicle re-identification[J].Journal of Computational Design and Engineering,2023,10(2):488-502.
[24] SUN K,PANG X,ZHENG M,et al.Heterogeneous context interaction network for vehicle re-identification[J].Neural Networks,2024,169:293-306.
[25] ZHANG W,JIA S,ZHU H.Single Pixel Imaging Based on Bicubic Interpolation Walsh Transform Matrix[J].IEEE Access,2024,12:138575-13858
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发