基于非对称增强注意力与特征交叉融合的行人重识别方法

doi:10.3969/j.issn.1000-1158.2022.12.08

摘要
图/表
参考文献(28)
相关文章 (15)

全文: PDF (2693 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对现有的行人重识别方法提取到的特征信息充分性与辨识性不足导致检索精度低的问题,提出一种基于非对称增强注意力与特征交叉融合的行人重识别方法。首先,构建非对称增强注意力模块,通过多重池化聚合的跨邻域通道交互注意力增强显著特征表示,使网络聚焦于图像中的行人区域;其次,考虑到网络各层特征间的差异性与关联性,构建特征交叉融合模块,利用交叉融合方式实现同层不同级特征的跨层级融合,进而实现多尺度融合;最后,水平切分输出特征以获取局部特征,从而实现在特定区域上描述行人。在Market1501、DukeMTMC-reID与CUHK03这3个公开数据集上对提出的方法进行了验证,首位命中率(Rank-1)分别达到了93.5%、85.1%和64.3%,证明了该方法在提升行人重识别性能上具有优越性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	金梅
	李媛媛
	郝兴军
	杨曼
	张立国

关键词 ：计量学, 行人重识别;非对称增强注意力, 特征交叉融合, 深度学习, 首位命中率

Abstract：In order to solve the problem of poor retrieval accuracy due to the lack of sufficient and recognizable feature information extracted by the existing pedestrian re-identification methods, a pedestrian re-identification method based on the asymmetric enhanced attention and feature cross fusion is proposed. First, an asymmetric enhanced attention module is constructed, and the salient feature representation is enhanced through the cross-neighbor channel interactive attention that is aggregated by multiple pooling, so that the pedestrian area in the image is focused by the network. Then, taking into account the differences and relevance between the features of each layer of the network, a feature cross fusion module is constructed, and the cross fusion method is used to achieve cross-level fusion of features at different levels in the same layer, thereby realizing multi-scale fusion. Finally, the output features are segmented horizontally to obtain local features, so as to describe pedestrians in a specific area. The proposed method is verified on the three public data sets of Market1501, DukeMTMC-reID and CUHK03. Rank-1 reach 93.5%, 85.1% and 64.3% respectively, proving that the method is superior in improving the performance of pedestrian re-identification.

Key words： metrology pedestrian re-identification asymmetric enhanced attention feature cross fusion deep learning Rank-1

收稿日期: 2021-06-23 发布日期: 2022-12-28

PACS:

TB96

基金资助:河北省科学技术研究与发展计划科技支撑计划(20310302D);河北省中央引导地方专项(199477141G)

通讯作者: 李媛媛(1997-),黑龙江肇东人,燕山大学硕士研究生,研究方向为计算机视觉图像处理。Email:13263516568@163.com E-mail: 13263516568@163.com

作者简介: 金梅(1977-),黑龙江明水人,燕山大学副教授,主要从事计算机视觉、智能信息处理等方面的研究。Email:meijin297@126.com

引用本文:

金梅,李媛媛,郝兴军,杨曼,张立国. 基于非对称增强注意力与特征交叉融合的行人重识别方法[J]. 计量学报, 2022, 43(12): 1573-1580.
JIN Mei,LI Yuan-yuan,HAO Xing-jun,YANG Man,ZHANG Li-guo. Pedestrian Re-identification Method Based on Asymmetric Enhanced Attention and Feature Cross Fusion. Acta Metrologica Sinica, 2022, 43(12): 1573-1580.

链接本文:

http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2022.12.08 或 http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2022/V43/I12/1573

［1］程淑红,程彦龙. 融合批量再标准化和YOLOv3的手势识别研究［J］. 计量学报, 2021, 42(1): 29-34.
Cheng S H,Cheng Y L. Hand Gesture Recognition Algorithm Combined with Batch Renormalization and YOLOv3［J］. Acta Metrologica Sinica, 2021, 42(1): 29-34.
［2］Michela F, Loris B, Alessandro P, et al. Person re-identification by symmetry-driven accumulation of local features［C］// 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Francisco, USA, 2010:2360-2367.
［3］Tetsu M, Takahiro O, Einoshin S, et al. Hierarchical Gaussian Descriptor for Person Re-identification［C］// 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA, 2016:1363-1372.
［4］谭芳,穆平安,马忠雪.基于YOLOv3检测和特征点匹配的多目标跟踪算法［J］.计量学报,2021,42(2):157-162.
Tan F, Mu P A, Ma Z X. Multi-target tracking algorithm based on YOLOv3 detection and feature point matching［J］. Acta Metrologica Sinica, 2021, 42(2):157-162.
［5］赵一中,刘文波.基于深度信念网络的非限制性人脸识别算法研究［J］. 计量学报,2017, 38(1):65-68.
Zhao Y Z, Liu W B. Research on Unrestricted Face Recognition Algorithm Based on Deep Belief Network［J］. Acta Metrologica Sinica, 2017, 38(1):65-68.
［6］Zhou K Y, Yang Y X, Andrea C, et al. Omni-Scale Feature Learning for Person Re-Identification［C］// 2019 IEEE/CVF International Conference on Computer Vision. Seoul, Korea, 2019:3701-3711.
［7］Luo H, Jiang W, Gu Y Z, et al. A Strong Baseline and Batch Normalization Neck for Deep Person Re-identification［J］. IEEE Transactions on Multimedia, 2019, 22(10):2597-2609.
［8］Wu L, Shen C H, Anton V D H. PersonNet: Person Re-identification with Deep Convolutional Neural Networks［DB］. arXiv:1601.07255.
［9］Ejaz A, Michael J, Tim K M, et al. An improved deep learning architecture for person re-identification［C］// 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, USA, 2015:3908-3916.
［10］薛震,于莲芝,胡婵娟. 基于图像序列的运动目标检测识别的关键技术研究［J］. 计量学报, 2020, 41(12): 1475-1481.
Xue Z,Yu L Z,Hu C J. Research on Key Techniques of Moving Target Detection and Recognition Based on Image Sequence［J］. Acta Metrologica Sinica, 2020, 41(12): 1475-1481.
［11］徐家臻, 李婷, 杨巍. 多尺度局部特征选择的行人重识别算法［J］.计算机工程与应用, 2020, 56(2):141-145.
Xu J Z, Li T, Yang W. Pedestrian re-recognition algorithm based on multi-scale local feature selection［J］. Computer Engineering and Applications, 2020, 56(2):141-145.
［12］Sun Y F, Zheng L, Yang Y, et al. Beyond Part Models: Person Retrieval with Refined Part Pooling (and A Strong Convolutional Baseline)［J］. Springer, Cham, 2017, 11208:501-518.
［13］Zhao H Y, Tian M Q, Sun S Y, et al. Spindle Net: Person Re-identification with Human Body Region Guided Feature Decomposition and Fusion［C］// 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA, 2017:907-915.
［14］Chen T L, Ding S J, Xie J Y, et al. ABD-Net: Attentive but Diverse Person Re-Identification［C］// 2019 IEEE/CVF International Conference on Computer Vision. Seoul, Korea, 2019:8350-8360.
［15］Li W, Zhu X T, Gong S G. Harmonious Attention Network for Person Re-identification［C］// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA, 2018:2285-2294.
［16］Xu J, Zhao R, Zhu F, et al. Attention-Aware Compositional Network for Person Re-identification［C］// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA, 2018: 2119-2128.
［17］Tay C T, Roy S, Yap K H. AANet: Attribute Attention Network for Person Re-Identifications［C］// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, USA, 2019: 7127-7136.
［18］Wang C, Song L C, Wang G L, et al. Multi-scale multi-patch person re-identification with exclusivity regularized softmax［J］. Neurocomputing, 2020, 382:64-70.
［19］Wang Y J, Zhang W, Huang D X, et al. Multi-level feature fusion and multi-loss learning for person re-identification［J］.Signal Processing Image Communication, 2021, 94(10):116197.
［20］Zhong Z, Zheng L, Cao D L, et al. Re-ranking person re-identification with k-reciprocal encoding［C］// 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA, 2017:3652-3661.
［21］Zheng Z D, Zheng L, Yang Y. Pedestrian Alignment Network for Large-scale Person Re-Identification［J］. IEEE Transactions on Circuits and Systems for Video Technology, 2019, 29(10):3037-3045.
［22］Lin Y T, Zheng L, Zheng Z D, et al. Improving person re-identification by attribute and identity learning［J］. Pattern Recognition, 2019, 95:151-161.
［23］Chen Y B, Zhu X T, Gong S G. Person Re-identification by Deep Learning Multi-scale Representations［C］// 2017 IEEE International Conference on Computer Vision Workshops. Venice, Italy, 2017:2590-2600.
［24］Li R, Zhang B P, Teng Z, et al. A divide-and-unite deep network for person re-identification［J］. Applied Intelligence, 2021, 51(7):1479-1491.
［25］Luo H, Jiang X, Zhang X, et al. AlignedReID++: Dynamically matching local information for person re-identification［J］. Pattern Recognition, 2019, 94:53-61.
［26］Li S Z, Yu H M, Huang W, et al. Attributes-aided part detection and refinement for person re-identification［J］. Pattern Recognition, 2020, 97: 107016.
［27］Chang X B, Timothy M H, Xiang T. Multi-level Factorisation Net for Person Re-identification［C］// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA, 2018:2109-2118.
［28］Ebetci A, Akgul Y S. End-to-end training of CNN ensembles for person re-identification［J］. Pattern Recognition, 2020, 104: 107319.