用于车辆重识别的视角感知局部注意力网络

doi:10.19678/j.issn.1000-3428.0062867

计算机工程 ›› 2022, Vol. 48 ›› Issue (10): 288-297,305. doi: 10.19678/j.issn.1000-3428.0062867

用于车辆重识别的视角感知局部注意力网络

代广昭¹, 孙伟^1,2, 徐凡¹, 张小瑞^2,3,4, 陈旋⁵, 常鹏帅¹, 汤毅¹, 胡亚华¹

1. 南京信息工程大学自动化学院, 南京 210044;
2. 南京信息工程大学江苏省大气环境与装备技术协同创新中心, 南京 210044;
3. 南京信息工程大学数字取证教育部工程研究中心, 南京 210044;
4. 南京信息工程大学无锡研究院, 江苏无锡 214100;
5. 南京信息工程大学计算机与软件学院, 南京 210044

收稿日期:2021-10-03 修回日期:2021-12-14 发布日期:2022-10-09
作者简介:代广昭(1995—),男,硕士研究生,主研方向为模式识别、计算机视觉;孙伟(通信作者),副教授、博士;徐凡,硕士研究生;张小瑞,教授、博士;陈旋、常鹏帅,硕士研究生;汤毅,本科生;胡亚华,硕士研究生。
基金资助:
国家自然科学基金（61304205）；江苏省自然科学基金（BK20191401，BK20201136）；江苏省研究生科研与实践创新计划项目（SJCX21_0363）；大学生创新创业训练项目（XJDC202110300601，202010300290，202010300211，202010300116E）。

View-Aware Part Attention Network for Vehicle Re-Identification

DAI Guangzhao¹, SUN Wei^1,2, XU Fan¹, ZHANG Xiaorui^2,3,4, CHEN Xuan⁵, CHANG Pengshuai¹, TANG Yi¹, HU Yahua¹

1. College of Automation, Nanjing University of Information Science and Technology, Nanjing 210044, China;
2. Jiangsu Collaborative Innovation Center on Atmospheric Environment and Equipment Technology, Nanjing University of Information Science and Technology, Nanjing 210044, China;
3. Engineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science and Technology, Nanjing 210044, China;
4. Wuxi Research Institute, Nanjing University of Information Science and Technology, Wuxi, Jiangsu 214100, China;
5. College of Computer and Software, Nanjing University of Information Science and Technology, Nanjing 210044, China

Received:2021-10-03 Revised:2021-12-14 Published:2022-10-09

摘要/Abstract

摘要： 车辆重识别的目的是从大型车辆数据库中找到与查询车辆相同特征的所有车辆图片。目前，由于同一车辆在不同视角下外观差异大或颜色、车型相同的不同车辆在特定视角下外观差异小，导致车辆重识别的准确度和鲁棒性均有待提高。提出一个视角感知局部注意力网络，采用弱监督注意力学习方式代替人工手动的车辆局部部件标注，自适应学习每个视角内所有显著性局部特征。通过局部注意力裁剪操作裁剪并放大该视角领域内部件细节信息，并基于局部注意力擦除操作擦除一些局部区域，以鼓励模型发掘该视角领域内其他更多的显著性局部线索。构建一种共同视角的注意力增强模块，以强化共同视角特征学习，并根据视角的相似度给每个视角分配相应的权重，使同一视角特征学习得到增强，不同视角特征学习受到抑制。实验结果表明，所提网络在VeRi-776数据集下的mAP为81.2%，在VehicleID数据集下的CMC@1、CMC@5分别为85.7%、98.0%，相较于PRN、PVEN、SAVER等重识别网络具有更高的识别精度和更强的泛化能力。

关键词: 车辆重识别, 注意力机制, 共同视角, 局部感知, 数据增强

Abstract: Vehicle re-identification aims to retrieve all same-identity images from querying vehicle images from a large image database. Currently, the appearance difference of same vehicles under different perspectives is large, whereas the appearance difference of different vehicles under specific perspectives is small probably due to the having same color and model, which leads to a need for the improvement of the accuracy and robustness of vehicle image recognition.A View-Aware Part-Attention Network(VPAN) is proposed, and a weakly-supervised attention-learning method is used to replace manual vehicle local component labeling to adaptively learn all significant local features in each perspective. The detail information of the internal parts in a perspective field is clipped and enlarged by a local attention-clipping operation, and some local areas are erased based on a local attention-erasing operation to encourage the model to discover more significant local clues in the perspective field.A common perspective attention enhancement module is constructed to strengthen common perspective feature learning.Each perspective is assigned a corresponding weight according to the similarity of perspectives, so that same perspective feature-learning is enhanced and different perspective feature-learning is suppressed.The experimental results show that a map of the proposed network with the viri-776 dataset is 81.2%, and that with the vehicleid datasets CMC@1 and CMC@5 are 85.7% and 98.0% respectively. Compared with PRN, PVEN, SARATR and other re-identification networks, the proposed network has higher recognition accuracy and generalization ability.

Key words: vehicle re-identification, attention mechanism, common view, part awareness, data augment

中图分类号:

TP391.41

代广昭, 孙伟, 徐凡, 张小瑞, 陈旋, 常鹏帅, 汤毅, 胡亚华. 用于车辆重识别的视角感知局部注意力网络[J]. 计算机工程, 2022, 48(10): 288-297,305.

DAI Guangzhao, SUN Wei, XU Fan, ZHANG Xiaorui, CHEN Xuan, CHANG Pengshuai, TANG Yi, HU Yahua. View-Aware Part Attention Network for Vehicle Re-Identification[J]. Computer Engineering, 2022, 48(10): 288-297,305.

http://www.ecice06.com/CN/Y2022/V48/I10/288

图/表 13

参考文献

[1] ŠPAŇHEL J, SOCHOR J, JURÁNEK R, et al.Holistic recognition of low quality license plates by CNN using track annotated data[C]//Proceedings of the 14th IEEE International Conference on Advanced Video and Signal Based Surveillance.Washington D.C., USA:IEEE Press, 2017:1-6.
[2] LIU X, LIU W, MEI T, et al.A deep learning-based approach to progressive vehicle re-identifification for urban surveillance[C]//Proceedings of European Conference on Computer Vision.Berlin, Germany:Springer, 2016:869-884.
[3] LIU X C, LIU W, MEI T, et al.PROVID:progressive and multimodal vehicle reidentification for large-scale urban surveillance[J].IEEE Transactions on Multimedia, 2018, 20(3):645-658.
[4] LIU H, TIAN Y, YANG Y, et al.Deep relative distance learning:tell the difference between similar vehicles[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:2167-2175.
[5] WANG Z D, TANG L M, LIU X H, et al.Orientation invariant feature embedding and spatial temporal regularization for vehicle re-identification[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2017:379-387.
[6] KHORRAMSHAHI P, KUMAR A, PERI N, et al.A dual-path model with adaptive attention for vehicle re-identification[C]//Proceedings of IEEE International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2019:6131-6140.
[7] HE B, LI J, ZHAO Y F, et al.Part-regularized near-duplicate vehicle re-identification[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2019:3992-4000.
[8] 李熙莹, 周智豪, 邱铭凯.基于部件融合特征的车辆重识别算法[J].计算机工程, 2019, 45(6):12-20. LI X Y, ZHOU Z H, QIU M K.Vehicle re-identification algorithm based on component fusion feature[J].Computer Engineering, 2019, 45(6):12-20.(in Chinese)
[9] LIU X C, LIU W, ZHENG J K, et al.Beyond the parts:learning multi-view cross-part correlation for vehicle re-identification[C]//Proceedings of the 28th ACM International Conference on Multimedia.New York, USA:ACM Press, 2020:907-915.
[10] CHEN X, SUI H G, FANG J, et al.Vehicle re-identification using distance-based global and partial multi-regional feature learning[J].IEEE Transactions on Intelligent Transportation Systems, 2021, 22(2):1276-1286.
[11] ZHENG B, LEI Z B, TANG C, et al.OERFF:a vehicle re-identification method based on orientation estimation and regional feature fusion[J].IEEE Access, 2021, 9:66661-66674.
[12] ZHOUY Y, SHAO L.Viewpoint-aware attentive multi-view inference for vehicle re-identification[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:6489-6498.
[13] GOODFELLOW J, POUGET-ABADIE J, MIRZA M, et al.Generative adversarial nets[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems.Cambridge, USA:MIT Press, 2014:2672-2680.
[14] GAO L, ZHANG J, ZHANG L F, et al.DSP:dual soft-paste for unsupervised domain adaptive semantic segmentation[C]//Proceedings of the 29th ACM International Conference on Multimedia.New York, USA:ACM Press, 2021:2825-2833.
[15] ZHUANG W M, WEN Y G, ZHANG S.Joint optimization in edge-cloud continuum for federated unsupervised person re-identification[C]//Proceedings of the 29th ACM International Conference on Multimedia.New York, USA:ACM Press, 2021:433-441.
[16] HU T, QI H G, HUANG Q M, et al.See better before looking closer:weakly supervised data augmentation network for fine-grained visual classification[EB/OL].[2021-09-01].https://arxiv.org/abs/1901.09891.
[17] CHU R H, SUN Y F, LI Y D, et al.Vehicle re-identification with viewpoint-aware metric learning[C]//Proceedings of IEEE/CVF International Conference on Computer Vision.Washington D.C., USA:IEEE Press, 2019:8281-8290.
[18] MENG D C, LI L, LIU X J, et al.Parsing-based view-aware embedding network for vehicle re-identification[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2020:7101-7110.
[19] RONNEBERGER O, FISCHER P, BROX T.U-net:convolutional networks for biomedical image segmentation[EB/OL].[2021-09-01].https://arxiv.org/abs/1505.04597v1.
[20] CHEN T S, LIU C T, WU C W, et al.Orientation-aware vehicle re-identification with semantics-guided part attention network[C]//Proceedings of the European Conference on Computer Vision.Berlin, Germany:Springer, 2020:330-346.
[21] HERMANS A, BEYER L, LEIBE B.In defense of the triplet loss for person re-identification[EB/OL].[2017-09-01].https://arxiv.org/abs/1703.07737.
[22] RUSSAKOVSKY O, DENG J, SU H, et al.ImageNet large scale visual recognition challenge[J].International Journal of Computer Vision, 2015, 115(3):211-252.
[23] HE K M, ZHANG X Y, REN S Q, et al.Deep residual learning for image recognition[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2016:770-778.
[24] HU J, SHEN L, SUN G.Squeeze-and-excitation networks[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA.IEEE Press, 2018:7132-7141.
[25] LUO H, GU Y Z, LIAO X Y, et al.Bag of tricks and a strong baseline for deep person re-identification[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops.Washington D.C., USA:IEEE Press, 2019:1487-1495.
[26] KINGMA D P, BA J.Adam:a method for stochastic optimization[EB/OL].[2021-09-01].https://arxiv.org/abs/1412.6980.
[27] HUANG H J, LI D W, ZHANG Z, et al.Adversarially occluded samples for person re-identification[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.New York, USA:ACM Press, 2019:5098-5107.
[28] ZHONG Z, ZHENG L, KANG G L, et al.Random erasing data augmentation[EB/OL].[2021-09-01].https://arxiv.org/abs/1708.04896.
[29] SUN W, ZHANG X R, HE X Z, et al.A two-stage vehicle type recognition method combining the most effective Gabor features[J].Computers, Materials & Continua, 2020, 65(3):2489-2510.
[30] SUN W, ZHANG X R, SHI S S, et al.Vehicle classification approach based on the combined texture and shape features with a compressive DL[J].IET Intelligent Transport Systems, 2019, 13(7):1069-1077.
[31] LIU X C, LIU W, MA H D, et al.Large-scale vehicle re-identification in urban surveillance videos[C]//Proceedings of IEEE International Conference on Multimedia and Expo.Washington D.C., USA:IEEE Press, 2016:1-6.
[32] BAI Y, LOU Y H, GAO F, et al.Group-sensitive triplet embedding for vehicle reidentification[J].IEEE Transactions on Multimedia, 2018, 20(9):2385-2399.
[33] ZHU Y C, ZHA Z J, ZHANG T Z, et al.A structured graph attention network for vehicle re-identification[C]//Proceedings of the 28th ACM International Conference on Multimedia.New York, USA:ACM Press, 2020:646-654.
[34] LOU Y H, BAI Y, LIU J, et al.Embedding adversarial learning for vehicle re-identification[J].IEEE Transactions on Image Processing, 2019, 28(8):3794-3807.
[35] KAMENOU E, DEL RINCON J M, MILLER P, et al.Multi-level deep learning vehicle re-identification using ranked-based loss functions[C]//Proceedings of 25th International Conference on Pattern Recognition.New York, USA:ACM Press, 2021:9099-9106.
[36] GUO H Y, ZHAO C Y, LIU Z W, et al.Learning coarse-to-fine structured feature embedding for vehicle re-identification[J].Proceedings of the AAAI Conference on Artificial Intelligence, 2018, 32(1):34-42.
[37] KHORRAMSHAHI P, PERI N, CHEN J C, et al.The devil is in the details:self-supervised attention for vehicle re-identification[C]//Proceedings of the European Conference on Computer Vision.Berlin, Germany:Springer, 2020:369-386.
[38] SUN Z R, NIE X S, XI X M, et al.CFVMNet:a multi-branch network for vehicle re-identification based on common field of view[C]//Proceedings of the 28th ACM International Conference on Multimedia.New York, USA:ACM Press, 2020:3523-3531.

选择文件类型/文献管理软件名称

选择包含的内容

用于车辆重识别的视角感知局部注意力网络

View-Aware Part Attention Network for Vehicle Re-Identification

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	丰芳宇, 罗晓曙, 蒙志明, 王广宇. 基于抗混叠残差注意力网络的人脸表情识别[J]. 计算机工程, 2023, 49(8): 190-198.
[2]	江雨燕, 陶承凤, 李平. 数据增强和自适应自步学习的深度子空间聚类算法[J]. 计算机工程, 2023, 49(8): 96-103, 110.
[3]	王书朋, 何引弟. 融合特征注意力机制的非均匀光照图像增强算法[J]. 计算机工程, 2023, 49(8): 232-239.
[4]	刘昊鑫, 董超, 勾智楠, 高凯. 融合混合表征的小样本关系抽取方法[J]. 计算机工程, 2023, 49(8): 63-68.
[5]	杨长沛, 廖列法. 基于门控空洞卷积特征融合的中文命名实体识别[J]. 计算机工程, 2023, 49(8): 85-95.
[6]	刘俊豪, 王美林, 谢兴, 宋烨兴, 许莉花. 基于改进YOLOv5的皮革瑕疵检测算法[J]. 计算机工程, 2023, 49(8): 240-249.
[7]	马娜, 温廷新, 贾旭, 李晓会. 复杂光照条件下自适应的车脸重识别模型[J]. 计算机工程, 2023, 49(8): 275-282, 290.
[8]	陈露萌, 曹彦彦, 黄民, 谢鑫钢. 基于改进YOLOv5的火焰检测方法[J]. 计算机工程, 2023, 49(8): 291-301, 309.
[9]	李强龙, 周新文, 位梦恩, 甘阳洲. 基于条形池化和注意力机制的街道场景红外目标检测算法[J]. 计算机工程, 2023, 49(8): 310-320.
[10]	张家熔, 苑津莎, 许珈宁, 罗志宏. 基于多元信息嵌入与协同神经网络的力学实体识别算法[J]. 计算机工程, 2023, 49(7): 125-134.
[11]	白明昌. 基于折叠路径聚合的属性网络节点嵌入方法[J]. 计算机工程, 2023, 49(7): 76-84.
[12]	费蓉, 马梦阳, 张晓, 黑新宏, 徐庆征, 邱原. 基于轨迹预测与冲突检测的自动驾驶碰撞检测模型[J]. 计算机工程, 2023, 49(7): 10-20.
[13]	刘豪, 吴红兰, 房宇轩. 结合全局上下文信息的高效人体姿态估计[J]. 计算机工程, 2023, 49(7): 102-109.
[14]	吴珊, 周凤. 基于改进SSD算法的小目标检测[J]. 计算机工程, 2023, 49(7): 179-188.
[15]	齐咏生, 杜晓旭, 朱俊峰, 高胜利, 刘利强. 基于增强型轻量深度网络的牧区牲畜高效检测[J]. 计算机工程, 2023, 49(7): 278-287.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

用于车辆重识别的视角感知局部注意力网络

View-Aware Part Attention Network for Vehicle Re-Identification

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 13

参考文献

相关文章 15

编辑推荐

Metrics

本文评价