Vehicle Reidentification Based on Multi-Attribute Adaptive Polymerization Network

doi:10.16183/j.cnki.jsjtu.2024.135

Abstract

Abstract:

To address the insufficient feature perception ability and reduced recognition accuracy in vehicle reidentification due to intra-class differences and inter-class similarity of vehicle targets, a vehicle reidentification method based on multi-attribute adaptive aggregation network (MaAPN) architecture is proposed. First, the ResNet-50 network is used as the backbone network for feature extraction, and the instance-batch normalization (IBN) adaptive module is introduced to extract feature representations with strong domain adaptability. Next, the attributes such as camera perspective, vehicle type, and vehicle color are integrated into the network to develop a multi-attribute self-attention feature enhancement model to enhance the robustness and discriminability of feature representation. Finally, a comprehensive loss function is designed to further improve the accuracy of the network by optimizing the feature distance between samples. The results show that the MaAPN architecture achieves an average accuracy of 87.3% and 86.9% on the VeRi-776 and VERI-Wild datasets respectively, and achieves optimal results on various evaluation indicators, effectively improving the accuracy of vehicle reidentification.

Key words: vehicle reidentification, feature adaptation, multi-attribute polymerization, circle loss function

CLC Number:

U491.2

FAN Xing, GE Fei, JIA Wenwen, XIAO Fangwei. Vehicle Reidentification Based on Multi-Attribute Adaptive Polymerization Network[J]. Journal of Shanghai Jiao Tong University, 2026, 60(1): 123-132.

Figures/Tables 10

Fig.1

Fig.2

Fig.3

Tab.1

Tab.2

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

References 30

[1]	LIU X C, LIU W, MEI T, et al. A deep learning-based approach to progressive vehicle re-identification for urban surveillance[C]// Computer Vision-ECCV 2016. Berlin, Germany: Springer, 2016: 869-884.
[2]	LIU H Y, TIAN Y H, WANG Y W, et al. Deep relative distance learning: Tell the difference between similar vehicles[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016: 2167-2175.
[3]	贺晓东, 王春艳, 孙昊, 等. 基于局部特征与视点感知的车辆重识别算法[J]. 仪器仪表学报, 2022, 43(10): 177-184.
	HE Xiaodong, WANG Chunyan, SUN Hao, et al. Local-features and viewpoint-aware for vehicle re-identification[J]. Chinese Journal of Scientific Instrument, 2022, 43(10): 177-184.
[4]	PENG J J, WANG H B, ZHAO T T, et al. Learning multi-region features for vehicle re-identification with context-based ranking method[J]. Neurocomputing, 2019, 359: 427-437. doi: 10.1016/j.neucom.2019.06.013 URL
[5]	陈冬艳, 彭锦佳, 蒋广琪, 等. 基于局部感知的车辆重识别算法[J]. 计算机工程与设计, 2022, 43(7): 2048-2054.
	CHEN Dongyan, PENG Jinjia, JIANG Guangqi, et al. Local-aware based vehicle re-identification perception[J]. Computer Engineering and Design, 2022, 43(7): 2048-2054.
[6]	ZHAO Y Z, SHEN C H, WANG H B, et al. Structural analysis of attributes for vehicle re-identification and retrieval[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(2): 723-734. doi: 10.1109/TITS.6979 URL
[7]	QIAN W, HE Z Q, CHEN C, et al. Navigating diverse salient features for vehicle re-identification[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(12): 24578-24587. doi: 10.1109/TITS.2022.3190959 URL
[8]	苏育挺, 陆荣烜, 张为. 基于注意力和自适应权重的车辆重识别算法[J]. 浙江大学学报(工学版), 2023, 57(4): 712-718.
	SU Yuting, LU Rongxuan, ZHANG Wei. Vehicle re-identification algorithm based on attention mechanism and adaptive weight[J]. Journal of Zhejiang University (Engineering Science), 2023, 57(4): 712-718.
[9]	GUO H Y, ZHU K, TANG M, et al. Two-level attention network with multi-grain ranking loss for vehicle re-identification[J]. IEEE Transactions on Image Processing, 2019, 28(9): 4328-4338. doi: 10.1109/TIP.2019.2910408 pmid: 30998466
[10]	WANG H B, PENG J J, CHEN D Y, et al. Attri-bute-guided feature learning network for vehicle re-identification[J]. IEEE MultiMedia, 2020, 27(4): 112-121. doi: 10.1109/MMUL.2020.2999464 URL
[11]	PAN X G, LIU X L, SONG B, et al. Vehicle re-identification approach combining multiple attention mechanisms and style transfer[C]// 2022 3rd International Conference on Pattern Recognition and Machine Learning. Chengdu, China: IEEE, 2022: 65-71.
[12]	孙伟, 胡亚华, 代广昭, 等. 用于车辆重识别的部件耦合Transformer网络[J]. 计算机辅助设计与图形学学报, 2023, 35(8): 1289-1298.
	SUN Wei, HU Yahua, DAI Guangzhao, et al. Part coupled Transformer network for vehicle re-identification[J]. Journal of Computer-Aided Design & Computer Graphics, 2023, 35(8): 1289-1298.
[13]	HUANG P X, HUANG R H, HUANG J J, et al. Deep feature fusion with multiple granularity for vehicle re-identification[C]// Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Re-cognition. Long Beach, USA: IEEE, 2019: 80-88.
[14]	ZHU J Q, ZENG H Q, HUANG J C, et al. Vehicle re-identification using quadruple directional deep learning features[J]. IEEE Transactions on Intelligent Transportation Systems, 2020, 21(1): 410-420. doi: 10.1109/TITS.6979 URL
[15]	汪琦, 雪心远, 闵卫东, 等. 跨域联合学习与共享子空间度量的车辆重识别[J]. 中国图象图形学报, 2024, 29(5): 1364-1380.
	WANG Qi, XUE Xinyuan, MIN Weidong, et al. Cross-domain joint learning and shared subspace me-tric for vehicle re-identification[J]. Journal of Image and Graphics, 2024, 29(5): 1364-1380. doi: 10.11834/jig.230405 URL
[16]	PAN X G, LUO P, SHI J P, et al. Two at once: Enhancing learning and generalization capacities via IBN-Net[C]// Proceedings of the European Conference on Computer Vision. Munich, Germany: Springer, 2018: 484-500.
[17]	SUN Y F, CHENG C M, ZHANG Y H, et al. Circle loss: A unified perspective of pair similarity optimization[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, WA, USA: IEEE, 2020: 6397-6406.
[18]	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016: 770-778.
[19]	DENG J, DONG W, SOCHER R, et al. ImageNet: A large-scale hierarchical image database[C]// 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, FL, USA: IEEE, 2009: 248-255.
[20]	HINTON G, VINYALS O, DEAN J. Distilling the knowledge in a neural network[DB/OL]. (2015-03-09)[2024-04-18]. https://arxiv.org/abs/1503.02531.
[21]	HU J, SHEN L, ALBANIE S, et al. Squeeze-and-excitation networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020, 42(8): 2011-2023. doi: 10.1109/TPAMI.2019.2913372 pmid: 31034408
[22]	SZEGEDY C, VANHOUCKE V, IOFFE S, et al. Rethinking the inception architecture for computer vision[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016: 2818-2826.
[23]	LOU Y H, BAI Y, LIU J, et al. VERI-Wild: A large dataset and a new method for vehicle re-identification in the wild[C]// 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, CA, USA: IEEE, 2019: 3230-3238.
[24]	HE L X, LIAO X Y, LIU W, et al. FastReID: A pytorch toolbox for general instance re-identification[C]// Proceedings of the 31st ACM International Conference on Multimedia. Ottawa, Canada: ACM, 2023: 9664-9667.
[25]	JIN X, LAN C L, ZENG W J, et al. Uncertainty-aware multi-shot knowledge distillation for image-based object re-identification[C]// Proceedings of the AAAI Conference on Artificial Intelligence. New York, USA: AAAI, 2020: 11165-11172.
[26]	MENG D C, LI L, LIU X J, et al. Parsing-based view-aware embedding network for vehicle re-identification[C]// 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Seattle, WA, USA: IEEE, 2020: 7101-7110.
[27]	KHORRAMSHAHI P, PERI N, CHEN J C, et al. The devil is in the details: Self-supervised attention for vehicle re-identification[C]// Computer Vision-ECCV 2020. Glasgow, UK: Springer, 2020: 369-386.
[28]	SHEN F, ZHU J Q, ZHU X B, et al. Exploring spatial significance via hybrid pyramidal graph network for vehicle re-identification[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(7): 8793-8804. doi: 10.1109/TITS.2021.3086142 URL
[29]	LI H C, LI C L, ZHENG A H, et al. Attribute and state guided structural embedding network for vehicle re-identification[J]. IEEE Transactions on Image Processing, 2022, 31: 5949-5962. doi: 10.1109/TIP.2022.3202370 pmid: 36083957
[30]	LI H C, LI C L, ZHENG A H, et al. MsKAT: Multi-scale knowledge-aware transformer for vehicle re-identification[J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23(10): 19557-19568. doi: 10.1109/TITS.2022.3166463 URL

对比方法	VeRi-776		VERI-Wild		VERI-Wild		VERI-Wild
	VeRi-776		Test3000(小型)		Test5000(中型)		Test10000(大型)
	mAP	R₁	mAP	R₁	mAP	R₁	mAP	R₁
FastReID^[24]	80.4	96.5	81.9	96.3	75.7	94.5	66.7	91.1
UMTS^[25]	75.9	95.8	72.7	84.5	66.1	79.3	54.2	72.8
PVEN^[26]	79.5	95.6	82.5	96.7	77.0	95.4	69.7	93.4
SAVER^[27]	79.6	96.4	80.9	94.5	75.3	92.7	67.7	89.5
HPGN^[28]	80.2	96.7	80.4	91.4	75.2	88.2	65.0	82.7
ASSEN^[29]	81.3	96.9	80.6	94.9	74.5	91.7	66.2	88.8
MsKAT^[30]	82.0	97.1	84.0	97.3	78.7	95.6	72.2	93.9
SVRN^[7]	84.5	97.2	85.5	97.1	81.5	95.9	76.3	93.9
MaAPN	87.3	97.6	86.9	98.7	83.0	97.1	79.2	96.8

骨干网络	IBN	MaFE	损失函数	VeRi-776		VERI-Wild		VERI-Wild		VERI-Wild
				VeRi-776		Test3000(小型)		Test5000(中型)		Test10000(大型)
				mAP	R₁	mAP	R₁	mAP	R₁	mAP	R₁
√	×	×	L_tri	77.4	91.8	72.3	89.5	64.3	84.9	53.6	80.7
√	√	×	L_tri	79.2	93.1	77.2	92.0	69.9	88.7	61.8	85.4
√	√	√	L_tri	84.9	96.2	84.4	96.5	79.3	94.8	75.2	93.4
√	√	√	L_cir	87.3	97.6	86.9	98.7	83.0	97.1	79.2	96.8