Attention Based Deep Learning for Vehicle Re-Identification

Hato, Siham

doi:Under indexing

Proceedings of International Conference on Applied Innovation in IT · 2026/06/12 · Vol. 14 · Issue 3 · pp. 115–124

Attention Based Deep Learning for Vehicle Re-Identification

Siham Hussein Ali and Eman Hato

📄 Download PDF DOI: Under indexing

Abstract

Vehicle Re-Identification (Re-ID) is a crucial task in intelligent transportation systems, where vehicles are matched from different camera angles. While deep learning has significantly improved this process, challenges related to varying camera angles, obscuring, lighting variations, and similar factors persist. Attention mechanisms assign different weights to different input elements, enabling the model to focus more precisely on the important information that improves its performance. This paper presents the impact of a multi-level attention fusion framework for vehicle identification, integrating the ResNet-50 architecture with multiple attention mechanisms, including spatial attention mechanisms, cross-channel attention, self-attention, and mutual attention. This design allows the network to learn both precise local identity indicators and overall contextual relationships in a unified and efficient manner. Extensive experiments on the VeRi-776 dataset demonstrate that attention fusion with a two-stage convolutional neural network mechanism outperformed previous methods in terms of accuracy and robustness across the dataset's metrics, achieving the highest scores of 85.1% on the mAP scale and 94.1% on the Rank-1 scale. Therefore, these results suggest that future work should incorporate adaptive, multi-level attention.

Keywords

Attention Mechanisms ResNet-50 Architecture VeRi-776 Dataset Mean Average Precision Rank-1 Accuracy Vehicle Re-Identification.

References

A. Farid, F. Hussain, K. Khan, M. Shahzad, U. Khan, and Z. Mahmood, “A fast and accurate real-time vehicle detection method using deep learning for unconstrained environments,” Applied Sciences, vol. 13, no. 5, 2023, doi: 10.3390/app13053059.
S. H. Mousa, N. M. Shati, and N. Sakthivadivel, “DeepRing: Convolution neural network based on blockchain technology,” Al-Mustansiriyah Journal of Science, vol. 35, no. 2, pp. 61-69, Jun. 2024.
S. Abdulhaleem and E. Hato, “Video skimming using object detection and relation-based importance score,” International Journal of Intelligent Engineering Systems, vol. 18, no. 8, pp. 485-498, 2025, doi: 10.22266/ijies2025.0930.30.
G. Szűcs, R. Borsodi, and D. Papp, “Multi-camera trajectory matching based on hierarchical clustering and constraints,” Multimedia Tools and Applications, vol. 83, no. 15, pp. 44879-44902, 2024.
M. R. Shuvo, M. S. Mekala, and E. Elyan, “Deep learning and attention-based methods for human activity recognition and anticipation: A comprehensive review,” Cognitive Computation, 2025, doi: 10.1007/s12559-025-10513-2.
C. M. Mylonakis, P. Velanas, P. I. Lazaridis, P. Sarigiannidis, S. K. Goudos, and Z. D. Zaharis, “Deep learning framework using spatial attention mechanisms for adaptable angle estimation across diverse array configurations,” Technologies, vol. 13, no. 2, 2025, doi: 10.3390/technologies13020046.
Y. Wang, W. Wang, Y. Li, Y. Jia, Y. Xu, Y. Ling, and J. Ma, “An attention mechanism module with spatial perception and channel information interaction,” Complex & Intelligent Systems, vol. 10, no. 4, pp. 5427-5444, 2024.
B. Alnasyan, M. Basheri, M. Alassafi, and K. Alnasyan, “Kanformer: An attention-enhanced deep learning model for predicting student performance in virtual learning environments,” Social Network Analysis and Mining, vol. 15, no. 1, 2025.
M. Gheini, X. Ren, and J. May, “Cross-attention is all you need: Adapting pretrained transformers for machine translation,” in Proc. 2021 Conference on Empirical Methods in Natural Language Processing, pp. 1754-1765, 2021.
Y. Qian, J. Barthelemy, E. Karuppiah, and P. Perez, “Identifying re-identification challenges: Past, current and future trends,” SN Computer Science, vol. 5, no. 7, 2024.
X. Sun, Y. Chen, Y. Duan, Y. Wang, J. Zhang, B. Su, and L. Li, “Vehicle re-identification method based on multi-attribute dense linking network combined with distance control module,” Frontiers in Neurorobotics, vol. 17, 2023, doi: 10.3389/fnbot.2023.1294211.
L. Fei and B. Han, “Multi-object multi-camera tracking based on deep learning for intelligent transportation: A review,” Sensors, 2023.
Y. Lou, Y. Bai, J. Liu, S. Wang, and L.-Y. Duan, “VERI-wild: A large dataset and a new method for vehicle re-identification in the wild,” in Proc. IEEE Conference on Computer Vision and Pattern Recognition, pp. 3230-3238, 2019.
H. Wang, J. Hou, and N. Chen, “A survey of vehicle re-identification based on deep learning,” IEEE Access, 2019, doi: 10.1109/ACCESS.2019.2956172.
Y. Bai, Y. Lou, F. Gao, S. Wang, Y. Wu, and L.-Y. Duan, “Group-sensitive triplet embedding for vehicle re-identification,” IEEE Transactions on Multimedia, vol. 20, no. 9, pp. 2385-2399, 2018.
Z. Sun, X. Wang, Y. Zhang, Y. Song, J. Zhao, J. Xu, W. Yan, and C. Lv, “A comprehensive review of pedestrian re-identification based on deep learning,” Complex & Intelligent Systems, vol. 10, no. 2, 2024, doi: 10.1007/s40747-023-01229-7.
S. Liu and S. S. Agaian, “VERI-D: A new dataset and method for multi-camera vehicle re-identification of damaged cars under varying lighting conditions,” APL Machine Learning, vol. 2, no. 1, 2024.
R. S. M. Saad, M. M. Moussa, N. S. Abdel-Kader, and H. Farouk, “Deep video-based person re-identification (Deep Vid-ReID): Comprehensive survey,” EURASIP Journal on Advances in Signal Processing, vol. 2024, no. 1, 2024.
A. Zheng, X. Lin, J. Dong, W. Wang, J. Tang, and B. Luo, “Multi-scale attention vehicle re-identification,” Neural Computing and Applications, 2020, doi: 10.1007/s00521-020-05108-x.
L. Rong, Y. Xu, X. Zhou, L. Han, L. Li, and X. Pan, “A vehicle re-identification framework based on the improved multi-branch feature fusion network,” Scientific Reports, vol. 11, no. 1, 2021.
H. Li, Y. Wang, Y. Wei, L. Wang, and L. Ge, “Discriminative-region attention and orthogonal-view generation model for vehicle re-identification,” Applied Intelligence, vol. 53, no. 1, pp. 186-203, 2023.
W. Pan, L. Huang, J. Liang, L. Hong, and J. Zhu, “Progressively hybrid transformer for multi-modal vehicle re-identification,” Sensors, vol. 23, no. 9, 2023.
F. Shen, Y. Xie, J. Zhu, X. Zhu, and H. Zeng, “GiT: Graph interactive transformer for vehicle re-identification,” IEEE Transactions on Image Processing, vol. 32, 2023.
Z. Wu, Z. Jin, and X. Li, “Two-stage progressive learning for vehicle re-identification in variable illumination conditions,” Electronics, vol. 12, no. 24, 2023.
X. Guo, J. Yang, X. Jia, C. Zang, and Z. Chen, “A novel dual-pooling attention module for UAV vehicle re-identification,” Scientific Reports, vol. 14, no. 1, 2024, doi: 10.1038/s41598-024-52225-x.
Y. Huang, H. Sheng, and W. Ke, “RANDnet: Vehicle re-identification with relation attention and nuance-disparity masks,” Applied Sciences, vol. 14, no. 11, 2024.
A. Holla B., M. M. Pai, U. Verma, and R. M. Pai, “MSFFT: Multi-scale feature fusion transformer for cross platform vehicle re-identification,” Neurocomputing, vol. 582, 2024.
L. Bai and L. Rong, “Vehicle re-identification with multiple discriminative features based on non-local-attention block,” Scientific Reports, vol. 14, no. 1, 2024, doi: 10.1038/s41598-024-82755-3.
Y. Wang, R. Li, and Y. Shao, “Vehicle re-identification method based on efficient self-attention CNN-transformer and multi-task learning optimization,” Sensors, vol. 25, no. 10, 2025, doi: 10.3390/s25102977.
T. Ma, K. Sun, X. Pang, W. Si, T. Liu, and C. Wang, “Multi-axis compression fusion network for vehicle re-identification,” Scientific Reports, vol. 15, no. 1, 2025, doi: 10.1038/s41598-025-15854-4.