MMCE-Net: A Multi-Module Collaborative Enhancement Network for Occluded Person Re-Identification

Mingjun Ma; Jun Li

doi:10.47297/taposatWSP2633-456934.20260702

PDF
导出
分享
收藏
专辑

MMCE-Net: A Multi-Module Collaborative Enhancement Network for Occluded Person Re-Identification

7卷第2期

浏览量: 0 Downloads: 0

MMCE-Net: A Multi-Module Collaborative Enhancement Network for Occluded Person Re-Identification
MMCE-Net: A Multi-Module Collaborative Enhancement Network for Occluded Person Re-Identification
College of Control Engineering, Northeast Forestry University, Harbin, Heilongjiang, 150040, China
2026年7卷第2期页码：172-179
纸质出版：2026-02-28
DOI：10.47297/taposatWSP2633-456934.20260702
Accepted：

Scan QR Code

Mingjun Ma & Jun Li. (2026). MMCE-Net: A Multi-Module Collaborative Enhancement Network for Occluded Person Re-Identification. 科技理论与实践（英文版）, 7(2), 172-179.

Ma Mingjun,Li Jun.MMCE-Net: A Multi-Module Collaborative Enhancement Network for Occluded Person Re-Identification[J].Theory and Practice of Science and Technology,2026,07(02):172-179.
Mingjun Ma & Jun Li. (2026). MMCE-Net: A Multi-Module Collaborative Enhancement Network for Occluded Person Re-Identification. 科技理论与实践（英文版）, 7(2), 172-179. DOI： 10.47297/taposatWSP2633-456934.20260702.

Ma Mingjun,Li Jun.MMCE-Net: A Multi-Module Collaborative Enhancement Network for Occluded Person Re-Identification[J].Theory and Practice of Science and Technology,2026,07(02):172-179. DOI： 10.47297/taposatWSP2633-456934.20260702.

摘要

Abstract

Person re-identification aims to match pedestrian images captured by different cameras and remains challenging due to occlusion

pose variation

and background interference in practical scenarios. This paper proposes a collaborative enhancement framework for person re-identification based on a ResNet-50 backbone

with the objective of improving robustness under partial occlusion. The main contribution of this work lies in the integration of a multi-scale attention mechanism that strengthens local feature perception and enhances sensitivity to critical body regions. In addition

an Enhanced Feature Generation Module is designed to fuse shallow and deep features through cross-layer interactions

enabling more complete and discriminative pedestrian representations. By jointly leveraging complementary information from different network stages

the proposed framework effectively mitigates the negative impact of occluded or misaligned regions. Extensive experiments conducted on multiple public datasets

including occluded and conventional person re-identification benchmarks

demonstrate that the proposed method achieves competitive and robust performance compared with representative state-of-the-art approaches

validating its effectiveness in complex real-world scenarios.

关键词

Keywords

references

Zheng, L.; Yang, Y.; Hauptmann, A.G. Person re-identification: Past, present and future. Int. J. Comput. Vis. 2016, 127, 211–238.

Liao, S.; Hu, Y.; Zhu, X.; Li, S.Z. Person re-identification by local maximal occurrence representation. Proc. CVPR 2015, 2197–2206.

Sun, Y.; Zheng, L.; Deng, W.; Wang, S. SVDNet for pedestrian retrieval. Proc. ICCV 2017, 3820–3828.

Wang, G.; Yang, S.; Liu, H.; Wang, Z.; Yang, Y.; Wang, S.; Yu, G.; Zhou, E.; Sun, J. High-order information matters: Learning relation and topology for occluded person re-identification. Proc. CVPR 2020, 6449–6458.

Miao, J.; Wu, Y.; Liu, P.; Ding, Y.; Yang, Y. Pose-guided feature alignment for occluded person re-identification. Proc. ICCV 2019, 542 – 551 .

Gao, S.; Wang, J.; Lu, H.; Liu, Z. Pose-guided visible part matching for occluded person re-identification. Proc. CVPR 2020, 11744–11752.

Fan, X.; Luo, H.; Zhang, X.; He, L.; Zhang, C.; Jiang, W. SCPNet: Spatial-channel parallelism network for joint holistic and partial person re-identification. ACCV 2018.

Sun, Y.; Zheng, L.; Yang, Y.; Tian, Q.; Wang, S. Beyond part models: Person retrieval with refined part pooling. Proc. ECCV 2018.

Fu, Y.; Wei, Y.; Wang, G.; Zhou, Y.; Shi, H.; Huang, T. Horizontal pyramid matching for person re-identification. Proc. AAAI 2019.

Hu, J.; Shen, L.; Sun, G. Squeeze-and-excitation networks. Proc. CVPR 2018.

Sun, Y.; Xu, Q.; Li, Y.; Zhang, C.; Li, Y.; Wang, S.; Sun, J. Perceive where to focus: Visibility-aware part-level features for partial person re-identification. Proc. CVPR 2019.

He, L.; Liang, J.; Li, H.; Sun, Z. Partial person re-identification with aligned spatial features. Proc. ICCV 2018

Liu, J.; Zha, Z.J.; Wu, F. Partial re-identification via probabilistic alignment. IEEE Trans. Image Process. 2020.

Jia, M.; Cheng, X.; Lu, S.; Zhang, J. Learning disentangled representation implicitly via transformer for occluded person re-identification. IEEE Trans. Multimed. 2023.

Mikriukov, G.; Ravanbakhsh, M.; Demir, B. Deep Unsupervised Contrastive Hashing for Large-Scale Cross-Modal Text-Image Retrieval in Remote Sensing. arXiv preprint, 2022, arXiv:2201.08125. https://arxiv.org/abs/2201.08125

Cui, X.; Liang, Y.; Zhang, W. Dual-branch person re-identification algorithm based on multi-feature representation. Electronics 2023.

Bai, X.; Wang, A.; Zhang, C.; Hu, H. Cross-domain person re-identification based on normalized IBN-Net. IEEE Access 2024.

Zhang, S.; Yang, J.; Schiele, B.Occluded person re-identification with occlusion-aware adversarial feature learning.Proc. ICCV, 2021.

Li, S.; Sun, S.; Li, Y.; et al.Learning discriminative representation via self-paced adversarial learning for occluded person re-identification.IEEE Trans. Image Process., 2020.

Wang, G.; Yuan, Y.; Chen, X.; et al. Learning discriminative features with multiple granularities for person re-identification. Proc. ACM MM 2018.

Yang, Y.; Yu, S.; Li, X.; et al. HOReID: High-order relation-aware network for occluded person re-identification. Proc. CVPR 2020.

Wang, T.; Liu, H.; Song, P.; et al. Pose-guided feature disentangling for occluded person re-identification based on Transformer. Proc. AAAI 2022.

Xu, B.; He, L.; Liang, J.; Sun, Z. Learning Feature Recovery Transformer for Occluded Person Re-Identification. IEEE Transactions on Image Processing, 2022, 31, 4651–4662.

文章被引用时，请邮件提醒。

Submit

Research on Network Transmission Anomaly Detection and Optimization Based on Signal Decomposition and LSTM

Related Author

Ning Pan Big Data Center, Hubei University

Related Institution

Big Data Center, Hubei University

⁰