Person re-identification method based on keypoint-based dynamic region partitioning and APLNet

Abstract

To promote the technology of person re-identification (Re-ID) in intelligent video analysis, a new segmentation method of keypoint-based dynamic region partitioning (KDRP) and an improved adaptive average pooling layer list network (APLNet) are proposed in this work. The KDRP addresses the limitations of traditional stripe segmentation methods avoiding the influence of shooting angles and pedestrian postures. The APLNet integrates the adaptive average pooling layer list (AAPLL) module and the priority circle loss (P-circle loss) to solve the problem of inconsistent size of feature map and promote the model performance respectively. Experimental results on different datasets have validated the effectiveness of the proposed method.

Keywords

A

References

[1] Yang B, Chen J, Ye M, et al. . Shallow-deep collaborative learning for unsupervised visible-infrared person re-identification. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 17–21, 2024, Seattle, Washington, USA. 2024, New York, IEEE: 1687016879[C]

[2] PENG C, WANG B, LIU D, et al. Masked attribute description embedding for cloth-changing person re-identification[EB/OL]. (2024-01-11) [2025-03-25]. https://arxiv.org/abs/2401.05646.

[3] WANG Q, QIAN X, LI B, et al. Content and salient semantics collaboration for cloth-changing person re-identification[EB/OL]. (2024-05-26) [2025-03-25]. https://arxiv.org/abs/2405.16597.

[4] Dong N, Yan S, Tang H, et al. . Multiview information integration and propagation for occluded person re-identification. Information fusion. 2024, 104: 102201 J]

[5] Tu H, Liu C, Peng Y, et al. . Clothing-change person re-identification based on fusion of RGB modality and gait features. Signal image and video processing. 2024, 18(3): 2367-2376 J]

[6] Si T, He F, Wu H, et al. . Spatial-driven features based on image dependencies for person re-identification. Pattern recognition. 2022, 124: 108462 J]

[7] Yang J, Zhang C, Li Z, et al. . Discriminative feature mining with relation regularization for person re-identification. Information processing & management. 2023, 60(3): 103295 J]

[8] Pan H, Liu Q, Chen Y, et al. . Pose-aided video-based person re-identification via recurrent graph convolutional network. IEEE transactions on circuits and systems for video technology. 2023, 33(12): 7183-7196 J]

[9] Chen Y, Wang H, Sun X, et al. . Deep attention aware feature learning for person re-identification. Pattern recognition. 2022, 126: 108567 J]

[10] JUNG H, LEE J, YOO J, et al. Paformer: part aware transformer for person re-identification[EB/OL]. (2024-08-12) [2025-03-25]. https://arxiv.org/abs/2408.05918.

[11] ZHU K, GUO H, LIU S, et al. Learning semantics-consistent stripes with self-refinement for person re-identification[J]. IEEE transactions on neural networks and learning systems, 2022.

[12] Zang X, Li G, Gao W. Multidirection and multiscale pyramid in transformer for video-based pedestrian retrieval. IEEE transactions on industrial informatics. 2022, 18(12): 8776-8785 J]

[13] ZHANG Z, HE D, LIU S, et al. Completed part transformer for person re-identification[J]. IEEE transactions on multimedia, 2023.

[14] Liu Z, Zhang L, Zhang D. Neural image parts group search for person re-identification. IEEE transactions on circuits and systems for video technology. 2022, 33(6): 2724-2737 J]

[15] XIE G, SU H, LUO Y, et al. Person re-identification by utilizing hierarchical spatial relation reasoning[J]. Image and vision computing, 2024: 105111.

[16] Sun Y, Zheng L, Li Y, et al. . Learning part-based convolutional features for person re-identification. IEEE transactions on pattern analysis and machine intelligence. 2019, 43(3): 902-917 J]

[17] Sun Y, Xu Q, Li Y, et al. . Perceive where to focus: learning visibility-aware part-level features for partial person re-identification. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 15–20, 2019, Long Beach, CA, USA. 2019, New York, IEEE: 393-402[C]

[18] Fu Y, Wei Y, Zhou Y, et al. . Horizontal pyramid matching for person re-identification. Proceedings of the AAAI Conference on Artificial Intelligence. 201982958302[C]

[19] Siv R, Mancas M, Gosselin B, et al. . People tracking and re-identifying in distributed contexts: extension study of PoseTReID. Proceedings of the 9th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI), October 6–7, 2022, Jakarta, Indonesia. 2022, New York, IEEE: 337342[C]

[20] Liu P, Ma Y. Key points and visible part fusion attention network for occluded pedestrian detection in traffic environments. Optoelectronics letters. 2024, 20(7): 430-436 J]

[21] Walter B. Analysis of convolutional neural network image classifiers in a hierarchical max-pooling model with additional local pooling. Journal of statistical planning and inference. 2023, 224: 109-126 J]

[22] Rafiq M, Parihar S S, Chauhan Y S, et al. . Efficient implementation of max-pooling algorithm exploiting history-effect in ferroelectric-finfets. IEEE transactions on electron devices. 2022, 69(11): 6446-6452 J]

[23] Xu R, Zheng Y, Wang X, et al. . Person re-identification based on improved attention mechanism and global pooling method. Journal of visual communication and image representation. 2023, 94: 103849 J]

[24] KIM B J, KIM S W. Stochastic subsampling with average pooling[EB/OL]. (2024-09-25) [2025-03-25]. https://arxiv.org/abs/2409.16630.

[25] Wu Y, Huang W, Hui S, et al. . Learning knowledge graph embedding with batch circle loss. Proceedings of the International Joint Conference on Neural Networks (IJCNN), July 18–23, 2022, Padua, Italy. 2022, New York, IEEE: 1-8[C]

[26] Che J, He Y, Wu J. Pedestrian multiple-object tracking based on fairmot and circle loss. Scientific reports. 2023, 13(1): 4525 J]

[27] Li H, Zhang T, Li S, et al. . Representation strategy for unsupervised domain adaptation on person re-identification. Optoelectronics letters. 2024, 20(12): 749-756 J]

RIGHTS & PERMISSIONS

Tianjin University of Technology