Development of a method of re-identification of a person

Authors

  • O.M. Kyrylenko Vinnytsia National Technical University

DOI:

https://doi.org/10.31649/1681-7893-2021-41-1-25-32

Keywords:

deep learning, human re-identification, OSNet, PyTorch, Market-1501, DukeMTMC-reID

Abstract

The review of OSNet neural network architecture is made for the purpose of training of own models of re-identification of the person. The structure of the neural network was also considered. Existing data sets for model training are investigated. Models were trained using PyTorch. The obtained own models were tested on the validation databases Market-1501 and DukeMTMC-reID. The results of learning neural network models are presented. The results are obtained in comparison with existing analogues.

References

O. Kyrylenko, R. Maslii, and Y. Marushchak «Analysis of methods of person reidentification in multi camera environment», Norwegian Journal of development of the International Science, №47, pp. 46-48, 2020.

O. Kyrylenko, R. Kvyetnyy and R. Maslii, "Research of human attributes for the problem of re-identification", Information Technology and Computer Engineering, vol. 49, issue 3, p. 4–13, 2020.

H. Yu and W. Zheng "Weakly supervised discriminative feature learning with state information for person identification" Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR) 2020.

K. Zhou, Yo. Yang, A. Cavallaro, Tao Xiang “Omni-Scale Feature Learning for Person Re-Identification”, Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 3702-3712.

K. Zhou, Yo. Yang, A. Cavallaro, Tao Xiang “Learning Generalisable Omni-Scale Representations for Person Re-Identification”, Proceedings of the IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021.

L. Zheng L. Shen L. Tian S. Wang J. Wang and Q. Tian "Scalable person re-identification: A benchmark" Proc. IEEE Int. Conf. Comput. Vis. pp. 1116-1124 Jun. 2015.

Wei Li, Rui Zhao, Tong Xiao, and Xiaogang Wang. Deepreid: Deep filter pairing neural network for person reidentification. In CVPR, 2014.

Ergys Ristani, Francesco Solera, Roger Zou, Rita Cucchiara, and Carlo Tomasi. Performance measures and a data set for multi-target, multi-camera tracking. In ECCVW, 2016.

Evaluation Metrics. Cumulative Matching Characteristics. [Electronic resource] – Access mode: https://cysu.github.io/open-reid/notes/evaluation_metrics.html

О. Bubenshchikov, E. Lepa. "The use of convolutional neural networks to identify a person" Bulletin of the Kherson National Technical University, №1 (68), 2019, p. 136-142.

Wu, S.; Gao, L. Multi-Level Joint Feature Learning for Person Re-Identification. Algorithms 2020, 13, 111. https://doi.org/10.3390/a13050111

K. Zhou and T. Xiang "Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch".

Zijun Zhang et al. Normalized direction-preserving Adam. arXiv:1709.04546v2, 2017

The Softmax function and its derivative. [Electronic resource] – Access mode: https://eli.thegreenplace.net/2016/the-softmax-function-and-its-derivative/

Z. Zhedong, N. Zheng and Yi Yang. "Parameter-efficient person re-identification in the 3d space." arXiv preprint arXiv:2006.04569, 2020.

Downloads

Abstract views: 130

Published

2022-05-02

How to Cite

[1]
O. Kyrylenko, “Development of a method of re-identification of a person”, Опт-ел. інф-енерг. техн., vol. 41, no. 1, pp. 25–32, May 2022.

Issue

Section

OptoElectronic/Digital Methods and Systems for Image/Signal Processing

Metrics

Downloads

Download data is not yet available.