i-manager Publications

Multi-View Pose Variation Using Aligned Person RE-Identification

Imam Ahamed*

Northern Border University, Saudi Arabia.

Periodicity:March - May'2021
DOI : https://doi.org/10.26634/jit.10.2.18432

Abstract

A person's body silhouette recorded by one or more cameras must be recognised across a wide range of potential targets to complete the person's re-identification job. The person re-identification issue is the process of matching together pictures of the same pedestrian taken with several cameras. The issue is further by the pictures' poor resolution, variations in lighting, and the appearance of held items like a bag from various vantage points. The major problems arise from the fact that several images of the same individual are taken at various points in time and with various cameras. The aligned re-id method may help with some of the problems associated with person re-identification. By using Aligned Re- ID, a global feature may be extracted and learned together with local ones. We integrate several losses to limit the model since the intra-class distance in person re-identification should be smaller than the inter-class distance. Using three common benchmark datasets, Market-1501, CUHK03 (detected and labelled), and CUHK-SYSU, we test our method's performance and find that it outperforms existing methods.

Keywords

Person re-identification, Aligned Re-ID, Pose variation, CNN.

How to Cite this Article?

Ahamed, I. (2021). Multi-View Pose Variation Using Aligned Person RE-Identification. i-manager's Journal on Information Technology, 10(2), 1-12. https://doi.org/10.26634/jit.10.2.18432

References

[1]. Cheng, D., Gong, Y., Zhou, S., Wang, J., & Zheng, N. (2016). Person re-identification by multi-channel partsbased cnn with improved triplet loss function. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 1335-1344).

[2]. Lin, Y., Zheng, L., Zheng, Z., Wu, Y., Hu, Z., Yan, C., & Yang, Y. (2019). Improving person re-identification by attribute and identity learning. Pattern Recognition, 95, 151-161.

[3]. Varior, R. R., Haloi, M., & Wang, G. (2016, October). Gated siamese convolutional neural network architecture for human re-identification. In European Conference on Computer Vision (pp. 791-808). Springer, Cham.

[4]. Xiao, Q., Luo, H., & Zhang, C. (2017). Margin sample mining loss: A deep learning based method for person reidentification. arXiv preprint arXiv:1710.00478.

[5]. Zhao, R., Oyang, W., & Wang, X. (2016). Person reidentification by saliency learning. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(2), 356-370.

[6]. Zheng, L., Huang, Y., Lu, H., & Yang, Y. (2019). Poseinvariant embedding for deep person re-identification. IEEE Transactions on Image Processing, 28(9), 4500-4509.

[7]. Zheng, L., Shen, L., Tian, L., Wang, S., Wang, J., & Tian, Q. (2015). Scalable person re-identification: A benchmark. In Proceedings of the IEEE International Conference on Computer Vision (pp. 1116-1124).

[8]. Zheng, L., Yang, Y., & Hauptmann, A. G. (2016). Person re-identification: Past, present and future. arXiv preprint arXiv, 1-20.

[9]. Zhou, X., Yang, C., & Yu, W. (2012). Moving object detection by detecting contiguous outliers in the low-rank representation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(3), 597-610.

[10]. Zhuang, L., Gao, H., Lin, Z., Ma, Y., Zhang, X., & Yu, N. (2012, June). Non-negative low rank and sparse graph for semi-supervised learning. In 2012, IEEE Conference on Computer Vision and Pattern Recognition (pp. 2328-2335). IEEE.

If you have access to this article please login to view the article or kindly login to purchase the article

Single Article

	North Americas,UK, Middle East,Europe		India	Rest of world
	USD	EUR	INR	USD-ROW
Online	15	15

ADD TO CART

Options for accessing this content:

If you would like institutional access to this content, please recommend the title to your librarian.
If you already have i-manager's user account: Login above and proceed to purchase the article.
New Users: Please register, then proceed to purchase the article.