Cross-dataset person re-identification method based on multi-pool fusion and background elimination network

被引：0
|
作者：
Li Y. [1 ]
Zhang B. [1 ]
Sun J. [1 ]
Chen H. [1 ]
Zhu J. [1 ]
机构：
[1] School of Electronic and Information Engineering, Beijing Jiaotong University, Beijing
来源：
Tongxin Xuebao/Journal on Communications | 2020年 / 41卷 / 10期
基金：
中国国家自然科学基金;
关键词：
Background elimination; Cross-dataset; Deep learning; Multi-pool fusion; Person re-identification;
D O I：
10.11959/j.issn.1000-436x.2020181
中图分类号：
TB18 [人体工程学]; Q98 [人类学];
学科分类号：
030303 ; 1201 ;
摘要：
The existing cross-dataset person re-identification methods were generally aimed at reducing the difference of data dis-tribution between two datasets, which ignored the influence of background information on recognition performance. In order to solve this problem, a cross-dataset person re-ID method based on multi-pool fusion and background elimination network was proposed. To describe both global and local features and implement multiple fine-grained representations, a multi-pool fusion network was constructed. To supervise the network to extract useful foreground features, a feature-level supervised background elimination network was constructed. The final network loss function was defined as a multi-task loss, which combined both person classification loss and feature activation loss. Three person re-ID benchmarks were employed to evaluate the proposed method. Using MSMT17 as the training set, the cross-dataset mAP for Market-1501 was 35.53%, which was 9.24% higher than ResNet50. Using MSMT17 as the training set, the cross-dataset mAP for DukeMTMC-reID was 41.45%, which was 10.72% higher than ResNet50. Compared with existing methods, the proposed method shows better cross-dataset person re-ID performance. © 2020, Editorial Board of Journal on Communications. All right reserved.
引用
收藏
页码：70 / 79
页数：9
相关论文
共 27 条
  • [1] CHEN K, CHEN Y, HAN C, Et al., Hard sample mining makes person re-identification more efficient and accurate, Nerocomputing, 2020, 382, pp. 259-267
  • [2] SERBETCI A, AKGUL Y S., End-to-end training of CNN ensembles for person re-identification, Pattern Recognition, Pattern Recognition, 104, (2020)
  • [3] LI Y J, ZHUO L, ZHANG J, Et al., Robust resource allo-cation algorithm for heterogeneous wireless network with SWIPT, ACTA Automatica Sinica, 44, 9, pp. 1554-1568, (2018)
  • [4] GOU M, ZHANG X, RATES-BORRAS A., Person re-identification in appearance impaired scenarios, Proceedings of British Machine Vision Con-ference, pp. 1-14, (2016)
  • [5] MATSUKAWA T, OKABE T, SUZUKI E., Hierarchical gaussian descriptor for person re-identification, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 1363-1372, (2016)
  • [6] KOESTINGER M, HIRZER M, WOHLHART P., Large scale metric learning from equivalence constraints, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2288-2295, (2012)
  • [7] ZHENG W S, GONG S, XIANG T., Re-identification by relative distance comparison, IEEE Transactions on Pattern Analysis and Machine Intelligence, 35, 3, pp. 653-668, (2013)
  • [8] YI D, LEI Z, LIAO S, Et al., Deep metric learning for person re-identification, International Conference on Pattern Recognition, pp. 34-39, (2014)
  • [9] YAO H, ZHANG S, ZHANG Y, Et al., Deep representation learning with part loss for person re-identification, IEEE Transactions on Image Processing, 28, 6, pp. 2860-2871, (2019)
  • [10] SUN Y, ZHENG L, YANG Y, Et al., Beyond part models: person retrieval with refined part pooling (and a strong convolutional baseline), Proceedings of the European Conference on Computer Vision, pp. 501-508, (2018)
123