logo

SCIENCE CHINA Information Sciences, Volume 64 , Issue 2 : 120111(2021) https://doi.org/10.1007/s11432-019-2943-6

Learning generalizable deep feature using triplet-batch-center loss for person re-identification

More info
  • ReceivedSep 12, 2019
  • AcceptedMay 9, 2020
  • PublishedOct 27, 2020

Abstract

There is no abstract available for this article.


Acknowledgment

This work was supported by Zhejiang Lab (Grant No. 2019NB0AB02) and National Natural Science Foundation of China (NSFC) (Grant Nos. 61876212, 61733007, 6157220).


References

[1] Yu Y, Ji Z, Guo J. Zero-Shot Learning via Latent Space Encoding. IEEE Trans Cybern, 2019, 49: 3755-3766 CrossRef Google Scholar

[2] He X W, Zhou Y, Zhou Z C, et al. Triplet-center loss for multi-view 3d object retrieval. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2018. 1945--1954. Google Scholar

[3] Song C F, Huang Y, Ouyang W L, et al. Mask-guided contrastive attention model for person re-identification. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2018. 1179--1188. Google Scholar

[4] Xu J, Zhao R, Zhu F, et al. Attention-aware compositional network for person re-identification. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2018. 2119--2128. Google Scholar

[5] Liu J X, Ni B B, Yan Y C, et al. Pose transferrable person re-identification. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2018. 4099--4108. Google Scholar

[6] Sarfraz M S, Schumann A, Eberle A, et al. A pose-sensitive embedding for person re-identification with expanded cross neighborhood re-rankin. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, 2018. 420--429. Google Scholar

[7] Li W, Zhu X T, Gong S G. Harmonious attention network for person re-identification. In: Proceedings of IEEE conference on computer vision and pattern recognition, 2018. 2285--2294. Google Scholar

[8] Wang C, Zhang Q, Huang C, et al. Mancs: a multi-task attentional network with curriculum sampling for person re-identification. In: Proceedings of European Conference on Computer Vision (ECCV), 2018. 365--381. Google Scholar

[9] Fan X, Jiang W, Luo H. SphereReID: Deep hypersphere manifold embedding for person re-identification. J Visual Communication Image Representation, 2019, 60: 51-58 CrossRef Google Scholar

  • Table 1  

    Table 1Performance (%) comparisons by state-of-the-art methods$^{\rm~a)}$

    2*Method Market1501CUHK03 (labeled)CUHK03 (detected)DukeMTMC-reID
    Rank-1 mAP Rank-1 mAP Rank-1 mAP Rank-1 mAP
    MGCAM [3]84.874.350.150.246.746.9
    AACN [4]85.966.976.859.3
    Pose transfer [5]87.768.933.830.530.128.268.648.4
    PSE [6]87.769.030.227.379.862.0
    HA-CNN [7]91.275.744.441.041.738.680.563.8
    Mancs [8]93.182.369.063.965.560.584.971.8
    SphereReID [9]94.483.666.8*65.7*66.1*63.6*83.968.5
    series TBCL94.284.875.973.372.770.585.874.0

    a) Bold numbers denote the best performance. $*$ represents the result of the implementation.