logo

SCIENCE CHINA Information Sciences, Volume 62, Issue 9: 199101(2019) https://doi.org/10.1007/s11432-018-9639-7

Joint horizontal and vertical deep learning feature for vehicle re-identification

More info
  • ReceivedMar 27, 2018
  • AcceptedOct 11, 2018
  • PublishedJun 12, 2019

Abstract

There is no abstract available for this article.


Acknowledgment

This work was supported in part by National Natural Science Foundation of China (Grant Nos. 61602191, 61871434, 61802136, 61672521), in part by Natural Science Foundation of Fujian Province (Grant Nos. 2018J01090, 2016J01308), in part by Promotion Program for Young and Middle-aged Teacher in Science and Technology Research of Huaqiao University (Grant Nos. ZQN-PY418, ZQN-YX403), and in part by Scientific Research Funds of Huaqiao University (Grant No. 16BS108).


Supplement

Detailed parameter configuration of the proposed method and performance comparison results.


References

[1] Jin X, Zhu S, Xiao C. 3D textured model encryption via 3D Lu chaotic mapping. Sci China Inf Sci, 2017, 60: 122107 CrossRef Google Scholar

[2] Guo L, Guo C, Li L. Two-stage local constrained sparse coding for fine-grained visual categorization. Sci China Inf Sci, 2018, 61: 018104 CrossRef Google Scholar

[3] Zhu J, Zeng H, Liao S. Deep Hybrid Similarity Learning for Person Re-Identification. IEEE Trans Circuits Syst Video Technol, 2018, 28: 3183-3193 CrossRef Google Scholar

[4] Chen Y C, Zheng W S, Lai J H. An Asymmetric Distance Model for Cross-View Feature Mapping in Person Reidentification. IEEE Trans Circuits Syst Video Technol, 2017, 27: 1661-1675 CrossRef Google Scholar

[5] Liao S C, Hu Y, Zhu X Y, et al. Person re-identification by local maximal occurrence representation and metric learning. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Boston, 2015. 2197--2206. Google Scholar

[6] Zhu J Q, Zeng H Q, Du Y Z, et al. Person re-identification based on novel triplet convolutional neural network. Journal of Electronics and Information Technology, 2018, 40: 1012--1016. Google Scholar

[7] Zhu J Q, Zeng H Q, Lei Z, et al. A shortly and densely connected convolutional neural network for vehicle re-identification. In: Proceedings of International Conference on Pattern Recognition, Beijing, 2018. Google Scholar

[8] Zheng L, Wang S J, Zhou W G, et al. Bayes merging of multiple vocabularies for scalable image retrieval. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Columbus, 2014. 1963--1970. Google Scholar

[9] Liu X, Liu W, Mei T. PROVID: Progressive and Multimodal Vehicle Reidentification for Large-Scale Urban Surveillance. IEEE Trans Multimedia, 2018, 20: 645-658 CrossRef Google Scholar

  • Figure 1

    (Color online) Diagram for the proposed method. The acronyms MP, HAP, VAP, T, SN and CAT denote max pooling, horizontal average pooling, vertical average pooling, transposition, spatial normalization and concatenation layers, respectively. (a) The packaged block of convolutional layer, batch normalization and leaky ReLU layers (CBLR) block;protect łinebreak (b) the short and dense unit (SDU); (c) the framework of joint horizontal and vertical deep feature learning.

Copyright 2020 Science China Press Co., Ltd. 《中国科学》杂志社有限责任公司 版权所有

京ICP备18024590号-1       京公网安备11010102003388号