Research on Person Re-Identification Based on Deep Learning under Big Data Environment

被引：0
作者：
Li P. [1 ]
Wang D.-Y. [1 ]
Shi W.-X. [1 ]
Jiang Z.-G. [2 ]
机构：
[1] China Academy of Electronics and Information Technology, Xinjiang Lianhai INA-INT Information Technology Ltd, Beijing
[2] Beihang University, School of Astronautics, Beijing
来源：
Beijing Youdian Daxue Xuebao/Journal of Beijing University of Posts and Telecommunications | 2019年 / 42卷 / 06期
关键词：
Capsule networks; Convolutional neural networks; Digital capsule; Person re-identification; Primary capsule;
D O I：
10.13190/j.jbupt.2019-124
中图分类号：
学科分类号：
摘要：
Convolutional neural networks produce higher probability of error for person re-identifications. To overcome the shortcomings, a new deep learning method based on capsule networks model for person re-identification was proposed. First, the standard convolutional layers are used to learn discriminative features. Then, several features in different layers are grouped together to form the primary capsules which represent a rich semantic features. After that, a dynamic routing algorithm which is an iterative routing process, is introduced to decide the attribution between primary capsule and digital capsule. To this end, the digital capsule layer is obtained and each capsule can learn to recognize the presence of persons. To highlight the superiorities of the proposed algorithm, extensive experiments are conducted on a series of challenging datasets and show that the algorithm favorably performs against the previous work. © 2019, Editorial Department of Journal of Beijing University of Posts and Telecommunications. All right reserved.
引用
收藏
页码：29 / 34
页数：5
相关论文
共 15 条
  • [1] Yuan Y., Fang J., Wang Q., Online anomaly detection in crowd scenes via structure analysis, IEEE Transactions on Cybernetics, 45, 3, pp. 548-561, (2015)
  • [2] Jin K.H., McCann M.T., Froustey E., Et al., Deep convolutional neural network for inverse problems in imaging, IEEE Transactions on Image Processing, 26, 9, pp. 4509-4522, (2017)
  • [3] Kruthiventi S.S.S., Ayush K., Babu R.V., DeepFix: a fully convolutional neural network for predicting human eye fixations, IEEE Transactions on Image Processing, 26, 9, pp. 4446-4456, (2017)
  • [4] Zhou S., Wang J., Shi R., Et al., Large margin learning in set to set similarity comparison for person Reidentification, IEEE Transactions on Multimedia, 20, 3, pp. 593-604, (2018)
  • [5] Su C., Zhang S., Yang F., Et al., Attributes driven tracklet-to-tracklet person re-identification using latent prototypes space mapping, Pattern Recognition, 66, pp. 4-15, (2017)
  • [6] Franco A., Oliveira L., Convolutional covariance features: Conception, integration and performance in person re-identification, Pattern Recognition, 61, pp. 593-609, (2017)
  • [7] Xian Y., Hu H., Enhanced multi-dataset transfer learning method for unsupervised person re-identification using co-training strategy, IET Computer Vision, 12, 8, pp. 1219-1227, (2018)
  • [8] Dong X., Thanou D., Frossard P., Et al., Learning laplacian matrix in smooth graph signal representations, IEEE Transactions on Signal Processing, 64, 23, pp. 6160-6173, (2016)
  • [9] Srivastava N., Hinton G.E., Krizhevsky A., Sutskever I., Salakhutdinov R., Dropout: a simple way to prevent neural networks from overfitting, Journal of Machine Learning Research, 15, 1, pp. 1929-1958, (2014)
  • [10] Sainath T.N., Kingsbury B., Saon G., Et al., Deep convolutional neural networks for large-scale speech tasks, Neural Networks, 64, pp. 39-48, (2015)
12