Person Re-identification Based on Multi-information Flow Convolutional Neural Network

SANG Hai-feng; WANG Chuan-zheng; L; Uuml; Ying-yu

doi:10.3969/j.issn.0372-2112.2019.02.014

您当前的位置：

首页 >

文章列表页 >

Person Re-identification Based on Multi-information Flow Convolutional Neural Network

更新时间：2025-07-16

- Person Re-identification Based on Multi-information Flow Convolutional Neural Network
- Acta Electronica Sinica Vol. 47, Issue 2, Pages: 351-357(2019)
- 作者机构：
  
  1. 沈阳工业大学信息科学与工程学院,辽宁,沈阳,110870
  2. 东北大学信息科学与工程学院,辽宁,沈阳,110819
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China (No.61773105);Natural Science Foundation of Liaoning Province (No.20170540675);Research Project of Education Department of Liaoning Province (No.LQGD2017023)
- DOI：10.3969/j.issn.0372-2112.2019.02.014
  CLC： TP391
- Published Online：25 February 2019，
  
  Published：2019
- 稿件说明：
移动端阅览
SANG Hai-feng, WANG Chuan-zheng, L, et al. Person Re-identification Based on Multi-information Flow Convolutional Neural Network[J]. Acta Electronica Sinica, 2019, 47(2): 351-357.
DOI：

SANG Hai-feng, WANG Chuan-zheng, L, et al. Person Re-identification Based on Multi-information Flow Convolutional Neural Network[J]. Acta Electronica Sinica, 2019, 47(2): 351-357. DOI： 10.3969/j.issn.0372-2112.2019.02.014.

摘要

行人再识别问题中，由于视角、光照和行人姿态等因素的变化，导致难以提取有效的行人特征，降低识别精度.而深度神经网络在训练样本较少的情况下较难训练，易出现过拟合现象.针对上述问题，本文提出一种多信息流动卷积神经网络（Multi-information Flow Convolutional Neural Network，MiF-CNN）模型，模型中包含一个特殊的卷积结构，该结构中每层卷积层提取到的特征与后续所有卷积层的输入相连接，增强了网络的特征信息流动性和梯度的反向传播效率，使得模型提取到的行人特征更具判别力.采用多损失函数组合方式训练网络模型，更好的区分行人类别.最后利用欧氏距离对行人特征相似性进行排序.在标准行人再识别数据集VIPeR和CUHK01上的实验表明，本文方法进一步提高了行人再识别精度，并有效改善了深度神经网络的过拟合现象.

Abstract

It is challenging to learn efficient features in person re-identification task due to complex variations of viewpoints

illumination

pose etc.In addition

deep neural network still suffers from overfitting with a small training set.To solve these problems

a Multi-information Flow Convolutional Neural Network(MiF-CNN)is designed for person re-identification which contains a specific convolutional architecture.In this architecture

features from each convolution layer are concatenated with all subsequent convolution layers so that it can improve the flowability of feature information and the efficiency of backpropagation gradient.In this way

the proposed network can learn more discriminative features.Moreover

a combination of three loss functions is used to train proposed network.Finally

the identification accuracy is obtained by ranking the similarity of extracted features measured by Euclidean distance.Experimental results on the VIPeR and CUHK01 datasets demonstrate that MiF-CNN outperform most of existing methods of person re-identification and reduce overfitting more effectively.

关键词

Keywords

references

Views

381

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Ship Target Detection in SAR Image Based on Selective Coordinate Attention

Abnormal Event Detection Based on Deep Learning

Survey of Object Detection Based on Deep Learning

A Survey of Research and Application of Small Object Detection Based on Deep Learning

Operator Fusion Method and Hardware Architecture Design Based on Non-Standard Operators

Related Author

YAN Chun-man

WANG Cheng

WEN Jia

WANG Hong-jun

DENG Jia

LIU Peng-fei

LUO Hui-lan

CHEN Hong-kun

Related Institution

College of Physics and Electronic Engineering， Northwest Normal University

Engineering Research Center of Gansu Province for Intelligent Information Technology and Application

School of Information Science and Engineering, Yanshan University

The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province

State Key Laboratory of Software Engineering of Hebei Province

⁰