基于DCN的无监督域自适应行人重识别方法

杨海伦; 王金聪; 任洪娥; 陶锐

doi:10.37188/CJLCD.2021-0095

您当前的位置：

首页 >

文章列表页 >

基于DCN的无监督域自适应行人重识别方法

目标识别与成像 | 更新时间：2021-11-08

- 基于DCN的无监督域自适应行人重识别方法
- DCN-Based unsupervised domain adaptive person re-identification method
- 液晶与显示 2021年36卷第11期页码：1573-1582
- 作者机构：
  
  1.东北林业大学信息与计算机工程学院, 黑龙江哈尔滨 150040
  2.东北林业大学机电工程学院, 黑龙江哈尔滨 150040
  3.黑龙江林业智能装备工程研究中心, 黑龙江哈尔滨 150040
  4.呼伦贝尔学院, 内蒙古呼伦贝尔 021008
- 作者简介：
  
  [ "杨海伦(1998-), 男, 辽宁锦州人, 硕士研究生, 2020年于长春理工大学获得学士学位, 主要从事图像识别与智能控制方面的研究。E-mail:yanghailun1998@163.com" ]
  [ "任洪娥(1962-), 女, 吉林白山人, 博士, 教授, 2009年于东北林业大学获得博士学位, 主要从事图像识别与智能控制方面的研究。E-mail: nefu_rhe@163.com" ]
- 基金信息：
  
  黑龙江省自然科学基金(LH2020F040);中央高校基本科研业务费专项资金资助项目(2572017PZ10)
- DOI：10.37188/CJLCD.2021-0095
  中图分类号： TP391.41
- 收稿日期：2021-04-12，
  
  修回日期：2021-05-17，
  
  纸质出版日期：2021-11
- 稿件说明：
移动端阅览
杨海伦, 王金聪, 任洪娥, 等. 基于DCN的无监督域自适应行人重识别方法[J]. 液晶与显示, 2021,36(11):1573-1582.

Hai-lun YANG, Jin-cong WANG, Hong-e REN, et al. DCN-Based unsupervised domain adaptive person re-identification method[J]. Chinese journal of liquid crystals and displays, 2021, 36(11): 1573-1582.
杨海伦, 王金聪, 任洪娥, 等. 基于DCN的无监督域自适应行人重识别方法[J]. 液晶与显示, 2021,36(11):1573-1582. DOI： 10.37188/CJLCD.2021-0095.

Hai-lun YANG, Jin-cong WANG, Hong-e REN, et al. DCN-Based unsupervised domain adaptive person re-identification method[J]. Chinese journal of liquid crystals and displays, 2021, 36(11): 1573-1582. DOI： 10.37188/CJLCD.2021-0095.

摘要

为解决无监督行人重识别研究中存在的遮挡、域间及相机间风格差异较大等问题，本文提出基于可变形卷积的无监督域自适应模型。针对特征提取过程中的遮挡问题提出基于可变形卷积的CNN模型；在预训练阶段提出应用SPGAN直接减小域间差异，训练过程中提出使用CycleGAN生成不同相机风格图像缓解相机风格差异性问题；提出多损失协同训练的方法实现CycleGAN和复用CNN模型的迭代优化进一步提高识别准确率。实验结果表明，本文提出的方法在源域DukeMTMC-reID/Market-1501和目标域Market-1501/DukeMTMC-reID下进行实验，mAP和Rank-1分别达到了68.7%、64.1%和88.2%、78.1%。本文所提出的模型有效缓解了行人被遮挡、域间及相机间风格差异较大等问题，与现有方法比较，有更好的识别效果。

Abstract

In order to solve the problems of occlusion

large differences in styles between domains and cameras in the research of unsupervised person re-recognition

this paper proposes an unsupervised domain adaptive model based on deformable convolution. Aiming at the occlusion problem in the feature extraction process

a CNN model based on deformable convolution is proposed. In the pre-training stage

it is proposed to apply SPGAN to directly reduce the difference between domains. During the training process

it is proposed to use CycleGAN to generate images of different camera styles to alleviate the problem of camera style differences. A multi-loss collaborative training method is proposed to realize the iterative optimization of CycleGAN and re-used CNN models to further improve the recognition accuracy. The experimental results show that the method proposed in this paper is tested in the source domain DukeMTMC-reID/Market-1501 and the target domain Market-1501/DukeMTMC-reID

and mAP and Rank-1 reach 68.7%

64.1% and 88.2%

78.1%

respectively. The model proposed in this paper effectively alleviates the problems of pedestrians being occluded

and large differences in styles between domains and cameras. Compared with the existing methods

it has a better recognition effect.

关键词

Keywords

references

YE M, SHEN J B, LIN G J, et al . Deep learning for person re-identification: a survey and outlook[J]. arXiv : 2001.04193V1.

黄乐弘 , 曹立华 , 李宁 , 等 . 深度学习的空间红外弱小目标状态感知方法 . 中国光学 , 2020 . 13 ( 3 ): 527 - 536 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202003014.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202003014.htm .

L H HUANG , L H CAO , N LI , 等 . Deep learning-based spatial infrared dim and small target state perception method . Chinese Optics , 2020 . 13 ( 3 ): 527 - 536 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202003014.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202003014.htm .

李勇 , 杨德东 , 毛宁 , 等 . 基于卷积神经网络的响应自适应跟踪 . 液晶与显示 , 2018 . 33 ( 7 ): 596 - 605 . DOI: 10.3788/YJYXS20183307.0596 http://doi.org/10.3788/YJYXS20183307.0596 .

Y LI , D D YANG , N MAO , 等 . Response adaptive tracking based on convolution neural network . Chinese Journal of Liquid Crystals and Displays , 2018 . 33 ( 7 ): 596 - 605 . DOI: 10.3788/YJYXS20183307.0596 http://doi.org/10.3788/YJYXS20183307.0596 .

陈晓冬 , 艾大航 , 张佳琛 , 等 . Gabor滤波融合卷积神经网络的路面裂缝检测方法 . 中国光学 , 2020 . 13 ( 6 ): 1293 - 1301 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202006014.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202006014.htm .

X D CHEN , D H AI , J C ZHANG , 等 . Gabor filter fusion network for pavement crack detection . Chinese Optics , 2020 . 13 ( 6 ): 1293 - 1301 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202006014.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202006014.htm .

YADAV A, VISHWAKARMA D K. Person re-identification using deep learning networks: a systematic review[J]. arXiv preprint arXiv : 2012.13318, 2020.

DENG W J, ZHENG L, YE Q X, et al . Image-image domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification[C]// Proceedings of 2018 IEEE / CVF Conference on Computer Vision and Pattern Recognition . Salt Lake City: IEEE, 2018: 994-1003.

WEI L H, ZHANG S L, GAO W, et al . Person transfer GAN to bridge domain gap for person re-identification[C]// Proceedings of 2018 IEEE / CVF Conference on Computer Vision and Pattern Recognition . Salt Lake City: IEEE, 2018: 79-88.

Z ZHONG , L ZHENG , Z D ZHENG , 等 . CamStyle: a novel data augmentation method for person re-identification . IEEE Transactions on Image Processing , 2019 . 28 ( 3 ): 1176 - 1190 . DOI: 10.1109/TIP.2018.2874313 http://doi.org/10.1109/TIP.2018.2874313 .

FU Y, WEI Y C, WANG G S, et al . Self-similarity grouping: a simple unsupervised cross domain adaptation approach for person re-identification[C]// Proceedings of 2019 IEEE / CVF International Conference on Computer Vision . Seoul, Korea: IEEE, 2019: 6112-6121.

L C SONG , C WANG , L F ZHANG , 等 . Unsupervised domain adaptive re-identification: theory and practice . Pattern Recognition , 2020 . 102 107173 DOI: 10.1016/j.patcog.2019.107173 http://doi.org/10.1016/j.patcog.2019.107173 .

ZHAI Y P, LU S J, YE Q X, et al . Ad-cluster: augmented discriminative clustering for domain adaptive person re-identification[C]// Proceedings of 2020 IEEE / CVF Conference on Computer Vision and Pattern Recognition . Seattle: IEEE, 2020: 9021-9030.

F X YANG , K LI , Z ZHONG , 等 . Asymmetric co-teaching for unsupervised cross-domain person re-identification . Proceedings of the AAAI Conference on Artificial Intelligence , 2020 . 34 ( 7 ): 12597 - 12604 . DOI: 10.1609/aaai.v34i07.6950 http://doi.org/10.1609/aaai.v34i07.6950 .

WANG D K, ZHANG S L. Unsupervised person re-identification via multi-label classification[C]// Proceedings of 2020 IEEE / CVF Conference on Computer Vision and Pattern Recognition . Seattle: IEEE, 2020: 10981-10990.

DAI J F, QI H Z, XIONG Y W, et al . Deformable convolutional networks[C]// Proceedings of 2017 IEEE International Conference on Computer Vision . Venice: IEEE, 2017: 764-773.

ZHU X Z, HU H, LIN S, et al . Deformable ConvNets V2: more deformable, better results[C]// Proceedings of 2019 IEEE / CVF Conference on Computer Vision and Pattern Recognition . Long Beach: IEEE, 2019: 9308-9316.

杨琦 , 车进 , 张良 , 等 . GAN网络混合编码的行人再识别 . 液晶与显示 , 2021 . 36 ( 2 ): 334 - 342 . DOI: 10.37188/CJLCD.2020-0167 http://doi.org/10.37188/CJLCD.2020-0167 .

Q YANG , J CHE , L ZHANG , 等 . Person re-identification of GAN network hybrid coding . Chinese Journal of Liquid Crystals and Displays , 2021 . 36 ( 2 ): 334 - 342 . DOI: 10.37188/CJLCD.2020-0167 http://doi.org/10.37188/CJLCD.2020-0167 .

HE K M, ZHANG X Y, REN S Q, et al . Deep residual learning for image recognition[C]// Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition . Las Vegas: IEEE, 2016: 770-778.

HERMANS A, BEYER L, LEIBE B. In defense of the triplet loss for person re-identification[J]. arXiv preprint arXiv : 1703.07737, 2017.

LIN S, LI H L, LI C T, et al . Multi-task mid-level feature alignment network for unsupervised cross-dataset person re-identification[J]. arXiv preprint arXiv : 1807.01440, 2018.

CHEN Y B, ZHU X T, GONG S G. Instance-guided context rendering for cross-domain person re-identification[C]// Proceedings of 2019 IEEE / CVF International Conference on Computer Vision . Seoul, Korea: IEEE, 2019: 232-242.

ZHANG X Y, CAO J W, SHEN C H, et al . Self-training with progressive augmentation for unsupervised cross-domain person re-identification[C]// Proceedings of 2019 IEEE / CVF International Conference on Computer Vision . Seoul, Korea: IEEE, 2019: 8222-8231.

Z ZHONG , L ZHENG , Z M LUO , 等 . Learning to adapt invariance in memory for person re-identification . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2021 . 43 ( 8 ): 2723 - 2738 . http://ieeexplore.ieee.org/abstract/document/9018132 http://ieeexplore.ieee.org/abstract/document/9018132 .

JIN X, LAN C L, ZENG W J, et al . Style normalization and restitution for generalizable person re-identification[C]// Proceedings of 2020 IEEE / CVF Conference on Computer Vision and Pattern Recognition . Seattle: IEEE, 2020: 3143-3152.

HUANG Y R, PENG P X, JIN Y, et al . Domain adaptive attention model for unsupervised cross-domain person re-identification[J]. arXiv preprint arXiv : 1905.10529, 2019.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于自监督注意和图像特征融合的文本生成图像方法

基于集成生成对抗网络的视频异常事件检测方法

基于双专用注意力机制引导的循环生成对抗网络

基于残差卷积注意力网络的视频修复