[1]李俊峰,楼 琼,钱亚冠,等.基于像素对齐和特征对齐的跨模态行人重识别[J].浙江科技学院学报,2022,(03):251-260.[doi:10.3969/j.issn.1671-8798.2022.03.007 ]
 LI Junfeng,LOU Qiong,QIAN Yaguan,et al.Cross-modality person re-identification based on pixel alignment and feature alignment[J].,2022,(03):251-260.[doi:10.3969/j.issn.1671-8798.2022.03.007 ]
点击复制

基于像素对齐和特征对齐的跨模态行人重识别(/HTML)
分享到:

《浙江科技学院学报》[ISSN:1001-3733/CN:61-1062/R]

卷:
期数:
2022年03期
页码:
251-260
栏目:
出版日期:
2022-06-16

文章信息/Info

Title:
Cross-modality person re-identification based on pixel alignment and feature alignment
文章编号:
1671-8798(2022)03-0251-10
作者:
李俊峰楼 琼钱亚冠孙安临
(浙江科技学院 理学院,杭州 310023)
Author(s):
LI Junfeng LOU Qiong QIAN Yaguan SUN Anlin
(School of Sciences, Zhejiang University of Science and Technology, Hangzhou 310023, Zhejiang, China)
关键词:
行人重识别 跨模态 像素对齐 特征对齐 非局部神经网络
分类号:
TP391.41
DOI:
10.3969/j.issn.1671-8798.2022.03.007
文献标志码:
A
摘要:
为了减少可见光-红外跨模态行人重识别中较大的跨模态差异,提出一种联合像素对齐和特征对齐的跨模态行人重识别方法。首先,从像素级角度出发,利用对齐生成对抗网络(alignment generative adversarial network,AlignGAN),将可见光图像转换为红外图像,减少可见光和红外图像之间的跨模态差距。其次,从特征级角度出发,通过交换可见光和红外图像的模态特定特征来生成跨模态配对图像,同时进行全局集合级对齐和细粒度实例级对齐。最后,运用基于非局部块的深度为50层的残差网络(50-layer residual nets,ResNet-50)捕获图像的长距离依赖关系。在SYSU-MM01数据集上进行了大量试验,我们的方法得到41.8%的识别准确率,在相比较的方法中准确率最高。可见,本方法可以有效地减少跨模态行人重识别中较大的跨模态差异。本研究结果可为跨模态行人重识别的研究提供参考。

参考文献/References:

[1] GONG S, CRISTANI M, YAN S, et al. Personre-identification[J].Advances in Computer Vision & Pattern Recognition,2014,42(7):301.
[2] LI W, ZHU X, GONG S. Harmonious attention network for person re-identification[C]//IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2018:2285.
[3] WANG Z, HU R, CHEN C, et al. Person reidentification via discrepancy matrix and matrix metric[J].IEEE Transactions on Cybernetics,2018,48(10):3006.
[4] WEI L, ZHANG S, GAO W, et al. Person transfer GAN to bridge domain gap for person reidentification[C]//IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2018:79.
[5] 陈丹,李永忠,于沛泽,等.跨模态行人重识别研究与展望[J].计算机系统应用,2020,29(10):20.
[6] WU A, ZHENG W S, YU H X, et al. Rgb-infrared cross-modality person re-identification[C]//IEEE International Conference on Computer Vision. Piscataway:IEEE,2017:5390.
[7] YE M, LAN X, LI J, et al. Hierarchical discriminative learning for visible thermal person re-identification[C]//AAAI Conference on Artificial Intelligence. Palo Alto:AAAI,2018:7501.
[8] YE M, WANG Z, LAN X, et al. Visible thermal person re-identification via dual-constrained Top-ranking[C]//International Joint Conference on Artificial Intelligence. San Francisco:Morgan Kaufmann,2018:1092.
[9] 冯敏,张智成,吕进,等.基于生成对抗网络的跨模态行人重识别研究[J].现代信息科技,2020,4(4):107.
[10] WANG Z, WANG Z, ZHENG Y, et al. Learning to reduce dual-level discrepancy for infrared-visible person reidentification[C]//IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2019:618.
[11] WANG G, ZHANG T, CHENG J, et al. Rgb-infrared cross-modality person re-identification via joint pixel and feature alignment[C]//IEEE International Conference on Computer Vision. Piscataway:IEEE,2019:3623.
[12] WANG G, YANG Y, ZHANG T, et al. Cross-modality paired-images generation and augmentation for RGB-infrared person re-identification[J].Neural Networks,2020,128:294.
[13] WANG X, GIRSHICK R, GUPTA A, et al. Non-local neural networks[C]//IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2018:1.
[14] GOODFELLOW I J, POUGET-ABADIE J, MIRZA M, et al. Generativeadversarial nets[C]//Advances in Neural Information Processing Systems(NIPS). Massachusetts:MIT Press,2014:2672.
[15] ZHU J Y, PARK T, ISOLA P, et al. Unpaired image-to-image translation using cycle-consistent adversarial networks[C]//2017 IEEE International Conference on Computer Vision. Piscataway:IEEE,2017:2242.
[16] ZHENG L, YANG Y, HAUPTMANN A G. Person re-identification:past, present and future[EB/OL].(2016-10-10)[2021-05-10].https://arxiv.org/abs/1610.02984.
[17] HERMANS A, BEYER L, LEIBE B. In defense of the triplet loss for person re-identification[EB/OL].(2017-03-22)[2021-05-10].https://arxiv.org/abs/1703.07737.
[18] KRHENBÜHL P, KOLTUN V. Efficient inference in fully connected crfs with gaussian edge potentials[J].Neural Information Processing Systems(NIPS). Massachusetts:MIT Press,2011:2.
[19] BUADES A, COLL B, MOREL J M. A non-local algorithm for image denoising[C]//IEEE International Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2005:1.
[20] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//IEEE International Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2016:770.
[21] DALAL N, RIGGS B. Histograms of oriented gradients for human detection[C]//IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2005:886.
[22] LIAO S, HU Y, ZHU X, et al. Person reidentification by local maximal occurrence representation and metric learning[C]//IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2015:2197.
[23] DAI P, JI R, WANG H, et al. Cross-modality person re-identification with generative adversarial training[C]//International Joint Conference on Artificial Intelligence. San Francisco:Morgan Kaufmann,2018:677.

备注/Memo

备注/Memo:
收稿日期:2021-05-24
基金项目:国家自然科学基金项目(11801511)
通信作者:钱亚冠(1976— ),男,浙江省绍兴人,教授,博士,主要从事机器学习与人工智能安全研究。E-mail:qianyaguan@zust.edu.cn。
更新日期/Last Update: 2022-06-30