Cross‐modality person re‐identification using hybrid mutual learning
Abstract Cross‐modality person re‐identification (Re‐ID) aims to retrieve a query identity from red, green, blue (RGB) images or infrared (IR) images. Many approaches have been proposed to reduce the distribution gap between RGB modality and IR modality. However, they ignore the valuable collaborati...
Main Authors: | , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2023-02-01
|
Series: | IET Computer Vision |
Online Access: | https://doi.org/10.1049/cvi2.12123 |
_version_ | 1797905155409051648 |
---|---|
author | Zhong Zhang Qing Dong Sen Wang Shuang Liu Baihua Xiao Tariq S. Durrani |
author_facet | Zhong Zhang Qing Dong Sen Wang Shuang Liu Baihua Xiao Tariq S. Durrani |
author_sort | Zhong Zhang |
collection | DOAJ |
description | Abstract Cross‐modality person re‐identification (Re‐ID) aims to retrieve a query identity from red, green, blue (RGB) images or infrared (IR) images. Many approaches have been proposed to reduce the distribution gap between RGB modality and IR modality. However, they ignore the valuable collaborative relationship between RGB modality and IR modality. Hybrid Mutual Learning (HML) for cross‐modality person Re‐ID is proposed, which builds the collaborative relationship by using mutual learning from the aspects of local features and triplet relation. Specifically, HML contains local‐mean mutual learning and triplet mutual learning where they focus on transferring local representational knowledge and structural geometry knowledge so as to reduce the gap between RGB modality and IR modality. Furthermore, Hierarchical Attention Aggregation is proposed to fuse local feature maps and local feature vectors to enrich the information of the classifier input. Extensive experiments on two commonly used data sets, that is, SYSU‐MM01 and RegDB verify the effectiveness of the proposed method. |
first_indexed | 2024-04-10T10:00:34Z |
format | Article |
id | doaj.art-e7b34ecee66c4d78b20f003f9ac74677 |
institution | Directory Open Access Journal |
issn | 1751-9632 1751-9640 |
language | English |
last_indexed | 2024-04-10T10:00:34Z |
publishDate | 2023-02-01 |
publisher | Wiley |
record_format | Article |
series | IET Computer Vision |
spelling | doaj.art-e7b34ecee66c4d78b20f003f9ac746772023-02-16T09:51:43ZengWileyIET Computer Vision1751-96321751-96402023-02-0117111210.1049/cvi2.12123Cross‐modality person re‐identification using hybrid mutual learningZhong Zhang0Qing Dong1Sen Wang2Shuang Liu3Baihua Xiao4Tariq S. Durrani5Tianjin Key Laboratory of Wireless Mobile Communications and Power Transmission Tianjin Normal University Tianjin ChinaTianjin Key Laboratory of Wireless Mobile Communications and Power Transmission Tianjin Normal University Tianjin ChinaTianjin Key Laboratory of Wireless Mobile Communications and Power Transmission Tianjin Normal University Tianjin ChinaTianjin Key Laboratory of Wireless Mobile Communications and Power Transmission Tianjin Normal University Tianjin ChinaThe State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing ChinaElectronic and Electrical Engineering University of Strathclyde Glasgow UKAbstract Cross‐modality person re‐identification (Re‐ID) aims to retrieve a query identity from red, green, blue (RGB) images or infrared (IR) images. Many approaches have been proposed to reduce the distribution gap between RGB modality and IR modality. However, they ignore the valuable collaborative relationship between RGB modality and IR modality. Hybrid Mutual Learning (HML) for cross‐modality person Re‐ID is proposed, which builds the collaborative relationship by using mutual learning from the aspects of local features and triplet relation. Specifically, HML contains local‐mean mutual learning and triplet mutual learning where they focus on transferring local representational knowledge and structural geometry knowledge so as to reduce the gap between RGB modality and IR modality. Furthermore, Hierarchical Attention Aggregation is proposed to fuse local feature maps and local feature vectors to enrich the information of the classifier input. Extensive experiments on two commonly used data sets, that is, SYSU‐MM01 and RegDB verify the effectiveness of the proposed method.https://doi.org/10.1049/cvi2.12123 |
spellingShingle | Zhong Zhang Qing Dong Sen Wang Shuang Liu Baihua Xiao Tariq S. Durrani Cross‐modality person re‐identification using hybrid mutual learning IET Computer Vision |
title | Cross‐modality person re‐identification using hybrid mutual learning |
title_full | Cross‐modality person re‐identification using hybrid mutual learning |
title_fullStr | Cross‐modality person re‐identification using hybrid mutual learning |
title_full_unstemmed | Cross‐modality person re‐identification using hybrid mutual learning |
title_short | Cross‐modality person re‐identification using hybrid mutual learning |
title_sort | cross modality person re identification using hybrid mutual learning |
url | https://doi.org/10.1049/cvi2.12123 |
work_keys_str_mv | AT zhongzhang crossmodalitypersonreidentificationusinghybridmutuallearning AT qingdong crossmodalitypersonreidentificationusinghybridmutuallearning AT senwang crossmodalitypersonreidentificationusinghybridmutuallearning AT shuangliu crossmodalitypersonreidentificationusinghybridmutuallearning AT baihuaxiao crossmodalitypersonreidentificationusinghybridmutuallearning AT tariqsdurrani crossmodalitypersonreidentificationusinghybridmutuallearning |