Person re‐identification via deep compound eye network and pose repair module

Abstract Person re‐identification is aimed at searching for specific target pedestrians from non‐intersecting cameras. However, in real complex scenes, pedestrians are easily obscured, which makes the target pedestrian search task time‐consuming and challenging. To address the problem of pedestrians...

Full description

Saved in:
Bibliographic Details
Main Authors: Hongjian Gu, Wenxuan Zou, Keyang Cheng, Bin Wu, Humaira Abdul Ghafoor, Yongzhao Zhan
Format: Article
Language:English
Published: Wiley 2024-09-01
Series:IET Computer Vision
Subjects:
Online Access:https://doi.org/10.1049/cvi2.12282
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846164012285296640
author Hongjian Gu
Wenxuan Zou
Keyang Cheng
Bin Wu
Humaira Abdul Ghafoor
Yongzhao Zhan
author_facet Hongjian Gu
Wenxuan Zou
Keyang Cheng
Bin Wu
Humaira Abdul Ghafoor
Yongzhao Zhan
author_sort Hongjian Gu
collection DOAJ
description Abstract Person re‐identification is aimed at searching for specific target pedestrians from non‐intersecting cameras. However, in real complex scenes, pedestrians are easily obscured, which makes the target pedestrian search task time‐consuming and challenging. To address the problem of pedestrians' susceptibility to occlusion, a person re‐identification via deep compound eye network (CEN) and pose repair module is proposed, which includes (1) A deep CEN based on multi‐camera logical topology is proposed, which adopts graph convolution and a Gated Recurrent Unit to capture the temporal and spatial information of pedestrian walking and finally carries out pedestrian global matching through the Siamese network; (2) An integrated spatial‐temporal information aggregation network is designed to facilitate pose repair. The target pedestrian features under the multi‐level logic topology camera are utilised as auxiliary information to repair the occluded target pedestrian image, so as to reduce the impact of pedestrian mismatch due to pose changes; (3) A joint optimisation mechanism of CEN and pose repair network is introduced, where multi‐camera logical topology inference provides auxiliary information and retrieval order for the pose repair network. The authors conducted experiments on multiple datasets, including Occluded‐DukeMTMC, CUHK‐SYSU, PRW, SLP, and UJS‐reID. The results indicate that the authors’ method achieved significant performance across these datasets. Specifically, on the CUHK‐SYSU dataset, the authors’ model achieved a top‐1 accuracy of 89.1% and a mean Average Precision accuracy of 83.1% in the recognition of occluded individuals.
format Article
id doaj-art-c87b5f7de71c4cbe820489df3eb77f9e
institution Kabale University
issn 1751-9632
1751-9640
language English
publishDate 2024-09-01
publisher Wiley
record_format Article
series IET Computer Vision
spelling doaj-art-c87b5f7de71c4cbe820489df3eb77f9e2024-11-18T16:52:09ZengWileyIET Computer Vision1751-96321751-96402024-09-0118682684110.1049/cvi2.12282Person re‐identification via deep compound eye network and pose repair moduleHongjian Gu0Wenxuan Zou1Keyang Cheng2Bin Wu3Humaira Abdul Ghafoor4Yongzhao Zhan5School of Computer Science and Technology University of Science and Technology of China Hefei Anhui ChinaSchool of Computer Science and Communication Engineering Jiangsu University Zhenjiang Jiangsu ChinaSchool of Computer Science and Communication Engineering Jiangsu University Zhenjiang Jiangsu ChinaSchool of Computer Science and Communication Engineering Jiangsu University Zhenjiang Jiangsu ChinaSchool of Software Engineering University of Sialkot Sialkot Punjab PakistanSchool of Computer Science and Communication Engineering Jiangsu University Zhenjiang Jiangsu ChinaAbstract Person re‐identification is aimed at searching for specific target pedestrians from non‐intersecting cameras. However, in real complex scenes, pedestrians are easily obscured, which makes the target pedestrian search task time‐consuming and challenging. To address the problem of pedestrians' susceptibility to occlusion, a person re‐identification via deep compound eye network (CEN) and pose repair module is proposed, which includes (1) A deep CEN based on multi‐camera logical topology is proposed, which adopts graph convolution and a Gated Recurrent Unit to capture the temporal and spatial information of pedestrian walking and finally carries out pedestrian global matching through the Siamese network; (2) An integrated spatial‐temporal information aggregation network is designed to facilitate pose repair. The target pedestrian features under the multi‐level logic topology camera are utilised as auxiliary information to repair the occluded target pedestrian image, so as to reduce the impact of pedestrian mismatch due to pose changes; (3) A joint optimisation mechanism of CEN and pose repair network is introduced, where multi‐camera logical topology inference provides auxiliary information and retrieval order for the pose repair network. The authors conducted experiments on multiple datasets, including Occluded‐DukeMTMC, CUHK‐SYSU, PRW, SLP, and UJS‐reID. The results indicate that the authors’ method achieved significant performance across these datasets. Specifically, on the CUHK‐SYSU dataset, the authors’ model achieved a top‐1 accuracy of 89.1% and a mean Average Precision accuracy of 83.1% in the recognition of occluded individuals.https://doi.org/10.1049/cvi2.12282computer visionneural nets
spellingShingle Hongjian Gu
Wenxuan Zou
Keyang Cheng
Bin Wu
Humaira Abdul Ghafoor
Yongzhao Zhan
Person re‐identification via deep compound eye network and pose repair module
IET Computer Vision
computer vision
neural nets
title Person re‐identification via deep compound eye network and pose repair module
title_full Person re‐identification via deep compound eye network and pose repair module
title_fullStr Person re‐identification via deep compound eye network and pose repair module
title_full_unstemmed Person re‐identification via deep compound eye network and pose repair module
title_short Person re‐identification via deep compound eye network and pose repair module
title_sort person re identification via deep compound eye network and pose repair module
topic computer vision
neural nets
url https://doi.org/10.1049/cvi2.12282
work_keys_str_mv AT hongjiangu personreidentificationviadeepcompoundeyenetworkandposerepairmodule
AT wenxuanzou personreidentificationviadeepcompoundeyenetworkandposerepairmodule
AT keyangcheng personreidentificationviadeepcompoundeyenetworkandposerepairmodule
AT binwu personreidentificationviadeepcompoundeyenetworkandposerepairmodule
AT humairaabdulghafoor personreidentificationviadeepcompoundeyenetworkandposerepairmodule
AT yongzhaozhan personreidentificationviadeepcompoundeyenetworkandposerepairmodule