EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Person Re-Identification Using Local Relation-Aware Graph Convolutional Network

Yu Lian, Wenmin Huang, Shuang Liu iD, Peng Guo, Zhong Zhang iD, Tariq S Durrani

Sensors · 28.09.2023 · 23 (19) · 8138

DOI: 10.3390/s23198138 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 28.09.2023. Verfasst von Yu Lian und 5 weiteren Autor:innen. Beteiligte Einrichtung: Tianjin Key Laboratory of Wireless Mobile Communications and Power Transmission, Tianjin Normal University, Tianjin 300387, China. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 73 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, atural Science Foundation of Tianjin, Scientific Research Project of Tianjin Educational Committee.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Local feature extractions have been verified to be effective for person re-identification (re-ID) in recent literature. However, existing methods usually rely on extracting local features from single part of a pedestrian while neglecting the relationship of local features among different pedestrian images. As a result, local features contain limited information from one pedestrian image, and cannot benefit from other pedestrian images. In this paper, we propose a novel approach named Local Relation-Aware Graph Convolutional Network (LRGCN) to learn the relationship of local features among different pedestrian images. In order to completely describe the relationship of local features among different pedestrian images, we propose overlap graph and similarity graph. The overlap graph formulates the edge weight as the overlap node number in the node’s neighborhoods so as to learn robust local features, and the similarity graph defines the edge weight as the similarity between the nodes to learn discriminative local features. To propagate the information for different kinds of nodes effectively, we propose the Structural Graph Convolution (SGConv) operation. Different from traditional graph convolution operations where all nodes share the same parameter matrix, SGConv learns different parameter matrices for the node itself and its neighbor nodes to improve the expressive power. We conduct comprehensive experiments to verify our method on four large-scale person re-ID databases, and the overall results show LRGCN exceeds the state-of-the-art methods.

Schlagwörter

Stichwörter der Autor:innen

graph convolutional networklocal feature relationshipperson re-identification

Autor:innen und Institutionen

  • Tianjin Key Laboratory of Wireless Mobile Communications and Power Transmission, Tianjin Normal University, Tianjin 300387, China.
  • CATARC (Tianjin) Automotive Engineering Research Institute Co., Ltd., Tianjin 300300, China.
  • Department of Electronic and Electrical Engineering, University of Strathclyde, Glasgow G1 1QE, UK.

ORCID-Profile: Shuang Liu, Zhong Zhang

Förderung

  • National Natural Science Foundation of China
  • atural Science Foundation of Tianjin
  • Scientific Research Project of Tianjin Educational Committee
  • Tianjin Normal University Research Innovation Project for Postgraduate Students

Zitieren

APA

Lian, Y., Huang, W., Liu, S., Guo, P., Zhang, Z., & Durrani, T. S. (2023). Person Re-Identification Using Local Relation-Aware Graph Convolutional Network. Sensors, 23(19), 8138. https://doi.org/10.3390/s23198138

Vancouver

Lian Y, Huang W, Liu S, Guo P, Zhang Z, Durrani TS. Person Re-Identification Using Local Relation-Aware Graph Convolutional Network. Sensors. 2023;23(19):8138. doi:10.3390/s23198138

BibTeX
@article{Lian2023,
  author = {Lian, Yu and Huang, Wenmin and Liu, Shuang and Guo, Peng and Zhang, Zhong and Durrani, Tariq S},
  title = {Person Re-Identification Using Local Relation-Aware Graph Convolutional Network},
  journal = {Sensors},
  year = {2023},
  volume = {23},
  number = {19},
  pages = {8138},
  doi = {10.3390/s23198138}
}
Kostenlos starten →