Integration graph attention network and multi-centre constrained loss for cross-modality person re‐identification

Tools

He, Di and Zhang, Jingrui and Zhang, Zhong and Liu, Shuang and Durrani, Tariq S. (2022) Integration graph attention network and multi-centre constrained loss for cross-modality person re‐identification. IET Computer Vision, 17 (1). pp. 76-87. ISSN 1751-9640 (https://doi.org/10.1049/cvi2.12132)

[thumbnail of He-etal-IET-CV-2022-Integration-graph-attention-network-and-multi-centre-constrained-loss-for-cross-modality]
Preview
 Text. Filename: He_etal_IET_CV_2022_Integration_graph_attention_network_and_multi_centre_constrained_loss_for_cross_modality.pdf
Final Published Version
License: Creative Commons Attribution-NonCommercial 4.0 logo
Download (1MB)| Preview

Abstract

Abstract: Cross‐modality person re‐identification is a challenging task due to the large visual appearance difference between RGB and infrared images. Existing studies mainly focus on learning local features and ignore the correlation between local features. In this paper, the Integration Graph Attention Network is proposed to learn the completed correlation between local features via the graph structure. To this end, the authors learn the coarse‐fine attention weights to aggregate the local features by considering local detail and global information. Furthermore, the Multi‐Centre Constrained Loss is proposed to optimise the feature similarity by constraining the centres of modality and identity. It simultaneously utilises three kinds of centre constraints, that is intra‐identity centre constraint, modality centre constraint, and inter‐identity centre constraint, in order to reduce the influence of modality information explicitly. The proposed method is evaluated on two standard benchmark datasets, that is SYSU‐MM01 and RegDB, and the results demonstrate that the authors’ method achieves better performance than the state‐of‐the‐art methods, for example, surpassing NFS by 4.8% and 6.0% mAP on the single‐shot setting in All‐search and Indoor‐search modes, respectively.

CORE (COnnecting REpositories)