Modality Synergy Complement Learning with Cascaded Aggregation for Visible-Infrared Person Re-Identification

Yiyuan Zhang, Sanyuan Zhao*, Yuhao Kang, Jianbing Shen

*此作品的通讯作者

科研成果: 书/报告/会议事项章节会议稿件同行评审

38 引用 (Scopus)

摘要

Visible-Infrared Re-Identification (VI-ReID) is challenging in image retrievals. The modality discrepancy will easily make huge intra-class variations. Most existing methods either bridge different modalities through modality-invariance or generate the intermediate modality for better performance. Differently, this paper proposes a novel framework, named Modality Synergy Complement Learning Network (MSCLNet) with Cascaded Aggregation. Its basic idea is to synergize two modalities to construct diverse representations of identity-discriminative semantics and less noise. Then, we complement synergistic representations under the advantages of the two modalities. Furthermore, we propose the Cascaded Aggregation strategy for fine-grained optimization of the feature distribution, which progressively aggregates feature embeddings from the subclass, intra-class, and inter-class. Extensive experiments on SYSU-MM01 and RegDB datasets show that MSCLNet outperforms the state-of-the-art by a large margin. On the large-scale SYSU-MM01 dataset, our model can achieve 76.99% and 71.64% in terms of Rank-1 accuracy and mAP value. Our code will be available at https://github.com/bitreidgroup/VI-ReID-MSCLNet.

源语言英语
主期刊名Computer Vision – ECCV 2022 - 17th European Conference, Proceedings
编辑Shai Avidan, Gabriel Brostow, Moustapha Cissé, Giovanni Maria Farinella, Tal Hassner
出版商Springer Science and Business Media Deutschland GmbH
462-479
页数18
ISBN(印刷版)9783031197802
DOI
出版状态已出版 - 2022
活动17th European Conference on Computer Vision, ECCV 2022 - Tel Aviv, 以色列
期限: 23 10月 202227 10月 2022

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
13674 LNCS
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

会议

会议17th European Conference on Computer Vision, ECCV 2022
国家/地区以色列
Tel Aviv
时期23/10/2227/10/22

指纹

探究 'Modality Synergy Complement Learning with Cascaded Aggregation for Visible-Infrared Person Re-Identification' 的科研主题。它们共同构成独一无二的指纹。

引用此

Zhang, Y., Zhao, S., Kang, Y., & Shen, J. (2022). Modality Synergy Complement Learning with Cascaded Aggregation for Visible-Infrared Person Re-Identification. 在 S. Avidan, G. Brostow, M. Cissé, G. M. Farinella, & T. Hassner (编辑), Computer Vision – ECCV 2022 - 17th European Conference, Proceedings (页码 462-479). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); 卷 13674 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-19781-9_27