Unsupervised person re-identification via multi-domain joint learning. (June 2023)

• Record Type: Journal Article
• Authors: Chen, Feng; Wang, Nian; Tang, Jun; Yan, Pu; Yu, Jun
• Is Part Of: Pattern recognition. Volume 138(2023)
• Publication Date: 2023-06
• Subjects: Person re-identification -- Data augmentation -- Domain adaptation -- Unsupervised learning; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2023.109369
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: Render the transferred model more generalizable by enhancing appearance diversity. Achieve feature decoupling with the aid of data augmentation and a multi-label assignment strategy. Improve the reliability of pseudo labels by exploiting the correlation of multiple clustering results. Abstract: Deep learning techniques have achieved impressive progress in the task of person re-identification. However, how to generalize a learned model from the source domain to the target domain remains a long-standing challenge. Inspired by the fact that the enrichment of data diversity and the utilization of miscellaneous semantic features can lead to better generalization ability, we design a model that integrates a novel data augmentation method with a multi-label assignment strategy to achieve semantic features decoupling in the source domain. The pre-trained model is employed to extract several kinds of semantic features from the target dataset, and each kind of semantic features is regarded as a specific domain. We then cluster features of each domain and exploit the connection between different clustering results to perform self-distillation for generating more reliable pseudo labels. Finally, the obtained pseudo labels are used to fine-tune the pre-trained model to achieve model transfer from the source domain to the target one. Extensive experiments demonstrate that our approach outperforms some state-of-the-art methods by a clear margin and even surpass some supervised methods. Source code is available at: https://www.github.com/flychen321/MDJL .