Dual Knowledge Distillation on Multiview Pseudo Labels for Unsupervised Person Re-Identification

Zhu, Wenjie
Peng, Bo
Yan, Wei Qi
Item type
Journal Article
Degree name
Journal Title
Journal ISSN
Volume Title
Institute of Electrical and Electronics Engineers (IEEE)

Unsupervised person re-identification (Re-ID) has made significant progress by leveraging valuable pseudo labels from completely unlabeled data. However, the predominant use of pseudo labels heavily relies on clustering results, which may lead to the accumulation of supervision deviation due to inevitable noise. In this paper, we propose a novel framework, namely Dual Knowledge Distillation on Multiview Pseudo Labels (DKD-MPL), to address this challenge. Specifically, the proposed DKD-MPL framework consists of two modules: Global Knowledge Distillation (GKD) and Self-Knowledge Distillation (SKD). In the GKD module, the pseudo labels obtained from the epoch-wise clustering procedure serve as the logits for the teacher model, while the mini-batch query images' pseudo labels act as the logits for the student model. Within the SKD module, we facilitate self-knowledge distillation by considering the pseudo labels generated by positive anchors and query images as two augmentations of the mini-batch data. As a result, DKD-MPL facilitates the exploitation of both global and local complementary knowledge across different views of pseudo labels, thereby mitigating supervision deviation. To demonstrate the effectiveness of DKD-MPL, we provide a theoretical analysis of the proposed loss and conduct extensive experiments on four popular datasets, e.g., Market-1501, DukeMTMC-reID, MSMT17, and VeRi-776. The results indicate that our method surpasses unsupervised approaches and achieves comparable performance to supervised person Re-ID methods.

08 Information and Computing Sciences , 09 Engineering , Artificial Intelligence & Image Processing , 40 Engineering , 46 Information and computing sciences
IEEE Transactions on Multimedia, ISSN: 1520-9210 (Print); 1941-0077 (Online), Institute of Electrical and Electronics Engineers (IEEE), 1-13. doi: 10.1109/tmm.2024.3366395
Rights statement
IEEE policy provides that authors are free to follow funder public access mandates to post accepted articles in repositories. When posting in a repository, the IEEE embargo period is 24 months. However, IEEE recognizes that posting requirements and embargo periods vary by funder. IEEE authors may comply with requirements to deposit their accepted articles in a repository per funder requirements where the embargo is less than 24 months. Information on specific funder requirements can be found here [https://open.ieee.org/for-authors/funders/]