Mar 13, 2023 · Unsupervised Re-ID methods aim at learning robust and discriminative features from unlabeled data. However, existing methods often ignore the ...

Unsupervised Re-ID methods aim at learning robust and discriminative features from unlabeled data. However, existing methods often ignore the noise from ...

The official repository for Dynamic Clustering and Cluster Contrastive Learning for Unsupervised Person Re-ID with Feature Distribution Alignment achieves ...

This work designs a dynamic clustering parameters scheduler (DCPS) which adjust the hyper-parameter of clustering to fit the variation of intra- and ...

To our knowledge, our work is the first detailed study around the parameter settings of the clustering algorithm in unsupervised Re-ID research. (2) We propose ...

Each feature is assigned a pseudo ID generated by a clustering algorithm. Dur- ing training, the contrastive loss is minimized to train the network and learn a.

Mar 13, 2023 · In this paper, we propose a novel unsupervised Re-ID framework based on dynamic clustering and dynamic clus- ter contrastive learning. We design ...

Dynamic Clustering and Cluster Contrastive Learning for Unsupervised Person Re-Id With Feature Distribution Alignment. April 2024. DOI:10.1109/ICASSP48485 ...

Mar 13, 2023 · Unsupervised Re-ID methods aim at learning robust and discriminative features from unlabeled data. However, existing methods often ignore ...

In this paper, we propose a Dynamic Hybrid Contrastive Learning (DHCL) method for unsupervised person re-ID.