Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox


Cluster Contrast for Unsupervised Person Re-Identification

Mar 22, 2021
Zuozhuo Dai, Guangyuan Wang, Siyu Zhu, Weihao Yuan, Ping Tan


Share this with someone who'll enjoy it:


Unsupervised person re-identification (re-ID) attractsincreasing attention due to its practical applications in in-dustry. State-of-the-art unsupervised re-ID methods trainthe neural networks using a memory-based non-parametricsoftmax loss. They store the pre-computed instance featurevectors inside the memory, assign pseudo labels to them us-ing clustering algorithm, and compare the query instancesto the cluster using a form of contrastive loss. Duringtraining, the instance feature vectors are updated. How-ever, due to the varying cluster size, the updating progressfor each cluster is inconsistent. To solve this problem, wepresent Cluster Contrast which stores feature vectors andcomputes contrast loss in the cluster level. We demonstratethat the inconsistency problem for cluster feature represen-tation can be solved by the cluster-level memory dictionary.By straightforwardly applying Cluster Contrast to a stan-dard unsupervised re-ID pipeline, it achieves considerableimprovements of 9.5%, 7.5%, 6.6% compared to state-of-the-art purely unsupervised re-ID methods and 5.1%, 4.0%,6.5% mAP compared to the state-of-the-art unsuperviseddomain adaptation re-ID methods on the Market, Duke, andMSMT17 datasets.Our source code is available at https://github.com/wangguangyuan/ClusterContrast.git.



   Access Paper Source



Share this with someone who'll enjoy it: