دورية أكاديمية
Threshold-Based Hierarchical Clustering for Person Re-Identification
العنوان: | Threshold-Based Hierarchical Clustering for Person Re-Identification |
---|---|
المؤلفون: | Minhui Hu, Kaiwei Zeng, Yaohua Wang, Yang Guo |
المصدر: | Entropy, Vol 23, Iss 5, p 522 (2021) |
بيانات النشر: | MDPI AG, 2021. |
سنة النشر: | 2021 |
المجموعة: | LCC:Science LCC:Astrophysics LCC:Physics |
مصطلحات موضوعية: | person re-identification, threshold-based hierarchical clustering, unsupervised domain adaptation, fully unsupervised method, Science, Astrophysics, QB460-466, Physics, QC1-999 |
الوصف: | Unsupervised domain adaptation is a challenging task in person re-identification (re-ID). Recently, cluster-based methods achieve good performance; clustering and training are two important phases in these methods. For clustering, one major issue of existing methods is that they do not fully exploit the information in outliers by either discarding outliers in clusters or simply merging outliers. For training, existing methods only use source features for pretraining and target features for fine-tuning and do not make full use of all valuable information in source datasets and target datasets. To solve these problems, we propose a Threshold-based Hierarchical clustering method with Contrastive loss (THC). There are two features of THC: (1) it regards outliers as single-sample clusters to participate in training. It well preserves the information in outliers without setting cluster number and combines advantages of existing clustering methods; (2) it uses contrastive loss to make full use of all valuable information, including source-class centroids, target-cluster centroids and single-sample clusters, thus achieving better performance. We conduct extensive experiments on Market-1501, DukeMTMC-reID and MSMT17. Results show our method achieves state of the art. |
نوع الوثيقة: | article |
وصف الملف: | electronic resource |
اللغة: | English |
تدمد: | 1099-4300 |
Relation: | https://www.mdpi.com/1099-4300/23/5/522; https://doaj.org/toc/1099-4300 |
DOI: | 10.3390/e23050522 |
URL الوصول: | https://doaj.org/article/661ace9171234cf5913df3e56da036dc |
رقم الأكسشن: | edsdoj.661ace9171234cf5913df3e56da036dc |
قاعدة البيانات: | Directory of Open Access Journals |
تدمد: | 10994300 |
---|---|
DOI: | 10.3390/e23050522 |