Person re-identification with intra-domain similarity grouping based on semantic fusion

Unsupervised cross-domain person re-identification aims to adapt a model trained on a labeled source-domain dataset to a target-domain dataset.However, the cluster-based unsupervised cross-domain pedestrian re-identification algorithm often generates noise due to the different input pedestrian pictu...

Popoln opis

Bibliografske podrobnosti

Principais autores:	Qiqi KOU, Ji HUANG, Deqiang CHENG, Yunlong LI, Jianying ZHANG
Format:	Artigo
Jezik:	Chinês
Izdano:	Editorial Department of Journal on Communications 2022-07-01
Serija:	Tongxin xuebao
Teme:	unsupervised cross domain;person re-identification;semantic fusion;adaptive perception;fine-grained information
Online dostop:	http://www.joconline.com.cn/thesisDetails#10.11959/j.issn.1000-436x.2022136
Oznake:	Označite Brez oznak, prvi označite!