arrow
返回

Collaborative Contrastive Refining for Weakly Supervised Person Search

delete2023-01-01
delete2
PRE
AI
C
Chengyou Jia
罗敏楠 封面图
罗敏楠 (Minnan Luo) *
C
Caixia Yan
L
Linchao Zhu
Xiaojun Chang 封面图
Xiaojun Chang (Xiaojun Chang)
Q
Qinghua Zheng
DOI:10.1109/TIP.2023.3308393delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Weakly supervised person search involves training a model with only bounding box annotations, without human-annotated identities. Clustering algorithms are commonly used to assign pseudo-labels to facilitate this task. However, inaccurate pseudo-labels and imbalanced identity distributions can result in severe label and sample noise. In this work, we propose a novel Collaborative Contrastive Refining (CCR) weakly-supervised framework for person search that jointly refines pseudo-labels and the sample-learning process with different contrastive strategies. Specifically, we adopt a hybrid contrastive strategy that leverages both visual and context clues to refine pseudo-labels, and leverage the sample-mining and noise-contrastive strategy to reduce the negative impact of imbalanced distributions by distinguishing positive samples and noise samples. Our method brings two main advantages: 1) it facilitates better clustering results for refining pseudo-labels by exploring the hybrid similarity; 2) it is better at distinguishing query samples and noise samples for refining the sample-learning process. Extensive experiments demonstrate the superiority of our approach over the state-of-the-art weakly supervised methods by a large margin (more than 3% mAP on CUHK-SYSU). Moreover, by leveraging more diverse unlabeled data, our method achieves comparable or even better performance than the state-of-the-art supervised methods.
Keyword:
Person search
unsupervised person Re-ID
weakly supervised learning
clustering algorithm

期刊

IEEE Transactions on Image Processing 封面图
IEEE Transactions on Image Processing
IF:
13.7
论文数:
1.0W
被引数:
8.4W

机构

X
xi'an jiaotong university
学者数:
9.3W
论文数: 6.7W
被引数: 75
U
university of technology sydney
学者数:
1.6W
论文数: 2.0W
被引数: 25
Z
zhejiang university
学者数:
17.7W
论文数: 12.1W
被引数: 152
学者 查看更多机构
引用论文

引用论文

err分享
err收藏
Conductivity Enhancement in Thin Silicon-on-Insulator Layer Embedding Artificial Dislocation Network
err2011-02-01
err0
PREAI
errYasuhiko Ishikawa; Kazuaki Yamauchi; Chihiro Yamamoto; Michiharu Tabe
err分享
err收藏
Comparison of multimicrophone probe design and processing methods in measuring acoustic intensity
err2014-05-01
err0
PREAI
errCurtis P. Wiederhold; Kent L. Gee; Jonathan D. Blotter; Scott D. Sommerfeldt; Jarom H. Giraud
err分享
err收藏
err分享
err收藏
学者 查看更多内容