模式识别与人工智能
2025年4月3日 星期四   首 页     期刊简介     编委会     投稿指南     伦理声明     联系我们                                                                English
模式识别与人工智能  2022, Vol. 35 Issue (12): 1064-1077    DOI: 10.16451/j.cnki.issn1003-6059.202212002
“基于深度学习的图像理解及应用”专题 最新目录| 下期目录| 过刊浏览| 高级检索 |
模态不变性特征学习和一致性细粒度信息挖掘的跨模态行人重识别
石林波1, 李华锋1, 张亚飞1, 谢明鸿1
1.昆明理工大学 信息工程与自动化学院 昆明 650504
Modal Invariance Feature Learning and Consistent Fine-Grained Information Mining Based Cross-Modal Person Re-identification
SHI Linbo1, LI Huafeng1, ZHANG Yafei1, XIE Minghong1
1. Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650504

全文: PDF (5413 KB)   HTML (1 KB) 
输出: BibTeX | EndNote (RIS)      
摘要 跨模态行人重识别方法主要通过对齐不同模态的像素分布或特征分布以缓解模态差异,却忽略具有判别性的行人细粒度信息.为了获取不受模态差异影响且更具判别性的行人特征,文中提出模态不变性特征学习和一致性细粒度信息挖掘的跨模态行人重识别方法.方法主要包括模态不变性特征学习模块和语义一致的细粒度信息挖掘模块,联合两个模块,使特征提取网络获取具有判别性的特征.具体地,首先利用模态不变性特征学习模块去除特征图中的模态信息,缓解模态差异.然后,使用语义一致的细粒度信息挖掘模块,对特征图分别进行通道分组和水平分块,在充分挖掘具有判别性的细粒度信息的同时实现语义对齐.实验表明,文中方法性能较优.
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
石林波
李华锋
张亚飞
谢明鸿
关键词 跨模态行人重识别模态差异细粒度信息语义一致性    
Abstract:In the existing cross-modal person re-identification methods, modal differences are lessened by aligning features or pixel distributions of different modalities. However, the discriminative fine-grained information of pedestrians is ignored in these methods. To obtain more discriminative pedestrian features independent of modal differences, a modal invariance feature learning and consistent fine-grained information mining based cross-modal person re-identification method is proposed. The proposed method is mainly composed of two modules, modal invariance feature learning and semantically consistent fine-grained information mining. The two modules are combined to drive the feature extraction network to obtain discriminative features. Specifically, the modal invariant feature learning module is utilized to remove the modal information from the feature map to reduce the modal differences. Channel grouping and horizontal segmentation are conducted on person feature maps via the semantic consistent fine-grained information mining module. Consequently, the semantic alignment is achieved and the discriminative fine-grained information is fully mined. Experimental results show that the performance of the proposed method is significantly improved compared with the state-of-the-art cross-modal person re-identification methods.
Key wordsCross-Modal Person Re-identification    Modal Difference    Fine-Grained Information    Semantic Consistency   
收稿日期: 2022-08-08     
ZTFLH: TP391.41  
基金资助:国家自然科学基金项目(No.62161015,62276120,61966021)资助
通讯作者: 张亚飞,博士,副教授,主要研究方向为图像处理、模式识别.E-mail:zyfeimail@163.com.   
作者简介: 石林波,硕士研究生,主要研究方向为计算机视觉、行人重识别.E-mail:1527467911@qq.com.李华锋,博士,教授,主要研究方向为图像处理、计算机视觉.E-mail:hfchina99@163.com.谢明鸿,博士,高级工程师,主要研究方向为计算机视觉.E-mail:minghongxie@163.com.
引用本文:   
石林波, 李华锋, 张亚飞, 谢明鸿. 模态不变性特征学习和一致性细粒度信息挖掘的跨模态行人重识别[J]. 模式识别与人工智能, 2022, 35(12): 1064-1077. SHI Linbo, LI Huafeng, ZHANG Yafei, XIE Minghong. Modal Invariance Feature Learning and Consistent Fine-Grained Information Mining Based Cross-Modal Person Re-identification. Pattern Recognition and Artificial Intelligence, 2022, 35(12): 1064-1077.
链接本文:  
http://manu46.magtech.com.cn/Jweb_prai/CN/10.16451/j.cnki.issn1003-6059.202212002      或     http://manu46.magtech.com.cn/Jweb_prai/CN/Y2022/V35/I12/1064
版权所有 © 《模式识别与人工智能》编辑部
地址:安微省合肥市蜀山湖路350号 电话:0551-65591176 传真:0551-65591176 Email:bjb@iim.ac.cn
本系统由北京玛格泰克科技发展有限公司设计开发 技术支持:support@magtech.com.cn