摘要
跨模态行人重识别,又称可见光-红外行人重识别,是一项在可见光图像与红外图像间进行行人识别与检索的技术。该技术作为视频智能监控系统的核心关键技术之一,目前在安防监控和疑犯追踪等领域被迫切需要。近年来,随着新型摄像头(可在弱光环境下自动切换到红外模式,拍摄红外图像)逐渐代替可见光摄像头并普及起来,跨模态行人重识别得到了学术界和工业界的广泛关注并取得显著的发展。对此,本文分别对跨模态行人重识别的概念以及相关研究进行介绍与分析,并对目前相关研究工作所存在的问题进行总结与展望。
Cross-modal person re-identification, also known as visible-infrared person re-identification, is a person recognition and retrieval technology between visible and infrared images. As one of the key technologies of video intelligent surveillance system, this technology is urgently needed in security monitoring and suspect tracking. In recent years, as cameras(which automatically switch to infrared modal in poor light environment and capture infrared images) are gradually replacing visible cameras and gradually becoming popular, Cross-modal person re-identification has been widely concerned by academia and industry,achieving remarkable development.This paper introduces the concept of Cross-modal person re-identification and analyzes related research, and summarizes and prospects the existing problems in the current research work.
作者
刘玉林
LIU Yulin(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)
出处
《电视技术》
2022年第5期9-11,共3页
Video Engineering
关键词
跨模态行人重识别
双流网络
生成对抗网络
模态差异
cross-modal person re-identification
dual-flow network
adversarial generation network
modal differences
