跨模态目标重识别研究综述
Survey on Cross-modality Object Re-identification Research
崔振宇 1周嘉欢 1彭宇新1
作者信息
- 1. 北京大学王选计算机研究所 北京100871;多媒体信息处理国家重点实验室 北京100871
- 折叠
摘要
目标重识别(ReID)技术旨在匹配不同区域摄像头在不同时间拍摄到的同一目标,其核心是通过目标间的细粒度差异实现不同 目标的有效区分.因此,目标重识别技术被广泛应用于安防布控、刑侦监控等领域并发挥了重要作用.传统的目标重识别技术通常适用于光照条件良好情况下的可见光模态数据,但在处理黑夜低光照条件下的 目标重识别任务时,其性能通常受到严重限制.红外摄像机因其卓越的夜视性能,通常被应用于在低光照条件下采集目标红外图像.因此,跨模态目标重识别技术旨在通过可见光图像匹配红外图像,实现全天候不间断的目标重识别.近年来,跨模态目标重识别技术取得了很大进展,然而,对于现有模型的归纳总结及深入分析仍然欠缺.为此,对跨模态目标重识别领域的相关研究和新颖方法进行了深入调研和总结,讨论了现有方法在实际场景中面临的挑战,并从模型分类和模型评价两个方面对现有方法进行归纳与分析.首先,围绕跨模态目标重识别问题的研究难点,将跨模态目标重识别分为生成式方法和非生成式方法两大类;然后,对当前跨模态重识别领域中广泛使用的评测数据集以及相关评价指标进行了综述与总结;最后,讨论了跨模态重识别领域仍然存在的挑战并对未来发展趋势进行了展望.
Abstract
Object re-identification(ReID)technology aims to match the same object captured by cameras across different areas at different time.The key is to distinguish different objects through fine-grained differences between different individuals,which is widely used in security control,criminal investigation and monitoring,etc.Traditional ReID technology is usually suitable for visi-ble cameras with good lighting conditions,but its performance is severely limited under low-light conditions.The infrared camera is often used to collect infrared images of objects under low light conditions due to its outstanding night vision performance.Therefore,cross-modality object re-identification technology focuses on achieving uninterrupted object ReID across day and night from visible images to infrared images(Ⅵ-ReID),and vice versa.In recent years,Ⅵ-ReID technology has made significant pro-gress.However,a comprehensive summary and in-depth analysis of existing models are still lacking.To this end,this paper con-ducts an in-depth investigation and summary of relevant research and novel methods in the field of Ⅵ-ReID.It discusses the chal-lenges faced by existing methods in actual scenarios,and categorizes them from two aspects:model classification and model evalu-ation.First,focusing on the research challenges,Ⅵ-ReID is categorized into generative methods and non-generative methods.Se-condly,the evaluation datasets and evaluation metrics are reviewed and summarized.Finally,the remaining challenges in Ⅵ-ReID are discussed and the future development trends are prospected.
关键词
计算机视觉/目标重识别/跨模态/细粒度特征/表征学习Key words
Computer vision/Object re-identification/Cross-modality/Fine-grained feature/Representation learning引用本文复制引用
基金项目
国家自然科学基金(61925201)
国家自然科学基金(62132001)
出版年
2024
NETL
NSTL
万方数据