Following the popularity of Unsupervised Domain Adaptation (UDA) in person re-identification, the recently proposed setting of Online Unsupervised Domain Adaptation (OUDA) attempts to bridge the gap toward practical applications by introducing a consideration of streaming data. However, this still falls short of truly representing real-world applications. This paper defines the setting of Real-world Real-time Online Unsupervised Domain Adaptation (\(\hbox {R}^2\)OUDA) for Person Re-identification. The \(\hbox {R}^2\)OUDA setting sets the stage for true real-world real-time OUDA, bringing to light four major limitations found in real-world applications that are often neglected in current research: system generated person images, subset distribution selection, time-based data stream segmentation, and a segment-based time constraint. To address all aspects of this new \(\hbox {R}^2\)OUDA setting, this paper further proposes Real-World Real-Time Online Streaming Mutual Mean Teaching (\(\hbox {R}^2\)MMT), a novel multi-camera system for real-world person re-identification. Taking a popular person re-identification dataset, \(\hbox {R}^2\)MMT was used to construct over 100 data subsets and train more than 3000 models, exploring the breadth of the \(\hbox {R}^2\)OUDA setting to understand the training time and accuracy trade-offs and limitations for real-world applications. \(\hbox {R}^2\)MMT, a real-world system able to respect the strict constraints of the proposed \(\hbox {R}^2\)OUDA setting, achieves accuracies within \(0.1\%\) of comparable OUDA methods that cannot be applied directly to real-world applications.

随着无监督域适应（UDA）在行人重识别中的流行，最近提出的在线无监督域适应（OUDA）设置试图通过引入流数据的考虑来弥合与实际应用的差距。然而，这仍然无法真正代表现实世界的应用程序。本文定义了用于人员重新识别的真实世界实时在线无监督域适应（\(\hbox {R}^2\) OUDA）的设置。\ (\hbox {R}^2\)OUDA 设置为真正的现实世界实时 OUDA 奠定了基础，揭示了现实世界应用中发现的四个主要限制，这些限制在当前研究中经常被忽视：系统生成的人物图像、子集分布选择、基于时间的数据流分割，以及基于段的时间约束。为了解决这个新的\(\hbox {R}^2\) OUDA设置的各个方面，本文进一步提出了真实世界实时在线流媒体互均值教学（\(\hbox {R}^2\) MMT） ，一种新颖的多摄像头系统，用于现实世界的人员重新识别。以流行的行人重识别数据集为例，使用\(\hbox {R}^2\) MMT 构建了 100 多个数据子集并训练了 3000 多个模型，探索了\(\hbox {R}^2 \)OUDA 设置用于了解训练时间和准确性的权衡以及实际应用的限制。\(\hbox {R}^2\) MMT 是一个现实世界的系统，能够遵守提议的\(\hbox {R}^2\) OUDA 设置的严格约束，实现了\(0.1\%\以内的精度）无法直接应用于实际应用的类似 OUDA 方法。