Abstract
Real-world surveillance systems are dynamically evolving, requiring a person Re-identification model to continuously handle newly incoming data from various domains. To cope with these dynamics, Lifelong ReID (LReID) has been proposed to learn and accumulate knowledge across multiple domains incrementally. However, LReID models need to be trained on large-scale labeled data for each unseen domain, which are typically inaccessible due to privacy and cost concerns. In this paper, we propose a new paradigm called Continual Few-shot ReID (CFReID), which requires models to be incrementally trained using few-shot data and tested on all seen domains. Under few-shot conditions, CFREID faces two core challenges: 1) learning knowledge from few-shot data of unseen domain, and 2) avoiding catastrophic forgetting of seen domains. To tackle these two challenges, we propose a Stable Distribution Alignment (SDA) framework from feature distribution perspective. Specifically, our SDA is composed of two modules, i.e., Meta Distribution Alignment (MDA) and Prototype-based Few-shot Adaptation (PFA). To support the study of CFReID, we establish an evaluation benchmark for CFReID on five publicly available ReID datasets. Extensive experiments demonstrate that our SDA can enhance the few-shot learning and anti-forgetting capabilities under few-shot conditions. Notably, our approach, using only 5\% of the data, i.e., 32 IDs, significantly outperforms LReID's state-of-the-art performance, which requires 700 to 1,000 IDs.
Abstract (translated)
现实世界中的监控系统在不断演变，需要一个人重识别模型能够持续处理来自不同领域的新的数据。为应对这种动态变化，提出了终身重识别（Lifelong ReID，简称LReID）方法，旨在跨多个领域逐步学习和累积知识。然而，LReID模型通常需要针对每个未见过的领域使用大量标记的数据进行训练，这在实际操作中由于隐私保护和成本问题往往难以实现。 为了解决这些问题，在本文中我们提出了一个名为连续少量样本重识别（Continual Few-shot ReID, 简称CFReID）的新范式。在这种框架下，模型需要通过少量样本数据逐步进行增量训练，并在所有已见过的领域上进行测试。在少量样本条件下，CFReID面临两个核心挑战：1) 从未见过领域的少量样本数据中学习知识；2) 避免对已见过领域的遗忘现象。 为应对这两个挑战，我们提出了一种名为稳定分布对齐（Stable Distribution Alignment, 简称SDA）的框架，并从特征分布的角度出发解决这些问题。具体来说，我们的SDA包括两个模块：元分布对齐（Meta Distribution Alignment, 简称MDA）和基于原型的少量样本适应（Prototype-based Few-shot Adaptation, 简称PFA）。为了支持CFReID的研究工作，我们建立了五个公开可用重识别数据集上的评估基准。大量的实验结果表明，在少量样本条件下，我们的SDA能够增强模型的学习能力和抗遗忘能力。 值得注意的是，通过仅使用5%的数据（即32个身份），我们提出的方法在性能上显著超越了LReID的最佳表现方法，后者需要700到1,000个身份的数据。
URL
https://arxiv.org/abs/2503.18469
