通过动态重标注与混合采样，提升跨摄像头人物识别泛化能力
DynaMix: Generalizable Person Re-identification via Dynamic Relabeling and Mixed Data Sampling
- 动态重标注伪标签，实时优化单摄像头身份信息
- 支持百万图像、数十万身份的大规模训练，性能超越现有方法
- 适合需要跨场景泛化的人物识别系统开发者
可泛化的行人重识别（Re-ID）旨在跨未见摄像头和环境识别个体。现有方法严重依赖有限的多摄像头标注数据，本文提出DynaMix，有效融合人工标注的多摄像头数据与大规模伪标注的单摄像头数据。不同于以往工作，DynaMix通过三个核心组件动态适应训练数据的结构与噪声：(1) 重标注模块，在线精炼单摄像头身份的伪标签；(2) 高效中心点模块，在大规模身份空间中保持稳健的身份表征；(3) 数据采样模块，精心组合混合数据小批量，平衡学习复杂度与批内多样性。所有组件均针对大规模高效设计，支持在百万级图像和数十万身份上有效训练。大量实验表明，DynaMix在可泛化行人Re-ID任务中持续优于当前最优方法。
原文摘要 · Abstract (English)
Generalizable person re-identification (Re-ID) aims to recognize individuals across unseen cameras and environments. While existing methods rely heavily on limited labeled multi-camera data, we propose DynaMix, a novel method that effectively combines manually labeled multi-camera and large-scale pseudo-labeled single-camera data. Unlike prior works, DynaMix dynamically adapts to the structure and noise of the training data through three core components: (1) a Relabeling Module that refines pseudo-labels of single-camera identities on-the-fly; (2) an Efficient Centroids Module that maintains robust identity representations under a large identity space; and (3) a Data Sampling Module that carefully composes mixed data mini-batches to balance learning complexity and intra-batch diversity. All components are specifically designed to operate efficiently at scale, enabling effective training on millions of images and hundreds of thousands of identities. Extensive experiments demonstrate that DynaMix consistently outperforms state-of-the-art methods in generalizable person Re-ID.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。