arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.03024cs.CV
收藏

重新思考无监督行人重识别中的采样策略

Rethinking Sampling Strategies for Unsupervised Person Re-identification

University of Chinese Academy of Sciences (UCAS)(中国科学院大学);Institute of North Electronic Equipment(北方电子设备研究所);Peng Cheng Laboratory(鹏城实验室);Tianjin University(天津大学)

Xumeng Han, Xuehui Yu, Guorong Li, Jian Zhao, Gang Pan, Qixiang Ye, Jianbin Jiao, Zhenjun Han

AI总结:

本文研究无监督行人重识别中的采样策略,提出 group sampling 分组采样以缓解过拟合并增强统计稳定性,在三个基准上取得与最先进方法相当或更优的性能。

摘要:

无监督行人重识别(re-ID)仍然是一项具有挑战性的任务。尽管已有大量研究聚焦于框架设计和损失函数,但本文表明采样策略同样发挥着重要作用。我们分析了在相同框架和损失函数下,不同采样策略之间产生性能差异的原因。我们认为,恶化的过拟合是导致性能不佳的一个重要因素,而增强统计稳定性可以纠正这一问题。受此启发,本文提出了一种简单而有效的方法,称为 group sampling(分组采样),它将来自同一类别的样本汇集为组。模型随后使用归一化的分组样本进行训练,这有助于减轻单个样本的负面影响。分组采样通过确保样本被更高效地划分到正确类别,更新了伪标签生成流程。它规范了表征学习过程,以渐进方式增强特征表征的统计稳定性。在 Market-1501、DukeMTMC-reID 和 MSMT17 上进行的大量实验表明,分组采样取得了与最先进方法相当的性能,并且在纯相机无关设置下优于现有技术。代码已发布于 https://github.com/ucas-vg/GroupSampling。

英文摘要:

Unsupervised person re-identification (re-ID) remains a challenging task. While extensive research has focused on the framework design and loss function, this paper shows that sampling strategy plays an equally important role. We analyze the reasons for the performance differences between various sampling strategies under the same framework and loss function. We suggest that deteriorated over-fitting is an important factor causing poor performance, and enhancing statistical stability can rectify this problem. Inspired by that, a simple yet effective approach is proposed, termed group sampling, which gathers samples from the same class into groups. The model is thereby trained using normalized group samples, which helps alleviate the negative impact of individual samples. Group sampling updates the pipeline of pseudo-label generation by guaranteeing that samples are more efficiently classified into the correct classes. It regulates the representation learning process, enhancing statistical stability for feature representation in a progressive fashion. Extensive experiments on Market-1501, DukeMTMC-reID and MSMT17 show that group sampling achieves performance comparable to state-of-the-art methods and outperforms the current techniques under purely camera-agnostic settings. Code has been available at https://github.com/ucas-vg/GroupSampling.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑