arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2209.09574cs.CV
收藏

面向长期身份识别的采样不可知特征表示

Sampling Agnostic Feature Representation for Long-Term Person Re-identification

Seoul National University of Science and Technology(首尔科学技术大学)

Seongyeop Yang, Byeongkeun Kang, Yeejin Lee

摘要:

人员重新识别是跨非重叠相机识别个体的问题。虽然重新识别问题已经取得了显着的进展,它仍然是一个具有挑战性的问题,由于外观变化的同一个人以及其他人的相似外观。一些先前的工作通过将正样本的特征与负样本的特征分离来解决该问题。然而,现有模型的性能在很大程度上取决于用于训练的样本的特征和统计。因此,我们提出了一个新的框架,称为采样独立的鲁棒特征表示网络(SirNet),学习从随机选择的样本解开特征嵌入。一个精心设计的抽样独立的最大差异损失被引入到模型的样本作为一个集群的同一个人。因此,所提出的框架可以使用所学习的特征来生成额外的硬否定/肯定,这导致与其他身份的更好的可辨别性。在大规模基准数据集上的大量实验结果验证了该模型比现有的最先进的模型更有效。

英文摘要:

Person re-identification is a problem of identifying individuals across non-overlapping cameras. Although remarkable progress has been made in the re-identification problem, it is still a challenging problem due to appearance variations of the same person as well as other people of similar appearance. Some prior works solved the issues by separating features of positive samples from features of negative ones. However, the performances of existing models considerably depend on the characteristics and statistics of the samples used for training. Thus, we propose a novel framework named sampling independent robust feature representation network (SirNet) that learns disentangled feature embedding from randomly chosen samples. A carefully designed sampling independent maximum discrepancy loss is introduced to model samples of the same person as a cluster. As a result, the proposed framework can generate additional hard negatives/positives using the learned features, which results in better discriminability from other identities. Extensive experimental results on large-scale benchmark datasets verify that the proposed model is more effective than prior state-of-the-art models.

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。

↑