论文标题
重新识别的集群级特征对齐
Cluster-level Feature Alignment for Person Re-identification
论文作者
论文摘要
实例级别的对齐被广泛利用以重新确定人，例如空间对准，潜在的语义对准和三重对齐。本文探讨了另一种特征对齐方式，即整个数据集的集群级特征对齐，在该数据集中，模型不仅可以在本地迷你批次中看到采样的图像，还可以从蒸馏锚中看到整个数据集的全局特征分布。为了实现这一目标，我们提出了锚定损失，并研究了集群级特征对齐的许多变体，该变体由数据集概述中的迭代聚合和对齐组成。我们的广泛实验表明，在传统培训饱和后，通过小型培训工作可以提供一致且显着的绩效提高。在理论方面和实验方面，我们提出的方法都可能导致更稳定，具有指导性的优化，以更好地表示和泛化，以使其对准良好的嵌入。
Instance-level alignment is widely exploited for person re-identification, e.g. spatial alignment, latent semantic alignment and triplet alignment. This paper probes another feature alignment modality, namely cluster-level feature alignment across whole dataset, where the model can see not only the sampled images in local mini-batch but the global feature distribution of the whole dataset from distilled anchors. Towards this aim, we propose anchor loss and investigate many variants of cluster-level feature alignment, which consists of iterative aggregation and alignment from the overview of dataset. Our extensive experiments have demonstrated that our methods can provide consistent and significant performance improvement with small training efforts after the saturation of traditional training. In both theoretical and experimental aspects, our proposed methods can result in more stable and guided optimization towards better representation and generalization for well-aligned embedding.
