Hierarchical Proxy-based Loss for Deep Metric Learning
Hierarchical Proxy-based Loss for Deep Metric Learning
复制标题
DOI：
10.1109/wacv51458.2022.00052
复制
发表时间：
2021-03
期刊：
影响因子：
--
通讯作者：
Zhibo Yang;M. Bastan;Xinliang Zhu;Douglas Gray;D. Samaras
中科院分区：
文献类型：
--
作者：
Zhibo Yang;M. Bastan;Xinliang Zhu;Douglas Gray;D. Samaras
基于代理的度量学习损失上级基于对的损失，因为它们的快速收敛和低训练复杂度。然而，现有的基于代理的损失集中在学习类判别功能，而忽略了跨类共享的共性，这是潜在的有用的描述和匹配样本。此外，它们忽略了真实世界数据集中类别的隐式层次结构，其中类似的从属类可以组合在一起。在本文中，我们提出了一个框架，利用这种隐式层次结构的代理上施加一个层次结构，并可用于任何现有的基于代理的损失。这使得我们的模型能够在不破坏隐式数据层次结构的情况下捕获类区分特征和类共享特征。我们评估我们的方法上五个既定的图像检索数据集，如在商店和SOP。结果表明，我们的分层代理为基础的损失框架，提高了现有的代理为基础的损失的性能，特别是在大型数据集，表现出较强的层次结构。
Proxy-based metric learning losses are superior to pair-based losses due to their fast convergence and low training complexity. However, existing proxy-based losses focus on learning class-discriminative features while overlooking the commonalities shared across classes which are potentially useful in describing and matching samples. Moreover, they ignore the implicit hierarchy of categories in real-world datasets, where similar subordinate classes can be grouped together. In this paper, we present a framework that leverages this implicit hierarchy by imposing a hierarchical structure on the proxies and can be used with any existing proxy-based loss. This allows our model to capture both class-discriminative features and class-shared characteristics without breaking the implicit data hierarchy. We evaluate our method on five established image retrieval datasets such as In-Shop and SOP. Results demonstrate that our hierarchical proxy-based loss framework improves the performance of existing proxy-based losses, especially on large datasets which exhibit strong hierarchical structure.