End-to-end distance metric learning (DML) has been applied to obtain features useful in many computer vision tasks. However, these DML studies have not provided equitable comparisons between features extracted from DML-based networks and softmax-based networks. In this paper, we present objective comparisons between these two approaches under the same network architecture.

中文翻译：

---

与基于距离度量学习的功能相比，基于Softmax的功能的重要性。

端到端距离度量学习（DML）已应用于获得许多计算机视觉任务中有用的功能。但是，这些DML研究没有提供从基于DML的网络和基于softmax的网络中提取的特征之间的公平比较。在本文中，我们提出了在相同网络架构下这两种方法之间的客观比较。