Improving Deep Metric Learning with Virtual Classes and Examples Mining

TL;DR

This paper introduces MIRAGE, a novel generation-based approach using virtual classes to improve deep metric learning by enhancing sampling and mining strategies, leading to better performance on standard datasets.

Contribution

The paper proposes MIRAGE, a new method that employs virtual classes composed of generated examples to improve training in deep metric learning.

Findings

Virtual classes significantly improve performance on benchmark datasets.

MIRAGE outperforms existing generation-based methods.

Empirical results demonstrate enhanced sampling and mining effectiveness.

Abstract

In deep metric learning, the training procedure relies on sampling informative tuples. However, as the training procedure progresses, it becomes nearly impossible to sample relevant hard negative examples without proper mining strategies or generation-based methods. Recent work on hard negative generation have shown great promises to solve the mining problem. However, this generation process is difficult to tune and often leads to incorrectly labelled examples. To tackle this issue, we introduce MIRAGE, a generation-based method that relies on virtual classes entirely composed of generated examples that act as buffer areas between the training classes. We empirically show that virtual classes significantly improve the results on popular datasets (Cub-200-2011, Cars-196 and Stanford Online Products) compared to other generation methods.