DeepAI AI Chat
Log In Sign Up

Construct Informative Triplet with Two-stage Hard-sample Generation

by   Chuang Zhu, et al.
Beijing University of Posts and Telecommunications

In this paper, we propose a robust sample generation scheme to construct informative triplets. The proposed hard sample generation is a two-stage synthesis framework that produces hard samples through effective positive and negative sample generators in two stages, respectively. The first stage stretches the anchor-positive pairs with piecewise linear manipulation and enhances the quality of generated samples by skillfully designing a conditional generative adversarial network to lower the risk of mode collapse. The second stage utilizes an adaptive reverse metric constraint to generate the final hard samples. Extensive experiments on several benchmark datasets verify that our method achieves superior performance than the existing hard-sample generation algorithms. Besides, we also find that our proposed hard sample generation method combining the existing triplet mining strategies can further boost the deep metric learning performance.


page 25

page 29


Symmetrical Synthesis for Deep Metric Learning

Deep metric learning aims to learn embeddings that contain semantic simi...

No Pairs Left Behind: Improving Metric Learning with Regularized Triplet Objective

We propose a novel formulation of the triplet objective function that im...

Learning Incremental Triplet Margin for Person Re-identification

Person re-identification (ReID) aims to match people across multiple non...

Improving Generalization of Metric Learning via Listwise Self-distillation

Most deep metric learning (DML) methods employ a strategy that forces al...

Hardness-Aware Deep Metric Learning

This paper presents a hardness-aware deep metric learning (HDML) framewo...

Robust Visual Tracking via Statistical Positive Sample Generation and Gradient Aware Learning

In recent years, Convolutional Neural Network (CNN) based trackers have ...

Improving Deep Metric Learning with Virtual Classes and Examples Mining

In deep metric learning, the training procedure relies on sampling infor...