Person Re-identification with Adversarial Triplet Embedding

TL;DR

This paper introduces Adversarial Triplet Embedding (ATE), a novel deep metric learning approach for person re-identification that generates adversarial triplets to improve feature embedding and overcome local optima issues.

Contribution

The paper proposes a unified framework that combines adversarial triplet generation with discriminative embedding, addressing limitations of traditional triplet loss methods.

Findings

ATE outperforms state-of-the-art methods on benchmark datasets.

Adversarial triplet generation enhances the robustness of feature embedding.

The minimax formulation provides a theoretically sound solution.

Abstract

Person re-identification is an important task and has widespread applications in video surveillance for public security. In the past few years, deep learning network with triplet loss has become popular for this problem. However, the triplet loss usually suffers from poor local optimal and relies heavily on the strategy of hard example mining. In this paper, we propose to address this problem with a new deep metric learning method called Adversarial Triplet Embedding (ATE), in which we simultaneously generate adversarial triplets and discriminative feature embedding in an unified framework. In particular, adversarial triplets are generated by introducing adversarial perturbations into the training process. This adversarial game is converted into a minimax problem so as to have an optimal solution from the theoretical view. Extensive experiments on several benchmark datasets demonstrate the effectiveness of the approach against the state-of-the-art literature.