Paper Reading on Person Re-Idnetification

2020.06.21
ludics

1. Deep Learning for Person Re-identification: A Survey and Outlook

1.1 Introduction

1.2 Image-based

1.2.1 Hand-crafted Systems

1.2.2 Deeply-learned Systems

2. Deep Learning for Person Re-identification: A Survey and Outlook

2.1 Introduction

2.2 封闭世界行人ReID

2.2.1 特征表示学习

2.2.2 度量学习

3. Transferable, Controllable, and Inconspicuous Adversarial Attacks on Person Re-identification With Deep Mis-Ranking

Wang, H., Wang, G., Li, Y., Zhang, D., & Lin, L. (2020). Transferable, Controllable, and Inconspicuous Adversarial Attacks on Person Re-identification With Deep Mis-Ranking. Retrieved from link

3.1 Introduction

3.2 方法

3.2.1 Learning-to-Mis-Rank

3.2.2 为攻击学习可迁移特征

3.2.3 控制攻击像素的数量

3.2.4 视觉质量的感知损失

3.2.5 目标函数

3.3 实验