[论文解读] Attention-Aware Compositional Network for Person Re-identification

Jing Xu, Rui Zhao+3 more|arXiv (Cornell University)|May 9, 2018

Video Surveillance and Tracking Methods参考文献 36被引用 82

一句话总结

本文提出一种注意力感知的组成网络，以提升行人重识别性能，在 CUHK03-NP 和 DukeMTMC-reID 上进行评估。

ABSTRACT

Person re-identification (ReID) is to identify pedestrians observed from different camera views based on visual appearance. It is a challenging task due to large pose variations, complex background clutters and severe occlusions. Recently, human pose estimation by predicting joint locations was largely improved in accuracy. It is reasonable to use pose estimation results for handling pose variations and background clutters, and such attempts have obtained great improvement in ReID performance. However, we argue that the pose information was not well utilized and hasn't yet been fully exploited for person ReID. In this work, we introduce a novel framework called Attention-Aware Compositional Network (AACN) for person ReID. AACN consists of two main components: Pose-guided Part Attention (PPA) and Attention-aware Feature Composition (AFC). PPA is learned and applied to mask out undesirable background features in pedestrian feature maps. Furthermore, pose-guided visibility scores are estimated for body parts to deal with part occlusion in the proposed AFC module. Extensive experiments with ablation analysis show the effectiveness of our method, and state-of-the-art results are achieved on several public datasets, including Market-1501, CUHK03, CUHK01, SenseReID, CUHK03-NP and DukeMTMC-reID.

研究动机与目标

• 在外观和视角多变的条件下推动鲁棒的行人重识别。
• 开发一个注意力感知的架构，用于将特征组成以实现精确识别。
• 利用 CUHK03-NP 和 DukeMTMC-reID 数据集来证明有效性。
• 探索注意力引导如何提升重识别任务中的特征表征。

提出的方法

• 引入一个注意力感知的组成网络架构用于行人重识别。
• 结合注意力机制以引导跨身体部位的特征组成。
• 利用组成策略整合局部与全局线索以获得鲁棒的身份表征。
• 结合适用于重识别基准的训练策略来优化网络。
• 在标准数据集上评估模型以证明相对于基线的改进。

实验结果

研究问题

• RQ1 注意力引导的特征组成对行人重识别性能的影响是什么？
• RQ2 注意力感知的组成方法是否能提高对姿态、光照和视角变化的鲁棒性？
• RQ3 所提出的方法与在 CUHK03-NP 和 DukeMTMC-reID 上的现有重识别方法相比如何？

主要发现

• 所提出的网络在标准基准上显示出改进的重识别性能。
• 注意力引导的组成在具有挑战性条件下提升了身份之间的辨别度。
• 该方法在 CUHK03-NP 和 DukeMTMC-reID 数据集上显示出有效性。
• 该方法通过注意机制整合局部与全局线索，获得鲁棒的表示。