[论文解读] Black Re-ID: A Head-shoulder Descriptor for the Challenging Problem of Person Re-Identification

Boqiang Xu, Lingxiao He+4 more|arXiv (Cornell University)|Aug 19, 2020

Video Surveillance and Tracking Methods参考文献 42被引用 8

一句话总结

本文提出了一种头肩自适应注意力网络（HAA），通过在黑色衣物或低光照条件下利用头肩特征来增强行人重识别（Re-ID）性能。该方法采用自适应注意力模块，动态加权全局特征与头肩特征，在Black-reID和标准Re-ID基准上均实现了最先进性能，在新提出的Black-reID数据集上实现了4.7%的mAP提升。

ABSTRACT

Person re-identification (Re-ID) aims at retrieving an input person image from a set of images captured by multiple cameras. Although recent Re-ID methods have made great success, most of them extract features in terms of the attributes of clothing (e.g., color, texture). However, it is common for people to wear black clothes or be captured by surveillance systems in low light illumination, in which cases the attributes of the clothing are severely missing. We call this problem the Black Re-ID problem. To solve this problem, rather than relying on the clothing information, we propose to exploit head-shoulder features to assist person Re-ID. The head-shoulder adaptive attention network (HAA) is proposed to learn the head-shoulder feature and an innovative ensemble method is designed to enhance the generalization of our model. Given the input person image, the ensemble method would focus on the head-shoulder feature by assigning a larger weight if the individual insides the image is in black clothing. Due to the lack of a suitable benchmark dataset for studying the Black Re-ID problem, we also contribute the first Black-reID dataset, which contains 1274 identities in training set. Extensive evaluations on the Black-reID, Market1501 and DukeMTMC-reID datasets show that our model achieves the best result compared with the state-of-the-art Re-ID methods on both Black and conventional Re-ID problems. Furthermore, our method is also proved to be effective in dealing with person Re-ID in similar clothing. Our code and dataset are avaliable on https://github.com/xbq1994/.

研究动机与目标

• 解决因黑色衣物或光照不足导致衣物特征被遮挡时的行人重识别挑战。
• 提出一种新型头肩描述符，以利用面部、发型和眼镜等在衣物线索缺失时具有判别性的特征。
• 设计一种自适应注意力机制，当输入图像中存在黑色衣物时，自动增强对头肩特征的依赖。
• 创建首个基准数据集Black-reID，包含1,274个身份且具有大量穿黑色衣物的样本，用于黑衣行人重识别问题。
• 证明该方法在Black-reID之外也具有泛化能力，在Market1501和DukeMTMC-reID等标准Re-ID数据集上也提升了性能。

提出的方法

• 设计双流网络：一个全局流用于全行人表征，一个头肩注意力（HSA）流用于局部头肩特征。
• 引入轻量级头肩定位层，替代现成的姿态估计器，提升效率与准确性。
• 实现自适应注意力模块，当输入图像包含黑色衣物时，学习分配更高的权重给头肩流。
• 在HSA流中使用GeM池化，学习动态池化参数p，平衡全局与局部特征聚合。
• 使用联合损失函数（结合三元组损失与交叉熵损失）端到端训练模型，实现判别性特征学习。
• 采用集成策略，根据输入图像内容自适应融合全局与头肩特征，提升鲁棒性。

实验结果

研究问题

• RQ1 在低光照或穿黑色衣物条件下，头肩特征能否作为衣物特征的可靠替代？
• RQ2 深度学习模型如何在衣物线索缺失或退化时动态优先使用头肩特征？
• RQ3 全局与头肩表征之间的自适应特征融合对不同光照与衣物条件下的Re-ID性能有何影响？
• RQ4 所提出的GeM池化机制与GAP和GMP相比，在捕捉判别性头肩特征方面表现如何？
• RQ5 当在专用的Black-reID数据集上训练时，该方法在标准Re-ID基准上的泛化能力如何？

主要发现

• 所提出的HAA模型在Black-reID数据集上相比最先进方法实现了4.7%的mAP提升和4.3%的rank-1准确率提升。
• 自适应注意力模块在Black-reID上使mAP提升0.7%、rank-1提升0.5%，在DukeMTMC-reID上分别提升1.1%和0.8%。
• GeM池化优于GAP与GMP，在Black-reID数据集上相比GAP实现3.8%的mAP提升，相比GMP实现1.7%的提升。
• 通过自适应注意力融合全局与头肩特征，相比仅使用全局特征，mAP提升7.8%，rank-1提升5.5%。
• 仅使用头肩特征在Black-reID上即达到34.2%的mAP和35.2%的rank-1，证明其即使在无全局特征时也具备强大判别能力。
• 模型泛化能力良好，在Market1501和DukeMTMC-reID上均达到最先进性能，证实其在Black-reID场景外也具有效性。