Progressive Cross-Camera Soft-Label Learning for Semi-Supervised Person Re-Identification
Progressive Cross-Camera Soft-Label Learning for Semi-Supervised Person Re-Identification
复制标题
用于半监督人员重新识别的渐进式跨摄像头软标签学习
DOI：
10.1109/tcsvt.2020.2983600
复制
发表时间：
2019-08
影响因子：
8.4
通讯作者：
Yang Gao
中科院分区：
文献类型：
--
作者：
Lei Qi;Lei Wang;Jing Huo;Yinghuan Shi;Yang Gao
在本文中，我们主要研究只有摄像机内(摄像机内)标签而没有摄像机间(摄像机交叉)标签的半监督人员重新识别(Re-ID)的情况。在现实应用中，与相机间标签相比，这些相机内标签可以很容易地通过跟踪算法或很少的人工注释来捕获。在这种情况下，由于缺乏跨镜头的标签信息，在训练阶段很难探索跨镜头的人之间的关系。针对这一问题，我们提出了一种新的用于半监督人员Re-ID任务的渐进式跨摄像机软标签学习(PCSL)框架，该框架可以生成跨摄像机软标签并利用它们来优化网络。具体地，我们计算了一个基于人级特征的亲和度矩阵，并对其进行调整以产生跨摄像机人物之间的相似性(即跨摄像机软标签)。为了利用这些软标签来训练网络，我们分别从分类和判别的角度研究了加权交叉熵损失和加权三重损失。特别是，该框架交替生成渐进的跨摄像机软标签，并在整个学习过程中逐步改进特征表示。在五个大规模基准数据集上的大量实验表明，PCSL的性能明显优于使用标记源域或基于Gans模型生成的图像的最先进的非监督方法。此外，该方法甚至具有与深度监督Re-ID方法相当的性能。
In this paper, we focus on the semi-supervised person re-identification (Re-ID) case, which only has the intra-camera (within-camera) labels but not inter-camera (cross-camera) labels. In real-world applications, these intra-camera labels can be readily captured by tracking algorithms or few manual annotations, when compared with cross-camera labels. In this case, it is very difficult to explore the relationships between cross-camera persons in the training stage due to the lack of cross-camera label information. To deal with this issue, we propose a novel Progressive Cross-camera Soft-label Learning (PCSL) framework for the semi-supervised person Re-ID task, which can generate cross-camera soft-labels and utilize them to optimize the network. Concretely, we calculate an affinity matrix based on person-level features and adapt them to produce the similarities between cross-camera persons (i.e., cross-camera soft-labels). To exploit these soft-labels to train the network, we investigate the weighted cross-entropy loss and the weighted triplet loss from the classification and discrimination perspectives, respectively. Particularly, the proposed framework alternately generates progressive cross-camera soft-labels and gradually improves feature representations in the whole learning course. Extensive experiments on five large-scale benchmark datasets show that PCSL significantly outperforms the state-of-the-art unsupervised methods that employ labeled source domains or the images generated by the GANs-based models. Furthermore, the proposed method even has a competitive performance with respect to deep supervised Re-ID methods.
登录
查看更多内容
DOI：
--
发表时间：
2017-03
期刊：
ArXiv
影响因子：
--
作者：
Alexander Hermans;Lucas Beyer;B. Leibe
通讯作者：
Alexander Hermans;Lucas Beyer;B. Leibe
DOI：
10.1109/iccv.2017.550
发表时间：
2017-09
期刊：
2017 IEEE International Conference on Computer Vision (ICCV)
影响因子：
--
作者：
Mang Ye;A. J. Ma;Liang Zheng;Jiawei Li;P. Yuen
通讯作者：
Mang Ye;A. J. Ma;Liang Zheng;Jiawei Li;P. Yuen
DOI：
10.1109/cvpr.2018.00242
发表时间：
2018-03
期刊：
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
影响因子：
--
作者：
Jingya Wang;Xiatian Zhu;S. Gong;Wei Li-
通讯作者：
Jingya Wang;Xiatian Zhu;S. Gong;Wei Li-
DOI：
10.1007/978-3-319-46466-4_52
发表时间：
2016-10
期刊：
--
影响因子：
--
作者：
Liang Zheng;Zhi Bie;Yifan Sun;Jingdong Wang;Chi Su;Shengjin Wang;Q. Tian
通讯作者：
Liang Zheng;Zhi Bie;Yifan Sun;Jingdong Wang;Chi Su;Shengjin Wang;Q. Tian
DOI：
10.1109/cvpr.2016.152
发表时间：
2016-06
期刊：
2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
影响因子：
--
作者：
Tetsu Matsukawa;Takahiro Okabe;Einoshin Suzuki;Yoichi Sato
通讯作者：
Tetsu Matsukawa;Takahiro Okabe;Einoshin Suzuki;Yoichi Sato