Online Joint Multi-Metric Adaptation From Frequent Sharing-Subset Mining for Person Re-Identification
Online Joint Multi-Metric Adaptation From Frequent Sharing-Subset Mining for Person Re-Identification
复制标题
DOI：
10.1109/cvpr42600.2020.00298
复制
发表时间：
2020-06
期刊：
影响因子：
--
通讯作者：
Jiahuan Zhou;Bing Su;Ying Wu
中科院分区：
文献类型：
--
作者：
Jiahuan Zhou;Bing Su;Ying Wu
人物身份识别（Person Re-Identification，P-RID）作为一个实例级的识别问题，在计算机视觉领域仍然是一个具有挑战性的问题。许多P-RID工作旨在从离线训练数据中学习忠实和有区别的特征/指标，并直接将其用于看不见的在线测试数据。然而，由于训练数据和测试数据之间存在严重的数据转移问题，它们的性能在很大程度上受到限制。因此，我们提出了一个在线联合多度量自适应模型，通过学习一系列的度量为所有的共享子集，以适应离线学习的P-RID模型的在线数据。每个共享子集是从所提出的新的频繁共享子集挖掘模块中获得的，并且包含一组测试样本，这些测试样本彼此之间具有很强的视觉相似性关系。与现有的在线P-RID方法不同，我们的模型同时考虑了测试样本之间的样本特定判别式和基于集合的视觉相似性，因此自适应的多个度量可以通过多核后期融合框架联合细化所有给定测试样本的判别式。我们提出的模型通常适用于任何离线学习的P-RID基线以进行在线增强，我们模型的性能改进不仅通过对几个广泛使用的P-RID基准测试（CUHK 03、Market 1501、DukeMTMC-reID和MSMT 17）的广泛实验来验证。）和最先进的P-RID基线，而且还得到了提供的深入理论分析的保证。
Person Re-IDentification (P-RID), as an instance-level recognition problem, still remains challenging in computer vision community. Many P-RID works aim to learn faithful and discriminative features/metrics from offline training data and directly use them for the unseen online testing data. However, their performance is largely limited due to the severe data shifting issue between training and testing data. Therefore, we propose an online joint multi-metric adaptation model to adapt the offline learned P-RID models for the online data by learning a series of metrics for all the sharing-subsets. Each sharing-subset is obtained from the proposed novel frequent sharing-subset mining module and contains a group of testing samples which share strong visual similarity relationships to each other. Unlike existing online P-RID methods, our model simultaneously takes both the sample-specific discriminant and the set-based visual similarity among testing samples into consideration so that the adapted multiple metrics can refine the discriminant of all the given testing samples jointly via a multi-kernel late fusion framework. Our proposed model is generally suitable to any offline learned P-RID baselines for online boosting, the performance improvement by our model is not only verified by extensive experiments on several widely-used P-RID benchmarks (CUHK03, Market1501, DukeMTMC-reID and MSMT17) and state-of-the-art P-RID baselines but also guaranteed by the provided in-depth theoretical analyses.