1 资源
2 名词释义
- Multi-target multi-camera tracking (MTMCT):多目标多相机跟踪;
- locality aware appearance metric (LAAM):局部邻域内目标外观的度量;
- re-identification (Re-ID):重识别;
- single camera tracking(SCT):单摄像头跟踪;
- multi-camera tracking (MCT):多摄像头跟踪;
3 摘要
Abstract: Multi-target multi-camera tracking (MTMCT) systems track targets across cameras. Due to the continuity of target trajectories, tracking systems usually restrict their data association within a local neighborhood. In single camera tracking, local neighborhood refers to consecutive frames;in multi-camera tracking, it refers to neighboring cameras that the target may appear successively. For similarity estimation, tracking systems often adopt appearance features learned from the re-identification (re-ID) perspective. Different from tracking, re-ID usually does not have access to the trajectory cues that can limit the search space to a local neighborhood. Due to its global matching property, the re-ID perspective requires to learn global appearance features. We argue that the mismatch between the local matching procedure in tracking and the global nature of re-ID appearance features may compromise MTMCT performance.
To fit the local matching procedure in MTMCT, in this work, we introduce locality aware appearance metric (LAAM). Specifically, we design an intra-camera metric for single camera tracking, and an inter-camera metric for multi-camera tracking. Both metrics are trained with data pairs sampled from their corresponding local neighborhoods, as opposed to global sampling in the re-ID perspective. We show that the locally learned metrics can be successfully applied on top of several globally learned reID features. With the proposed method, we report new stateof-the-art performance on the DukeMTMC dataset, and a substantial improvement on the CityFlow dataset.
摘要: 多目标多摄像机跟踪(MTMCT)系统跨摄像机跟踪目标。由于目标轨迹的连续性,跟踪系统通常将其数据关联限制在局部邻域内。在单摄像机跟踪中,局部邻域是指连续的帧;在多摄像机跟踪中,局部邻域是指目标可能连续出现的相邻摄像机。对于相似度估计,跟踪系统通常采用从再识别(re-ID)角度学习的外观特征。与跟踪不同的是,re-ID通常不能访问轨迹线索,而这些线索可以将搜索空间限制在一个本地社区。由于其全局匹配属性,reid透视图需要学习全局外观特性。我们认为,跟踪中的局部匹配过程与re-ID外观特征的全局性质之间的不匹配可能会影响MTMCT的性能。
为了适应MTMCT中的局部匹配过程,在本文中,我们引入了局域感知的外观度量(locality - aware appearance metric, LAAM)。具体来说,我们设计了用于单相机跟踪的相机内度量,以及用于多相机跟踪的相机间度量。这两个指标都使用从其相应的本地邻居中采样的数据对进行训练,而不是在re-ID透视图中进行全局采样。我们证明了局部学习的度量可以成功地应用于几个全局学习的reID特性之上。通过提出的方法,我们报告了DukeMTMC数据集的最新性能,以及CityFlow数据集的重大改进。
4 Overview
图1:重识别(ReID) 与多目标多相机跟踪(MTMCT) 任务之间的区别。给定一个查询,重识别在所有相机的图库中全局搜索真实的匹配图。相比之下,多目标多相机的跟踪在单摄像机跟踪(SCT) 只考虑相邻帧的匹配,在 多摄像机跟踪(MCT) 中只考虑相邻相机间的匹配。具体来说,多摄像头跟踪时,当目标出现在摄像头2中,就不考虑摄像头3,因为目标从未出现在这两个摄像头(摄像头可能太远了)。
图6:全局度量和LAAM之间的错误匹配比较。我们报告了验证集的假阳性率和假阴性率。
表3:IDF1在DukeMTMC验证集上的准确性。三个重识别特征在不同方法下的评估。
表4:CityFlow在线测试集结果。请注意,CityFlow数据集仅评估多摄像机跟踪。该方法具有较高的精度。
5 理解概述