[论文阅读 2019 综述 目标跟踪]Deep Learning for Visual Tracking: A Comprehensive Survey

admin管理员组

文章数量:1130349

简介

paper:Deep Learning for Visual Tracking: A Comprehensive Survey
github:MMarvasti/Deep-Learning-for-Visual-Tracking-Survey

这篇论文是之前看的，主要是对深度学习下跟踪算法的一个综述。

主要内容

由于是综述，内容涵盖较多，这里截取部分我感兴趣的内容。

首先，作者根据多个方面对跟踪算法进行了一个分类，如下图所示:

然后，论文中还介绍了一些在单目标跟踪中常用的评价指标:

• Center Location error(CLE): The CLE or precision metric is defined as the average Euclidean distance between the precise ground-truth locations of the target and estimated locations by a visual tracker(简单说就是预测位置与标签的欧式距离误差)
• Accuracy: The accuracy is then calculated by the average overlap scores (AOS) during the tracking when a visual tracker’s estimations have overlap with the ground-truth ones(即average IOU)
• Robustness/failure score:The robustness or failure score is defined as the number of required reinitializations when a tracker loses (or drifts) the target during the tracking task. The failure is detected when the overlap score drops to zero.(Robustness是跟踪器跟踪失败后重新初始化的次数，failure是失败总数，当IOU score为0时判定为failure)
• Expected average overlap(EAO):the EAO score is calculated as Φ ^ N s = ⟨ 1 N s ∑ i = 1 N s Φ i ⟩ \widehat{\Phi}_{N_{s}}=\left\langle\frac{1}{N_{s}} \sum_{i=1}^{N_{s}} \Phi_{i}\right\rangle Φ Ns​​=⟨Ns​1​∑i=1Ns​​Φi​⟩,where Φ i \Phi_{i} Φi​ is defined as the average of perframe overlaps until the end of sequences, even if failure leads to zero overlaps.

论文中还列举出了近些年SOTA的跟踪器:

(PS:”革命尚未成功，同志仍需努力“，还有好多论文要看o(╥﹏╥)o)

小结

大佬太多了!(溜~)

简介

paper:Deep Learning for Visual Tracking: A Comprehensive Survey
github:MMarvasti/Deep-Learning-for-Visual-Tracking-Survey

这篇论文是之前看的，主要是对深度学习下跟踪算法的一个综述。

主要内容

由于是综述，内容涵盖较多，这里截取部分我感兴趣的内容。

首先，作者根据多个方面对跟踪算法进行了一个分类，如下图所示:

然后，论文中还介绍了一些在单目标跟踪中常用的评价指标:

• Center Location error(CLE): The CLE or precision metric is defined as the average Euclidean distance between the precise ground-truth locations of the target and estimated locations by a visual tracker(简单说就是预测位置与标签的欧式距离误差)
• Accuracy: The accuracy is then calculated by the average overlap scores (AOS) during the tracking when a visual tracker’s estimations have overlap with the ground-truth ones(即average IOU)
• Robustness/failure score:The robustness or failure score is defined as the number of required reinitializations when a tracker loses (or drifts) the target during the tracking task. The failure is detected when the overlap score drops to zero.(Robustness是跟踪器跟踪失败后重新初始化的次数，failure是失败总数，当IOU score为0时判定为failure)
• Expected average overlap(EAO):the EAO score is calculated as Φ ^ N s = ⟨ 1 N s ∑ i = 1 N s Φ i ⟩ \widehat{\Phi}_{N_{s}}=\left\langle\frac{1}{N_{s}} \sum_{i=1}^{N_{s}} \Phi_{i}\right\rangle Φ Ns​​=⟨Ns​1​∑i=1Ns​​Φi​⟩,where Φ i \Phi_{i} Φi​ is defined as the average of perframe overlaps until the end of sequences, even if failure leads to zero overlaps.

论文中还列举出了近些年SOTA的跟踪器:

(PS:”革命尚未成功，同志仍需努力“，还有好多论文要看o(╥﹏╥)o)

小结

大佬太多了!(溜~)

本文标签： 目标论文deeplearningTracking