Efficient Minimum Error Bounded Particle Resampling L1 Tracker With Occlusion Detection
Efficient Minimum Error Bounded Particle Resampling L1 Tracker With Occlusion Detection
复制标题
具有遮挡检测功能的高效最小误差有界粒子重采样 L1 跟踪器
DOI：
10.1109/tip.2013.2255301
复制
发表时间：
2013-07-01
影响因子：
10.6
通讯作者：
Bai, Li
中科院分区：
文献类型：
--
作者：
Mei, Xue;Ling, Haibin;Bai, Li
近年来，稀疏表示被应用于视觉跟踪，以从目标模板子空间中找到具有最小重建误差的目标。虽然有效，但这些L1跟踪器需要高的计算成本，这是由于大量的计算l（1）最小化。此外，l（1）最小化的固有遮挡不敏感性尚未完全表征。在本文中，我们提出了一个有效的L1跟踪器，命名为有界粒子reserve（BPR）-L1跟踪器，具有最小误差界和遮挡检测。首先，从线性最小二乘方程计算最小误差界，并作为粒子滤波器（PF）框架中的粒子resception的指导。在两步测试中，在解决计算上昂贵的l（1）最小化之前，去除大多数不重要的样本。第一步，称为tau测试，将样本观测可能性与一组有序阈值进行比较，以在不损失恢复精度的情况下删除不重要的样本。第二步，命名为最大测试，确定相对于目标的最大样本概率，以进一步去除不重要的样本，而不改变当前帧的跟踪结果。虽然在重新测试期间牺牲了最小的精度，但max测试在tau测试的基础上实现了显著的速度提升。BPR-L1技术还可以有益于PF框架中具有最小误差界限的其他跟踪器，特别是对于基于稀疏表示的跟踪器。在误差界计算之后，BPR-L1通过研究l（1）最小化中的平凡系数来执行遮挡检测。通过设计，这些系数包含关于图像损坏（包括遮挡）的丰富信息。然后使用检测到的遮挡来增强模板更新。为了进行评估，我们进行了三个视频应用程序的实验：生物识别（头部运动，手持物体，舞台上的歌手），行人（城市旅行，走廊监控），和交通中的汽车（广域运动图像，地面安装的角度）。所提出的BPR-L1方法表现出优异的性能相比，9个国家的最先进的跟踪器在11个具有挑战性的基准序列。
Recently, sparse representation has been applied to visual tracking to find the target with the minimum reconstruction error from a target template subspace. Though effective, these L1 trackers require high computational costs due to numerous calculations for l(1) minimization. In addition, the inherent occlusion insensitivity of the l(1) minimization has not been fully characterized. In this paper, we propose an efficient L1 tracker, named bounded particle resampling (BPR)-L1 tracker, with a minimum error bound and occlusion detection. First, the minimum error bound is calculated from a linear least squares equation and serves as a guide for particle resampling in a particle filter (PF) framework. Most of the insignificant samples are removed before solving the computationally expensive l(1) minimization in a two-step testing. The first step, named tau testing, compares the sample observation likelihood to an ordered set of thresholds to remove insignificant samples without loss of resampling precision. The second step, named max testing, identifies the largest sample probability relative to the target to further remove insignificant samples without altering the tracking result of the current frame. Though sacrificing minimal precision during resampling, max testing achieves significant speed up on top of tau testing. The BPR-L1 technique can also be beneficial to other trackers that have minimum error bounds in a PF framework, especially for trackers based on sparse representations. After the error-bound calculation, BPR-L1 performs occlusion detection by investigating the trivial coefficients in the l(1) minimization. These coefficients, by design, contain rich information about image corruptions, including occlusion. Detected occlusions are then used to enhance the template updating. For evaluation, we conduct experiments on three video applications: biometrics (head movement, hand holding object, singers on stage), pedestrians (urban travel, hallway monitoring), and cars in traffic (wide area motion imagery, ground-mounted perspectives). The proposed BPR-L1 method demonstrates an excellent performance as compared with nine state-of-the-art trackers on eleven challenging benchmark sequences.