Visual search for traffic signs: the effects of clutter, luminance, and aging

Latency and eye movement measures were used to examine the effects of aging, clutter, and luminance on visual search for traffic signs embedded in digitized images of driving scenes. Initially 14 older and 14 younger observers classified daytime and nighttime traffic scenes as containing low or high amounts of clutter. Next, an independent sample of 14 younger and 14 older participants searched for traffic signs contained within these scenes. Errors were more common among the elderly. Search efficiency declined with increased clutter and with aging. However, relative to the young, older adults did not suffer disproportionately as a result of increased clutter. The methods developed might be profitably employed to assess sign conspicuity and sign acquisition during driving.     

Smoothing innovations and data association with IPDA

Surveillance sensors return detections from targets as well as the clutter measurements. Data association algorithms often use innovations to discriminate between the target and the clutter measurements. Reducing the covariance of innovations reduces the surveillance area from which measurements are used, reducing the number of clutter measurements. This paper introduces smoothing innovations which reduce innovation covariance, and improve the data association performance. This concept is applied to the Integrated Probabilistic Data Association (IPDA) to produce a Smoothing IPDA (sIPDA). sIPDA trajectory estimation errors are reduced with a smoothing delay. A surprising outcome is that sIPDA improves the false track discrimination in real time (without the smoothing delay).     

Tracking Multiple Occluding People by Localizing on Multiple Scene Planes

Occlusion and lack of visibility in crowded and cluttered scenes make it difficult to track individual people correctly and consistently, particularly in a single view. We present a multi-view approach to solving this problem. In our approach we neither detect nor track objects from any single camera or camera pair; rather evidence is gathered from all the cameras into a synergistic framework and detection and tracking results are propagated back to each view. Unlike other multi-view approaches that require fully calibrated views our approach is purely image-based and uses only 2D constructs. To this end we develop a planar homographic occupancy constraint that fuses foreground likelihood information from multiple views, to resolve occlusions and localize people on a reference scene plane. For greater robustness this process is extended to multiple planes parallel to the reference plane in the framework of plane to plane homologies. Our fusion methodology also models scene clutter using the Schmieder and Weathersby clutter measure, which acts as a confidence prior, to assign higher fusion weight to views with lesser clutter. Detection and tracking are performed simultaneously by graph cuts segmentation of tracks in the space-time occupancy likelihood data. Experimental results with detailed qualitative and quantitative analysis, are demonstrated in challenging multi-view, crowded scenes.     

改进的KCF算法在车辆跟踪中的应用

针对核相关滤波算法(KCF)在复杂道路场景下难以应对因车辆尺度变化,遮挡及旋转而不能继续跟踪的问题,提出了一种新的跟踪方法来更好地实现复杂道路场景下的车辆跟踪。该方法借鉴快速分类尺度空间跟踪器(fDDST),采用一维尺度相关滤波器进行尺度估计。同时融合Kalman滤波器形成预测-跟踪-校准的跟踪机制。该机制结合遮挡处理能够保证系统在目标被严重遮挡时跟踪的准确性。在模型更新方面,在目标被遮挡时,自适应的调节学习率参数,及时纠正模型偏移、特征丢失等问题。实验结果表明,在复杂道路场景下车辆旋转 、遮挡及尺度变化时,均能有效地跟踪目标车辆,且具有良好的鲁棒性。     

GM‐PHD Filter with State‐Dependent Clutter

In traditional filtering methods, clutter is often assumed to obey a uniform distribution over the entire monitoring area. For many sensors, however, clutter may concentrate in target‐containing regions. Under this condition, the performance of the traditional multi‐target tracking filter can be degraded. In an effort to solve this problem, this paper proposes an improved algorithm based on a Gaussian Mixture probability hypothesis density (GM‐PHD) filter to deal with state‐dependent clutter. First, the relationship between state and clutter is modeled using the uniform distribution centered on the target state. Then, the clutter intensity is calculated according to the distribution of clutter in the whole monitoring area and is used to update the filter. The simulation results show that the improved filter can track targets’ trajectories more effectively in an environment of state‐dependent clutter than the standard GM‐PHD filter.     

视频序列中面向人的多目标跟踪算法

针对视频序列中人的跟踪问题，提出一种基于运动检测的多目标跟踪算法．跟踪系统由运动目标检测、关联矩阵建立、特殊情况判断及处理以及轨迹关联4部分构成．提出一种基于改进的c-均值聚类的自适应运动分割方法；不同情况下建立不同的关联矩阵，以准确判断实际场景状况；对遮挡问题作出处理，在两个目标遮挡不严重的情况下，分别采用均值漂移算法对其进行跟踪．实验结果表明，该算法具有较强的鲁棒性，能有效实现复杂场景下多目标跟踪．     

LoVPE 3维局部特征描述子

目的 局部特征描述子在3维目标识别等任务中能够有效地克服噪声、不同点云分辨率、局部遮挡、点云散乱分布等因素的干扰,但是已有3维描述子难以在性能和效率之间取得平衡,为此提出LoVPE（局部多视点投影视图相关编码）特征描述子用于复杂场景中的3维目标识别。方法 首先构建局部参考坐标系,将世界坐标系下的局部表面变换至关键点局部参考坐标系下的局部表面;然后绕局部参考坐标系各坐标轴旋转K个角度获得多视点局部表面,将局部表面内的点投影至局部参考系各坐标平面内,投影平面分成N×N块,统计每块内投影点的散布信息生成特征描述向量;最后将各视点特征描述向量进行两两视图对相关编码得到低维度特征描述向量,采用ZCA（零项分量分析）白化降低特征描述向量各维间相关性得到LoVPE描述子。结果 在公用数据集上进行不同描述子对噪声、不同分辨率、遮挡及杂波等干扰鲁棒性的特征匹配实验,实验结果表明,提出的描述子特征匹配率与现有最佳描述子处于同等水平,但保持了较低的特征维度和较高的计算效率,维度降低约1半、特征构建及特征匹配时间缩短为现有最佳描述子的1/4。结论 提出一种新的3维局部特征描述子,具有强描述能力,对噪声、不同网格分辨率、遮挡及杂波等具有强鲁棒性,存储消耗较少且计算效率较高,该方法适用于模型点云及真实点云数据,可用于复杂场景中的3维目标识别。     

基于自适应模板更新的改进孪生卷积网络目标跟踪算法

现有的孪生网络目标跟踪算法采用边界框模板进行跟踪,在目标形变、遮挡等干扰下很容易导致跟踪漂移。在轮廓检测网络和孪生卷积网络(Siamese)跟踪网络的基础上,提出一种基于深度轮廓模板更新的改进孪生卷积网络目标跟踪算法。利用轮廓检测网络获取目标边缘轮廓,降低背景杂波干扰;利用改进的Siamese网络获得轮廓模板和搜索区域的深度特征;通过相似性匹配获得最优跟踪目标。仿真实验结果表明,所提出的改进模型能够提高目标形变、遮挡等干扰下目标跟踪性能,具有较高的工程应用价值。     

抗遮挡目标跟踪的模型学习综述

视觉目标跟踪任务中的遮挡问题是最具挑战的场景属性之一,研究有效的抗遮挡模型学习方案,对构建适应复杂场景的长期鲁棒跟踪模型具有重要意义.剖析了遮挡影响跟踪性能的本质原因,以抗遮挡性能较好的先进跟踪算法为研究对象,系统分析了模型学习中有效抗遮挡机制,并对其改善长短期遮挡问题的有效性进行比较分析,包括以硬负样本挖掘、有效样本...     

一种复杂场景下的运动目标跟踪算法

提出了一种基于跟踪窗口自适应和抗遮挡的目标跟踪算法。采用Mean Shift算法确定当前帧的目标位置,最优选取核函数带宽,使跟踪窗口能够根据目标尺寸大小作出自适应调整。利用Bhattacharyya系数作为遮挡的判断依据,当目标遮挡时引入卡尔曼滤波器估计目标的运动信息,进行后续状态预测。实验表明,该算法能有效跟踪复杂场景下的运动目标。     

依赖模板匹配纠错的混合跟踪算法*

经典均值漂移跟踪器(MST)和模板匹配跟踪器(TMT)在性能的许多方面具有强互补性。为利用这种互补性提高跟踪的可靠性,提出了一种混合跟踪算法(HTA)。HTA维持了颜色直方图和亮度模板两种目标模型,并分别以TMT和MST为主、辅跟踪器。当目标被遮挡或周围出现相似颜色模式的干扰物而造成主跟踪器错跟时,算法自动切换到辅跟踪器进行纠错;干扰消失或目标被重新捕获时,可自动切换到主跟踪器。实验结果表明,HTA对复杂跟踪环境适应性强且计算量不大。     

基于杂波特性的自动跟踪起始试验统计方法

Two test statistics based on clutter characteristics are derived. A tentative track is confirmed when the track-is-on-clutter hypothesis is rejected. A constant and known false track rate results when the assumptions of the null hypothesis are true. The first test statistic is based on the clutter density. A high probability of target detection is resulted when the expected distance to the nearest target peak is less than the expected distance to the nearest clutter peak. The second test statistic is based on the clutter amplitude. A high probability of target detection is resulted when the expected amplitude of the target peak is greater than the expected amplitude of clutter peaks. The behavior of the clutter-based test statistics is compared with the target visibility method, using simulated data. All track initiation methods are applied using a track updater based on probabilistic data association (PDA), extended to incorporate peak amplitude information, which is available.     

基于DSmT的机载目标断续航迹关联

针对机载目标跟踪中常见的目标航迹断续问题,将多传感器获得的目标属性与状态等多特征信息相结合,建立中断前后各航迹段组合的关联统计量,以各特征关联度为证据并基于DSm理论进行合成,根据识别结果对同一目标中断前后的航迹段进行关联,提高航迹连续性并降低战场态势复杂度。仿真结果表明了该方法的有效性与可行性。     

加权间隔结构化支持向量机目标跟踪算法

目的 目标在跟踪过程中,各种因素的干扰使得跟踪结果存在不确定性。因此,将跟踪过程中所提取样本的可靠性融入跟踪模型中,有助于克服低可靠性样本对跟踪算法的影响。为此,基于最近的结构化支持向量机（SSVM）跟踪算法,提出一种包含样本置信度的加权间隔结构化支持向量机跟踪模型（WMSSVM）,以增强SSVM跟踪算法性能。方法 首先,基于打分和位置重合率估计样本可靠性;其次,建立WMSSVM模型处理具有不同置信度的跟踪样本训练问题,并采用对偶坐标下降优化算法求解跟踪模型。结果 在包含100个视频的OTB100跟踪数据集上进行测试,提出的WMSSVM跟踪器与基准跟踪器Scale-DLSSVM相比,在精准度和成功率两个指标上分别提高了1%和2%。与最近的跟踪算法相比,提出的方法也表现出更好的性能。结论 本文首次将样本的可靠性融入结构化支持向量机跟踪模型,并提出一种基于加权间隔的结构化支持向量机跟踪模型及其优化求解方法,在包含100个视频序列的跟踪数据集上验证了提出方法的有效性,本文提出的算法能够适应复杂场景下的跟踪任务,并在背景混杂、目标形变、遮挡、运动模糊、目标出界、快速位移等类别的视频中表现出优异的性能。     

基于粒子滤波与改进水平集的人手跟踪

提出一种基于肤色信息的改进水平集分割算法,给出整个算法的推导和实现过程,实现复杂背景下的精确人手轮廓分割,在进行人手跟踪时,使用粒子滤波对手的位置和大小进行跟踪,并用跟踪结果初始化水平集函数,以此加快轮廓曲线的收敛速度,获得手的轮廓后,对指尖位置进行定位。实验结果表明,该算法能够在复杂背景下实时、准确地跟踪人手轮廓和指尖位置。     

Surface signatures: an orientation independent free-form surface representation scheme for the purpose of objects registration and matching

This paper introduces anew free-form surface representation scheme for the purpose of fast and accurate registration and matching. Accurate registration of surfaces is a common task in computer vision. The proposed representation scheme captures the surface curvature information (seen from certain points) and produces images, called "surface signatures," at these points. Matching signatures of different surfaces enables the recovery of the transformation parameters between these surfaces. We propose using template matching to compare the signature images. To enable partial matching, another criterion, the overlap ratio is used. This representation scheme can be used as a global representation of the surface as well as a local one and performs near real-time registration. We show that the signature representation can be used to recover scaling transformation as well as matching objects in 3D scenes in the presence of clutter and occlusion. Applications presented include: free-form object matching, multimodal medical volumes registration, and dental teeth reconstruction from intraoral images.     

Event-averaged maximum likelihood estimation and mean-field theoryin multitarget tracking

This paper presents a novel type of Kalman filter for track maintenance in multitarget tracking using thresholded sensor data at high target/clutter densities and low detection levels. The filter is robust against tracking errors induced by crossing tracks, clutter, and missed detections, and the computational complexity of the filter scales well with problem size. There are two key features that differentiate this approach from earlier work. First, to reduce computational load, the filter exploits techniques from statistical field theory to simplify measurement to track association by using a mean-field approximation to sum over associations. Second, to enhance tracking of close together targets, the filter explicitly models the error correlations that occur between such target pairs. These error correlations are caused by measurement to track association ambiguities that arise when target separations are comparable to sensor measurement errors     

融合SPA遮挡分割的多目标跟踪方法

复杂环境下的多目标视频跟踪是计算机视觉领域的一个难点,有效处理目标间遮挡是解决多目标跟踪问题的关键。将运动分割方法引入目标跟踪领域,提出一种融合骨架点指派(SPA)遮挡分割的多目标跟踪方法。由底层光流信息得到骨架点,并估计骨架点遮挡状态;综合使用目标外观、运动、颜色信息等高级语义信息,将骨架点指派给各个目标;最后以骨架点为核,对运动前景密集分类,得到准确的目标前景像素;在粒子滤波器跟踪框架下,使用概率外观模型进行多目标跟踪。在PETS2009数据集上的实验结果表明,文中方法能够改进现有多目标跟踪方法对目标间交互适应性较差的缺点,更好地处理动态遮挡问题。