一种新的车辆遮挡检测与分割方法

遮挡是图像处理和运动物体跟踪过程中比较难解决的问题。针对交通视频中车辆遮挡问题,提出了一种基于统计模型的车辆遮挡分割算法。该算法在提取出运动车辆的基础上,使用横向-纵向扫描方法对运动车辆前景模板进行去空洞处理,得到完整的运动车辆区域;利用统计模型判断是否发生遮挡,如果判断车辆发生遮挡,在加入纠错机制的前提下得到正确的遮挡点,同时确定遮挡区域;利用边缘提取方法分割出遮挡区域,得到完全分隔的车辆。实验结果表明,该方法能够很好地解决车辆部分遮挡问题。     

在线多类别逐点式多目标跟踪与分割

现有基于深度学习的多目标跟踪算法大多利用目标检测任务预测的边界框跟踪目标,当目标间存在遮挡时,边界框会产生重叠进而影响跟踪准确度,针对这个问题,提出了一种在线多类别逐点式多目标跟踪与分割（category-free point-wise multi-object tracking and segmentation,CPMOTS）算法。该算法摒弃了边界框的目标表征方式,利用实例分割的像素级掩码表征目标进行跟踪,网络采用并行结构同时分割与跟踪多类别目标,并保证了运行效率,这在真实场景中有很强的实用性。CPMOTS首先利用实例分割网络得到实例分割掩码,对其采样得到无序点集;然后将点集的特征输入跟踪网络得到判别性的实例级嵌入向量;最后将该嵌入向量通过直观高效的注意力模块以显式建模其通道间的依赖关系,自适应学习每个特征通道的重要程度,依照这个重要程度选择性地强化有用的特征,抑制无用的特征,实现通道特征重标定,从而提高算法的性能。在多目标跟踪与分割基准数据集KITTI MOTS的实验表明,CPMOTS跟踪的精度优于大部分其他对比方法,并达到了16 frame/s的近实时速度。     

一种基于贪心搜索的实时多目标遮挡处理算法

提出了一种固定摄像机遮挡条件下的多目标跟踪算法,包括基于区域相关的运动前景分割、基于合并--分裂检测的数据关联和基于贪心搜索的遮挡目标定位三部分. 该算法的主要特点表现在: 1)将遮挡条件下的目标跟踪问题转化为一个已知目标数量和特征的图像分类问题; 2)用贪心搜索和积分图算法快速定位遮挡中的目标,保证了算法的实时性; 3)对目标数量无约束,能够处理多目标相互遮挡下的跟踪问题(发生遮挡的目标数量大于等于2), 且对目标的遮挡程度和目标运动模式无约束,具有良好的可扩展性. 采用手工标定的IBM多人遮挡数据库的测试结果证明了算法的有效性.     

时空特征融合网络的多目标跟踪与分割

目的 多目标跟踪与分割是计算机视觉领域一个重要的研究方向。现有方法多是借鉴多目标跟踪领域先检测然后进行跟踪与分割的思路,这类方法对重要特征信息的关注不足,难以处理目标遮挡等问题。为了解决上述问题,本文提出一种基于时空特征融合的多目标跟踪与分割模型,利用空间三坐标注意力模块和时间压缩自注意力模块选择出显著特征,以此达到优异的多目标跟踪与分割性能。方法 本文网络由2D编码器和3D解码器构成,首先将多幅连续帧图像输入到2D编码层,提取出不同分辨率的图像特征,然后从低分辨率的特征开始通过空间三坐标注意力模块得到重要的空间特征,通过时间压缩自注意力模块获得含有关键帧信息的时间特征,再将两者与原始特征融合,然后与较高分辨率的特征共同输入3D卷积层,反复聚合不同层次的特征,以此得到融合多次的既有关键时间信息又有重要空间信息的特征,最后得到跟踪和分割结果。结果 实验在YouTube-VIS(YouTube video instance segmentation)和KITTI MOTS(multi-object tracking and segmentation)两个数据集上进行定量评估。在YouTub...     

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．     

一种基于运动信息的多目标跟踪方法研究

为解决多目标跟踪中的遮挡问题,提出一种基于目标运动信息的方法。采用混合高斯模型结合背景差法获取初始运动信息,根据目标短时间内状态的稳定性,对其进行预测,再结合视觉特征达到精确跟踪。由于使用速度和视觉特征信息对目标单独跟踪,从而巧妙地避免遮挡的处理。实验结果表明,该方法实时有效,同时对遮挡问题的处理也有较好的效果。     

卫星视频中的单目标分割和跟踪

对于卫星视频图像中存在的目标与背景对比性低、缺乏目标特征信息等问题,提出一种结合目标运动信息、时空背景和外观模型的目标分割和跟踪方法.根据首帧定位得到目标区域,首先对目标使用方向梯度直方图方法提取特征利用核相关滤波器得到目标跟踪区域1;接着利用颜色空间特征建立目标与其周围区域上下文信息的空间模型得到目标跟踪区域2;然后利用视觉背景提取算法以像素为单位在目标区域上检测运动目标得到单目标的分割区域3;最后分别对3个区域进行相关计算得到最优区域作为最终目标跟踪位置和模板更新样本.实验结果表明,本文算法与KCF算法相比,跟踪的成功率和准确率有很大的提高,同时实现了单目标分割.     

基于数据关联矩阵的多目标跟踪算法

针对视频中的多目标跟踪问题,提出一种改进的基于数据关联矩阵的多目标跟踪算法,实现视频场景复杂环境下的多个目标跟踪。使用区间分布模型获取图像的背景和前景,对前景目标建立相应的运动模型。根据运动模型和Kalman滤波器的位置预测,建立相关的匹配代价函数、关联矩阵和匹配链表。实验结果表明,该算法对目标在场景中的频繁出现和消失、交叉运动和短暂遮挡等均有较好的处理效果。     

基于形状上下文和粒子滤波的多目标跟踪

目标跟踪是计算机视觉领域里研究的热点和难点。提出一种基于形状上下文和粒子滤波的多目标跟踪算法,通过在跟踪过程中融入目标检测信息来处理目标进入与离开场景问题和目标重叠与分离问题。首先,采用自适应增强检测算法对视频区域中的目标进行检测;然后,利用形状上下文特征来建立被跟踪目标的外观模型;最后,利用粒子滤波方法进行粒子的选择和目标的跟踪。实验证明,提出的算法能够有效处理目标进入与离开场景的问题和目标重合与分离的问题,在单一背景和复杂背景下都能进行较为准确的跟踪,还能一定程度上处理部分遮挡问题。     

基于姿态模型的人群目标分割方法

多相机间运动目标的跟踪与识别需要获得尽可能准确的目标区域。针对人群目标的粘连问题,提出一种基于姿态模型的人群目标分割方法。依据人体在运动过程中姿态的变化规律,构造7种出现频率较高的姿态模型。依次对单个目标和联合目标进行模型匹配,获得各个目标的位置、大小以及运动姿态信息。实验结果表明,该方法能有效解决相互遮挡情况下的目标分割问题。     

A color-based particle filter for multiple object tracking in an outdoor environment

Tracking multiple objects is more challenging than tracking a single object. Some problems arise in multiple-object tracking that do not exist in single-object tracking, such as object occlusion, the appearance of a new object and the disappearance of an existing object, updating the occluded object, etc. In this article, we present an approach to handling multiple-object tracking in the presence of occlusions, background clutter, and changing appearance. The occlusion is handled by considering the predicted trajectories of the objects based on a dynamic model and likelihood measures. We also propose target-model-update conditions, ensuring the proper tracking of multiple objects. The proposed method is implemented in a probabilistic framework such as a particle filter in conjunction with a color feature. The particle filter has proven very successful for nonlinear and non-Gaussian estimation problems. It approximates a posterior probability density of the state, such as the object’s position, by using samples or particles, where each state is denoted as the hypothetical state of the tracked object and its weight. The observation likelihood of the objects is modeled based on a color histogram. The sample weight is measured based on the Bhattacharya coefficient, which measures the similarity between each sample’s histogram and a specified target model. The algorithm can successfully track multiple objects in the presence of occlusion and noise. Experimental results show the effectiveness of our method in tracking multiple objects.     

Motion from occlusions

In computer vision, occlusions are almost always seen as undesirable singularities that pose difficult challenges to image motion analysis problems, such as optic flow computation, motion segmentation, disparity estimation, or egomotion estimation. However, it is well known that occlusions are extremely powerful cues for depth or motion perception, and could be used to improve those methods.

In this paper, we propose to recover camera motion information based uniquely on occlusions, by observing two specially useful properties: occlusions are independent of the camera rotation, and reveal direct information about the camera translation.

We assume a monocular observer, undergoing general rotational and translational motion in a static environment. We present a formal model for occlusion points and develop a method suitable for occlusion detection. Through the classification and analysis of the detected occlusion points, we show how to retrieve information about the camera translation (FOE). Experiments with real images are presented and discussed in the paper.     

Image-Based Modeling by Joint Segmentation

The paper first traces the image-based modeling back to feature tracking and factorization that have been developed in the group led by Kanade since the eighties. Both feature tracking and factorization have inspired and motivated many important algorithms in structure from motion, 3D reconstruction and modeling. We then revisit the recent quasi-dense approach to structure from motion. The key advantage of the quasi-dense approach is that it not only delivers the structure from motion in a robust manner for practical modeling purposes, but also it provides a cloud of sufficiently dense 3D points that allows the objects to be explicitly modeled. To structure the available 3D points and registered 2D image information, we argue that a joint segmentation of both 3D and 2D is the fundamental stage for the subsequent modeling. We finally propose a probabilistic framework for the joint segmentation. The optimal solution to such a joint segmentation is still generally intractable, but approximate solutions are developed in this paper. These methods are implemented and validated on real data set.     

Multiple-Target Tracking by Spatiotemporal Monte Carlo Markov Chain Data Association

We propose a framework for tracking multiple targets, where the input is a set of candidate regions in each frame, as obtained from a state-of-the-art background segmentation module, and the goal is to recover trajectories of targets over time. Due to occlusions by targets and static objects, as also by noisy segmentation and false alarms, one foreground region may not correspond to one target faithfully. Therefore, the one-to-one assumption used in most data association algorithms is not always satisfied. Our method overcomes the one-to-one assumption by formulating the visual tracking problem in terms of finding the best spatial and temporal association of observations, which maximizes the consistency of both motion and appearance of trajectories. To avoid enumerating all possible solutions, we take a data-driven Markov Chain Monte Carlo (DD-MCMC) approach to sample the solution space efficiently. The sampling is driven by an informed proposal scheme controlled by a joint probability model combining motion and appearance. Comparative experiments with quantitative evaluations are provided.     

粒子滤波的视频目标跟踪算法研究

目标跟踪是计算机视觉和图像处理的一个重点课题,在视频监控、机器人视觉导航以及智能交通控制中具有广泛的应用前景.通过粒子滤波技术,研究了如何整合颜色特征、前景信息和积分图运算等技术实现视频目标跟踪的粒子滤波算法.在对目标进行分割中采用了混合高斯背景建模方法;同时结合积分直方图的计算方法对颜色特征进行分段统计及相互遮挡的判断,实现基于粒子滤波的目标跟踪算法的优化,解决跟踪中诸如遮挡、光照变化、背景干扰、尺寸变化等难以解决的问题.实验结果表明提出的方法达到了预期目标.     

2D human body tracking with Structural Kalman filter

Tracking moving objects is one of the most important techniques in motion analysis and understanding, and it has many difficult problems to solve. Especially, estimating and identifying moving objects, when the background and moving objects vary dynamically, are very difficult. It is possible under such a complex environment that targets may disappear totally or partially due to occlusion by other objects. The Kalman filter has been used to estimate motion information and use the information in predicting the appearance of targets in the succeeding frames. In this paper, we propose another version of the Kalman filter, to be called Structural Kalman filter, which can successfully work its role of estimating motion information under such a deteriorating condition as occlusion. Experimental results show that the suggested approach is very effective in estimating and tracking non-rigid moving objects reliably.     

Multiple objects tracking in the presence of long-term occlusions

We present a robust object tracking algorithm that handles spatially extended and temporally long object occlusions. The proposed approach is based on the concept of “object permanence” which suggests that a totally occluded object will re-emerge near its occluder. The proposed method does not require prior training to account for differences in the shape, size, color or motion of the objects to be tracked. Instead, the method automatically and dynamically builds appropriate object representations that enable robust and effective tracking and occlusion reasoning. The proposed approach has been evaluated on several image sequences showing either complex object manipulation tasks or human activity in the context of surveillance applications. Experimental results demonstrate that the developed tracker is capable of handling several challenging situations, where the labels of objects are correctly identified and maintained over time, despite the complex interactions among the tracked objects that lead to several layers of occlusions.     

Tracking multiple moving objects by binary object forest segmentation

In an earlier study it was shown that the low level image segmentation technique known as binary object forest (BOF) analysis could be successfully used to extract one or two moving objects from complex backgrounds, even when the motion involved was very large. The method involved performing BOF analysis on each of a pair of images from a sequence and then matching the vertices of the resulting graphs. In the present study the problem of tracking multiple objects in complex backgrounds and in difficult circumstances such as partial occlusion, is considered. The approach taken is once again to perform an initial BOF analysis of each image but now to attempt matching over subgraphs of the BOF rather than simply on individual vertices. It is shown theoretically and experimentally that this results in a much more robust matching scheme. This increase in robustness not only allows multiple objects to be tracked but facilitates correct matching even when partial object occlusion occurs and when motion towards the sensor results in large (apparent) size changes between frames.     

自适应在线判别外观学习的分层关联多目标跟踪

目的 复杂场景下目标频繁且长时间的遮挡、跟踪目标外观相似引起身份转换等问题给多目标跟踪带来许多挑战。针对多目标跟踪在复杂场景中因长时间遮挡引起身份转换和轨迹分段的问题,提出一种基于自适应在线判别外观学习的分层关联多目标跟踪算法。方法 利用轨迹置信度将多目标跟踪分为局部关联和全局关联两个层次。在局部关联中,置信度高的可靠轨迹利用外观、位置-大小相似度与当前帧检测点进行关联;在全局关联中,置信度低的不可靠轨迹引入运动模型和有效关联范围进一步关联分段的轨迹。在提取目标外观特征时引入增量线性可判别分析方法以解决身份转换问题,依据新增样本与目标样本均值的外观特征差异自适应地更新目标外观模型。结果 在公开数据集2D MOT2015中的PETS09-S2L1、TUD-Stadmitte、Town-Center 3个数据集中与当前10种多目标跟踪算法进行比较,该方法对各个数据集身份转换和轨迹分段都有减少,其中在Town-Center数据集中,身份转换减少了60个,轨迹分段减少了84个,跟踪准确度提高了5.2%以上。结论 本文多目标跟踪方法,能够在复杂场景中稳定有效地实现多目标跟踪,减少轨迹分段现象,其中引入的在线线性可判别外观学习对遮挡产生的身份转换具有良好的解决效果。