Spatially constrained level-set tracking and segmentation of non-rigid objects

Level-set is a widely used technique in segmentation-based tracking due to its flexibility in handling 2D topological changes and computational efficiency. Most existing level-set models aim at grouping pixels that have similar features into a region, without consideration of the spatial relationship of these pixels. In this paper, we present a novel level-set tracking method that incorporates spatial information to improve the robustness and accuracy of tracking non-rigid objects. Both tracking and segmentation are performed in a unified probabilistic framework, with additional spatial constraints from a part-based model—the Hough Forests. In the stage of tracking, the rigid motion of the target object is estimated by rigid registration in both the color space and the Hough voting space. Then in the stage of segmentation, some support points are obtained from back-projection, and guide the level-set evolution to capture the shape deformation. We conduct quantitative evaluation on two recently proposed public benchmarks: a non-rigid object tracking dataset and the CVPR2013 online tracking benchmark, involving 61 sequences in total. The experimental results demonstrate that our tracking method performs comparably to the state-of-the-arts in the CVPR2013 benchmark, while shows significantly improved performance in tracking non-rigid objects.     

一种融合遮挡分割的多目标跟踪算法

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。提出了一种融合遮挡分割的多目标跟踪算法,计算每个目标的光流速度概率直方图,反映其运动统计信息;综合使用外观、运动、颜色信息构造新的像素距离表达,借助分阶段分类思想及K均值聚类技术进行遮挡分割,得到准确的运动前景像素;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,更好地处理动态遮挡问题。实验表明,所提算法解决了复杂环境下的多目标跟踪问题。     

目标遮挡情况下的压缩跟踪算法

针对经典压缩跟踪算法在目标被遮挡时容易导致目标丢失的问题,提出了一种基于目标遮挡情况下的压缩跟踪算法.该方法首先依据分类器的最大响应值判断目标是否被遮挡.若发生遮挡则利用基于颜色直方图特征的粒子滤波算法进行跟踪预测,即将遮挡前提取的目标颜色直方图与粒子的颜色直方图进行相似性比较.为确保目标再现时能及时准确地捕捉其位置,再利用Harris角点特征进一步验证,并将预测的位置作为目标位置继续压缩跟踪.仿真结果表明,该算法能够准确地判断遮挡的发生,平均跟踪成功率较经典的压缩跟踪算法提高了24%,有效提高了跟踪的鲁棒性.     

APPOS: An adaptive partial occlusion segmentation method for multiple vehicles tracking

In traffic surveillance videos, it is common that the vehicles are occluded partially by each other. Such kind of occlusion situation is a challengeable task in multiple vehicles tracking. Various solutions in dealing with the occlusion for vehicles tracking have been proposed in many literatures. However, most of them are specialized on one tracking method and cannot flexibly adapt to the others. In this paper, we propose an adaptive partial occlusion segmentation method (APPOS) for multiple vehicles tracking. In this method, the occlusion detection process is firstly conducted to discover the occlusion. After that, the candidate regions of the respective occluded vehicles are roughly evaluated by the contour’s optical flow. Finally, the line scanning which uses color contrast among regions is adopted to accurately locate the vehicles. We evaluate the effectiveness and accuracy of APPOS by the experiments on practical and simulating videos.     

2-D mesh-based video object segmentation and tracking with occlusion resolution

This paper integrates fully automatic video object segmentation and tracking including detection and assignment of uncovered regions in a 2-D mesh-based framework. Particular contributions of this work are (i) a novel video object segmentation method that is posed as a constrained maximum contrast path search problem along the edges of a 2-D triangular mesh, and (ii) a 2-D mesh-based uncovered region detection method along the object boundary as well as within the object. At the first frame, an optimal number of feature points are selected as nodes of a 2-D content-based mesh. These points are classified as moving (foreground) and stationary nodes based on multi-frame node motion analysis, yielding a coarse estimate of the foreground object boundary. Color differences across triangles near the coarse boundary are employed for a maximum contrast path search along the edges of the 2-D mesh to refine the boundary of the video object. Next, we propagate the refined boundary to the subsequent frame by using motion vectors of the node points to form the coarse boundary at the next frame. We detect occluded regions by using motion-compensated frame differences and range filtered edge maps. The boundaries of detected uncovered regions are then refined by using the search procedure. These regions are either appended to the foreground object or tracked as new objects. The segmentation procedure is re-initialized when unreliable motion vectors exceed a certain number. The proposed scheme is demonstrated on several video sequences.     

基于特征融合和注意力机制的目标分割跟踪算法

在当前的目标跟踪领域,现有的基于分割的算法没有充分利用目标的长距离依赖信息和各个特征层的不同特性,前背景判别能力不强,对目标的多尺度估计不足。针对此问题,提出了自适应特征融合模块和混合域注意力模块,以提高网络对目标的多尺度估计能力和对目标的前背景辨别能力,并将其集成到当前基于视频分割的算法中,提出了一种新的目标跟踪算法,在各大公开数据集上的实验结果证明其达到了领先水平。     

基于改进的混合高斯模型的运动目标跟踪研究

对传统混合高斯背景模型作了改进,消除了缓慢运动目标对背景模型的影响,其中提出了目标间差分方法区分出前后帧变化区,对不同区域采用不同的学习权重更新策略。通过实验证明,该改进算法提高了背景模型的健壮性,在跟踪系统中获得较好效果。     

Segmentation and tracking of moving objects for content-based videocoding

To enable content-based functionalities in video coding, a decomposition of the scene into physical objects is required. Such objects are normally not characterised by homogeneous colour, intensity, or optical flow. Therefore, conventional techniques based on these low-level features cannot perform the desired segmentation. The authors address segmentation and tracking of moving objects and present a new video object plane (VOP) segmentation algorithm that extracts semantically meaningful objects. A morphological motion filter detects physical objects by identifying areas that are moving differently from the background. A new filter criterion is introduced that measures the deviation of the estimated local motion from the synthesised global motion. A two-dimensional binary model is derived for the object of interest and tracked throughout the sequence by a Hausdorff object tracker. To accommodate for rotations and changes in shape, the model is updated every frame by a two-stage method that accounts for rigid and non-rigid moving parts of the object. The binary model then guides the actual VOP extraction, whereby a novel boundary post-processor ensures high boundary accuracy. Experimental results demonstrate the performance of the proposed algorithm     

Automatic video segmentation and tracking for content-based applications

Advanced multimedia applications have to provide content-related functionalities such as search and retrieval of meaningful objects, detection and analysis of events, and understanding of scenes, which allow the user to access and manipulate the multimedia content with greater flexibility. This greatly depends on automatic techniques for extracting such objects from multimedia data. In this article we intend to provide a tutorial on the state-of-the-art in video segmentation and tracking technology with particular attention paid to the recent developments in attention-based object extraction. Performance results are included to highlight this emerging technology     

Performance measures for video object segmentation and tracking

We propose measures to evaluate quantitatively the performance of video object segmentation and tracking methods without ground-truth (GT) segmentation maps. The proposed measures are based on spatial differences of color and motion along the boundary of the estimated video object plane and temporal differences between the color histogram of the current object plane and its predecessors. They can be used to localize (spatially and/or temporally) regions where segmentation results are good or bad; and/or they can be combined to yield a single numerical measure to indicate the goodness of the boundary segmentation and tracking results over a sequence. The validity of the proposed performance measures without GT have been demonstrated by canonical correlation analysis with another set of measures with GT on a set of sequences (where GT information is available). Experimental results are presented to evaluate the segmentation maps obtained from various sequences using different segmentation approaches.     

Heuristic algorithm for visual tracking of deformable objects

Many vision problems require fast and accurate tracking of objects in dynamic scenes. These problems can be formulated as exploration problems and thus can be expressed as a search into a state space based approach. However, these problems are hard to solve because they involve search through a space of transformations corresponding to all the possible motion and deformation. In this paper, we propose a heuristic algorithm through the space of transformations for computing target 2D motion. Three features are combined in order to compute efficient motion: (1) a quality of function match based on a holistic similarity measurement, (2) Kullback–Leibler measure as heuristic to guide the search process and (3) incorporation of target dynamics into the search process for computing the most promising search alternatives. Once 2D motion has been calculated, the result value of the quality of function match computed is used with the purpose of verifying template updates. A template will be updated only when the target object has evolved to a transformed shape dissimilar with respect to the actual shape. Also, a short-term memory subsystem is included with the purpose of recovering previous views of the target object. The paper includes experimental evaluations with video streams that illustrate the efficiency and suitability for real-time vision based tasks in unrestricted environments.     

基于三维分层图模型的复杂情况下多运动目标跟踪

复杂情况下的多运动目标跟踪是视频监控中的关 键问题,在多目标运动中相互遮挡时二维视觉信息容易丢失而造 成无法准确跟踪识别目标。本文采用kinect摄像机获取三维视觉信息,从节点、边、空间结 构三个层次上建立目标的三维分 层图模型表征目标的三维特征,并在每层进行通过视频帧间的匹配从而获得三维分层图模型 匹配结果,并根据匹配结果先初 步分析目标跟踪情况,如发生遮挡则通过遮挡区域聚类块与三维分层图模型中各特征匹配确 定其匹配结果,从而得到多运动 目标在复杂运动情况中的跟踪结果。实验表明,在实验室kinect拍摄的视频序列上当目标出 现遮挡等复杂情况,也能取得较 好的跟踪结果,在实验视频中比经典方法的跟踪总体性能指标改善约3%,说明本方法能较好 地实现复杂情况下的多运动目标跟踪。     

一种简单易行的运动对象分割方法

在兼顾运动图像分割效果和实时性的原则上,提出了一种简单有效的分割方法。首先利用二次帧差的方法求出图像帧的运动区域,去除结果中不准确的小区域,最后利用二次扫描的方法填充该区域。实验证明：利用该方法得到了较好的分割效果并节省了处理时间。     

复杂地面场景下的红外运动目标跟踪

复杂地面场景下的红外目标易受背景影响并经常出现遮挡情况,难以简单地依靠亮度或梯度信息检测并跟踪目标。根据复杂背景下红外运动目标与背景的速度场差异,提出了利用光流对目标进行跟踪的算法。首先对图像进行配准,保证在随动跟踪时背景的相对静止;然后在目标的跟踪波门内计算改进的Horn-Schunck 光流;最后根据目标的光流特征,优化粒子滤波算法中粒子的转移概率,实现对目标的稳健跟踪。实验结果表明,该跟踪算法能对复杂地面场景下的红外运动目标持续跟踪,并不受目标被短时遮挡的影响。     

Computational sensor for visual tracking with attention

This paper presents a VLSI embodiment of an optical tracking computational sensor which focuses attention on a salient target in its field of view. Using both low-latency massive parallel processing and top-down sensory adaptation, the sensor suppresses interference front features irrelevant for the task at hand, and tracks a target of interest at speeds of up to 7000 pixels/s. The sensor locks onto the target to continuously provide control for the execution of a perceptually guided activity. The sensor prototype, a 24×24 array of cells, is built in 2-μm CMOS technology. Each cell occupies 62 μm×62 μm of silicon, and contains a photodetector and processing electronics     

Learning multiple instance deep representation for objects tracking

Object tracking has been widely used in various intelligent systems, such as pedestrian tracking, autonomous vehicles. To solve the problem that appearance changes and occlusion may lead to poor tracking performance, we propose a multiple instance learning (MIL) based method for object tracking. To achieve this task, we first manually label the first several frames of video stream in image level, which can indicate that whether a target object in the video stream. Then, we leverage a pre-trained convolutional neural network that has rich prior information to extract deep representation of target object. Since the location of the same object in adjacent frames is similar, we introduce a particle filter to predict the location of target object within a specific region. Comprehensive experiments have shown the effectiveness of our proposed method.     

Real-time vessel segmentation and tracking for ultrasound imaging applications

A method for vessel segmentation and tracking in ultrasound images using Kalman filters is presented. A modified Star-Kalman algorithm is used to determine vessel contours and ellipse parameters using an extended Kalman filter with an elliptical model. The parameters can be used to easily calculate the transverse vessel area which is of clinical use. A temporal Kalman filter is used for tracking the vessel center over several frames, using location measurements from a handheld sensorized ultrasound probe. The segmentation and tracking have been implemented in real-time and validated using simulated ultrasound data with known features and real data, for which expert segmentation was performed. Results indicate that mean errors between segmented contours and expert tracings are on the order of 1%-2% of the maximum feature dimension, and that the transverse cross-sectional vessel area as computed from estimated ellipse parameters a, b as determined by our algorithm is within 10% of that determined by experts. The location of the vessel center was tracked accurately for a range of speeds from 1.4 to 11.2 mm/s.     

Improved tracking of limb occlusion pressure for surgicaltourniquets

Constant-pressure tourniquets are widely used to occlude blood flow into a patient's limb to facilitate the performance of a wide variety of surgical procedures. Adaptive tourniquets that automatically adjust the cuff pressure to the minimum necessary for occlusion (limb occlusion pressure) as a function of the patient's changing systolic blood pressure are expected to reduce the incidence of tourniquet-related injuries. However, these devices have not been widely used, largely due to problems in tracking the systolic blood pressure safely, accurately, and reliably in clinical environments with noise present. Initial lab trials and clinical trials compared the performance in tracking limb occlusion pressure during varying noise conditions of a typical oscillometric blood pressure monitor with that of a prototype system. The prototype system functions by detecting noise and rapidly estimating limb occlusion pressure using only data uncorrupted by noise. Results showed that the prototype consistently estimated limb occlusion pressure more rapidly, more accurately, and more reliably than the oscillometric monitor in noisy conditions typical of surgical procedures. The results also indicate that the prototype is feasible for incorporation into an adaptive tourniquet     

运动目标的自动分割与跟踪

该文提出了一种对视频序列中的运动目标进行自动分割的算法。该算法分析图像在L U V空间中的局部变化,同时使用运动信息来把目标从背景中分离出来。首先根据图像的局部变化,使用基于图论的方法把图像分割成不同的区域。然后,通过度量合成的全局运动与估计的局部运动之间的偏差来检测出运动的区域,运动的区域通过基于区域的仿射运动模型来跟踪到下一帧。为了提高提取的目标的时空连续性,使用Hausdorff跟踪器对目标的二值模型进行跟踪。对一些典型的MPEG-4测试序列所进行的评估显示了该算法的优良性能。