RGBT双模态加权相关滤波跟踪算法

与基于可见光的单模态跟踪算法相比,基于可见光/热红外(RGB-Thermal, RGBT)的双模态跟踪算法对光照变化具有更强的鲁棒性。但在实际应用场景中,双模态跟踪算法仍然会受到局部遮挡目标形变的影响。为了解决以上问题,本文提出了一种可见光/热红外RGBT双模态加权相关滤波跟踪算法。该算法首先利用可见光图像和热红外图像联合求解权重图,然后利用权重图引导相关滤波器求解过程,最后根据权重图推断前景目标是否被遮挡。该算法在公开数据集RGBT234上的结果表明,本文提出的RGBT双模态加权相关滤波跟踪算法能够有效处理目标局部遮挡和目标畸变等情况,实现复杂场景下鲁棒持续的目标跟踪。      

Robust online multi-target visual tracking using a HISP filter with discriminative deep appearance learning

We propose a novel online multi-target visual tracker based on the recently developed Hypothesized and Independent Stochastic Population (HISP) filter. The HISP filter combines advantages of traditional tracking approaches like MHT and point-process-based approaches like PHD filter, and it has linear complexity while maintaining track identities. We apply this filter for tracking multiple targets in video sequences acquired under varying environmental conditions and targets density using a tracking-by-detection approach. We also adopt deep CNN appearance representation by training a verification-identification network (VerIdNet) on large-scale person re-identification data sets. We construct an augmented likelihood in a principled manner using this deep CNN appearance features and spatio-temporal information. Furthermore, we solve the problem of two or more targets having identical label considering the weight propagated with each confirmed hypothesis. Extensive experiments on MOT16 and MOT17 benchmark data sets show that our tracker significantly outperforms several state-of-the-art trackers in terms of tracking accuracy.     

基于稀疏稠密结构表示与在线鲁棒字典学习的视觉跟踪

L1跟踪对适度的遮挡具有鲁棒性,但是存在速度慢和易产生模型漂移的不足。为了解决上述两个问题,该文首先提出一种基于稀疏稠密结构的鲁棒表示模型。该模型对目标模板系数和小模板系数分别进行L2范数和L1范数正则化增强了对离群模板的鲁棒性。为了提高目标跟踪速度,基于块坐标优化原理,用岭回归和软阈值操作建立了该模型的快速算法。其次,为降低模型漂移的发生,该文提出一种在线鲁棒的字典学习算法用于模板更新。在粒子滤波框架下,用该表示模型和字典学习算法实现了鲁棒快速的跟踪方法。在多个具有挑战性的图像序列上的实验结果表明:与现有跟踪方法相比,所提跟踪方法具有较优的跟踪性能。     

Adaptive online performance evaluation of video trackers

We propose an adaptive framework to estimate the quality of video tracking algorithms without ground-truth data. The framework is divided into two main stages, namely, the estimation of the tracker condition to identify temporal segments during which a target is lost and the measurement of the quality of the estimated track when the tracker is successful. A key novelty of the proposed framework is the capability of evaluating video trackers with multiple failures and recoveries over long sequences. Successful tracking is identified by analyzing the uncertainty of the tracker, whereas track recovery from errors is determined based on the time-reversibility constraint. The proposed approach is demonstrated on a particle filter tracker over a heterogeneous data set. Experimental results show the effectiveness and robustness of the proposed framework that improves state-of-the-art approaches in the presence of tracking challenges such as occlusions, illumination changes, and clutter and on sequences containing multiple tracking errors and recoveries.     

Data fusion for ballistic targets tracking using least squares

The paper deals with the problem of state estimation of continuous-time nonlinear system using discrete-time measurements from multiple sensors. In particular, the problem of multi-radar tracking of artillery ballistic objects is considered. A batch estimator based on the iterative least squares approach is developed using simplified and accurate models of ballistic flight. The estimator is applied to process the sequences of measurements from radars tracking the same ballistic target. Estimates of the target state over time are computed and their accuracy is compared to the estimates yielded by the extended Kalman filter. Partial estimates from multiple radars are combined using track fusion approach and propagated using the 3 degree of freedom model of ballistic flight. Accuracy of target's firing point estimation is also analysed with respect to the data rates and locations of the radars with respect to the target. Practical aspects of the proposed method are also discussed.     

Dynamic representation-based tracker for long-term pedestrian tracking with occlusion

This paper presents a dynamic representation-based tracker (DRT) to handle occlusions in the long-term pedestrian tracking of a single target. In our DRT, an adaptive representation network (ARN) is first constructed to extract multiple features, including classical features such as appearance and pose as well as some vector-format deep features. These features are then stacked to form a dynamic representation so as to convert the target tracking into a matching problem between the target features and candidate features, where the Euclidean distance (ED) and locality-constrained linear coding (LLC) are used as measurements in the decision-making. Next, the target state is determined through a voting procedure according to the feature matching error. Finally, a pose supervised module (PSM) and an IOU filtering module (IFM) are applied, respectively, to refine the target state and to filter out some invalid candidate targets that have been detected. Experimental results on public benchmark datasets show that our DRT is quite robust to complex environments with long-term pedestrian occlusions, and outperforms several existing state-of-the-arts trackers as it produces the best performance on both the pedestrian tracking dataset with occlusion (PTDO) and the pedestrian tracking dataset with occlusion plus (PTDO Plus).     

Prediction based collaborative trackers (PCT): a robust and accurate approach toward 3D medical object tracking

Robust and fast 3D tracking of deformable objects, such as heart, is a challenging task because of the relatively low image contrast and speed requirement. Many existing 2D algorithms might not be directly applied on the 3D tracking problem. The 3D tracking performance is limited due to dramatically increased data size, landmarks ambiguity, signal drop-out or complex nonrigid deformation. In this paper, we present a robust, fast, and accurate 3D tracking algorithm: prediction based collaborative trackers (PCT). A novel one-step forward prediction is introduced to generate the motion prior using motion manifold learning. Collaborative trackers are introduced to achieve both temporal consistency and failure recovery. Compared with tracking by detection and 3D optical flow, PCT provides the best results. The new tracking algorithm is completely automatic and computationally efficient. It requires less than 1.5 s to process a 3D volume which contains millions of voxels. In order to demonstrate the generality of PCT, the tracker is fully tested on three large clinical datasets for three 3D heart tracking problems with two different imaging modalities: endocardium tracking of the left ventricle (67 sequences, 1134 3D volumetric echocardiography data), dense tracking in the myocardial regions between the epicardium and endocardium of the left ventricle (503 sequences, roughly 9000 3D volumetric echocardiography data), and whole heart four chambers tracking (20 sequences, 200 cardiac 3D volumetric CT data). Our datasets are much larger than most studies reported in the literature and we achieve very accurate tracking results compared with human experts' annotations and recent literature.     

Learning discriminative update adaptive spatial-temporal regularized correlation filter for RGB-T tracking

The RGB-T trackers based on correlation filter framework have been extensively investigated for that they can track targets more accurately in most complex scenes. However, the performance of these trackers is limited when facing some specific challenging scenarios, such as occlusion and background clutter. For different tracking targets, most of these trackers utilize fixed regularization constraint to build the filter model, which is obviously unreasonable to effectively present the appearance changes and characteristics of a specific target. In addition, they adopt a simple model update mechanism based on linear interpolation, which can easily lead to model degradation in challenging scenarios, resulting in tracker drift. To solve the above problems, we propose a novel adaptive spatial-temporal regularized correlation filter model to learn an appropriate regularization for achieving robust tracking and a relative peak discriminative method for model updating to avoid the model degradation. Besides, to make better integrate the unique advantages of the two modes and adapt the changing appearance of the target, an adaptive weighting ensemble scheme and a multi-scale search mechanism are adopted, respectively. To optimize the proposed model, we designed an efficient ADMM algorithm, which greatly improved the efficiency. Extensive experiments have been carried out on two available datasets, RGBT234 and RGBT210, and the experimental results indicate that the tracker proposed by us performs favorably in both accuracy and robustness against the state-of-the-art RGB-T trackers.     

Bayesian methods for multiaspect target tracking in image sequences

In this paper, we introduce new algorithms for automatic tracking of multiaspect targets in cluttered image sequences. We depart from the conventional correlation filter/Kalman filter association approach to target tracking and propose instead a nonlinear Bayesian methodology that enables direct tracking from the image sequence incorporating the statistical models for the background clutter, target motion, and target aspect change. Proposed algorithms include 1) a batch hidden Markov model (HMM) smoother and a sequential HMM filter for joint multiframe target detection and tracking and 2) two mixed-state sequential importance sampling trackers based on the sampling/importance resampling (SIR) and the auxiliary particle filtering (APF) techniques. Performance studies show that the proposed algorithms outperform the association of a bank of template correlators and a Kalman filter in adverse scenarios of low target-to-clutter ratio and uncertainty in the true target aspect.     

Dynamic swarm particle for fast motion vehicle tracking

Nowadays, the broad availability of cameras and embedded systems makes the application of computer vision very promising as a supporting technology for intelligent transportation systems, particularly in the field of vehicle tracking. Although there are several existing trackers, the limitation of using low‐cost cameras, besides the relatively low processing power in embedded systems, makes most of these trackers useless. For the tracker to work under those conditions, the video frame rate must be reduced to decrease the burden on computation. However, doing this will make the vehicle seem to move faster on the observer's side. This phenomenon is called the fast motion challenge. This paper proposes a tracker called dynamic swarm particle (DSP), which solves the challenge. The term particle refers to the particle filter, while the term swarm refers to particle swarm optimization (PSO). The fundamental concept of our method is to exploit the continuity of vehicle dynamic motions by creating dynamic models based on PSO. Based on the experiments, DSP achieves a precision of 0.896 and success rate of 0.755. These results are better than those obtained by several other benchmark trackers.     

Robust state estimator design for a solar battery charging system

This article describes a robust state estimator design for a solar battery charger where there is significant noise in the output measurements of the solar array voltage that causes degradation in the performance of the maximum power point tracking. The application of the extended Kalman filter, to the photovoltaic system, can lead to enhanced state estimation results so that a recursive solution can be obtained to achieve the most accurate estimate from the noisy signals. Additionally, as a consequence of applying the Kalman filter (KF), the immeasurable state of the inductor current can be estimated without a current sensor. The proposed controller uses the estimated solar array voltage for maximum power point tracker, and the estimated inductor current for determining the battery current controller. The methods for system modeling and design of the extended KF are presented, and the experimental results verify the validity of the proposed system.     

Real-time and robust visual tracking with scene-perceptual memory

Unmanned aerial vehicle (UAV) based aerial visual tracking is one of the research hotspots in computer vision. However, the mainstream trackers for UAV still have two shortcomings: (1) the accuracy of correlation filter tracker is greatly improved with more complex model, it impedes accuracy-speed trade-off. (2) object occlusion and camera motion in the aerial tracking scene also seriously restrict the application of aerial tracking. To address these problems, and inspired by AutoTrack tracker, we propose an aerial correlation filtering tracker based on scene-perceptual memory, Fast-AutoTrack. Firstly, to perceive and judge tracking anomalies, such as object occlusion and camera motion, inspired by the peak sidelobe ratio and AutoTrack, a confidence score is designed by perceiving and remembering the changing trend of the confidence and the local historical confidence. Secondly, after tracking anomaly occurring, several search regions are predicted based on the local object motion trend and the Spatio-temporal context information for object re-detection. Finally, to accelerate the model updating, the perceptual hashing algorithm (PHA) is used to obtain the similarity of the search regions between two adjacent frames. On typical aerial tracking datasets UAVDT, UAV123@10fps, and DTB70, Fast-AutoTrack run 71.4% faster than AutoTrack with almost equal accuracy and show favorable accuracy-speed trade-off.     

基于主分量寻踪的鲁棒视觉跟踪

传统子空间跟踪易受到模型漂移的影响而导致跟踪失败.针对此问题,本文提出一种基于主分量寻踪的鲁棒视觉跟踪方法.该方法以多个模板张成的子空间作为目标表观模型,利用主分量寻踪求解候选目标的误差分量,在粒子滤波框架下利用候选目标的误差分量估计最优状态参数.为了适应目标表观变化并克服模型漂移,本文提出一种模板更新方法.当跟踪结果与目标模板相似时,该方法利用跟踪结果更新目标模板,否则利用跟踪结果的低秩分量更新目标模板.在多个具有挑战性的图像序列上的实验结果表明:与现有跟踪方法相比,文中的跟踪方法具有较优的跟踪性能.     

Sampled-data filtering framework for cardiac motion recovery: optimal estimation of continuous dynamics from discrete measurements

Quantitative and noninvasive estimation of cardiac kinematics has significant physiological and clinical implications. In this paper, a sampled-data filtering framework is presented for the recovery of cardiac motion and deformation functions from periodic medical image sequences. Cardiac dynamics is a continuously evolving physical/physiological process, whereas the imaging data can provide only sampled measurements at discrete time instants. Given such a hybrid paradigm, stochastic multiframe filtering frameworks are constructed to couple the continuous dynamics with the discrete measurements, and to coordinately deal with the parameter uncertainty of the biomechanical constraining model and the noisy nature of the imaging data. The state estimates are predicted according to the continuous-time biomechanically constructed state equation between observation time points, and then updated with the new imaging-derived measurements at discrete time instants, yielding physically more meaningful and more accurate estimation results. Both continuous-discrete Kalman filter and sampled-data Hinfinity filter are applied for motion recovery. While Kalman filter is the optimal estimator under Gaussian noises, the Hinfinity scheme can give robust estimation results when the types and levels of model uncertainties and data disturbances are not available a priori. The strategies are validated through synthetic data experiments to illustrate their advantages and on canine MR phase contrast images and human MR tagging data to show their clinical potential.     

Visual tracking and recognition using appearance-adaptive models in particle filters

We present an approach that incorporates appearance-adaptive models in a particle filter to realize robust visual tracking and recognition algorithms. Tracking needs modeling interframe motion and appearance changes, whereas recognition needs modeling appearance changes between frames and gallery images. In conventional tracking algorithms, the appearance model is either fixed or rapidly changing, and the motion model is simply a random walk with fixed noise variance. Also, the number of particles is typically fixed. All these factors make the visual tracker unstable. To stabilize the tracker, we propose the following modifications: an observation model arising from an adaptive appearance model, an adaptive velocity motion model with adaptive noise variance, and an adaptive number of particles. The adaptive-velocity model is derived using a first-order linear predictor based on the appearance difference between the incoming observation and the previous particle configuration. Occlusion analysis is implemented using robust statistics. Experimental results on tracking visual objects in long outdoor and indoor video sequences demonstrate the effectiveness and robustness of our tracking algorithm. We then perform simultaneous tracking and recognition by embedding them in a particle filter. For recognition purposes, we model the appearance changes between frames and gallery images by constructing the intra- and extrapersonal spaces. Accurate recognition is achieved when confronted by pose and view variations.     

Occlusion-robust online multi-object visual tracking using a GM-PHD filter with CNN-based re-identification

We propose a novel online multi-object visual tracker using a Gaussian mixture Probability Hypothesis Density (GM-PHD) filter and deep appearance learning. The GM-PHD filter has a linear complexity with the number of objects and observations while estimating the states and cardinality of time-varying number of objects, however, it is susceptible to miss-detections and does not include the identity of objects. We use visual-spatio-temporal information obtained from object bounding boxes and deeply learned appearance representations to perform estimates-to-tracks data association for target labeling as well as formulate an augmented likelihood and then integrate into the update step of the GM-PHD filter. We also employ additional unassigned tracks prediction after the data association step to overcome the susceptibility of the GM-PHD filter towards miss-detections caused by occlusion. Extensive evaluations on MOT16, MOT17 and HiEve benchmark data sets show that our tracker significantly outperforms several state-of-the-art trackers in terms of tracking accuracy and identification.     

基于相关滤波的目标重检测跟踪

近年来，相关滤波方法由于具备运算速度快，鲁棒性强的优势，在目标跟踪领域发展迅速。然而，面对复杂场景时，现有模型难以满足实际需求。针对背景感知相关滤波方法(BACF)在目标发生自身旋转、尺度变换、运动出视野等挑战下，相关滤波器最大响应值减弱，造成跟踪精度下降的问题，提出了一种基于相关滤波的目标重检测跟踪方法。在原有背景感知相关滤波方法的基础上，引入滤波器响应检测机制，当判定到相关滤波跟踪结果不可信时，利用粒子滤波采样策略生成大量粒子，感知目标状态，重新确定目标中心位置。在此基础上，利用自适应尺度估计机制重新计算目标尺度信息，从而实现对目标的重新跟踪。为了验证改进算法的有效性，实验选取了OTB2013、OTB2015、VOT2016共3个公开数据集进行测试，同时与相关滤波及深度学习方法进行对比，从视频属性、跟踪精确度、算法鲁棒性等角度展示所有算法的性能。实验结果表明:基于相关滤波的目标重检测跟踪方法在3个公开数据集中取得较好的实验结果，并在目标发生旋转，尺度变换及运动超出视野的情况下，有效提高了BACF的准确率和成功率。     

L2范数正则化鲁棒编码视觉跟踪

针对基于稀疏表示的视觉跟踪计算效率低和易于产生模型漂移的不足,该文提出一种基于L2范数正则化鲁棒编码的视觉跟踪方法。该方法利用L2范数正则化鲁棒编码求解候选目标的编码系数,以粒子滤波为框架,利用候选目标的加权重建误差建立似然模型跟踪目标。为了适应目标的变化并克服模型漂移问题,利用L2范数正则化鲁棒编码估计当前目标的加权矩阵用于遮挡检测,根据遮挡检测结果实现模型更新。对提出的跟踪方法进行实验的结果表明:与现有跟踪方法相比,该方法具有较优的跟踪性能。