Visual object tracking via online sparse instance learning

Sparse representation has been attracting much more attention in visual tracking. However most sparse representation based trackers only focus on how to model the target appearance and do not consider the learning of sparse representation when the training samples are imprecise, and hence may drift or fail in the challenging scene. In this paper, we present a novel online tracking algorithm. The tracker integrates the online multiple instance learning into the recent sparse representation scheme. For tracking, the integrated sparse representation combining texture, intensity and local spatial information is proposed to model the target. This representation takes both occlusion and appearance change into account. Then, an efficient online learning approach is proposed to select the most distinguishable features to separate the target from the background samples. In addition, the sparse representation is dynamically updated online with respect to the current context. Both qualitative and quantitative evaluations on challenging benchmark video sequences demonstrate that the proposed tracking algorithm performs favorably against several state-of-the-art methods.     

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.     

Incremental object learning and robust tracking of multiple objects from RGB-D point set data

In this paper, we propose a novel model-free approach for tracking multiple objects from RGB-D point set data. This study aims to achieve the robust tracking of arbitrary objects against dynamic interaction cases in real-time. In order to represent an object without prior knowledge, the probability density of each object is represented by Gaussian mixture models (GMM) with a tempo-spatial topological graph (TSTG). A flexible object model is incrementally updated in the pro-posed tracking framework, where each RGB-D point is identified to be involved in each object at each time step. Furthermore, the proposed method allows the creation of robust temporal associations among multiple updated objects during split, complete occlusion, partial occlusion, and multiple contacts dynamic interaction cases. The performance of the method was examined in terms of the tracking accuracy and computational efficiency by various experiments, achieving over 97% accuracy with five frames per second computation time. The limitations of the method were also empirically investigated in terms of the size of the points and the movement speed of objects.     

基于HOG及在线多实例学习的目标跟踪算法

传统的基于色彩直方图或空间色彩直方图的跟踪算法在跟踪目标出现尺度变化的复杂条件下,因无法显著区分颜色相近的目标和背景,不能得到准确跟踪结果.提出基于HOG及在线多实例学习的目标跟踪算法.此算法采用HOG特征值提取方式,结合在线多实例学习技术,对目标远离场景、平移、旋转、遮挡等情况进行跟踪.实验结果表明,该算法能够对各种复杂情况下的动态目标进行有效跟踪,具有良好的鲁棒性和准确性.     

Robust object representation by boosting-like deep learning architecture

This paper presents a new deep learning architecture for robust object representation, aiming at efficiently combining the proposed synchronized multi-stage feature (SMF) and a boosting-like algorithm. The SMF structure can capture a variety of characteristics from the inputting object based on the fusion of the handcraft features and deep learned features. With the proposed boosting-like algorithm, we can obtain more convergence stability on training multi-layer network by using the boosted samples. We show the generalization of our object representation architecture by applying it to undertake various tasks, i.e. pedestrian detection and action recognition. Our approach achieves 15.89% and 3.85% reduction in the average miss rate compared with ACF and JointDeep on the largest Caltech dataset, and acquires competitive results on the MSRAction3D dataset.     

Adaptive model updating for robust object tracking

In this paper, we exploit features extracted from convolutional neural network (CNN) to be better utilized for visual tracking. It is observed that CNN features in higher levels provide semantic information which is robust to appearance variations. Thus we integrate the hierarchical features in different layers of a deep model to correlation filter tracking framework. More specifically, correlation filters are learned on each layer to encode the object appearance. The peak-to-sidelobe ratio (PSR) is employed to measure the differences between image patches. To leverage the robustness of our model, we develop an adaptive model updating scheme to train the correlation filters according to different response maps. Extensive experimental results on three large scale benchmark datasets show that the proposed algorithm performs favorably against state-of-the-art methods.     

Multiple kernel-based multi-instance learning algorithm for image classification

In this paper, a novel multi-instance learning (MIL) algorithm based on multiple-kernels (MK) framework has been proposed for image classification. This newly developed algorithm defines each image as a bag, and the low-level visual features extracted from its segmented regions as instances. This algorithm is started from constructing a “word-space” from instances based on a collection of “visual-words” generated by affinity propagation (AP) clustering method. After calculating the distance between a “visual-word” and the bag (image), a nonlinear mapping mechanism is introduced for registering each bag as a coordinate point in the “word-space”. In this case, the MIL problem is transformed into a standard supervised learning problem, which allows multiple-kernels support vector machine (MKSVM) classifiers to be trained for the image categorization. Compared with many popular MIL algorithms, the proposed method, named as MKSVM-MIL, shows its satisfactorily experimental results on the COREL dataset, which highlights the robustness and effectiveness for image classification applications.     

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.     

Re-identification framework for long term visual object tracking based on object detection and classification

In this paper, we address the problem of long-term visual object tracking and we present an efficient real-time single object tracking system suitable for integration in autonomous platforms that need to encompass intelligent capabilities. We propose a novel long-term tracking framework for classification based re-detection and tracking, that incorporates state estimation, object re-identification and automated management of tracking and detection results. Our method integrates a novel object re-identification technique which efficiently filters a number of detection candidates and systematically corrects the tracking results. Through extensive experimental validation on the UAV123, UAV20L and TLP datasets, we demonstrate the effectiveness of the proposed system and its advantage over several state-of-the art trackers. The results furthermore highlight the proposed tracker’s ability to handle challenges arising from real-world and long-term scenarios, such as variations in pose, scale, occlusions and out-of-view situations. Furthermore, we propose a variant that is suitable for deployment on autonomous robots, such as Unmanned Aerial Vehicles.     

Effective appearance model for probabilistic object tracking

This paper presents a robust object tracking approach via a spatially constrained colour model. Local image patches of the object and spatial relation between these patches are informative and stable during object tracking. So, we propose to partition an object into patches and develop a Spatially Constrained Colour Model (SCCM) by combining the colour distributions and spatial configuration of these patches. The likelihood of the candidate object is given by estimating the confidences of the pixels in the ...     

多元假设检验GMPHD轨迹跟踪

由于在军事和民事领域逐步广泛的应用,数目不定的多目标跟踪技术正受到越来越多的关注。概率假设密度(PHD)滤波方法,特别是具有闭式递归的高斯混合概率假设密度(GMPHD)技术,在噪声和漏警等影响下仍能形成优越的群目标跟踪性能。然而PHD滤波器并不能实现多目标航迹跟踪,而其与传统数据互联的结合,复杂度高且跟踪效果不尽如人意。在该文中,各目标的航迹信息以假设形式表述,数据互联则是通过使用经典的多元假设检测方法判决假设矩阵实现。其与GMPHD的结合不仅实现了数据互联和轨迹管理,还因为积累时间信息大大降低了杂波干扰的影响。实验结果证明,该算法可以对多个目标所形成的轨迹实施正确跟踪,同时,计算量的大幅度降低带来了跟踪系统可实现性的提高。     

Local maximum edge binary patterns: A new descriptor for image retrieval and object tracking

A new algorithm meant for content based image retrieval (CBIR) and object tracking applications is presented in this paper. The local region of image is represented by local maximum edge binary patterns (LMEBP), which are evaluated by taking into consideration the magnitude of local difference between the center pixel and its neighbors. This LMEBP differs from the existing LBP in a manner that it extracts the information based on distribution of edges in an image. Further, the effectiveness of our algorithm is confirmed by combining it with Gabor transform. Four experiments have been carried out for proving the worth of our algorithm. Out of which three are meant for CBIR and one for object tracking. It is further mentioned that the database considered for first three experiments are Brodatz texture database (DB1), MIT VisTex database (DB2), rotated Brodatz database (DB3) and the fourth contains three observations. The results after being investigated show a significant improvement in terms of their evaluation measures as compared to LBP and other existing transform domain techniques.     

Network in network based weakly supervised learning for visual tracking

One of the key limitations of the many existing visual tracking method is that they are built upon low-level visual features and have limited predictability power of data semantics. To effectively fill the semantic gap of visual data in visual tracking with little supervision, we propose a tracking method which constructs a robust object appearance model via learning and transferring mid-level image representations using a deep network, i.e., Network in Network (NIN). First, we design a simple yet effective method to transfer the mid-level features learned from NIN on the source tasks with large scale training data to the tracking tasks with limited training data. Then, to address the drifting problem, we simultaneously utilize the samples collected in the initial and most previous frames. Finally, a heuristic schema is used to judge whether updating the object appearance model or not. Extensive experiments show the robustness of our method.     

Robust appearance feature learning using pixel‐wise discrimination for visual tracking

Considering the high dimensions of video sequences, it is often challenging to acquire a sufficient dataset to train the tracking models. From this perspective, we propose to revisit the idea of hand‐crafted feature learning to avoid such a requirement from a dataset. The proposed tracking approach is composed of two phases, detection and tracking, according to how severely the appearance of a target changes. The detection phase addresses severe and rapid variations by learning a new appearance model that classifies the pixels into foreground (or target) and background. We further combine the raw pixel features of the color intensity and spatial location with convolutional feature activations for robust target representation. The tracking phase tracks a target by searching for frame regions where the best pixel‐level agreement to the model learned from the detection phase is achieved. Our two‐phase approach results in efficient and accurate tracking, outperforming recent methods in various challenging cases of target appearance changes.     

Penalized Gaussian mixture probability hypothesis density filter for multiple target tracking

Bayesian multi-target filter develops a theoretical framework for estimating the full multi-target posterior which is intractable in practice. The probability hypothesis density (PHD) is a practical solution for Bayesian multi-target filter which propagates the first order moment of the multi-target posterior instead of the full version. Recently, the Gaussian Mixture PHD (GM-PHD) has been proposed as an implementation of the PHD filter which provides a close form solution. The performance of this filter degrades when targets are moving near each other such as crossing targets. In this paper, we propose a novel approach called penalized GM-PHD (PGM-PHD) filter to improve this drawback. The simulation results provided for various probabilities of detection, clutter rates, targets velocities and frame rates indicate that the proposed method achieves better performance compared to the GM-PHD filter.     

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。     

基于多示例学习和随机蕨丛检测的在线目标跟踪

基于检测的目标跟踪方法目前在计算机视觉领域受到了广泛的关注,这类方法通过训练判别分类器将目标对象从背景中分离出来;分类器的训练是根据当前的跟踪状态从当前帧中提取正负样本来进行,但训练样本的不准确将导致分类器退化产生漂移。该文提出一种能够有效克服目标漂移的跟踪算法,采用检测器和跟踪器相结合的框架,利用中值流算法作为跟踪器,提高跟踪点的可靠性;级联若干个随机蕨弱分类器构成强分类器作为检测器;用在线多示例学习方法更新检测器,提高检测精度;最后将检测器、跟踪器的结果相融合得到最终的目标位置。实验结果表明,与其它方法相比,该方法对目标漂移有更强的鲁棒性。     

基于特征融合的粒子滤波目标跟踪新方法

针对传统粒子滤波(PF)算法采用单一颜色特征建模 跟踪目标性能差的缺陷,提出一种颜色特征与纹理特 征相融合的PF目标跟踪新算法。首先,采用一种具有抗噪声和保护纹理边缘的全局中值二值 模式 (GMBP)纹理算子,对模板图像进行局部差绝对值处理,得到幅 值序列模板,将幅值序列模板内的中值作为模板的阈值,与模板邻域比较获得新的纹理图像 ;然后,与 具有光照不变特性的局部二值模式(LBP)纹理算子结合,形成一种(GMLBP)新的纹理描述算子 。最后,分别计算GMLBP纹理特征粒子权值和HSV颜色特征粒子权 值,并依据权值大小确定融合系数,对纹理特征粒子权值和颜色特征粒子权值进行线 性融合,再对融合后粒子权值进行归一化处理,从而得到目标位置状态的最终估计值。对比 实验结果表明, 相对于单一颜色特征的目标跟踪算法,所提算法捕捉目标位置准确且具有更低的平均跟踪误 差,其平均误差降低了近2倍。     

基于增强RPN的孪生网络目标跟踪算法

目前孪生网络跟踪器已经具有比较良好的表现,但是对于卷积神经网络所提取的特征仍没有较好地利用其特点,同时孪生网络通过相似性学习进行跟踪的特性使跟踪器的准确性和鲁棒性存在不足。提出了一种金字塔式特征融合的方法,根据骨干网络特征提取层不同深度具有不同侧重的特点提高网络对目标的表征能力,然后使用注意力机制对区域推荐网络（Region Proposal Network,RPN）进行增强,最终实现更精准更鲁棒的跟踪。在OTB100数据集的实验中,新提出的SiamERPN(Siamese Enhanced RPN)算法分别得到了0.668的成功率和0.876的精度,测试结果好于基线算法和其他对比算法。     

采用频闪成像技术的柔性气动外形运动物体SIFT特征跟踪(英文)

近年来,有关微型飞行器的结构动力学、材料气动弹性以及飞行控制方面的研究受到了高度的重视。为了提高飞行器的空气动力学检测能力,提出了一种基于频闪成像技术的柔性气动外形物体跟踪方法。利用尺度不变特征转换(SIFT)算法提取出的特征对于图像的尺度变换、旋转以及光照变化和局部图像变形等具有的不变特性,提出了一种利用感兴趣区域中SIFT特征对柔性气动外形运动物体进行探测与跟踪的方法。该方法在获取物体的状态、位置以及空间转换关系等方面表现出良好的性能,并且为同一场景中同一物体在不同位置之间的相互匹配提供了可靠保证。实验表明:在该实验系统中,基于频闪成像技术、利用SIFT特征作为柔性气动外形物体探测的方法具有一定的可行性。