基于多任务最小软阈值回归方法的目标跟踪*

在视频跟踪中,模型表示是直接影响跟踪效率的核心问题之一.在随时间和空间变化的复杂数据中学习目标外观模型表示所需的有效模板,从而适应内在或外在因素所引起的目标状态变化是非常重要的.文中详细描述较为鲁棒的目标外观模型表示策略,并提出一种新的多任务最小软阈值回归跟踪算法(MLST).该算法框架将候选目标的观测模型假设为多任务线性回归问题,利用目标模板和独立同分布的高斯-拉普拉斯重构误差线性表示候选目标不同状态下的外观模型,从而跟踪器能够很好地适应各种复杂场景并准确预测每一时刻的真实目标状态.大量实验证明,文中在线学习策略能够充分挖掘目标在不同时刻的特殊状态信息以提高模型表示精度,使得跟踪器保持最佳的状态,从而在一定程度上提高跟踪性能.实验结果显示,本文算法体现较好的鲁棒性并优于一些目前较先进的跟踪算法.     

Online multi-view subspace learning via group structure analysis for visual object tracking

In this paper, we focus on incrementally learning a robust multi-view subspace representation for visual object tracking. During the tracking process, due to the dynamic background variation and target appearance changing, it is challenging to learn an informative feature representation of tracking object, distinguished from the dynamic background. To this end, we propose a novel online multi-view subspace learning algorithm (OMEL) via group structure analysis, which consistently learns a low-dimensional representation shared across views with time changing. In particular, both group sparsity and group interval constraints are incorporated to preserve the group structure in the low-dimensional subspace, and our subspace learning model will be incrementally updated to prevent repetitive computation of previous data. We extensively evaluate our proposed OMEL on multiple benchmark video tracking sequences, by comparing with six related tracking algorithms. Experimental results show that OMEL is robust and effective to learn dynamic subspace representation for online object tracking problems. Moreover, several evaluation tests are additionally conducted to validate the efficacy of group structure assumption.     

Incremental Tensor Subspace Learning and Its Applications to Foreground Segmentation and Tracking

Appearance modeling is very important for background modeling and object tracking. Subspace learning-based algorithms have been used to model the appearances of objects or scenes. Current vector subspace-based algorithms cannot effectively represent spatial correlations between pixel values. Current tensor subspace-based algorithms construct an offline representation of image ensembles, and current online tensor subspace learning algorithms cannot be applied to background modeling and object tracking. In this paper, we propose an online tensor subspace learning algorithm which models appearance changes by incrementally learning a tensor subspace representation through adaptively updating the sample mean and an eigenbasis for each unfolding matrix of the tensor. The proposed incremental tensor subspace learning algorithm is applied to foreground segmentation and object tracking for grayscale and color image sequences. The new background models capture the intrinsic spatiotemporal characteristics of scenes. The new tracking algorithm captures the appearance characteristics of an object during tracking and uses a particle filter to estimate the optimal object state. Experimental evaluations against state-of-the-art algorithms demonstrate the promise and effectiveness of the proposed incremental tensor subspace learning algorithm, and its applications to foreground segmentation and object tracking.     

Learned Models for Estimation of Rigid and Articulated Human Motion from Stationary or Moving Camera

We propose an approach for modeling, measurement and tracking of rigid and articulated motion as viewed from a stationary or moving camera. We first propose an approach for learning temporal-flow models from exemplar image sequences. The temporal-flow models are represented as a set of orthogonal temporal-flow bases that are learned using principal component analysis of instantaneous flow measurements. Spatial constraints on the temporal-flow are then incorporated to model the movement of regions of rigid or articulated objects. These spatio-temporal flow models are subsequently used as the basis for simultaneous measurement and tracking of brightness motion in image sequences. Then we address the problem of estimating composite independent object and camera image motions. We employ the spatio-temporal flow models learned through observing typical movements of the object from a stationary camera to decompose image motion into independent object and camera motions. The performance of the algorithms is demonstrated on several long image sequences of rigid and articulated bodies in motion.     

Robust feature learning for online discriminative tracking without large-scale pre-training

Owing to the inherent lack of training data in visual tracking, recent work in deep learning-based trackers has focused on learning a generic representation offline from large-scale training data and transferring the pre-trained feature representation to a tracking task. Offline pre-training is time-consuming, and the learned generic representation may be either less discriminative for tracking specific objects or overfitted to typical tracking datasets. In this paper, we propose an online discriminative tracking method based on robust feature learning without large-scale pre-training. Specifically, we first design a PCA filter bank-based convolutional neural network (CNN) architecture to learn robust features online with a few positive and negative samples in the high-dimensional feature space. Then, we use a simple soft-thresholding method to produce sparse features that are more robust to target appearance variations. Moreover, we increase the reliability of our tracker using edge information generated from edge box proposals during the process of visual tracking. Finally, effective visual tracking results are achieved by systematically combining the tracking information and edge box-based scores in a particle filtering framework. Extensive results on the widely used online tracking benchmark (OTB-50) with 50 videos validate the robustness and effectiveness of the proposed tracker without large-scale pre-training.     

结合Boosting方法与SVM的多核学习跟踪算法

针对传统算法在外界环境及目标运动导致外形变化的影响下跟踪效果不稳定的问题,提出一种鲁棒的多核学习跟踪算法,将Boosting提升方法引入到多核学习框架中,用比传统多核学习算法更少的样本训练,构建出基于互补性特征集和核函数集的弱分类器池,从中将多个单核的弱分类器组合出一个多核的强分类器,从而在出现较强背景干扰、目标被遮挡的情况下仍能正确地对候选图块中的背景和目标进行分类。对不同视频序列的测试结果表明,与同样采用Boosting方法的OAB算法及近年跟踪精度高的LOT算法相比,该算法能够在复杂环境下更准确地跟踪到目标。     

Kernel-based object tracking using asymmetric kernels with adaptive scale and orientation selection

Kernel-based object tracking refers to computing the translation of an isotropic object kernel from one video frame to the next. The kernel is commonly chosen as a primitive geometric shape and its translation is computed by maximizing the likelihood between the current and past object observations. In the case when the object does not have an isotropic shape, kernel includes non-object regions which biases the motion estimation and results in loss of the tracked object. In this paper, we propose to use an asymmetric object kernel for improving the tracking performance. An important advantage of an asymmetric kernel over an isotropic kernel is its precise representation of the object shape. This property enhances tracking performance due to discarding the non-object regions. The second contribution of our paper is the introduction of a new adaptive kernel scale and orientation selection method which is currently achieved by greedy algorithms. In our approach, the scale and orientation are introduced as additional dimensions to the spatial image coordinates, in which the mode seeking, hence tracking, is achieved simultaneously in all coordinates. Demonstrated in a set of experiments, the proposed method has better tracking performance with comparable execution time then kernel tracking methods used in practice.     

A kernel learning framework for domain adaptation learning

Domain adaptation learning(DAL) methods have shown promising results by utilizing labeled samples from the source(or auxiliary) domain(s) to learn a robust classifier for the target domain which has a few or even no labeled samples.However,there exist several key issues which need to be addressed in the state-of-theart DAL methods such as sufficient and effective distribution discrepancy metric learning,effective kernel space learning,and multiple source domains transfer learning,etc.Aiming at the mentioned-above issues,in this paper,we propose a unified kernel learning framework for domain adaptation learning and its effective extension based on multiple kernel learning(MKL) schema,regularized by the proposed new minimum distribution distance metric criterion which minimizes both the distribution mean discrepancy and the distribution scatter discrepancy between source and target domains,into which many existing kernel methods(like support vector machine(SVM),v-SVM,and least-square SVM) can be readily incorporated.Our framework,referred to as kernel learning for domain adaptation learning(KLDAL),simultaneously learns an optimal kernel space and a robust classifier by minimizing both the structural risk functional and the distribution discrepancy between different domains.Moreover,we extend the framework KLDAL to multiple kernel learning framework referred to as MKLDAL.Under the KLDAL or MKLDAL framework,we also propose three effective formulations called KLDAL-SVM or MKLDAL-SVM with respect to SVM and its variant μ-KLDALSVM or μ-MKLDALSVM with respect to v-SVM,and KLDAL-LSSVM or MKLDAL-LSSVM with respect to the least-square SVM,respectively.Comprehensive experiments on real-world data sets verify the outperformed or comparable effectiveness of the proposed frameworks.     

张量核范数回归的目标跟踪

目的 视觉目标跟踪中,不同时刻的目标状态是利用在线学习的模板数据线性组合近似表示。由于跟踪中目标受到自身或场景中各种复杂干扰因素的影响,跟踪器的建模能力很大程度地依赖模板数据的概括性及其误差的估计精度。很多现有算法以向量形式表示样本信号,而改变其原始数据结构,使得样本数据各元素之间原有的自然关系受到严重破坏;此外,这种数据表述机制会提高数据的维度,而带来一定的计算复杂度和资源浪费。本文以多线性分析的角度更进一步深入研究视频跟踪中的数据表示及其建模机制,为其提供更加紧凑有效的解决方法。方法 本文跟踪框架中,候选样本及其重构信号以张量形式表示,从而保证其数据的原始结构。跟踪器输出候选样本外观状态时,以张量良好的多线性特性来组织跟踪系统的建模任务,利用张量核范数及L₁范数正则化其目标函数的相关成分,在多任务状态学习假设下充分挖掘各候选样本外观表示任务的独立性及相互依赖关系。结果 用结构化张量表示的数据原型及其多任务观测模型能够较为有效地解决跟踪系统的数据表示及计算复杂度难题。同时,为候选样本外观模型的多任务联合学习提供更加简便有效的解决途径。这样,当跟踪器遇到破坏性较强的噪声干扰时,其张量核范数约束的误差估计机制在多任务联合学习框架下更加充分挖掘目标全面信息,使其更好地适应内在或外在因素所引起的视觉信息变化。在一些公认测试视频上的实验结果表明,本文算法在候选样本外观模型表示方面表现出更为鲁棒的性能。因而和一些优秀的同类算法相比,本文算法在各测试序列中跟踪到的目标图像块平均中心位置误差和平均重叠率分别达到4.2和0.82,体现出更好的跟踪精度。结论 大量实验验证本文算法的张量核范数回归模型及其误差估计机制能够构造出目标每一时刻状态更接近的最佳样本信号,在多任务学习框架下严格探测每一个候选样本的真实状态信息,从而较好地解决模型退化和跟踪漂移问题。     

Accurate visual representation learning for single object tracking

As a fundamental visual task, single object tracking has witnessed astonishing improvements. However, there still existing many factors should be to addressed for accurately tracking performance. Among them, visual representation is one of important influencers suffer from complex appearance changes. In this work, we propose a rich appearance representation learning strategy for tracking. First, by embedding the saliency feature extractor module, we try to improve the visual representation ability by fusing the saliency information learning from different convolution lays. With leveraging lightweight Convolutional Neural Network VGG-M as the features extractor backbone, we can attain robust appearance model by deep features with fruitful semantic information. Second, as for the classifier has significant complementary guidance for location prediction, we propose to generate diverse feature instances of the target by introducing the adversarial learning strategy. Given the generated diverse instances, many complex situations in the tracking process can be effectively simulated, especially the occlusion that conformed to the long tail distribution. Third, to optimize the bounding boxes refinement, we employ a precise pooling strategy for attaining feature maps with high resolution. Then, our approach can capture the subtle appearance changes effectively over a long time range. Finally, extensive experiments was conducted on several benchmark datasets, the results demonstrate that the proposed approach performs favorably against many state-of-the-art algorithms.

     

Shape based appearance model for kernel tracking

This paper investigates kernel based tracking using shape information. A kernel based tracker typically models an object with a primitive geometric shape, and then estimates the object state by fitting the kernel such that the appearance model is optimized. Most of the appearance models in kernel based tracking utilize the textural information within the kernel, although a few of them also make use of the gradient information along the kernel boundary. Interestingly, shape information of a general form has never been fully exploited in kernel tracking, despite the fact that shape has been widely used in silhouette tracking at the cost of intensive computation. In this paper, we propose an original way to incorporate shape knowledge into the appearance model of kernel based trackers while preserving their computational advantage versus silhouette based trackers. Experimental results demonstrate that kernel tracking is strongly improved by exploiting the proposed shape cue through comparisons to both kernel and silhouette trackers.     

在线复合模板模型表示的视觉目标跟踪

目的 视觉目标跟踪中,目标往往受到自身或场景中各种复杂干扰因素的影响,这对正确捕捉所感兴趣的目标信息带来极大的挑战。特别是,跟踪器所用的模板数据主要是在线学习获得,数据的可靠性直接影响到候选样本外观模型表示的精度。针对视觉目标跟踪中目标模板学习和候选样本外观模型表示等问题,采用一种较为有效的模板组织策略以及更为精确的模型表示技术,提出一种新颖的视觉目标跟踪算法。方法 跟踪框架中,将候选样本外观模型表示假设为由一组复合模板和最小重构误差组成的线性回归问题,首先利用经典的增量主成分分析法从在线高维数据中学习出一组低维子空间基向量(模板正样本),并根据前一时刻跟踪结果在线实时采样一些特殊的负样本加以扩充目标模板数据,再利用新组织的模板基向量和独立同分布的高斯—拉普拉斯混合噪声来线性拟合候选目标外观模型,最后估计出候选样本和真实目标之间的最大似然度,从而使跟踪器能够准确捕捉每一时刻的真实目标状态信息。结果 在一些公认测试视频序列上的实验结果表明,本文算法在目标模板学习和候选样本外观模型表示等方面比同类方法更能准确有效地反映出视频场景中目标状态的各种复杂变化,能够较好地解决各种不确定干扰因素下的模型退化和跟踪漂移问题,和一些优秀的同类算法相比,可以达到相同甚至更高的跟踪精度。结论 本文算法能够在线学习较为精准的目标模板并定期更新,使得跟踪器良好地适应内在或外在因素(姿态、光照、遮挡、尺度、背景扰乱及运动模糊等)所引起的视觉信息变化,始终保持其最佳的状态,使得候选样本外观模型的表示更加可靠准确,从而展现出更为鲁棒的性能。     

Adaptive multi-cue tracking by online appearance learning

This paper proposes a multi-cue based appearance learning algorithm for object tracking. In each frame, the target object is represented by different cues in the image-as-matrix form. This representation can describe the target from different perspectives and can preserve the spatial correlation information inside the target region. Based on these cues, multiple appearance models are learned online by bilinear subspace analysis to account for the target appearance variations over time. Tracking is formulated within the Bayesian inference framework, in which the observation model is constructed by fusing all the learned appearance models. The combination of online appearance modeling and weight update of each appearance model can adapt our tracking algorithm to both the target and background changes. We test our algorithm on a variety of challenging sequences by tracking car, face, pedestrian, and so on. Experimental results and comparisons to several state-of-the-art methods show improved tracking performance.     

A motion observable representation using color correlogram and its applications to tracking

This paper presents a special form of color correlogram as representation for object tracking and carries out a motion observability analysis to obtain the optimal correlogram in a kernel based tracking framework. Compared with the color histogram, where the position information of each pixel is ignored, a simplified color correlogram (SCC) representation encodes the spatial information explicitly and enables an estimation algorithm to recover the object orientation. In this paper, based on the SCC representation, the mean shift algorithm is developed in a translation–rotation joint domain to track the positions and orientations of objects. The ability of the SCC in detecting and estimating object motion is analyzed and a principled way to obtain the optimal SCC as object representation is proposed to ensure reliable tracking. Extensive experimental results demonstrate SCC as a viable object representation for tracking.     

Robust online appearance models for visual tracking

We propose a framework for learning robust, adaptive, appearance models to be used for motion-based tracking of natural objects. The model adapts to slowly changing appearance, and it maintains a natural measure of the stability of the observed image structure during tracking. By identifying stable properties of appearance, we can weight them more heavily for motion estimation, while less stable properties can be proportionately downweighted. The appearance model involves a mixture of stable image structure, learned over long time courses, along with two-frame motion information and an outlier process. An online EM-algorithm is used to adapt the appearance model parameters over time. An implementation of this approach is developed for an appearance model based on the filter responses from a steerable pyramid. This model is used in a motion-based tracking algorithm to provide robustness in the face of image outliers, such as those caused by occlusions, while adapting to natural changes in appearance such as those due to facial expressions or variations in 3D pose.     

Robust object tracking with RGBD-based sparse learning

Robust object tracking has been an important and challenging research area in the field of computer vision for decades. With the increasing popularity of affordable depth sensors, range data is widely used in visual tracking for its ability to provide robustness to varying illumination and occlusions. In this paper, a novel RGBD and sparse learning based tracker is proposed. The range data is integrated into the sparse learning framework in three respects. First, an extra depth view is added to the color image based visual features as an independent view for robust appearance modeling. Then, a special occlusion template set is designed to replenish the existing dictionary for handling various occlusion conditions. Finally, a depth-based occlusion detection method is proposed to efficiently determine an accurate time for the template update. Extensive experiments on both KITTI and Princeton data sets demonstrate that the proposed tracker outperforms the state-of-the-art tracking algorithms, including both sparse learning and RGBD based methods.     

高斯核函数卷积神经网络跟踪算法

针对深度学习跟踪算法训练样本缺少、训练费时、算法复杂度高等问题,引入高斯核函数进行加速,提出一种无需训练的简化卷积神经网络跟踪算法。首先,对初始帧目标进行归一化处理并聚类提取一系列初始滤波器组,跟踪过程中结合目标背景信息与前景候选目标进行卷积;然后,提取目标简单抽象特征;最后,将简单层的卷积结果进行叠加得到目标的深层次特征表达。通过高斯核函数加速来提高算法中全部卷积运算的速度,利用目标的局部结构特征信息,对网络各阶段滤波器进行更新,结合粒子滤波跟踪框架实现跟踪。在CVPR2013跟踪数据集上的实验表明,本文方法脱离了繁琐深度学习运行环境,能克服低分辨率下目标局部遮挡与形变等问题,提高复杂背景下的跟踪效率。     

Multi-scale 2D tracking of articulated objects using hierarchical spring systems

This paper presents a flexible framework to build a target-specific, part-based representation for arbitrary articulated or rigid objects. The aim is to successfully track the target object in 2D, through multiple scales and occlusions. This is realized by employing a hierarchical, iterative optimization process on the proposed representation of structure and appearance. Therefore, each rigid part of an object is described by a hierarchical spring system represented by an attributed graph pyramid. Hierarchical spring systems encode the spatial relationships of the features (attributes of the graph pyramid) describing the parts and enforce them by spring-like behavior during tracking. Articulation points connecting the parts of the object allow to transfer position information from reliable to ambiguous parts. Tracking is done in an iterative process by combining the hypotheses of simple trackers with the hypotheses extracted from the hierarchical spring systems.