一种新型多特征融合粒子滤波视觉跟踪算法

针对单一视觉信息在动态变化环境下描述目标不够充分、跟踪目标不够稳定的缺点,提出了一种基于粒子滤波框架的新型多特征融合的视觉跟踪算法。采用颜色和形状信息来描述运动模型,通过民主合成策略将两种信息融合在一起,使得跟踪算法能根据当前跟踪形势自适应调整两种信息的权重以期达到最佳的最大似然比,实现信息间的优势互补。在设计粒子滤波跟踪算法时,利用自适应信息融合策略构建似然模型,提高了粒子滤波跟踪算法在复杂场景下的稳健性。实验结果表明,多特征融合跟踪算法不仅能准确、高效地跟踪目标,而且对光照、姿态变化引起的目标表观变化具有良好的鲁棒性。     

Adaptive Rao–Blackwellized Particle Filter and Its Evaluation for Tracking in Surveillance

Particle filters can become quite inefficient when being applied to a high-dimensional state space since a prohibitively large number of samples may be required to approximate the underlying density functions with desired accuracy. In this paper, by proposing an adaptive Rao-Blackwellized particle filter for tracking in surveillance, we show how to exploit the analytical relationship among state variables to improve the efficiency and accuracy of a regular particle filter. Essentially, the distributions of the linear variables are updated analytically using a Kalman filter which is associated with each particle in a particle filtering framework. Experiments and detailed performance analysis using both simulated data and real video sequences reveal that the proposed method results in more accurate tracking than a regular particle filter     

Adaptive Rao-Blackwellized particle filter and its evaluation for tracking in surveillance.

Particle filters can become quite inefficient when being applied to a high-dimensional state space since a prohibitively large number of samples may be required to approximate the underlying density functions with desired accuracy. In this paper, by proposing an adaptive Rao-Blackwellized particle filter for tracking in surveillance, we show how to exploit the analytical relationship among state variables to improve the efficiency and accuracy of a regular particle filter. Essentially, the distributions of the linear variables are updated analytically using a Kalman filter which is associated with each particle in a particle filtering framework. Experiments and detailed performance analysis using both simulated data and real video sequences reveal that the proposed method results in more accurate tracking than a regular particle filter.     

基于加权子空间拟合的声源定位与跟踪方法

麦克风阵列声源定位可为复杂环境下的说话人空间位置估计问题提供一种有效的解决方案。该文基于粒子滤波框架,提出了一种加权子空间拟合声源定位与跟踪方法。该方法将窄带子空间拟合算法的代价函数推广至宽带情形,构建了一种适用于宽带语音信号的似然函数,并结合说话人的运动模型估计声源的位置。计算机仿真与实测结果验证了该方法的有效性。     

Real-time speaker tracking using particle filter sensor fusion

Sensor fusion for object tracking has become an active research direction during the past few years. But how to do it in a robust and principled way is still an open problem. In this paper, we propose a new fusion framework that combines both the bottom-up and top-down approaches to probabilistically fuse multiple sensing modalities. At the lower level, individual vision and audio trackers are designed to generate effective proposals for the fuser. At the higher level, the fuser performs reliable tracking by verifying hypotheses over multiple likelihood models from multiple cues. Unlike traditional fusion algorithms, the proposed framework is a closed-loop system where the fuser and trackers coordinate their tracking information. Furthermore, to handle nonstationary situations, the proposed framework evaluates the performance of the individual trackers and dynamically updates their object states. We present a real-time speaker tracking system based on the proposed framework by fusing object contour, color and sound source location. We report robust tracking results.     

Tracking in Wireless Sensor Networks Using Particle Filtering: Physical Layer Considerations

In this paper, a new framework for target tracking in a wireless sensor network using particle filters is proposed. Under this framework, the imperfect nature of the wireless communication channels between sensors and the fusion center along with some physical layer design parameters of the network are incorporated in the tracking algorithm based on particle filters. We call this approach ldquochannel-aware particle filtering.rdquo Channel-aware particle filtering schemes are derived for different wireless channel models and receiver architectures. Furthermore, we derive the posterior Cramer-Rao lower bounds (PCRLBs) for our proposed channel-aware particle filters. Simulation results are presented to demonstrate that the tracking performance of the channel-aware particle filters can reach their theoretical performance bounds even with relatively small number of sensors and they have superior performance compared to channel-unaware particle filters.     

基于融合策略自适应的多线索跟踪方法

基于多线索融合的跟踪是跟踪领域近年来的研究热点之一,该文结合两种常用的线索融合方式:乘性融合及加权和融合,提出一种融合策略自适应的鲁棒跟踪方法。该方法使用粒子滤波技术,统计样本的二阶中心矩并求Frobenius范数以表征线索的受噪声污染程度,最后适时切换两种融合策略。实践证明,新的融合策略比传统单一的融合方式更鲁棒。     

Abrupt motion tracking via intensively adaptive Markov-chain Monte Carlo sampling

The robust tracking of abrupt motion is a challenging task in computer vision due to its large motion uncertainty. While various particle filters and conventional Markov-chain Monte Carlo (MCMC) methods have been proposed for visual tracking, these methods often suffer from the well-known local-trap problem or from poor convergence rate. In this paper, we propose a novel sampling-based tracking scheme for the abrupt motion problem in the Bayesian filtering framework. To effectively handle the local-trap problem, we first introduce the stochastic approximation Monte Carlo (SAMC) sampling method into the Bayesian filter tracking framework, in which the filtering distribution is adaptively estimated as the sampling proceeds, and thus, a good approximation to the target distribution is achieved. In addition, we propose a new MCMC sampler with intensive adaptation to further improve the sampling efficiency, which combines a density-grid-based predictive model with the SAMC sampling, to give a proposal adaptation scheme. The proposed method is effective and computationally efficient in addressing the abrupt motion problem. We compare our approach with several alternative tracking algorithms, and extensive experimental results are presented to demonstrate the effectiveness and the efficiency of the proposed method in dealing with various types of abrupt motions.     

基于分层采样粒子滤波的麦克风阵列说话人跟踪方法

针对噪声与混响环境下的说话人跟踪问题,本文提出了一种基于粒子滤波的麦克风阵列声源定位与跟踪方法.该方法在粒子滤波框架下,将无混响影响的语音建立信号作为观测信息,通过计算麦克风阵列波束形成器的输出能量来构建似然函数,同时考虑语音信号不同频率成分在声源定位中的作用,利用分层采样方法提高粒子的采样效率.实验结果表明,本文方法提高了说话人声源跟踪系统的抗噪声与抗混响能力.     

A generic framework for tracking using particle filter with dynamic shape prior.

Tracking deforming objects involves estimating the global motion of the object and its local deformations as functions of time. Tracking algorithms using Kalman filters or particle filters (PFs) have been proposed for tracking such objects, but these have limitations due to the lack of dynamic shape information. In this paper, we propose a novel method based on employing a locally linear embedding in order to incorporate dynamic shape information into the particle filtering framework for tracking highly deformable objects in the presence of noise and clutter. The PF also models image statistics such as mean and variance of the given data which can be useful in obtaining proper separation of object and background.     

Gaussian particle filtering

Sequential Bayesian estimation for nonlinear dynamic state-space models involves recursive estimation of filtering and predictive distributions of unobserved time varying signals based on noisy observations. This paper introduces a new filter called the Gaussian particle filter. It is based on the particle filtering concept, and it approximates the posterior distributions by single Gaussians, similar to Gaussian filters like the extended Kalman filter and its variants. It is shown that under the Gaussianity assumption, the Gaussian particle filter is asymptotically optimal in the number of particles and, hence, has much-improved performance and versatility over other Gaussian filters, especially when nontrivial nonlinearities are present. Simulation results are presented to demonstrate the versatility and improved performance of the Gaussian particle filter over conventional Gaussian filters and the lower complexity than known particle filters.     

Embedding motion in model-based stochastic tracking.

Particle filtering is now established as one of the most popular methods for visual tracking. Within this framework, there are two important considerations. The first one refers to the generic assumption that the observations are temporally independent given the sequence of object states. The second consideration, often made in the literature, uses the transition prior as the proposal distribution. Thus, the current observations are not taken into account, requiring the noise process of this prior to be large enough to handle abrupt trajectory changes. As a result, many particles are either wasted in low likelihood regions of the state space, resulting in low sampling efficiency, or more importantly, propagated to distractor regions of the image, resulting in tracking failures. In this paper, we propose to handle both considerations using motion. We first argue that, in general, observations are conditionally correlated, and propose a new model to account for this correlation, allowing for the natural introduction of implicit and/or explicit motion measurements in the likelihood term. Second, explicit motion measurements are used to drive the sampling process towards the most likely regions of the state space. Overall, the proposed model handles abrupt motion changes and filters out visual distractors, when tracking objects with generic models based on shape or color distribution. Results were obtained on head tracking experiments using several sequences with moving camera involving large dynamics. When compared against the Condensation Algorithm, they have demonstrated the superior tracking performance of our approach.     

Visual tracking in high-dimensional state space by appearance-guided particle filtering

In this paper, we propose a new approach, appearance-guided particle filtering (AGPF), for high degree-of-freedom visual tracking from an image sequence. This method adopts some known attractors in the state space and integrates both appearance and motion-transition information for visual tracking. A probability propagation model based on these two types of information is derived from a Bayesian formulation, and a particle filtering framework is developed to realize it. Experimental results demonstrate that the proposed method is effective for high degree-of-freedom visual tracking problems, such as articulated hand tracking and lip-contour tracking.     

Combination of Multiple Measurement Cues for Visual Face Tracking

Visual face tracking is an important building block for all intelligent living and working spaces, as it is able to locate persons without any human intervention or the need for the users to carry sensors on themselves. In this paper we present a novel face tracking system built on a particle filtering framework that facilitates the use of non-linear visual measurements on the facial area. We concentrate on three different such non-linear visual measurement cues, namely object detection, foreground segmentation and colour matching. We derive robust measurement likelihoods under a unified representation scheme and fuse them into our face tracking algorithm. This algorithm is complemented with optimum selection of the particle filter’s object model and a target handling scheme. The resulting face tracking system is extensively evaluated and compared to baseline ones.     

粒子流粒子滤波检测前跟踪方法

检测前跟踪通过在连续多帧观测中对目标信号进行非相参积累以检测和跟踪微弱目标。积累的关键在于对目标轨迹的准确估计和多帧迭代滤波。传统粒子滤波器过于依赖建议分布，对目标轨迹的估计不够准确。新提出的粒子流滤波器是一种很好的替代方法，但其过于依赖当前时刻的量测而弱化多帧迭代滤波。本文提出一种在粒子滤波框架下采用粒子流的检测前跟踪方法:采用粒子滤波器进行多帧迭代滤波，但在每一帧内，采用Localized Exact Daum-Huang粒子流进行滤波。为了应对目标量测的不确定性，本文改造了Localized Exact Daum-Huang滤波器，为每个粒子在其邻域内寻找最大似然量测，并利用该量测更新粒子状态。Rayleigh分布杂波下Swerling1型起伏目标的检测和跟踪实验证明了所提算法的性能。      

自适应位置融合的目标跟踪算法

为提高分层卷积特征目标跟踪算法的实时性和鲁棒性,文中提出了一种基于多个相关滤波器预测位置自适应融合的实时目标跟踪算法。该算法首先提取VGG-19网络的Pool4层卷积特征,通过特征均值比对多通道的特征图进行裁剪,提高算法速度。然后利用不同高斯样本分布训练多个相关滤波分类器,并对所有分类器预测的目标位置进行自适应融合,提高算法对目标姿态变化的鲁棒性;最后采用稀疏模型更新策略,进一步提高算法速度。在OTB100标准数据集上测试本文算法, 实验结果表明,该算法的平均距离精度为86.3%,比原分层卷积特征跟踪算法提高了2.6个百分点,在目标发生遮挡、形变、相似背景干扰等情况时具有很好的鲁棒性;平均跟踪速度为45.2帧/s,是原算法的4倍,实时性能良好。     

Target tracking by fusion of random measures

In this paper we propose fusion methods for tracking a single target in a sensor network. The sensors use sequential Monte Carlo (SMC) techniques to process the received measurements and obtain random measures of the unknown states. We apply standard particle filtering (SPF) and cost-reference particle filtering (CRPF) methods. For both types of filtering, the random measures contain particles drawn from the state space. Associated to the particles, the SPF has weights representing probability masses, while the CRPF has user-defined costs measuring the quality of the particles. Summaries of the random measures are sent to the fusion center which combines them into a global summary. Similarly, the fusion center may send a global summary to the individual sensors that use it for improved tracking. Through extensive simulations and comparisons with other methods, we study the performance of the proposed algorithms. This work has been supported by the National Science Foundation under Award CCF-0515246 and the Office of Naval Research under Award N00014-06-1-0012.     

Information Regularized Sensor Fusion: Application to Localization With Distributed Motion Sensors

We propose the information regularization principle for fusing information from sets of identical sensors observing a target phenomenon. The principle basically proposes an importance-weighting scheme for each sensor measurement based on the mutual information based pairwise statistical similarity matrix between sensors. The principle is applied to maximum likelihood estimation and particle filter based state estimation. A demonstration of the proposed regularization scheme in centralized data fusion of dense motion detector networks for target tracking is provided. Simulations confirm that the introduction of information regularization significantly improves localization accuracy of both maximum likelihood and particle filter approaches compared to their baseline implementations. Outlier detection and sensor failure detection capabilities, as well as possible extensions of the principle to decentralized sensor fusion with communication constraints are briefly discussed.     

Temporal filtering for restoration of wavelet-compressed motion imagery

Temporal filtering of motion imagery can alleviate the effects of noise and artifacts in the data by incorporating observations of the imagery data from several distinct frames. If the noise that is expected to occur in the data is well-modeled by independent and identically distributed (IID) Gaussian noise, then straightforward algorithms can be designed that filter along motion trajectories in an optimal fashion. This paper addresses the restoration of motion imagery that have been compressed by scalar quantization of the data's discrete wavelet transform coefficients. Noise due to compression in such situations is neither independent nor identically distributed, and thus straightforward filters designed for the IID case are suboptimal. This paper provides a statistical characterization of the quantization error and shows how the improved noise modeling can be used in temporal filtering to improve visual quality of the decompressed motion imagery. Example restoration results include the cases where the data have been compressed by quantization of two- and three-dimensional wavelet transform coefficients. Although not developed in this work, the noise model is also directly applicable to other restoration algorithms that incorporate information from other time instants, such as super-resolution.