Tracking Multiple Visual Targets via Particle-Based Belief Propagation

Xue, Jianru; Zheng, Nanning; Geng, Jianhua; Zhong, Xiaopin

doi:10.1109/tsmcb.2007.910533

Cited by 28 publications

(19 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A variety of applications which involve distributed in-network statistical inference tasks among spatially inter-connected agents or sensors have been widely studied in many online/offline systems. In sensor networks, where the knowledge of statistical dependencies among sensed data is given, the tasks of target tracking [20,21,22], detection [23], parameter estimation [24,2] are the examples, see [4] for a survey. In social networks, where the underlying social phenomenon of interest such as voting models, rumor/opinion propagation [7] evolves over a given social interaction graph, the inference tasks of distributed consensus-based estimation [6], deanonymization of community-structured social network [8] and distributed observability [5] are studied.Message-passing has manifested as an efficient procedure for inference over graphical models that provide the framework of succinct model of the statistical uncertainty of multi-agents.…”

Section: Related Workmentioning

confidence: 99%

Learning Data Dependency with Communication Cost

Jang

Song

2018

Proceedings of the Eighteenth ACM International Symposium on Mobile Ad Hoc Networking and Computing

View full text Add to dashboard Cite

In this paper, we consider the problem of recovering a graph that represents the statistical data dependency among nodes for a set of data samples generated by nodes, which provides the basic structure to perform an inference task, such as MAP (maximum a posteriori). This problem is referred to as structure learning. When nodes are spatially separated in different locations, running an inference algorithm requires a non-negligible amount of message passing, incurring some communication cost. We inevitably have the trade-off between the accuracy of structure learning and the cost we need to pay to perform a given message-passing based inference task because the learnt edge structures of data dependency and physical connectivity graph are often highly different. In this paper, we formalize this trade-off in an optimization problem which outputs the data dependency graph that jointly considers learning accuracy and message-passing costs. We focus on a distributed MAP as the target inference task due to its popularity, and consider two different implementations, ASYNC-MAP and SYNC-MAP that have different message-passing mechanisms and thus different cost structures. In ASYNC-MAP, we propose a polynomial time learning algorithm that is optimal, motivated by the problem of finding a maximum weight spanning tree. In SYNC-MAP, we first prove that it is NP-hard and propose a greedy heuristic. For both implementations, we then quantify how the probability that the resulting data graphs from those learning algorithms differ from the ideal data graph decays as the number of data samples grows, using the large deviation principle, where the decaying rate is characterized by some topological structures of both original data dependency and physical connectivity graphs as well as the degree of the trade-off, which provides some guideline on how many samples are necessary to obtain a certain learning accuracy. We validate our theoretical findings through extensive simulations, which confirms that it has a good match.

show abstract

Section: Related Workmentioning

confidence: 99%

Learning Data Dependency with Communication Cost

Jang

Song

2018

Proceedings of the Eighteenth ACM International Symposium on Mobile Ad Hoc Networking and Computing

View full text Add to dashboard Cite

show abstract

“…Unlike Kalman filter tracks the evolution of only a single Gaussian, the mixture of Gaussian method tracks multiple Gaussian distribution simultaneously (Xue et al, 2008). In this paper, we use modified version of the usual pixel-level background subtraction method (Zirkovic and Heijden, 2005) which uses Gaussian mixture probability density.…”

Section: Adaptive Background Subtractionmentioning

confidence: 99%

An implementation of a novel vision‐based robotic tracking system

Talu

Soygüder²,

Aydoğmuş³

2010

Sensor Review

View full text Add to dashboard Cite

If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information. About Emerald www.emeraldinsight.comEmerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online products and additional customer resources and services.Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation. AbstractPurpose -The purpose of the paper is to present an approach to detect and isolate the sensor failures, using a bank of extended Kalman filters (EKFs) using an innovative initialization of covariance matrix using system dynamics. Design/methodology/approach -The EKF is developed for nonlinear flight dynamic estimation of a spacecraft and the effects of the sensor failures using a bank of Kalman filters in investigated. The approach is to develop fast convergence Kalman filter algorithm based on covariance matrix computation for rapid sensor fault detection. The proposed nonlinear filter has been tested and compared with the classical Kalman filter schemes via simulations performed on the model of a space vehicle; this simulation activity has shown the benefits of the novel approach. Findings -In the simulations, the rotational dynamics of a spacecraft dynamic model are considered, and the sensor failures are detected and isolated. Research limitations/implications -A novel fast convergence Kalman filter for detection and isolation of faulty sensors applied to the three axis spacecraft attitude control problem is examined and an effective approach to isolate the faulty sensor measurements is proposed. Advantages of using innovative initialization of covariance matrix are presented in the paper. The proposed scheme enhances the improvement in estimation accuracy. The proposed method takes advantage of both the fast convergence capability and the robustness of numerical stability. Quaternion-based initialization of the covariance matrix is not considered in this paper. Originality/value -A new fast converging Kalman filter for sensor fault detection and isolation by innovative initialization of covariance matrix applied to a nonlinear spacecraft dynamic model is examined and an effective approach to isolate the measurements from failed sensors is proposed. An EKF has been developed for the nonlinear dynamic estimation of an orbiting spacecraft. The proposed methodology detects and decides if and where a sensor fault has occurred, isolates the faulty sensor, and outputs the corresponding healthy sensor measurement.

show abstract

“…PF is a Sequential Monte Carlo (SMC) approach, and uses, recursively, Monte Carlo integration in its structure to solve the filtering problem. Particle filters have successfully been used in a wide variety of applications, such as adaptive change detection (Matsumoto and Yosui, 2007), human motion modeling and analysis (Zhang and Fan, 2010;del Rincón et al, 2011), analysis of facial expression (Pantic and Patras, 2006), joint audio-visual tracking (Talantzis et al, 2009), multiple target visual tracking (Pernkopf, 2008;Xue et al, 2008), and robotics (Duan et al, 2007;Duan and Cai, 2008).…”

Section: Introductionmentioning

confidence: 99%