Visual Tracking by Continuous Density Propagation in Sequential Bayesian Filtering Framework

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 31; no. 5; pp. 919 - 930
• Main Authors: Bohyung Han, Ying Zhu, Comaniciu, D., Davis, L.S.
• Format: Journal Article
• Language: English
• Published: Los Alamitos, CA IEEE 01.05.2009; IEEE Computer Society; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithmics. Computability. Computer arithmetics; Algorithms; Applied sciences; Artificial Intelligence; Bayes Theorem; Bayesian filtering; Bayesian methods; Computer science; control theory; systems; Costs; Density; density approximation; Density functional theory; density interpolation; density propagation; Dynamical systems; Exact sciences and technology; Filtering; Image Enhancement - methods; Image Interpretation, Computer-Assisted - methods; Interpolation; Mathematical analysis; mean shift; Monte Carlo methods; Nonlinear dynamics; particle filter; Particle filters; Particle tracking; Pattern Recognition, Automated - methods; Pattern recognition. Digital image processing. Computational geometry; Probability density function; Reproducibility of Results; Sampling; Sampling methods; Sensitivity and Specificity; Signal Processing, Computer-Assisted; Subtraction Technique; Theoretical computing; visual tracking; Computer vision; Motion; Tracking; Video analysis; mean shift; Image processing; visual tracking; Dynamical system; Modeling; density interpolation; Analytic representation; Multidimensional analysis; density propagation; Image sequence; Probability density function; Sampling; Pattern analysis; Particle filter; Analytical function; density approximation; Bayes estimation; Mixed distribution; Monte Carlo method; Filtering; Motion estimation; Non gaussian noise; Real time; Bayesian filtering; Gaussian process; Density function; Function representation; Artificial intelligence
• ISSN: 0162-8828; 1939-3539
• DOI: 10.1109/TPAMI.2008.134

Particle filtering is frequently used for visual tracking problems since it provides a general framework for estimating and propagating probability density functions for nonlinear and non-Gaussian dynamic systems. However, this algorithm is based on a Monte Carlo approach and the cost of sampling and measurement is a problematic issue, especially for high-dimensional problems. We describe an alternative to the classical particle filter in which the underlying density function has an analytic representation for better approximation and effective propagation. The techniques of density interpolation and density approximation are introduced to represent the likelihood and the posterior densities with Gaussian mixtures, where all relevant parameters are automatically determined. The proposed analytic approach is shown to perform more efficiently in sampling in high-dimensional space. We apply the algorithm to real-time tracking problems and demonstrate its performance on real video sequences as well as synthetic examples.