Algorithmic and Architectural Optimizations for Computationally Efficient Particle Filtering

TitleAlgorithmic and Architectural Optimizations for Computationally Efficient Particle Filtering
Publication TypeJournal Articles
Year of Publication2008
AuthorsSankaranarayanan AC, Srivastava A, Chellappa R
JournalImage Processing, IEEE Transactions on
Pagination737 - 748
Date Published2008/05//
ISBN Number1057-7149
KeywordsAutomated;Reproducibility of Results;Sensitivity and Specificity;Signal Processing, Computer-Assisted;Models, Computer-Assisted;Video Recording;, convex program;independent Metropolis Hastings sampler;nonGaussian noise process;nonlinear dynamical system filtering;particle filtering algorithm;pipelined architectural optimization;video sequences;visual tracking;convex programming;image sequences;opti, Statistical;Pattern Recognition

In this paper, we analyze the computational challenges in implementing particle filtering, especially to video sequences. Particle filtering is a technique used for filtering nonlinear dynamical systems driven by non-Gaussian noise processes. It has found widespread applications in detection, navigation, and tracking problems. Although, in general, particle filtering methods yield improved results, it is difficult to achieve real time performance. In this paper, we analyze the computational drawbacks of traditional particle filtering algorithms, and present a method for implementing the particle filter using the Independent Metropolis Hastings sampler, that is highly amenable to pipelined implementations and parallelization. We analyze the implementations of the proposed algorithm, and, in particular, concentrate on implementations that have minimum processing times. It is shown that the design parameters for the fastest implementation can be chosen by solving a set of convex programs. The proposed computational methodology was verified using a cluster of PCs for the application of visual tracking. We demonstrate a linear speedup of the algorithm using the methodology proposed in the paper.