Facial Features Tracking Using Auxiliary Particle Filtering and Observation Model Based on Bhattacharyya Distance

Shirinzadeh, Fatemeh; Seyedarabi, Hadi; Aghagolzadeh, Ali
October 2012
International Journal of Computer Theory & Engineering;Oct2012, Vol. 4 Issue 5, p758
Academic Journal
Recently, particle filtering has become an effective algorithm for facial feature tracking. One problem with particle filtering is that as the dimensionality of the state space increases, a large number of particles that are propagated from the previous time are wasted in area where they have low observation probability, hence a very large number of particles are necessary to track the state and, as a result the complexity increases and the speed of algorithm reduces. In this research, auxiliary particle filter with factorized likelihoods is used in order to overcome this problem. In a tracking approach, the estimated state is updated by incorporating the new observations. Therefore an observation model is needed. In this paper a novel color-based observation model that is invariant to changes in illumination intensity is proposed. The proposed observation model employs the Bhattacharyya distance to update a prior distribution calculated by the particle filter. In this paper experimentally is showed that the proposed algorithm clearly outperforms multiple independent template tracking.


Related Articles

  • A MONTE CARLO APPROACH TO CALCULATING PROBABILITIES FOR CONTINUOUS IDENTITY BY DESCENT DATA. Browning, Sharon // Journal of Applied Probability;Sep2000, Vol. 37 Issue 3, p850 

    Deals with a study which presented a Monte Carlo approach to calculating probabilities for continuous identity by descent data. Simulation study; Discussion.

  • Monte Carlo methods for estimating depletion potentials in highly size-asymmetrical hard sphere mixtures. Ashton, D. J.; Sánchez-Gil, V.; Wilding, N. B. // Journal of Chemical Physics;Oct2013, Vol. 139 Issue 14, p144102 

    We investigate Monte Carlo simulation strategies for determining the effective ('depletion') potential between a pair of hard spheres immersed in a dense sea of much smaller hard spheres. Two routes to the depletion potential are considered. The first is based on estimates of the insertion...

  • A new determination of the γπ → ππ anomalous amplitude via Ï€e- → Ï€-e-Ï€0 data. Giller, I.; Ocherashvili, A.; Ebertshäuser, T.; Moinester, M.; Scherer, S. // European Physical Journal A -- Hadrons & Nuclei;Aug2005, Vol. 25 Issue 2, p229 

    We discuss the reaction π- e- → π- e-π0 with the purpose of obtaining information on the γπ → ππ anomalous amplitude $ \cal {F}$3π. We compare a full calculation at $ \cal {O}$( p6) in chiral perturbation theory and various phenomenological...

  • A Glace at Monte Carlo Integration. Singh, Udayan // Mathematics in School;May2013, Vol. 42 Issue 3, p20 

    The article provides information on the Monte Carlo Integration method. A brief background of the Monte Carlo Method, often considered by students as a fascinating method of problem solving that involves the use of probability. It outlines the basic steps for integration or finding area under...

  • Homotopy probability theory II. Drummond-Cole, Gabriel; Park, Jae-Suk; Terilla, John // Journal of Homotopy & Related Structures;Sep2015, Vol. 10 Issue 3, p623 

    This is the second of two papers that introduce a deformation theoretic framework to explain and broaden a link between homotopy algebra and probability theory. This paper outlines how the framework can assist in the development of homotopy probability theory, where a vector space of random...

  • A statistical model of hydrogen bond networks in liquid alcohols. Sillrén, Per; Bielecki, Johan; Mattsson, Johan; Börjesson, Lars; Matic, Aleksandar // Journal of Chemical Physics;3/7/2012, Vol. 136 Issue 9, p094514 

    We here present a statistical model of hydrogen bond induced network structures in liquid alcohols. The model generalises the Andersson-Schulz-Flory chain model to allow also for branched structures. Two bonding probabilities are assigned to each hydroxyl group oxygen, where the first is the...

  • A thermal conductivity model of nanofluids based on particle size distribution analysis. Dengqing Zhou; Huiying Wu // Applied Physics Letters;8/25/2014, Vol. 105 Issue 8, p1 

    A model for predicting the thermal conductivity of nanofluids is proposed, in which the influence of nanoparticle clusters has been considered through the analysis of particle size distribution (PSD). The relation between PSD and thermal conductivity of nanofluids is established under the...

  • Stochastic frontier models with threshold efficiency. Lee, Sungwon; Lee, Young // Journal of Productivity Analysis;Aug2014, Vol. 42 Issue 1, p45 

    This paper proposes a tail-truncated stochastic frontier model that allows for the truncation of technical efficiency from below. The truncation bound implies the inefficiency threshold for survival. Specifically, this paper assumes a uniform distribution of technical inefficiency and derives...

  • Assessing the evidence for response time mixture distributions. Dixon, P. // Behavior Research Methods;Sep2012, Vol. 44 Issue 3, p706 

    I describe a technique for comparing two simple accounts of a distribution of response times: A mixture model and a generalized-shift model. In the mixture model, a target distribution is assumed to be a mixture of response times from two other (reference) distributions. In the generalized-shift...


Read the Article


Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics