Statistical inefficiency often limits the effectiveness of particle filters for high-dimensional Bayesian tracking problems. To improve sampling efficiency on continuous domains, we propose the use of a particle filter with hybrid Monte Carlo (HMC), an MCMC (Markov chain Monte Carlo) method that follows posterior gradients toward. high probability states, while ensuring a properly weighted approximation to the posterior. We use HMC filtering to infer the 3D shape and motion of people from natural, monocular image sequences. The approach currently uses an empirical, edge-based likelihood function, and a second-order dynamic model with soft biomechanical joint constraints.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Hybrid Monte Carlo filtering: edge-based people tracking


    Contributors:
    Poon, E. (author) / Fleet, D.J. (author)


    Publication date :

    2002-01-01


    Size :

    1896262 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Hybrid Monte Carlo Filtering: Edge-Based People Tracking

    Poon, E. / Fleet, D. J. | British Library Conference Proceedings | 2002


    People Tracking Using Hybrid Monte Carlo Filtering

    Choo, K. / Fleet, D. / IEEE | British Library Conference Proceedings | 2001


    People tracking using hybrid Monte Carlo filtering

    Kiam Choo, / Fleet, D.J. | IEEE | 2001


    Improved sequential Monte Carlo filtering for ballistic target tracking

    Bruno, M.G.S. / Pavlov, A. | Tema Archive | 2005