Abstract
Tracking human motion from monocular video sequences has attracted a great deal of interests in recent years. The difficulty in solving this problem is largely due to the nonlinear property of human dynamics and the high dimensionality of the state vector space required to model human motion. Traditional particle filtering methods usually fail in this situation because the distributions they sample from are ill-defined. In this thesis we propose a novel tracking algorithm, namely the Differential Evolution - Markov Chain (DE-MC) particle filtering. It is based on the particle filter framework but makes substantial changes to its core, i.e. the sampling strategy. In this new approach, the Differential Evolution algorithm and the Markov Chain Monte Carlo algorithm are integrated, aiming at improving both the accuracy and efficiency in approximating the posterior distribution. Global optimization and importance sampling are spirits of the proposed method. To apply the DE-MC particle filter to articulated model-based human motion tracking, we also integrate multiple image cues including the area of silhouettes, color histograms and boundaries to measure the image likelihoods. We find the Fourier Descriptor (FD) to be a new and effective image feature in human motion tracking applications. Our other contributions, such as a modified color cue-based measurement function and a simple adaptive strategy for sampling, also help to improve the performance of the human tracker. Experimental results including the comparison with the performance of other particle filtering methods demonstrate the power of the proposed approach.
Highlights
1.1 General BackgroundOne of the most important characteristics of human resides in their extended ability to com municate
We carry out experiments with the proposed Differential Evolution - Markov Chain (DE-Markov chain (MC)) particle filter and measurement func tion
We will show the general tracking results first and some comparison results concerning the performance of the Differential Evolution Algorithm (DE)-MC particle filter and proposed measurement function
Summary
1.1 General BackgroundOne of the most important characteristics of human resides in their extended ability to com municate. Computers receive signals from humans through low bandwidth devices such as a keyboard or mouse These devices have turned out to be the bottleneck in communications between human and computer. For the work in which multiple image cues are combined, the overall measurement func tion is often defined as the multiplication of individual measurement functions (or the sum of their logarithmic version) It is based on the assumption of statistical independence (naive Bayesian model) of these cues. In place of the naive Bayesian model, they develop new image likelihood model based on Gibbs sampling theory Their experimental results show that the measurement function obtained with the Gibbs model yields a distribution much more approximative to the ground-truth distribution than naive Bayesian model
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
