摘要 :
In sign-language or gesture recognition, articulated hand motion tracking is usually a prerequisite to behaviour understanding. However the difficulties such as non-rigidity of the hand, complex background scenes, and occlusion et...
展开
In sign-language or gesture recognition, articulated hand motion tracking is usually a prerequisite to behaviour understanding. However the difficulties such as non-rigidity of the hand, complex background scenes, and occlusion etc make tracking a challenging task. In this paper we present a hybrid HMM/Particle filter tracker for simultaneously tracking and recognition of non-rigid hand motion. By utilising separate image cues, we decompose complex motion into two independent (non-rigid/rigid) components. A generative model is used to explore the intrinsic patterns of the hand articulation. Non-linear dynamics of the articulation such as fast appearance deformation can therefore be tracked without resorting to a complex kinematic model. The rigid motion component is approximated as the motion of a planar region, where a standard particle filter method suffice. The novel contribution of the paper is that we unify the independent treatments of non-rigid motion and rigid motion into a robust Bayesian framework. The efficacy of this method is demonstrated by performing successful tracking in the presence of significant occlusion clutter.
收起
摘要 :
In sign-language or gesture recognition, articulated hand motion tracking is usually a prerequisite to behaviour understanding. However the difficulties such as non-rigidity of the hand, complex background scenes, and occlusion et...
展开
In sign-language or gesture recognition, articulated hand motion tracking is usually a prerequisite to behaviour understanding. However the difficulties such as non-rigidity of the hand, complex background scenes, and occlusion etc make tracking a challenging task. In this paper we present a hybrid HMM/Particle filter tracker for simultaneously tracking and recognition of non-rigid hand motion. By utilising separate image cues, we decompose complex motion into two independent (non-rigid/rigid) components. A generative model is used to explore the intrinsic patterns of the hand articulation. Non-linear dynamics of the articulation such as fast appearance deformation can therefore be tracked without resorting to a complex kinematic model. The rigid motion component is approximated as the motion of a planar region, where a standard particle filter method suffice. The novel contribution of the paper is that we unify the independent treatments of non-rigid motion and rigid motion into a robust Bayesian framework. The efficacy of this method is demonstrated by performing successful tracking in the presence of significant occlusion clutter.
收起
摘要 :
An attempt has been made to analyze impact of Martian atmosphere parameter uncertainties on entry vehicle aerodynamics for hypersonic rarefied conditions with a DSMC code. The code has been validated by comparing Viking vehicle fl...
展开
An attempt has been made to analyze impact of Martian atmosphere parameter uncertainties on entry vehicle aerodynamics for hypersonic rarefied conditions with a DSMC code. The code has been validated by comparing Viking vehicle flight data with present computational results. Then, by simulating flows around the Mars Science Laboratory, the impact of errors of free stream parameter uncertainties on aerodynamics is investigated. The validation results show that the present numerical approach can show good agreement with the Viking flight data. The physical and chemical properties of CO_2 has strong impact on aerodynamics of Mars entry vehicles, so it is necessary to make proper corrections to the data obtained with air model in hypersonic rarefied conditions, which is consistent with the conclusions drawn in continuum regime. Uncertainties of free stream density and velocity weakly influence aerodynamics and pitching moment. However, aerodynamics appears to be little influenced by free stream temperature, the maximum error of what is below 0.5%. Center of pressure position is not sensitive to free stream parameters.
收起
摘要 :
Successful tracking of articulated hand motion is the first step in many computer vision applications such as gesture recognition. However the nonrigidity of the hand, complex background scenes, and occlusion make tracking a chall...
展开
Successful tracking of articulated hand motion is the first step in many computer vision applications such as gesture recognition. However the nonrigidity of the hand, complex background scenes, and occlusion make tracking a challenging task. We divide and conquer tracking by decomposing complex motion into nonrigid motion and rigid motion. A learning-based algorithm for analyzing nonrigid motion is presented. In this method, appearance-based models are learned from image data, and underlying motion patterns are explored using a generative model. Nonlinear dynamics of the articulation such as fast appearance deformation can thus be analyzed without resorting to a complex kinematic model. We approximate the rigid motion as planar motion which can be approached by a filtering method. We unify our treatments of nonrigid motion and rigid motion into a single, robust Bayesian framework and demonstrate the efficacy of this method by performing successful tracking in the presence of significant occlusion clutter.
收起
摘要 :
In Direct Simulation Monte Carlo (DSMC) method, it is difficulty to keep grid size smaller than local mean free path, that influences calculation accuracy much. In this paper, a new modification technique of collision probability ...
展开
In Direct Simulation Monte Carlo (DSMC) method, it is difficulty to keep grid size smaller than local mean free path, that influences calculation accuracy much. In this paper, a new modification technique of collision probability with least square method and sub-relaxation technique is introduced to improve simulation accuracy in DSMC with coarse grid size. The new method is applied to direct simulation Monte Carlo method based on unstructured tetrahedral cells, which can be easily used for other grid topology. A numerical example with different grid size is implemented to validate the method discussed here, and the numerical results confirm feasibility of the modification technique.
收起
摘要 :
The two flux approximation model is usually used to calculate the heat transfer of radiation for porous materials. In this paper, the parameters identification method of the extinction coefficient and the albedo of scattering in t...
展开
The two flux approximation model is usually used to calculate the heat transfer of radiation for porous materials. In this paper, the parameters identification method of the extinction coefficient and the albedo of scattering in the two flux model is set up by using the identification theory according to the data of the back temperature. The simulated process shows that the convergence rate is fast and the results by the parameters identification are very close to the origin values.
收起