Voxel Data Based Marker-less Human Motion Capture Using Geometry Model
نویسندگان
چکیده
In this paper, we propose a novel approach for human posture estimation using geometry model. A volumetric reconstruction of a participant is obtained from multi-camera images. After definition of body model, the geometry model is fitted into the 3D human reconstructed volume. The gray theory, which is applicable to the prediction problem of a time-varying nonlinear system, is utilized to perform the forecasting job during tracking. Moreover, a hierarchical estimation is applied to avoid local optimum. Finally the posture is gained from the geometric parameter. Experimental results show the efficiency of the proposed algorithm and precision of posture estimation.
منابع مشابه
Prediction of Human Vertebral Compressive Strength Using Quantitative Computed Tomography Based Nonlinear Finite Element Method
Introduction: Because of the importance of vertebral compressive fracture (VCF) role in increasing the patients’ death rate and reducing their quality of life, many studies have been conducted for a noninvasive prediction of vertebral compressive strength based on bone mineral density (BMD) determination and recently finite element analysis. In this study, QCT-voxel based nonlinear finite eleme...
متن کاملHuman Body Model Acquisition and Motion Capture Using Voxel Data
In this paper we present a system for human body model acquisition and tracking of its parameters from voxel data. 3D voxel reconstruction of the body in each frame is computed from silhouettes extracted from multiple cameras. The system performs automatic model acquisition using a template based initialization procedure and a Bayesian network for refinement of body part size estimates. The twi...
متن کاملModel-based Human Pose Estimation Using Labelled Voxels by ICP
We present a system for markerless motion capture by using labelled voxels which can recover human posture of the subsequent frames robustly and precisely on temporal coherence. The system uses 3D voxel data reconstructed from multiple synchronized video streams as input, and initialize the model posture by segmented silhouette, and then labeling the voxel data of next frame by fitting the huma...
متن کاملRegression-Based Human Motion Capture From Voxel Data
A regression based method is proposed to recover human body pose from 3D voxel data. In order to do this we need to convert the voxel data into a feature vector. This is done using a Bayesian approach based on Mixture of Probabilistic PCA that transforms a collection of 3D shape context descriptors, extracted from the voxels, to a compact feature vector. For the regression, the newly-proposed M...
متن کاملA system for marker-less motion capture
In this contribution we present a silhouette based human motion capture system. The system components contain silhouette extraction based on level sets, a correspondence module, which relates image data to model data and a pose estimation module. Experiments are done in different camera setups and we estimate the model components with 21 degrees of freedom in up to two frames per second. To eva...
متن کامل