[Company Logo Image]  (*)

Home Up Feedback Contents Search

Human Motion   

HeadWheelchair ClipOn EMG SmartSpeakingKeyboard Virtual Multi Stereopsis Plant Phenotype 3D using Octree Limb-Volume measurement using Infrared Depth Sensor Object_Recognition Scene Understanding Field Phenotyping Background Subtraction Human Motion Calibration of VSNs Image-Based Servoing Robot Navigation Path Planning Inverse Kinematics Target Tracking Target Geolocation ODI Virtual Dermatologist CNN Virtual Machines for Image Processing New Models for Parallel Soft Computing

Home
Up

 

Human-Motion Capture

by Yuanqiang Evan Dong

Introduction

    Human motion capture has numerous application in human-robot interaction, law enforcement, surveillance, entertainment, sports, medicine, etc. Various methods have been developed to date and they can be categorized into: marker-based or markerless; articulated model-based or appearance-based; single view or multiple view; and so on. Marker-based methods are the simplest ones and therefore are also the methods with most success so far. However, it is obviously not always possible to add markers to the human subjects, and markerless approaches are without a doubt the most general and desirable methods.

    In our work, we explore markerless method for human motion capture. We propose a Bayesian estimation based method which falls into single view and articulated model-based category. The estimator, derived from Particle Filters, was expanded to a hierarchical model by introducing a new coarse-to-fine framework to deal with the computational complexity inherent to Particle Filters.

    Results

    Since the estimation of human pose from single view is restrictive to specific class of human motion, we tested our proposed method using partial frames of "Combo" and "Gesture" sequence in HumanEva I Dataset.

    1. Estimation results for "Combo_S1"

Pose estimation at "coarse" level

Pose estimation at "fine" level

 

    2. Estimation results for "Combo_S2"

    3. Estimation results for "Combo_S3"

    4. Estimation results for "Combo_S4"

    5. Estimation results for "Gesture_S1"

    6. Estimation results for "Gesture_S2"

    7. Estimation results for "Gesture_S3"

    8. Estimation results for "Gesture_S4"

 

    References

  1. Dong, Y., Conrad, D. and DeSouza, G. N., " Wii Using only 'We': Using Background Subtraction and Human Pose Recognition to Eliminate Game Controllers", in the Proceedings of the 2011 IEEE International Conference on Robotics and Automation (IEEE-ICRA). (submitted)
     

  2. Y. Dong and G. N. DeSouza, " A New Approach to the Analysis and Aggregation of Hierarchical Particle Filters for Human Motion Capture", Journal of Computer Vision and Image Understanding (submitted).
     

  3. Dong Y. and DeSouza, G.N., "A New Hierarchical Particle Filtering for Markerless Human Motion Capture", in the Proceedings of the 2009 IEEE Workshop on Computational Intelligence for Visual Intelligence and IEEE Symposium Series on Computational Intelligence (CIVI), pp. 14-21, Nashville, TN.

 

 

Home ] Up ]

Send mail to webmaster@ee.missouri.edu with questions or comments about this web site.
Last modified: 06/26/16
(*) Logo created by James Wong