Abstract
We present a new approach for tracking both the human body shape and the whole body motion with complete six DOF of each body limb without imposing rotation or translation constraints. First, a surface mesh with highly improved quality is obtained by using our new silhouette-based visual hull reconstruction method for each frame of multi-view videos. Then, a skinned mesh model is fitted to the data using hierarchical weighted ICP (HWICP) algorithm, where an easy-toadjust strategy for selecting the set of ICP registration points is given based on the weights of the skinned model and the Approximate Nearest Neighbors (ANN) method is applied for fast searching nearest neighbors. By comparing HWICP with the general hierarchical ICP (Iterative Closest Point) method based on synthetic data, we demonstrate the power of weighting corresponding point pairs in HWICP, especially when adjacent body segments of target are near 'cylindrical-shaped'.
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.
