Articulate Object Model Acquisition from Visual Hull in Time Sequences

Masaaki IIYAMA  Koh KAKUSHO  Michihiko MINOH  

Publication
D - Abstracts of IEICE TRANSACTIONS on Information and Systems (Japanese Edition)   Vol.J89-D   No.6   pp.1379-1390
Publication Date: 2006/06/01
Online ISSN: 1881-0225
DOI: 
Print ISSN: 1880-4535
Type of Manuscript: PAPER
Category: 
Keyword: 
pose estimation,  articulate object,  EM algorithm,  volume intersection,  

Full Text(in Japanese): PDF(907.4KB)
>>Buy this Article


Summary: 
We propose a method to estimate articulate object's shape and motion from visual hulls in time sequences. Visual hulls contains superfluous region which is not a part of object shape, and such region has undesirable effects on the estimation. We use a voxel feature which reduces the undesirable effects. The experimental results show our method can estimate articulate object's shape and motion correctly.