Accurate multi-view reconstruction using robust binocular stereo and surface meshing
University of British Columbia
IEEE Conference on Computer Vision and Pattern Recognition, 2008. CVPR 2008, p.1-8
@conference{bradley2008accurate,
title={Accurate multi-view reconstruction using robust binocular stereo and surface meshing},
author={Bradley, D. and Boubekeur, T. and Heidrich, W.},
booktitle={Computer Vision and Pattern Recognition, 2008. CVPR 2008. IEEE Conference on},
pages={1–8},
issn={1063-6919},
year={2008},
organization={IEEE}
}
This paper presents a new algorithm for multi-view reconstruction that demonstrates both accuracy and efficiency. Our method is based on robust binocular stereo matching, followed by adaptive point-based filtering of the merged point clouds, and efficient, high-quality mesh generation. All aspects of our method are designed to be highly scalable with the number of views. Our technique produces the most accurate results among current algorithms for a sparse number of viewpoints according to the Middlebury datasets. Additionally, we prove to be the most efficient method among non-GPU algorithms for the same datasets. Finally, our scaled-window matching technique also excels at reconstructing deformable objects with high-curvature surfaces, which we demonstrate with a number of examples.
January 17, 2011 by hgpu
