Partial model of the UNC reading room, a real environment digitized with a 3D laser scanner. The floor texture has been replaced to emphasize the holes in the chair model. (a) Chair model reconstructed from the original samples only. Notice the big holes and missing outer surface on its left. (b) Model obtained from (a) using the symmetry-based techniques described in [40]. Despite the clear improvement, many small holes are still visible. (c) Complete model obtained applying our hole-filling algorithm to the model shown in (b). |
jianning@cs.sunysb.edu |
|
oliveira@inf.ufrgs.br |
Image and Vision Computing.
Volume 25, Issue 1, January 2007. pp. 103-113. [DOI]
Abstract | Downloads | Reference | Acknowledgments |
Creating models of real scenes is a complex task for which the use of traditional modeling techniques is inappropriate. For this task, laser rangefinders are frequently used to sample the scene from several viewpoints, with the resulting range images integrated into a final model. In practice, due to surface reflectance properties, occlusions and accessibility limitations, certain areas of the scenes are usually not sampled, leading to holes and introducing undesirable artifacts in the resulting models. We present an algorithm for filling holes on surfaces reconstructed from point clouds. The algorithm is based on moving least squares and can interpolate both geometry and shading information. The reconstruction process is mostly automatic and the sampling rate of the given samples is preserved in the reconstructed areas. We demonstrate the use of the algorithm on both real and synthetic datasets to obtain complete geometry and plausible shading.
Jianning Wang and Manuel M. Oliveira. "Filling holes on locally smooth surfaces reconstructed from point clouds". Image and Vision Computing. Volume 25 (2007), Number 1, pp. 103-113.
@article{Wang2007103, title = {Filling holes on locally smooth surfaces reconstructed from point clouds}, author = {Jianning Wang and Manuel M. Oliveira}, journal = {Image and Vision Computing}, volume = {25}, number = {1}, pages = {103 - 113}, year = {2007} }
Hole filling, Surface reconstruction, Moving least squares, 3D scanners.
The partial model of the UNC reading room was kindly provided by the UNC IBR Group. |