Skip to main content

Feature-enhanced Surfaces from Incomplete Point Cloud with Segmentation and Curve Skeleton Information.

Wang, M., Fan, Y., Guo, S., Liao, M., Chang, J. and He, D., 2017. Feature-enhanced Surfaces from Incomplete Point Cloud with Segmentation and Curve Skeleton Information. In: 2017 International Conference on Virtual Reality and Visualization (ICVRV), 21-22 October 2017, Zhengzhou, China, 97 - 102.

Full text available as:

[img]
Preview
PDF (©2017 IEEE)
08719135.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.

1MB

DOI: 10.1109/ICVRV.2017.00027

Abstract

Raw data of point cloud is often noisy and with topological defects (such as holes), which cause problems including faulty connection and inaccurate structure. As a result, the surface reconstruction of point cloud data is a highly challenging problem. This work proposes a novel method, which improves the surface quality compared with existing methods. Our method combines both the local detailed features and the global topological information during the reconstruction process. To facilitate the feature refinement, we first pre-process the point cloud data by relocating each point, upsampling the point data, and optimizing normals to enhance the features and geometric details. We then identify the topological information by segmenting the geometry and constructing curve skeletons for each part and guide the surface reconstruction with the skeletons by minimal user interaction. We demonstrate the effectiveness of our methods with various examples, where our reconstruction can fill out missing data and preserve sharp features.

Item Type:Conference or Workshop Item (Paper)
Uncontrolled Keywords:surface reconstruction; segmentation; enhanced features; curve skeleton
Group:Faculty of Media & Communication
ID Code:32429
Deposited By: Unnamed user with email symplectic@symplectic
Deposited On:25 Jun 2019 09:22
Last Modified:25 Oct 2019 10:58

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -