Name
Abstract | ||
|---|---|---|
AbstractWe present TreePartNet, a neural network aimed at reconstructing tree geometry from point clouds obtained by scanning real trees. Our key idea is to learn a natural neural decomposition exploiting the assumption that a tree comprises locally cylindrical shapes. In particular, reconstruction is a two-step process. First, two networks are used to detect priors from the point clouds. One detects semantic branching points, and the other network is trained to learn a cylindrical representation of the branches. In the second step, we apply a neural merging module to reduce the cylindrical representation to a final set of generalized cylinders combined by branches. We demonstrate results of reconstructing realistic tree geometry for a variety of input models and with varying input point quality, e.g., noise, outliers, and incompleteness. We evaluate our approach extensively by using data from both synthetic and real trees and comparing it with alternative methods. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1145/3478513.3480486 | ACM Transactions on Graphics |
Keywords | DocType | Volume |
3D Reconstruction, Procedural Modeling, Deep Learning, Optimization, Procedural Generation, Geometric Modeling | Journal | 40 |
Issue | ISSN | Citations |
6 | 0730-0301 | 0 |
PageRank | References | Authors |
0.34 | 48 | 6 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yanchao Liu | 1 | 0 | 0.34 |
| Jianwei Guo | 2 | 27 | 11.29 |
| Bedrich Benes | 3 | 1276 | 80.15 |
| Oliver Deussen | 4 | 2852 | 205.16 |
| Xiaopeng Zhang | 5 | 372 | 36.34 |
| Hui Huang | 6 | 694 | 52.19 |