Polyhedral Surface: Self-supervised Point Cloud Reconstruction Based on Polyhedral Surface
This addresses the problem of accurate 3D modeling and rendering in computer graphics, offering an incremental improvement over previous methods.
The paper tackles point cloud reconstruction by proposing a polyhedral surface representation to better capture sharp features and surface boundaries without requiring a local coordinate system, achieving state-of-the-art results on ShapeNetCore, ABC, and ScanNet datasets.
Point cloud reconstruction from raw point cloud has been an important topic in computer graphics for decades, especially due to its high demand in modeling and rendering applications. An important way to solve this problem is establishing a local geometry to fit the local curve. However, previous methods build either a local plane or polynomial curve. Local plane brings the loss of sharp feature and the boundary artefacts on open surface. Polynomial curve is hard to combine with neural network due to the local coordinate consistent problem. To address this, we propose a novel polyhedral surface to represent local surface. This method provides more flexible to represent sharp feature and surface boundary on open surface. It does not require any local coordinate system, which is important when introducing neural networks. Specifically, we use normals to construct the polyhedral surface, including both dihedral and trihedral surfaces using 2 and 3 normals, respectively. Our method achieves state-of-the-art results on three commonly used datasets (ShapeNetCore, ABC, and ScanNet). Code will be released upon acceptance.