Spline Positional Encoding for Learning 3D Implicit Signed Distance Fields
This work addresses the challenge of detailed 3D shape reconstruction for applications in computer graphics and vision, though it appears incremental as it builds upon existing positional encoding methods.
The paper tackled the problem of recovering fine-scale geometric details in 3D signed distance fields from unorganized point clouds by proposing Spline Positional Encoding, a novel positional encoding scheme that maps input coordinates to a high-dimensional space before processing with MLPs, and demonstrated its superiority over other encoding schemes in tasks like 3D shape reconstruction and shape space learning.
Multilayer perceptrons (MLPs) have been successfully used to represent 3D shapes implicitly and compactly, by mapping 3D coordinates to the corresponding signed distance values or occupancy values. In this paper, we propose a novel positional encoding scheme, called Spline Positional Encoding, to map the input coordinates to a high dimensional space before passing them to MLPs, for helping to recover 3D signed distance fields with fine-scale geometric details from unorganized 3D point clouds. We verified the superiority of our approach over other positional encoding schemes on tasks of 3D shape reconstruction from input point clouds and shape space learning. The efficacy of our approach extended to image reconstruction is also demonstrated and evaluated.