Shape As Points: A Differentiable Poisson Solver

TL;DR

This paper introduces Shape-As-Points, a differentiable point-to-mesh layer based on Poisson Surface Reconstruction, enabling fast, explicit, and interpretable 3D shape representation and reconstruction from point clouds.

Contribution

It presents a novel differentiable Poisson Surface Reconstruction layer that efficiently bridges point clouds and meshes for end-to-end shape optimization.

Findings

Accelerates inference by an order of magnitude compared to neural implicit methods.

Produces topology-agnostic, watertight manifold surfaces.

Effective in surface reconstruction from unoriented point clouds.

Abstract

In recent years, neural implicit representations gained popularity in 3D reconstruction due to their expressiveness and flexibility. However, the implicit nature of neural implicit representations results in slow inference time and requires careful initialization. In this paper, we revisit the classic yet ubiquitous point cloud representation and introduce a differentiable point-to-mesh layer using a differentiable formulation of Poisson Surface Reconstruction (PSR) that allows for a GPU-accelerated fast solution of the indicator function given an oriented point cloud. The differentiable PSR layer allows us to efficiently and differentiably bridge the explicit 3D point representation with the 3D mesh via the implicit indicator field, enabling end-to-end optimization of surface reconstruction metrics such as Chamfer distance. This duality between points and meshes hence allows us to represent shapes as oriented point clouds, which are explicit, lightweight and expressive. Compared to neural implicit representations, our Shape-As-Points (SAP) model is more interpretable, lightweight, and accelerates inference time by one order of magnitude. Compared to other explicit representations such as points, patches, and meshes, SAP produces topology-agnostic, watertight manifold surfaces. We demonstrate the effectiveness of SAP on the task of surface reconstruction from unoriented point clouds and learning-based reconstruction.