Points2Surf: Learning Implicit Surfaces from Point Cloud Patches

TL;DR

Points2Surf is a novel patch-based deep learning framework that accurately reconstructs surfaces from raw point clouds, outperforming existing methods especially on unseen shapes, by leveraging local and global information.

Contribution

It introduces a patch-based learning approach that improves generalization and reconstruction accuracy over prior deep learning methods for surface reconstruction from point clouds.

Findings

Reduces reconstruction error by 30% compared to SPR.

Achieves over 270% improvement over previous deep learning methods.

Demonstrates superior performance on synthetic and real data, especially on unseen classes.

Abstract

A key step in any scanning-based asset creation workflow is to convert unordered point clouds to a surface. Classical methods (e.g., Poisson reconstruction) start to degrade in the presence of noisy and partial scans. Hence, deep learning based methods have recently been proposed to produce complete surfaces, even from partial scans. However, such data-driven methods struggle to generalize to new shapes with large geometric and topological variations. We present Points2Surf, a novel patch-based learning framework that produces accurate surfaces directly from raw scans without normals. Learning a prior over a combination of detailed local patches and coarse global information improves generalization performance and reconstruction accuracy. Our extensive comparison on both synthetic and real data demonstrates a clear advantage of our method over state-of-the-art alternatives on previously unseen classes (on average, Points2Surf brings down reconstruction error by 30% over SPR and by 270%+ over deep learning based SotA methods) at the cost of longer computation times and a slight increase in small-scale topological noise in some cases. Our source code, pre-trained model, and dataset are available on: https://github.com/ErlerPhilipp/points2surf