Scalable Surface Reconstruction with Delaunay-Graph Neural Networks

TL;DR

This paper presents a scalable, learning-based surface reconstruction method for large, defected point clouds using Delaunay-graph neural networks, effectively handling real-world MVS data and outperforming existing algorithms.

Contribution

It introduces a novel visibility-aware, graph neural network-based approach that combines deep learning with energy models for large-scale surface reconstruction.

Findings

Outperforms existing reconstruction algorithms on benchmarks.

Learns visibility models from synthetic data that generalize to real-world data.

Handles large-scale, defect-laden point clouds effectively.

Abstract

We introduce a novel learning-based, visibility-aware, surface reconstruction method for large-scale, defect-laden point clouds. Our approach can cope with the scale and variety of point cloud defects encountered in real-life Multi-View Stereo (MVS) acquisitions. Our method relies on a 3D Delaunay tetrahedralization whose cells are classified as inside or outside the surface by a graph neural network and an energy model solvable with a graph cut. Our model, making use of both local geometric attributes and line-of-sight visibility information, is able to learn a visibility model from a small amount of synthetic training data and generalizes to real-life acquisitions. Combining the efficiency of deep learning methods and the scalability of energy based models, our approach outperforms both learning and non learning-based reconstruction algorithms on two publicly available reconstruction benchmarks. Our code and data is available at https://github.com/raphaelsulzer/dgnn.