Automated Reconstruction of 3D Open Surfaces from Sparse Point Clouds

TL;DR

This paper introduces IPVNet, a novel learning-based implicit model that reconstructs open 3D surfaces from sparse point clouds by combining raw data and voxel representations, outperforming existing methods.

Contribution

IPVNet is the first model to effectively reconstruct open surfaces by predicting unsigned distances using both raw point clouds and voxel data, reducing outliers.

Findings

IPVNet outperforms state-of-the-art methods in accuracy.

Reconstruction produces fewer outliers.

Effective on both synthetic and real-world datasets.

Abstract

Real-world 3D data may contain intricate details defined by salient surface gaps. Automated reconstruction of these open surfaces (e.g., non-watertight meshes) is a challenging problem for environment synthesis in mixed reality applications. Current learning-based implicit techniques can achieve high fidelity on closed-surface reconstruction. However, their dependence on the distinction between the inside and outside of a surface makes them incapable of reconstructing open surfaces. Recently, a new class of implicit functions have shown promise in reconstructing open surfaces by regressing an unsigned distance field. Yet, these methods rely on a discretized representation of the raw data, which loses important surface details and can lead to outliers in the reconstruction. We propose IPVNet, a learning-based implicit model that predicts the unsigned distance between a surface and a query point in 3D space by leveraging both raw point cloud data and its discretized voxel counterpart. Experiments on synthetic and real-world public datasets demonstrates that IPVNet outperforms the state of the art while producing far fewer outliers in the reconstruction.