NESI: Shape Representation via Neural Explicit Surface Intersection

TL;DR

NESI introduces a novel shape representation method using neural explicit surface intersections, enabling efficient, accurate, and versatile 3D shape processing suitable for various digital media applications.

Contribution

The paper proposes a new explicit surface-based neural representation that supports occupancy and parametric queries, outperforming implicit methods in accuracy and compactness.

Findings

Reduces approximation error compared to state-of-the-art methods.

Supports a wider range of processing operations including occupancy and parametric access.

Achieves high-quality shape approximations with fewer parameters.

Abstract

Compressed representations of 3D shapes that are compact, accurate, and can be processed efficiently directly in compressed form, are extremely useful for digital media applications. Recent approaches in this space focus on learned implicit or parametric representations. While implicits are well suited for tasks such as in-out queries, they lack natural 2D parameterization, complicating tasks such as texture or normal mapping. Conversely, parametric representations support the latter tasks but are ill-suited for occupancy queries. We propose a novel learned alternative to these approaches, based on intersections of localized explicit, or height-field, surfaces. Since explicits can be trivially expressed both implicitly and parametrically, NESI directly supports a wider range of processing operations than implicit alternatives, including occupancy queries and parametric access. We represent input shapes using a collection of differently oriented height-field bounded half-spaces combined using volumetric Boolean intersections. We first tightly bound each input using a pair of oppositely oriented height-fields, forming a Double Height-Field (DHF) Hull. We refine this hull by intersecting it with additional localized height-fields (HFs) that capture surface regions in its interior. We minimize the number of HFs necessary to accurately capture each input and compactly encode both the DHF hull and the local HFs as neural functions defined over subdomains of R^2. This reduced dimensionality encoding delivers high-quality compact approximations. Given similar parameter count, or storage capacity, NESI significantly reduces approximation error compared to the state of the art, especially at lower parameter counts.