Patch-Grid: An Efficient and Feature-Preserving Neural Implicit Surface Representation

TL;DR

Patch-Grid is a novel neural implicit surface representation that efficiently captures complex shapes with sharp features, open boundaries, and thin structures, significantly reducing training time while maintaining high fidelity.

Contribution

It introduces a unified patch-grid approach with CSG-based merging and an octree structure, enabling fast training and accurate modeling of intricate geometric details.

Findings

Achieves state-of-the-art reconstruction quality for complex shapes.

Trains within seconds due to simplified CSG operations.

Effectively preserves sharp features and handles open/thin structures.

Abstract

Neural implicit representations are widely used for 3D shape modeling due to their smoothness and compactness, but traditional MLP-based methods struggle with sharp features, such as edges and corners in CAD models, and require long training times. To address these limitations, we propose Patch-Grid, a unified neural implicit representation that efficiently fits complex shapes, preserves sharp features, and handles open boundaries and thin geometric structures. Patch-Grid learns a signed distance field (SDF) for each surface patch using a learnable patch feature volume. To represent sharp edges and corners, it merges the learned SDFs via constructive solid geometry (CSG) operations. A novel merge grid organizes patch feature volumes within a shared octree structure, localizing and simplifying CSG operations. This design ensures robust merging of SDFs and significantly reduces computational complexity, enabling training within seconds while maintaining high fidelity. Experimental results show that Patch-Grid achieves state-of-the-art reconstruction quality for shapes with intricate sharp features, open surfaces, and thin structures, offering superior robustness, efficiency, and accuracy.