Faithful Contouring: Near-Lossless 3D Voxel Representation Free from Iso-surface

TL;DR

Faithful Contouring introduces a high-resolution, near-lossless voxel representation for 3D meshes that preserves geometric details without relying on iso-surface extraction, improving accuracy and efficiency in 3D reconstruction.

Contribution

The paper presents a novel sparse voxelized representation supporting high resolutions and a dual-mode autoencoder, enabling scalable, accurate, and flexible 3D shape reconstruction without mesh-to-field conversions.

Findings

Achieves distance errors at the 10^{-5} level.

Reduces Chamfer Distance by 93%.

Improves F-score by 35% over baselines.

Abstract

Accurate and efficient voxelized representations of 3D meshes are the foundation of 3D reconstruction and generation. However, existing representations based on iso-surface heavily rely on water-tightening or rendering optimization, which inevitably compromise geometric fidelity. We propose Faithful Contouring, a sparse voxelized representation that supports 2048+ resolutions for arbitrary meshes, requiring neither converting meshes to field functions nor extracting the isosurface during remeshing. It achieves near-lossless fidelity by preserving sharpness and internal structures, even for challenging cases with complex geometry and topology. The proposed method also shows flexibility for texturing, manipulation, and editing. Beyond representation, we design a dual-mode autoencoder for Faithful Contouring, enabling scalable and detail-preserving shape reconstruction. Extensive experiments show that Faithful Contouring surpasses existing methods in accuracy and efficiency for both representation and reconstruction. For direct representation, it achieves distance errors at the $10^{-5}$ level; for mesh reconstruction, it yields a 93\% reduction in Chamfer Distance and a 35\% improvement in F-score over strong baselines, confirming superior fidelity as a representation for 3D learning tasks.