SVRecon: Sparse Voxel Rasterization for Surface Reconstruction

2025-11-21

AI Summary

To address local minima and SDF discontinuities arising from voxel decoupling and boundary over-sharpening in sparse voxel-based SDF reconstruction, this work introduces sparse voxel rasterization into SDF-based surface reconstruction for the first time. We propose a visual-geometric initialization strategy to enhance convergence robustness and design intra-voxel-group structural consistency constraints alongside hierarchical spatial smoothing losses to explicitly enforce geometric continuity of the SDF field. Our method integrates sparse voxel representation, signed distance function modeling, and geometry-prior-driven optimization. Evaluated on multiple benchmarks, it achieves high-fidelity, high-accuracy 3D surface reconstruction with rapid convergence and significantly improved stability compared to state-of-the-art sparse SDF approaches.

Technology Category

• Computer Vision: 3D Computer Vision
• Search and Optimization: Mixed Discrete/Continuous Search
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

We extend the recently proposed sparse voxel rasterization paradigm to the task of high-fidelity surface reconstruction by integrating Signed Distance Function (SDF), named SVRecon. Unlike 3D Gaussians, sparse voxels are spatially disentangled from their neighbors and have sharp boundaries, which makes them prone to local minima during optimization. Although SDF values provide a naturally smooth and continuous geometric field, preserving this smoothness across independently parameterized sparse voxels is nontrivial. To address this challenge, we promote coherent and smooth voxel-wise structure through (1) robust geometric initialization using a visual geometry model and (2) a spatial smoothness loss that enforces coherent relationships across parent-child and sibling voxel groups. Extensive experiments across various benchmarks show that our method achieves strong reconstruction accuracy while having consistently speedy convergence. The code will be made public.

Problem

• Extending sparse voxel rasterization for high-fidelity surface reconstruction
• Maintaining geometric smoothness across independently parameterized sparse voxels
• Preventing local minima during optimization of spatially disentangled voxels

Innovation

• Sparse voxel rasterization with SDF integration
• Geometric initialization using visual geometry model
• Spatial smoothness loss for voxel coherence