Resolution Where It Counts: Hash-based GPU-Accelerated 3D Reconstruction via Variance-Adaptive Voxel Grids

TL;DR

This paper introduces a GPU-friendly, variance-adaptive voxel grid for real-time 3D reconstruction that improves memory efficiency and speed while maintaining accuracy, using hash tables and parallel processing techniques.

Contribution

It presents a novel multi-resolution voxel grid with variance-based adaptation and a hash table structure supporting GPU acceleration, surpassing traditional fixed-resolution and octree methods.

Findings

Achieves up to 13x speedup over fixed-resolution methods

Uses 4x less memory than baseline approaches

Maintains comparable reconstruction accuracy

Abstract

Efficient and scalable 3D surface reconstruction from range data remains a core challenge in computer graphics and vision, particularly in real-time and resource-constrained scenarios. Traditional volumetric methods based on fixed-resolution voxel grids or hierarchical structures like octrees often suffer from memory inefficiency, computational overhead, and a lack of GPU support. We propose a novel variance-adaptive, multi-resolution voxel grid that dynamically adjusts voxel size based on the local variance of signed distance field (SDF) observations. Unlike prior multi-resolution approaches that rely on recursive octree structures, our method leverages a flat spatial hash table to store all voxel blocks, supporting constant-time access and full GPU parallelism. This design enables high memory efficiency and real-time scalability. We further demonstrate how our representation supports GPU-accelerated rendering through a parallel quad-tree structure for Gaussian Splatting, enabling effective control over splat density. Our open-source CUDA/C++ implementation achieves up to 13x speedup and 4x lower memory usage compared to fixed-resolution baselines, while maintaining on par results in terms of reconstruction accuracy, offering a practical and extensible solution for high-performance 3D reconstruction.