Dynamic Voxel Grid Optimization for High-Fidelity RGB-D Supervised Surface Reconstruction

TL;DR

This paper presents a dynamic voxel grid optimization method for high-fidelity RGB-D surface reconstruction that adaptively refines complex regions, achieving detailed 3D models efficiently without prior knowledge.

Contribution

It introduces a novel dynamic grid refinement scheme that allocates finer voxels to complex regions, improving detail capture and computational efficiency in 3D surface reconstruction.

Findings

Produces high-quality detailed 3D reconstructions

Faster than baseline NeuralRGBD method

Effective on both synthetic and real-world data

Abstract

Direct optimization of interpolated features on multi-resolution voxel grids has emerged as a more efficient alternative to MLP-like modules. However, this approach is constrained by higher memory expenses and limited representation capabilities. In this paper, we introduce a novel dynamic grid optimization method for high-fidelity 3D surface reconstruction that incorporates both RGB and depth observations. Rather than treating each voxel equally, we optimize the process by dynamically modifying the grid and assigning more finer-scale voxels to regions with higher complexity, allowing us to capture more intricate details. Furthermore, we develop a scheme to quantify the dynamic subdivision of voxel grid during optimization without requiring any priors. The proposed approach is able to generate high-quality 3D reconstructions with fine details on both synthetic and real-world data, while maintaining computational efficiency, which is substantially faster than the baseline method NeuralRGBD.