DistGrid: Scalable Scene Reconstruction with Distributed Multi-resolution Hash Grid

TL;DR

DistGrid introduces a scalable scene reconstruction approach using joint multi-resolution hash grids, effectively handling large-scale scenes without redundant background models, and outperforming existing methods in quality and efficiency.

Contribution

The paper presents a novel multi-resolution hash grid method for large-scale scene reconstruction that eliminates the need for multiple background models, improving scalability and efficiency.

Findings

Outperforms existing methods on large-scale scenes

Provides visually plausible and high-quality reconstructions

Eliminates redundancy of background NeRFs

Abstract

Neural Radiance Field~(NeRF) achieves extremely high quality in object-scaled and indoor scene reconstruction. However, there exist some challenges when reconstructing large-scale scenes. MLP-based NeRFs suffer from limited network capacity, while volume-based NeRFs are heavily memory-consuming when the scene resolution increases. Recent approaches propose to geographically partition the scene and learn each sub-region using an individual NeRF. Such partitioning strategies help volume-based NeRF exceed the single GPU memory limit and scale to larger scenes. However, this approach requires multiple background NeRF to handle out-of-partition rays, which leads to redundancy of learning. Inspired by the fact that the background of current partition is the foreground of adjacent partition, we propose a scalable scene reconstruction method based on joint Multi-resolution Hash Grids, named DistGrid. In this method, the scene is divided into multiple closely-paved yet non-overlapped Axis-Aligned Bounding Boxes, and a novel segmented volume rendering method is proposed to handle cross-boundary rays, thereby eliminating the need for background NeRFs. The experiments demonstrate that our method outperforms existing methods on all evaluated large-scale scenes, and provides visually plausible scene reconstruction. The scalability of our method on reconstruction quality is further evaluated qualitatively and quantitatively.