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

arXiv:2405.04416v1 Announce Type: new
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 …

abstract arxiv capacity challenges cs.cv distributed grid hash however memory mlp nerf network neural radiance field object quality resolution scalable scale type while