PlanarSplatting: Accurate Planar Surface Reconstruction in 3 Minutes

TL;DR

PlanarSplatting is a novel, ultra-fast method for accurate indoor surface reconstruction from multiview images, leveraging 3D planes and CUDA acceleration to outperform existing techniques in speed and accuracy.

Contribution

It introduces a plane-based explicit optimization framework that eliminates the need for plane detection and matching, achieving rapid and precise indoor scene reconstruction.

Findings

Reconstructs indoor scenes in 3 minutes with high geometric accuracy

Outperforms existing methods on ScanNet and ScanNet++ datasets

Demonstrates significant speed and accuracy improvements

Abstract

This paper presents PlanarSplatting, an ultra-fast and accurate surface reconstruction approach for multiview indoor images. We take the 3D planes as the main objective due to their compactness and structural expressiveness in indoor scenes, and develop an explicit optimization framework that learns to fit the expected surface of indoor scenes by splatting the 3D planes into 2.5D depth and normal maps. As our PlanarSplatting operates directly on the 3D plane primitives, it eliminates the dependencies on 2D/3D plane detection and plane matching and tracking for planar surface reconstruction. Furthermore, the essential merits of plane-based representation plus CUDA-based implementation of planar splatting functions, PlanarSplatting reconstructs an indoor scene in 3 minutes while having significantly better geometric accuracy. Thanks to our ultra-fast reconstruction speed, the largest quantitative evaluation on the ScanNet and ScanNet++ datasets over hundreds of scenes clearly demonstrated the advantages of our method. We believe that our accurate and ultrafast planar surface reconstruction method will be applied in the structured data curation for surface reconstruction in the future. The code of our CUDA implementation will be publicly available. Project page: https://icetttb.github.io/PlanarSplatting/