Abstract

This paper proposes a novel 3D reconstruction approach, dubbed Neural Procedural Reconstruction (NPR). NPR infers a sequence of shape grammar rule applications and reconstructs CAD-quality models with procedural structure from 3D points. While most existing methods rely on low-level geometry analysis to extract primitive structures, our approach conducts global analysis of entire building structures by deep neural networks (DNNs), enabling the reconstruction even from incomplete and sparse input data. We demonstrate the proposed system for residential buildings with aerial LiDAR as the input. Our 3D models boast compact geometry and semantically segmented architectural components. Qualitative and quantitative evaluations on hundreds of houses demonstrate that the proposed approach makes signicant improvements over the existing state-of-the-art.

Paper

Main paper

Code

Code (still updating)

Raw LiDAR Data

Can be downloaded here

Pre-processed Data and Annotation

Rotated Data

Rotated Annotations

Acknowledgement

This research is partially supported by National Science Foundation under grant IIS 1540012 and IIS 1618685, and Google Faculty Research Award. We thank Nvidia for a generous GPU donation.