# Automatic normal orientation in point clouds of building interiors

## TL;DR

This paper presents a novel method for automatically orienting surface normals in unstructured indoor point clouds, addressing challenges in urban scene measurements and improving accuracy for applications like visualization and reconstruction.

## Contribution

The proposed approach is specifically designed for complex indoor point clouds, overcoming limitations of existing methods that assume simpler data or topology.

## Key findings

- High accuracy in normal orientation on real-world datasets
- Fast processing suitable for large indoor scans
- Effective in multi-story, multi-room building environments

## Abstract

Orienting surface normals correctly and consistently is a fundamental problem in geometry processing. Applications such as visualization, feature detection, and geometry reconstruction often rely on the availability of correctly oriented normals. Many existing approaches for automatic orientation of normals on meshes or point clouds make severe assumptions on the input data or the topology of the underlying object which are not applicable to real-world measurements of urban scenes. In contrast, our approach is specifically tailored to the challenging case of unstructured indoor point cloud scans of multi-story, multi-room buildings. We evaluate the correctness and speed of our approach on multiple real-world point cloud datasets.