Gravity-aligned Rotation Averaging with Circular Regression

TL;DR

This paper presents a novel rotation averaging method that incorporates gravity information using circular regression, significantly improving accuracy and speed in 3D reconstruction from images, especially when gravity data is available.

Contribution

The paper introduces a gravity-aligned rotation averaging algorithm based on circular regression, supporting partial gravity data and refining gravity estimates, with state-of-the-art results.

Findings

Achieves 13 AUC@$1^\u00b0$ improvement over SfM baseline.

Runs eight times faster than existing methods.

Outperforms planar pose graph optimization by 23 AUC@$1^\u00b0$.

Abstract

Reconstructing a 3D scene from unordered images is pivotal in computer vision and robotics, with applications spanning crowd-sourced mapping and beyond. While global Structure-from-Motion (SfM) techniques are scalable and fast, they often compromise on accuracy. To address this, we introduce a principled approach that integrates gravity direction into the rotation averaging phase of global pipelines, enhancing camera orientation accuracy and reducing the degrees of freedom. This additional information is commonly available in recent consumer devices, such as smartphones, mixed-reality devices and drones, making the proposed method readily accessible. Rooted in circular regression, our algorithm has similar convergence guarantees as linear regression. It also supports scenarios where only a subset of cameras have known gravity. Additionally, we propose a mechanism to refine error-prone gravity. We achieve state-of-the-art accuracy on four large-scale datasets. Particularly, the proposed method improves upon the SfM baseline by 13 AUC@$1^\circ$ points, on average, while running eight times faster. It also outperforms the standard planar pose graph optimization technique by 23 AUC@$1^\circ$ points. The code is at https://github.com/colmap/glomap.