Perspective-n-Point in Practice: Performance, Robustness, and Accuracy for Mesh-Based Localisation

Visual localisation, the task of determining camera poses from images, has matured significantly, offering various solutions for handheld device localisation. This paper investigates the Perspective-n-Point (PnP) problem, a crucial step in visual localisation that is often underexplored in practical...

Full description

Saved in:
Bibliographic Details
Main Authors: F. Vultaggio, P. Fanta-Jende, M. Gerke
Format: Article
Language:English
Published: Copernicus Publications 2025-06-01
Series:The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Online Access:https://isprs-archives.copernicus.org/articles/XLVIII-1-W4-2025/131/2025/isprs-archives-XLVIII-1-W4-2025-131-2025.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849421732718313472
author F. Vultaggio
F. Vultaggio
P. Fanta-Jende
M. Gerke
author_facet F. Vultaggio
F. Vultaggio
P. Fanta-Jende
M. Gerke
author_sort F. Vultaggio
collection DOAJ
description Visual localisation, the task of determining camera poses from images, has matured significantly, offering various solutions for handheld device localisation. This paper investigates the Perspective-n-Point (PnP) problem, a crucial step in visual localisation that is often underexplored in practical applications. We evaluate the performance of state-of-the-art PnP algorithms with real-world data, analysing their impact on localisation accuracy and robustness. Using a dataset comprising a large-scale aerial mesh and smartphone images, we conduct experiments to assess PnP algorithm performance. Specifically, we examine the effects of PnP algorithms in isolation, followed by the incorporation of RANSAC for outlier rejection, and finally, the addition of non linear pose refinement. By maintaining a fixed set of 2D-3D correspondences, this approach allows us to: assess the true outlier rejection capabilities of PnP algorithms, quantify the accuracy improvement achievable with non linear pose refinement, and identify superior PnP algorithms for robust visual localisation.
format Article
id doaj-art-39d7e091da89431ea265920259178df5
institution Kabale University
issn 1682-1750
2194-9034
language English
publishDate 2025-06-01
publisher Copernicus Publications
record_format Article
series The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
spelling doaj-art-39d7e091da89431ea265920259178df52025-08-20T03:31:23ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342025-06-01XLVIII-1-W4-202513113810.5194/isprs-archives-XLVIII-1-W4-2025-131-2025Perspective-n-Point in Practice: Performance, Robustness, and Accuracy for Mesh-Based LocalisationF. Vultaggio0F. Vultaggio1P. Fanta-Jende2M. Gerke3Unit Assistive and Autonomous Systems, Center for Vision, Automation and Control, AIT Austrian Institute of Technology, Vienna, AustriaInstitute of Geodesy and Photogrammetry, Technische Universität Braunschweig, Braunschweig, GermanyUnit Assistive and Autonomous Systems, Center for Vision, Automation and Control, AIT Austrian Institute of Technology, Vienna, AustriaInstitute of Geodesy and Photogrammetry, Technische Universität Braunschweig, Braunschweig, GermanyVisual localisation, the task of determining camera poses from images, has matured significantly, offering various solutions for handheld device localisation. This paper investigates the Perspective-n-Point (PnP) problem, a crucial step in visual localisation that is often underexplored in practical applications. We evaluate the performance of state-of-the-art PnP algorithms with real-world data, analysing their impact on localisation accuracy and robustness. Using a dataset comprising a large-scale aerial mesh and smartphone images, we conduct experiments to assess PnP algorithm performance. Specifically, we examine the effects of PnP algorithms in isolation, followed by the incorporation of RANSAC for outlier rejection, and finally, the addition of non linear pose refinement. By maintaining a fixed set of 2D-3D correspondences, this approach allows us to: assess the true outlier rejection capabilities of PnP algorithms, quantify the accuracy improvement achievable with non linear pose refinement, and identify superior PnP algorithms for robust visual localisation.https://isprs-archives.copernicus.org/articles/XLVIII-1-W4-2025/131/2025/isprs-archives-XLVIII-1-W4-2025-131-2025.pdf
spellingShingle F. Vultaggio
F. Vultaggio
P. Fanta-Jende
M. Gerke
Perspective-n-Point in Practice: Performance, Robustness, and Accuracy for Mesh-Based Localisation
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
title Perspective-n-Point in Practice: Performance, Robustness, and Accuracy for Mesh-Based Localisation
title_full Perspective-n-Point in Practice: Performance, Robustness, and Accuracy for Mesh-Based Localisation
title_fullStr Perspective-n-Point in Practice: Performance, Robustness, and Accuracy for Mesh-Based Localisation
title_full_unstemmed Perspective-n-Point in Practice: Performance, Robustness, and Accuracy for Mesh-Based Localisation
title_short Perspective-n-Point in Practice: Performance, Robustness, and Accuracy for Mesh-Based Localisation
title_sort perspective n point in practice performance robustness and accuracy for mesh based localisation
url https://isprs-archives.copernicus.org/articles/XLVIII-1-W4-2025/131/2025/isprs-archives-XLVIII-1-W4-2025-131-2025.pdf
work_keys_str_mv AT fvultaggio perspectivenpointinpracticeperformancerobustnessandaccuracyformeshbasedlocalisation
AT fvultaggio perspectivenpointinpracticeperformancerobustnessandaccuracyformeshbasedlocalisation
AT pfantajende perspectivenpointinpracticeperformancerobustnessandaccuracyformeshbasedlocalisation
AT mgerke perspectivenpointinpracticeperformancerobustnessandaccuracyformeshbasedlocalisation