3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments

The field of visual localization has been researched for several decades and has meanwhile found many practical applications. Despite the strong progress in this field, there are still challenging situations in which established methods fail. We present an approach to significantly improve the accur...

Full description

Saved in:
Bibliographic Details
Main Authors: V. Ress, J. Meyer, W. Zhang, D. Skuddis, U. Soergel, N. Haala
Format: Article
Language:English
Published: Copernicus Publications 2025-07-01
Series:The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Online Access:https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1283/2025/isprs-archives-XLVIII-G-2025-1283-2025.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849706961628561408
author V. Ress
J. Meyer
J. Meyer
W. Zhang
D. Skuddis
U. Soergel
N. Haala
author_facet V. Ress
J. Meyer
J. Meyer
W. Zhang
D. Skuddis
U. Soergel
N. Haala
author_sort V. Ress
collection DOAJ
description The field of visual localization has been researched for several decades and has meanwhile found many practical applications. Despite the strong progress in this field, there are still challenging situations in which established methods fail. We present an approach to significantly improve the accuracy and reliability of established visual localization methods by adding rendered images. In detail, we first use a modern visual SLAM approach that provides a 3D Gaussian Splatting (3DGS) based map to create reference data. We demonstrate that enriching reference data with images rendered from 3DGS at randomly sampled poses significantly improves the performance of both geometry-based visual localization and Scene Coordinate Regression (SCR) methods. Through comprehensive evaluation in a large industrial environment, we analyze the performance impact of incorporating these additional rendered views.
format Article
id doaj-art-3e3a1b96cb8a4540a57091b0e2b669b8
institution DOAJ
issn 1682-1750
2194-9034
language English
publishDate 2025-07-01
publisher Copernicus Publications
record_format Article
series The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
spelling doaj-art-3e3a1b96cb8a4540a57091b0e2b669b82025-08-20T03:16:03ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342025-07-01XLVIII-G-20251283129010.5194/isprs-archives-XLVIII-G-2025-1283-20253D Gaussian Splatting aided Localization for Large and Complex Indoor-EnvironmentsV. Ress0J. Meyer1J. Meyer2W. Zhang3D. Skuddis4U. Soergel5N. Haala6Institute for Photogrammetry and Geoinformatics, University of Stuttgart, GermanyInstitute for Photogrammetry and Geoinformatics, University of Stuttgart, GermanyInstitute of Geomatics, University of Applied Sciences and Arts Northwestern Switzerland, SwitzerlandInstitute for Photogrammetry and Geoinformatics, University of Stuttgart, GermanyInstitute for Photogrammetry and Geoinformatics, University of Stuttgart, GermanyInstitute for Photogrammetry and Geoinformatics, University of Stuttgart, GermanyInstitute for Photogrammetry and Geoinformatics, University of Stuttgart, GermanyThe field of visual localization has been researched for several decades and has meanwhile found many practical applications. Despite the strong progress in this field, there are still challenging situations in which established methods fail. We present an approach to significantly improve the accuracy and reliability of established visual localization methods by adding rendered images. In detail, we first use a modern visual SLAM approach that provides a 3D Gaussian Splatting (3DGS) based map to create reference data. We demonstrate that enriching reference data with images rendered from 3DGS at randomly sampled poses significantly improves the performance of both geometry-based visual localization and Scene Coordinate Regression (SCR) methods. Through comprehensive evaluation in a large industrial environment, we analyze the performance impact of incorporating these additional rendered views.https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1283/2025/isprs-archives-XLVIII-G-2025-1283-2025.pdf
spellingShingle V. Ress
J. Meyer
J. Meyer
W. Zhang
D. Skuddis
U. Soergel
N. Haala
3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
title 3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments
title_full 3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments
title_fullStr 3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments
title_full_unstemmed 3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments
title_short 3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments
title_sort 3d gaussian splatting aided localization for large and complex indoor environments
url https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1283/2025/isprs-archives-XLVIII-G-2025-1283-2025.pdf
work_keys_str_mv AT vress 3dgaussiansplattingaidedlocalizationforlargeandcomplexindoorenvironments
AT jmeyer 3dgaussiansplattingaidedlocalizationforlargeandcomplexindoorenvironments
AT jmeyer 3dgaussiansplattingaidedlocalizationforlargeandcomplexindoorenvironments
AT wzhang 3dgaussiansplattingaidedlocalizationforlargeandcomplexindoorenvironments
AT dskuddis 3dgaussiansplattingaidedlocalizationforlargeandcomplexindoorenvironments
AT usoergel 3dgaussiansplattingaidedlocalizationforlargeandcomplexindoorenvironments
AT nhaala 3dgaussiansplattingaidedlocalizationforlargeandcomplexindoorenvironments