GS-ReLoc: A Gaussian-Splatting Relocalization Method for Robust and Accurate Mono Camera Pose Estimation

Relocalization is a critical challenge in visual SLAM and autonomous navigation, where precise initial pose estimation is essential for robust system performance. Common approaches to visual relocalization rely on image retrieval to find the most similar image in a database, followed by 2D-2D featur...

Full description

Saved in:
Bibliographic Details
Main Authors: Karoly Fodor, Andras Rovid
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11039628/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850168860560326656
author Karoly Fodor
Andras Rovid
author_facet Karoly Fodor
Andras Rovid
author_sort Karoly Fodor
collection DOAJ
description Relocalization is a critical challenge in visual SLAM and autonomous navigation, where precise initial pose estimation is essential for robust system performance. Common approaches to visual relocalization rely on image retrieval to find the most similar image in a database, followed by 2D-2D feature matching to estimate the query image’s pose. However, these methods heavily depend on feature matching, which can be challenging when there is a significant spatial gap between the retrieved and query images or when local visual features are insufficient to establish reliable correspondences. To address image-retrieval database sparsity and reliability we present GS-ReLoc, a novel method that leverages 3D Gaussian Splat (3DGS) models to augment image retrieval databases. This augmentation increases the likelihood of initializing the rendering-based pose refinement process closer to the ground truth (GT) pose, leading to improved final pose estimates. The method begins by constructing a 3DGS model using Structure-from-Motion (SfM) reconstruction, which serves as the foundation for rendering novel virtual keyframes from novel poses. These keyframes enrich the database with diverse viewpoints through an efficient keyframe pose generation strategy. For a query frame, the algorithm identifies the best-matching database entry using a kd-tree structure, providing an initial pose estimate informed by the augmented database. This improved initial pose estimation strategy reduces the risk of the rendering-based pose refinement process converging to local minima, while also yielding higher final pose accuracy. The proposed method is evaluated on the indoor 7Scenes and outdoor Cambridge Landmark datasets, and achieves state-of-the-art pose estimation accuracy while maintaining robustness and computational efficiency, demonstrating its practical applicability in real-world relocalization scenarios.
format Article
id doaj-art-335ca195795246c3aa51afcf2405fc7b
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-335ca195795246c3aa51afcf2405fc7b2025-08-20T02:20:52ZengIEEEIEEE Access2169-35362025-01-011310708010709210.1109/ACCESS.2025.358106811039628GS-ReLoc: A Gaussian-Splatting Relocalization Method for Robust and Accurate Mono Camera Pose EstimationKaroly Fodor0https://orcid.org/0009-0000-5135-9230Andras Rovid1https://orcid.org/0000-0002-9044-1760Department of Automotive Technologies, Faculty of Transportation Engineering and Vehicle Engineering, Budapest University of Technology and Economics (BME), Budapest, HungaryDepartment of Automotive Technologies, Faculty of Transportation Engineering and Vehicle Engineering, Budapest University of Technology and Economics (BME), Budapest, HungaryRelocalization is a critical challenge in visual SLAM and autonomous navigation, where precise initial pose estimation is essential for robust system performance. Common approaches to visual relocalization rely on image retrieval to find the most similar image in a database, followed by 2D-2D feature matching to estimate the query image’s pose. However, these methods heavily depend on feature matching, which can be challenging when there is a significant spatial gap between the retrieved and query images or when local visual features are insufficient to establish reliable correspondences. To address image-retrieval database sparsity and reliability we present GS-ReLoc, a novel method that leverages 3D Gaussian Splat (3DGS) models to augment image retrieval databases. This augmentation increases the likelihood of initializing the rendering-based pose refinement process closer to the ground truth (GT) pose, leading to improved final pose estimates. The method begins by constructing a 3DGS model using Structure-from-Motion (SfM) reconstruction, which serves as the foundation for rendering novel virtual keyframes from novel poses. These keyframes enrich the database with diverse viewpoints through an efficient keyframe pose generation strategy. For a query frame, the algorithm identifies the best-matching database entry using a kd-tree structure, providing an initial pose estimate informed by the augmented database. This improved initial pose estimation strategy reduces the risk of the rendering-based pose refinement process converging to local minima, while also yielding higher final pose accuracy. The proposed method is evaluated on the indoor 7Scenes and outdoor Cambridge Landmark datasets, and achieves state-of-the-art pose estimation accuracy while maintaining robustness and computational efficiency, demonstrating its practical applicability in real-world relocalization scenarios.https://ieeexplore.ieee.org/document/11039628/Gaussian splat modelmono camera based relocalizationNetVLADimage-retrievalrendering-based pose refinement
spellingShingle Karoly Fodor
Andras Rovid
GS-ReLoc: A Gaussian-Splatting Relocalization Method for Robust and Accurate Mono Camera Pose Estimation
IEEE Access
Gaussian splat model
mono camera based relocalization
NetVLAD
image-retrieval
rendering-based pose refinement
title GS-ReLoc: A Gaussian-Splatting Relocalization Method for Robust and Accurate Mono Camera Pose Estimation
title_full GS-ReLoc: A Gaussian-Splatting Relocalization Method for Robust and Accurate Mono Camera Pose Estimation
title_fullStr GS-ReLoc: A Gaussian-Splatting Relocalization Method for Robust and Accurate Mono Camera Pose Estimation
title_full_unstemmed GS-ReLoc: A Gaussian-Splatting Relocalization Method for Robust and Accurate Mono Camera Pose Estimation
title_short GS-ReLoc: A Gaussian-Splatting Relocalization Method for Robust and Accurate Mono Camera Pose Estimation
title_sort gs reloc a gaussian splatting relocalization method for robust and accurate mono camera pose estimation
topic Gaussian splat model
mono camera based relocalization
NetVLAD
image-retrieval
rendering-based pose refinement
url https://ieeexplore.ieee.org/document/11039628/
work_keys_str_mv AT karolyfodor gsrelocagaussiansplattingrelocalizationmethodforrobustandaccuratemonocameraposeestimation
AT andrasrovid gsrelocagaussiansplattingrelocalizationmethodforrobustandaccuratemonocameraposeestimation