UDGS-SLAM: UniDepth Assisted Gaussian Splatting for Monocular SLAM

Recent advancements in monocular neural depth estimation, particularly those achieved by the UniDepth network, have prompted the investigation of integrating UniDepth within a Gaussian splatting framework for monocular SLAM. This study presents UDGS-SLAM, a novel approach that eliminates the necessi...

Full description

Saved in:
Bibliographic Details
Main Authors: Mostafa Mansour, Ahmed Abdelsalam, Ari Happonen, Jari Porras, Esa Rahtu
Format: Article
Language:English
Published: Elsevier 2025-07-01
Series:Array
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S259000562500027X
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850218994780340224
author Mostafa Mansour
Ahmed Abdelsalam
Ari Happonen
Jari Porras
Esa Rahtu
author_facet Mostafa Mansour
Ahmed Abdelsalam
Ari Happonen
Jari Porras
Esa Rahtu
author_sort Mostafa Mansour
collection DOAJ
description Recent advancements in monocular neural depth estimation, particularly those achieved by the UniDepth network, have prompted the investigation of integrating UniDepth within a Gaussian splatting framework for monocular SLAM. This study presents UDGS-SLAM, a novel approach that eliminates the necessity of RGB-D sensors for depth estimation within Gaussian splatting framework. UDGS-SLAM employs statistical filtering to ensure local consistency of the estimated depth and jointly optimizes camera trajectory and Gaussian scene representation parameters. The proposed method achieves high-fidelity rendered images and low ATE-RMSE of the camera trajectory. The performance of UDGS-SLAM is rigorously evaluated using the TUM RGB-D dataset and benchmarked against several baseline methods, demonstrating superior performance across various scenarios. Additionally, an ablation study is conducted to validate design choices and investigate the impact of different network backbone encoders on system performance.
format Article
id doaj-art-20ec5fc6f0254079954d676c8badbfd5
institution OA Journals
issn 2590-0056
language English
publishDate 2025-07-01
publisher Elsevier
record_format Article
series Array
spelling doaj-art-20ec5fc6f0254079954d676c8badbfd52025-08-20T02:07:31ZengElsevierArray2590-00562025-07-012610040010.1016/j.array.2025.100400UDGS-SLAM: UniDepth Assisted Gaussian Splatting for Monocular SLAMMostafa Mansour0Ahmed Abdelsalam1Ari Happonen2Jari Porras3Esa Rahtu4Faculty of Engineering and Natural Sciences, Tampere University, Finland; Corresponding authors.School of Engineering Science, LUT University, Finland; Corresponding authors.School of Engineering Science, LUT University, FinlandSchool of Engineering Science, LUT University, Finland; School of Electrical Engineering, Aalto University, FinlandFaculty of Information Technology and Communication Sciences, Tampere University, FinlandRecent advancements in monocular neural depth estimation, particularly those achieved by the UniDepth network, have prompted the investigation of integrating UniDepth within a Gaussian splatting framework for monocular SLAM. This study presents UDGS-SLAM, a novel approach that eliminates the necessity of RGB-D sensors for depth estimation within Gaussian splatting framework. UDGS-SLAM employs statistical filtering to ensure local consistency of the estimated depth and jointly optimizes camera trajectory and Gaussian scene representation parameters. The proposed method achieves high-fidelity rendered images and low ATE-RMSE of the camera trajectory. The performance of UDGS-SLAM is rigorously evaluated using the TUM RGB-D dataset and benchmarked against several baseline methods, demonstrating superior performance across various scenarios. Additionally, an ablation study is conducted to validate design choices and investigate the impact of different network backbone encoders on system performance.http://www.sciencedirect.com/science/article/pii/S259000562500027XUniDepthGaussian splittingMonocular SLAMDense SLAMMappingScene representation
spellingShingle Mostafa Mansour
Ahmed Abdelsalam
Ari Happonen
Jari Porras
Esa Rahtu
UDGS-SLAM: UniDepth Assisted Gaussian Splatting for Monocular SLAM
Array
UniDepth
Gaussian splitting
Monocular SLAM
Dense SLAM
Mapping
Scene representation
title UDGS-SLAM: UniDepth Assisted Gaussian Splatting for Monocular SLAM
title_full UDGS-SLAM: UniDepth Assisted Gaussian Splatting for Monocular SLAM
title_fullStr UDGS-SLAM: UniDepth Assisted Gaussian Splatting for Monocular SLAM
title_full_unstemmed UDGS-SLAM: UniDepth Assisted Gaussian Splatting for Monocular SLAM
title_short UDGS-SLAM: UniDepth Assisted Gaussian Splatting for Monocular SLAM
title_sort udgs slam unidepth assisted gaussian splatting for monocular slam
topic UniDepth
Gaussian splitting
Monocular SLAM
Dense SLAM
Mapping
Scene representation
url http://www.sciencedirect.com/science/article/pii/S259000562500027X
work_keys_str_mv AT mostafamansour udgsslamunidepthassistedgaussiansplattingformonocularslam
AT ahmedabdelsalam udgsslamunidepthassistedgaussiansplattingformonocularslam
AT arihapponen udgsslamunidepthassistedgaussiansplattingformonocularslam
AT jariporras udgsslamunidepthassistedgaussiansplattingformonocularslam
AT esarahtu udgsslamunidepthassistedgaussiansplattingformonocularslam