Globally consistent dense 3D reconstruction with loop closure under fast camera motion

This paper introduces a novel globally consistent dense RGB-D SLAM system designed to address the challenges of online dense reconstruction under fast camera motion. To achieve robust pose tracking and globally consistent map building, we present a random optimization pose estimation framework that...

Full description

Saved in:
Bibliographic Details
Main Authors: Haigang Sui, Xuanhao Wang, Guohua Gou, Yang Ye, Han Li, Hao Zhang, Weicheng Jiang, Mingting Zhou
Format: Article
Language:English
Published: Taylor & Francis Group 2025-05-01
Series:Geo-spatial Information Science
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/10095020.2025.2505549
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850185007388164096
author Haigang Sui
Xuanhao Wang
Guohua Gou
Yang Ye
Han Li
Hao Zhang
Weicheng Jiang
Mingting Zhou
author_facet Haigang Sui
Xuanhao Wang
Guohua Gou
Yang Ye
Han Li
Hao Zhang
Weicheng Jiang
Mingting Zhou
author_sort Haigang Sui
collection DOAJ
description This paper introduces a novel globally consistent dense RGB-D SLAM system designed to address the challenges of online dense reconstruction under fast camera motion. To achieve robust pose tracking and globally consistent map building, we present a random optimization pose estimation framework that integrates both geometric and photometric information. Furthermore, we propose a voxel revisitation-based loop closure detection scheme that relies solely on depth maps. Innovatively, we introduce a local curvature feature, to address loop closure detection when photometric information is unavailable, efficiently extracting 3D geometric information directly from 2D images. In addition, we leverage voxel revisitation observations to accelerate the search for potential loop closures. This is achieved by maintaining a dynamic active map and detecting voxel outflow or revisits, thereby enhancing loop closure detection efficiency. This method is the first voxel revisitation-based loop closure detection scheme that relies entirely on depth information. Combined with a random optimization pose estimation framework utilizing both geometric and photometric features, it adapts well to fast camera motion, enabling accurate and globally consistent dense map construction. Extensive evaluations on public datasets demonstrate the effectiveness of our method under fast camera motion and highlight its significant advantages over existing methods.
format Article
id doaj-art-8ee8e96135704f13ba8a358fb6927ea1
institution OA Journals
issn 1009-5020
1993-5153
language English
publishDate 2025-05-01
publisher Taylor & Francis Group
record_format Article
series Geo-spatial Information Science
spelling doaj-art-8ee8e96135704f13ba8a358fb6927ea12025-08-20T02:16:50ZengTaylor & Francis GroupGeo-spatial Information Science1009-50201993-51532025-05-0112310.1080/10095020.2025.2505549Globally consistent dense 3D reconstruction with loop closure under fast camera motionHaigang Sui0Xuanhao Wang1Guohua Gou2Yang Ye3Han Li4Hao Zhang5Weicheng Jiang6Mingting Zhou7State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaState Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing, Wuhan University, Wuhan, ChinaThis paper introduces a novel globally consistent dense RGB-D SLAM system designed to address the challenges of online dense reconstruction under fast camera motion. To achieve robust pose tracking and globally consistent map building, we present a random optimization pose estimation framework that integrates both geometric and photometric information. Furthermore, we propose a voxel revisitation-based loop closure detection scheme that relies solely on depth maps. Innovatively, we introduce a local curvature feature, to address loop closure detection when photometric information is unavailable, efficiently extracting 3D geometric information directly from 2D images. In addition, we leverage voxel revisitation observations to accelerate the search for potential loop closures. This is achieved by maintaining a dynamic active map and detecting voxel outflow or revisits, thereby enhancing loop closure detection efficiency. This method is the first voxel revisitation-based loop closure detection scheme that relies entirely on depth information. Combined with a random optimization pose estimation framework utilizing both geometric and photometric features, it adapts well to fast camera motion, enabling accurate and globally consistent dense map construction. Extensive evaluations on public datasets demonstrate the effectiveness of our method under fast camera motion and highlight its significant advantages over existing methods.https://www.tandfonline.com/doi/10.1080/10095020.2025.2505549Dense reconstructionfast camera motion trackingdepth-only loop closurevoxel revisitsurface patches
spellingShingle Haigang Sui
Xuanhao Wang
Guohua Gou
Yang Ye
Han Li
Hao Zhang
Weicheng Jiang
Mingting Zhou
Globally consistent dense 3D reconstruction with loop closure under fast camera motion
Geo-spatial Information Science
Dense reconstruction
fast camera motion tracking
depth-only loop closure
voxel revisit
surface patches
title Globally consistent dense 3D reconstruction with loop closure under fast camera motion
title_full Globally consistent dense 3D reconstruction with loop closure under fast camera motion
title_fullStr Globally consistent dense 3D reconstruction with loop closure under fast camera motion
title_full_unstemmed Globally consistent dense 3D reconstruction with loop closure under fast camera motion
title_short Globally consistent dense 3D reconstruction with loop closure under fast camera motion
title_sort globally consistent dense 3d reconstruction with loop closure under fast camera motion
topic Dense reconstruction
fast camera motion tracking
depth-only loop closure
voxel revisit
surface patches
url https://www.tandfonline.com/doi/10.1080/10095020.2025.2505549
work_keys_str_mv AT haigangsui globallyconsistentdense3dreconstructionwithloopclosureunderfastcameramotion
AT xuanhaowang globallyconsistentdense3dreconstructionwithloopclosureunderfastcameramotion
AT guohuagou globallyconsistentdense3dreconstructionwithloopclosureunderfastcameramotion
AT yangye globallyconsistentdense3dreconstructionwithloopclosureunderfastcameramotion
AT hanli globallyconsistentdense3dreconstructionwithloopclosureunderfastcameramotion
AT haozhang globallyconsistentdense3dreconstructionwithloopclosureunderfastcameramotion
AT weichengjiang globallyconsistentdense3dreconstructionwithloopclosureunderfastcameramotion
AT mingtingzhou globallyconsistentdense3dreconstructionwithloopclosureunderfastcameramotion