PointSLAM++: Robust Dense Neural Gaussian Point Cloud-based SLAM

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Xu, Han, Boyao, Chen, Xiaojun, Liu, Ying, Li, Ruihui
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914260254195712
author Wang, Xu
Han, Boyao
Chen, Xiaojun
Liu, Ying
Li, Ruihui
author_facet Wang, Xu
Han, Boyao
Chen, Xiaojun
Liu, Ying
Li, Ruihui
contents Real-time 3D reconstruction is crucial for robotics and augmented reality, yet current simultaneous localization and mapping(SLAM) approaches often struggle to maintain structural consistency and robust pose estimation in the presence of depth noise. This work introduces PointSLAM++, a novel RGB-D SLAM system that leverages a hierarchically constrained neural Gaussian representation to preserve structural relationships while generating Gaussian primitives for scene mapping. It also employs progressive pose optimization to mitigate depth sensor noise, significantly enhancing localization accuracy. Furthermore, it utilizes a dynamic neural representation graph that adjusts the distribution of Gaussian nodes based on local geometric complexity, enabling the map to adapt to intricate scene details in real time. This combination yields high-precision 3D mapping and photorealistic scene rendering. Experimental results show PointSLAM++ outperforms existing 3DGS-based SLAM methods in reconstruction accuracy and rendering quality, demonstrating its advantages for large-scale AR and robotics.
format Preprint
id arxiv_https___arxiv_org_abs_2601_11617
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PointSLAM++: Robust Dense Neural Gaussian Point Cloud-based SLAM
Wang, Xu
Han, Boyao
Chen, Xiaojun
Liu, Ying
Li, Ruihui
Computer Vision and Pattern Recognition
Graphics
Robotics
Real-time 3D reconstruction is crucial for robotics and augmented reality, yet current simultaneous localization and mapping(SLAM) approaches often struggle to maintain structural consistency and robust pose estimation in the presence of depth noise. This work introduces PointSLAM++, a novel RGB-D SLAM system that leverages a hierarchically constrained neural Gaussian representation to preserve structural relationships while generating Gaussian primitives for scene mapping. It also employs progressive pose optimization to mitigate depth sensor noise, significantly enhancing localization accuracy. Furthermore, it utilizes a dynamic neural representation graph that adjusts the distribution of Gaussian nodes based on local geometric complexity, enabling the map to adapt to intricate scene details in real time. This combination yields high-precision 3D mapping and photorealistic scene rendering. Experimental results show PointSLAM++ outperforms existing 3DGS-based SLAM methods in reconstruction accuracy and rendering quality, demonstrating its advantages for large-scale AR and robotics.
title PointSLAM++: Robust Dense Neural Gaussian Point Cloud-based SLAM
topic Computer Vision and Pattern Recognition
Graphics
Robotics
url https://arxiv.org/abs/2601.11617