Stereo 3D Gaussian Splatting SLAM for Outdoor Urban Scenes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Xiaohan, Gong, Ziren, Tosi, Fabio, Poggi, Matteo, Mattoccia, Stefano, Liu, Dong, Wu, Jun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915420139683840
author Li, Xiaohan
Gong, Ziren
Tosi, Fabio
Poggi, Matteo
Mattoccia, Stefano
Liu, Dong
Wu, Jun
author_facet Li, Xiaohan
Gong, Ziren
Tosi, Fabio
Poggi, Matteo
Mattoccia, Stefano
Liu, Dong
Wu, Jun
contents 3D Gaussian Splatting (3DGS) has recently gained popularity in SLAM applications due to its fast rendering and high-fidelity representation. However, existing 3DGS-SLAM systems have predominantly focused on indoor environments and relied on active depth sensors, leaving a gap for large-scale outdoor applications. We present BGS-SLAM, the first binocular 3D Gaussian Splatting SLAM system designed for outdoor scenarios. Our approach uses only RGB stereo pairs without requiring LiDAR or active sensors. BGS-SLAM leverages depth estimates from pre-trained deep stereo networks to guide 3D Gaussian optimization with a multi-loss strategy enhancing both geometric consistency and visual quality. Experiments on multiple datasets demonstrate that BGS-SLAM achieves superior tracking accuracy and mapping performance compared to other 3DGS-based solutions in complex outdoor environments.
format Preprint
id arxiv_https___arxiv_org_abs_2507_23677
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Stereo 3D Gaussian Splatting SLAM for Outdoor Urban Scenes
Li, Xiaohan
Gong, Ziren
Tosi, Fabio
Poggi, Matteo
Mattoccia, Stefano
Liu, Dong
Wu, Jun
Robotics
3D Gaussian Splatting (3DGS) has recently gained popularity in SLAM applications due to its fast rendering and high-fidelity representation. However, existing 3DGS-SLAM systems have predominantly focused on indoor environments and relied on active depth sensors, leaving a gap for large-scale outdoor applications. We present BGS-SLAM, the first binocular 3D Gaussian Splatting SLAM system designed for outdoor scenarios. Our approach uses only RGB stereo pairs without requiring LiDAR or active sensors. BGS-SLAM leverages depth estimates from pre-trained deep stereo networks to guide 3D Gaussian optimization with a multi-loss strategy enhancing both geometric consistency and visual quality. Experiments on multiple datasets demonstrate that BGS-SLAM achieves superior tracking accuracy and mapping performance compared to other 3DGS-based solutions in complex outdoor environments.
title Stereo 3D Gaussian Splatting SLAM for Outdoor Urban Scenes
topic Robotics
url https://arxiv.org/abs/2507.23677