VIGS-SLAM: Visual Inertial Gaussian Splatting SLAM

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhu, Zihan, Zhang, Wei, Li, Moyang, Haala, Norbert, Pollefeys, Marc, Barath, Daniel
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915858301845504
author Zhu, Zihan
Zhang, Wei
Li, Moyang
Haala, Norbert
Pollefeys, Marc
Barath, Daniel
author_facet Zhu, Zihan
Zhang, Wei
Li, Moyang
Haala, Norbert
Pollefeys, Marc
Barath, Daniel
contents We present VIGS-SLAM, a visual-inertial 3D Gaussian Splatting SLAM system that achieves robust real-time tracking and high-fidelity reconstruction. Although recent 3DGS-based SLAM methods achieve dense and photorealistic mapping, their purely visual design degrades under challenging conditions such as motion blur, low texture, and exposure variations. Our method tightly couples visual and inertial cues within a unified optimization framework, jointly optimizing camera poses, depths, and IMU states. It features robust IMU initialization, time-varying bias modeling, and loop closure with consistent Gaussian updates. Experiments on five challenging datasets demonstrate our superiority over state-of-the-art methods. Project page: https://vigs-slam.github.io
format Preprint
id arxiv_https___arxiv_org_abs_2512_02293
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle VIGS-SLAM: Visual Inertial Gaussian Splatting SLAM
Zhu, Zihan
Zhang, Wei
Li, Moyang
Haala, Norbert
Pollefeys, Marc
Barath, Daniel
Robotics
Computer Vision and Pattern Recognition
We present VIGS-SLAM, a visual-inertial 3D Gaussian Splatting SLAM system that achieves robust real-time tracking and high-fidelity reconstruction. Although recent 3DGS-based SLAM methods achieve dense and photorealistic mapping, their purely visual design degrades under challenging conditions such as motion blur, low texture, and exposure variations. Our method tightly couples visual and inertial cues within a unified optimization framework, jointly optimizing camera poses, depths, and IMU states. It features robust IMU initialization, time-varying bias modeling, and loop closure with consistent Gaussian updates. Experiments on five challenging datasets demonstrate our superiority over state-of-the-art methods. Project page: https://vigs-slam.github.io
title VIGS-SLAM: Visual Inertial Gaussian Splatting SLAM
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.02293