MG-SLAM: Structure Gaussian Splatting SLAM with Manhattan World Hypothesis

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Shuhong, Deng, Tianchen, Zhou, Heng, Li, Liuzhuozheng, Wang, Hongyu, Wang, Danwei, Li, Mingrui
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917195256168448
author Liu, Shuhong
Deng, Tianchen
Zhou, Heng
Li, Liuzhuozheng
Wang, Hongyu
Wang, Danwei
Li, Mingrui
author_facet Liu, Shuhong
Deng, Tianchen
Zhou, Heng
Li, Liuzhuozheng
Wang, Hongyu
Wang, Danwei
Li, Mingrui
contents Gaussian Splatting SLAMs have made significant advancements in improving the efficiency and fidelity of real-time reconstructions. However, these systems often encounter incomplete reconstructions in complex indoor environments, characterized by substantial holes due to unobserved geometry caused by obstacles or limited view angles. To address this challenge, we present Manhattan Gaussian SLAM, an RGB-D system that leverages the Manhattan World hypothesis to enhance geometric accuracy and completeness. By seamlessly integrating fused line segments derived from structured scenes, our method ensures robust tracking in textureless indoor areas. Moreover, The extracted lines and planar surface assumption allow strategic interpolation of new Gaussians in regions of missing geometry, enabling efficient scene completion. Extensive experiments conducted on both synthetic and real-world scenes demonstrate that these advancements enable our method to achieve state-of-the-art performance, marking a substantial improvement in the capabilities of Gaussian SLAM systems.
format Preprint
id arxiv_https___arxiv_org_abs_2405_20031
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MG-SLAM: Structure Gaussian Splatting SLAM with Manhattan World Hypothesis
Liu, Shuhong
Deng, Tianchen
Zhou, Heng
Li, Liuzhuozheng
Wang, Hongyu
Wang, Danwei
Li, Mingrui
Robotics
Computer Vision and Pattern Recognition
Gaussian Splatting SLAMs have made significant advancements in improving the efficiency and fidelity of real-time reconstructions. However, these systems often encounter incomplete reconstructions in complex indoor environments, characterized by substantial holes due to unobserved geometry caused by obstacles or limited view angles. To address this challenge, we present Manhattan Gaussian SLAM, an RGB-D system that leverages the Manhattan World hypothesis to enhance geometric accuracy and completeness. By seamlessly integrating fused line segments derived from structured scenes, our method ensures robust tracking in textureless indoor areas. Moreover, The extracted lines and planar surface assumption allow strategic interpolation of new Gaussians in regions of missing geometry, enabling efficient scene completion. Extensive experiments conducted on both synthetic and real-world scenes demonstrate that these advancements enable our method to achieve state-of-the-art performance, marking a substantial improvement in the capabilities of Gaussian SLAM systems.
title MG-SLAM: Structure Gaussian Splatting SLAM with Manhattan World Hypothesis
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2405.20031