Understanding while Exploring: Semantics-driven Active Mapping

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Liyan, Zhan, Huangying, Yin, Hairong, Xu, Yi, Mordohai, Philippos
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911262816862208
author Chen, Liyan
Zhan, Huangying
Yin, Hairong
Xu, Yi
Mordohai, Philippos
author_facet Chen, Liyan
Zhan, Huangying
Yin, Hairong
Xu, Yi
Mordohai, Philippos
contents Effective robotic autonomy in unknown environments demands proactive exploration and precise understanding of both geometry and semantics. In this paper, we propose ActiveSGM, an active semantic mapping framework designed to predict the informativeness of potential observations before execution. Built upon a 3D Gaussian Splatting (3DGS) mapping backbone, our approach employs semantic and geometric uncertainty quantification, coupled with a sparse semantic representation, to guide exploration. By enabling robots to strategically select the most beneficial viewpoints, ActiveSGM efficiently enhances mapping completeness, accuracy, and robustness to noisy semantic data, ultimately supporting more adaptive scene exploration. Our experiments on the Replica and Matterport3D datasets highlight the effectiveness of ActiveSGM in active semantic mapping tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2506_00225
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Understanding while Exploring: Semantics-driven Active Mapping
Chen, Liyan
Zhan, Huangying
Yin, Hairong
Xu, Yi
Mordohai, Philippos
Robotics
Computer Vision and Pattern Recognition
Effective robotic autonomy in unknown environments demands proactive exploration and precise understanding of both geometry and semantics. In this paper, we propose ActiveSGM, an active semantic mapping framework designed to predict the informativeness of potential observations before execution. Built upon a 3D Gaussian Splatting (3DGS) mapping backbone, our approach employs semantic and geometric uncertainty quantification, coupled with a sparse semantic representation, to guide exploration. By enabling robots to strategically select the most beneficial viewpoints, ActiveSGM efficiently enhances mapping completeness, accuracy, and robustness to noisy semantic data, ultimately supporting more adaptive scene exploration. Our experiments on the Replica and Matterport3D datasets highlight the effectiveness of ActiveSGM in active semantic mapping tasks.
title Understanding while Exploring: Semantics-driven Active Mapping
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.00225