ElasticFusion: Dense SLAM without a pose graph

Thomas Whelan, Stefan Leutenegger, Renato F. Salas-Moreno, Ben Glocker, Andrew J. Davison

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

615 Scopus citations

Abstract

We present a novel approach to real-time dense visual SLAM. Our system is capable of capturing comprehensive dense globally consistent surfel-based maps of room scale environments explored using an RGB-D camera in an incremental online fashion, without pose graph optimisation or any post-processing steps. This is accomplished by using dense frame-to-model camera tracking and windowed surfel-based fusion coupled with frequent model refinement through non-rigid surface deformations. Our approach applies local model-to-model surface loop closure optimisations as often as possible to stay close to the mode of the map distribution, while utilising global loop closure to recover from arbitrary drift and maintain global consistency.

Original languageEnglish
Title of host publicationRobotics
Subtitle of host publicationScience and Systems XI, RSS 2015
EditorsJonas Buchli, David Hsu, Lydia E. Kavraki
PublisherMIT Press Journals
ISBN (Electronic)9780992374716
DOIs
StatePublished - 2015
Externally publishedYes
Event2015 Robotics: Science and Systems Conference, RSS 2015 - Rome, Italy
Duration: 13 Jul 201517 Jul 2015

Publication series

NameRobotics: Science and Systems
Volume11
ISSN (Electronic)2330-765X

Conference

Conference2015 Robotics: Science and Systems Conference, RSS 2015
Country/TerritoryItaly
CityRome
Period13/07/1517/07/15

Fingerprint

Dive into the research topics of 'ElasticFusion: Dense SLAM without a pose graph'. Together they form a unique fingerprint.

Cite this