LSD-SLAM: Large-Scale Direct monocular SLAM

Jakob Engel, Thomas Schöps, Daniel Cremers

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

2472 Zitate (Scopus)


We propose a direct (feature-less) monocular SLAM algorithm which, in contrast to current state-of-the-art regarding direct methods, allows to build large-scale, consistent maps of the environment. Along with highly accurate pose estimation based on direct image alignment, the 3D environment is reconstructed in real-time as pose-graph of keyframes with associated semi-dense depth maps. These are obtained by filtering over a large number of pixelwise small-baseline stereo comparisons. The explicitly scale-drift aware formulation allows the approach to operate on challenging sequences including large variations in scene scale. Major enablers are two key novelties: (1) a novel direct tracking method which operates on , thereby explicitly detecting scale-drift, and (2) an elegant probabilistic solution to include the effect of noisy depth values into tracking. The resulting direct monocular SLAM system runs in real-time on a CPU.

TitelComputer Vision, ECCV 2014 - 13th European Conference, Proceedings
Herausgeber (Verlag)Springer Verlag
AuflagePART 2
ISBN (Print)9783319106045
PublikationsstatusVeröffentlicht - 2014
Veranstaltung13th European Conference on Computer Vision, ECCV 2014 - Zurich, Schweiz
Dauer: 6 Sept. 201412 Sept. 2014


NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NummerPART 2
Band8690 LNCS
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349


Konferenz13th European Conference on Computer Vision, ECCV 2014


Untersuchen Sie die Forschungsthemen von „LSD-SLAM: Large-Scale Direct monocular SLAM“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren