TY - GEN
T1 - LSD-SLAM
T2 - 13th European Conference on Computer Vision, ECCV 2014
AU - Engel, Jakob
AU - Schöps, Thomas
AU - Cremers, Daniel
PY - 2014
Y1 - 2014
N2 - 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.
AB - 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.
UR - http://www.scopus.com/inward/record.url?scp=84906489259&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-10605-2_54
DO - 10.1007/978-3-319-10605-2_54
M3 - Conference contribution
AN - SCOPUS:84906489259
SN - 9783319106045
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 834
EP - 849
BT - Computer Vision, ECCV 2014 - 13th European Conference, Proceedings
PB - Springer Verlag
Y2 - 6 September 2014 through 12 September 2014
ER -