An evaluation of the RGB-D SLAM system

Endres, Felix; Hess, Jürgen; Engelhard, Nikolas; Sturm, Jürgen; Cremers, Daniel; Burgard, Wolfram

doi:10.1109/icra.2012.6225199

Cited by 602 publications

(414 citation statements)

References 30 publications

Supporting

Mentioning

402

Contrasting

Unclassified

Order By: Relevance

“…For comparison we show respective results from semi-dense mono-VO [9], keypoint-based mono-SLAM [15], direct RGB-D SLAM [14] and keypointbased RGB-D SLAM [7]. Note that [14] and [7] use depth information from the sensor, while the others do not. (3) direct image alignment with and without ESM minimization (indicated by light / dark) for a different number of pyramid levels (color).…”

Section: Quantitative Evaluationmentioning

confidence: 99%

LSD-SLAM: Large-Scale Direct Monocular SLAM

Engel¹,

Schöps²,

Cremers³

2014

Computer Vision – ECCV 2014

Self Cite

2,537

1,904

View full text Add to dashboard Cite

Abstract. 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 sim(3), 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.

show abstract

Section: Quantitative Evaluationmentioning

confidence: 99%

LSD-SLAM: Large-Scale Direct Monocular SLAM

Engel¹,

Schöps²,

Cremers³

2014

Computer Vision – ECCV 2014

Self Cite

2,537

1,904

View full text Add to dashboard Cite

show abstract

“…The Kinect sensor is using structured light and a visual camera to build a scene, specifically a so-called point cloud, and in combination with the Simultaneous Localization and Mapping (SLAM) algorithm provides an estimate of the position and orientation of the sensor in this scene [4]. Fig.…”

Section: Methodsmentioning

confidence: 99%

Advanced Concepts in Multi-Dimensional Radiation Detection and Imaging

Vetter

Haefner

Barnowski

et al. 2016

Proceedings of International Symposium on Radiation Detectors and Their Uses (ISRD2016)

View full text Add to dashboard Cite

Recent developments in the detector fabrication, signal readout, and data processing enable new concepts in radiation detection that are relevant for applications ranging from fundamental physics to medicine as well as nuclear security and safety. We present recent progress in multi-dimensional radiation detection and imaging in the Berkeley Applied Nuclear Physics program. It is based on the ability to reconstruct scenes in three dimensions and fuse it with gamma-ray image information. We are using the High-Efficiency Multimode Imager HEMI in its Compton imaging mode and combining it with contextual sensors such as the Microsoft Kinect or visual cameras. This new concept of volumetric imaging or scene data fusion provides unprecedented capabilities in radiation detection and imaging relevant for the detection and mapping of radiological and nuclear materials. This concept brings us one step closer to the seeing the world with gamma-ray eyes.

show abstract

“…The whole map is kept consistent using pose graph optimization (PGO). The open source system RGBD-SLAM 1 described in [2] works in a similar fashion: It registers RGBD frames by matching 3D point pairs using various available image features but without the successive ICP refinement step. There are also some approaches that rely on depth images alone, e.g.…”

Section: Related Workmentioning

confidence: 99%

Efficient onbard RGBD-SLAM for autonomous MAVs

Scherer

Zell

2013

2013 IEEE/RSJ International Conference on Intelligent Robots and Systems

View full text Add to dashboard Cite

Abstract-We present a computationally inexpensive RGBD-SLAM solution taylored to the application on autonomous MAVs, which enables our MAV to fly in an unknown environment and create a map of its surroundings completely autonomously, with all computations running on its onboard computer. We achieve this by implementing efficient methods for both tracking its current location with respect to a heavily processed previously seen RGBD image (keyframe) and efficient relative registration of a set of keyframes using bundle adjustment with depth constraints as a front-end for pose graph optimization. We prove the accuracy and efficiency of our system based on a public benchmark dataset and demonstrate that the proposed method enables our quadrotor to fly autonomously.

show abstract

An evaluation of the RGB-D SLAM system

Cited by 602 publications

References 30 publications

LSD-SLAM: Large-Scale Direct Monocular SLAM

LSD-SLAM: Large-Scale Direct Monocular SLAM

Advanced Concepts in Multi-Dimensional Radiation Detection and Imaging

Efficient onbard RGBD-SLAM for autonomous MAVs

Contact Info

Product

Resources

About