Towards a benchmark for RGB-D SLAM evaluation

By Jürgen Sturm, Stéphane Magnenat, Nikolas Engelhard, François Pomerleau, Francis Colas, Wolfram Burgard, Daniel Cremers, Roland Yves Siegwart. 2011

141 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

RGB-D datasets using microsoft kinect or similar sensors: a survey (2016, background): “Six research publications about evaluating ego-motion estimation and SLAM over TUM Benchmark dataset are [21, 38, 45, 84, 86, 87].”
A benchmark for RGB-D visual odometry, 3D reconstruction and SLAM (2014, methodology): “We evaluate 5 different odometers; (i) DVO [13] which performs full frame alignment between pairs of RGB-D frames making use of a robust cost function to improve robustness to outliers; (ii) FOVIS [14] which uses FAST feature correspondences between pairs of RGB-D frames to resolve camera motion, re”
A benchmark for the evaluation of RGB-D SLAM systems (2012, unknown): “We verified that the Kinect and the motion capture system do not interfere: The motion capture LEDs appear as dim lamps in the Kinect infrared image with no influence on the produced depth maps, while the projector of the Kinect is not detected at all by the motion capture cameras.”
ElasticFusion: Dense SLAM Without A Pose Graph (2015, methodology): “We use the absolute trajectory (ATE) root-mean-square error metric (RMSE) in our comparison, which measures the root-mean-square of the Euclidean distances between all estimated camera poses and the ground truth poses associated by timestamp [22].”
Evaluating Egomotion and Structure-from-Motion Approaches Using the TUM RGB-D Benchmark (2012, background): “…Attribution license (CC-BY 3.0) at http://vision.in.tum.de/data/datasets/ rgbd-dataset Since the launch of the benchmark website in July 2011 [6], six peer-reviewed scientiﬁc publications (plus one master’s thesis) have appeared in which our benchmark has been used to evaluate egomotion…”
An evaluation of the RGB-D SLAM system (2012, methodology): “WE EVALUATED OUR SYSTEM ON A LARGE SET OF SEQUENCES FROM THE RGB-D SLAM DATASET [29].”
Real-time visual odometry from dense RGB-D images (2011, methodology): “To evaluate our algorithm quantitatively, we use the publicly available RGB-D dataset by Sturm et al. [10].”
3D Object Recognition in Cluttered Scenes with Local Surface Features: A Survey (2014, background): “14 RGB-D SLAM [55] Depth Image Kinect videos o, c Available”
Are we ready for autonomous driving? The KITTI vision benchmark suite (2012, unknown): “Visual Odometry / SLAM setting #sequences length #frames resolution ground truth metric TUM RGB-D [43] indoor 27 0.”
ElasticFusion: Real-time dense SLAM and light source estimation (2016, methodology): “To evaluate the trajectory estimation performance of our approach we test our system on the RGB-D benchmark of Sturm et al. (2012).”
The role of RGB-D benchmark datasets: an overview (2013, methodology): “The annotated point clouds are then simply reprojected into the input stream using the camera pose for the Kinect sensor at each frame.”