Improving the Agility of Keyframe-Based SLAM

By Georg S. W. Klein, David William Murray. European Conference on Computer Vision, 2008

404 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/b5b8dd48d7300e700292af9503dfa325c2710fc5

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Robust monocular SLAM in dynamic environments (2013, methodology): “To improve the agility of PTAM, they also add edge features to the map to enhance the feature tracking under fast motion, and an inter-frame rotation estimator to aid tracking under rapid panning [18].”
Real-time monocular SLAM: Why filter? (2010, methodology): “Now with much more processing power available, it is not surprising, when looking at our accuracy/cost analysis, that the most accurate real-time monocular SLAM system available [14] is based on bundle adjusting a sparse number of keyframes.”
Visual SLAM algorithms: a survey from 2010 to 2016 (2017, background): “Moreover, feature points and edgelets are combined to achieve robust estimation against to motion-blurred input images [42].”
Evaluation of Interest Point Detectors and Feature Descriptors for Visual Tracking (2011, background): “Further research explored the use of particle filters (Eade and Drummond 2006b), features at multiple scales (Chekhlov et al. 2006, 2007), and edge-based features (Eade and Drummond 2006a; Klein and Murray 2008 ), as well as recovery from tracking failure (Williams et al. 2007).”
SVO: Semidirect Visual Odometry for Monocular and Multicamera Systems (2017, background): “For incremental computation of the posterior, the authors of [55] show that (8) can be approximated by the product of a Gaussian distribution for the depth and a Beta distribution for the inlier ratio: where a k and b k are the parameters controlling the Beta distribution.”
PL-SVO: Semi-direct Monocular Visual Odometry by combining points and line segments (2016, methodology): “Solutions for the VO problem has been addressed employing diﬀerent sensors, such as monocular or stereo cameras [2] [3] [4], RGB-D cameras [5] [6], or a combination of any of them with an Inertial Measurement Unit (IMU) [7].”
DTAM: Dense tracking and mapping in real-time (2011, methodology): “A similar step is performed before feature matching in PTAM’s tracker, computing first the interframe 2D image transform and fitting a 3D rotation [7].”
ORB-SLAM: A Versatile and Accurate Monocular SLAM System (2015, methodology): “The original version was later improved with edge features, a rotation estimation step during tracking, and a better relocalization method [32].”
Learning to Navigate the Energy Landscape (2016, methodology): “We refer the interested reader to [15, 14, 24, 10, 48, 36, 47] and [41, 38] for relevant litterature on RGB camera relocalization and Hand Pose Estimation, respectively.”
Parallel Tracking and Mapping on a camera phone (2009, methodology): “Previous experience with the keyframe-based PTAM system [6, 7] on the PC has shown that it is able to generate room-sized maps and track a moving camera, while (crucially) offering a high level robustness to rapid motion, poor imaging conditions, and inevitable data association failures.”
Scene Coordinate Regression Forests for Camera Relocalization in RGB-D Images (2013, methodology): “The tiny-image baseline uses the parameters from [11, 17].”
Visual simultaneous localization and mapping: a survey (2012, background): “One way of alleviating this problem to some extent is by the use of keyframes (see “Appendix I”) (Mouragnon et al. 2006; Klein and Murray 2008).”
PL-SLAM: A Stereo SLAM System Through the Combination of Points and Line Segments (2017, methodology): “successful proposals, we can highlight FastSLAM [23], PTAM [24] [25], SVO [26] [10], and, more recently, ORB-SLAM [13], which relies on a fast and continuous tracking of ORB features [27], and a local bundle adjustment (LBA) step with the continuous observations of the point features.”
PL-SLAM: Real-time monocular visual SLAM with points and lines (2017, positive): “In [14] the accuracy was improved with edge features together with a rotation estimation step during tracking that provided better relocalization results, and even reducing the computational cost [24].”