Visual SLAM: Why filter?

By Hauke Strasdat, José M. M. Montiel, Andrew J. Davison. Image and Vision Computing, 2012

487 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/72683e699f7e33ad43062ba41d77a9068bebb85d

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

Simultaneous Localization and Mapping: A Survey of Current Trends in Autonomous Driving (2017, unknown): “This comparison effort has been proposed in [83] and then extended in [84] for monocular approaches.”
Differentiable SLAM-net: Learning Particle SLAM for Visual Navigation (2021, methodology): “Optimization-based approaches apply bundle adjustment on a set of keyframes and local maps; and they are popular for both visual [53; 47; 46; 38] and lidar-based SLAM [28].”
Visual SLAM algorithms: a survey from 2010 to 2016 (2017, methodology): “3 Comparison between MonoSLAM and PTAM The difference between the EKF based mapping in MonoSLAM and the BA-based mapping with keyframes in PTAM was discussed in the literature [34].”
ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras (2016, background): “Most modern stereo SLAM systems are keyframe based [7] and perform BA optimization in a local area to achieve scalability.”
Geometry-Aware Learning of Maps for Camera Localization (2017, background): “In visual SLAM, 3D landmarks (with feature descriptors of the corresponding image patches) are often defined as maps in both Bayesian filtering approaches [14, 41, 53] and key-frame based approaches [34, 42, 49].”
A constant-time SLAM back-end in the continuum between global mapping and submapping: application to visual stereo SLAM (2016, methodology): “A thorough comparison between ﬁltering and BA solutions is presented in the works (Strasdat et al. 2010) and (Strasdat et al. 2012), ultimately advocating for the application of BA as it outperforms ﬁltering methods in terms of accuracy per unit of computation time.”