Information-Driven Direct RGB-D Odometry

Fontan, Alejandro; Civera, Javier; Triebel, Rudolph

doi:10.1109/cvpr42600.2020.00498

Cited by 29 publications

(30 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So that we choose these representative systems, according to the methods and attributes of them, and enumerate their attribute in Table 1. We have also introduced 3 relatively advanced RGB-D SLAM systems, include two direct-based tracking methods: RGBDTAM [13], ID-RGBDO [14] and one indirect-based methods Plane-Edge-SLAM [18], and we have also introduced their tracking method in the survey part. Due to the lack of open source code or implementation difficulties, we did not conduct experiments and only refer to the experimental data provided in their articles.…”

Section: Discussionmentioning

confidence: 99%

“…While most direct methods take as many pixels as possible to ensure the accuracy, Alejandro Fontán et al proposed ID-RGBDO in 2020 [14], which takes only 24 points for tracking and can achieve an accuracy similar to the state of the art (some times out perform it). Based on an information matrix, ID-RGBDO selects only the most informative points in the local Bundle Adjustment and pose tracking optimizations.…”

Section: ) Id-rgbdomentioning

confidence: 99%

See 1 more Smart Citation

Survey and Evaluation of RGB-D SLAM

2021

View full text Add to dashboard Cite

The traditional visual SLAM systems take the monocular or stereo camera as input sensor, with complex map initialization and map point triangulation steps needed for 3D map reconstruction, which are easy to fail, computationally complex and can cause noisy measurements. The emergence of RGB-D camera which provides RGB image together with depth information breaks this situation. While a number of RGB-D SLAM systems have been proposed in recent years, the current classification research on RGB-D SLAM is very lacking, and their advantages and shortcomings remain unclear regarding different applications and perturbations, such as illumination transformation, noise and rolling shutter effect of sensors. In this paper, we mainly introduced the basic concept and structure of the RGB-D SLAM system, and then introduced the differences between the various RGB-D SLAM systems in the three aspects of tracking, mapping, and loop detection, and we make a classification study on different RGB-D SLAM algorithms according to the three aspect. Furthermore, we discuss some advanced topics and open problems of RGB-D SLAM, hoping that it will help for future exploring. In the end, we conducted a large number of evaluation experiments on multiple RGB-D SLAM systems, and analyzed their advantages and disadvantages, as well as performance differences in different application scenarios, and provided references for researchers and developers.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: ) Id-rgbdomentioning

confidence: 99%

Survey and Evaluation of RGB-D SLAM

2021

View full text Add to dashboard Cite

show abstract

“…Motion-based feature selection methods have been widely applied in visual SLAM [16], [25]- [27], and many of them are based on the covariance or information matrices that capture uncertainties of poses [28]. These methods formulate the feature selection as a submatrix selection problem and aim to find a subset of features with the objective to preserve the information matrix's spectral attributes.…”

Section: B Feature Selectionmentioning

confidence: 99%

Greedy-Based Feature Selection for Efficient LiDAR SLAM

Jiao¹,

Zhu²,

Ye³

et al. 2021

Preprint

View full text Add to dashboard Cite

Modern LiDAR-SLAM (L-SLAM) systems have shown excellent results in large-scale, real-world scenarios. However, they commonly have a high latency due to the expensive data association and nonlinear optimization. This paper demonstrates that actively selecting a subset of features significantly improves both the accuracy and efficiency of an L-SLAM system. We formulate the feature selection as a combinatorial optimization problem under a cardinality constraint to preserve the information matrix's spectral attributes. The stochasticgreedy algorithm is applied to approximate the optimal results in real-time. To avoid ill-conditioned estimation, we also propose a general strategy to evaluate the environment's degeneracy and modify the feature number online. The proposed feature selector is integrated into a multi-LiDAR SLAM system. We validate this enhanced system with extensive experiments covering various scenarios on two sensor setups and computation platforms. We show that our approach exhibits low localization error and speedup compared to the state-of-the-art L-SLAM systems. To benefit the community, we have released the source code: https://ram-lab.com/file/site/m-loam.

show abstract

“…The uncertainty so obtained has been extensively studied in photometric computer vision [11]. It has also found applicability in a wide variety of tasks, such as RGB-D SLAM [10], radial distortion model selection [42] in SFM, skeletal images selection for efficient SFM [51], height map fusion [79], 3D reconstruction [17], and camera calibration [39]. Polic et al recently make efforts [41,40] towards efficient uncertainty computation in large-scale 3D reconstruction.…”

Section: Related Workmentioning

confidence: 99%

Fusing the Old with the New: Learning Relative Camera Pose with Geometry-Guided Uncertainty

Zhuang¹,

Chandraker²

2021

Preprint

View full text Add to dashboard Cite

Learning methods for relative camera pose estimation have been developed largely in isolation from classical geometric approaches. The question of how to integrate predictions from deep neural networks (DNNs) and solutions from geometric solvers, such as the 5-point algorithm [37], has as yet remained under-explored. In this paper, we present a novel framework that involves probabilistic fusion between the two families of predictions during network training, with a view to leveraging their complementary benefits in a learnable way. The fusion is achieved by learning the DNN uncertainty under explicit guidance by the geometric uncertainty, thereby learning to take into account the geometric solution in relation to the DNN prediction. Our network features a self-attention graph neural network, which drives the learning by enforcing strong interactions between different correspondences and potentially modeling complex relationships between points. We propose motion parmeterizations suitable for learning and show that our method achieves state-of-the-art performance on the challenging DeMoN [61] and ScanNet [8] datasets. While we focus on relative pose, we envision that our pipeline is broadly applicable for fusing classical geometry and deep learning.

show abstract

Information-Driven Direct RGB-D Odometry

Cited by 29 publications

References 27 publications

Survey and Evaluation of RGB-D SLAM

Survey and Evaluation of RGB-D SLAM

Greedy-Based Feature Selection for Efficient LiDAR SLAM

Fusing the Old with the New: Learning Relative Camera Pose with Geometry-Guided Uncertainty

Contact Info

Product

Resources

About