论文阅读 [TPAMI-2022] RGB-D SLAM in Dynamic Environments Using Point Correlations

论文阅读 [TPAMI-2022] RGB-D SLAM in Dynamic Environments Using Point Correlations

论文搜索(studyai.com)

搜索论文: RGB-D SLAM in Dynamic Environments Using Point Correlations

搜索论文: http://www.studyai.com/search/whole-site/?q=RGB-D+SLAM+in+Dynamic+Environments+Using+Point+Correlations

关键字(Keywords)

Dynamics; Correlation; Simultaneous localization and mapping; Robustness; Motion estimation; Cameras; Motion segmentation; SLAM; motion estimation; dynamic environments

机器视觉

运动捕捉; 运动估计; 视觉SLAM; RGBD数据

摘要(Abstract)

In this paper, a simultaneous localization and mapping (SLAM) method that eliminates the influence of moving objects in dynamic environments is proposed.

本文提出了一种消除动态环境中运动目标影响的同步定位与映射(SLAM)方法。.

This method utilizes the correlation between map points to separate points that are part of the static scene and points that are part of different moving objects into different groups.

该方法利用贴图点之间的相关性,将静态场景中的点和不同移动对象中的点分成不同的组。.

A sparse graph is first created using Delaunay triangulation from all map points.

首先从所有地图点使用Delaunay三角剖分创建稀疏图。.

In this graph, the vertices represent map points, and each edge represents the correlation between adjacent points.

在该图中,顶点表示贴图点,每条边表示相邻点之间的相关性。.

If the relative position between two points remains consistent over time, there is correlation between them, and they are considered to be moving together rigidly.

如果两点之间的相对位置随时间保持一致,则它们之间存在相关性,并且被认为是刚性地一起移动。.

If not, they are considered to have no correlation and to be in separate groups.

如果没有,则认为它们没有相关性,属于不同的组。.

After the edges between the uncorrelated points are removed during point-correlation optimization, the remaining graph separates the map points of the moving objects from the map points of the static scene.

在点相关优化过程中去除不相关点之间的边后,剩余的图将移动对象的贴图点与静态场景的贴图点分离。.

The largest group is assumed to be the group of reliable static map points.

假设最大的组是可靠的静态地图点组。.

Finally, motion estimation is performed using only these points.

最后,仅使用这些点执行运动估计。.

The proposed method was implemented for RGB-D sensors, evaluated with a public RGB-D benchmark, and tested in several additional challenging environments.

该方法已在RGB-D传感器上实现,使用公共RGB-D基准进行了评估,并在其他一些具有挑战性的环境中进行了测试。.

The experimental results demonstrate that robust and accurate performance can be achieved by the proposed SLAM method in both slightly and highly dynamic environments.

实验结果表明,在轻微和高度动态的环境中,所提出的SLAM方法都可以实现鲁棒性和精确性。.

Compared with other state-of-the-art methods, the proposed method can provide competitive accuracy with good real-time performance…

与其他最先进的方法相比,该方法可以提供具有竞争力的精度和良好的实时性能。。.

作者(Authors)

[‘Weichen Dai’, ‘Yu Zhang’, ‘Ping Li’, ‘Zheng Fang’, ‘Sebastian Scherer’]

你可能感兴趣的:(人工智能,计算机视觉,CVPR,深度学习,机器学习)