3.6 地图优化 Map optimization
The map, consisting of a set of keyframes and tracked -constraints, is continuously optimized in the background using pose graph optimization [18].
地图是由关键帧(作为节点)和帧间约束关系(相连接构成的边)组成,并且持续在(SLAM)后端通过姿态图优化进行修正。
The error function that is minimized is – in accordance with the left-multiplication convention from Sec. 2.2 – defined by ( defining the world frame)
误差函数使用左乘规则(参见2.2小节),进行最小化(优化),这个误差函数定义如下:见公式21.( 定义为世界坐标系):
image copy right belongs to engel14eccv paper, 图像摘录自 engel14eccv论文
Fig. 7: Loop closure for a long and challenging outdoor trajectory (after the loop closure on the left, before on the right).
图示7:(顶部)显示的是难度较大的室外长距离SLAM(大)闭环(相机)轨迹示图(其中左边是闭环之后的图示,右边是闭环之前的图示)。
Also shown are three selected close-ups of the generated pointcloud, and semi-dense depth maps for selected keyframes.
(中间)显示的是三个(不同区域的半稠密)局部点云地图特写,以及(底部是对应的)逆深度在关键帧图像上的显示。