There are also non-trivial engineering decisions on how to process the video stream.

在处理视频流方面，研究中也使用了一些重要的工程决策。

And so, I designed an algorithm that could sort of use both of those cues to process a stream from a single camera and make a map of a 3D environment that it could then use to avoid obstacles.

所以, 我设计了一种算法，它可以利用这两种线索来处理来自单个相机的信号，并绘制出一个3D环境的地图，这样它就可以用来躲避障碍物。