A vehicle includes one or more cameras that capture a plurality of two-dimensional images of a three-dimensional object. A light detector and/or semantic classifier searches for light of the three-dimensional object within these images. A vehicle signal detection module fuses information from the light detector and/or the semantic classifier to produce a semantic meaning for the light. The vehicle may be controlled based on the semantic meaning. Further, the vehicle may include a depth sensor and an object projector. The object projector may determine a region of interest within the two-dimensional image based on the depth sensor. A light detector and/or semantic classifier may use these regions of interest to efficiently perform a search for light.
一种载具包括捕获三维对象的多个二维图像的一个或多个相机。光检测器和/或语义分类器在这些图像内搜索三维对象的光。载具信号检测模块融合来自光检测器和/或语义分类器的信息以产生针对光的语义含义。可以基于语义含义来控制载具。此外,载具可以包括深度传感器和对象投影仪。对象投影仪可以基于深度传感器来确定二维图像内的感兴趣区域。光检测器和/或语义分类器可以使用这些感兴趣区域来高效地执行对光的搜索。
Multi-modal, multi-technology vehicle signal detection
多模态、多技术载具信号检测
2022-09-30
Patent
Electronic Resource
Chinese