Method for generating object information and electronic device

Abstract

The present application relates to a method and an electronic device for generating object information. The method comprises: generating a global map comprising a point cloud based on a plurality of local frame data comprising the point cloud; generating a three-dimensional object map comprising three-dimensional semantic information on a global coordinate system based on two-dimensional semantic segmentation information corresponding to key frame data selected by filtering the local frame data; and generating object information based on the three-dimensional semantic information included in the three-dimensional object map.