3 3 Semantic segmentation system in 3D point cloud and semantic segmentation method in 3D point cloud using the same

摘要

The present invention relates to a system for semantic segmentation of a three-dimensional space and a method for semantic segmentation of a three-dimensional space using the same, wherein the system for semantic segmentation of the three-dimensional space includes a point cloud including at least one object. An input module to be input, a local feature information extraction unit for extracting local feature information from the point cloud input from the input module, and a global feature from the point cloud from which the local feature information is extracted and a global feature information extracting unit for extracting information, and a segmentation result extracting unit for generating segmentation result information on the object in the point cloud based on the local feature information and global feature information. The semantic segmentation system of the three-dimensional space according to the present invention and the semantic segmentation method of the three-dimensional space using the same are based on the PointNet architecture and use PointSIFT, so that the semantic segmentation of the three-dimensional point cloud is used. It has the advantage of improving the accuracy and processing speed of the work.