Research and application of point cloud completion technology
Point cloud data contains a large amount of spatial information,which is acquired by LiDAR,3D sensors and other devices,and is widely used in computer vision fields such as autonomous driving,3D reconstruction,and medical image process-ing.However,due to factors such as limited sensor field of view and object occlusion,there are missing parts in the acquired point cloud data.Point cloud completion aims to infer the missing parts from incomplete point cloud data and restore the complete point cloud data.Based on this,we classify deep learning-based point cloud completion methods,analyze the advantages and disadvan-tages of the models,describe several commonly used public datasets and evaluation criteria in the field of point cloud completion,and discuss the future research direction of point cloud completion technology.
point cloudpoint cloud completioncomputer visiondeep learning
万方数据
