Three-dimensional reconstruction method and device

Abstract

The invention provides a three-dimensional reconstruction method and device, and the method comprises the steps: calculating the first direction information of each first data point in a generated point cloud, and calculating the second direction information of each second data point in a real point cloud; calculating a direction difference between each first data point and each second data point; calculating the distance difference between each first data point and each second data point; carrying out fusion processing to obtain a directional chamfer distance loss function; and training the reconstruction network through the chamfering distance loss function with the direction, and performing three-dimensional reconstruction on the to-be-constructed object through the trained reconstruction network. Therefore, when the difference between the generated point cloud and the real point cloud is measured, not only the distance difference is considered, but also the direction difference is added, so that the difference between the generated point cloud and the real point cloud can be measured more accurately, the generated point cloud is more helpful to approach the real point cloud in a network training process, and the quality of three-dimensional point cloud reconstruction is improved.

Description

technical field [0001] The invention relates to the technical field of point clouds, in particular to a three-dimensional reconstruction method and a three-dimensional reconstruction device. Background technique [0002] With the development of science and technology and the progress of society, 3D reconstruction technology plays an indispensable role in industry and life, such as unmanned driving, VR / AR, face recognition, etc. At the same time, they have more and more requirements for the quality of 3D reconstruction. higher. In recent years, affected by the rapid development of deep learning technology, the research on 3D object reconstruction methods based on deep learning has developed rapidly, which has indeed improved the surface quality of 3D object reconstruction. Although learning-based 3D representations are diverse, such as voxels, point clouds, patch meshes, etc. However, the network framework of the 3D reconstruction method based on learning is similar. It for...

AI Technical Summary

Problems solved by technology

The CD loss function only considers the distance difference between points when measuring the difference between two point clouds, resulting in poor quality of 3D reconstruction

Images