A 3D point cloud reconstruction method based on deep learning

Abstract

The invention discloses a three-dimensional point cloud reconstruction method based on deep learning, which includes: a point cloud predictor predicts coordinate information of a three-dimensional point cloud according to an input potential feature representation; Representation is used as input to learn complementary features that combine information from other branches; by applying cross-view interaction units, each sparse point cloud reconstruction subnet captures cross-view complementary information, which is fed back to the point cloud predictor to generate sparse point clouds; A global-guided dense point cloud reconstruction module composed of a point cloud feature extraction subnet, a globally guided feature learning subnet and a generation layer, each point cloud feature extraction subnet is composed of a series of multi-layer perceptrons with shared weights. The described multilayer perceptron extracts point cloud features from the generated sparse point cloud; takes the chamfering distance loss as the geometric consistency constraint, and constructs the semantic consistency constraint to optimize the generation of dense point clouds.

Description

technical field [0001] The invention relates to the field of three-dimensional point cloud reconstruction, in particular to a three-dimensional point cloud reconstruction method based on deep learning. Background technique [0002] As one of the hot research tasks in the field of computer vision, the goal of 3D model reconstruction is to generate a real 3D model of the object contained in the 2D image by the information of the given 2D image. 3D models are represented in several ways, including: voxels, meshes, and 3D point clouds. As a typical representative of 3D model, 3D point cloud has been used in many fields such as autonomous driving and virtual reality. Therefore, the task of 3D point cloud reconstruction has attracted extensive attention of researchers. Furthermore, the quality of the generated point cloud models can significantly affect the performance of subsequent tasks, such as 3D model retrieval, classification, and segmentation. However, due to the irregul...

AI Technical Summary

Problems solved by technology

[0005] Although the existing 3D model reconstruction methods based on deep learning can predict a reasonable 3D shape from a limited number of input views, these methods usually directly generate relatively sparse and rough 3D point clouds.

Images