Point cloud generation method, system and computer storage medium

Abstract

The present invention provides a point cloud generation method, a system and a computer storage medium. The method includes initializing spatial parameters of each pixel in a reference image (S101); updating the spatial parameters of each pixel in the reference image by using the propagation of adjacent pixels (S102); comparing whether the spatial parameters of each pixel have changed before and after the spread, if there is a change, classifying each pixel according to the difference between the depth and the score before and after the spread, and changing the spatial parameter of each pixelwithin a predetermined range according to the different classification situations (S103); calculating the score of the spatial parameters of each pixel after the change, and updating the spatial parameters of at least part of the pixel to the spatial parameters whose score reaches the preset threshold range (S104); according to the updated spatial parameters of each pixel in the image, determiningthe depth map corresponding to the reference image (S105); generating a dense point cloud image according to the depth map corresponding to the reference image (S106). Through the method, the calculation speed is increased and the depth map and dense point cloud image noise are reduced.

Description

technical field [0001] The present invention generally relates to the field of information technology, and more specifically relates to a point cloud generation method, system and computer storage medium. Background technique [0002] Densifying the sparse point cloud is an important link in the 3D reconstruction algorithm. It inherits the sparse point cloud generated in the previous link. After obtaining the accurate spatial pose between images, it encrypts the sparse point cloud to restore the details of the scene. , which is of great significance for the subsequent generation of a complete grid structure. The mainstream 3D reconstruction densification algorithm includes the local block matching (patchmatch) algorithm, which is a multi-view stereo matching algorithm, and calculates the depth map corresponding to the reference image through the reference image and multiple neighboring images. In the traditional patchmatch algorithm, a variety of random values ​​are combine...

AI Technical Summary

Problems solved by technology

In the traditional patchmatch algorithm, a variety of random values ​​are combined into alternative update combinations, which requires a large amount of calculation; to save memory resources, the original image is usually down-sampled by half, and then the depth map of the same size is calculated, but Many details in the original image are lost as a result, resulting in a lower quality depth map

Images