Depth map optimization method and device for large-scale scene reconstruction and storage medium

Abstract

The invention discloses a depth map optimization method for large-scale scene reconstruction, and relates to the technical field of image processing, and the method comprises the steps: carrying out the preprocessing of N multi-view images, and determining the depth estimation range of each multi-view image; selecting one of the N multi-view images as a reference image and M images as source images, and determining an initial depth map and an initial depth confidence map of the reference image according to the depth estimation range of the reference image and the depth estimation range of the source images; calculating a confidence mask value of a pixel point in the reference image according to the initial depth confidence map of the reference image; according to the confidence mask value, optimizing the depth value of each pixel point in the initial depth map of the reference image, and determining an optimized depth map; and filtering the optimized depth map, and determining the depth map, thereby solving the problem that the fusion degree of the texture region is not high, realizing that the depth map of the weak texture region can fill the missing position, and effectively reconstructing the large outdoor scene.

Description

Technical field [0001] The present application relates to image processing technology, and particularly relates to a depth map optimization method, apparatus, and storage medium for large-scale reconstruction of the scene. Background technique [0002] Depth map based on the traditional method of fusion in some environments with relatively high accuracy is obtained, but the realization of the process due to the severe environment dependent texture photographs to calculate the consistency, texture in weak physical scene depth fusion when a high degree of integration is not a problem may occur, for example, can not meet the demand for precision in the building and integration of a large number of glass material having a surface, the depth map traditional fusion method. [0003] Currently, if the texture of how to rebuild the region, is needed to solve the problem. Inventive content [0004] Depth map optimization method embodiment of the present embodiment by providing a large-sca...

AI Technical Summary

Problems solved by technology

[0004] The embodiment of the present application provides a depth map optimization method for large-scale scene reconstruction, which solves the problem in the prior art that the fusion degree of weak texture regions is not high, and realizes that the depth map of weak texture regions can correct missing Filling in places, effectively reconstructing large outdoor scenes

Images