Three-dimensional reconstruction method of scene, terminal and readable storage medium

Abstract

The embodiment of the invention relates to the field of computer vision, and discloses a three-dimensional reconstruction method of a scene, a terminal and a readable storage medium. The three-dimensional reconstruction method of the scene comprises the steps: after it is detected that new image data is added to a first storage space, screening data corresponding to each frame of image data in thefirst storage space is acquired, wherein the screening data is used for representing the importance degree of the corresponding image data; after the screened data meets a first data migration condition corresponding to the cache storage space, migrating the image data to the cache storage space; and performing three-dimensional reconstruction on the target scene according to the image data in the first storage space after data migration. According to the embodiment, the three-dimensional reconstruction speed of the scene can be improved, and the scene range of three-dimensional reconstruction is expanded.

Description

technical field [0001] The embodiments of the present invention relate to the field of computer vision, and in particular to a method for three-dimensional reconstruction of a scene, a terminal and a readable storage medium. Background technique [0002] When a robot performs operations such as navigation and obstacle avoidance, it needs to fully understand the scene it is in. Three-dimensional reconstruction of the scene is one of the core technologies for the robot to fully understand the scene it is in. [0003] The current 3D reconstruction method is expressed through voxels. This method does not need to estimate the topological structure of the object surface, but only uses the corresponding depth information to perform a high-precision 3D reconstruction. For example: the data structure of the Kinect fusion system is to use a conventional 3D grid to evenly divide it into a group of voxels, and map it to a predefined physical dimension to achieve a description within a c...

AI Technical Summary

Problems solved by technology

[0004] The inventors found that there are at least the following problems in related technologies: currently, as the scope of the scene to be reconstructed increases, the amount of reconstruction data involved continues to increase, resulting in slower and slower 3D reconstruction speed, and real-time reconstruction of large scenes cannot be realized , such as: a room-wide scene

Images