Indoor scene three-dimensional reconstruction method based on single depth vision sensor

Abstract

The invention relates to an indoor scene three-dimensional reconstruction method based on a single depth vision sensor. The method is technically characterized by including the following steps of firstly, continuously scanning a whole indoor scene through the single depth vision sensor; secondly, conducting preprocessing including denoising, hole repairing and the like on collected depth data to obtain smooth depth data; thirdly, calculating point cloud data corresponding to the current depth frame according to the depth data collected in the second step; fourthly, conducting registration on point cloud obtained through different viewpoint depth frames to obtain complete point cloud of the indoor scene; fifthly, conducting plane fitting, achieving segmentation of the special point cloud, and establishing an independent and complete three-dimensional model of each object in the indoor scene. Scanning devices used by the method are simple; scanned data information is comprehensive, and the point cloud registration accuracy calculation efficiency is effectively improved; finally, a complete and high-quality three-dimensional model set with a geographic structure and a color map can be established for the indoor scene.

Description

technical field [0001] The invention belongs to the technical field of three-dimensional reconstruction of indoor scenes, in particular to a method for three-dimensional reconstruction of indoor scenes based on a single depth vision sensor. Background technique [0002] Building a high-quality 3D model of an indoor scene, especially creating an independent 3D model for each object in the room is a very challenging task. At present, many 3D reconstruction methods for indoor scenes focus on reconstructing local models in indoor scenes, resulting in the method itself: easy to lose details of many indoor scenes, requiring cumbersome user interaction operations, and requiring large-scale reconstruction such as laser scanners. Disadvantages such as expensive hardware equipment. [0003] Commercial depth cameras can realize 3D model reconstruction of objects in the scene, but building a 3D model of an indoor scene is different from building a 3D model of a single object. Within th...

AI Technical Summary

Problems solved by technology

However, the early kinectfusion system uses a planar voxel grid, which can only be used in a small volume range; although the visual odometry algorithm can improve the accuracy of the kinectfusion system, it can only be used in a relatively narrow space and compared with the camera motion trajectory In simple cases, complex camera motion paths will affect the 3D model reconstruction effect of the entire indoor scene

Images