Method, system and device for three-dimensional reconstruction of monocular image based on reference plane

Abstract

The invention relates to the measurement field, in particular to a method, a system and a device for three-dimensional reconstruction of a monocular image based on a reference plane. The method for three-dimensional reconstruction of a monocular image comprise that following step of: capturing a two-dimensional image; dividing the two-dimensional image to obtain a plurality of reference planes ina corresponding three-dimensional space; completing three-dimensional reconstruction corresponding to the two-dimensional image according to a plurality of reference planes; extracting projection points of all objects in the two-dimensional image on their corresponding reference planes, and completing three-dimensional reconstruction of all objects. The invention provides a method, a system and adevice for three-dimensional reconstruction of a monocular image based on a reference plane, which firstly creates a reference plane by utilizing the three-dimensional reconstruction ability of the monocular to the plane, and then maps each object point in the image to each simulation plane based on the reference plane to obtain the depth of each object point, thereby completing three-dimensionalreconstruction of the whole image. The structure is simple, easy to implement and easy to use.

Description

technical field [0001] The present invention relates to the field of measurement, in particular to a method, system and device for three-dimensional reconstruction of a monocular image based on a reference plane. Background technique [0002] Image 3D reconstruction is used in various fields. Monocular vision has a simple structure and is easy to apply. Without the help of known standard objects, it can only perform three-dimensional reconstruction of objects on a single plane specified in the image. Binocular stereo vision simulates the function of human eyes and completes 3D reconstruction through parallax. Compared with monocular vision, it can perform 3D reconstruction of all objects in the image, but binocular stereo vision has complex structure, difficult precise calibration process, and corresponding point matching errors. The biggest disadvantage is that in scenes with sparse feature points on the surface of objects, it is difficult to obtain accurate shapes to comp...

AI Technical Summary

Problems solved by technology

Binocular stereo vision simulates the function of human eyes and completes 3D reconstruction through parallax. Compared with monocular vision, it can perform 3D reconstruction of all objects in the image, but binocular stereo vision has complex structure, difficult precise calibration process, and corresponding point matching errors. The big disadvantage is that in scenes where the feature points on the surface of the object are sparse, it is difficult to obtain an accurate shape to complete the 3D reconstruction

Structured light cameras need to cooperate with core components such as laser projectors, optical diffraction elements, and infrared cameras to capture diffuse infrared speckle through infrared cameras and calculate the depth information of each point. bear

Images