Object three-dimensional reconstruction method and system

Abstract

The invention provides an object three-dimensional reconstruction method and system, and the method comprises the steps: enabling a processor to receive original three-dimensional point cloud data, and determining whether a first point cloud point corresponding to the first point cloud data is in an interested region corresponding to preset interested region data according to the first point clouddata and the preset interested region data; when the first point cloud point is in the region of interest, storing the first point cloud data in the three-dimensional point cloud data of the object;when the first point cloud point is not in the region of interest, determining first mirror surface equation data, performing reflection transformation processing according to the first point cloud data and the first mirror surface equation data to obtain first reflection point cloud data, and when it is determined that the first reflection point cloud point is in the region of interest, generating second reflection point cloud data; and when the first reflection point cloud data, the second reflection point cloud data and the camera center position data determine that the first reflection point cloud data are reliable point cloud data, storing the first reflection point cloud data in the object three-dimensional point cloud data.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method and system for three-dimensional reconstruction of an object. Background technique [0002] In recent years, with the development of computer technology and the rapid development of data processing and other fields, 3D modeling of targets (eg, objects of interest) from image data is a popular computer vision topic. Reconstruction of 3D models is an important process in many fields such as computer animation, medical imaging, computer graphics, etc. [0003] Reconstruction was performed using a series of images taken at different positions according to the camera. An obvious disadvantage of this approach is that the object of interest must be static. Such systems are not well suited to work on dynamic objects and require prior knowledge such as the rotational speed of the turntable. Another approach is to perform shape extraction from silhouette with the support of multi...

AI Technical Summary

Problems solved by technology

Another approach is to perform shape extraction from silhouette with the support of multiple cameras to retrieve the visual hull of the object, but this method lacks concave areas in the reconstructed model

There are also solutions using a system of multiple color cameras and depth sensors. The main disadvantage of this approach is the need for a synchronous protocol. When working on moving objects, sometimes each camera and sensor must be connected to a unique computer, which can obtain good Accuracy, but expensive equipment and unnecessary resource redundancy in processing

Also, using multiple depth cameras can cause severe IR interference

Images