A Method of Object Reconstruction Based on Geometric Constraints

Abstract

The invention relates to a method for object reconstruction based on geometric constraints, which belongs to the field of computer vision, including obtaining initial point clouds by the structure of motion recovery (SFM) method; image feature point clustering to obtain the classification results of feature points, that is, similar parts in the image Neighborhood relationship; for the normal feature clustering of the initial point cloud, the geometric structure of the initial point cloud is defined by using the correspondence between the image feature point classification results and the initial point cloud clustering results; the initial point cloud is obtained from the geometric structure For the sparser part, define this part as a "hole", and then use the combined structural constraints of the "hole" area to fit the space plane and surface through the RANSAC method and the least square method; sample the fitted surface, Add the obtained 3D points to the initial point cloud to obtain a dense point cloud model, and finally use Poisson surface reconstruction to obtain the 3D model of the target. The experimental results of the present invention verify the practicability of the method and achieve better results.

Description

technical field [0001] The invention relates to a method for object reconstruction based on geometric constraints, which uses the geometric structure constraints existing in the point cloud of the object's three-dimensional space to perform three-dimensional reconstruction on the object; the method solves the lack of reconstruction results caused by sparse texture in traditional algorithms Part of the repair problem further improves the reconstruction effect of smoother or overexposed surfaces. The invention belongs to the field of computer vision. Background technique [0002] Image-based 3D reconstruction has always been a research hotspot in the direction of computer vision. In 1983, Martin proposed the method of using the outline of the object in the image to reconstruct, and then in 1986 and 1987, Chien and Potsmesil respectively proposed the use of orthogonal A method for projecting and extracting an object model and a method for constructing an object model using mult...

AI Technical Summary

Problems solved by technology

The present invention effectively improves the limitations of the existing methods that easily generate holes for missing texture parts, and combines the geometric constraints of image neighborhood information corresponding to three-dimensional points, and achieves better experimental results

Images