3D mesh optimization based on contour recovery and least-squares vertex movement allocation

In order to solve the problems of edge and plane passivation and inaccurate distribution of pixel power to vertex in 3D mesh optimization using image matching best surface estimation method, a 3D mesh method based on contour restoration and least square vertex movement is proposed to optimize the initial mesh. Firstly, large scale simplification and subdivision method are combined, the global contour of the level of detail in mesh is restored, and then the vertex gradient accumulation of the least square method is used to allocate the moving amount of the energy of the object in the triangle to the three vertices more accurately. Finally, the smoothing term based on curvature is used to replace the homogenized Laplacian smoothing. The mesh shape is retained while the matching noise is filtered to achieve more refined mesh optimization. After the experimental verification, the feasibility of this method is proved.