Laplace cutout matrix method by using moving least square method

Abstract

The invention discloses a Laplace cutout matrix method by using a moving least square method. The Laplace cutout matrix method by using the moving least square method uses a KNN neighbourhood to replace a space neighborhood, a linear relationship of non-neighborhood pixels on an alpha chart is acquired, the moving least square method is used for replacing a least square method, a moving Laplace matrix is calculated, and an alpha chart is obtained. The cutout Laplace matrix calculation method is improved, the moving least square method is used for replacing the least square method to deduce the moving Laplace matrix, and compared with the least square method, the moving least square method solution linear condition is more accurate. The KNN neighbourhood is used for replacing the space neighborhood, the Laplace matrix can reflect the relationship of alpha values of the non-neighborhood pixels, and a more accurate cutout Laplace matrix calculation method is obtained.

Description

technical field [0001] The invention belongs to the technical field of interactive matting, and in particular relates to a Laplacian matting matrix method using a moving least square method. Background technique [0002] The prospect of interactive matting technology to extract images under limited user interaction is widely used in image and video editing, 3D reconstruction and other fields, and has extremely high application value. In the matting technology in recent years, the Laplace matrix gives the linear relationship between the pixels on the alpha map, which plays an important role in the estimation of the alpha map. Interactive matting is to calculate the alpha map of the foreground under limited user interaction, so as to separate the foreground from the background. The input of the matting problem is the original image I and the tripartite map provided by the user, and the output is the alpha map, the foreground F, and the background B. Therefore, it is a typical...

AI Technical Summary

Problems solved by technology

[0004] The object of the present invention is to provide a kind of Laplacian matting matrix method using moving least squares method, aiming to solve the existing problem in the prior art that the Laplace matrix cannot reflect the relationship between pixels in non-neighborhoods; In the area where the foreground and background components change suddenly, the Laplace matrix is ​​difficult to obtain the desired effect

Images