A 3D face reconstruction method based on a single image

Abstract

The invention discloses a three-dimensional human face reconstruction method based on a single picture. The method comprises the steps of 1) performing human face reconstruction (face reconstruction)based on a FLAME model; 2) performing FLAME grid face detail enhancement (detail enhancement); and 3) performing FLAME grid chartlet complementation (chartlet complementation). According to the main processes, an input human face image is given; based on an algorithm, firstly, by using human face feature points extracted from the image and in combination with three-dimensional feature points of aFLAME grid, feature point matching energy is established, and the shape of a human face is obtained; secondly, grid vertexes of a human face height field are guided to move by means of high-frequencyinformation of a human face area in the image, and fine face details are reconstructed; and finally, a human face chartlet is complemented through a built FLAME albedo parameterization model, and illumination information is removed. The method solves the problem of three-dimensional human face reconstruction for the single picture, and can be applied to the human face reconstruction and the humanface animation production.

Description

technical field [0001] The invention relates to the technical field of computer graphics and three-dimensional animation production, in particular to a method for reconstructing a three-dimensional human face based on a single picture. Background technique [0002] Human beings are not only the subject of various activities, but also the object of many applications, so most virtual reality applications are inevitably human-centered. The reconstruction of human surface shape and its motion plays a key role in such applications. From daily life, game animation, education and training to industrial design, it is increasingly important to construct highly realistic personalized virtual human static models and dynamic geometry. sex. For example, in a virtual fitting, in addition to body shape and motion information, the fitting person's facial expressions, fluttering hair, etc. will have an impact on the experience. In many virtual reality applications, the realistic appearance...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problem of insufficient completeness of current face reconstruction work, the present invention proposes a three-dimensional face reconstruction method based on a single picture, which can reconstruct a relatively good face model according to the face image, with better face details , and the number of grid points is small, which is conducive to the application in real-time environment

Images