Method for generating facial animation from single image
AssigneesZhejiang University, Faceunity Technology Co., Ltd.
InventorsKun ZHOU, Jiahao GENG
Life of this patent
Estimating the term…
Abstract
A method for generating a facial animation from a single image is provided. The method is mainly divided into four steps: generation of facial feature points in an image, global two-dimensional deformation of the image, optimization of details of a facial area, and generation of texture of an oral cavity area. The present disclosure can generate a facial animation in real time according to a change of the facial feature points, and the animation quality reaches a level of current state-of-art facial image animation technology. The present disclosure can be used in a series of applications, such as facial image editing, portrait animation generation based on a single image, and facial expression editing in videos.
Claims5
A method for generating a facial animation from a single image, comprising following steps:
- (1) generation of facial feature points of the image:
- calculating feature points of a facial area in the image and feature points of a background area in the image;
- (2) global two-dimensional deformation of the image:
- based on initial feature points obtained in the step 1 and changes of the feature points specified by a user or a program, generating, through global two-dimensional deformation, a deformation image that conforms to constraint of the feature points;
- (3) optimization of details of the facial area:
- optimizing texture of the facial area in the deformation image through generative adversarial neural network, the facial area not comprising an oral cavity area;
- (4) generation of texture of the oral cavity area:
- synthesizing texture of the oral cavity area through a generative adversarial neural network, and generating a final facial animation image.
Dependencies are inferred from the claim text.