Face animation synthesis method and system

Abstract

The invention discloses a face animation synthesis method and system, and belongs to the technical field of artificial intelligence, and the method comprises the steps: obtaining a face sketch sequence and a previous frame of face image; performing feature extraction on the previous frame of face image and the face sketch sequence, and fusing the extracted features to obtain a fused feature map; on the basis of the fused feature map, predicting to obtain an intermediate image, an optical flow between video frames and a mask; and synthesizing a face video frame based on the intermediate image, the optical flow between the video frames, the mask and the face image of the previous frame. A local-global face collaborative face animation synthesis method is adopted, and high-reality-sense face animation synthesis is achieved.

Description

technical field [0001] The invention relates to the technical field of artificial intelligence, in particular to a method and system for synthesizing human face animation. Background technique [0002] With the explosive development of artificial intelligence, high-realistic facial animation synthesis, as an emerging research hotspot, has opened up a new model of virtual human intelligence and is widely used in film and television, entertainment, education, medical care, social networking, e-commerce, and content marketing. , artistic creation, scientific research and many other fields, its important academic value and application prospects have attracted widespread attention from academia and industry. However, synthesizing high-realistic facial animation is extremely challenging. It is necessary to ensure the synchronization of lip motion and speech / text input, the coordination of global motion of the face area, and the coordination of local motion of different parts of th...

AI Technical Summary

Problems solved by technology

However, synthesizing high-realistic facial animation is extremely challenging. It is necessary to ensure the synchronization of lip motion and speech / text input, the coordination of global motion of the face area, and the coordination of local motion of different parts of the face.

[0003] However, the existing synthesis methods often treat this task as a single image synthesis problem and ignore the global deformation between adjacent frames, or only focus on lip synchronization and ignore the local synergy between different regions of the face, resulting in synthetic Face animation timing is choppy, with noticeable video judder issues, and texture compositing resulting in non-coordinated, stiff faces

Images