Implementation of Synchronized and Coordinated Motion in 3D Facial Animation Based on Weighted Dirichlet Algorithm

Abstract

With the continuous advancement of Virtual Reality (VR) technology, people's requirements for human-computer interaction are becoming increasingly high. To achieve realistic facial expression animation in VR human-computer interaction, a method based on weighted Dirichlet algorithm is proposed to effectively synchronize and coordinate complex facial expressions and mouth shape changes in 3D facial animation. This method achieves natural transitions between different facial expressions and mouth shapes by applying weighted processing to key facial points and combining Dirichlet energy minimization technology. It also combines a mouth shape prediction model based on a Transformer deep learning architecture to achieve speech-driven lip animation synchronization and coordination. The findings denoted that the predicted animation trajectory of the weighted Dirichlet algorithm highly overlapped with the actual motion trajectory on the corners of the mouth and middle of the lips, with maximum errors reduced to 3.0mm and 1.8mm, respectively. In addition, the errors in the lip and nasolabial groove areas were reduced by 46.49% and 54.12%, respectively. In summary, the 3D facial animation synchronized and coordinated motion based on the weighted Dirichlet algorithm effectively achieves precise simulation and coordinated motion of facial animation actions, significantly improving the performance of 3D facial animation.