3D Face Dynamic Expression Synthesis System Based on DFFD
Hu Ni, Jingen Liu · 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) · 2019
Synthesis of highly realistic facial expressions with speech communication function is one of the most challenging problems in computer graphics. In recent years, virtual characters have been widely used in the game industry, e-commerce, film and television production, medical treatment, military and other fields. How to build realistic 3d face model and synthesize real facial expression is the main problem of 3d facial expression animation. In this paper, a 3d face dynamic expression synthesis algorithm driven by DFFD data is proposed. By collecting real facial expressions to drive a universal neutral 3d virtual face model, a realistic 3d facial expression animation consistent with the facial movements of real performers is synthesized. In addition, the 3d virtual face synthesized by this method presents the consistency between visual, auditory and the content to be expressed in the process of speech.