Beneath the Skin: Bringing Digital Faces to Life with Intuitive Physics

Zhiyun Peng1, Michael Tao2, Eugene Fiume1
1Simon Fraser University, Canada
2Augmenta, United States
Contact information for the corresponding author: zhiyun_peng [at] sfu [dot] ca
Intended audience: Academics outside my discipline. Friends and family members. Animators.
[Submitted, under review]
Abstract

Have you ever noticed when a digital character’s face just looks... wrong? Human eyes are highly sensitive to fake expressions. Creating realistic digital faces is a massive challenge: traditional animation offers artistic freedom but often lacks physical realism, while true physical simulations are incredibly complex, mathematically heavy, and painfully slow to compute. Our research bridges this gap. We have developed a digital "musculoskeletal" system that takes the heavy lifting out of biomechanics. Instead of wrestling with dense equations and long rendering times, animators and researchers can intuitively "play" with our model like a digital puppet. By automatically translating simple artistic controls into complex, underlying muscle movements, our tool allows creators to generate highly realistic, physically accurate facial expressions—including speech. We are putting the power of advanced physics directly into the hands of artists.

Keywords: Digital Puppeteering, Facial Animation, Biomechanics, Physics, Computer Graphics.
Reference
[1] Zhao Q, Long P, Zhang Q, Qin D, Liang H, Zhang L, Zhang Y, Yu J, Xu L. Media2face: Co-speech facial animation generation with multi-modality guidance. In ACM SIGGRAPH 2024 conference papers 2024 Jul 13 (pp. 1-13).