The National Science Foundation (NSF) awarded a $166,289 Project Grant to the University of Massachusetts (UMass) under the Engineering program (CFDA 47.041) to develop a novel single-ear biosensing system for continuous facial and eye movement tracking and 3D facial reconstruction. The research aims to enable camera-free facial landmark detection and 3D facial animation through cross-modal transfer learning between visual and biosignal data. Key project objectives include designing low-power facial muscle activity sensing circuits, developing compression algorithms to optimize power and computational efficiency, and integrating inertial sensors to enhance the robustness of the 3D facial reconstruction models. The research team plans to build demonstration applications in areas like mobile virtual/augmented reality and speech enhancement. This award supports UMass's expertise in advancing wearable biosensing and transfer learning techniques to overcome limitations of existing camera-based facial analysis solutions. The project period runs from October 1, 2023 through January 31, 2025.