An end-to-end method that produces smooth, identity-preserving animation on arbitrary 3D face meshes from in-the-wild video.

A data-free method that generalizes realistic, lip-synced 3D speech animations for arbitrary characters using video diffusion models.

A method that incorporates interactions between joints, frames, and words for text-driven motion generation and zero-shot editing.

A method that enables direct retargeting between two facial meshes with different shapes and mesh structures.

Simulation tool that leverages Generative AI to assist in staging musical scenes by simulating stage direction and design.

A rendering technique for generating ink-painting style edge images based on relative depth differences between object boundaries.

Edge-computing cloud system that automates tasks like entry/exit tracking and unauthorized parking detection.

Development of customizable AI avatars that generate responsive, conversational facial animations, enabling lifelike interaction and feedback through intuitive user inputs.