Search papers, labs, and topics across Lattice.
5
0
6
3
Achieving state-of-the-art lip-sync accuracy while maintaining natural expressiveness, ReFree-S2V redefines the standards for speech-driven character animation.
Recovering muscle deformations from surface observations could revolutionize how we create realistic virtual humans, making high-fidelity animations accessible and efficient.
Imagine telepresence where your avatar convincingly blends into any environment, relit in real-time based on the scene's actual lighting, all from a single headset.
AI-generated videos can now respect physics, thanks to a framework that uses a physical simulator to guide diffusion models, resulting in more realistic and coherent motion.
Finally, realistic and expressive full-body gestures for conversational agents can be generated in real-time, thanks to a novel causal framework that models both temporal dynamics and part-level motion hierarchy.
