Skip to content
← Back to Research

preprint · arXiv (2026) · Jan 26, 2026

Splat-Portrait: Generalizing Talking Heads with Gaussian Splatting

Tong Shi, Melonie de Almeida, Daniela Ivanova, Nicolas Pugeault, Paul Henderson

In plain English

Builds a 3D talking head from a single portrait using Gaussian splatting, separating the head from the background and generating lip motion from audio without 3D supervision or facial landmarks.

talking head generation
gaussian splatting
lip sync
Read the source ↗

Related research