KAIST's AnyTalk: video diffusion models generate 3D speech animation for arbitrary characters
KAIST · hf · 2026-08-18
KAIST presents AnyTalk, which generates 3D speech animations for arbitrary characters without any animation data.
The approach: adapt video diffusion models via character-specific fine-tuning to synthesize talking-head videos, then optimize blendshape parameters from the synthesized footage to obtain drivable 3D speech animation. A distilled real-time variant is also provided.
More from Multimodal
- Turn photos into 3D models using MeshyAI — filiksyos · 2026-08-18
- AI-Edited Photos Are Polluting Citizen Science: Gemini Gave a Heron a Third Leg — kscottz · 2026-08-18
- Talking avatars' tell has moved: jaw and expression coherence, not lip sync — admrys · 2026-08-18
- Miniature Monsoon Cloud Over Mumbai: Photorealistic AI Prompt Scenes — CurieuxExplorer · 2026-08-18
- Turn Static Images into Editable 3D Characters with Pure Three.js Code — Scobleizer · 2026-08-18
- Seedance 2.5 Generates Bioluminescent Waves in Hokusai Style — CurieuxExplorer · 2026-08-18