Research map / Video / Video generation
GGTalker: Talking Head Systhesis with Generalizable Gaussian Priors and Identity-Specific Adaptation
Wentao Hu, Shunkai Li, Ziqiao Peng, Haoxian Zhang, Fan Shi, Xiaoqiang Liu, Pengfei Wan, Di Zhang, Hui Tian
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Audio-Expression priors (train from scratch) (ablation) | GGTalker · Self-reenactment (talking head) | LPIPS: 0.0306 -> 0.0281 | Improved |
| add Audio-Expression priors (train from scratch) (ablation) | GGTalker · Self-reenactment (talking head) | LMD: 2.741 -> 2.328 | Improved |
| add Audio-Expression priors (train from scratch) (ablation) | GGTalker · Cross-language reenactment | LSE-C: 3.268 -> 5.769 | Improved |
| add Audio-Expression fine-tuning (ablation) | GGTalker · Self-reenactment (talking head) | LPIPS: 0.0385 -> 0.0281 | Improved |
| add Audio-Expression fine-tuning (ablation) | GGTalker · Self-reenactment (talking head) | LMD: 3.287 -> 2.328 | Improved |
| add Audio-Expression fine-tuning (ablation) | GGTalker · Cross-language reenactment | LSE-C: 4.780 -> 5.769 | Improved |
| add Expression-Visual priors (ablation) | GGTalker · Self-reenactment (talking head) | LPIPS: 0.0438 -> 0.0281 | Improved |
| add Expression-Visual priors (ablation) | GGTalker · Self-reenactment (talking head) | LMD: 3.826 -> 2.328 | Improved |
| add Expression-Visual priors (ablation) | GGTalker · Cross-language reenactment | LSE-C: 4.682 -> 5.769 | Improved |
| add Expression-Visual fine-tuning (ablation) | GGTalker · Self-reenactment (talking head) | LPIPS: 0.0473 -> 0.0281 | Improved |
| add Expression-Visual fine-tuning (ablation) | GGTalker · Self-reenactment (talking head) | LMD: 2.925 -> 2.328 | Improved |
| add Expression-Visual fine-tuning (ablation) | GGTalker · Cross-language reenactment | LSE-C: 4.524 -> 5.769 | Improved |
Closest papers by meaning
- GoHD: Gaze-oriented and Highly Disentangled Portrait Animation with Rhythmic Poses and Realistic Expressions (AAAI 2025 · similarity 0.88)
- PointTalk: Audio-Driven Dynamic Lip Point Cloud for 3D Gaussian-based Talking Head Synthesis (AAAI 2025 · similarity 0.88)
- Hierarchically Controlled Deformable 3D Gaussians for Talking Head Synthesis (AAAI 2025 · similarity 0.87)
- FaceTalk: Audio-Driven Motion Diffusion for Neural Parametric Head Models (CVPR 2024 · similarity 0.87)
- Monocular and Generalizable Gaussian Talking Head Animation (CVPR 2025 · similarity 0.87)
- InsTaG: Learning Personalized 3D Talking Head from Few-Second Video (CVPR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
