Research map / Video / Video generation
AudioAvatar: Personalized Audio-driven Whole-body Talking Avatars
Seungeun Lee, SeungJun Moon, Hah Min Lew, Ji-Su Kang, Gyeong-Moon Park
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-particle motion embedding (ablation) | AudioAvatar · Test set (30 subjects) | SyncC: 7.05 -> 7.20 | Improved |
| add patchified alignment strategy (ablation) | AudioAvatar · Test set (30 subjects) | FVD: 252 -> 240 | Improved |
| add hand and face refinement module (ablation) | AudioAvatar · Test set (30 subjects) | HKC: 0.860 -> 0.897 | Improved |
| add hybrid talking-video data synthesis (ablation) | AudioAvatar · Test set (30 subjects) | SyncC: 6.85 -> 7.20 | Improved |
| add video score distillation loss (ablation) | AudioAvatar · Test set (30 subjects) | FVD: 290 -> 240 | Improved |
| add trajectory alignment loss (ablation) | AudioAvatar · Test set (30 subjects) | SyncD: 6.20 -> 5.42 | Improved |
| drive avatar directly from audio instead of audio-to-pose converter with LHM/PERSONA | Gaussian avatars · Test set (30 subjects) | SyncD: 6.80 -> 5.42 (vs PERSONA) | Improved |
| use Gaussian avatar with distilled priors instead of video diffusion models | AudioAvatar vs HunyuanVideo-Avatar · Test set (30 subjects) | FVD: 320 -> 240 | Improved |
Closest papers by meaning
- From Audio to Photoreal Embodiment: Synthesizing Humans in Conversations (CVPR 2024 · similarity 0.91)
- TaoAvatar: Real-Time Lifelike Full-Body Talking Avatars for Augmented Reality via 3D Gaussian Splatting (CVPR 2025 · similarity 0.90)
- Teller: Real-Time Streaming Audio-Driven Portrait Animation with Autoregressive Motion Generation (CVPR 2025 · similarity 0.89)
- Towards High-fidelity 3D Talking Avatar with Personalized Dynamic Texture (CVPR 2025 · similarity 0.89)
- GoHD: Gaze-oriented and Highly Disentangled Portrait Animation with Rhythmic Poses and Realistic Expressions (AAAI 2025 · similarity 0.89)
- DAWN: Dynamic Frame Avatar with Non-autoregressive Diffusion Framework for Talking head Video Generation (ICLR 2025 · similarity 0.88)
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.
