Research map / 3D vision / Face and shape
GUAVA: Generalizable Upper Body 3D Gaussian Avatar
Dongbin Zhang, Yunfei Liu, Lijian Lin, Ye Zhu, Yang Li, Minghan Qin, Yu Li, Haoqian Wang
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 neural refiner (ablation) | GUAVA · Self-reenactment (ablation) | LPIPS: 0.1060 -> 0.0813 | Improved |
| add neural refiner (ablation) | GUAVA · Self-reenactment (ablation) | PSNR: 24.93 -> 25.87 | Improved |
| add inverse texture mapping (ablation) | GUAVA · Self-reenactment (ablation) | LPIPS: 0.0864 -> 0.0813 | Improved |
| add inverse texture mapping (ablation) | GUAVA · Self-reenactment (ablation) | PSNR: 25.65 -> 25.87 | Improved |
| add UV Gaussians (ablation) | GUAVA · Self-reenactment (ablation) | LPIPS: 0.0877 -> 0.0813 | Improved |
| add UV Gaussians (ablation) | GUAVA · Self-reenactment (ablation) | PSNR: 25.82 -> 25.87 | No clear effect |
| use EHM instead of SMPLX (ablation) | GUAVA · Self-reenactment (ablation) | PSNR: 25.60 -> 25.87 | Improved |
| use EHM instead of SMPLX (ablation) | GUAVA · Self-reenactment (ablation) | LPIPS: 0.0846 -> 0.0813 | Improved |
| single-image feed-forward GUAVA vs 2D diffusion MimicMotion | GUAVA · Self-reenactment (vs 2D methods) | FPS: 0.21 -> 52.21 | Improved |
| single-image feed-forward GUAVA vs per-ID trained GART | GUAVA · Self-reenactment (vs 3D methods) | PSNR: 24.46 -> 25.70 | Improved |
| use GUAVA instead of MimicMotion for identity | GUAVA · Cross-reenactment | IPS: 0.1310 -> 0.5554 | Improved |
Later papers that compared with this method
3 later accepted papers report GUAVA as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Bringing Your Portrait to 3D Presence (CVPR 2026 · better on 6 of 6 shared results)
- SplatCtrlA: Generalizable Single Image to Fully Controllable 3D Avatar (ECCV 2026 · better on 6 of 6 shared results)
- FlexiAvatar: Unified 3D Gaussian Human Avatars Under Arbitrary Body Visibility (ECCV 2026 · better on 2 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- Champ — Champ: Controllable and Consistent Human Image Animation with 3D Parametric Guidance (ECCV 2024) · better on 4 of 4 shared results
- GART — GART: Gaussian Articulated Template Models (CVPR 2024) · better on 2 of 2 shared results
- GaussianAvatar — GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians (CVPR 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Expressive Gaussian Human Avatars from Monocular RGB Video (NeurIPS 2024 · similarity 0.89)
- RGBAvatar: Reduced Gaussian Blendshapes for Online Modeling of Head Avatars (CVPR 2025 · similarity 0.88)
- FRESA: Feedforward Reconstruction of Personalized Skinned Avatars from Few Images (CVPR 2025 · similarity 0.88)
- Generalizable and Animatable Gaussian Head Avatar (NeurIPS 2024 · similarity 0.88)
- Gaussian Head & Shoulders: High Fidelity Neural Upper Body Avatars with Anchor Gaussian Guided Texture Warping (ICLR 2025 · similarity 0.88)
- AniGS: Animatable Gaussian Avatar from a Single Image with Inconsistent Gaussian Reconstruction (CVPR 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.
