Research map / 3D vision / Face and shape
FlexiAvatar: Unified 3D Gaussian Human Avatars Under Arbitrary Body Visibility
Yihalem Yimolal Tiruneh, Muhammad Salman Ali, Uyoung Jeong, Muneeb Khan, MD Sayem, ALLANUR BAYRAMGELDIYEV, Binod Bhattarai, Seungryul Baek
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 part-specific residual refinement heads for face and hands (ablation) | FlexiAvatar · NeuMan | PSNR: 34.90 -> 35.77 | Improved |
| add diffusion-generated auxiliary views (ablation) | FlexiAvatar · NeuMan | PSNR: 35.29 -> 35.77 | Improved |
| add diffusion-generated auxiliary views (ablation) | FlexiAvatar · NeuMan | LPIPS x100: 0.92 -> 0.83 | Improved |
| add visibility-aware optimization (ablation) | FlexiAvatar · INSTA head-only | PSNR: 32.15 -> 33.04 | Improved |
| add visibility-aware optimization (ablation) | FlexiAvatar · INSTA head-only | LPIPS x100: 5.80 -> 5.63 | Improved |
| add occlusion-robust confidence-masked SMPL-X optimization (ablation) | FlexiAvatar · INSTA head-only | PSNR: 31.07 -> 33.04 | Improved |
| add visibility-aware pruning of Gaussians (head-only) | FlexiAvatar vs ExAvatar · INSTA head-only | number of Gaussians: 167,390 -> 84,095 | Improved |
| add visibility-aware pruning of Gaussians (upper-body) | FlexiAvatar vs ExAvatar · TalkShow upper-body | FPS: 31.06 -> 33.87 | Improved |
| lower synthetic-view loss weight from 1.0 to 0.1 (ablation) | FlexiAvatar · NeuMan | PSNR: 33.59 -> 35.77 | Improved |
| let residual module also refine Gaussian mean positions (ablation) | FlexiAvatar · NeuMan | PSNR: 35.77 -> 35.14 | Hurt |
| extend residual refinement to whole body (ablation) | FlexiAvatar · NeuMan | PSNR: 35.77 -> 35.78 | No clear effect |
| use Otsu automatic threshold instead of fixed 0.400 threshold (ablation) | FlexiAvatar · INSTA Biden head-only | PSNR: 33.52 -> 33.50 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 6 of 6 on most shared results.
- GUAVA — GUAVA: Generalizable Upper Body 3D Gaussian Avatar (ICCV 2025) · better on 2 of 3 shared results
- RGBAvatar — RGBAvatar: Reduced Gaussian Blendshapes for Online Modeling of Head Avatars (CVPR 2025) · better on 1 of 1 shared result
- Vid2AvatarPro — Vid2Avatar-Pro: Authentic Avatar from Videos in the Wild via Universal Prior (CVPR 2025) · better on 1 of 1 shared result
- GauHuman — GauHuman: Articulated Gaussian Splatting from Monocular Human Videos (CVPR 2024) · better on 1 of 1 shared result
- GaussianAvatar — GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians (CVPR 2024) · better on 1 of 1 shared result
- Vid2Avatar — Vid2Avatar: 3D Avatar Reconstruction From Videos in the Wild via Self-Supervised Scene Decomposition (CVPR 2023) · better on 1 of 1 shared result
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
- GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians (CVPR 2024 · similarity 0.90)
- Vid2Avatar-Pro: Authentic Avatar from Videos in the Wild via Universal Prior (CVPR 2025 · similarity 0.90)
- MPMAvatar: Learning 3D Gaussian Avatars with Accurate and Robust Physics-Based Dynamics (NeurIPS 2025 · similarity 0.90)
- Real-time High-fidelity Gaussian Human Avatars with Position-based Interpolation of Spatially Distributed MLPs (CVPR 2025 · similarity 0.90)
- 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting (CVPR 2024 · similarity 0.89)
- FRESA: Feedforward Reconstruction of Personalized Skinned Avatars from Few Images (CVPR 2025 · similarity 0.89)
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.
