Research map / 3D vision / Face and shape
PERSONA: Personalized Whole-Body 3D Avatar with Pose-Driven Deformations from a Single Image
Geonhee Sim, Gyeongsik Moon
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 |
|---|---|---|---|
| model pose-driven deformations with mean offsets from MLPs (ablation) | PERSONA · NeuMan test | PSNR: 28.02 -> 29.20 | Improved |
| model pose-driven deformations (ablation) | PERSONA · X-Humans 00034 | PSNR: 26.32 -> 27.60 | Improved |
| model pose-driven deformations (ablation) | PERSONA · X-Humans, mask IoU | mask IoU %: 88.60 -> 90.06 | Improved |
| lower image loss weight from 1 to 0.1 in geometry-weighted optimization (ablation) | PERSONA · NeuMan test | PSNR: 28.18 -> 29.20 | Improved |
| lower image loss weight further from 0.1 to 0.01 (ablation) | PERSONA · NeuMan test | PSNR: 29.20 -> 29.00 | Hurt |
| use MimicMotion instead of Champ or StableAnimator for training videos (ablation) | PERSONA · NeuMan test | PSNR: 29.13 / 28.98 -> 29.20 | No clear effect |
| use Sapiens 1B instead of 0.3B geometry estimator (ablation) | PERSONA · NeuMan test | PSNR: 28.98 -> 29.20 | No clear effect |
| oversample input image with 1:1 balanced sampling with generated frames (ablation) | PERSONA · qualitative ablation | identity preservation | Improved |
| use only mean offsets, not scale offsets, for pose-driven deformation (ablation) | PERSONA · qualitative ablation | sharpness | Improved |
| render with PERSONA instead of diffusion-based animators | PERSONA · rendering speed | frames per second: 0.24-0.88 -> 25.56 | Improved |
Later papers that compared with this method
1 later accepted paper reports PERSONA as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Zero-Shot Reconstruction of Animatable 3D Avatars with Cloth Dynamics from a Single Image (CVPR 2026 · better on 9 of 9 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- AniGS — AniGS: Animatable Gaussian Avatar from a Single Image with Inconsistent Gaussian Reconstruction (CVPR 2025) · better on 3 of 3 shared results
- Champ — Champ: Controllable and Consistent Human Image Animation with 3D Parametric Guidance (ECCV 2024) · better on 2 of 3 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
- Visual Persona: Foundation Model for Full-Body Human Customization (CVPR 2025 · similarity 0.89)
- FRESA: Feedforward Reconstruction of Personalized Skinned Avatars from Few Images (CVPR 2025 · similarity 0.89)
- PEGASUS: Personalized Generative 3D Avatars with Composable Attributes (CVPR 2024 · similarity 0.88)
- Vid2Avatar-Pro: Authentic Avatar from Videos in the Wild via Universal Prior (CVPR 2025 · similarity 0.88)
- PERSE: Personalized 3D Generative Avatars from A Single Portrait (CVPR 2025 · similarity 0.88)
- En3D: An Enhanced Generative Model for Sculpting 3D Humans from 2D Synthetic Data (CVPR 2024 · 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.
