Research map / 3D vision / Face and shape
Learning Neural Parametric Head Models
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the proposed local neural-field ensemble with symmetry | NPHM · Identity reconstruction | L1-Chamfer: 0.200e-2 -> 0.182e-2 | Helped |
| Use the proposed local neural-field ensemble with symmetry | NPHM · Expression reconstruction | L1-Chamfer: 0.299e-2 -> 0.272e-2 | Helped |
| Reduce the local ensemble to 12 regions (ablation) | NPHM · Identity reconstruction | L1-Chamfer: 0.206 -> 0.289 | Hurt |
| Reduce the local ensemble to 12 regions (ablation) | NPHM · Identity reconstruction | F-Score@1.5: 0.938 -> 0.876 | Hurt |
| Use 26 regions instead of the full 39-region ensemble (ablation) | NPHM · Identity reconstruction | L1-Chamfer: 0.206 -> 0.237 | Hurt |
| Use 26 regions instead of the full 39-region ensemble (ablation) | NPHM · Identity reconstruction | F-Score@1.5: 0.938 -> 0.913 | Hurt |
| Remove symmetry sharing and the symmetry regularizer (ablation) | NPHM · Identity reconstruction | L1-Chamfer: 0.206 -> 0.230 | Hurt |
| Remove symmetry sharing and the symmetry regularizer (ablation) | NPHM · Identity reconstruction | Normal consistency: 0.976 -> 0.974 | Hurt |
| Remove symmetry sharing and the symmetry regularizer (ablation) | NPHM · Identity reconstruction | F-Score@1.5: 0.938 -> 0.917 | Hurt |
| Replace the local ensemble with a single anchor region (ablation) | NPHM · Identity reconstruction | L1-Chamfer: 0.206 -> 0.254 | Hurt |
| Replace the local ensemble with a single anchor region (ablation) | NPHM · Identity reconstruction | F-Score@1.5: 0.938 -> 0.906 | Hurt |
Closest papers by meaning
- ID-to-3D: Expressive ID-guided 3D Heads via Score Distillation Sampling (NeurIPS 2024 · similarity 0.88)
- FaceTalk: Audio-Driven Motion Diffusion for Neural Parametric Head Models (CVPR 2024 · similarity 0.87)
- DiffusionAvatars: Deferred Diffusion for High-fidelity 3D Head Avatars (CVPR 2024 · similarity 0.87)
- 3D Facial Expressions through Analysis-by-Neural-Synthesis (CVPR 2024 · similarity 0.86)
- MonoNPHM: Dynamic Head Reconstruction from Monocular Videos (CVPR 2024 · similarity 0.86)
- Gaussian Eigen Models for Human Heads (CVPR 2025 · similarity 0.86)
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.