Research map / 3D vision / Face and shape
Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars
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 FENeRF instead of pi-GAN | 3D-aware face generation · CelebA-HQ | FID: 14.7 -> 12.1 | Helped |
| Use FENeRF instead of Giraffe | 3D-aware face generation · FFHQ | FID: 31.9 -> 28.2 | Helped |
| Use FENeRF instead of pi-GAN | 3D-aware face generation · CelebA-HQ | KID (× 10^3): 3.9 -> 1.6 | Helped |
| Use FENeRF instead of pi-GAN | 3D-aware face generation · FFHQ | KID (× 10^3): 23.5 -> 17.3 | Helped |
| Add image rendering to the semantic-only model (ablation) | FENeRF · CelebAMask-HQ and FFHQ | geometry quality | Helped |
| Add semantic rendering to the image-rendering model (ablation) | FENeRF · CelebAMask-HQ and FFHQ | geometry quality | Helped |
| Add a learnable coordinate embedding (ablation) | FENeRF · CelebAMask-HQ and FFHQ | image detail | Helped |
| Inject coordinate embedding at the beginning of the MLP (ablation) | FENeRF · CelebAMask-HQ and FFHQ | image and semantic quality | Hurt |
| Inject coordinate embedding into the color branch (ablation) | FENeRF · CelebAMask-HQ and FFHQ | image and semantic quality | Helped |
Closest papers by meaning
- AniFaceGAN: Animatable 3D-Aware Face Image Generation for Video Avatars (NeurIPS 2022 · similarity 0.92)
- PanoHead: Geometry-Aware 3D Full-Head Synthesis in 360deg (CVPR 2023 · similarity 0.90)
- 3D Gaussian Head Avatars with Expressive Dynamic Appearances by Compact Tensorial Representations (CVPR 2025 · similarity 0.90)
- LeGO: Leveraging a Surface Deformation Network for Animatable Stylized Face Generation with One Example (CVPR 2024 · similarity 0.90)
- 3DToonify: Creating Your High-Fidelity 3D Stylized Avatar Easily from 2D Portrait Images (CVPR 2024 · similarity 0.88)
- What You See is What You GAN: Rendering Every Pixel for High-Fidelity Geometry in 3D GANs (CVPR 2024 · 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.