Research map / 3D vision / Face and shape
MoFaNeRF: Morphable Facial Neural Radiance Field
Yiyu Zhuang, Hao Zhu, Xusen Sun, Xun Cao
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 |
|---|---|---|---|
| use PCA shape code instead of one-hot or learnable (ablation) | MoFaNeRF · Ablation | PSNR (dB): 26.27 -> 26.57 vs one-hot; 25.24 -> 26.57 vs learnable | Improved |
| use learnable expression code instead of one-hot or PCA (ablation) | MoFaNeRF · Ablation | PSNR (dB): 25.59 -> 26.57 vs one-hot; 25.79 -> 26.57 vs PCA | Improved |
| add appearance code and texture encoding module (TEM) (ablation) | MoFaNeRF · Ablation | PSNR (dB): 25.73 -> 26.57 | Improved |
| use modulation-based architecture instead of deformation-based (ablation) | MoFaNeRF · Ablation | PSNR (dB): 24.22 -> 25.69 | Improved |
| use modulation-based architecture instead of hybrid (ablation) | MoFaNeRF · Ablation | PSNR (dB): 24.42 -> 25.69 | Improved |
| add identity-specific modulation (ISM) (ablation) | MoFaNeRF · Ablation | PSNR (dB): 25.69 -> 26.57 | Improved |
| add landmark-based sampling instead of uniform sampling (ablation) | MoFaNeRF · Ablation | PSNR (dB): 25.67 -> 26.57 | Improved |
| add RefineNet (GAN and perceptual loss) (ablation) | MoFaNeRF · FaceScape representation ability | LPIPS / PSNR / SSIM: LPIPS 0.061 -> 0.034; PSNR 31.49 -> 30.17; SSIM 0.951 -> 0.935 | Mixed |
| add RefineNet (ablation) | MoFaNeRF · Identity consistency (changing view) | DFID: 0.687 -> 0.707 | Hurt |
| MoFaNeRF versus FaceScape bilinear model | MoFaNeRF · FaceScape representation ability | PSNR (dB): 27.96 -> 31.49 | Improved |
Later papers that compared with this method
3 later accepted papers report MoFaNeRF as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- GPHM: Gaussian Parametric Head Model for Monocular Head Avatar Reconstruction (TPAMI 2025 · better on 3 of 3 shared results)
- 3D Gaussian Parametric Head Model (ECCV 2024 · better on 2 of 2 shared results)
- Head360: Learning a Parametric 3D Full-Head for Free-View Synthesis in 360° (ECCV 2024 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- i3DMM — i3DMM: Deep Implicit 3D Morphable Model of Human Heads (CVPR 2021) · better on 3 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
- HeadNeRF: A Real-Time NeRF-Based Parametric Head Model (CVPR 2022 · similarity 0.87)
- Learning Dynamic Facial Radiance Fields for Few-Shot Talking Head Synthesis (ECCV 2022 · similarity 0.87)
- Dynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstruction (CVPR 2021 · similarity 0.87)
- StrokeFaceNeRF: Stroke-based Facial Appearance Editing in Neural Radiance Field (CVPR 2024 · similarity 0.87)
- FFaceNeRF: Few-shot Face Editing in Neural Radiance Fields (CVPR 2025 · similarity 0.86)
- LAENeRF: Local Appearance Editing for Neural Radiance Fields (CVPR 2024 · 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.
