Research map / 3D vision / Face and shape
Learning a 3D Morphable Face Reflectance Model From Low-Cost Data
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 |
|---|---|---|---|
| Represent spatially varying specular intensity and shininess with weights over multiple Blinn-Phong BRDFs rather than one global exponent | Face reflectance model · Multi-PIE | LPIPS: 0.096 -> 0.083 | Helped |
| Represent spatially varying specular intensity and shininess with weights over multiple Blinn-Phong BRDFs rather than one global exponent | Face reflectance model · Multi-PIE | SSIM: 0.937 -> 0.942 | Helped |
| Represent spatially varying specular intensity and shininess with weights over multiple Blinn-Phong BRDFs rather than one global exponent | Face reflectance model · Multi-PIE | PSNR: 24.82 -> 25.00 | Helped |
| Add FFHQ update-by-reconstruction finetuning (ablation) | Initial Multi-PIE reflectance model · CelebA-HQ | LPIPS: 0.1256 -> 0.110 | Helped |
| Add FFHQ update-by-reconstruction finetuning (ablation) | Initial Multi-PIE reflectance model · CelebA-HQ | SSIM: 0.886 -> 0.896 | Helped |
| Add FFHQ update-by-reconstruction finetuning (ablation) | Initial Multi-PIE reflectance model · CelebA-HQ | PSNR: 23.26 -> 24.21 | Helped |
| Use the proposed reflectance model instead of AlbedoMM | Face reflectance reconstruction · 3D-RFE | Diffuse Albedo PSNR: 19.31 -> 20.13 | Helped |
| Use the proposed reflectance model instead of AlbedoMM | Face reflectance reconstruction · 3D-RFE | Specular Albedo PSNR: 26.51 -> 19.14 | Hurt |
| Use the proposed reflectance model and reconstruction pipeline instead of BFM09 | Face geometry reconstruction · NoW challenge validation set | Median geometry error (mm): 1.44 -> 1.51 | Hurt |
| Constrain predicted high-order SH lighting with a lighting PCA model rather than directly predicting 273 coefficients (ablation) | In-the-wild reflectance reconstruction · FFHQ qualitative lighting prediction | Lighting prediction quality | Helped |
| Add more mirror-like specular terms (ablation) | Face reflectance representation · Eye specularity | Eye reflectance modeling | No clear effect |
Closest papers by meaning
- S^3-Face: SSS-Compliant Facial Reflectance Estimation via Diffusion Priors (CVPR 2025 · similarity 0.88)
- Monocular Identity-Conditioned Facial Reflectance Reconstruction (CVPR 2024 · similarity 0.88)
- SFDM: Robust Decomposition of Geometry and Reflectance for Realistic Face Rendering from Sparse-view Images (CVPR 2025 · similarity 0.87)
- Diffusion Reflectance Map: Single-Image Stochastic Inverse Rendering of Illumination and Reflectance (CVPR 2024 · similarity 0.85)
- DiLiGenRT: A Photometric Stereo Dataset with Quantified Roughness and Translucency (CVPR 2024 · similarity 0.85)
- SpecNeRF: Gaussian Directional Encoding for Specular Reflections (CVPR 2024 · similarity 0.84)
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.