Research map / 3D vision / Face and shape
High-Res Facial Appearance Capture From Polarized Smartphone Images
Dejan Azinović, Olivier Maury, Christophe Hery, Matthias Nießner, Justus Thies
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 the proposed polarized smartphone capture and texture reconstruction method instead of NLT | Facial appearance reconstruction · Held-out validation frames, 10 subjects | PSNR: 31.51 -> 32.37 | Improved |
| Use the proposed polarized smartphone capture and texture reconstruction method instead of NLT | Facial appearance reconstruction · Held-out validation frames, 10 subjects | SSIM: 0.96 -> 0.96 | No clear effect |
| Use the proposed polarized smartphone capture and texture reconstruction method instead of NLT | Facial appearance reconstruction · Held-out validation frames, 10 subjects | LPIPS: 0.11 -> 0.10 | Improved |
| Use the proposed polarized smartphone capture and texture reconstruction method instead of NextFace | Facial appearance reconstruction · Held-out validation frames, 10 subjects | PSNR: 22.85 -> 32.37 | Improved |
| Use the proposed polarized smartphone capture and texture reconstruction method instead of NextFace | Facial appearance reconstruction · Held-out validation frames, 10 subjects | SSIM: 0.89 -> 0.96 | Improved |
| Use the proposed polarized smartphone capture and texture reconstruction method instead of NextFace | Facial appearance reconstruction · Held-out validation frames, 10 subjects | LPIPS: 0.31 -> 0.10 | Improved |
| Add direction-dependent flashlight attenuation modeling (ablation) | Facial texture reconstruction · Re-rendered images | image error | Improved |
| Use a non-Lambertian diffuse BRDF that models the Fresnel effect (ablation) | Diffuse texture reconstruction · Face texture reconstruction | diffuse texture accuracy | Improved |
| Add cross-polarized capture (ablation) | Facial appearance reconstruction · Face texture reconstruction | diffuse/specular separation | Improved |
| Use two-step texture optimization instead of jointly optimizing all textures (ablation) | Facial appearance reconstruction · Purely diffuse render | diffuse texture fidelity | Improved |
| Use coarse-to-fine optimization rather than direct highest-resolution optimization (ablation) | Facial appearance reconstruction · Face texture reconstruction | texture sharpness | Improved |
| Use photographs in texture optimization rather than video frames only (ablation) | Facial appearance reconstruction · Face texture reconstruction | texture quality | Improved |
Closest papers by meaning
- WildCap: Facial Albedo Capture in the Wild via Hybrid Inverse Rendering (CVPR 2026 · similarity 0.88)
- Practical and Scalable Desktop-Based High-Quality Facial Capture (ECCV 2022 · similarity 0.87)
- Monocular Facial Appearance Capture in the Wild (ICCV 2025 · similarity 0.86)
- A Morphable Face Albedo Model (CVPR 2020 · similarity 0.86)
- S^3-Face: SSS-Compliant Facial Reflectance Estimation via Diffusion Priors (CVPR 2025 · similarity 0.84)
- Glossy Object Reconstruction with Cost-effective Polarized Acquisition (CVPR 2025 · 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.
