Research map / 3D vision / Face and shape
SfSNet: Learning Shape, Reflectance and Illuminance of Faces `in the Wild'
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 |
|---|---|---|---|
| Fine-tune SfSNet using Photoface ground-truth normals and albedo | SfSNet · Photoface | pixels <20°: 43.6% -> 83.7% | Helped |
| Train with SfS-supervision on synthetic and real images instead of synthetic data only (ablation) | SfSNet · CelebA | reconstruction MAE: 48.54 -> 10.99 | Helped |
| Train with SfS-supervision on synthetic and real images instead of synthetic data only (ablation) | SfSNet · CelebA | lighting rank-1 accuracy: 63.88% -> 78.44% | Helped |
| Use the residual-block decomposition and estimate lighting from image, normal, and albedo features rather than an image-only skip network (ablation) | SfSNet vs SkipNet+ · CelebA | lighting top-1 accuracy: 67.70% -> 78.44% | Helped |
| Use the residual-block decomposition and estimate lighting from image, normal, and albedo features rather than an image-only skip network (ablation) | SfSNet vs SkipNet+ · CelebA | reconstruction MAE: 11.33 -> 10.99 | Helped |
| Use SfSNet instead of LDAN for lighting estimation | MultiPIE lighting classification · MultiPIE | top-1 lighting classification accuracy: 65.87 -> 78.44 | Helped |
| Use SfSNet instead of Pix2Vertex | Normal estimation · Photoface | pixels <20°: 24.8% -> 43.6% | Helped |
Closest papers by meaning
- SFDM: Robust Decomposition of Geometry and Reflectance for Realistic Face Rendering from Sparse-view Images (CVPR 2025 · similarity 0.89)
- InverseFaceNet: Deep Monocular Inverse Face Rendering (CVPR 2018 · similarity 0.88)
- Learning to Decouple the Lights for 3D Face Texture Modeling (NeurIPS 2024 · similarity 0.87)
- Low-Light Face Super-resolution via Illumination, Structure, and Texture Associated Representation (AAAI 2024 · similarity 0.87)
- S^3-Face: SSS-Compliant Facial Reflectance Estimation via Diffusion Priors (CVPR 2025 · similarity 0.87)
- Monocular Identity-Conditioned Facial Reflectance Reconstruction (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.