Research map / 3D vision / Face and shape
On Learning 3D Face Morphable Model from In-the-Wild Images
Luan Tran, Xiaoming Liu
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 |
|---|---|---|---|
| add lighting modeling (disentangle shading from albedo) (ablation) | Nonlinear 3DMM · AFLW2000-3D face alignment | NME: 4.70 -> 4.30 | Improved |
| represent shape as a UV position map with CNN decoder instead of vector with MLP (ablation) | Nonlinear 3DMM · AFLW2000-3D face alignment | NME: 4.30 -> 4.12 | Improved |
| use nonlinear decoder instead of linear BFM shape basis | 3DMM shape · Basel scans shape representation (l_S=160) | NME: 0.0241 -> 0.0146 | Improved |
| use nonlinear albedo decoder instead of linear BFM albedo | 3DMM albedo · Albedo representation power | L1 error: 0.062 -> 0.053 | Improved |
| use nonlinear 3DMM instead of FaceWarehouse bilinear model | 3DMM shape · BU-3DFE expression scans (2500 scans) | Chamfer distance: 0.00197 -> 0.00083 | Improved |
| add albedo symmetry and constancy regularization (ablation) | Nonlinear 3DMM · qualitative shading/albedo separation | visual quality | Improved |
| add shape smoothness regularization (ablation) | Nonlinear 3DMM · qualitative shape quality | visual quality | Improved |
Closest papers by meaning
- Nonlinear 3D Face Morphable Model (CVPR 2018 · similarity 0.96)
- 3D Reconstruction of “In-the-Wild” Faces in Images and Videos (TPAMI 2018 · similarity 0.93)
- 3D Face Morphable Models "In-The-Wild" (CVPR 2017 · similarity 0.92)
- Learning Complete 3D Morphable Face Models From Images and Videos (CVPR 2021 · similarity 0.92)
- Beyond 3DMM: Learning to Capture High-Fidelity 3D Face Shape (TPAMI 2022 · similarity 0.91)
- Regressing Robust and Discriminative 3D Morphable Models With a Very Deep Neural Network (CVPR 2017 · similarity 0.91)
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.
