Research map / 3D vision / Face and shape
Disentangled Representation Learning for 3D Face Shape
Check 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 |
|---|---|---|---|
| Use deformation representation instead of Euclidean coordinates (ablation) | Ours · FaceWarehouse test | E_avd mean error: 0.939 -> 0.472 | Helped |
| Use deformation representation instead of Euclidean coordinates (ablation) | Ours · FaceWarehouse test | E_sed mean error: 0.447 -> 0.00333 | Helped |
| Add the fusion module (ablation) | Ours · FaceWarehouse test | E_exp mean error: 0.0582 -> 0.0388 | Helped |
| Add the fusion module (ablation) | Ours · FaceWarehouse test | E_sed mean error: 0.00283 -> 0.00333 | Hurt |
| Use deformation representation and fusion together (ablation) | Ours · FaceWarehouse test | E_avd mean error: 0.981 -> 0.472 | Helped |
| Use deformation representation and fusion together (ablation) | Ours · FaceWarehouse test | E_exp mean error: 0.170 -> 0.0388 | Helped |
| Use the proposed disentangled representation rather than FLAME | 3D face expression extrapolation · COMA | Average vertex distance mean error: 2.001 -> 1.643 | Helped |
| Use the proposed disentangled representation rather than FLAME | 3D face expression extrapolation · COMA | Average vertex distance median error: 1.615 -> 1.536 | Hurt |
| Increase training data with DR-based identity augmentation | Ours · FaceWarehouse and Bosphorus data comparison | E_avd: 18.3/18.0 -> 4.7/3.8 | Helped |
Closest papers by meaning
- Disentangling Features in 3D Face Shapes for Joint Face Reconstruction and Recognition (CVPR 2018 · similarity 0.90)
- 3D Shape Variational Autoencoder Latent Disentanglement via Mini-Batch Feature Swapping for Bodies and Faces (CVPR 2022 · similarity 0.88)
- ImFace: A Nonlinear 3D Morphable Face Model With Implicit Neural Representations (CVPR 2022 · similarity 0.87)
- DiLO: Disentangled Latent Optimization for Learning Shape and Deformation in Grouped Deforming 3D Objects (AAAI 2026 · similarity 0.87)
- Learning Complete 3D Morphable Face Models From Images and Videos (CVPR 2021 · similarity 0.86)
- ID-to-3D: Expressive ID-guided 3D Heads via Score Distillation Sampling (NeurIPS 2024 · similarity 0.85)
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.